Effects of Food Assistance and Nutrition Programs on Nutrition and Health: Volume 3, Literature Review

Effects of Food Assistance and Nutrition Programs on Nutrition and Health: Volume 3, Literature Review. Edited by Mary Kay Fox and William Hamilton, Abt Associates Inc., and Biing-Hwan Lin, Food and Rural Economics Division, Economic Research Service, U.S. Department of Agriculture. Food Assistance and Nutrition Research Report No. 19-3.

Abstract This report provides a comprehensive review and synthesis of published research on the impact of USDA’s domestic food and nutrition assistance programs on participants’ nutrition and health outcomes. The outcome measures reviewed include food expenditures, household nutrient availability, dietary intake, other measures of nutrition status, food security, birth outcomes, breastfeeding behaviors, immunization rates, use and cost of health care services, and selected nonhealth outcomes, such as academic achievement and school performance (children) and social isolation (elderly). The report is one of four volumes produced by a larger study that includes Volume 1, Research Design; Volume 2, Data Sources; Volume 3, Literature Review; and Volume 4, Executive Summary of the Literature Review. The review examines the research on 15 USDA food assistance programs but tends to focus on the largest ones for which more research is available: food stamps, school feeding programs, and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). Over half of USDA's budget—$41.6 billion in fiscal year 2003—was devoted to food assistance and nutrition programs that provide low-income families and children with access to a healthy diet.

Keywords: Dietary intake, food expenditures, nutrient availability, nutrient intake, nutritional status, nutrition and health outcomes, USDA's food assistance and nutrition programs

Washington, DC 20036 October 2004 Acknowledgments Many individuals deserve recognition for their roles in making this report a reality. First and foremost are the chapter authors. Without their tireless efforts, this report would not exist. Authors include current and former Abt Associates staff—Joy Behrens, Nancy Burstein, David Connell, Mary Kay Crepinsek, Mary Kay Fox, Frederic Glantz, Cristofer Price, and William Hamilton—as well as consultants— Virginia Casey, John Cook, Peter H. Rossi, and Joanne Tighe.

We also owe a debt of gratitude to a cadre of reviewers who offered thoughtful review and comment on one or more chapters of the report in one or more drafts. Their contributions improved the report. Reviewers included a team of senior advisors (Johanna Dwyer, Peter H. Rossi, and Rick Trowbridge), a group of external technical experts (Janet Currie, Barbara Deveney, Shiriki Kumanyika, Suzanne Murphy, Barry Popkin, and Mike Puma), other Abt staff members (Nancy Cole, Joan McLaughlin), and staff at USDA’s Economic Research Service (Jane Reed, Betsy Frazao, Linda Ghelfi, Craig Gundersen, Joanne Guthrie, Bill Levedahl, Vic Oliveira, Mark Prell, David Smallwood, Laura Tiehen, Jay Variyam, and Parke Wilde), Food and Nutrition Service (Lisa Ramirez-Branum, Steven Carlson, Edward Herzog, Jay Hirshman, Patricia McKinney, Anita Singh, and Tracy von Ins), and Center for Nutrition Policy and Promotion (Peter Basiotis and Andrea Carlson). We offer special thanks to Johanna Dwyer and Betsy Frazao who took on reviewing tasks that were larger than most and who offered extensive and useful feedback.

Sharon Christenson and Daniel Singer deserve special recognition for coordinating the literature search as well as the document retrieval process. And, finally, several people at Abt Associates and the Economic Research Service (ERS) deserve our gratitude for managing production and editing of the report. Eileen MacEnaney and Eileen Fahey coordinated production of the report at Abt Associates. Linda Hatcher, Courtney Knauth, Sharon Lee, Tom McDonald, Victor Phillips, Mary Reardon, Dale Simms, Priscilla Smith, and John Weber completed the cover design and final editing and production at ERS.

We sincerely appreciate the efforts of all these colleagues.

Mary Kay Fox William Hamilton Biing-Hwan Lin

ii Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Contents Page

List of Tables...... vi

Chapter 1, Introduction...... 1 Mary Kay Fox, William Hamilton, and Sharon Christenson Identifying Relevant Research for Review ...... 3 Organization of This Report ...... 7 References...... 12

Chapter 2, Research Methods...... 13 William Hamilton, Mary Kay Fox, and Peter H. Rossi Evaluation Design...... 13 Outcome Measures ...... 19 References...... 27

Chapter 3, Food Stamp Program...... 30 Nancy Burstein, Cristofer Price, Peter H. Rossi, and Mary Kay Fox Program Overview...... 30 Assessing Impacts of the Food Stamp Program...... 33 Food Expenditures...... 35 Household Nutrient Availability...... 47 Individual Dietary Intake...... 56 Other Nutritional and Health Outcomes...... 77 Summary...... 84 References...... 86

Chapter 4, WIC Program...... 91 Mary Kay Fox Program Overview...... 91 Research Overview...... 94 Impacts of WIC Prenatal Participation...... 97 Impacts of WIC Participation on Infants and Children...... 131 Impacts of WIC Participation on WIC Households or Undifferentiated WIC Participants ...... 165 Summary...... 166 References...... 168

Chapter 5, National School Lunch Program ...... 175 Mary Kay Crepinsek and Mary Kay Fox Program Overview...... 175 Research Overview...... 178 Impacts on Dietary Intake ...... 179 Impacts on Other Nutrition- and Health-Related Outcomes...... 197 Summary...... 206 References...... 208

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 iii Page

Chapter 6, School Breakfast Program ...... 211 David Connell and Mary Kay Fox Program Overview...... 211 Research Overview...... 214 Summary...... 233 References...... 234

Chapter 7, Child and Adult Care Food Program ...... 236 Frederic Glantz Program Overview...... 236 Review of Research on the Child Care Component of CACFP ...... 239 Review of Research on the Adult Care Component of CACFP ...... 245 Summary...... 246 References...... 248

Chapter 8, Summer Food Service Program ...... 250 Joy Behrens and Mary Kay Fox Program Overview...... 250 Research Review ...... 252 Summary...... 254 References...... 255

Chapter 9, The Emergency Food Assistance Program...... 256 John Cook Program Overview...... 256 Research Review ...... 256 Summary...... 259 References...... 260

Chapter 10, Nutrition Services Incentive Program ...... 261 Virginia Casey and Mary Kay Crepinsek Program Overview...... 261 Research Overview...... 263 Research Results...... 268 Summary...... 282 References...... 284

Chapter 11, Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas...... 286 Cristofer Price Program Overview...... 286 Research Review ...... 287 Summary...... 290 References...... 292

iv Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Page

Chapter 12, Commodity Supplemental Food Program...... 293 Joy Behrens Program Overview...... 293 Research Review ...... 294 Summary...... 295 References...... 296

Chapter 13, Food Distribution Program on Indian Reservations...... 297 John Cook Program Overview...... 297 Research Review ...... 298 Summary...... 301 References...... 302

Chapter 14, WIC Farmers’ Market Nutrition Program ...... 304 Joy Behrens Program Overview...... 304 Research Review ...... 305 Summary...... 306 References...... 307

Chapter 15, Special Milk Program...... 308 Joy Behrens Program Overview...... 308 Research Review ...... 309 Summary...... 310 References...... 311

Chapter 16, Team Nutrition Initiative and Nutrition Education and Training Program ...... 312 Joanne Tighe Overview of the Team Nutrition Initiative ...... 312 Research Review of the Team Nutrition Initiative...... 313 Overview of the Nutrition Education and Training Program...... 315 Research Review of the Nutrition Education and Training Program...... 316 Summary...... 318 References...... 320

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 v List of Tables Table Page

1 Federal food assistance and nutrition programs...... 2 2 Searchable databases used in computerized literature search...... 4 3 Program names, acronyms, and variants used in computerized ...... literature search...... 5 4 Keywords used in querying searchable databases ...... 6 5 Studies that examined the impact of the Food Stamp Program on dietary intakes of individuals [SAMPLE TABLE]...... 9 6 Findings from studies that examined the impact of the Food Stamp Program on dietary intakes of individuals [SAMPLE TABLE]...... 10 7 Maximum monthly food stamp benefits before deductions, FY 2003 ...... 32 8 Studies that examined the impact of the Food Stamp Program on household food expenditures ...... 37 9 Findings from studies that examined the impact of the Food Stamp Program on household food expenditures using participant vs. nonparticipant comparisons ...... 43 10 Findings from studies that examined the impact of the Food Stamp Program on household food expenditures using dose-response analyses.....44 11 Findings from studies that examined the impact of food stamp cashout on household food expenditures...... 46 12 Studies that examined the impact of the Food Stamp Program on household availability of food energy and nutrients ...... 48 13 Findings from studies that examined the impact of the Food Stamp Program on household availability of food energy and nutrients ...... 52 14 Studies that examined the impact of the Food Stamp Program on dietary intakes of individuals ...... 58 15 Findings from studies that examined the impact of the Food Stamp Program on dietary intakes of individuals...... 63 16 Studies that examined the impact of the Food Stamp Program on other nutrition and health outcomes ...... 78 17 Studies that examined the impact of prenatal WIC participation on birth outcomes, including associated health care costs...... 98 18 Findings from studies that examined the impact of prenatal WIC participation on birth outcomes, including associated health care costs.....109 19 Studies that examined the impact of the WIC program on breastfeeding .....116 20 Findings from studies that examined the impact of the WIC program on breastfeeding...... 119 21 Studies that examined the impact of the WIC program on nutrition and health outcomes of pregnant women...... 123 22 Findings from studies that examined the impact of the WIC program on the dietary intakes of pregnant women ...... 127 23 Studies that examined the impact of the WIC program on nutrition and health outcomes of infants and children...... 132 24 Findings from studies that examined the impact of the WIC program on the dietary intakes of children and/or infants ...... 143 25 Findings from studies that examined the impact of the WIC program on other nutrition, health, and developmental outcomes of infants and/or children ...... 153 26 Studies that examined the impact of the National School Lunch Program on students' dietary intakes ...... 180 vi Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Table Page

27 Findings from studies that examined the impact of the National School Lunch Program on students' dietary intakes at lunch...... 186 28 Findings from studies that examined the impact of the National School Lunch Program on students' dietary intakes over 24 hours ...... 192 29 Findings from studies that examined the impact of the National School Lunch Program on students' food consumption patterns ...... 198 30 Studies that examined the impact of the National School Lunch Program on other nutrition and health outcomes ...... 201 31 Findings from studies that examined the impact of the National School Lunch Program on other nutrition and health outcomes...... 203 32 Studies that examined the impact of the School Breakfast Program on students’ dietary intakes...... 215 33 Low-income students' breakfast consumption patterns by SBP availability ...... 218 34 Findings from studies that examined the impact of the School Breakfast Program on students' dietary intakes at breakfast...... 219 35 Findings from studies that examined the impact of the School Breakfast Program on students' dietary intakes over 24 hours ...... 223 36 Studies that examined the impact of universal-free breakfast programs on school performance and behavioral/cognitive outcomes ...... 229 37 Findings from studies that examined the impact of universal-free breakfast programs on school performance and behavioral/cognitive outcomes ...... 232 38 Meal reimbursement for homes and centers, July 1, 2002-June 30, 2003 ....238 39 Monthly administrative cost reimbursement rates for sponsors of family child care homes, July 1, 2002-June 30, 2003...... 238 40 Studies that examined the nutrient content of meals and snacks offered in the Child and Adult Care Food Program and/or the nutrient contribution of meals and snacks consumed by program participants...... 240 41 Studies that examined the impact of the Elderly Nutrition Program on nutrition and health outcomes ...... 264 42 Findings from studies that examined the impact of the Elderly Nutrition Program on participants' dietary intakes...... 270 43 Findings from studies that examined the impact of the Elderly Nutrition Program on biochemical indicators of nutritional status...... 277 44 Findings from studies that examined the impact of the Elderly Nutrition Program on participants' weight status...... 280 45 Level 1 nutrition screen from the Nutrition Screening Initiative ...... 282 46 Studies that examined the impact of the Nutrition Assistance Program in Puerto Rico on household food expenditures and/or nutrient availability...... 288

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 vii

Chapter 1 Introduction

Since the mid-1940s, when concerns about the nutri- Work Opportunity Reconciliation Act of 1996 tional status of young men drafted for service in World (PRWORA)—resulted in significant changes to several War II led to establishment of the National School FANPs. Most of these changes tightened eligibility Lunch Program (NSLP), the U.S. Government has standards and/or reduced benefit levels. committed to ensuring that its citizens neither go hungry nor suffer the consequences of inadequate dietary The continued pressures of welfare reform, and the intake.1,2 Over the years, many Federal programs have increased accountability encompassed in the been deployed to meet this commitment. Today, the Government Performance and Results Act (GPRA), Federal nutrition safety net includes 16 distinct food are certain to lead to heightened scrutiny of all Federal assistance and nutrition programs (FANPs) (table 1). assistance programs. In the past, much of the assessment Administered by the Food and Nutrition Service of FANPs centered on issues related to program opera- (FNS), U.S. Department of Agriculture (USDA), the tions, such as whether only eligible participants received 16 programs together were funded at approximately benefits. Future program reviews are likely to be more $38 billion in fiscal year (FY) 2002.3 An estimated one broadly based, to focus on program effectiveness, and in five Americans participated in one or more FANPs to ask if the program is achieving its objectives. at some point during FY 2002 (Oliveira, 2003). Recent program policies have emphasized the nutrition Although FANPs vary greatly in size, target popula- focus of the FANPs, which separates them from other tion, and benefit-delivery strategy, all provide children federally sponsored income support programs. Indeed, or low-income households with food, the means to in FY 1998, FNS made a “renewed commitment to purchase food, and/or nutrition education. Several pro- nutrition education in all FNS programs” and established grams also provide avenues for disbursement of sur- a special staff within the agency to “refocus efforts plus agricultural commodities. All FANPs share the toward nutrition and nutrition education” (USDA/FNS, main goal of ensuring the health of vulnerable 2003). The growing emphasis on nutrition education in Americans by providing access to a nutritionally ade- the Food Stamp Program (FSP) is one example of this quate diet. renewed commitment. In FY 1992, only five States had approved State plans for FSP nutrition education, In recent years, the efficacy of the web of programs that and the Federal share of expenditures for FSP nutrition make up the nutrition safety net has been questioned. In education was $661,000. In FY 2002, 48 State agen- 1996, during the throes of welfare reform, Congress cies had approved FSP nutrition education plans and seriously considered abolishing key components of the Federal expenditures for FSP nutrition education current Federal system in favor of block grants to States. exceeded $174 million (USDA/FNS, 2003). Most of While this initiative was ultimately defeated, welfare this increase occurred after 1998 (Speshock, 1999). reform—specifically the Personal Responsibility and A further example of the renewed focus on nutrition in

1Many World War II draftees who were rejected had nutrition-related the FANPs is the set of goals and core objectives problems, including stunted growth, missing or rotted teeth, and physical defined in the FNS strategic plan for 2000-05 deformities associated with rickets or other severe nutritional deficiencies (USDA/FNS, 2000). One of two key goals is during infancy and childhood. “improved nutrition for children and low-income peo- 2The earliest version of a federally operated food assistance and nutrition programs was actually the New Deal food stamp program (operated in ple.” Core objectives under this goal include improv- the 1930s). This program allowed poor households to purchase stamps that ing food security, promoting healthy food choices were redeemable for most foods. Households also received a supply of free among FANP participants, and improving the quality bonus stamps that were redeemable for selected surplus commodities. The New Deal food stamp program was discontinued during World War II. of meals, food packages, commodities, and other pro- 3The list of FANPs used in this report differs slightly from the list used gram benefits. by FNS. FNS considers the Nutrition Education and Training Program and Team Nutrition to be part of the National School Lunch and School In recognition of both the renewed emphasis on nutri- Breakfast Programs. FNS also operates the Disaster Relief Program, a program that is not considered in this review because its role in the nutrition tion and nutrition education in the FANPs and the safety net is substantively different from that of the other FANPs. increasing Federal focus on program accountability,

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 1 Chapter 1: Introduction

Table 1—Federal food assistance and nutrition programs Year FY 2002 1 2 2 Program begun costs FY 2002 participation

$ millions 3 4 National School Lunch Program (NSLP) 1946 6,857 28,006,873 lunches per day Special Milk Program (SMP) 1955 16 112,781,614 total half-pints

Commodity Supplemental Food Program (CSFP) 1968 110 427,444 participants per month Summer Food Service Program (SFSP) 1968 263 121,865,417 total meals and snacks Food Stamp Program (FSP) 1974 20,677 19,099,524 participants 5 per month 6 Special Supplemental Nutrition Program 1975 4,319 7,490,841 participants for Women, Infants, and Children (WIC) per month 4 School Breakfast Program (SBP) 1975 1,566 8,144,384 breakfasts per day 7 8 Nutrition Services Incentive Program (NSIP) 1975 152 252,748,643 total meals Nutrition Education and Training Program (NET) 1977 0 0 Food Distribution Program on 1977 69 110,122 participants per month Indian Reservations (FDPIR) 9 4 Child and Adult Care Food Program (CACFP) 1978 1,852 1,691,448,979 total child meals and snacks; 44,570,764 total adult meals and snacks 10 Nutrition Assistance Program for Puerto Rico, 1981 1,362 Not available American Samoa, and the Northern Marianas (NAP) 11 12 The Emergency Food Assistance Program (TEFAP) 1981 435 611 million total pounds of food distributed 13 13 WIC Farmers’ Market Nutrition Program (FMNP) 1992 25 2+ million total participants 14 Team Nutrition Initiative (TN) 1995 10 Not available 15 Senior Farmers’ Market Nutrition Program (SFMNP) 2002 13 Not available 1 Year of permanent authorization. Several food assistance and nutrition programs started as pilot projects before being established as permanent programs. 2 Unless otherwise noted, data on costs and participation were obtained from USDA/FNS administrative data for FY 2002 (http://www.fns.usda.gov/pd, accessed April 2003). Reported costs include all cash benefits/reimbursements, food/commodity costs (as applicable), and administrative costs. 3 In 1998, the program began covering snacks served in after-school programs. In FY 2002, a total of 122,914,873 snacks were served. 4 In FY 2002, an additional $124 million was spent on State administrative expenses for the NSLP, the SBP, and the CACFP. 5 Individuals in participating households. 6 Excludes estimated cost of WIC Farmers’ Market Nutrition Program (FMNP), based on FY 2002 appropriation for FMNP. 7 Formerly known as the Nutrition Program for the Elderly (NPE). In FY 2003, administration for the program was transferred to the U.S. Department of Health and Human Services. FNS continues to supply commodities and financial support to the program. 8 Total meals for FY 2001, the latest year for which FNS collected data. 9 The adult day care component was added in 1989. In 1999, the program expanded to serve children living in homeless shelters. 10 The FY 2002 grant for Puerto Rico was $1,351 million, the grant for American Samoa was $5.3 million, and the grant for the Northern Marianas was $6.1 million. 11 Until 1996, FNS operated a separate Commodity Distribution Program for Charitable Institutions, Soup Kitchens, and Food Banks. Under the Personal Responsibilities and Work Opportunities Reconciliation Act (PRWORA), this program was merged into TEFAP. 12 In FY 2002, FNS donated an additional $16 million in commodities to disaster relief and charitable institutions. 13 Cost reflects FY 2003 appropriation. Source: http://www.fns.usda.gov/wic/FMNP/FMNPfags.htm, accessed April 2003. 14 FY 2002 appropriation. Source: L. French (2002). Personal communication. 15 Based on FY 2002 appropriation ($15 million) and residual carried over into FY 2003 ($1.7 million). Source: http://www.fns.usda.gov/wic/ Senior FMNP/SFMNPFY02.htm and SFMNPFY03.htm, accessed April 2003.

2 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 1: Introduction

USDA’s Economic Research Service (ERS) contracted The search completed for this summary emphasized with Abt Associates Inc. to conduct the Nutrition and recall over precision. In essence, it was accepted that Health Outcomes Study. A major focus of the study staff would need to weed through numerous irrelevant was a comprehensive review and synthesis of existing citations to identify literature that was truly representa- research on the impact of FANPs on nutrition- and tive of the existing research. The search was highly health-related outcomes. This report presents results of inclusive and used overlapping search methods. The that effort.4 selection of searchable databases and search terms (keywords) were both carefully considered, as described below. The actual search was carried out by Identifying Relevant a research librarian with extensive experience in sup- Research for Review porting social science research. The objective of the literature review was to summa- Selecting Searchable Databases rize current knowledge about the effects on FANP participation on nutrition- and health-related outcomes. The first step in selecting databases was to define rele- The first step was a comprehensive literature search. vant disciplines (or fields of study) and research sub- The approach to identifying empirical studies to be ject areas. After a careful review of available databases included in the research summary followed principles and their topical coverage, the following list of disci- in The Handbook of Research Synthesis (Cooper and plines/subject areas was defined: Hedges, 1994). This text is generally accepted as a definitive reference on research synthesis. The corner- • Medicine and health stone of the process is a comprehensive computerized • Nutrition search of bibliographic databases. The following sec- • Nursing and allied health tions describe the methods used to conduct the com- • Health economics puterized search and the steps taken to cross-check and • Health education expand the resulting list of citations. • Social science research • Agricultural research, economics, and policy Computerized Literature Search • Education research • Social services and public welfare In defining parameters for a literature search, two key • Public health concerns are recall and precision (White, 1992). Recall refers to the hypothetical percentage of all rele- These subject areas were used to select a group of vant citations that are actually identified through the searchable databases. The initial subject-specific list search. Precision refers to the percentage of identified was expanded to include a number of more general citations that are ultimately judged relevant to the databases targeted toward “gray” or unpublished research synthesis. Precision and recall tend to vary research, including those that cover dissertations, con- inversely. A search designed to yield a high recall will ferences, foundation grants, ongoing research projects, invariably have less precision—that is, it will yield and government documents. A total of 26 databases numerous irrelevant references. On the other hand, a was included in the online search (table 2). search designed to be highly precise will yield fewer, more focused references but will run a greater risk of The Dialog Information Retrieval Service (Dialog) missing relevant research. was selected as the main vehicle for the search. Among information retrieval services, Dialog provides access to the largest number of social science research 4A separate summary report (Fox and Hamilton, 2004) presents major databases via a single, integrated user interface. findings from each of the detailed chapters included in this report. In addi- Indeed, as noted in table 2, Dialog provided direct tion, the Nutrition and Health Outcomes Study produced six other reports. One report reviews the research designs available to researchers interested access to all but three of the selected databases. It also in studying the effects of FANPs (Hamilton and Rossi, 2002) and another provides such special features as the capability to describes existing data sources that might be useful in these endeavors search multiple databases simultaneously and to (Logan et al., 2002). The four other reports summarize the nutrition and health characteristics of low-income populations, using data from the third remove duplicates as they occur across databases. National Health and Nutrition Examination Survey (NHANES-III). The reports cover FSP participants and nonparticipants (Fox and Cole, 2004a), Defining Search Parameters participants and nonparticipants in the Special Supplemental Nutrition Program for Women, Infants, and Children (Cole and Fox, 2004a), school- Because the search was so large and complex, it was age children (Fox and Cole, 2004b), and older adults (Cole and Fox, 2004b). completed in two waves. The 26 databases were divided

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 3 Chapter 1: Introduction

Table 2—Searchable databases used in computerized literature search 1 Database name Database producer Subject category

Ageline American Association of Social services and Retired Persons public welfare Agricultural Online Access U.S. National Agricultural Agricultural research; (AGRICOLA) Research Library economics; policy Biological and Agricultural Index (BAI) H.W. Wilson Company Agricultural research

Combined Health Information U.S. National Institutes of Health Health education; 1 Database (CHID) public health Computer Retrieval of Information U.S. National Institutes of Health Public health; medicine 2 on Scientific Projects (CRISP) and health Conferences Papers Index Cambridge Scientific Abstracts General

Current Research Information System (CRIS) U.S. Department of Agriculture Nutrition

Dissertation Abstracts Online University Microfilms, Inc. General

Economic Literature Index (EconLit) American Economic Association Health economics

Education Research U.S. Department of Education Education research Information Center (ERIC) Excerpta Medica (EMBASE) Elsevier Science; Netherlands Medicine and health; health economics; public health Federal Research in Progress (FEDRIP) U.S. National Technical General Information Service Foundation Grants Index The Foundation Center General

GPO Monthly Catalogue U.S. Government Printing Office General

Health and Wellness Database (HPD) Information Access Company Medicine and health; nutrition

HealthStar U.S. National Library of Medicine Health economics

Inside Conferences British Library General

MEDLINE U.S. National Library of Medicine Medicine and health; nutrition

National Technical Information Service U.S. National Technical General Bibliographic Database Information Service 3 Nursing and Allied Health Database Cinahl Information Systems Nursing and allied health; medicine and health; nutrition Nutrition Abstracts and Reviews, Series A: CAB International; England Nutrition Human and Experimental PAIS International Public Affairs Information Service Social science research

Social Sciences Index H.W. Wilson Company Social science research

Social Sciences Abstracts H.W. Wilson Company Social science research

Social SciSearch Institute for Scientific Information Social science research

Sociological Abstracts Sociological Abstracts, Inc. Social services and public welfare

1 Searched via Dialog, except as noted. 2 Searched via the Worldwide Web. 3 Searched via Data Star.

4 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 1: Introduction into two groups and each group was searched inde- sets of citations were created by merging results of the pendently. Databases were grouped to minimize over- two search waves and removing duplicate records. lap; that is, those likely to yield duplicate records were grouped together to permit removal of duplicates Identifying Relevant References before citations were downloaded. All of the citations generated by the search were initially captured in a “browsing format” that provided For each set of databases, 16 separate searches were title and indexing information (keywords used in conducted—one for each program listed in table 3, as indexing the citation in the database) without the cost well as one using the generic terms “nutrition assis- of retrieving a full citation. These abbreviated citations tance,” “food assistance,” “nutrition supplementation,” were manually reviewed by chapter authors to identify and “nutrition education.” Each search included all of sources that were potentially relevant for the research the search terms identified in table 4. review. Because the focus of the literature review was the impact/effect of FANPs on nutrition and health Searches were limited to English language documents outcomes, citations deemed potentially relevant were and to records from 1973 to 2002.5 Program-specific those that appeared to summarize research comparing program participants with nonparticipants. All citations 5The initial search was conducted in 1999. The bibliography was updat- selected for further review were downloaded in full ed in 2002, before preparation of the final version of the report. The 2002 update included only published research. Additional published research format, consisting of a complete citation and, where was incorporated before final publication in 2004. available, an abstract.

Table 3—Program names, acronyms, and variants used in computerized literature search

3 Child and Adult Care Food Program (CACFP) Nutrition Program for the Elderly (NPE) Child Care Feeding/Food Program (CCFP) Elderly Feeding Program Adult Care Feeding/Food Program Elderly Nutrition Program 1 Homeless Children Nutrition Program 1 Child Nutrition Homeless Demonstration Project School Breakfast Program (SBP) Breakfast Program Commodity Distribution to Charitable 2 Institutions, Soup Kitchens, and Food Banks Special Milk Program (SMP) 2 Commodity Distribution Program Supplemental Milk Program 2 Commodity Donation Program Special Supplemental Nutrition Program for Commodity Supplemental Food Program (CSFP) Women, Infants, and Children (WIC) Special Supplemental Food Program for Food Distribution Program on Indian Reservations (FDPIR) Women, Infants, and Children WIC program Food Stamp Program (FSP) Food Stamps Summer Food Service Program (SFSP) Summer Feeding Program National School Lunch Program (NSLP) School Lunch Program Team Nutrition (TN) Team Nutrition Initiative (TNI) Nutrition Assistance Program for Puerto Rico and the Northern Marianas (NAP) Temporary Emergency Food Puerto Rico/Puerto Rican Nutrition Assistance Program (TEFAP) Assistance Program Emergency Feeding Program Emergency Food Program Nutrition Education and Training (NET) Nutrition Education and Training Program (NETP) WIC Farmers’ Market Nutrition Program 4 Farmers’ Market Nutrition Program 1 In July 1999, the Homeless Children Nutrition Program was discontinued as a separate program and formally incorporated into the CACFP. 2 Under PRWORA, the previously separate Commodity Distribution to Charitable Institutions, Soup Kitchens, and Food Banks Program was combined with the Temporary Emergency Food Assistance Program to form The Emergency Food Assistance Program (TEFAP). 3 In 2001, the Nutrition Program for the Elderly (NPE) was renamed the Nutrition Services Incentive Program (NSIP). 4 The Senior Farmers’ Market Nutrition Program was not included in the search because the program was not established until 2002.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 5 Chapter 1: Introduction

Table 4—Keywords used in querying searchable databases General terms Specific terms

Food/nutrient availability Breakfast consumption Dietary trends Food purchases Food/nutrient intake Diet Dietary variety Food selections Food/nutrient consumption Dietary adequacy Eating behaviors Food use Dietary effects Eating practices Healthy Eating Index (HEI) Dietary impacts Folic acid Nutrient availability Dietary intake Food choices Nutrient content Dietary outcomes Food consumption Nutrient intake Dietary quality Food costs Nutritional adequacy Dietary patterns Food expenditures Nutritional intake Dietary practices Food intake

Health-related behaviors Alcohol use Cow’s milk (use of) Infant feeding practices Health-related practices Breastfeeding Drug abuse Perinatal care Breast feeding Drug use Prenatal care Cigarette (tobacco) use Immunizations Smoking

Pregnancy and Birthweight Length of gestation Neural tube defects birth outcomes Birth weight Light-for-date infants Perinatal morbidity Fetal growth Low birthweight Pregnancy Fetal outcomes Low birth weight Pregnancy outcome(s) Gestational age Low birth-weight Prematurity Head circumference Maternal morbidity Preterm delivery Infant morbidity Maternal mortality Preterm infants Infant mortality Maternal weight gain Very low birthweight Intrauterine growth retardation Neonatal morbidity Very low birth weight Neonatal mortality Very low-birthweight

Nutrition/health status Allergies Health outcome(s) Mortality Nutrition outcomes Anemia Health status Nutrition Health outcomes Body Mass Index (BMI) Height Nutritional status Body measurements Hematocrit Obesity Body weight Hemoglobin Overnutrition Bone density Iron deficiency Overweight Fertility Iron-deficiency Postnatal growth Folacin status Iron deficient Skinfold(s) Food intolerances Iron-deficient Stature Growth Iron status Undernutrition Growth rate Length Underweight Growth velocity Malnutrition Weight Health Morbidity Weight gain

Other relevant outcomes Behavioral development Food security School performance Cognitive development Functional status Social isolation Cognitive performance Hunger Quality of life Food insecurity School attendance

Health economics Healthcare (access, utilization, needs, costs) Medical (care, costs, needs) Medicaid Medicare Medicaid costs Medicare costs

6 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 1: Introduction

Citations flagged as irrelevant for the research review they were retained in the bibliography to ensure that included: the final report would provide general information about the type of research that has been done on the • General program descriptions. FANP in question.

• Program manuals and guidance materials. Though the computer searches were comprehensive, as tables 2-4 demonstrate, any such search is imperfect. • Descriptive research on program participation To guard against important omissions, initial lists of and/or costs. program-specific citations from the computer searches (minus the exclusions noted above) were cross- • Descriptive research on participant characteristics. checked against several existing research reviews • Research on issues related to program operations, (Nelson et al., 1981; Rush et al., 1988; Fraker, 1990; such as use of electronic benefits transfer (EBT) in Rossi, 1998; Besharov and Germanis, 2001), as well the Special Supplemental Nutrition Program for as against a listing of recent FNS research publica- Women, Infants, and Children (WIC). tions. A summary of preliminary citations was submitted to ERS and was reviewed by staff at ERS, FNS, • Research related to program accountability, fraud, and members of the project’s expert panel. Additional or abuse. citations provided by these reviewers were incorporated before documents were retrieved and reviewed. • Research related to determinants of outcomes of interest with no mention of impact or effect of pro- Documents were obtained from Abt’s in-house library, gram participation (for example, research on factors local university libraries, interlibrary loan, relevant that influence decisions about breastfeeding). Federal agencies, and, when necessary, from primary authors. All retrieved citations were reviewed by chap- In addition, research that involved FANP participants ter authors. Using the exclusion criteria described pre- but did not explicitly compare participants and nonpar- viously, as well as a review of research design and ticipants was excluded. For example, studies that methodology, authors identified research that provided examined the effectiveness of a specific smoking ces- empirical information on the effect of FANP participa- sation or breastfeeding promotion program among tion on nutrition- and/or health-related outcomes. WIC participants were excluded, as were studies that These documents formed the foundation of the examined specific interventions designed to decrease research review. Other relevant references were identi- the fat content of school lunches. Although useful for fied by authors as they reviewed papers and reports other purposes, this type of research sheds no light on and cross-checked bibliographies. the impact of FANP participation on nutrition- and health-related outcomes.6 Organization of This Report Not surprisingly, numerous relevant citations were locat- The next chapter provides an overview of the research ed for the flagship FANPs (FSP, WIC, and NSLP). designs and outcome measures used in the literature Many fewer citations were located for the smaller pro- reviewed.7 All readers are encouraged to read grams. Exclusion criteria were relaxed somewhat for chapter 2 before reading any of the program-specif- programs that generated few relevant citations. ic chapters that follow it. Although the citations considered under these relaxed standards were not expected to include information on The remainder of the report consists of 14 chapters program effects or to lead to other relevant research, that summarize available research for all of the FANPs identified in table 1, with the exception of the Senior 6Much of this research on FANP participants (without nonparticipant Farmers’ Market Nutrition Program, which was not controls) involved nutrition education interventions. Readers interested in established until 2002. The Team Nutrition Initiative general information on the effectiveness of such interventions are referred (TN) and the Nutrition Education and Training to a comprehensive series of literature reviews prepared by FNS. These reviews summarize research on the effectiveness of nutrition education for Program (NET) are covered in a single chapter. six population groups: pregnant women and caretakers of infants, preschool-age children, school-age children, adults, older adults, and intermediaries, paraprofessionals, and professionals. Complete citations for these 7A more comprehensive discussion of the strengths and weaknesses of reports are provided in the reference list at the end of this chapter (high- the various designs, as well as descriptions of other possible designs, can lighted with asterisks). be found in a separate report (Hamilton and Rossi, 2002).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 7 Chapter 1: Introduction

Each program-specific chapter includes the following: detailed information. Summary tables include all differences reported to be significant at the 5 percent Program Overview—A summary of the program’s leg- level or better. islative history and its benefits and eligibility requirements, with current information on program costs and Second, nonsignificant results are reported in the participation, and, as appropriate, on current policy interest of providing a comprehensive picture of the issues. body of research. A consistent pattern of nonsignificant findings may indicate a true underlying effect, Research Review—A description and synthesis of even though no single study’s results would be inter- research on the impact of the relevant FANP on nutri- preted that way. tion- and health-related outcomes. Where no such research was identified, there is a description of the Third, to give a complete picture, summary tables type of research that has been done and important or present findings for all studies reviewed, including interesting findings from the most recent or most rele- older studies and those with comparatively weak vant research. designs. However, when discussing conclusions that can be drawn from the available research, the authors Summary—A review of what is and is not known intentionally avoid the simplistic and flawed approach about the nutrition- and health-related impacts of the of “vote counting” (adding up the number of studies FANP, with areas for future research identified. that report differences favorable to participants). Rather, the authors give greater weight to findings For FANPs that have been widely studied, two types from studies that have the strongest research designs of tabular presentations are used to provide an and are most recent. overview of the breadth of existing studies and the relative consistency of their results: Finally, as in table 6, summaries of findings related to impacts on dietary intake show whether participants (1) Tables that summarize the important characteris- consumed more or less food energy or nutrients than tics of each study, including the year published nonparticipants, which is consistent with the general (or written, for nonpublished reports), data approach in the reviewed literature. Comparisons of sources, population studied, sample size, research participants and nonparticipants were most often based design, measure of program participation, and on mean intakes as a percentage of age- and gender- analysis method(s). Table 5 is an example. appropriate Recommended Dietary Allowances (RDAs), and study authors generally interpreted (2) Tables that summarize research results for a specific greater mean intakes among participants as evidence outcome or set of outcomes. These tables provide of a positive program impact. a visual overview of the patterns of research findings, using a format similar to that in table 6. This approach to assessing dietary intakes of groups was common practice at the time most of the studies As with any distillation of complex data, these tabular reviewed in this report were completed. Readers are summaries involved compromise. It is important that cautioned to avoid this “more is better” interpretation, readers understand four aspects of this compromise however. The reality is that a significant difference in before reading the program-specific chapters. the mean intakes of two groups does not necessarily First, summaries do not provide information on the mean that the two groups differ in the proportion of size of any effects detected or on the level of statistical individuals with inadequate diets. In recent years, significance reported. This information would greatly methods to assess dietary intakes have improved sub- increase the size and complexity of the summary table, stantially. For many nutrients, researchers can now making it harder for the reader to see the general pat- reliably estimate the prevalence of inadequate intakes tern of statistically significant effects. Interested read- in specific population subgroups, which is discussed in ers should refer to original papers and reports for more more detail in chapter 2.

8 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 1: Introduction n n od d h e t s l s and t ance u e gressio gressio n j of e d s a on m - t u te te re te re n s o a a a ison o g mod u n i j is of vari uare test s e d s Analysis met g a e o r s d g a n i e Multivari Compar Multivari means analys r Multivari e switchin with selecti b Chi-sq mmy mmy mmy u u u unt e on d on d on d Measure of participation us valu duals Participation dummy; Participati Participati Benefit amo Participati bon se n onse o t vs. t vs. t vs. pant pant pant p p n n n kes of indivi Design partici partici partici tration. Participa non Participa non Participa non Dose-res Dose-res s s ed l ) 16 z a i e t Demon derly n u d i n v y. bsid 2% i latio es 20 d es 12- n Cashou als i als i y lder y ng (8 ncome ncome ncome ncome el l idu idu Popu ,542) ,358) ,935) ,093) r g in su ses onl 1 1 3 1 (sample siz e alysis of national surveys d ONLY l Program on dietary inta rpo Adults ag S female) (n=200) housi and o Low-i livin Low-i e Youth ag (n= (n= Low-i indiv (n= Low-i indiv rural areas (n= al surveys an u n ve p rity Income/Elderl ustrati cu e v udies y Food Stamp E PURPOSE ll uti ll ll lls for ill ntified on ne enc ency aire ua the ed method f d entary Se pho onsec -pers m onn ry analysis of natio onq viduals. Data collection n our reca our reca our reca our reca , inclu Indi mparisons—State and local studies mparisons—Secondary y le and i 24-h via tele 24-h food frequ 24-h and n Food frequ questi 2 nonc 24-h o o e Supple s b tab t c c ILLUSTRATIV tual 1 trition Examination Survey. u n Servi icipant c d Intake ge Survey. ce Experiment. the ac Foo f mined the impact o nparticipan Nutritio o n ncome areas 6-97) exa 0-81 FNS 9-73 RIME 8-94 4-96 8-94 Data source t NHANES-III 198 NHANES-III Low-i in Connecticut (199 199 CSFII/DHKS 198 SSI/ECD and 196 198 nal Health and N Health Knowled Food and ng Survey of come Maintenan rtial versio a p dies tha nd se-response estimates—Seconda se-response estimates—State and local st s: a et al. a ce n 2) r d 000) et al. 00 d IIA: Do IB: Participant vs. no IA: Participant vs. nonpart IIB: Do n : this is a CSFII = Continui DHKS = Diet and FNS SSI/ECD = NHANES = Natio RIME = Rural In 3) 0) 6) Data sou 1 Note SAMPLE TABLE—INCLUDED FOR Table 5—Stu Study Group Dixon (2 Bhattachary Currie (2 Group Fey-Yensa (200 Group Gleaso (200 Group Butler an Raymon (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 9 Chapter 1: Introduction less d ct a Continued— nsume ant imp ts co n Signific Participa Elderly Butler (1996) [6 sites; D-R] individuals f ] less d l; D-R l; P-N] a nal; P-N] nsume [nationa [nation State; D-R] ) 0 ts co e} 0) [natio n 7a) l-ag (200 n n ez (198 st (1978) [1 Participa e {schoo t impact Children Gleaso W Elderly Lop Wome Fraker (199 n ifica ] ] p Program on dietary intakes o d l; P-N] No sign l; D-R l; D-R l; P-N] l; P-N] a 0) l; P-N] nal; P-N] nsume a nal; D-R] [nationa [nationa [nationa ts co [nation ) a (200 [6 sites; P-N] ation [nationa s ) ) 003) e Food Stam ONLY ll n 4 0 0 8) S (2 2) [natio a) [natio 98 7a) more/same n 987) of th (200 (200 995) [n 998 n n (199 usehold p Participa ez (198 imer (199 e [1 State; P-N] [2 sites; P-N] {preschool} Posner (1 Lop Gregorio (1 W Butler (1985) [6 sites; P-N] Adults Gleaso Rose (1 All ho Whitfield (1982) [1 city; D-R] Bisho Children Gleaso Cook (1 Elderly Fey-Yensa Perez-Escami PURPOSE e impact E RATIV more d xamined th ct LUST a IL nal; P-N] nsume ant imp ts co 0) [natio n udies that e t Signific Participa Children Fraker (199 of table. dings from s at end ergy energy and macronutrients See notes Table 6—Fin SAMPLE TABLE—INCLUDED FOR Outcome Food Food en

10 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 1: Introduction y less ses d s ct a nsume ext discu adding up all the ant imp (P-N = participant ts co ch n The t significant findings ma ting,” or Signific pproa ch a Participa r han others. Elderly Butler (1996) [6 sites; D-R] sea attern of non “vote coun re tice of individuals ideration t c istent p f ] s s con e), and the less d re con l; D-R l; P-N] l; P-N] 1, a a 0) r oid the pra v 1 Stat r nal; P-N] nsume hapte ty o merit mo [nationa d to a [nation a (200 [nationa State; D-R] ) 003) ci ll 0 8) ts co (2 n d in c 7a) 2) [natio n studies 00 (200 cautione n note s some ez (198 st (1978) [1 imer (199 Participa e e [2 sites; P-N] [1 State; P-N] t impact ers are Lop Adults Dixon (2 W Elderly Fey-Yensa W Perez-Escami Children Gleaso n ifica research. A ample, national vs. 1 ings from x y. Read ] p Program on dietary intakes o y of wa , find d s n y (for e l; P-N] No sign l; D-R l; P-N] a l; P-N] nsume nal; D-R] nal; P-N] a ted in that the stud nsideratio ly. [nationa [nationa ts co [nation ) [6 sites; P-N] s ) n e Food Stam ONLY cture of the bod ation co n 4 e of 0 r S a) [natio 0) [natio 98 7a) more/same 987) ses o scop of th sive pi (200 995 [n 998 n n rpo usehold u Participa ez (198 prehen ate, the ign and othe st studies. s lts would be interpre Elderly Lop Wome Fraker (199 Butler (1985) [6 sites; P-N] Adults Gleaso Posner (1 Children Rose (1 Gregorio (1 Cook (1 Rural Butler (1996) [2 sites; D-R] All ho Whitfield (1982) [1 city; D-R] tive p PURPOSE e com e impact E on d ra nge t a lus stro y’s resu earch d s RATIV d for il the publicati f providing more y). es in re d xamined th ct s from the ude c LUST a single stud stud nal; P-N] IL nal; P-N] name, incl s se finding nsume gh no differen s s ant imp table, spon 2) [natio ts co 0) [natio author’ r n d in the interest o en thou tual udies that e se re phasize v c t Signific (199 usehold senio p porte ct, e the a s. Because of R = do - Participa e re of r Children Fraker (199 All ho Bisho y, D sults a derlying effe t stud version re cular result dings from s un rticipan Cell entries show the with parti a partial nonpa Notes: Nonsignificant Table 6—Fin SAMPLE TABLE—INCLUDED FOR Outcome Protein vs. indicate a true studies methodological limitations and em This is

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 11 Chapter 1: Introduction

References *Lytle, L. 1994. Nutrition Education for School-Aged Children: A Review of Research. USDA, Food and *Balch, G.I. 1994. Nutrition Education for Adults: A Nutrition Service. Review of Research. USDA, Food and Nutrition Service. *Maloney, S., and S. White. 1994. Nutrition Education Besharov, D., and P. Germanis. 2001. Rethinking WIC. for Older Adults: A Review of Research. USDA, Food Washington, DC: The AEI Press. and Nutrition Service. *Bronner, Y.L., D.M. Paige, S.M. Gross, et al. 1994. Nelson, K., J. Vermeersch, L. Jordan, et al. 1981. The Nutrition Education for Pregnant Women and National Evaluation of School Nutrition Programs, Caretakers of Infants: A Review of Research. USDA, Review of Research: Vol. 2. Santa Monica, CA: Food and Nutrition Service. System Development Corporation. Cole, N., and M.K. Fox. 2004a. Nutrition and Health Oliveira, V. 2003. The Food Assistance Landscape: Characteristics of Low-Income Populations: Volume March 2003. FANRR-28-2. USDA, Economic II, WIC Participants and Nonparticipants. Research Service. E-FAN-04-014-2. USDA, Economic Research Service. *Olson, C.M. 1994. A Review of Research on the Cole, N., and M.K. Fox. 2004b. Nutrition and Health Effects of Training in Nutrition Education on Characteristics of Low-Income Populations: Volume Intermediaries, Paraprofessionals, and Professionals. IV, Older Adults. E-FAN-04-014-4. USDA, Economic USDA, Food and Nutrition Service. Research Service. Rossi, P.H. 1998. Feeding the Poor: Assessing Federal Cooper, H., and L.V. Hedges (eds.). 1994. The Handbook Food Aid. Washington, DC: The AEI Press. of Research Synthesis, Part III: Searching the Literature. New York, NY: Russell Sage Foundation. Rush, D., J. Leighton, and N.L. Sloan. 1988. “Review of Past Studies of WIC,” American Journal of Clinical Fox, M.K., and N. Cole. 2004a. Nutrition and Health Nutrition 48:394-411. Characteristics of Low-Income Populations: Volume I, Food Stamp Program Participants and Nonparticipants. Speshock, E. 1999. Personal communication. E-FAN-04-014-1. USDA, Economic Research Service. *Swadener, S.S. 1994. Nutrition Education for Fox, M.K., and N. Cole. 2004b. Nutrition and Health Preschool-Age Children: A Review of Research. Characteristics of Low-Income Populations: Volume USDA, Food and Nutrition Service. III, School-Age Children. E-FAN-04-014-3. USDA, Economic Research Service. U.S. Department of Agriculture, Food and Nutrition Service. 2000. “Food and Nutrition Service (FNS) Fox, M.K., W.L. Hamilton, and B.-H. Lin. 2004. Strategic Plan 2000 to 2005.” Available: http://www. Effects of Food Assistance and Nutrition Programs on fns.usda.gov/oane/MENU/gpra/FNSStrategicplan.htm. Nutrition and Health: Volume 4, Executive Summary Accessed April 2002. of the Literature Review. FANRR-19-4. USDA, Economic Research Service. U.S. Department of Agriculture, Food and Nutrition Service. 2003. “Food Stamp Program Nutrition Fraker, T.M. 1990. Effects of Food Stamps on Food Education Fact Sheet.” Available: http://www.fns.usda. Consumption: A Review of the Literature. USDA, gov/fsp/menu/admin/nutritioned/fsheet.htm. Accessed Food and Nutrition Service. April 2003. Hamilton, W.L., and P.H. Rossi. 2002. Effects of Food White, H.D. 1992. “Reference Books, Databases, and Assistance and Nutrition Programs on Nutrition and the Repertoire,” in H.D. White, M.J. Bates, and Health: Volume 1, Research Design. FANRR-19-1. P. Wilson (eds.), For Information Specialists: USDA, Economic Research Service. Interpretation of Reference and Bibliographic Work. Norwood, NJ: Ablex Publishing Corporation. Logan, C., M.K. Fox, and B.-H. Lin. 2002. Effects of Food Assistance and Nutrition Programs on Nutrition *Asterisked citations are literature reviews prepared by and Health: Volume 2, Data Sources. FANRR-19-2. FNS (see footnote 6 in the text). These reports summa- USDA, Economic Research Service. rize information on the effectiveness of nutrition education interventions for specific population groups.

12 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2 Research Methods

This chapter provides an overview of the research the time and place in which people live, which means methods used in the studies summarized in this report. that similar people in a different time or place may not There are two main sections: appropriately represent the counterfactual. All of these influences may contribute to selection bias, which dis- (1) Evaluation design—Much of this discussion is torts the evaluation of a program’s impact. adapted from Volume 1 of this series, Research Design (Hamilton and Rossi, 2002). The sections that follow describe key research designs encountered in FANP research and their various (2) Outcome measures. strengths and limitations. In the program-specific chapters that constitute the remainder of this report, Readers with limited knowledge of research design or the research design used in each study is clearly identi- measurement issues in nutrition and health-related fied. The text generally includes some discussion of research are encouraged to read this chapter before any design limitations; however, the present chapter serves of the program-specific chapters that follow and to use as the primary source of information on research it as a technical resource, as needed. methodology.

Evaluation Design The Randomized Experiment There is a strong consensus in the scientific communi- The studies reviewed in this report attempted to meas- ty that only randomized experiments are fully capable ure the impact of specific food and nutrition assistance of providing reliable estimates of a program’s impacts. programs (FANPs) on nutrition- and health-related The randomized experiment is the “gold standard” of outcomes. The impact of a program or other interven- program evaluation. tion is defined as the difference between what happens in the presence of the intervention and what would In the simplest randomized design, potential partici- have happened in its absence, generally called the pants are randomly assigned to either an experimental “counterfactual.” (or treatment) group, which will be subject to the program being assessed, or to a control group, from which Establishing the counterfactual—that is, estimating what the program will be withheld. The program’s impact is would have happened without a given program—is then estimated by comparing the average outcomes in usually accomplished by examining a population that the experimental group, after sufficient exposure to the has not been subjected to the program. What makes program, with control group outcomes measured at the the task difficult is the fact that people who become same time. participants in a social program are often quite different from those who do not because they either have Because the experimental and control groups differ at been selected for participation or have selected them- 8 the outset only by chance, they are considered to be selves (Campbell and Stanley, 1963). These selective fully comparable at that point. In other words, the two processes may make participants different in important groups are considered to be equivalent, in the statisti- ways from those who do not participate. These differ- cal aggregate, on all permanent and transitory charac- ences include not only people’s permanent characteris- teristics. Subsequently, the only systematic difference tics, such as their gender or race, but also transitory between the groups is exposure to the program. ones like their current income or employment, the Accordingly, it is credible to infer that any post-pro- opportunities they face, and the experiences they have gram differences between the two groups are caused had. Many of the transitory characteristics result from by the program, provided that the differences are greater than what might occur by chance. 8Evaluation designs often focus on units other than people, either aggregations of people (households, students in a school, the population of a The fundamental requirement of randomized experi- county) or operating entities (program offices, schools, businesses). For simplicity of presentation, the present discussion generally refers to indi- mentation is that program services be deliberately viduals rather than aggregations or other entities. withheld from some people who are otherwise like the

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 13 Chapter 2: Research Methods people receiving the service. Such a practice is generally program services but are not offered the new services prohibited in entitlement programs because law and specified in the intervention. In the case of the food regulation require that program benefits be provided to stamp cash-out demonstrations, for example, the eval- everyone who meets eligibility requirements and takes uations estimated the effects of receiving benefits in the necessary steps to qualify. Many FANPs are enti- the form of checks rather than as food stamps, but not tlement programs. the overall impact of the Food Stamp Program (FSP) itself. Saturation programs—those with sufficient funding and infrastructure to serve essentially all eligible people— Quasi-Experiments pose similar problems. Whether a potentially eligible Virtually all the research that has examined the impact person can receive benefits from a nonentitlement pro- of FANPs on nutrition and health outcomes has identi- gram depends on the local availability of program fied counterfactual conditions without random selec- funding and infrastructure. For many nonentitlement tion into treatment and control groups. Such impact programs that approach full saturation, like the Special evaluation designs are known as quasi-experiments. Supplemental Nutrition Program for Women, Infants, That is, they resemble experiments in providing a spe- and Children (WIC), it can be virtually impossible to cific representation of the counterfactual, but the coun- find a reasonably representative set of potential partici- terfactual is identified through some means other than pants to whom the program could be considered random selection. Most of the FANP research unavailable. If program services would normally be reviewed in this report used one of four quasi-experi- provided to everyone who applies and is eligible, it mental designs. may be considered unethical to withhold services for research purposes from people who might apply. Quasi-Experiment 1: Comparing Participants With Nonparticipants Given these challenges, it is not surprising that the literature reviewed for this report included only one This design, referred to as “participant vs. nonpartici- study that used a randomized experiment to evaluate pant” in the program-specific chapters, is the one most the impacts of a specific FANP. This study was com- commonly used in the research summarized in this pleted by Metcoff and his colleagues (1985) during the report. It calls for identifying comparable groups of early years of the WIC program. Random assignment participants and nonparticipants and interpreting the was feasible because, at the time, the demand for WIC average difference in outcomes between the groups as participation at the site in which the study was con- the effect of the program. Nonparticipants must be ducted exceeded the available funding. potentially eligible—that is, people who apparently could have applied and qualified for the program, but A few studies have used randomized experiments to did not—to be a credible representation of the counter- estimate the impact of demonstrations or pilot pro- factual. In most, but not all, FANP studies, researchers grams, rather than of the FANPs per se. These demon- apply an approximation of the means test to identify strations typically represented policy initiatives that nonparticipants with incomes below the eligibility cut- were tested on a limited scale before full-scale imple- off for the program in question. mentation. The most prominent examples are demonstrations of cashing out food stamps (the so-called Selection Bias in Participant/ “cash-out” studies (Fraker et al., 1992; Ohls et al., Nonparticipant Comparisons 1992) and studies of pilot projects in which school The major problem with this quasi-experimental design breakfasts were offered free to all school children (uni- is that identified nonparticipants may not be sufficiently versal free breakfast projects) (for example, Peterson comparable to participants. This problem, known as et al., 2003; McLaughlin et al., 2002; Murphy and selection bias, is a difficult issue in all quasi-experimen- Pagans, 2001). tal designs and is especially troublesome when people who have taken the actions necessary to participate in Keep in mind when interpreting results of evaluations a program are compared with people who have not. of demonstration projects that, in these evaluations, the counterfactual is not the absence of the program. Selection bias often occurs because participants are Rather, it is the status quo, or the program as it exists more highly motivated to achieve the program-relevant without the innovation or modification introduced by outcomes than nonparticipants. Suppose, for example, the demonstration. Control subjects experience usual that the women who seek WIC benefits for themselves

14 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods or their children tend to be very concerned about the did not qualify for the program, and at some point, effect of diet on their children’s health. Such women they will regain that level. These two types of people may well take other actions with the same objective, might have similar incomes at the time they enter the such as following nutrition advice included in program, but their subsequent outcomes, in the brochures they pick up in the doctor’s office—or get- absence of the program, might not be at all similar. ting to a doctor’s office at all. If this supposition were true, one would expect the children of mothers who Approaches To Dealing With Selection Bias seek WIC benefits to have better nutrition and health Researchers have used a variety of approaches to attempt outcomes—even in the absence of the program—than to counteract selection bias, the most common of which children of mothers who are less motivated and do not are described below.9 All have the basic objective of seek WIC benefits. A simple comparison of WIC and making the participant and nonparticipant groups non-WIC children would therefore reveal that the WIC “alike” on certain specified dimensions. However, all children had more positive outcomes even if the pro- leave open the possibility that bias remains. gram had no effect at all. Regression Adjustment. A prime example of this Sometimes selection bias operates in the opposite approach is the WIC-Medicaid study conducted by direction. Mothers of children with nutrition-related Devaney et al. (1990 and 1991) to assess the impact of problems might be especially motivated to seek WIC prenatal WIC participation on birth outcomes. Taking benefits, for example, whereas mothers of healthy chil- advantage of the fact that all Medicaid recipients were dren might be less inclined to participate. WIC might automatically eligible for WIC benefits, Devaney and improve the participating children’s condition, but the her colleagues contrasted birth outcomes of Medicaid children might not catch up to their nonparticipating, recipients who had participated in WIC during preg- healthier counterparts. In this example, the simple nancy with those who had not. The relevant dataset comparison would find WIC children to have less pos- was assembled by linking Medicaid records to WIC itive outcomes even though the program had a positive participation records and birth registration records. effect. Birth registration records provided information on the critical outcome of birthweight, WIC records identified Motivation of participants toward the program out- WIC participants, and Medicaid records identified come is one of the most common sources of potential those who gave birth during the period of study. The bias, and one of the most difficult to counteract. Other resulting linked WIC-Medicaid database included common sources of self-selection bias include need approximately 112,000 births to Medicaid mothers in (often proxied by income), potential for gain (often five States over a 2-year period. proxied by the dollar value of the benefit), and the individual’s desire not to depend on public assistance. To minimize selection bias, Devaney and her associates used regression adjustments. The equations included Selection bias may also result from program rules or variables that were likely to capture ways in which procedures. In nonentitlement programs, local staff participants and nonparticipants might differ, including often decide which applicants will be approved for educational attainment, prenatal medical care, gestational participation based on a combination of program poli- age, race, mother’s age, and birth parity. As typically cies and individual judgment. In all programs, out- happens, the researchers were limited to the variables reach practices, referral networks, office locations and available in existing datasets, which seldom measure hours, and community customs may make some peo- all of the factors that might create different outcomes ple more likely to participate than others. for participants and nonparticipants. Alternative attempts to counter selection biases led to quite drastic Finally, some selection bias occurs when program par- changes in estimates of the effects, without any clear ticipation is based on transitory characteristics. For indications of which attempt was more sensible. example, some people who qualify for means-tested programs are permanently poor, or nearly so, and would be income-eligible for program participation for 9Another technique for dealing with selection bias is the use of propen- many years. Other people who qualify for the same sity scores. Propensity scoring allows a more comprehensive and complex programs are not permanently poor, but are at a tem- treatment of covariates than is possible with regression adjustment (Hamilton and Rossi, 2002). However, though propensity score methods porary low point in a fluctuating income pattern. In an have been used extensively in public health research, they were not used in earlier period, their income was high enough that they the literature reviewed for this report.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 15 Chapter 2: Research Methods

Matched Pairs. Sometimes researchers construct a The dose-response model requires that benefits vary comparison group by matching participants and non- across households that are similar in terms of the factors participants on characteristics thought to be related to expected to affect their health and nutrition outcomes. selection tendencies. For each participant in the The food stamp situation appears to meet that condi- research sample, the researcher identifies a nonpartici- tion. Households of a given size with a given amount pant with identical or closely similar characteristics on of cash income receive differing benefit amounts key variables. Because the matching procedure can depending on, for example, how much of the income normally consider only a few variables, regression is earned and their allowable deductions. However, adjustment is still needed to estimate impacts. because the underlying logic driving benefit rules is that the benefit amount should be responsive to need, The matched-pair approach is advantageous mainly it would be desirable to see more extensive analysis of when there is a substantial marginal cost for including the extent to which food stamp benefit variation actu- subjects in the evaluation, typically when significant ally meets the requirements of dose-response analysis. new data collection is to be carried out. If the analysis is based on existing administrative or survey datasets, Two-Stage Models. Some researchers use a two-stage the matched-pairs approach excludes otherwise usable approach in which they first model the likelihood that observations and thus reduces the sample size avail- an individual will be a participant in the program. The able for analysis. model yields a predicted probability of participation for each participant and nonparticipant. The second More general matching procedures may identify more stage of analysis models the outcome as a function of than one nonparticipant (perhaps even many) similar some measure of participation. enough to each participant. When combined with regression adjustment, matched sampling is one of the One class of solutions simply uses the predicted proba- most effective methods for reducing bias from imbal- bility of participation in place of actual observed par- ances in observed covariates (Rubin, 1979). ticipation as an explanatory variable in the second-stage model. Another includes observed participation along Dose-Response. If program rules prescribe different with an inverse Mills ratio, which is a function of the amounts of the program benefit or service for different predicted probability of participation (Heckman, 1979). participants, a dose-response analytic model may be applicable. The underlying hypothesis is that greater In order for two-stage approaches to offer a material benefits will lead to greater effects on outcomes. The gain over simple regression adjustment, the participa- dose-response relationship may be estimated with a tion model must include one or more “instruments”— sample that consists only of participants, which elimi- variables that predict participation but are not correlat- nates the issue of whether participants differ from non- ed with the outcomes of interest. Finding an appropri- participants in unmeasurable ways. If this relationship ate instrument is often impossible, however, especially can be estimated, then the program’s impact may be when the researcher is working with existing datasets. described as the difference between the effect at any Participation is typically related to demographic char- given level of benefits (typically the average benefit) acteristics, need or potential benefit, motivation, and and the projected effect at the zero-benefit level (what pre-program measures of relevant outcomes, such as participants would receive if they did not participate). nutrition or health status. These same factors usually influence post-program outcomes. And many factors The FSP, with benefits measured in dollars and a large that initially seem like good instruments turn out, on number of actual benefit amounts, is the main candidate closer examination, to be related to outcomes. For for dose-response analysis among the FANPs. A number example, living close to a program office might be of researchers have used this approach, although with expected to make an individual more likely to partici- considerable variation in the way it was applied. Some pate and initially seems unrelated to health and nutrition researchers have estimated models that exclude non- outcomes, but the program’s location may have been participants (for example, Neenan and Davis, 1978; selected to give easy access to a high-risk community. Levedahl, 1991; Kramer-LeBlanc et al., 1997), while others include nonparticipants and specify the model In addition to the instrumental variable, some two-stage to include both a variable representing the benefit approaches use functional form to achieve identifica- amount and a variable representing participation per se tion in the models. In a procedure known as the two- (for example, Fraker, 1990; Devaney and Fraker, 1989). step Heckman method, the participation model uses a

16 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods nonlinear functional form (Heckman, 1979; Heckman (Heckman and Hotz, 1989). Nonetheless, after decades and Hotz, 1989). Alternatively, the participation and of research and debate, the statistical community has outcome equations can be estimated simultaneously not yet reached a consensus that any particular using a maximum likelihood approach. In both cases, approach will consistently remove selection bias. the effectiveness of the method depends on the validity of assumptions made about the error terms in the In addition, data limitations hamper nearly all attempts model, assumptions that cannot be verified empirically. to counter selection bias. Careful theorizing about the determinants of participation usually suggests many All of these two-stage approaches have been used in factors that are not measured in existing datasets. Even evaluating FANPs, but with no clear consensus that any with special data collection, many of the factors per- of them can be considered generally reliable. For tain to the time period before the individual began par- example, Gordon and Nelson (1995) used three ticipating (or not participating) and cannot be meas- approaches (instrumental variables, Heckman two-step, ured reliably on a retrospective basis. and simultaneous equations) and a rich dataset to estimate WIC effects on birthweight. They found that the Although the extent of remaining bias cannot be approaches to selection bias correction yielded “unsta- known for sure, testing the robustness of the results is ble and implausible results, [possibly] because the fac- usually informative. A program impact estimate that tors affecting WIC participation and birthweight are remains stable under various alternative specifications very nearly identical, since WIC targets low-income is somewhat more credible than one that varies dra- women at risk for poor pregnancy outcomes.” Ponza et matically. Of course, if several specifications fail al. (1996) similarly used multiple approaches to selec- equally to remove the bias, their results will be consis- tion-bias adjustment in evaluating the Elderly tent with one another but inaccurate. Nutrition Program (ENP). The authors rejected all of the two-stage approaches and based their conclusions Quasi-Experiment 2: Comparing Participants Before and After Program Participation on the results of the simple, one-stage regression adjustment. This simple design (referred to as “participants, before vs. after” in the program-specific chapters) eliminates Caveats to Selection-Bias Adjustment some dimensions of selection bias but has other major The most troubling aspect of statistical approaches to vulnerabilities. Subjects are selected into the study adjusting for selection bias is that one cannot be cer- before they have been meaningfully exposed to the tain whether the procedure has, in fact, eliminated program—for example, when they apply for program selection bias. Well-conceived applications of selection- services. They are clearly aware of the program at this bias adjustment models have yielded some plausible point and have already taken some action to respond to and some implausible results in evaluating FANPs. its requirements, but they have not normally been The situations that produce implausible results cannot “exposed” to any of the program’s benefits in ways be identified a priori, and none of the approaches has that would affect their status on the outcome dimen- 10 consistently yielded plausible results. Moreover, a sions of interest. The subjects’ status on the outcome plausible selection-bias adjustment has not necessarily dimensions is measured upon their selection into the accomplished its purpose just because it is plausible. study and again after program exposure (long enough after exposure that effects are expected to be visible). When researchers have compared the effects estimated in randomized experimental evaluations with those The subjects’ preparticipation status serves as the derived from comparing participants with nonpartici- counterfactual. The design assumes that, without the pants, the two sets of findings have often been divergent. program, the individual’s preprogram status would not For example, when La Londe and Maynard (1987) com- change. If this assumption is valid, the before vs. after pared the findings from a randomized experiment with difference represents the effect of the program. those obtained by using comparable nonparticipants as the counterfactual, they found that none of several meth- A prime example of the “participants before vs. after” ods for identifying comparable nonparticipants produced design in FANP research is the work done by Yip et al. results consistent with the findings from the random- 10 ized experiment. However, subsequent work argued that This may not be true if the program requires some action before enrollment that may itself affect the person's status on outcome variables of specification tests could have led to a result approach- interest. Examples would be preenrollment requirements, such as looking ing the estimate from the randomized experiment for a job or visiting a doctor.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 17 Chapter 2: Research Methods

(1987) on anemia among preschool children. Yip and Although this design is usually applied prospectively, his colleagues studied infants and preschool children it can be applied retrospectively if panel datasets pro- participating in WIC and contrasted hematocrit levels vide appropriate information. The researcher must be at the time of admission into the program with levels able to identify people who participated in the pro- at a followup visit a few months later. The data gram, determine when they began participating, and showed a marked decrease in anemia over the few have comparable measures of the key outcome dimen- intervening months. Because the time frame was so sions for both the pre- and post-program periods. short, it is unlikely that the effects were attributable to natural developmental processes or to long-term secu- Quasi-Experiment 3: Comparing lar declines in anemia among American children. Participants to Nonparticipants Before and After Program Participation When program effects are not expected to occur quickly, This design (“participant vs. nonparticipant, before and the assumptions of the before vs. after design become after”) combines the strengths of the two previous more tenuous because forces other than program par- quasi-experiments. It has less vulnerability to selection ticipation might cause changes in participants’ status. bias than a simple comparison of participants to non- For example, normal patterns of child development participants and less vulnerability to bias from tempo- involve substantial changes in many variables over rel- ral effects than a before vs. after comparison. atively short periods. A related issue is that some conditions improve naturally over time without interven- In this design, outcomes for participants and nonpartici- tion, a phenomenon known in medical treatment as pants are measured once before participation begins and spontaneous remission and in some statistical circum- again after the effects of participation are expected to be stances as regression toward the mean.11 Many people visible. Conceptually, the program’s impact is estimat- become eligible for means-tested programs because ed as the post-program difference in outcomes, sub- they have experienced a temporary drop in income. tracting out the difference that already existed before Over time, many such people have an improved income, participation. This design is therefore commonly called even if they do not enroll in a program. Accordingly, it a difference in differences or double difference design. would be a mistake to assume that the program causes such post-participation gains in income—or in any In practice, this design is usually applied with multi- conditions affected by income, such as many dimen- variate modeling. The dependent variable in the model sions of nutrition and health status. is often the post-program outcome, with the pre-program outcome measure as a predictor variable, along General societal trends may also improve conditions of with participation status. As in the regression adjust- a target population. These include not only long-term ment model discussed previously, the model adjusts trends, like the general reduction in nutrient deficien- for the differing composition of the participant and cies in the United States, but such short-term phenom- nonparticipant populations by incorporating covariates ena as swings in the unemployment rate or changes in that are expected to be related to the outcome measure Medicaid coverage. Any before vs. after period that or to the likelihood of participation. lasts more than a few months is potentially vulnerable to such temporal effects, and seasonal effects can A noteworthy example of this design is a study con- sometimes occur within a few months. ducted by Kennedy and Gershoff (1982). The authors compared changes in hemoglobin and hematocrit lev- Given this vulnerability, the participants before vs. after els of pregnant WIC participants and nonparticipants design is useful mainly for evaluating impacts that are between the first and final prenatal visits. expected to be fully visible within a brief period. If temporal effects might also occur, the design can nei- Although this variation is the strongest of the quasi- ther refute the possibility nor control for it statistically. experimental designs, it is rarely used to evaluate ongoing entitlement or saturation programs. Because the design calls for pre- and post-participation measures 11A related issue is measurement error. If a measure is not fully reliable, that is, is not capable of producing the same result in repeated application, on both participants and nonparticipants, data collection a before vs. after design may indicate negative results for an individual can be complicated and very costly. Moreover, existing simply because of measurement error. Special measurement efforts may national surveys or administrative datasets that collect therefore have to be made with this design. For example, infant development studies often require two independent measures of infant length at substantial amounts of nutrition and health outcome data each time point because infant length is difficult to measure accurately. are cross-sectional rather than longitudinal in design.

18 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods

Quasi-Experiment 4: Time Series Analysis Estimating impacts for the target population has both advantages and disadvantages. An impact estimate for Time series analyses are an important extension of the target population combines the program’s effec- before-and-after studies that can be employed when tiveness in reaching people (its penetration or partici- many observations of outcomes exist for periods pation rate) with its effectiveness in helping those it before and after program implementation. Unlike sim- does reach (the impact on participants). Because ple before-and-after designs, time series analyses take FANPs are designed to ameliorate problems in speci- trends into account. Observations that occur before the fied target populations, this kind of analysis addresses program is in place are used to model outcome trends the question of how well the program is achieving its in the absence of the program. The predicted trend rep- ultimate objective. However, it risks the possibility resents the counterfactual, and is contrasted with the that a positive impact on program participants may be trend actually observed after the program is in place. so diluted by nonparticipants that it is invisible in the The difference between the two trends is attributed to analysis. If the data represent the entire population of the program. an area, including those outside the program’s target The version of time series analysis that has been population, the dilution problem is exacerbated. used in FANP research is the cross-sectional time series. This approach uses time series on multiple Outcome Measures units, such as series for individual States or counties. A good example is the study undertaken by Rush Existing research has examined the impact of FANP and colleagues (1988) to assess effects of the WIC participation on a number of different outcomes. The program. Taking advantage of the rapid growth of outcomes are logically sequential, as summarized the WIC program in the 1970s, Rush and his col- below, using the FSP as an example. leagues conducted a time series analysis of the effect of the program’s growth on birth outcomes. They • Household food expenditures is the first outcome in related the growth of the WIC program between 1972 the sequence. The FSP, which provides earmarked and 1980 in a large number of counties to county- economic benefits, can be expected to have a direct aggregate data on birth outcomes. The research impact on the amount of money a household spends strategy was based on the expectation that if WIC is on food. effective in improving birth outcomes, improvements Household nutrient availability is the second out- ought to be proportional over time to its expansion. • come. If a household increases the amount of money Using birth registration records and State WIC it spends for food, it is expected to increase the records, Rush found that the growth of WIC over this availability to household members of food energy period led to increased average birthweight, longer and at least some nutrients. average duration of gestation, and decreased fetal mortality. These effects were over and above the secular • Individual dietary intake is the next outcome in the trends for this time period and were especially pro- sequence. For the FSP, the hypothesis is that nounced for births to less-well-educated and minority increased availability of nutrients in the household women. The analysis covered 19 States and almost leads to increased nutrient intake by individual 1,400 counties. household members. Programs like WIC and the school nutrition programs, which provide specific Unlike all of the preceding research designs, time foods or meals to participants, are hypothesized to series analyses do not focus on outcomes for individ- have a direct impact on individual dietary intake. ual program participants. Rather, they focus on a more broadly defined population that can be examined both • Measures of nutrition and health status other than before and after the program is introduced. Because dietary intake, which FANP participation may influ- the unit of aggregation in most data series is some ence through the above pathways. Such measures geographic unit, the analysis estimates the program’s include, for example, birth outcomes, nutritional bio- impact on the overall population of that area. Where a chemistries, linear growth in children, and body data series is available for a relevant subpopulation, weight. Relatively recent research on the School such as low-income households or pregnant women, Breakfast Program has expanded this set of out- the analysis can speak to the impact on that more spe- comes to include measures of school and academic cific target population. performance.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 19 Chapter 2: Research Methods

With the exception of the WIC program and the ENP, have focused on the FSP. However, a handful of stud- relatively few FANP studies have examined the last ies have assessed impacts on food expenditures rela- group of outcomes. Moreover, conclusions from stud- tive to participation in the WIC program, the National ies that have examined these outcomes must be inter- School Lunch Program (NSLP), and the Nutrition preted with caution. Establishing causality between Assistance Program (NAP) in Puerto Rico. FANP participation and long-term nutrition and health outcomes requires that data support a logical time Although studies of the impact of FANP participation sequence. For long-term outcomes (measures that on food expenditures are conceptually similar, they develop over time, such as linear growth and body vary substantially in how food expenditures were weight), FANP participation must precede the outcome measured. Some studies were based on money spent for a reasonable period and be of sufficient intensity to on food for at-home use over the course of a week (or provide a plausible basis for a hypothesized impact. In weekly food purchases), while others used the mone- addition, reliable assessment of impacts on measures tary value of food eaten out of household supplies over such as linear growth and nutritional biochemistries a week or a month. The former measure includes requires at least two measurements, one before partici- expenditures for foods not necessarily eaten during the pation and one after. Finally, nutrition and health status week of purchase and excludes the value of foods used are influenced by a complex interplay of diet, heredity, from household inventories during the recall period. and environment, making the task of determining the specific impacts of FANPs on these long-term out- Another important difference relates to whether the comes a challenge. measure considered expenditures only for food eaten at home or total food expenditures, including meals A few studies have examined the impact of FANP par- and snacks eaten away from home. Finally, some ticipation on health-related behaviors, including, measures included the value of purchased food only, specifically, the impact of the WIC program on breast- while others also included nonpurchased food (for feeding and child immunizations and the impact of the example, home-grown foods and food received as gifts). ENP on socialization among the elderly. Some researchers analyzed expenditures for the house- A potential limitation for all outcome measures used in hold as a whole, while others normalized expenditures FANP research is the problem of measurement error. to account for the household’s size, its age/sex compo- Estimation of key outcomes—including household sition, meals eaten away from home, meals served to food expenditures, household nutrient availability, and guests, and/or economies of scale. Commonly used individual dietary intake—involves collecting detailed approaches standardize food expenditures based on data over a day, multiple days, a week, or a month. “equivalent adults” (EAs), counting additional family The data are subject to errors associated with respon- members less heavily because of economies of scale, dents’ abilities, cooperation, and recall. These errors “adult male equivalents” (AMEs), counting family are assumed to affect participants and nonparticipants members according to caloric requirements, and in FANP studies equally; however, the overall effect is “equivalent nutrition units” (ENUs), counting family a reduction in measurement reliability. In turn, reduced members according to caloric requirements and per- reliability increases the likelihood that differences centage of meals eaten at home. In general, the more between participants and nonparticipants will be factors considered in normalizing expenditure data, the obscured (Rossi, 1998). better. That is, ENUs provide a more precise assessment of expenditures per household member than the The next sections of this chapter describe key outcome more basic EA measure. measures used in existing FANP research. Later program-specific chapters also include some discussion of In examining the impact of FANP participation on the strengths and limitations of various outcome meas- food expenditures, researchers have used both primary ures; however, the present chapter serves as the pri- data collection and secondary analysis of data collect- mary source of such information. ed in national surveys, such as the Consumer Expenditure Survey (CES), the Panel Study of Income Household Food Expenditures Dynamics (PSID), the Nationwide Food Consumption Survey (NFCS), and the Continuing Survey of Food Most of the studies that have examined the impact of Intakes by Individuals (CSFII). The latter two surveys FANP participation on household food expenditures are no longer conducted.

20 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods

Household Nutrient Availability All studies of impacts on household nutrient availability have focused on the FSP. Research has included Assessment of household nutrient availability is based both primary data collection and secondary analysis of on detailed records of household food use for an national survey data. Most of the secondary analyses extended period, usually 1 week. Information on quan- used data from the 1977-78 NFCS (low-income sup- tities of food withdrawn from the household food supplemental sample) or data from a followup NFCS low- ply is translated into nutrient equivalents to represent income sample that was collected in 1979-80. the food energy and nutrients available to household members. Although household nutrient availability Individual Dietary Intake excludes the nutrient content of food eaten away from home, it is still an important measure because A number of techniques can be used to assess individual the FSP is specifically intended to improve in-home dietary intake (Thompson and Byers, 1994). In research food consumption. on FANP impacts, the technique used most often is a single 24-hour recall or a single-day food record. Nonetheless, nutrient availability at the household Some studies collected multiple days of data, ranging level is not equivalent to nutrient intake at the individ- from 2 to 7 days, using recalls, records, or a combina- ual level. The relationship between the two measures tion approach. Respondents usually reported on their is weakened by several considerations. own intakes, but parents or other caregivers served as proxy respondents for infants and young children. • Some household members will get nutrients from foods eaten away from home. Although all dietary data collection techniques have limitations, it is generally accepted that the more days • Some of the food used from household supplies is of data available, the better the measure.12 In addition, wasted. food records are generally believed to be more accurate than recall-only methods because respondents, at least • Household members may unequally consume nutri- in theory, record food intake on a prospective basis ents from household food supplies, relative to their rather than recalling it retrospectively and have the needs, depending on their tastes and appetites. opportunity to measure or carefully observe portions. Food records impose a significant response burden, Moreover, increased availability of food energy and however, and are particularly problematic for respon- nutrients at the household level does not necessarily dents with limited literacy. Moreover, the need to record translate into better diets—for example, lower intakes food intake may alter respondents’ eating behavior. For of nutrients and food components that tend to be over- these reasons, recall-based data collection is preferred consumed by many Americans (fat, saturated fat, cho- for assessment of low-income populations. lesterol, and sodium) or greater adherence to recommended patterns of food intake (for example, eating The 24-hour recall has three key disadvantages. First fruits and vegetables or whole grains). For these rea- and most obvious, the method relies on memory, sons, one must examine the dietary intakes of individ- which tends to be imperfect. Second, 24-hour recalls ual household members to adequately assess nutrition- have been shown to be subject to systematic underre- related impacts of the FSP. porting by some subgroups, including individuals who are overweight (Briefel et al., 1997) and the elderly In assessing household nutrient availability, the (Madden et al., 1976). Third, because intakes vary so amount of energy and nutrients available in the foods much from day to day in highly industrialized coun- withdrawn from the household food supply is evaluat- tries, such as the United States, a single day’s intake is ed relative to the Recommended Dietary Allowances unlikely to be representative of the respondent’s usual (RDAs) and the household’s size and composition. diet (Beaton, 1983). Household nutrient requirements are generally defined based on adult male equivalents (AMEs), which take The accuracy of 24-hour recall data can be improved into consideration the number of individuals in the by careful, standardized interviewing techniques. household and their differing nutrient requirements based on age, gender, and pregnancy/lactation status, 12 or equivalent nutritional units (ENUs), which further There are limitations, however. Experience has shown that quality and completeness of data decrease as the number of days increases. adjust for the number of meals each family member Respondents tend to fill out records less carefully as time goes on, after eats at home and the number of meals served to guests. approximately 4 or 5 days (Gersovitz et al., 1978).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 21 Chapter 2: Research Methods

Computer-assisted interviewing is one way to achieve established intake standards rather than just comparing a high level of standardization. One of the first appli- raw intakes in kilocalories, milligrams (mg) or grams cations of computer-assisted 24-hour recalls was (gm). At the time most of these studies were conduct- developed for the third National Health and Nutrition ed, the standards used were the Recommended Dietary Examination Survey (NHANES-III) (McDowell et al., Allowances (RDAs) (National Research Council 1989). The approach was refined and improved, based (NRC), 1989a). More recent studies have also used the on methodological research, to better engage respon- Dietary Guidelines for Americans (U.S. Departments dents in the interview process and to provide memory of Agriculture (USDA) and Health and Human cues for accurate recall of food and beverage con- Services (HHS), 2000). A few studies used the Healthy sumption (Moshfegh et al., 2001). A version of the Eating Index (HEI) as a summary measure of dietary improved system was used to collect data for the quality (Kennedy et al., 1995). Each of these reference 1994-98 CSFII, and the final version is being used to standards is discussed in turn below. collect data in NHANES-IV. A comparable system is included in the Nutrition Data System (NDS), man- Recommended Dietary Allowances. Most FANP aged by the Nutrition Coordinating Center (NCC) at researchers compared mean intakes of participants and the University of Minnesota (NCC, 2001). nonparticipants, expressed as a percentage of age- and gender-appropriate RDAs. Some researchers compared Recent guidelines for dietary assessment issued by the the proportion of individuals in each group with Institute of Medicine (IOM, 2001) recommend that intakes below a defined cutoff, generally between 70 studies examining dietary intakes of groups collect a and 100 percent of the RDA. The latter approach is minimum of 2 nonconsecutive days or 3 consecutive less common, perhaps because an expert panel con- days of data for a subgroup of the population(s) being vened in the early 1980s by USDA specifically recom- studied. The additional data for the subgroup(s) can be mended against the use of fixed cutoffs relative to the used to adjust intake distributions for day-to-day, with- RDAs as a means of assessing the prevalence of inade- in-person variation (IOM, 2001).13 The adjustments quate intakes (NRC, 1986). provide reliable estimates of usual energy and nutrient intakes. These improved dietary assessment methods In assessing program impacts, researchers generally are just beginning to appear in FANP research deemed a significantly greater mean intake among par- (McLaughlin et al., 2002). ticipants or a significantly greater percentage of participants with intakes above a specified cutoff as evi- Nutrient estimates generated from dietary intake data dence of a positive program effect. Effects were char- generally include only the nutrients provided by the acterized as program participation leading to foods and beverages consumed. While studies may “increased intake(s).” collect information on use of vitamin and mineral supplements, the contributions of supplements are seldom Although these interpretations are common in the included in the estimates.14 None of the studies available literature, differences in the mean percentage reviewed for this report included contributions from of the RDA consumed, or in the proportion of individ- supplements. uals consuming some percentage of the RDA, do not provide information on the underlying question: Is the Comparison to Reference Standards percentage of FANP participants with adequate diets Most studies that have examined the impact of FANPs different than the percentage of nonparticipants with on nutrient intakes assessed intakes in reference to adequate diets? Even when mean nutrient intake of a group approximates or exceeds the RDA, significant proportions of the population may have inadequate 13Adjustment of intake distributions is necessary to develop accurate estimates of the proportion of the population with inadequate intakes. If intakes. On the other hand, use of RDA-based cutoffs research goals are limited to estimates of mean intake for each group, addi- seriously overestimates the proportion of a group at tional days of data are not necessary as long as sample sizes are sufficient risk of inadequate intake because, by definition, the (IOM, 2001). 14This trend may be changing. There is increasing interest in basing RDA exceeds the needs of nearly all (97-98 percent) assessment of nutrient intake on complete intake data, including vitamins healthy individuals in the group (IOM, 2001). and minerals provided by supplements. However, because supplement use can be intermittent and because most extant data have inconsistent refer- Thus, the available research provides an imperfect pic- ence time periods for dietary intake data and supplement data (previous 24- hours vs. use during preceding month or week), combining the two sources ture of both the prevalence of inadequate intakes and of data is not a straightforward task. the substantive significance of differences in intakes of

22 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods

FANP participants and nonparticipants. That is, the always leads to an overestimation of the problem available data provide information on whether FANP (IOM, 2001).16 Similarly, using observed intakes participants have “increased intakes” of food energy or rather than usual intakes tends to overestimate the per- key nutrients relative to nonparticipants but do not centage of individuals falling below a given cutoff provide any information on whether these differences because the distribution of observed intakes is usually affect the likelihood that FANP participants consume wider than the distribution of usual intakes. adequate amounts of food energy or nutrients. At the time this report was finalized, only one FANP This imperfect picture of the risk of inadequacy reflects study had used the EAR cut-point method to estimate a limitation in the reference standards and dietary assess- the effect of FANP participation on the prevalence of ment methods available when most of the existing FANP inadequate intakes (McLaughlin et al., 2002). research was conducted, rather than shortcomings in Applying the EAR, in combination with data on usual the research per se. This limitation has been addressed intakes, is not as straightforward as one might expect in the Dietary Reference Intakes (DRIs), a revised set because (1) the procedures used to estimate usual of nutrient intake standards that has replaced the intakes adjust distributions rather than individual esti- RDAs (IOM, 1999, 2000a, 2000b, 2002a, 2002b). mates and (2) the IOM specifically cautions against using a binary variable to represent inadequacy in a The development of the DRIs has led to statistically standard regression model (IOM, 2001). The DRI based guidance on estimating the prevalence of inade- applications report outlines an analysis strategy for quate intakes of population groups (IOM, 2001). The assessing the impact of FANP participation on the recommended approach, referred to as the “EAR cut- prevalence of inadequate intakes (IOM, 2001). point method,” differs in two important ways from the approach used in previous research. First, assessment Dietary Guidelines for Americans. The Dietary of adequacy is based on the Estimated Average Guidelines for Americans (DGAs) were developed Requirement (EAR) rather than the RDA. The EAR is specifically to provide consumers with recommenda- the level of intake estimated to meet the requirements tions that could be used to plan healthful diets of half of the healthy individuals in a given gender and (USDA/HHS, 2000). The DGAs have been revised life-stage group.15 It was developed specifically to over the years but have always stipulated moderate provide a better standard for assessing the adequacy of intake of fat, saturated fat, cholesterol, and sodium. nutrient intakes than is possible with the RDA. Relatively few FANP studies have used the DGAs to Second, assessment is based on estimates of usual assess dietary intakes of program participants vs. non- rather than observed intakes. As discussed above, esti- participants. Most research that has used the DGAs mation of usual intakes requires collecting 2 noncon- compared intakes of total fat and/or saturated fat, as a secutive or 3 consecutive days of intake data for a sub- percentage of total energy intake, to DGA recommen- group of the population(s) under study. These data are dations. Because early versions of the DGAs did not then used to adjust the distribution of intakes to include quantitative recommendations for cholesterol remove within-person variation and better represent and sodium intake, most studies used recommenda- usual intake patterns. tions from the NRC, which include a maximum of 300 mg per day for cholesterol and a maximum of 2,400 Compared with estimates from previous research, the mg per day for sodium (NRC, 1989b). The NRC rec- recommended approach to estimating the prevalence of ommendations for these two nutrients, which were inadequate intakes is likely to yield lower estimates of incorporated into guidelines for nutrition labeling, are the prevalence of inadequacy because, as noted, using the ones now included in the DGAs. the RDA as a reference point for assessing adequacy The DRIs have defined a new reference standard for intake of total fat, referred to as an Acceptable 15For some nutrients, most notably calcium, available data were insufficient to establish an EAR. In these instances, a different DRI—an Adequate Intake or AI—was established. The AI is a level of intake that is assumed to be adequate, based on observed or experimentally determined 16For some nutrients, the estimated prevalence of inadequate intakes estimates of intake. The DRIs also define ULs (Tolerable Upper Intake would be lower even if the old approach was replicated using the latest Levels) for selected nutrients. The UL is the highest intake likely to pose RDAs because the new RDAs for some nutrients differ substantially from no risk of adverse health effects. The DRI applications report provides previous RDAs. For example, for children ages 1-3, the 1989 RDAs for zinc guidance on appropriate uses of AIs and ULs in assessing nutrient intakes and vitamin C were, respectively, 10 mg and 40 mg. The new RDAs for of groups (IOM, 2001). these nutrients are substantially lower, at 3 mg (zinc) and 15 mg (vitamin C).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 23 Chapter 2: Research Methods

Macronutrient Distribution Range (AMDR) (IOM, friends, (3) having no one to confide in, (4) having 2002b). AMDRs have also been defined for carbohy- children that do not visit them, and (5) feeling lonely drates, protein, and specific types of polyunsaturated more often. A third study defined socialization based fatty acids. AMDRs have not been defined for saturat- on number of social contacts (with relatives and/or ed fat or cholesterol because these dietary components friends) per month. have no known beneficial effect in preventing chronic disease and are not required at any level in the diet Other Measures of Nutrition (IOM, 2002b). DRIs for electrolytes, including sodi- and Health Status um, are currently in development (IOM, 2003). While the majority of studies of the impact of FANPs on nutrition- and health-related outcomes have focused The Healthy Eating Index. Very few FANP studies on food expenditures, household nutrient availability, have examined impacts of FANP participation on the and/or individual dietary intake, some studies have HEI. Developed by USDA’s Center for Nutrition examined impacts on longer term measures of nutri- Policy and Promotion (CNPP), the HEI is a summary tion and health status. The most studied outcome in measure of overall diet quality (Kennedy et al., 1995). this group is birthweight (and related measures). Others It is based on 10 component scores, all of which are include measures of food sufficiency/security/insecuri- weighted equally in the total score. The component ty, nutritional biochemistries, linear growth in chil- scores measure different aspects of a healthy diet, dren, body weight, and school/academic performance. based on current public health recommendations. Five of the component scores are food-based and evaluate Birthweight and Related Measures food consumption compared with the recommendations of the Food Guide Pyramid for grains, vegeta- Impacts of FANP participation on birthweight—perhaps bles, fruits, dairy, and meat (USDA/CNPP, 1996). Four the most fundamental measure of nutrition and health component scores are nutrient-based and assess com- status in infants—has focused almost exclusively on pliance with the DGA recommendations for intake of the WIC program. This is an obvious and appropriate fat and saturated fat (USDA/HHS, 2000), as well as focus, given that one of the issues WIC was specifical- with the NRC recommendations for intake of choles- ly designed to address is birth outcomes among low- terol and sodium (NRC, 1989b). The 10th component income pregnant women. Note, however, that birth- score is food-based and assesses the level of variety in weight reflects multiple influences exerted both before the diet. Dietary variety is stressed in the Food Guide and during a pregnancy. These include, but are not Pyramid, the Dietary Guidelines, and the NRC’s diet limited to, maternal health and nutrition, intrauterine and health recommendations (Basiotis et al., 2002). exposures (tobacco, drugs, alcohol), and genetic factors.

Health-Related Behaviors Compared with the measures discussed in the previous sections, reliable and complete data on birthweight, Breastfeeding which is routinely measured at birth and recorded on A handful of studies have examined breastfeeding ini- the birth certificate, is easy to collect. However, proper tiation and duration among WIC participants and non- interpretation of data on birthweight depends on relat- participants. Initiation is generally defined as ever having birthweight to the expected weight for the infant’s ing breastfed, regardless of frequency or duration. gestational age (duration of pregnancy). Infants who Duration is measured as total length of time and/or as are below the expected weight are classified as having the percentage of mothers who breastfed for 6 months intrauterine growth retardation (IUGR). IUGR infants or more. are at increased risk for adverse birth outcomes, compared with those of low birthweight whose weight is Socialization Among the Elderly appropriate for their gestational age. Infants born at The ENP was designed to address the psychological full term (39+ weeks) with a birth weight of less than and sociological needs of the elderly as well as their 2,500 gm (5.5 pounds) are classified as IUGR. nutritional needs. Studies that have compared social- Another issue that affects interpretation of data on ization among ENP participants and nonparticipants birthweight is the simultaneity of WIC participation have used two different approaches. Two studies clas- and gestational age. Women who deliver early have sified respondents based on a five-point isolation less chance of enrolling in WIC than women who go index: (1) living alone, (2) reporting having too few to term. Consequently, both the decision to participate

24 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods in WIC and the length of participation are inexorably Impacts on nutritional biochemistries are best assessed linked with gestational age, an important predictor of using a design that compares participants (and poten- most birth outcomes. Moreover, women who enroll tially nonparticipants) before and after FANP partici- late in pregnancy will automatically have better out- pation. As described earlier, Yip and his colleagues comes than other women by virtue of their increased (1987) conducted a widely recognized study of the gestation. This simultaneity means that assessments of impact of WIC on the prevalence of anemia among the impact of WIC on birthweight that rely on a binary young children, using a classic “participants, before indicator of participation are likely to overstate the vs. after” design. impact of the program. Moreover, because the duration of WIC participation is also simultaneous with gesta- Studies that rely on single measures of iron status (or tional age, a traditional dose-response approach other nutritional biochemistries) are subject to signifi- employed by several studies—estimating WIC impacts cant selection bias, particularly WIC studies because based on number of months of WIC participation—is low blood levels of iron and other nutrients are used to not a viable solution to the problem. define eligibility for WIC participation.

Gordon and Nelson (1995) studied several approaches Linear Growth in Children to addressing the relationship between the timing of One of the most fundamental measures of health status WIC enrollment and gestational age. These approaches in preschool children is the attainment of normal included omitting very late WIC enrollees (enrolled growth. Failure to attain normal linear growth (stunt- after the eighth month), including gestational age as an ing) is a highly sensitive indicator of underlying nutri- independent variable in the birthweight regression, and tional deficits or other health problems. Height-for-age defining several cohorts of WIC participants by gesta- is used to assess the adequacy of linear growth, rela- tional age (pregnancy duration) at the time of WIC tive to growth curves established by the Center for enrollment. The authors found, however, that these Disease Control and Prevention (CDC) (Kuczmarski et approaches systematically underestimated the impact al., 2002). Height-for-age below the fifth percentile is of WIC and suggested that results from analyses using indicative of growth retardation (HHS, 2000). a binary indicator (participant vs. nonparticipant) and results of analyses that compare various cohorts of Similar to nutritional biochemistries, proper assess- WIC participants (e.g., early vs. late enrollees) bound ment of the impact of FANP participation on measures the likely magnitude of the effect. of linear growth in children requires at least two measurements, ideally collected for both treatment and con- Food Security trol groups (World Health Organization, 1995). For In 1997, USDA released the 18-item Federal food secu- example, children of Asian descent, many of whom rity module, the currently accepted standard for measur- came into the United States as refugees in the late ing household and individual food security (Price et al., 1970s and early 1980s, had an increased prevalence of 1997; Bickel et al., 2000). Studies completed before growth stunting relative to other children in the WIC 1997 used one or more of the questions included in the program. Over time, coincident with participation in early food security assessment work done by Wehler et WIC, the prevalence of stunting decreased significant- al. (1991) and by Radimer and her colleagues at Cornell ly to levels approaching those of other low-income University (1992). Studies completed after 1997 used children served by WIC (Yip et al., 1993). either the early questions or the 18-item module. Body Weight Nutritional Biochemistries The substantial increase in the prevalence of over- Several studies have examined the impact of FANP weight and obesity in the United States over the past participation on blood levels of key nutrients. The several decades has heightened interest in this aspect nutrient studied most often is iron. Iron deficiency is of nutritional status among low-income Americans. the most common known form of nutritional deficien- Few studies have attempted to estimate the impact of cy, affecting the entire age span from infancy to old FANP participation on this indicator, and none has age. In infancy and early childhood, iron deficiency is studied the issue adequately. Development of over- an especially important problem that may be associated weight and obesity is a complex process that takes with anemia as well as with delayed psychomotor place over a long period and is influenced by a number development (de Andraca et al., 1997). of factors other than dietary intake, including levels of

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 25 Chapter 2: Research Methods physical activity/inactivity and genetics. Moreover, overweight in childhood, however, because childhood low-income and food-insecure individuals are more mortality is not associated with weight, and weight- likely to be overweight or obese than higher income related morbidity in childhood is too infrequent to and food-secure individuals. This confounding makes define meaningful cutoffs (Cole, 2001). Therefore, it difficult to assess relationships between these char- children are classified as overweight by comparing acteristics using cross-sectional data. their weights and heights with appropriate reference populations. For children, overweight is defined as a For adults, overweight and obesity are defined based BMI at or above the 95th percentile on CDC growth on body mass index (BMI), a measure of the relation- charts, which define BMI percentile distributions by ship between height and weight that is commonly age and gender (CDC, 2003). Children with BMIs accepted for classifying adiposity (or fatness) in adults between the 85th and 95th percentiles are considered 17 (CDC, 2003). For adults, a healthy weight is defined to be at risk of overweight. as a BMI of at least 18.5 but less than 25. Overweight is defined as a BMI between 25 and 30, and obesity as School Performance a BMI of 30 or more. A BMI of less than 18.5 indi- A relatively recent body of research has examined cates extreme thinness or underweight. impacts of breakfast consumption on school performance. Virtually all of these studies have evaluated the Classifying children as overweight is fundamentally issue within the context of demonstration projects of different from classifying adults (Cole, 2001). Adults “universal free” school breakfast programs—that made have traditionally been classified as overweight based breakfast available to all students free of charge, on life insurance mortality data and data relating regardless of household income. Measures examined weight status to morbidity and mortality (Troiano and include attendance and tardiness, academic achieve- Flegal, 1998). These criteria cannot be used to define ment—generally measured with standardized test scores—cognitive functioning, student behavior, and 17BMI is equal to [weight in kilograms] / [height in meters] 2. referrals to school nurses.

26 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods

References Fraker, T.M. 1990. Effects of Food Stamps on Food Consumption: A Review of the Literature. USDA, Basiotis, P., A. Carlson, S. Gerrior, et al. 2002. The Food and Nutrition Service. Healthy Eating Index: 1999-2000. CNPP-12. USDA, Center for Nutrition Policy and Promotion. Fraker, T.M., A.P. Martini, J.C. Ohls, et al. 1992. The Evaluation of the Alabama Food Stamp Cash-out Beaton, G.H., J. Milner, V. McGuire, et al. 1983. Demonstration: Volume 1: Recipient Impacts. USDA, “Sources of Variation in 24-hour Recall Data: Food and Nutrition Service. Implications for Nutrition Study Design and Interpretation—Carbohydrate Sources, Vitamins and Gersovitz, M., J.P. Madden, and H. Smiciklas-Wright. Minerals,” American Journal of Clinical Nutrition 1978. “Validity of the 24-hour Dietary Recall and 37:895-96. Seven-day Food Record for Group Comparisons,” Journal of the American Dietetic Association 73:48-55. Bickel, G., M. Nord, C. Price, et al. 2000. Guide to Measuring Household Food Security: Revised 2000. Gleason, P., A. Rangarajan, and C. Olson. 2000. USDA, Food and Nutrition Service. Dietary Intake and Dietary Attitudes Among Food Stamp Participants and Other Low-Income Briefel, R.R., C.T. Sempos, M.A. McDowell, et al. Individuals. USDA, Food and Nutrition Service. 1997. “Dietary Methods Research in the Third National Health and Nutrition Examination Survey: Gleason, P., and C. Suitor. 2001. Children’s Diets in Underreporting of Energy Intake,” American Journal the Mid-1990s: Dietary Intake and Its Relationship of Clinical Nutrition 65(4, Supplement):1203S-09S. with School Meal Participation. USDA, Food and Nutrition Service. Campbell, J.T., and J.C. Stanley. 1963. Experimental and Quasi-Experimental Designs for Research. Gordon, A., and L. Nelson. 1995. Characteristics and Chicago, IL: Rand McNally. Outcomes of WIC Participants and Nonparticipants: Analysis of the 1988 National Maternal and Infant Centers for Disease Control and Prevention. 2003. Health Survey. USDA, Food and Nutrition Service. “CDC Growth Chart Training Modules.” Available: http://www.cdc.gov/nccdphp/dnpa. Accessed May 2003. Hamilton, W.L., and P.H. Rossi. 2002. Effects of Food Assistance and Nutrition Programs on Nutrition and Cole, N. 2001. The Prevalence of Overweight Among Health: Volume 1, Research Design. FANRR-19-1. WIC Children. USDA, Food and Nutrition Service. USDA, Economic Research Service. de Andraca, I., M. Castillo, and T. Walter. 1997. Heckman, J. 1979. “Sample Bias as a Specification “Psychomotor Development and Behavior in Iron- Error,” Econometrica 47:153-62. Deficient Anemic Infants,” Nutrition Reviews 55(4):125-32. Heckman, J., and V.J. Hotz. 1989. “Choosing Among Alternative Nonexperimental Methods for Estimating Devaney, B., L. Bilheimer, and J. Schore. 1990. The the Impact of Social Programs: The Case of Savings in Medicaid Costs for Newborns and Their Manpower Training,” Journal of the American Mothers from Prenatal WIC Participation in the WIC Statistical Association 84:862-80. Program: Volume 1. USDA, Food and Nutrition Service. Institute of Medicine. 2003. “The Current Project Devaney, B., L. Bilheimer, and J. Schore. 1991. The System—Project Title: Dietary Reference Intakes for Savings in Medicaid Costs for Newborns and Their Electrolytes and Water.” Available: http://www.iom. Mothers from Prenatal WIC Participation in the WIC edu/project.asp?id=3969. Accessed July 2003. Program: Volume 2. USDA, Food and Nutrition Service. Institute of Medicine, Food and Nutrition Board. Devaney, B., and T. Fraker. 1989. “The Effect of Food 2002a. Dietary Reference Intakes: Vitamin A, Vitamin Stamps on Food Expenditures: An Assessment of K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Findings from the Nationwide Food Consumption Manganese, Molybdenum, Nickel, Silicon, Vanadium, Survey,” American Journal of Agricultural Economics and Zinc. Washington, DC: National Academy Press. 71(1):99-104.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 27 Chapter 2: Research Methods

Institute of Medicine, Food and Nutrition Board. Levedahl, J.W. 1991. The Effect of Food Stamps and 2002b. Dietary Reference Intakes: Energy, Income on Household Food Expenditures. TB-1794. Carbohydrates, Fiber, Fats, Fatty Acids, Cholesterol, USDA, Economic Research Service. Protein, and Amino Acids (Macronutrients). Washington, DC: National Academy Press. Madden, J.P., S.J. Goodman, and H.A. Guthrie. 1976. “Validity of the 24-hour Recall: Analysis of Data Institute of Medicine, Food and Nutrition Board. 2001. Obtained from Elderly Subjects,” Journal of the Dietary Reference Intakes: Application in Dietary American Dietetic Association 68:143-147. Assessment. Washington, DC: National Academy Press. McDowell, M.A., R.R. Briefel, R.A. Warren, et al. Institute of Medicine, Food and Nutrition Board. 1989. “The Dietary Data Collection System: An 2000a. Dietary Reference Intakes: Thiamin, Automated Interview and Coding System for

Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, NHANES-III,” in P.J. Stumbo (ed.), Proceedings of Pantothenic Acid, Biotin, and Choline. Washington, the Fourteenth National Nutrient Databank DC: National Academy Press. Conference. Ithaca, NY: The CBORD Group, Inc.

Institute of Medicine, Food and Nutrition Board. McLaughlin, J., L. Bernstein, M.K. Crepinsek, et al. 2000b. Dietary Reference Intakes: Vitamin C, Vitamin 2002. Evaluation of the School Breakfast Program E, Selenium, and Carotenoids. Washington, DC: Pilot Project: Findings from the First Year of National Academy Press. Implementation. USDA, Food and Nutrition Service.

Institute of Medicine, Food and Nutrition Board. 1999. Metcoff, J., P. Costiloe, W.M. Crosby, et al. 1985. Dietary Reference Intakes: Calcium, Phosphorus, “Effect of Food Supplementation (WIC) During Magnesium, Vitamin D, Fluoride. Washington, DC: Pregnancy on Birth Weight,” American Journal of National Academy Press. Clinical Nutrition 41:933-47.

Kennedy, E.T., and S. Gershoff. 1982. “Effect of WIC Moshfegh, A.J., N.R. Raper, L.A. Ingwersen, et al. Supplemental Feeding on Hemoglobin and Hematocrit 2001. “An Automated Approach to 24-hour Dietary of Prenatal Patients,” Journal of the American Dietetic Recall Methodology,” Annals of Nutrition and Association 80:227-30. Metabolism 45(1, Supplement):156 (2.04.157).

Kennedy, E.T., J. Ohls, S. Carlson, et al. 1995. “The Murphy, J.M., and M. Pagano. 2001. Effects of a Healthy Eating Index: Design and Applications,” Universally Free, In-Classroom School Breakfast Journal of the American Dietetic Association Program: Final Report from the Third Year of the 95(10):1103-09. Maryland Meals for Achievement Evaluation. Boston, MA: Massachusetts General Hospital. Kramer-LeBlanc, C., P. Basiotis, and E. Kennedy. 1997. “Maintaining Food and Nutrition Security in the National Research Council. 1989a. Recommended United States with Welfare Reform,” American Dietary Allowances, 10th edition. Washington, DC: Journal of Agricultural Economics 79(5):1600-07. National Academy Press.

Kuczmarski, R., C. Ogden, L. Guo, et al. 2002. 2000 National Research Council. 1989b. Diet and Health: CDC Growth Charts for the United States: Methods Implications for Reducing Chronic Disease Risk. and Development. Vital and Health Statistics Series 11, Washington, DC: National Academy Press. No. 246. Centers for Disease Control and Prevention, National Center for Health Statistics. National Research Council. 1986. Nutrient Adequacy: Assessment Using Food Consumption Surveys. La Londe, R., and R. Maynard. 1987. “How Precise Washington, DC: National Academy Press. are Evaluations of Employment and Training Programs: Evidence from a Field Experiment,” Neenan, P.H., and C.G. Davis. 1978. Impact of the Evaluation Review 11(4):428-51. Food Stamp Program on Low-income Household Food Consumption in Florida. Gainesville, FL: University of Florida, Institute of Food and Agricultural Sciences.

28 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 2: Research Methods

Nutrition Coordinating Center. 2001. Nutrition Data Thompson, F.E., and T. Byers. 1994. “Dietary System for Research. Minneapolis, MN: University of Assessment Resource Manual,” Journal of Nutrition Minnesota. 124(11, Supplement).

Ohls, J.C., T.M. Fraker, A.P. Martini, et al. 1992. The Troiano, R.P., and Flegal, K.M. 1998. “Overweight Effects of Cash-out on Food Use by Food Stamp Children and Adolescents: Description, Epidemiology, Program Participants in San Diego. USDA, Food and and Demographics,” Pediatrics 101:497-504. Nutrition Service. U.S. Department of Agriculture, Center for Nutrition Peterson, K., M. Davison, K. Wahlstrom, et al. 2003. Policy and Promotion. 1996. The Food Guide Fast Break to Learning School Breakfast Program: A Pyramid. Home and Garden Bulletin No. 252. Report of the Third Year Results, 2001-2002. Minneapolis, MN: University of Minnesota. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 2000. Nutrition and Ponza, M., J.C. Ohls, B.E. Millen, et al. 1996. Serving Your Health: Dietary Guidelines for Americans, 5th Elders at Risk: The Older Americans Act Nutrition edition. Home and Garden Bulletin No. 232. Programs, National Evaluation of the Elderly Nutrition Program, 1993-1995, Volumes I, II, and III. U.S. Department of Health and Human Services. 2000. U.S. Department of Health and Human Services, Healthy People 2010: Understanding and Improving Administration on Aging. Health, 2nd Edition.

Price, C., W.L. Hamilton, and J.C. Cook. 1997. Guide Wehler, C.A., R.I. Scott, and J.J. Anderson. 1991. A to Implementing the Core Food Security Module. Survey of Childhood Hunger in the United States. USDA, Food and Nutrition Service. (The guide was Community Childhood Hunger Identification Project. revised and updated in 2000; see Bickel et al., 2000). Washington, DC: Food Research and Action Center.

Radimer, K., C. Olson, J.C. Greene, et al. 1992. World Health Organization. 1995. Physical Status: The “Understanding Hunger and Developing Indicators to Use and Interpretation of Anthropometry. Geneva, Assess it in Women and Children,” Journal of Switzerland: World Health Organization, WHO Nutrition Education 24(1, Supplement):36s-44s. Technical Report Series, No. 854.

Rossi, P.H. 1998. Feeding the Poor: Assessing Federal Yip, R., N. Binkin, L. Fleshood, et al. 1987. “Declining Food Aid. Washington, DC: The AEI Press. Prevalence of Anemia Among Low-income Children in the United States,” Journal of the American Medical Rubin, D.B. 1979. “Using Multivariate Matched Association 258(12):1619-23. Sampling and Regression Adjustment To Control Bias in Observational Studies,” Journal of the American Yip, R., K. Scanlon, and F. Trowbridge. 1993. “Trends Statistical Association 74:318-28. and Patterns in Height and Weight Status of Low- income U.S. Children,” Critical Reviews in Food Rush, D., J. Leighton, N. Sloan, et al. 1988. Science and Nutrition 33(4/5):409-21. “Historical Study of Pregnancy Outcomes,” American Journal of Clinical Nutrition 48:412-28.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 29 Chapter 3 Food Stamp Program

The Food Stamp Program (FSP) stands at the intersec- as long as the pooled income of the remainder of the tion of two sets of Federal programs: those for whom household is less than 165 percent of poverty. Monthly the primary goal is improving access to adequate nutri- benefit levels increase with the number of people in tion and those for whom it is income maintenance. The the household but not at a flat rate per person. FSP is particularly important because of its universali- ty; it is an entitlement program with eligibility require- Program Eligibility ments based almost solely on financial need, while the To be eligible for the FSP, a household must meet cer- other major food and nutrition assistance programs tain financial, work-related, and categorical require- (FANPs) are targeted toward certain types of individu- ments. Financial requirements include a gross income als or households. Food stamp benefits are distributed limit of 130 percent of poverty, a net income limit as electronic transfers with an explicit cash value, (gross income less allowable deductions) of 100 percent which can be used only to purchase food for home of poverty, and a countable assets limit of $2,000. consumption. Households with elderly or disabled members are not subject to the gross income limit, are eligible for The FSP is the cornerstone of the Nation’s nutrition deductions for medical expenses and increased deduc- safety net. In FY 2002, the total Federal expenditure tions for shelter costs, and have a countable assets limit for the FSP was $20.7 billion, or about 54 percent of of $3,000. Households in which all members receive the $38 billion Federal expenditure for FANPs. The Temporary Assistance for Needy Families (TANF), program served more than 19 million participants per Supplemental Security Income (SSI), or general assis- month (U.S. Department of Agriculture (USDA), Food tance are exempt from both income and asset tests. and Nutrition Service (FNS), 2003a). Work-related eligibility conditions require certain Program Overview household members to register for work, accept suitable job offers, and comply with State welfare agency The goal of the FSP is to “safeguard the health and work or training programs. Finally, a few groups are well-being of the Nation’s population by raising the categorically ineligible for the FSP, including strikers, level of nutrition among low-income individuals.” To most people who are not citizens or permanent resi- achieve this objective, the FSP provides electronic dents, postsecondary students, and people living in benefits that can be used at most retail grocery institutional settings. stores.18 Program Participation The FSP began as a small pilot program in 1961.19 The program expanded during the 1960s and early Because the FSP is available to most people who meet 1970s, finally reaching nationwide coverage in 1975. income and resource standards, the households that The FSP specifies the household rather than any indi- participate in the program are quite diverse and repre- vidual living in the household as the program partici- sent a broad spectrum of the needy population (Rosso, pant. A household includes all people living together 2003). In FY 2001, almost all FSP participants lived in in a dwelling who normally purchase food and prepare poverty. The gross monthly income of 89 percent of meals as a unit. Eligibility is based on the pooled FSP households was less than or equal to 100 percent income, resources, and expenditures of all members of of the poverty guideline. More than half of all FSP the household. Elderly and disabled individuals who households had incomes that were less than or equal to cannot prepare and purchase food because of a sub- 75 percent of the poverty guideline, and one-third had stantial disability may apply as a separate household, incomes that were less than or equal to 50 percent of the poverty guideline (Rosso, 2003).

18FSP benefits can be used only to purchase food or seeds and plants Administrative data for FY 2001 (Rosso, 2003; Tuttle, used to produce food. 2002) indicated that the vast majority (88 percent) of 19An earlier version of the FSP, which distributed surplus commodities to needy families, came to an end in 1943. For a detailed description of the FSP households included either a child, an elderly per- program and its history, see, for example, Ohls and Beebout (1993). son (60 or older), or a disabled person. More than half

30 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

(54 percent) of all FSP households had children. Of placed strict limits on participation for “able-bodied these, more than two-thirds (67 percent) were single- adults without dependents” (ABAWDs). (Eligibility parent households. Twenty percent of FSP households restrictions for some resident aliens and several other included one or more elderly individuals. The majority groups were rescinded in 1998.) Since the PRWORA (80 percent) of these households were elderly individ- reforms, participation in the Aid to Families With uals living alone. More than a quarter (28 percent) of Dependent Children (AFDC)/TANF programs has all FSP households included a disabled individual, and decreased dramatically, and such families are account- 58 percent of these households were disabled people ing for a decreasing share of all FSP households.20 living alone. Overall, 51 percent of all FSP participants Between 1995 and 2001, TANF-recipient households in FY 2001 were children, 10 percent were elderly, and fell from 38 percent to 26 percent of all FSP house- 13 percent were disabled. holds (Rosso, 2003).

Participation in the FSP has changed dramatically in While economic factors and program policies explain recent years. The number of participants increased by a substantial portion of the decline in FSP participa- about 47 percent between 1989 and 1994 (from 18.9 tion, other factors clearly were also involved. From the million in 1989 to a record high of 28.0 million in mid- to late 1990s, FSP participation declined not only March 1994) (Tuttle, 2002). After that, participation because fewer individuals were eligible, but also because declined steadily through 2000. Between 1994 and of a noteworthy drop in the percentage of eligible indi- 2000, the number of individuals participating in the viduals who actually elected to participate. Indeed, the FSP decreased from 28.0 million to 16.9 million, or by rate of FSP participation among income-eligible peo- 40 percent (Tuttle, 2002). Between 2000 and 2001, ple declined from 75 percent in 1994 to 58 percent in participation increased for the first time in 6 years, by 1999 (Cunnyngham, 2002). Factors that may have approximately 1 million people, or 6 percent. contributed to this decline include confusion about eligibility, erroneous termination of FSP benefits when A number of investigators have studied the shifts in TANF cases terminated, effects of TANF diversion FSP participation, particularly the unprecedented programs on the FSP application process, and shorten- decline in the mid- to late 1990s. (See, for example, ing of FSP certification periods (Kornfeld, 2002). In USDA/FNS, 2001; Jacobsen et al., 2001; Figlio et al., 2000, FSP participation rates increased slightly for the 2000; Wilde et al., 2000a, 2000b; Wallace and Blank, first time in 5 years, from 58 to 59 percent 1999.) There is strong evidence that economic condi- (Cunnyngham, 2002). tions played a role in the shifts seen in FSP participation levels over the past 10 to 15 years. The dramatic Program Benefits increase in participation in the early 1990s went hand- Food stamp benefits per household are determined by in-hand with a declining economy (Tuttle, 2002). a schedule of maximum benefits per household size. Similarly, the drop in participation between 1994 and Individual households receive the maximum benefit 2000 was consistent with an improving economy. The less 30 percent of the household’s net income (house- recent upswing in participation may be associated with holds are expected to set aside 30 percent of their non- the latest economic downturn. food stamp disposable income for food). Benefit levels The relationship between FSP participation and eco- are based on the Thrifty Food Plan, an estimate of nomic indicators does not tell the whole story, howev- what it costs for a household of a given size to pur- er. FSP participation and unemployment rates diverge chase the foods required for a nutritious diet. USDA at some points in time, indicating that factors other annually determines the cost of the Thrifty Food Plan. than the economy have been in play (Wilde, 2001). Maximum monthly food stamp allotments for FY Key changes in program policies and regulations may 2003, before deductions, are shown in table 7. also have contributed to fluctuating FSP rolls, although A key feature of the program before 1979 was the pur- it is generally believed that the impact of program chase requirement. The benefit allotment for house- policies is substantially less than that of economic con- holds of a given size had a fixed value. Participating ditions. The most notable policy changes in recent years households paid cash for their allotment, with the pay- include reforms enacted in 1996 as part of the Personal ment amount depending on household income. The Responsibility and Work Opportunity Reconciliation Act (PRWORA). These changes restricted program participation for resident aliens and other subgroups and 20Under PRWORA, the AFDC program was replaced by TANF.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 31 Chapter 3: Food Stamp Program difference between the amount paid and the value of Nutrition Education food stamps received was termed the “bonus.” The Nutrition education is a relatively recent, though purchase requirement was eliminated in 1979. increasing, emphasis in the FSP. In FY 1998, FNS Subsequently, eligible households simply received made a “renewed commitment to nutrition education” what had previously been the bonus amount of in the FSP (and all FANPs) and established a special coupons. staff within the agency to “refocus efforts toward The FSP originally issued benefits in the form of paper nutrition and nutrition education” (USDA/FNS, coupons of various denominations. Recipients redeemed 2003b). The focus on nutrition education as an these coupons for food at authorized stores. After a adjunct to the economic benefits provided by the series of demonstration projects, FNS authorized FSP reflects an important shift in the overarching States to use electronic benefits transfer (EBT) sys- mission and objectives of the program. As stated tems in place of paper coupons. In an EBT system, the in FNS’s strategic plan for 2000-05, there is a recipient receives a credit on a computerized account “growing awareness that making sure people have for the amount of the monthly benefit. To make a pur- enough food is not enough; people must have the chase, the recipient presents an EBT card and enters a knowledge and motivation to make food choices personal identification number (PIN) on a point-of-sale that promote health and prevent disease” (USDA/ (POS) terminal. The terminal verifies the amount of FNS, 2000). benefits available, debits the amount of the purchase This growing awareness is based on accumulated from the recipient’s balance, and records a credit for scientific evidence that dietary patterns are associated the retailer. The retailer receives daily an electronic with 4 of the 10 leading causes of death—coronary bank deposit for the net amount of FSP redemptions. heart disease, certain types of cancer, stroke, and Nearly all States use online EBT systems, in which the diabetes—and with the development of obesity and POS terminal communicates with a central computer hypertension (both of which contribute to these and to obtain authorization for each transaction. These other chronic diseases) (Frazao, 1999). In addition, online EBT systems use the same technology, and diet plays an important role in several other health often the same POS equipment, as commercial debit conditions, including osteoporosis, iron-deficiency and credit payment systems. Ohio and Wyoming use anemia, and neural-tube birth defects. Most offline EBT systems, in which a computer chip on the important, low-income individuals, the target card maintains the recipient’s balance and authorizes population for the FANPs, are at increased risk of the transaction. developing many of these health problems (U.S. Department of Health and Human Services PRWORA mandated that all FSP benefits be distrib- (HHS), 2000). uted via electronic transfers. The nationwide changeover from coupons to EBT was completed in The goal of food stamp nutrition education is to pro- June 2004 (USDA, 2004). mote healthy food choices and active lifestyles among FSP participants. Four core elements have been defined for nutrition education efforts: dietary Table 7—Maximum monthly food stamp benefits quality, food security, food safety, and shopping before deductions, FY 2003 behavior/food resource management. Although nutrition education is still a very small part of the Number in Maximum household monthly benefit overall program (less than 1 percent of total program expenditures in FY 2002), efforts in this area have Dollars increased substantially in the past decade. In FY 1992, 1 141 only five States applied for and received optional 2 259 funding for nutrition education activities in the 3 371 FSP, and the Federal share of the expenditure for 4 471 5 560 these activities was $661,000. In FY 2002, 48 States 6 672 had approved nutrition education plans, and Federal 7 743 expenditures for FSP nutrition education exceeded 8 849 $174 million (USDA/FNS, 2003b). Most of this Each additional person +106 increase occurred after FY 1998, when FNS made a

32 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program renewed commitment to nutrition education in the Assessing Impacts of the FSP. Virtually all of the research discussed in this Food Stamp Program chapter was conducted before the increased, and still growing, focus on nutrition education in FSP benefits are expected to directly affect household the FSP. food expenditures. By increasing food expenditures, the FSP is expected to increase the nutrients available Recent Legislative Changes to participating households, and therefore the nutrient intake of individuals in those households. Through this The FSP has been legislatively revised several times path, the FSP may improve other nutrition and health since its inception, but the basic nature of the benefit outcomes, such as food security, birthweight, and iron and the eligible population have remained relatively status. stable. As mentioned, the PRWORA legislation of 1996 placed a time limit on benefits for ABAWDs. This chapter summarizes existing research on the ABAWDS can receive benefits for only 3 months in a impact of the FSP in each of these areas. Three basic 36-month period unless they are working or are partic- approaches have been used to assess FSP impacts on ipating in certain types of qualified work experience or nutrition- and health-related outcomes: workforce programs. States can get approval to exempt ABAWDs from work requirements in designated geo- • Participant vs. nonparticipant designs that compare graphic areas, however, and the legislation provides mean outcomes. for other types of exemptions. In addition, PRWORA made most legal immigrants ineligible for the FSP, but • Dose-response analysis of the effect of the FSP per such households accounted for only a small percentage dollar of benefits. of all recipients, and later legislation in 1998 restored benefits to many of them. Other changes include the • Cashout demonstrations that estimate the impact of a introduction and expansion of employment-related single component of the FSP (the use of coupons) to requirements for various types of households and the obtain lower-bound estimates of impacts. replacement of food stamp coupons with electronic benefit transfers. As described in chapter 2, dose-response analysis is a variant of the “classic” participant vs. nonparticipant More recently, the Food Stamp Reauthorization Act of design. Each of these research approaches, and their 2002 included several provisions to improve access to relative strengths and weaknesses, is now discussed. the FSP and simplify program administration. The 2002 Act removed the prohibition on benefits for sev- Participant vs. Nonparticipant Comparisons eral categories of legally resident aliens, including Several studies have estimated impacts of the FSP by children, elderly or disabled people, and others legally comparing outcomes for FSP participants and nonpar- residing for 5 years. To make benefits more responsive ticipants. These studies generally (but not always) to household circumstances, the 2002 Act modified the compared FSP participants and FSP-eligible nonpartic- standard deduction applied to income when determin- ipants, so that the comparison was limited to people ing benefits, so that the deduction is scaled to family with similar incomes. The comparison is done with size and indexed to inflation. The 2002 Act also multivariate analysis to control for the characteristics authorized a transitional benefit alternative (TBA) for of FSP participants and nonparticipants. An indicator households leaving TANF and wider use of semiannu- of FSP participation captures the direct impact of the al income reporting. Several provisions of the act give FSP—that is, the difference in outcomes between FSP States more flexibility and encourage efforts to pro- participants and nonparticipants that is unexplained by mote FSP access. Most notably, the act lowered the other characteristics. standards for benefit accuracy, replacing the system of enhanced matching tied to payment accuracy with Comparisons between FSP participants and income- bonuses for a broader range of performance objectives. eligible nonparticipants yield direct estimates of the Finally, the 2002 Act repealed the requirement of impacts of the FSP. As discussed in chapter 1, howev- PRWORA that EBT systems be cost-neutral (that is, er, such estimates are subject to selection-bias prob- no more expensive than the inflation-adjusted cost of lems because unmeasured characteristics of FSP par- paper coupon issuance). ticipants may be correlated with both FSP participation and the outcomes of interest. For example, households

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 33 Chapter 3: Food Stamp Program choosing to participate in the FSP may give food receiving zero benefits, rather than nonparticipants. expenditures higher priority (compared with house- Alternatively, nonparticipants may be included in the holds choosing not to participate) even in the absence analysis (with zero benefits). In this case, the coeffi- of the program. In this case, participant vs. nonpartici- cient on the FSP participation indicator, if included in pant comparisons would overstate the impact of the the model, indicates the presence of selection bias.21 FSP, attributing higher food expenditures to FSP participation when, in fact, households participating in Dose-response analysis is not, however, a panacea. FSP have higher food expenditures even in the absence First, functional form is crucial. Because no FSP par- of the program. Conversely, participant vs. nonpartici- ticipants actually receive zero benefits, this approach pant comparisons could understate the impact of the relies on the researcher’s ability to extrapolate the rela- FSP if FSP households are especially needy in unmea- tionship from very low observed benefit levels down sured ways that are unrelated to food (for example, to zero. As will be seen later in this chapter, alternative high medical expenses). Such households, in the functional form assumptions can lead to different esti- absence of the FSP, would spend less on food than mates of FSP impacts. otherwise-similar nonparticipant households. Second, some selection bias may remain because those Several studies, including most of the more recent households that choose to participate when the “dose” ones, have used econometric techniques to attempt to is low—that is, households that receive only a small control for selection bias in estimating program FSP benefit—may be unlike households that partici- impacts. The standard approach is to identify and con- pate in order to receive a large benefit. This difference trol for variables (instruments) that affect FSP partici- seems a less serious matter, however, than the potential pation but do not affect the outcomes of interest. differences between participants and nonparticipants. However, most FSP studies rely on national survey data that have a limited number of potentially useful Similarly, unmeasured household characteristics likely variables. Moreover, these methods provide no guaran- affect both the FSP benefit and food expenditures (as tee that bias has actually been eliminated, and few well as other outcomes). When households that have the valid instruments have been identified in the literature. same measured characteristics but different FSP benefits are compared, one is tempted to think of the compari- Dose-Response Analysis son as an experiment in which Household A, which is essentially similar to Household B, receives more food A key feature of the FSP is that the benefit varies stamps and spends some amount more on food as a across participating households according to estimated consequence. However, if the reason Household A is need (based on the cost of the Thrifty Food Plan for a getting more food stamps than Household B is that given household size and income, minus various Household A is receiving an excess shelter cost deduc- exclusions and deductions). The benefit received by a tion while Household B is living in a rent-free situa- household can be as little as $10 or, in FY 2002 for an tion, one cannot expect outcomes absent the FSP to be eight-person household, as much as $838. Benefits can the same for both households. vary among households of the same size because of differences in total income, in whether income is Despite these caveats, dose-response analysis holds earned or unearned, and in deductions for housing, promise for assessing the impact of the FSP. While this child care, and medical expenses. approach is not as strong as randomly assigning FSP benefits to households, dose-response analysis is Several researchers have taken advantage of the varia- stronger than participant vs. nonparticipant compar- tion in FSP benefit amounts and used dose-response isons because it is less subject to (although not free analysis to identify the marginal impact of FSP bene- from) selection-bias problems. fits. Dose-response studies generally estimate the impact of the FSP based on variations in benefits and impacts among participants only, ignoring nonparticipants entirely. The overall impact of the FSP is esti- 21Selection bias may be said to occur if the expected value of the outcome mated as the impact per dollar of FSP benefits multi- absent the FSP, conditional on the other variables in the model, is different plied by the average FSP benefit. This approach for FSP participants than for nonparticipants. Omitting an indicator of FSP arguably removes a major source of selection bias participation from the specification when it should be present (i.e., when outcomes would be different even in the absence of the program) subjects because the implicit comparison group is households the coefficient on the FSP benefit amount to an omitted-variables bias that that have chosen to participate in the FSP but are is proportional to the true coefficient on FSP participation.

34 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Cashout Demonstrations shows that coupons are more effective than cash in increasing food expenditures. This idea is often The FSP provides to eligible households monthly cash expressed in terms of the marginal propensity to spend value benefits, which can be spent only for food. In on food, or MPS .23 This quantity represents the increase the cashout demonstrations, participating households F in food expenditures per dollar increase in income. The were given checks rather than food stamp coupons, MPS has been found to vary between different types of eliminating the restriction that benefits can be spent F income, being higher for food stamps than for other only for food. Impacts of cashout can be interpreted as sources. Explanations for this difference are as follows: lower-bound estimates of the FSP impact, corresponding to the effects of just one program component— 22 • For some households, the amount of the benefit is namely, the earmarking of benefits. greater than desired food expenditures. These households are “constrained” because they are unable to Lower-bound estimates would not be particularly useful, spend food stamp benefits on nonfood items, MPS =1. given the many available estimates of the impacts of the F FSP as a whole, except that two of the cashout demon- • In multiple-adult households, food stamps are under strations were randomized experiments. If these studies the control of the “food manager” in the household, find that coupon recipients spend significantly more on while a cash benefit can be co-opted by other adults food than cash-benefit recipients, the conclusion (without to purchase other items.24 fear of selection bias) is that the FSP does affect food expenditures. Moreover, if the measured difference is, • When food stamp benefits are received as a lump sum say, $0.20 per dollar of benefits, the conclusion is that the at the beginning of the month, the household has many effect of food stamp coupons on household food expendi- urgent and competing needs. The food stamps can be tures is at least $0.20 on the dollar—and, in fact, that it is used only for food, and so are promptly spent for at least $0.20 more on the dollar than the presumably food.25 An equivalent cash benefit received at the positive effect on food expenditures of ordinary income. beginning of the month, in contrast, might be spent in Similarly, the effect of cashout on household nutrient part on other things, such as health insurance or rent. availability, as measured in the two randomized experi- As the month proceeds, the household cannot go with- ments, may represent the effect of the FSP in general. out food altogether, so more non-food-stamp income is allocated for this purpose, even though the household Food Expenditures spent heavily on food at the beginning of the month. The FSP is virtually certain to result in increased food • Because food stamps are a steady and reliable income purchases, if for no other reason than that the program source for low-income households, they are treated as increases participating households’ incomes and the “permanent income.”26 Hence, they have more power income elasticity for food is positive. That is, increasing a household’s income by $1,000 per year would 23 Some authors use the notation MPCF (marginal propensity to consume always be expected to increase its food expenditures food). This refers not to the consuming (or eating) of food, but to households by some fraction of that amount. allocating their income to consumption goods of various kinds instead of to

savings. To avoid any confusion, the MPSF notation is used here. Economists have debated whether giving households 24This explanation was tested using data from the San Diego cashout experiment by comparing impacts between one- and multiple-adult house- coupons that must be spent on food consumed at home holds (Breunig et al., 2001). The “food manager” hypothesis would suggest is more effective at increasing food expenditures than that cashout would reduce food expenditures by a greater amount in multi- simply giving them a non-earmarked income supple- ple-adult households, which was indeed found to be the case. The authors ment. (See, for example, Southworth, 1945; Senauer and remark that although the household as a whole is unconstrained in its food expenditures, one of the adults may be constrained if he or she does not Young, 1986; Moffitt, 1989.) A simple theory of rational spend anything on food. Giving the household cash instead of food stamps behavior implies that coupons should have the same leads to the constrained adult’s controlling a greater fraction of the house- effect as cash because households can use the coupons to hold’s resources. 25A study in Reading, PA, found that food stamp recipients using elec- free up the money they would otherwise have spent on tronic benefits transfers spent 19 percent of their monthly benefits on the groceries. Nonetheless, a substantial body of evidence day of issuance and 70 percent within the first week (Bartlett and Hart, 1987). Quite similarly, a more recent study in Maryland found that recipients spent 23 percent of their benefits on the day of disbursement and 71 22The households still may have treated these benefits as lightly ear- percent within the first week (Cole, 1997). marked because they were formally identified as a food assistance benefit. 26Permanent income refers to normal or expected income over a long If so, the cashout impacts are an even lower underestimate of the total period of time. Current income is the sum of permanent income and (posi- impact of the FSP. tive or negative) transitory income (see Friedman, 1957).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 35 Chapter 3: Food Stamp Program

to affect routine and nonpostponable expenditures like Consumption Survey (NFCS) and the Bureau of Labor food than do income sources that fluctuate greatly. Statistics’ Consumer Expenditure Survey (BLS-CES). The other two studies that used participant vs. nonpar- • Finally, the psychological effect of earmarked bene- ticipant comparisons are based on State and local data. fits cannot be ignored. It seems to be human nature Fifteen of the 20 dose-response studies used national not to treat food stamps in the same way as cash. survey data and 5 used State and local data. Finally, two When constrained to spend a certain minimum of the five studies of food stamp cashout are based on amount on food, even if the constraint is not bind- cashout demonstrations that used experimental designs ing, households evidently end up allocating more of (in Alabama and San Diego) and three are based on 27 their budget to food. demonstrations that used quasi-experimental designs (in Washington State, Alabama, and Puerto Rico). Research Overview Since the mid-1970s, dozens of researchers have investi- Most studies of the impact of the FSP on food expendi- gated the impact of the FSP on household food expen- tures measured household food expenditures as expendi- ditures. The literature search identified 32 such studies tures for foods used at home, although some studies also completed since 1973. Key characteristics of these examined impacts on total food expenditures (food studies are summarized in table 8. Studies have been used at home and away from home). Food stamp bene- classified by the three alternative research approaches fits can be applied only toward food used at home, but discussed above: participant vs. nonparticipant com- several authors who examined both measures conclud- parisons (Group I), dose-response estimates of the ed that FSP participation induces households to substi- tute food at home for food away from home. MPSF (Group II), and cashout demonstrations (Group III). Participant vs. nonparticipant and dose-response Some studies defined food expenditures as food pur- studies are further subdivided by data source (national chases during a specified period, while others also survey data or State and local studies). Cashout studies included the value of nonpurchased food. A small are separated on the basis of design (randomized number of studies measured food expenditures as food experiment (“pure” cashout) or quasi-experiment). actually used during a particular period. Of the 32 studies, 7 used participant vs. nonparticipant The bulk of the impact estimates are derived from comparisons to estimate the impact of the FSP on models of the form: household food expenditures, 20 used dose-response analyses to estimate the marginal impact of FSP bene- 28 FOOD_EXP = b0 + b1 FSP + b2 BENEFIT + b3 fits, and 5 estimated impacts of food stamp cashout. OTHER INC + b X + u In addition to varying in the basic research approach, 4 these studies varied with respect to data source, defini- Where: tion and measurement of food expenditures, and model specification. With just a few exceptions (Kisker and FOOD_EXP is household expenditure on food; Devaney, 1988; Lane, 1978), researchers used some form of multivariate modeling in their analysis. FSP is an indicator of participation in the Food Stamp Program; Five of the seven participant vs. nonparticipant studies are based on secondary analyses of data collected in BENEFIT is the size of the food stamp benefit (zero national surveys, including the Nationwide Food for nonparticipants);

OTHER INC is the amount of other income available 27A classic example of the effect of earmarking is the difference in to the household; and behavior between a person who loses a $40 concert ticket and a person who loses $40 en route to buying a concert ticket. The loss of the ticket X is a vector of household characteristics. (earmarked) is much more likely to result in the person’s forgoing the concert than is the loss of the money. (This example is taken from Amos Tversky, a cognitive psychologist who studied human-choice behavior and Three main variations on this model have been used: the limits of the rational choice model.) Similarly, a recipient whose food Models may include FSP, BENEFIT, or both. Four of stamp benefit is cut by $40 is likely to curtail food expenditures more than the seven participant vs. nonparticipant studies esti- one whose cash assistance is curtailed by $40. 28Three of the studies in Group I also include dose-response estimates. mated models with FSP but not BENEFIT, and three These studies have not been double-counted as part of the 20. of these studies included both FSP and BENEFIT. Half

36 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program n n n e t 3 od a i h r food a tion rtion of Continued— rient n v t arisons a o s i gressio gressio gressio t u l p u /nu mp us food e o m te re te re te re pent o eo a a a d on lity eq e t on pro h Analysis met d nditur abi g i e Bivariate t-test Multivari expe avail income s regressi base W Multivari Multivari Bivariate c Simultan mmy; mmy; mmy; mmy mmy mmy mmy u u u u u u u nt on d on d on d on d on d on d on d Measure of participation efit amount efit amount us amou Participati Participati Participati Participati bon Participati ben Participati ben Participati t vs. t vs. t vs. t vs. t vs. t vs. t vs. pant; pant; pant; pant pant pant pant 4 n n n n n n n Design partici partici partici partici partici partici partici Participa non Participa non Participa non Participa non also dose- response Participa non also dose- response Participa non also dose- response Participa non sehold food expenditures ) s h h r e n 12 olds le le le le le le latio b b b b b b 9) lds lds lds lds lds wit lds wit ge 8- useh ly membe eho eho eho eho eho eho Popu ,900) ,254) ,562) ,454) 2 2 3 1 10,35 329) 332) (sample siz alysis of national surveys Program on hou FSP-eligi hous (n~ (n= elder FSP-eligi hous (n= child a (n= FSP-eligi FSP-eligi hous hous (n= FSP-eligi hous All ho (n= FSP-eligi hous (n= an ek 2 s e only only r per we , total eek month week week d d h u t i ld d ent ent n the Food Stamp ased food ased food ased food ased food ased food n per n per n per e eho f ed foo ed foo r mont r week ival ival p d d d d d e x e Measure of e e e e e e urch urch urch urch urch d d d d d u u u u u l l l l l c c c c c mparisons—State and local studies mparisons—Secondary n n n n n At-home, away At-home At-home Nonp i Per ENU per w At-home Nonp i At-home Nonp Per perso i Purchas At-home Nonp i Per perso Nonp Purchas i Per perso Per equ adult p Per hous Per equ adult p At-home o o t c 1 -LI -LI icipant c CES CES ly t n on State nparticipan e unty, CA m e l 2-73) 2-73) 3-74 BLS- 9-80 NFCS 7-78 NFCS 7-78 3-74 BLS- Data source p shingt p a u Kern Co 197 197 (197 197 NFCS elder s (197 197 W 197 of table. ) dies that examined the impact o 8 0) 8 9 at end 8) 1 8) ( d 83) 98 nd y IB: Participant vs. no IA: Participant vs. nonpart e e (198 n 3) a e (197 st et al. (1978) v e See notes e Table 8—Stu Study Group Hama a Chern (1 Kisker an D Basiotis et al. (198 Price (19 Salath Group Lan W

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 37 Chapter 3: Food Stamp Program e n n n n n n n n od ations h Continued— ent intak ction- i gressio gressio gressio gressio gressio gressio gressio gressio e us equ ip te re te re te re te re te re te re te re te re n sel eo a a a a a a a a del lity/nutr sh n Analysis met abi uatio Multivari Multivari Multivari Multivari Simultan for food cost/nutrient avail Multivari relatio Multivari 2-eq bias mo Multivari Multivari nued ly dummy; dummy; dummy; dummy; dummy; dummy; unt Conti e e e e e lue lue lue d week Measure of participation us valu us valu us valu us valu us valu efit amount Benefit amo Bonus va Bonus va Bonus va Participation Participation Participation Participation Participation Participation bon bon bon expecte bon bon ben se se se se n n n n onse onse onse onse onse onse o o o o p p p p p p p p p p ood expenditures— Design Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res s d l ) s o t t t t e n h n n n n C- e I l their s le le le le le u latio b b b b b o lds lds lds lds lds lds lds lds h ed al e l eho eho eho eho eho eho eho eho Popu ,473) ,000) ,809) ,210) ,582) ,407) ,850) b i 4 1 3 1 3 2 3 573) 790) 515) (sample siz g i l Program on household f al surveys FSP-eligi hous (n= hous (n= hous (n~ FSP participa FSP participa food stamps (n= who us FSP-eligi FSP-eligi FSP participa FSP- and W e hous (n= hous (n= FSP-eligi hous (n= hous (n= FSP participa (n= FSP-eligi hous (n= n ek ek ek Stamp 2 s ys ys food e only only only only only only 2 r a a per we per we per we per , total n , total week d d d d d d s u t i ld ld ld ld of food of food d l l ent iou n ased food n per the Food e eho eho eho eho f ed foo ed foo ed foo ed foo ed foo ed foo r week ival p last 7 d last 7 d d x e Measure of ry analysis of natio ditures o n n e e urch g prev d d recal d recal u l c n At-home, away At-home, total At home, away At-home, total Aide At-home At-home Per hous Purchas used i Purchas Purchas Purchas Purchas Per hous month used i Nonp Per hous durin i Per perso Per equ adult p months Purchas Per hous Aide At-home Expen 1 -LI -LI -LI -LI -LI -LI CES 9-80 NFCS 7-78 NFCS 9-91 CSFII 7-78 NFCS 3-74 BLS- 5 CSFII 7-78 NFSC 7-78 NFCS 7-78 NFCS 8 PSID Data source 197 197 198 197 197 197 198 197 197 197 of table. on se-response estimates—Seconda dies that examined the impact o c 1) s d n la 9) d 6) at end 3) Gad l (199 II A: Do ood an 198 uer an g (198 dah 0) 7) 3) st (1984) e See notes Table 8—Stu Study Group Kramer-LeB et al. (1997) Leve Fraker et al. (199 Devaney and Fraker (198 Basiotis et al. (198 Sena Youn Smallw Blaylock (1985) W Allen and (198 Chen (

38 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program 5 s n n n n n n e l d od b e h t a i a r l Continued— ment d ween a t e del, gressio gressio gressio gressio gressio gressio r v n regression c jus i u e an d h te re te re te re te re te re te re y l p a a a a a a on mo ons bet g a r n i g Analysis met o m us valu e m e e Multivari Multivari Multivari S Multivari Multivari bon d Multivari Dynamic a model regressi interacti Multivariate nued mmy u d unt unt Conti e n e e e e u o lue l i t a a v p Measure of i participation s c i u us valu us valu us valu us valu t r n a o Benefit amo Benefit amo Bonus va Participation dummy; Participation dummy; Participation dummy; Participation dummy; bon bon bon bon P B e e se se se s s n n n n n onse onse onse onse o o o o o p p p p p p p p p s s ood expenditures— e e r r - - Design e e s s o o Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res D D ) t t t t e ons ons n n n n n n o up up olds olds le le latio b b lds, lds lds lds lds lds lds n regi ng co ng co useh useh ncome eho eho eho eho eho eho eho Popu ,300) ,535) ,300) 3 4 3 896) 494) 911) 487) 123) 659) (sample siz Program on household f FSP-eligi hous (n= All ho (n~ hous receivi (n= Low-i hous (n= FSP participa All ho (n~ hous (n= hous receivi (n= FSP participa FSP participa hous (n= FSP participa FSP-eligi hous souther (n= ek ek Stamp 2 s ys e only only only only tures for tures for me me r a per we per we per i i udies month month d d d d o o u d d t i ld ld ld n n of food d e e l n at h at h ased food ased food n per n per the Food e eho eho eho d d f ed foo ed foo ed foo ed foo p last 7 d d d x Measure of n e e e urch urch al exp al exp d d d recal u u l l c c n n Annu Annu At-home At-home Per hous food use Purchas food use At-home Purchas At-home Nonp At-home Purchas i used i At-home Purchas Nonp Per hous i Per perso Per hous month Per perso Aide 1 n -LI -LI hout hout ana, CES i d ups i unty, FL s an stration stration ego cas ego cas nia, Ind e n n 0) 9-89) 6) 0) 8-72 PSID 8-72 PSID 7-78 NFCS 7-78 NFCS 2-73 BLS- Data source 197 196 demo (199 Counti county gro Califor Ohio, Virginia (197 Polk Co (197 197 197 196 San Di demo (199 San Di of table. dies that examined the impact o 7) se-response estimates—State and local st 5) 2) 76) at end d nd nd l (199 n (198 IIB: Do g et al. on et al. ey an an a iro (19 g (198 dah 2) 1) 6) 1) See notes Table 8—Stu Study Brown et al. (198 Chavas and Yeun Johns (198 Benus et al. (197 Hymans a Shap Group Breuni (200 Leve Rann Kushma Neen Davis (1977)

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 39 Chapter 3: Food Stamp Program n n n n n od h Continued— gressio gressio gressio gressio gressio te re te re te re te re te re a a a a a Analysis met Multivari Multivari Multivari Multivari Multivari nued ship ship ship ship t t t t er er er er efi efi efi efi Conti lue Measure of nt nt nt nt participation Bonus va Group memb dummy; ben amou Group memb dummy; ben amou Group memb Group memb dummy; ben amou dummy; ben amou se n o p on on ent of ent of ison of ison of ood expenditures— ants to ants to s s e e om om Design Rand assignm particip check or coupon Rand assignm particip check or coupon Dose-res Compar treatment and matched comparis counti Compar treatment and matched comparis counti 6 7 12 ) s s s h - t t t t e i n o n n n n h o w s s latio ges 8 d d l l o o entati h h d after FIP en a e e Popu ,386) ,143) ,371) ie s s 2 1 1 995) 780) (sample siz u u o o Program on household f FSP participa (n= (n= FSP participa H H (n= implem participating in AFDC and w appl (n= (n= ASSETS and FSP participa childr Stamp 2 d s only ed food ed food e e only only only r ENU, ENU, ENU, d and s s , total , total , total , total s design d d d h h u month month month a a a t i ld, ld, ld, ld h d c ent ed r n ased food urch urch u the Food e eho eho eho eho f ed foo ed foo ed foo ed foo p r mont ival p lud d d d n e x Measure of e e e onp onp o e urch d d d n u u u l l l d c c c n n n n Per hous food inc and AME per Per hous and AME per Per hous and AME per adult p Per equ At-home, away Purchas At-home, away Purchas i At-home, away Purchas i At-home, away Purchas a AME per mont At-home Nonp i and n and n Per hous 1 hout out ETS ions—Experimental design ions—Nonexperimental on State on State trat trat s s stration stration stration stration ego cas n n n n 0) 0) 0) 0) 2-73) ama cash Data source shingt shingt a a W cashout demo (199 W Alab demo (199 demo (199 Alabama ASS demo (199 (197 San Di of table. 2) dies that examined the impact o shout demon shout demon 3) 3) at end d 76) IIIA: Ca IIIB: Ca n an g (199 2) st and rner (199 e e See notes Table 8—Stu Study W Price (19 Group Fraker et al. (199 Ohls et al. (199 Group Cohe Youn Davis and W

40 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program per od h sehold, ction- e ent only food n sel s dels , per hou Analysis met uatio rson bias mo 2-eq per pe easure(s) repre ured n r m s nued ship t patio er e mea r efi Conti ll food; whethe Measure of nt participation a r nditures a pe x Group memb dummy; partici dummy; ben amou er e h ) from home, o with ay out c.; whet d tures. ood expenditures— s, et Design 7 vs. 1984 pendi x umed aw s Pre-cash compare cashout (197 con n food, gift w o e unit for ) r ng e 95) n ility 3,9 le t and the tim latio pant ib b n lds usi g . s d at home, food eho Popu e 7 eli partici (sample siz ic NU); and Program on household f value of home-g 197 criteria (n= Participa FSP-eligi hous non (E consume Stamp upplement. Training Serv 2 urvey. s s to food timated dollar e s r and nutrition unit e u t i ld come S d n ment and ased food the Food e eho f p d quivalent x Measure of e e clude the urch viduals. Expenditure S - Low In d mploy y u l . . c y Indi riable correspond y cipant group. n y or per e s or in AME per week Per hous At-home, total Nonp i va cipant group. s b on rough E th y adult, coup cs’ Consumer 1 CS co 4 ption Surve ption Surve umption Surve namics. d Intake F 8 s y m m 9 u ficienc D u on s t to the e dependent s us. o H Foo in the nonparti d in the nonparti n gram. equivalent . od stamp r eno come fo eme r olated ce Pro 7 Puerto Ri Data source s to Self-Suf re isolate whether th Puerto Ric NFCS and 1 suppl 197 au of Labor Statisti ide Food Con nwide Food C ues ng Survey of old Food Con credit, o en eh SP as endog ons of s re not reported dependen F dies that examined the impact o we s: I = Natio eat r ce r d with cash, ible participants we ible participants not i NFCS = Nationw HFCS = Hou NFCS-L BLS-CES = Bure CSFII = Continui ASSETS = Av PSID = Panel Study of In out et al. ase 5) Data sou Includes indicati Does not t Elig Main effects Elig FIP = Family In 1 2 3 4 5 6 7 Table 8—Stu Study Beeb (198 purch adult male equivalent (AME), pe

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 41 Chapter 3: Food Stamp Program of the dose-response models included BENEFIT only, researchers using the 1968-72 PSID were limited to and half included both FSP and BENEFIT. annual expenditures for food used at home, which is not likely to be a very precise measure. When only FSP is included in the model, a direct estimate of the impact of the program is obtained from the Normalization of household food expenditures to value of b1, the coefficient on the participation account for household size and composition is usually dummy. When BENEFIT is included in the model, b2 done by standardizing food expenditure on a per capita is the MPSF out of food stamps while b3 is the MPSF basis, or by one of several alternatives that reflect rela- out of nonfood stamp income. In models with both tive nutritional needs of household members, including

FSP and BENEFIT, b1 represents the impact of the “equivalent adults” (EAs), counting additional family FSP on food expenditures that is independent of the members less heavily because of economies of scale; benefit level—for example, FSP nutrition education “adult male equivalents” (AMEs), counting family may have a fixed effect on food expenditures regard- members according to caloric requirements; or “equiv- less of the FSP benefit amount. Alternatively, b1 may alent nutrition units” (ENUs), counting family membe interpreted in these models as the selection effect, bers according to caloric requirements and percentage or the expected difference in expenditures absent the of meals eaten at home. FSP (or if FSP benefit levels were zero) between individuals with similar characteristics who do and do not Research Results choose to participate in the FSP. Some researchers The following sections summarize findings from excluded this term when including nonparticipants in research that examined the impact of the FSP on food their samples, risking a bias in the estimated MPSF if expenditures. The discussion addresses results, in turn, there is indeed a selection effect (Kramer-LeBlanc et for each of the three design/analysis approaches. al., 1997; Chavas and Yeung, 1982). Other researchers excluded nonparticipants altogether, analyzing only Participant vs. Nonparticipant Comparisons variations in benefit levels and dropping the FSP term Seven studies used participant vs. nonparticipant com- (Levedahl, 1995, 1991; Senauer and Young, 1986; parisons to directly estimate the impact of the FSP on Neenan and Davis, 1977). food expenditures. As expected, all of these studies Numerous variations on these model specifications are found that FSP participants spent more on food than did found in the literature. For example: nonparticipants (table 9). Although the studies were conceptually similar, they varied substantially in how • Household expenditures on food may be dollars spent they measured food expenditures. Some used money over a particular period or the monetary value of food spent on food for at-home use over the course of a week, consumed from household supplies during the period. while others used the monetary value of food consumed out of household supplies over a week or a month.29 • Household food expenditures may be normalized to Furthermore, some studies analyzed total household food account for the household’s size, age/sex composition, expenditures, while others normalized household food meals eaten away from home, and/or economies of expenditures to account for household composition. scale; or alternatively, household food expenditures that have not been normalized may be analyzed with The numerical estimates shown in table 9 are taken household size and composition included as covariates. directly from the cited studies and hence vary in their units. Some pertain to food expenditures per week, oth- • Other income may be subdivided to estimate the ers per month, and so on. To achieve some roughly com- separate effects of different income sources on food parable measure across studies, the last column in table expenditures. 9 shows the estimated impacts as a percentage of food expenditures. Depending on how the authors reported • The food stamp benefit and income may enter the sample characteristics, these values were calculated equation nonlinearly, for example, in quadratic or either as a percentage of sample mean food expenditure logarithmic form. or as a percentage of the “counterfactual”—the amount participants would have spent on food absent the FSP. The measure of food expenditures is often determined by the data. For example, researchers using national 29Authors analyzing national survey data did not have a choice in this survey data often do not have a choice because avail- regard. The studies conducted by Lane (1978) and West et al. (1978), how- able measures are limited. As shown in table 8, ever, were based on data collected specifically for this purpose.

42 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Hama and Chern (1988) estimated a simultaneous Although the potential for selection bias remains, the model of food expenditure, but treated FSP participa- strongest evidence in this group of studies comes from tion as exogenous. Price (1983) estimated a model the work done by Basiotis et al. (1983), Price (1983), based on nonparticipants and then compared predicted and Salathe (1980). Putting aside differences in values (evaluated at the mean values of participants’ methodology and measurement and assuming that an characteristics) with participants’ actual expenditures. FSP household contains, on average, two people, esti- Basiotis et al. (1983), Salathe (1980), and West et al. mates from these three studies suggest that FSP partic- (1978) simply used FSP participation dummies. ipation increases household food expenditures by $2- $4 per week. The absolute effect corresponds to 18-20 Four of the available studies cannot be generalized to percent of at-home food expenditures. the FSP population as a whole. Studies by West et al. (1978) and Hama and Chern (1988) used samples that Dose-Response Studies made up only part of the eligible population—house- Of the 23 of the 32 identified studies, 23 used dose- holds with children ages 8-12 and households with one response models to study the impact of FSP participation or more elderly members, respectively. In addition, on household food expenditures, including the 20 studies West et al. (1978) and Lane (1978) used samples that in Group II (table 8), as well as 3 studies from Group I were geographically restricted—to the State of (Price, 1983; Salathe, 1980; West et al., 1978) that used Washington and to a single county in California, both direct and dose-response estimates. The dose- respectively. Findings from the studies completed by response studies related food expenditures to the FSP Kisker and Devaney (1988) and Lane (1978) are limit- benefit amount, calculating the MPSF out of food stamps. ed because the authors did not estimate multivariate Table 10 shows the MPS from food stamps, as estimated models. F

Table 9—Findings from studies that examined the impact of the Food Stamp Program on household food expenditures using participant vs. nonparticipant comparisons Estimated impact As a share of food 1 Study Population Measure Absolute expenditures

Dollars Percent Hama and Households with 1 or Per capita at-home food 0.64 3.7 Chern (1988) more people 65+ expenditures per week Kisker and FSP-eligible households Money value of food used 2.49 10.8 Devaney (1988) at home per “equivalent nutrition unit” per week Basiotis et al. (1983) FSP-eligible households At home food cost per 3.70 20.4 household per week Price (1983) All households Expenditures for at-home 2.01 18.2 food per week per adult equivalent Salathe (1980) FSP-eligible households Per capita at home food At home: 1.45 18.8 purchases per week Total: .88 9.4 Lane (1978) FSP-eligible households At home food expenditures 3.26 10.9 + value of food in-kind, per person per month West et al. (1978) FSP-eligible households Value of food consumed at 5.14 13.0 with child ages 8-12 home per month per “equivalent adult” 1 These percentages were calculated relative to either the sample mean as reported by the author (Basiotis et al., $18.11; Hama and Chern, $17.48; Kisker and Devaney, $23.14), or the author’s estimated counterfactual value—that is, what participants would have spent on food if they did not receive food stamps or what nonparticipants actually did spend on food (Lane, $30.00; West et al., $39.63; Salathe, $7.71 and $9.28; Price, $11.03).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 43 Chapter 3: Food Stamp Program in these studies. This table relies heavily on table IV.1 in estimate is 0.47 and four other estimates are in the Fraker (1990), which summarized 17 studies. range of 0.42-0.45.

Fraker completed a careful analysis of the bulk of this Fraker goes on to explain why the two highest estimates research. He remarked that the estimates of the MPSF are so different from the others. One of the estimates, varied greatly in size, ranging from 0.17 at the low end obtained from a dynamic-adjustment model, represents to 0.64 and 0.86 at the high end.30 The two highest “the full long-run or steady-state responses of house- estimates are clearly outliers, since the third-highest holds to changes in food stamp (and other food subsidy) benefits.” The other estimate is based on an unstable 30The estimate of 0.64, which is from Hymans and Shapiro (1976), is not model that yields vastly different estimates for two included in table 10. Where Fraker’s table IV.1 gave multiple estimates from half-samples of the data. Both estimates rely on a the same study, table 10 includes only the most general estimate—in this measure of non-food stamp income that excludes wel- case, the estimate from the full sample and not those from two half-samples. The estimate of 0.69 shown in table 10 (Levedahl, 1991) was not included in fare and nonwelfare transfer payments but includes some the research reviewed by Fraker. imputed income elements, and both estimates mingle other FANP benefits (such as school lunches) with the Table 10—Findings from studies that examined the FSP benefit. Consequently, these two estimates can be impact of the Food Stamp Program on household discounted, leaving a set of estimates “roughly evenly food expenditures using dose-response analyses1 distributed over the range of 0.17 to 0.47, indicating that a $1.00 increase in the value of the food stamp benefit Estimated MPSF Study from food stamps of a typical recipient household would lead to additional food expenditures of between $0.17 and $0.47.” Breunig et al. (2001)2 0.40 2 Kramer-LeBlanc et al. (1997) .35 The studies listed in table 10 span the period before and Levedahl (1995)2 .26 Levedahl (1991)2 .69 after the elimination of the purchase requirement (EPR) Fraker et al. (1990) .29 in the FSP. Before the EPR, participants were required Devaney and Fraker (1989) Weighted:3 .42 to use the food stamps they paid for, as well as the bonus Unweighted: .21 stamps, to purchase food. After the EPR, only the bonus Basiotis et al. (1987) .17 2 amount was given in stamps. Fraker stated that estimates Ranney and Kushman (1987) .40 based on data collected before the EPR are likely to be Senauer and Young (1986) Pre-EPR:4 .30 4 Post-EPR: .26 biased upward, relative to the current MPSF, because Smallwood and Blaylock (1985) .23 the EPR should have led to many more participants West (1984) Participants: .17 being unconstrained in their food purchases—that is, Eligibles: .47 treating their food stamp allotment as cash. Their Allen and Gadson (1983) .30 4 MPSF should therefore be much lower, close to that of Chen (1983) Pre-EPR: .20 31 Post-EPR:4 .23 non-food stamp income. Yet, Fraker notes that “the Price (1983)2 .42 three estimates that are based on post-EPR data range Brown et al. (1982) .45 from 0.23 and 0.29 and are only slightly toward the Chavas and Yeung (1982) .37 low end of the distribution of all estimates.”32 Johnson et al. (1981) .17 Salathe (1980) .36 Four of the more recent post-EPR estimates that were West et al. (1978) .31 not available to Fraker (Breunig et al., 2001; Kramer- Neenan and Davis (1977) .45 Benus et al. (1976) .86 LeBlanc et al., 1997; Levedahl, 1995, 1991) do not Hymans and Shapiro (1976) .29 support the notion that the MPSF has declined since West and Price (1976) .305 1979. Their values are 0.40, 0.35, 0.26, and 0.69, 1 Adapted and expanded from Fraker (1990), table IV.1.The MPSF is respectively. A possible explanation for this apparent the fraction of each additional dollar of income that is spent on food. paradox is that the EPR substantially increased partici- 2These studies were not included in Fraker (1990). 3Using sample weights from the NFCS. pation, drawing households into the program that were 4EPR = Elimination of the purchase requirement. 5Fraker reports this value as 0.37, citing p. 729 of West and Price. 31 This appears to be an error on Fraker's part. The text there reads: “The Fraker also presents estimates of the MPSF out of non-food stamp marginal propensity to obtain food out of bonus stamp income (0.30) is income, which are not discussed here. They range from 0.05 to 0.24 and still below the average propensity of food stamp recipients to consume are invariably lower than the corresponding MPSF out of food stamps from out of all income (0.37).” The latter value is apparently the ratio of food the same study. expenditures to total income for food stamp recipients. Data reported in 32These estimates come from Chen (1983), Senauer and Young (1986), the article are not sufficient, however, to make this calculation directly. and Fraker et al. (1990).

44 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program not willing to spend as much on food as the purchase • Faulty accounting for the effects of household size requirement necessitated. These new participants and composition on food expenditures may lead to a might indeed be constrained in their food purchases, biased estimate. Blaylock (1991) estimated food even if the constraint was removed for those who expenditure elasticities of 0.778 when both food would have participated under the old system. expenditures and income were measured on a per household basis, 0.687 when both were measured on All of the estimates reported in table 10 are subject to a per capita basis, and 0.521 when food expenditures caveats. Most studies have criticized their predecessors were measured on a basis that accounted for and further criticism has been applied in review arti- economies of scale and income was measured on a cles. Among the issues affecting some or all of the per capita basis. Assuming that the last of the cited estimates are the following: estimates applies, the household-based estimates are too large by nearly 50 percent.34 • Early studies used data collected before 1975, when uniform national standards for food stamp eligibility The Levedahl (1991) estimate of 0.69 is so distant and benefits were implemented. from the others that it requires further comment. In a later article (1995), Levedahl stated: • Many studies used data that are not nationally representative samples of FSP eligibles—that is, that were The theoretical and empirical results presented in this paper restricted to a particular geographic area or demo- demonstrate that, except for the specification used by graphic subgroup. Senauer and Young, approximations used to estimate the food expenditure equation of food stamp recipients are mis- • The functional form of the relationship between food specified. ...Given the availability of this specification, it would be difficult to justify using a functional form that was stamps and food expenditures may be misspecified. not flexible when estimating the food expenditure equation (Levedahl (1991) reestimates the expenditures equa- of food stamp recipients. tion with three common functional forms plus the one he believes is correct and gets alternative values The Senauer and Young specification that Levedahl of the MPSF, ranging from 0.29 to 0.69.) was recommending is the double-log form, which gave Levedahl an MPSF out of food stamps of 0.29 in his • Many researchers identify constrained households as 1991 paper and 0.26 in the 1995 paper (using San those in which monthly food expenditures exceed Diego cashout demonstration data). One, therefore, their allotment by no more than a small margin and can reasonably conclude that the 0.69 estimate, based exclude these households from the analysis. No fur- on translog specification, is an outlier. ther mention is then made of the constrained households for which, indeed, the FSP increases food The Cashout Demonstrations expenditures markedly. Finally, findings from the five cashout studies (table 11) provide lower-bound estimates of the impact of the • If, as seems plausible, FSP households have a higher FSP. Included in this group are the following: MPSF out of non-food-stamp income than nonparticipant households, a model that includes both partici- • Two studies of “pure” cashout demonstrations in pants and nonparticipants and does not fully account Alabama and San Diego, in which the only differ- for selection bias will overestimate the MPS from F ence between groups was the form of the food stamp food stamps. benefit (cash vs. check).

• Sample weights may have been used improperly (or • Two studies of other cashout demonstrations— not at all). Devaney and Fraker (1989) found that Alabama Avenues to Self-Sufficiency Through using weights in the NFCS nearly doubled the esti- Employment and Training Services (ASSETS) mated MPS .33 F and Washington Family Independence Program

33In a comment on the Devaney and Fraker (1989) paper, Kott (1990) suggested that the effect of weights could be due to differences in the 34 The elasticity of food expenditures with respect to income, ηF, is the MPSF between low-income households that lived in high-poverty vs. low- percentage increase in food expenditures associated with a 1-percent poverty areas, which was a sample stratifier. The latter group was under- increase in income. If a household spends one third of its income on food, sampled, and if its MPS is substantially higher than that of the former F then its MPSF is equal to ηF X 1/3. Blaylock's analysis, based on the 1982 group, then a weighted estimate of the overall MPSF would be higher than CES, used total expenditures as the measure of income and did not break the unweighted version. out the effects of food stamps per se.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 45 Chapter 3: Food Stamp Program

(FIP)—in which other programmatic changes were are possible, however, as the Puerto Rico study used a made simultaneously. pre-/post-design, with a 7-year interval, and both the Alabama ASSETS and Washington State demonstra- • One study of the conversion in Puerto Rico from tions were based on matched treatment and compari- food stamps to the cashed-out Nutrition Assistance son counties. The pure cashout demonstrations in Program (NAP). Alabama and San Diego were, however, true experiments. An additional limitation of the cashout studies The impact of cashout may be interpreted as the effect is their limited generalizability. While many of the of one of the two components of food stamp benefits, studies discussed were based on national surveys, each namely the coupon format. The cashout effect is a cashout evaluation reports results from a single State. lower bound of the total impact of the FSP because it excludes the effect on food expenditures of giving The estimated impacts on expenditures per AME or households more money. Note that the cashout effects ENU per month for food used at home range from are expected to be negative: They represent the effect -$0.34 (Alabama “pure” cashout) to -$25 (Alabama of not providing benefits in coupon form. ASSETS).35 In percentage terms, the range is from -0.3 to -21.9 percent. It is generally acknowledged that The direct estimates of differences in food expenditures provide comparisons that are free of a major 35Estimate based on 4.3 weeks per month. Results are discussed on an potential source of selection bias: Both check and AME or ENU basis, so the Puerto Rico study can be included in the coupon recipients are FSP participants. Other biases comparison.

Table 11—Findings from studies that examined the impact of food stamp cashout on household food expenditures

Study/demonstration Fraker Ohls Beebout Cohen and Davis and et al. et al. et al. Young Werner (1992)/ (1992)/ (1985)/ (1993)/ (1993)/ Alabama San Diego Puerto Rico Washington Alabama (pure (pure (FSP Estimated impact (FIP) (ASSETS) cashout) cashout) conversion)

On purchased food used at home per household per month: Absolute (dollars) -28.08 -56.44 2.66 -22.25 Percent -12.1 -26.8 1.1 -7.5

On purchased food used at home per AME/ENU per month: -22.12 -25.43 -.34 -9.39 -2.95 Absolute (dollars) -17.2 -21.9 -.3 -6.9 -2.4 Percent

On total food expenditures per household per month: Absolute (dollars) -25.60 -54.47 2.16 -23.85 Percent -7.3 -23.6 .9 -7.3

On total food per AME/ENU per month: Absolute (dollars) -26.62 -23.62 -.99 -10.98 -1.00 Percent -13.4 -18.5 -.7 -7.3 -.5

On MPSF out of food stamp benefits .01 -.17 -.06 Notes: AME = Adult Male Equivalent. ASSETS = Avenue to Self-Sufficiency through Employment and Training Services. ENU = Equivalent Nutrition Unit. FIP = Family Independence Program. 46 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program the Puerto Rico conversion and the Alabama “pure” food energy and at least some nutrients. This seeming- cashout demonstration were not realistic tests of the ly obvious effect may not occur for several reasons, differences between checks and coupons. In Puerto however, particularly for nutrients that are in short Rico, food stamps were used as a second currency supply. Participating households may increase expen- even before the changeover, so they were, in a sense, ditures on food in ways that actually reduce the avail- already cashed out. In Alabama, the issues were that ability of some nutrients, for example, by choosing cashout was introduced with little publicity as a short- foods that are convenient or especially palatable, but term demonstration, and food assistance was issued as lower in nutrients.36 They may also purchase more a separate check that was not combined with AFDC. expensive forms of the same food, resulting in no net Hence, check recipients were still likely to treat their gain in nutrients. In addition, nonparticipants may food stamp benefits as earmarked for food. The San obtain more of their food from nonpaid sources, such Diego result, an impact of -$9.39 (-6.9 percent), seems as friends, relatives, soup kitchens, and food pantries the strongest, being unconfounded with other changes (Gleason et al., 2000). and based on an experimental design. Moreover, even if increased food expenditures lead to Four of the five studies reviewed also estimated increased nutrient availability, there is no guarantee impacts on total food expenditures. The estimated that this effect will be consistently positive. For exam- impacts were quite similar to those for food at home, ple, increased expenditures may lead to greater avail- indicating that offering food stamps as coupons rather ability of nutrients and food components that than cash reduces expenditures on food away from Americans consume to excess, including fats, choles- home only slightly, if at all. terol, sodium, and added sugars.

The authors of three of the cashout studies also estimat- Assessment of household nutrient availability is based ed the MPSF for food stamp checks vs. coupons. The on detailed records of household food use for an difference between the two estimates is again a lower- extended period, usually 1 week. Information on quan- bound estimate of the impact of the FSP. These differ- tities of food withdrawn from the household food sup- ences were quite small in Puerto Rico and the “pure” ply is translated into nutrient equivalents to represent cashout demonstration in Alabama, but an impact of the amount of food energy and nutrients available to 0.17 was found in San Diego. Because of its strong household members. Although household nutrient design, the San Diego study settles, in the affirmative, availability thus excludes the nutrient content of food the question of whether the FSP increases food expen- consumed away from home, it is still an important ditures more than would a cash grant. As an aside, the measure because the FSP is intended to have its bene-

MPSF for food stamp coupons, per se, was estimated ficial effects specifically through improving in-home as 0.28 in this study, typical of other estimates. food consumption.

The amount of energy and nutrients available is evalu- Household Nutrient Availability ated relative to the Recommended Dietary Allowances Most studies that examined nutrition-related impacts (RDAs) and the household’s size and composition. of the FSP, especially the more recent ones, focused on Household nutrient requirements are generally defined impacts on the dietary intake of individuals residing in based on AMEs, which take into consideration the FSP households. A smaller number of studies exam- number of individuals in the household and their dif- ined nutrient availability at the household level. These fering nutrient requirements based on age, gender, and two outcomes are logically sequential. The hypothesis pregnancy/lactation status, or ENUs, which further is that the FSP benefit leads to increased food spend- adjust for the number of meals each family member ing, which leads to increased household nutrient avail- eats at home and the number of meals served to guests. ability, which, in turn, leads to increased intake by Research Overview individual household members. This section focuses on the middle, or household, link in this chain. The literature search identified 14 studies that examined the impact of the FSP on household nutrient As discussed in the preceding section, FSP participa- availability (table 12). All but three of these studies tion definitely leads to an increase in food expendi- (Bishop et al., 2000; Devaney and Moffitt, 1991; tures. One would suppose that, by spending more on food, households would increase the availability of 36See, for example, Prato and Bagali (1976).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 47 Chapter 3: Food Stamp Program e ; n n n n n 2 od ations h tion Continued— rient ent intak i t arisons a s gressio gressio gressio gressio gressio u /nu mp as models us equ us food e o ip te re te re te re te re te re eo eo a a a a a lity/nutr lity eq n-bi sh n nts Analysis met nditur abi abi Multivari Bivariate t-test Multivari Multivari Multivari expe avail Simultan for food cost/nutrient avail Multivari relatio selectio Bivariate c Simultan and nutrie y mmy mmy mmy mmy mmy mmy u u u u u u erg dummy; dummy; unt e e on d on d on d on d on d on d Measure of participation food en us valu us valu f Benefit amo Participation Participation Participati Participati Participati Participati Participati bon bon Participati se n onse onse o t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant p p p n n n n n n availability o Design partici partici partici partici partici partici Participa non Participa non Participa Participa Participa Dose-res Dose-res non non non Dose-res Participa non ) s h r e n n o le, le le le le le le le latio b b b b b b b b lds lds lds lds lds lds lds lds wit n regi ncome ly membe eho eho eho eho eho eho eho eho Popu ,900) ,850) ,562) ,360) ,000) ,535) ,925) ,454) 2 3 3 1 3 4 2 1 329) Program on household (sample siz alysis of national surveys surveys FSP-eligi hous (n~ (n= FSP-eligi elder FSP-eligi hous (n= FSP-eligi hous (n= souther (n= FSP-eligi hous (n~ Low-i hous (n= FSP-eligi hous FSP-eligi FSP-eligi hous (n= hous (n= an onal ly ly ly ly ly nt d d diary me ehol ehol Food Stamp e o of food ys) ys) d supp d supp d supp d supp d supp a a ll ry amou for food use for food use for food use for food use for food use o l l l l l hol hol hol hol hol the ditur f method ed at h of hous of hous pen Data collection our reca d recal d recal d recal d recal d recal ry analysis of nati pact o mparisons—Local studies mparisons—Secondary m Food categ food use (7 d (7 days) (7 days) (7 days) (7 days) from house from house and ex from house from house 24-h consum Aide from house (7 days) food use (7 d Aide Aide Aide Aide Record Record o o ipant c 1 -LI -LI -LI -LI -LI -LI icipant c CES ly

t n amined the i npartic e unty, CA m ex e l t 2-73) Data source 9-80 NFCS 9-80 NFCS 7-78 NFCS 7-78 NFCS 2-73 BLS- 7-78 NFCS 7-78 NFCS 7-78 p p u Kern Co 197 197 NFCS elder s 197 197 197 (197 197 197 197 of table.

-response estimates—Seconda ) udies tha e t 8 8 9) 9 at end 8) 983) 1 8) nd ( d

98 nd y II: Dos IB: Participant vs. no IA: Participant vs. nonpart (197 e on et al. n n 3) 7) 1) a e (197 v See notes e Table 12—S Study Group Hama a Chern (1 Kisker an D Allen and Gadson (1 Basiotis et al. (198 Scearce a Jense Group Lan Group Devaney and Moffitt (1991) Basiotis et al. (198 Johns (198

48 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program n n n e od h inued t ction- gressio gressio gressio e on minanc C te re te re te re n sel a a a dels nts— Analysis met uatio 2-eq bias mo Stochastic do methods Multivari Multivari Multivari n and nutrie y ship ship ship ship ship t t t t patio erg er er er er er efi efi efi efi Measure of nt nt nt nt participation food en f Group memb dummy; partici dummy; ben amou Group memb dummy Group memb Group memb Group memb dummy; ben amou dummy; ben amou dummy; ben amou ) with on out d availability o ent of ent of ent of ison of ants to ants to ants to s e om om om Design 7 vs. 1984 Pre-cash compare cashout (197 Rand assignm particip check or coupon Rand assignm particip check or coupon Rand assignm particip check or coupon Compar treatment and matched comparis counti 3 ) s s ng t t e 95) P n o n n n h o

ility s 3,9 le t and latio am on household pant d ib b l n lds usi g o entati ants ants ego FS h d after FIP e eho Popu ,184) ,386) ,143) ama FSP ie s 7 eli partici 2 2 1 935) 780) (sample siz u Progr o 197 criteria (n= Participa FSP-eligi non hous particip (n= San Di Alab particip (n= (n= (n= implem FSP participa FSP participa H (n= participating in AFDC and w appl y. upplement. ud Food Stamp from from from from from design come S use use use use use the f d d d d d method Expenditure St Data collection - Low In y . pact o . y y m records and recall records and recall 7-day foo 7-day foo records and recall records and recall records and recall 7-day foo 7-day foo 7-day foo 1 cs’ Consumer CS co 4 ption Surve hout hout ption Surve umption Surve out out F 8 s m m ions—Experimental design ions—Nonexperimental 9 u u on s t to the s us. o H n gram. on State trat trat amined the i s s eno stration stration stration stration stration ego cas ego cas n n n n n ex eme t 0) and 0) 0) 0) 0) ama cash ama cash Data source ce Pro 7 Puerto Ri shingt a Alab demo (199 San Di demo (199 W suppl Puerto Ric cashout demo (199 Alab demo (199 demo (199 197 NFCS and 1 San Di au of Labor Statisti ide Food Con nwide Food C old Food Con eh SP as endog s dependen F 2) shout demon shout demon udies tha t s: I = Natio eat r 3) ce r d IIIA: Ca IIIB: Ca et al. p BLS-CES = Bure HFCS = Hou NFCS = Nationw NFCS-L n an out et al. g (199 0) 2) 5) Data sou Does not t FIP = Family In 1 2 3 Table 12—S Study Group Bisho (200 Fraker et al. (199 Ohls et al. (199 Group Cohe Youn Beeb (198

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 49 Chapter 3: Food Stamp Program

Scearce and Jensen, 1979) were included in the previ- design used in the San Diego study, in particular, ous section on impacts on food expenditures. Six of makes that study’s evidence particularly powerful the identified studies (Group I) used participant vs. when it indicates positive impacts. If one program nonparticipant comparisons. Five of these studies used component has a positive impact, then the program as national survey data, and one used local data. Group II a whole must have a positive impact. However, when includes three dose-response studies, all of which are no significant impact is detected, one cannot conclude based on secondary analysis of national survey data. that the overall program has no impact.

The studies in Groups I and II, most of which are With the exception of the cashout studies, all of the described in Fraker’s (1990) excellent review, employed studies that examined the impact of the FSP on house- a variety of modeling approaches. Some used structural hold nutrient availability are based on data that were models that estimated the FSP effect on expenditures collected between the early 1970s and 1980. Applying and then the effect of expenditures on nutrient availabili- findings from these studies to today’s FSP population ty. Other researchers estimated reduced-form models, must be done with some caution. treating nutrient availability as a function of FSP benefits without regard to any intermediate mechanisms. Although the same general caution can be raised about research on food expenditures, a compelling argument Group III includes the four cashout demonstrations can be made that impacts on nutrition-related out- described previously, as well as a more recent study comes are more sensitive to temporal considerations that involved secondary analysis of data from the than impacts on food expenditures. For example, the Alabama and San Diego demonstrations.37 As American food supply has changed dramatically in the described in the preceding section, two of the cashout past 20-25 years, with important implications for both studies used random assignment (Fraker et al., 1992; nutrient availability and individual dietary intake. Ohls et al., 1992), one used matched treatment and Americans are eating substantially more grains than control groups (Cohen and Young, 1993), and one they were two decades ago, particularly refined grains, used a pre-/ post-design to compare households in as well as record-high amounts of caloric sweeteners Puerto Rico before and after the FSP was cashed out and some dairy products, and near-record amounts of (Beebout et al., 1985). Of the two randomized experi- added fats (Putnam and Gerrior, 1999). ments, the San Diego study (Ohls et al., 1992) is generally considered to be the strongest because it did not In addition to myriad new products in the market and suffer from implementation problems encountered in changes in food enrichment policies and standards the Alabama study (Fraker et al., 1992). over time, a number of sociodemographic trends may have influenced food-purchasing behaviors. These The estimation approach for the San Diego, Alabama, trends include, for example, a rise in the amount of and Washington cashout studies was to compare food eaten away from home, smaller households, more regression-adjusted mean nutrient availability for two-earner and single-parent households, an aging households in the treatment and control or comparison population, and increased ethnic and racial diversity groups. In the Puerto Rico cashout study, a structural (Putnam and Gerrior, 1999). modeling approach was used to estimate the effect of cashout on expenditures and then the effect of expen- The data on household nutrient availability are also ditures on nutrient availability (Beebout et al. 1985). subject to the limitations that affect much of the available research on nutrition-related impacts of FANPs, In interpreting findings from the cashout studies, one as discussed in chapter 2. In assessing impacts on should remember that these studies were designed to household nutrient availability, most researchers used measure only the effect of the form of the FSP benefit— the “more is better” approach that was the state of the food coupons or cash—rather than the full program art at the time. However, increased availability of ener- impact, including the dollar value of the benefit and gy or nutrients at the household level may or may not the form in which it was delivered. The randomized influence the likelihood that individual household members consume adequate diets. And, in the case of food energy, fat, cholesterol, and sodium, increased 37Excluded from this table is a recent study of food security and nutrient availability may not be a positive effect. (Only one availability by Cohen et al. (1999). The authors analyzed only variations in nutrient availability among participant households, so program impacts study examined impacts on the availability of fat, and could not be estimated. none looked at availability of cholesterol or sodium.)

50 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Finally, two features of data on household nutrient comparisons. The authors included tests of selection- availability tend to impart a substantial amount of bias adjustment models and found that these had little measurement error to the estimates. First, the transla- effect on their results. tion of foods into nutrients is only an approximation. Second, the samples of data on foods withdrawn from Substantial weight is also given to significant findings stocks and used are small and subject to sampling vari- from the San Diego cashout study (Ohls et al., 1992). ability. These characteristics may obscure differences Nonsignificant findings from this study are not given between participant and nonparticipant households. the same weight because, as previously noted, the cashout studies assessed the impact of the form of the Research Results FSP benefit rather than of the overall program. Thus, the absence of a significant effect in the cashout stud- Table 13 summarizes findings of studies that examined ies does not provide convincing evidence than an the impact of the FSP on household nutrient availabili- effect does not exist. ty. The table focuses on the question of whether the FSP had any statistically significant impact on the Food Energy and Macronutrients availability of a given nutrient and does not present information on the estimated amount of the FSP Findings from the strongest available research suggest impact. Because one cannot assume that increased that FSP participation increases the amount of food food expenditures automatically translate into energy available at the household level. The San Diego increased availability of any particular nutrient, the cashout study found a significant effect of food stamp first and most important question is whether any sig- coupons on the availability of food energy, whether nificant effect exists. In addition, the variety of ways measured as mean percentage of the Recommended in which individual study authors analyzed and report- Energy Allowance (REA) or as the percentage of ed nutrient impacts makes finding a common metric households that had less than 100 percent of the REA for characterizing results difficult. for energy available in the household food supply (Ohls et al., 1992). Devaney and Moffitt (1991) report- Table 13 is divided into four sections: food energy and ed similar results. macronutrients, vitamins, minerals, and summary measures. The text follows this general organization, Overall findings for the availability of protein (in but discusses findings for vitamins and minerals in one absolute terms, not as a percentage of total food ener- section. gy) were quite similar. Both Devaney and Moffitt (1991) and Ohls et al. (1992) found that the FSP sig- In the interest of providing a comprehensive picture of nificantly increased protein availability. Three of the the body of research, both significant and nonsignifi- four other studies that assessed protein availability cant results are reported in table 13 and in all other reported similar results. The only exception was the “findings” tables. As noted in chapter 1, a consistent Alabama cashout study in which implementation was pattern of nonsignificant findings may indicate a true weak (Fraker et al., 1992). underlying effect, even though no single study’s results would be interpreted in that way. Readers are cau- Allen and Gadson (1983) conducted the only study to tioned, however, to avoid the practice of “vote count- examine availability of carbohydrates and fat, and they ing,” or adding up all the studies with particular did so in absolute terms rather than as a percentage of results. Because of differences in research design and total food energy. They found that the FSP significant- other considerations, findings from some studies merit ly increased the availability of both nutrients at the more consideration than others. The text discusses household level. methodological limitations and emphasizes findings from the strongest studies. In this case, the greatest Given the age of most of the available studies, the weight is given to the study by Devaney and Moffitt paucity of information about the impact of the FSP on (1991) (shown in the table, as with all the studies, by the relative availability of carbohydrates and fat is not primary author’s name (Devaney, 1991). This is the surprising. Until the 1990s, almost all empirical only non-cashout study that is based on data collected research on FANPs focused on nutritional adequacy. after the elimination of the purchase requirement. In Since that time, studies have begun to focus on nutri- addition, the study used a dose-response model to tional concerns related to overconsumption of fat, sat- assess FSP impacts, an approach less prone to problems urated, fat, cholesterol, and sodium, and/or on food of selection bias than participant vs. nonparticipant consumption patterns (for example, consumption of

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 51 Chapter 3: Food Stamp Program e l b a l i a v ct a

a Continued— utrients ant imp ergy/n Signific Less en e ] l b ila d availability of l ego; CO utrients ava s 0) [San Di ergy/n (200 usehold p t impact Less en All ho Bisho n ifica e p Program on househo l b m ila No sign l; P-N] l; P-N] l; P-N] ava utrients [nationa s s s ) ) ) [nationa ) [nationa 5 5 8 8 2) [1 State; CO] 2) [1 State; CO] 2) [1 State; CO] 9 9 979 979 93) [1 State; CO] 983) of the Food Sta 1 1 ergy/n ( (

t t u u usehold usehold usehold n (19 o o b b e e [Puerto Rico; CO] [Puerto Rico; CO] e e More en All ho Cohe All ho Bishop (2000) [Alabama; CO] Fraker (199 B All ho Fraker (199 Fraker (199 Ohls (1992) [1 city; CO] B Scearce (1 Basiotis (1 Scearce (1 e impact e ] ] ] ] l R R R b - - - ila D D D

; ; ; l l l l; D-R l; P-N] l; D-R] a a a ct examined th n n n a t o o o l; P-N] l; P-N] l; P-N] l; P-N] l; P-N] i i i t t t a a a a a nal; P-N] a a a n n n tha [ [ [

s [nationa ) ) ) [nationa ) s s s s s utrients ava ) [nationa ant imp natio 1 1 1 1 9 9 9 8 9 9 9 979 93) [1 State; CO] 93) [1 State; CO] 983) 1 1 1 3) [nation 3) [nation 3) [nation 3) [nation 3) [nation 88) [ ( ( (

ergy/n studie y y y 98 98 98 98 98 Signific e e e usehold usehold usehold usehold usehold on (19 n (19 n (19 n n n a a a v v v e e e More en Allen (1 All ho Cohe Ohls (1992) [1 city; CO] D Cohe Ohls (1992) [1 city; CO] D Allen (1 D All ho All ho All ho All ho Basiotis (1 Allen (1 Scearce (1 Allen (1 Allen (1 Johns Elderly Hama (19 of table. and nutrients s Findings from e t at end a r ergy d ergy y h energy and macronutrients o b r See notes a Table 13— food en Outcome Food Food en Protein C Fat Vitamins Vitamin A

52 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program e l b a l i a v ct a

a Continued— utrients ant imp ergy/n Signific Less en e l b ila l; P-N] d availability of l utrients ava [nationa s s s 2) [1 State; CO] 2) [1 State; CO] 983) ergy/n usehold usehold usehold t impact Less en All ho Fraker (199 All ho Basiotis (1 All ho Fraker (199 n ifica e p Program on househo l ] b m ila No sign l; P-N] l; P-N] l; P-N] ava ego; CO utrients s s s s s s ) ) ) [nationa ) [nationa ) [nationa 5 5 8 8 0) [San Di 2) [1 State; CO] 9 9 979 979 979 93) [1 State; CO] 93) [1 State; CO] of the Food Sta 1 1 ergy/n ( (

t t (200 u u usehold usehold usehold usehold usehold usehold p n (19 n (19 o o b b e e [Puerto Rico; CO] [Puerto Rico; CO] e e More en All ho B All ho Cohe All ho Scearce (1 Ohls (1992) [1 city; CO] All ho Bisho Cohe All ho Scearce (1 All ho Ohls (1992) [1 city; CO] Fraker (199 Scearce (1 Ohls (1992) [1 city; CO] B e impact e ] ] ] l ] R R R b R - - - - D D D ila

D ; ; ;

l l l ; l; P-N] l l; P-N] l; P-N] a a a ct a d examined th n n n a t n e o o o l; P-N] l; P-N] l; P-N] l; P-N] l; P-N] l; P-N] i i i o t t t i a a a a a a nal; P-N] t a a a a n n n tha [ [ [ n

[ s ) ) ) ) [nationa [nationa s s s s s s s utrients ava ) [nationa natio ant imp 1 1 1 1 9 9 9 9 9 9 9 9 979 93) [1 State; CO] 983) 983) 1 1 1 1 3) [nation 3) [nation 3) [nation 3) [nation 3) [nation 3) [nation 88) [ ( ( ( (

ergy/n studie y y y y 98 98 98 98 98 98 Signific e e e e usehold usehold usehold usehold usehold usehold usehold n (19 n n n n a a a a v v v v e e e e More en Bishop (2000) [Alabama; CO] D Allen (1 D All ho All ho All ho All ho Cohe D All ho Bishop (2000) [Alabama; CO] All ho Allen (1 All ho D Allen (1 Elderly Hama (19 Allen (1 Allen (1 Basiotis (1 Basiotis (1 Allen (1 Scearce (1 of table. and nutrients—Continu Findings from at end 6 12 ergy vin a n See notes Table 13— food en Vitamin B Vitamin B Vitamin C Vitamin E Folate Niaci Ribofl Thiamin Outcome

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 53 Chapter 3: Food Stamp Program e l b a l i a v ct a

a Continued— utrients ant imp ergy/n Signific Less en e l b ila l; P-N] d availability of l utrients ava [nationa s s 2) [1 State; CO] 2) [1 State; CO] 983) ergy/n usehold usehold t impact Less en All ho Fraker (199 Basiotis (1 All ho Fraker (199 Ohls (1992) [1 city; CO] n ifica e p Program on househo l ] b m ila No sign l; D-R] l; D-R] l; P-N] ava ego; CO utrients [nationa [nationa [nationa s s s s s ) ) ) 5 5 5 8 8 8 0) [San Di 2) [1 State; CO] 9 9 9 987) 987) 983) of the Food Sta 1 1 1 ergy/n ( ( (

t t t (200 u u u usehold usehold usehold usehold usehold p o o o b b b e e e [Puerto Rico; CO] [Puerto Rico; CO] [Puerto Rico; CO] e e e More en All ho Fraker (199 Ohls (1992) [1 city; CO] All ho Ohls (1992) [1 city; CO] B All ho B All ho Bisho All ho Basiotis (1 B Basiotis (1 Basiotis (1 e impact e ] ] ] ] l R R R R b - - - - D D D D ila

; ; ; ; l l l l l; P-N] l; P-N] a a a a ct d examined th n n n n a t e o o o o l; P-N] l; P-N] l; P-N] l; P-N] i i i i t t t t a a a a nal; P-N] nal; P-N] nal; P-N] a a a a n n n n tha [ [ [ [

s ) ) ) ) s s s s s utrients ava ) [nationa ) [nationa natio natio natio ant imp 1 1 1 1 9 9 9 9 9 9 9 9 979 979 93) [1 State; CO] 93) [1 State; CO] 93) [1 State; CO] 1 1 1 1 3) [nation 3) [nation 3) [nation 3) [nation 88) [ 88) [ 88) [ ( ( ( (

ergy/n studie y y y y 98 98 98 98 Signific e e e e usehold usehold usehold usehold usehold n (19 n (19 n (19 n n n n a a a a v v v v e e e e More en All ho Cohe D D D All ho All ho All ho Cohe D All ho Cohe Allen (1 Scearce (1 Allen (1 Elderly Hama (19 Allen (1 Elderly Hama (19 Allen (1 Scearce (1 Elderly Hama (19 of table. and nutrients—Continu Findings from at end ergy

horus m esium u i c l See notes a Table 13— food en C Iron Magn Phosp Zinc Outcome Minerals

54 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program y s, wer ses o s ct sh and a , and iron). een ca ch (P-N = ext discu adding up all the w ant imp Diego sample ts scored l n bet The t s pproa e final estimate is not significant findings ma ting,” or th Signific Participa thiamin, calcium nce of difference han others. abama and San he research a attern of non “vote coun tice of ideration t c istent p ] s s atistical significa min C, riboflavin, con State), and t nd no significant wer st ined for both Al l; D-R o re con m d availability of fou l one 1, a y r r oid the pra v exa nd the city o hapte tamin A, vita merit mo was go. [nationa d to a on. The ) s ts scored l 1 n 8 mates a in, vi i d in c t nd ir s studies phorus e s r cautione r San Die note usehold s on (19 some Participa t impact ers are All ho Johns d only fo n n nutrients (prote of paramete ample, national vs. one ifica research. A m, magnesium, a ard indicated. e ings from p Program on househo x ve y. Read tamin E and pho m i reporte m all. se and y of c wa , find s of v for n were y (for e er/sa No sign s he st rd/re gh i co y). ailability ted in that y re andard the stud of a combination n, thiamin, calciu a st nsideratio d s above t - stud cture of the bod co or the latter e of r s f RDA ted out ts scored h c o of the Food Sta hat wa scop u n shout r sive pi ca values t ll rather than 7 prehen , vitamin C, niaci s const ate, the ign and othe ailability t 12 st studies. s v Participa ca lts would be interpre d CO = e e impact wa addition, while the av point estimate rient com re on d nge In a

. y, an d stro tamin B e i y’s resu -hour earch d d i s ] stud v o r se er p ed 24

the estimate of actual nut the publicati otein, v t s f providing r e gh o l; D-R s i sehold nutrient a s es in re n ct s from the spon or Alabama and d

y u c examined th single stud a t e e of p r cau e ratio nal; P-N] nal; P-N] name, stud se re w th hou s

tha finding s gh no se differen s e [nationa s ed only f t ) s s s ant imp a ailability 1 ts scored h 8 m i n t author’ av r s the sum of becau 88) [natio 88) [natio d d in the interest o en thou e seholds wi

phasize v studie y, D-R = do t per 1,000 calories). as Signific n not reported be ere report i usehold usehold usehold on (19 senio o porte ct, e amine p cluded

x s. Because of , Participa r e e re e r o rmer w v Johns Kisker (19 Kisker (19 All ho All ho All ho e s w e o r et al. (1987) oportion of hou H ratio (nutrient sults a u r 00) als 978) not in

. r derlying effe s nparticipant stud and nutrients—Continu re cular result s or the fo o t a 0 0 un n no e Findings from s f rient siotis e i et m ane (1 p d the p

i ergy y c 3 3 2 di Cell entries show the r e d r sse with parti ts ts a

n n 1 n % RDA f m hop et al. (20 gy and 1 gy and 1 o p m Modified diet score is defined Lowest nut Asse u Notes: Nonsignificant Data for L Data for Ba Bis 1 2 3 u o Table 13— food en Modifie score Minimum nutrient diet ratio 100+ ener nutrie 80+ % RDA for ener nutrie participant vs. indicate a true studies methodological limitations and em clear. c point estimate Outcome S

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 55 Chapter 3: Food Stamp Program fruits and vegetables and whole grains). All of this was limited to bivariate comparisons of participants research, however, has focused on the dietary intakes vs. nonparticipants, however, so the results must be of individual FSP participants rather than availability considered suggestive only. at the household level. Johnson et al. (1981) constructed two summary meas- Vitamins and Minerals ures. The first was a Modified Diet Score (MDS) that Evidence of an FSP effect on the availability of vita- aggregated individual RDA “scores” (percentage mins and minerals is weaker than it is for food energy RDA) for food energy and seven nutrients. Values for and protein. Some nutrients were not assessed by each nutrient were truncated at 1.2 to avoid the possi- Devaney and Moffitt (1991) or Ohls et al. (1992), and bility of large excesses in one nutrient compensating for the nutrients that were assessed in both studies, for shortages in another. The authors also assessed the significant results were divergent. Devaney and nutrient density of the foods used from the home food Moffitt reported several significant impacts, while supply (nutrients per 1,000 calories), using a measure Ohls et al. reported none. As noted, lack of a signifi- called the Minimum Nutrient Diet Ratio (MNDR). The cant effect in the cashout study (Ohls et al., 1992) is first measure showed a statistically significant positive not definitive evidence that an FSP effect does not effect in their dose-response analysis, but the effect for exist. Therefore, findings from Devaney and Moffitt the second measure was nonsignificant. (1991) provide the strongest available evidence about Finally, Basiotis et al. (1987), also using a dose- the impact of the FSP on household availability of response approach, found a positive effect on house- vitamins and minerals. hold nutrient availability as measured by an index that was the first principal component of 11 individual Devaney and Moffit (1991) found that the FSP signifi- 38 cantly increased household availability of a broad nutrients. array of vitamins and minerals: vitamins A, B6, C, riboflavin, thiamin, calcium, iron, magnesium, and Individual Dietary Intake phosphorus. The authors estimated that the FSP increased the amount of these nutrients available to the The food eaten by individuals is primarily determined household by between about 20 and 40 percent of the by the food available in the households to which they RDA. The estimated MPS out of food stamp benefits belong. However, the relationship between nutrient was substantially higher than the MPS out of other availability at the household level and nutrient intake income—that is, a dollar of food stamp benefits had a at the individual level is weakened by several consid- greater impact on nutrient availability than a dollar of erations: cash income. • Household members may unequally consume nutri- Using participant vs. nonparticipant comparisons, ents from the food supplies, relative to their needs, Allen and Gadson (1983) estimated comparable effects depending on their tastes and appetites. across roughly the same range of nutrients, adding • Some household food supplies are consumed by vitamin B12 and niacin to the list. The remaining studies in all three groups found a mix of results. guests or are wasted.

Summary Measures • Some household members may consume food from other sources, including restaurants, school cafete- Three studies used composite indices to assess the over- rias, and other nonhome sources. all effect of the FSP on household nutrient availability. The results are inconclusive but generally consistent Moreover, increased availability of food energy and with the pattern of findings for individual nutrients. selected nutrients at the household level does not necessarily translate into better diets at the individual Kisker and Devaney (1988) examined the percentage level—for example, to lower intakes of dietary compo- of households whose at-home food use provided 100 nents overconsumed by many Americans (fat, saturat- percent of the REA as well as the RDAs for each of 10 ed fat, cholesterol, and sodium) or to healthier patterns nutrients. A comparable summary statistic was computed using a cutoff of 80 percent rather than 100 per- 38Because the estimate is constructed out of a combination of parameter cent. The authors report a favorable and significant estimates, the statistical significance of the final estimate is not clear and is FSP impact for both summary measures. The analysis therefore not reported in table 13.

56 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program of food intake (for example, eating more fruits and through 1979-80 (table 12). The same is true of 18 of vegetables or whole grains).39 For these reasons, one the 20 studies that used national survey data to investi- must examine the dietary intakes of individual house- gate impacts on food expenditures (data collection hold members to adequately assess nutrition-related periods from 1968-72 through 1979-80) (table 8), impacts of the FSP. although, as noted, temporal considerations are less important for this outcome. Research Overview In contrast, 11 of the 17 studies that used national sur- The literature search identified 26 relevant studies vey data to assess FSP impacts on individual dietary (table 14). Only four of these studies (Kramer-Le intake used data collected in the mid-1980s through Blanc et al., 1997; Fraker et al., 1990; Basiotis et al., the mid-1990s (data collection periods from 1985 1987; West et al., 1978) were included in the previous through 1994-96) (table 14). Indeed, the studies by sections on impacts on food expenditures and/or Dixon (2002) and Bhattacharya and Currie (2000), as household nutrient availability. Most of the identified well as those by Gleason et al. (2000) and Wilde et al. studies focused on impacts within subgroups of the (1999) used national survey data that were the most population, most often children or the elderly. recent available at the time the literature search was Sixteen of the identified studies used a participant vs. completed (NHANES-III for the Dixon and Bhattacharya and Currie studies and CSFII 1994-96 nonparticipant design (Group I). Of these, 10 involved 40 secondary analysis of data from national surveys. The for the study by Wilde et al.). other six participant vs. nonparticipant studies used In addition, research on the impacts of the FSP on State or local samples. Two of these studies used data dietary intake addresses, albeit to a limited extent, from the FNS/SSI Elderly Cashout Demonstration nutrition-related concerns that were not addressed in (1980-81), but not in the context of a cashout study, the research on household nutrient availability. These per se. The researchers who used these data (Posner et concerns include consumption of fat, saturated fat, al., 1987; Butler et al., 1985) combined data across cholesterol, fiber, and sodium, as well as dietary intake cash and coupon sites because no significant differ- patterns, or the extent to which food consumption ences were detected between the two groups. They behaviors conform with recommendations made in defined participants as those receiving FSP benefits, USDA’s Food Guide Pyramid. whether in the form of cash or coupons, and nonparticipants as individuals who were income-eligible but not Nonetheless, the majority of research on FSP impacts participating in the FSP. on dietary intake is subject to the limitations discussed in chapter 2. Ten studies used intake data for a single Ten studies used a dose-response approach to estimate day and therefore provide weak estimates of individu- FSP impacts (Group II). Seven of these studies used als’ usual dietary intake. Seventeen studies used multi- national survey data and the remaining three used ple days of data or food frequency instruments to bet- State or local data. None of the cashout studies (Group ter capture usual dietary intake behaviors; however, III in the preceding two sections) examined impacts on none used the approach to estimating usual intake that individual dietary intake. was recently recommended by the Institute of 41 The data used in studies that assessed impacts of the FSP Medicine (IOM, 2001). (Some studies used more on individual dietary intake are generally more recent than one method to assess dietary intake.) than the data used in studies of impacts on food expen- Similarly, in assessing intakes of food energy, vitamins, ditures and household nutrient availability. For exam- and minerals, researchers generally compared mean ple, all eight studies that used national survey data to intakes of participants and nonparticipants relative to estimate impacts of the FSP on household nutrient the RDAs, or compared the proportion of individuals availability used data collected mainly in the 1970s, in each group with intakes below a defined cutoff and with data collection periods ranging from 1972-73

40In June 2002 and February 2003, data files for NHANES-IV 1999- 39At the time most of these data were collected, the FSP offered little to 2000, including the first 2 years of data from the 6-year NHANES data no nutrition education to program participants to encourage such dietary collection cycle, were released by the National Center for Health Statistics. patterns. However, whether providing nutrition education would have led to 41Gleason et al. (2000) used these methods to describe dietary intakes of different results is not clear. For example, Gleason et al. (2000) demonstrated low-income populations. However, in assessing differences in the dietary that the dietary knowledge and attitudes of low-income individuals did not intakes of FSP participants and nonparticipants, they compared regression- mediate the relationship between FSP participation and dietary intake. adjusted mean intakes rather than usual intakes.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 57 Chapter 3: Food Stamp Program n n n n n n n e od ed tests d h Continued— ed) hoo alysis of gressio gressio gressio gressio gressio gressio quar minanc eli i-s h ate selectio te re te re te re te re te re te re te an a a a a a a a e ts not us vari i h Analysis met Multivari Multivari Multivari Multivari Multivari methods Bivariate and multivari model estimation (weig and b varianc Multivari Stochastic do Bivariate c Maximum lik racted mmy; mmy mmy mmy mmy mmy mmy mmy mmy mmy e dex ratio u u u u u u u u u u int n on d on d on d on d on d on d on d on d on d on d tio erty in a Measure of participation duals Participati particip with pov Participati Participati Participati Participati Participati Participati Participati Participati Participati t vs. t vs. t vs. t vs. t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant pant pant pant pant n n n n n n n n n n kes of indivi Design partici partici partici partici partici partici partici partici partici partici Participa non Participa non Participa non Participa Participa Participa non non non Participa Participa Participa non non non Participa non

n 5 e -50 ) - 16 r 4 09) and eir 9 e 5) d n 8 l 1 ,5 i 6 3 , h c 1 le le 0,54 % ed 88)

2 latio olds ges 1 nd th l es 20 = b b es 12- 1 o n als als als ( o 1,3

l-ag h y en (n= ncome ncome en a l useh idu idu idu c Popu ,774), ,358) ,901) ,566) ,590) r dren ages 1-5 dren ages 1-5 er 125 s 2 381) a 1 1 1 2 818) 800) (sample siz e Program on dietary inta e alysis of national surveys d r l Adults ag P Low-i Schoo childr (n= older (n= Youth ag (n= Low-i indiv (n= Elderly indiv (n= und (n= e Chil in ho of poverty Chil (n= FSP-eligi indiv (n= WIC-eligi women ages childr (n= and n= an ys ys ys ys e e Food Stamp v v a a a a

l ll ll ll ll ll ll lls lls l ll uti uti ntified a the c ency f ua e method r

r onsec onsec u ed by 2 d ed by 2 d ed by 2 d ed by 2 d onq Data collection our reca our reca our reca our reca our reca our reca our reca our reca our reca o h - pact o mparisons—Secondary 4 m 24-h 24-h 24-h follow 24-h follow follow 24-h food frequ and n 24-h follow of food records 24-h of food records of food records 24-h of food records 24-h 2 2 nonc 4 nonc o

I - S 1 ES-I E -LI icipant c N AN A H H 80 amined the i N

3 7 ex 976- - t 6 CSFII-LI 9-91 CSFII 9-91 CSFII 1-73 N 1 4-96 CSFII 7-78 NFCS 5 CSFII 8-94 8-94 Data source 7 9 198 198 198 1 NHANES-III 198 NHANES-III 199 197 198 197 and 1 NHANES-II 198 of table. , nd udies tha 95) 95) t a 8) 7a a at end 2) 000) 00 IA: Participant vs. nonpart t (198 et al. h p 9) 2) 0) ez and 7b) imer (199 ilde et al. e See notes Table 14—S Study Group Dixon (2 Bhattachary Currie (2 W (199 W Cook et al. (19 Rose et al. (19 Bisho (199 Fraker et al. (199 Gregorio and Marshall (1984) Lop Habic 198

58 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program n n n n

od d h e t Continued— s s and ance s s u bias gressio gressio gressio gressio j of d a on- - te re te re te re te re n a a a a ison o i is of vari uare test s que s Analysis met e r g e Multivari Compar Multivari Multivari Bivariate t-test Bivariate t-test analys means r Multivari with selecti techni Chi-sq tinued mmy mmy mmy mmy mmy mmy u u u u u u on dummy; C unt on d on d on d on d on d on d Measure of participation efit amount duals— Participation Participati Participati Participati Participati Benefit amo ben Participati Participati se n onse o t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant p p n n n n n n kes of indivi Design partici partici partici partici partici partici Participa non Participa Participa non Participa non non Dose-res Dose-res Participa non Participa non ed ) ) WIC z ars i e derly derly past n 6 n n 9 = es bsid 2% n latio 102) dre (

lds ad ted i e 99) als als 5 a - (n= 4 ng (8 ncome ncome ncome el ncome el eho

idu idu Popu ,684) ,935) ,379) ,900) 5 g in su dren ages 8 s men ag 1 3 1 1 (sample siz Program on dietary inta e o g Black chil W Chil months to 5 ye who wer participating in or who h particip year (n= female) (n=200) housi (n= Low-i livin 18-4 Elderly (n= Low-i indiv a (n= Low-i indiv Low-i hous (n= al surveys n amp e ys v Food St a y cy and ll uti ll ll lls ll ll ne ne enc uen the rd f aires aire o method pho pho onsec onn onn ry analysis of natio ed by 2 d Data collection our reca our reca our reca our reca our reca our reca pact o mparisons—State and local studies m 2 food freq questi 24-h Food frequ 24-h 24-h 24-h 4-day rec 2 nonc questi via tele via tele follow 24-h of food records 24-h o t c 1 cs in in not i eas of n n nter ates for amined the i nparticipan ectio l d) n family ex iatric cli ncome areas t 6-97) 9-91 CSFII 0-81 0-81 4-96 Data source low-income ar Hartford, CT (1999) 2 ped Low-i in Connecticut (199 1 urba practice ce Florida (d data col reporte 198 FNS SSI/ECD 198 FNS SSI/ECD 1 county in Mississippi (1971) 199 CSFII/DHKS 198 of table. udies tha se-response estimates—Seconda t et al. at end n et al. IIA: Do IB: Participant vs. no n 3) 8) 7) 5) 5) 0) 8) See notes Table 14—S Study Group Fey-Yensa (200 Perez-Escamilla et al. (2000) Perkin et al. (198 Posner et al. (198 Butler et al. (198 Futrell et al. (197 Group Gleaso (200 Basiotis et al. (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 59 Chapter 3: Food Stamp Program ip ; sh bias ty/ n n n i n od bil h

Continued— l aila t lection itching relatio del e gressio gressio gressio n e

s us on- m t u te re te re te re te te sw s o eo a a a a a on mo g mod u ated se n j e d Analysis met g a

ations for food o s d a n i Multivari Multivari nutrient intake equ cost/nutrient av Multivari Simultan regressi Multivari investig Multivari e switchin with selecti b tinued racted rity e u on dummy; dummy; dummy; int C unt unt e e e; e n tio a Measure of participation us valu us valu us valu us valu duals— Benefit amo Benefit amo Participation dummy; Participation Participation Participation bon bon particip bon with social sec income bon se se n n onse onse onse onse o o p p p p p p kes of indivi Design Dose-res Dose-res Dose-res Dose-res Dose-res Dose-res

s l ) ng a i e n u d d i n v le le i latio ers b b d n als i als als als als i ool

eastfee y ncome ncome l idu idu idu idu idu Popu ,000) ,542) ,615) ,315) ,093) r 3 1 5 1 1 499) 793) (sample siz e Program on dietary inta d l Nonbr presch FSP-eligi Elderly indiv (n= Elderly indiv (n= Low-i e (n= indiv (n= FSP-eligi indiv (n= (n= Low-i indiv rural areas (n= amp ys ys ys ys ys Food St udies a a a a a ll ll ll ll ll ll on ne the f method pho -pers ed by 2 d ed by 2 d ed by 2 d ed by 2 d ed by 2 d Data collection n our reca our reca our reca our reca our reca our reca pact o m 24-h 24-h follow 24-h follow 24-h via tele and i follow follow of food records follow of food records of food records of food records of food records 24-h 24-h 1 ME -LI I ment ment le le p p 73 R amined the i ex 969- ly sup ly sup t 9-91 CSFII 9-91 CSFII 7-78 NFCS 7-78 NFCS 7-78 NFCS 0-81 Data source 198 198 elder 197 elder 198 FNS SSI/ECD and 1 197 197 of table. c 7) 5) udies tha se-response estimates—State and local st t n la at end d d IIB: Do 7) 6) 8a) See notes Table 14—S Study Rose et al. (199 Kramer-LeB et al. (1997) Akin et al. (198 Basiotis et al. (198 Akin et al. (198 Group Butler an Raymon (199

60 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program n n od h gressio gressio te re te re a a Analysis met Multivari Multivari tinued on C e lue Measure of participation us valu duals— Bonus va Participation dummy; bon se n onse o p p kes of indivi Design tration. Dose-res Dose-res s ) t Demon e n le latio b Cashou als y idu Popu dren ages 8-12 728) 195) (sample siz Program on dietary inta FSP-eligi (n= indiv (n= Chil amp upplement. rity Income/Elderl cu Food St ll come S the d f method entary Se ifie m viduals. Data collection - Low In our reca y . pact o y Indi y m Unspec 24-h e Supple s b c 1 trition Examination Survey. 8) ption Surve umption Surve u n Servi d Intake ge Survey. s ce Experiment. m u on s Foo on State amined the i Nutritio ex t 2-73) Data source shingt a Tulsa, OK (197 W (197 nal Health and N Health Knowled ide Food Con Food and nwide Food C ng Survey of come Maintenan stated. ) udies tha t 2 s: I = Natio 8 ce r CSFII = Continui DHKS = Diet and FNS SSI/ECD = NFCS = Nationw NFCS-L NHANES = Natio RIME = Rural In 8) st et al. itfield (19 Data sou Sample size not h e 1 2 Table 14—S Study W W (197

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 61 Chapter 3: Food Stamp Program used a “more is better” approach in interpreting findings. knowledge and attitudes, self-assessed general health None of the identified studies used the approach recently status, indicators of self-reported health problems, and recommended by the IOM, which calls for use of data indicators for exercise frequency, smoking status, and on usual intake in conjunction with defined Estimated use of vitamin and mineral supplements. Average Requirements (EARs) (IOM, 2001).42 The authors tested the robustness of their results by Consequently, the available research provides an imper- estimating effects separately for subgroups of the pop- fect picture of the substantive significance of observed ulation defined by age, gender, race/ethnicity, health differences in the dietary intakes of FSP participants and status, income, and (for adults) dietary attitudes. In nonparticipants. The available research provides infor- addition, they estimated several alternative models, mation on whether FSP participants consumed more or including a model that used a quadratic specification less energy and nutrients than nonparticipants. However, of FSP benefit amounts, a model that used a single this information cannot be used to draw conclusions binary variable to represent FSP participation, and about whether FSP participants were more or less like- quantile regression models that examined the effects of ly than nonparticipants to have adequate intakes. FSP participation on different parts of the nutrient intake distribution (5th, 10th, 25th, 50th, 75th, and 90th Finally, previous caveats about measurement error also percentiles). Results of all of these alternative analyses apply. The estimation of food and nutrient intake is an were qualitatively similar to results of the main analysis. elaborate process that is subject to significant measurement error. This error may make it difficult to detect dif- Food Energy and Macronutrients ferences between participant and nonparticipant groups. Seventeen different studies assessed the impact of the FSP on the intake of food energy in one or more sub- Research Results groups of the population. Only 2 of the 17 studies found Table 15 summarizes findings from available research a significant difference between FSP participants and on impacts of the FSP on dietary intake. Two studies nonparticipants (Fraker et al., 1990, for preschool chil- have been omitted from this tabulation because the dren; Butler and Raymond, 1996, for the elderly), and papers did not present detailed impact estimates (Akin the direction of the effect was not consistent. et al., 1987; Akin et al., 1985). A similar pattern was noted for protein. Seventeen dif- Overall, the literature strongly suggests that the FSP ferent studies assessed the impact of FSP participation has little to no impact on individuals’ dietary intake. In on protein intake. Only four studies (Fraker et al., the discussion that follows, no single study is empha- 1990, for preschool children; Bishop et al., 1992, for sized because of the general consistency of results across all FSP households; Butler and Raymond, 1996, for studies. Where results are inconsistent, findings from the elderly; Perkin et al., 1988 for women) reported a the study by Gleason et al. (2000), which examined significant FSP impact, and the direction of the effect impacts on preschool children, school-age children, was not consistent across studies.43 and adults, are given the most weight. This study is based on the most recent CSFII data and used a dose- Only a few studies looked at the impact of FSP partici- response approach. The authors elected not to estimate pation on the intake of carbohydrates, fat, or saturated selection-correction models because they believed that fat. None of these studies, which assessed intake based neither the CSFII nor the companion Diet and Health on contribution to total energy intake rather than in Knowledge Survey (DHKS), which was also used in absolute terms, reported significant differences in the analysis, included good candidates for identifying mean intakes of FSP participants and nonparticipants. variables. Instead, the authors included in their model Gleason et al. (2000) found, however, that preschool a wide variety of variables that may affect dietary FSP participants were significantly less likely than intake and/or may be correlated with FSP participation comparably aged nonparticipants to meet the Dietary or benefits. This included a number of variables not Guidelines recommendation of less than 10 percent of used in other research, including measures of dietary total energy from saturated fat.

42Gleason et al. (2000) used these methods to describe dietary intakes of 43Gleason et al. (2000) found no significant differences in mean intakes of low-income populations. However, in assessing differences in intakes of protein, expressed as a percentage of the RDA, for any of the three popula- FSP participants and nonparticipants, they compared regression-adjusted tions studied (preschool children, school-age children, and adults). For pre- percentages of individuals with intakes above specific RDA cutoffs rather school children, however, they found that FSP participants consumed signifi- than the percentage of individuals with usual intakes below the EAR. cantly less protein than nonparticipants, as a percentage of total energy intake.

62 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program less d ct a Continued— nsume ant imp ts co n Signific Participa Elderly Butler (1996) [6 sites; D-R] dividuals ] 2 es of in less d l; D-R l; P-N] a ntak nal; P-N] nsume site; P-N] [nationa [nation 1 State; D-R] ) 0 ts co e} 0) [natio n 1) [1 county; P-N] 7a) 988) [ l-ag (200 n n ez (198 st (1978) [1 Participa e {schoo t impact Wome Fraker (199 Perkin (1 Children Gleaso W Futrell (197 Elderly Lop n

] ifica p Program on dietary i N ] ] - 1 m P

; d e t l; P-N] No sign l; D-R l; D-R a l; P-N] t l; P-N] a S 0)

l; P-N] nal; P-N] 1 nsume a nal; D-R] [

) 3 [nationa 0 [nationa [nationa ts co ) [nation a (200 [6 sites; P-N] ation [nationa s ) ) 0 ll n 4 0 0 2 8) ( 2) [natio a) [natio 98 7a) more/same n of the Food Sta 987) a (200 (200 s 995) [n 998 n n n (199 usehold e p Participa Y ez (198 - imer (199 y e [2 sites; P-N] {preschool} e Posner (1 Lop W Gregorio (1 Rose (1 Children Gleaso Cook (1 Elderly F Butler (1985) [6 sites; P-N] Adults Gleaso Perez-Escami All ho Whitfield (1982) [1 city; D-R] Bisho e impact more d ct examined th a t nal; P-N] tha nsume s ant imp ts co 0) [natio n studie Signific Participa Children Fraker (199 of table. Findings from at end ergy energy and macronutrients See notes Table 15— Outcome Food Food en

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 63 Chapter 3: Food Stamp Program less d ct a Continued— nsume site; P-N] 1 ant imp ts co ontinued n C 988) [ Signific n Participa {Whites} Elderly Butler (1996) [6 sites; D-R] Wome Perkin (1 dividuals—

] 3 N ] ] - 2 P

P-N] ; es of in e less t e; t l; P-N] d l; D-R l; D-R a l; P-N] t l; P-N] a S ntak 0) Sta

1 [ nal; P-N]

) nsume 3 site; P-N] [nationa 0 [nationa [nationa 03) [1 [nation ) a (200 [nationa 1 State; D-R] ) ) 0 ll 4 0 0 2 8) ts co ( e} n 98 7a) 2) [natio n a 988) [ l-ag 00 (200 (200 s n n n n e Y ez (198 - st (1978) [1 imer (199 Participa y e e {schoo [2 sites; P-N] e t impact Children Gregorio (1 Elderly Fey-Yensan (20 W Elderly F Perez-Escami Adults Dixon (2 Adults Gleaso Wome Perkin (1 Children Gleaso W Lop n ifica p Program on dietary i ] ] 1 m d {Blacks} l; P-N] l; P-N] No sign l; D-R l; D-R l; P-N] a l; P-N] nsume nal; D-R] nal; P-N] a e Food Sta [nationa [nationa [nationa [nationa ts co ) [nation ) [6 sites; P-N] s ) ) ation n 4 4 0 0 a) [natio 0) [natio 98 98 7a) more/same 987) (200 (200 995 [n 998 n n n usehold Participa ez (198 {preschool} Children Gleaso Gregorio (1 Children Rose (1 Cook (1 Gregorio (1 Posner (1 Butler (1985) [6 sites; P-N] Adults Gleaso Elderly Lop Wome Fraker (199 Perkin (1988) [1 site; P-N] Rural Butler (1996) [2 sites; D-R] All ho Whitfield (1982) [1 city; D-R] e impact of th more d ct examined th a t nal; P-N] nal; P-N] tha nsume s s ant imp 2) [natio ts co 0) [natio n studie Signific (199 usehold p Participa All ho Bisho Children Fraker (199 of table. s Findings from e t at end a r d y h o b r See notes a Table 15— Outcome Protein C

64 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program less d ct a Continued— te; P-N] nsume si ant imp ts co ontinued n C s} Signific n (1988) [1 Participa {Black Wome Perkin dividuals— ] ] es of in less d l; D-R l; D-R ntak 0) nsume site; P-N] [nationa [nationa a (200 1 ) ) ll 0 0 4 ts co n 988) [ (200 (200 n n n Participa {preschool} {preschool} [2 sites; P-N] t impact Children Gleaso Children Gleaso Perez-Escami Wome Perkin (1 n ifica p Program on dietary i 5 5 ] ] ] ] m P-N] d e; t l; P-N] No sign l; D-R l; D-R l; D-R l; D-R l; D-R] l; D-R] l; P-N] Sta nsume nal; D-R] nal; D-R] site; P-N] e Food Sta [nationa [nationa [nationa [nationa [nationa 03) [1 ts co ) [nationa [nationa [nationa s s 1 ) ) ) ) n 4 0 0 0 0 8) e} e} a) [natio a) [natio 98 more/same 998) 998) 988) [ l-ag l-ag (200 (200 (200 (200 998 998 n n n n n usehold usehold Participa imer (199 e {schoo {schoo {Whites} Children Gleaso Children Gleaso Rose (1 Rose (1 Gregorio (1 Adults Gleaso Wome Perkin (1 Elderly Fey-Yensan (20 Adults Gleaso All ho Basiotis (1 W All ho Basiotis (1 e impact of th more d ct examined th a t tha nsume s ant imp ts co n studie Signific Participa of table. Findings from at end See notes Table 15— Outcome Fat Saturated fat

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 65 Chapter 3: Food Stamp Program less d ct a Continued— nal; P-N] nsume ant imp ts co 0) [natio ontinued n C Signific n Participa Children Whitfield (1982) [1 city; D-R] Wome Fraker (199 dividuals— ] ] ] ] ] ] es of in less d l; D-R l; D-R l; D-R l; D-R l; D-R l; D-R ntak 0) nal; P-N] nal; P-N] nsume site; P-N] [nationa [nationa [nationa [nationa [nationa [nationa a (200 1 ) ) ) ) ) ) ll 0 0 0 0 0 0 ts co 0) [natio n 1) [1 county; P-N] 2) [natio 988) [ 00 (200 (200 (200 (200 (200 (200 n n n n n n n Participa {preschool} [2 sites; P-N] t impact Children Gleaso Children Gleaso Children Gleaso Gleaso Fraker (199 Futrell (197 Adults Gleaso Adults Dixon (2 Perez-Escami Adults Gleaso Wome Perkin (1 n ifica p Program on dietary i 6 6 6 ] m d l; P-N] No sign l; D-R l; D-R] l; D-R] l; D-R] l; P-N] 0) l; P-N] l; P-N] nsume a a nal; D-R] nal; P-N] e Food Sta [nationa [nationa ts co ) [nationa [nationa [nationa a (200 [6 sites; P-N] ation ation [nationa s s s State; D-R] ) ll n 4 0 8) e} a) [natio 0) [natio 98 more/same 987) 987) 987) 987) l-ag (200 995) [n 995) [n 998 n n usehold usehold usehold Participa st (1978) [1 imer (199 e e [2 sites; P-N] {schoo Gregorio (1 Cook (1 Butler (1985) [6 sites; P-N] All ho Basiotis (1 W Elderly Posner (1 Children Gleaso Children Perez-Escami Wome Fraker (199 Elderly W Rose (1 All ho Basiotis (1 Cook (1 All ho Basiotis (1 e impact of th more d ct examined th a t 0) l; P-N] a nal; D-R] tha nsume s a (200 ation ant imp ll ts co a) [natio n studie 995) [n 998 Signific [2 sites; P-N] Participa Children Perez-Escami Children Cook (1 Children Rose (1 of table. Findings from at end 6 12 See notes Table 15— Outcome Vitamins Vitamin A Vitamin B Vitamin B

66 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program less d ct a Continued— nal; P-N] nsume s ant imp ts co 0) [natio ontinued n C Signific usehold Participa All ho Whitfield (1982) [1 city; D-R] Children Fraker (199 dividuals— ] ] es of in less l; P-N] d l; D-R l; D-R l; P-N] ntak 0) nal; P-N] nal; P-N] nal; P-N] nsume site; P-N] [nationa [nationa [nationa ) a (200 [nationa 1 ) ) ll 4 0 0 8) ts co 0) [natio n 98 2) [natio 2) [natio 988) [ 00 00 (200 (200 n n n imer (199 Participa e {preschool} [2 sites; P-N] {preschool} t impact Gleaso Adults Dixon (2 Children Perez-Escami Children Gleaso Gregorio (1 Elderly W Adults Dixon (2 Wome Fraker (199 Perkin (1 n ifica p Program on dietary i 6 ] ] ] m d l; P-N] No sign l; D-R l; D-R l; D-R l; D-R] l; P-N] l; P-N] l; P-N] nsume a a nal; D-R] nal; D-R] nal; P-N] nal; P-N] e Food Sta [nationa [nationa [nationa [nationa ts co ) [nationa [6 sites; P-N] ation ation [nationa s State; D-R] ) ) ) 1 n 4 0 0 0 8) e} a) [natio a) [natio 0) [natio 0) [natio 98 more/same 987) 987) l-ag (200 (200 (200 995) [n 995) [n 998 998 n n n n usehold Participa st (1978) [1 imer (199 e e {schoo {preschool} Wome Fraker (199 Elderly W Rose (1 Cook (1 Children Gleaso Fraker (199 Children Gleaso Rose (1 Cook (1 Gregorio (1 W Elderly Posner (1 Butler (1985) [6 sites; P-N] Adults Gleaso All ho Basiotis (1 e impact of th more d ct examined th a t tha nsume s ant imp ts co n studie Signific Participa of table. Findings from at end See notes Table 15— Outcome Vitamin C Vitamin E

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 67 Chapter 3: Food Stamp Program less d ct a Continued— nsume ant imp ts co ontinued n C Signific Participa Elderly Butler (1996) [6 sites; D-R] dividuals— ] ] ] ] es of in less d l; D-R l; D-R l; D-R l; D-R ntak nal; P-N] nal; P-N] nsume site; P-N] [nationa [nationa [nationa [nationa 1 ) ) ) ) 0 0 0 0 ts co 0) [natio n 2) [natio 988) [ 00 (200 (200 (200 (200 n n n n n n Participa {preschool} {Whites} t impact Children Gleaso Gleaso Children Gleaso Adults Dixon (2 Adults Gleaso Wome Fraker (199 Wome Perkin (1 n ifica p Program on dietary i 6 ] cks} m d No sign l; D-R l; D-R] l; P-N] 0) 0) l; P-N] nsume a nal; D-R] te; P-N] {Bla e Food Sta [nationa si ts co [nationa a (200 a (200 [6 sites; P-N] ation [nationa s State; D-R] ) ll ll n 0 8) a) [natio more/same 987) 987) (200 995) [n 998 n n usehold (1988) [1 Participa st (1978) [1 imer (199 e e [2 sites; P-N] [2 sites; P-N] Posner (1 Butler (1985) [6 sites; P-N] Adults Gleaso W Elderly W Wome Perkin Cook (1 All ho Basiotis (1 Children Perez-Escami Children Rose (1 Children Perez-Escami e impact of th ] more l; D-R d ct examined th a t 0) l; P-N] a nal; D-R] tha nsume s [nationa a (200 ation ) ant imp ll 0 e} ts co a) [natio n l-ag studie (200 995) [n 998 Signific n [2 sites; P-N] {schoo Participa Cook (1 Children Gleaso Children Rose (1 Perez-Escami of table. id Findings from at end nic ac n See notes Table 15— Outcome Folate Niaci Pantothe

68 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program less d ct a Continued— nsume site; P-N] 1 ant imp ts co ontinued n C 988) [ Signific n Participa {Whites} Wome Perkin (1 Elderly Butler (1996) [6 sites; D-R] dividuals— ] ] cks} es of in less d l; D-R l; D-R ntak nsume te; P-N] {Bla site; P-N] [nationa [nationa si 1 ) ) 0 0 ts co n 988) [ (200 (200 n n n n (1988) [1 Participa {preschool} {preschool} t impact Wome Perkin (1 Children Gleaso Children Gleaso Wome Perkin n ifica p Program on dietary i 6 6 ] ] ] ] m d No sign l; D-R l; D-R l; D-R l; D-R l; D-R] l; D-R] 0) 0) l; P-N] l; P-N] nsume a a nal; D-R] e Food Sta [nationa [nationa [nationa [nationa ts co [nationa [nationa a (200 a (200 [6 sites; P-N] [6 sites; P-N] ation ation s s State; D-R] State; D-R] ) ) ) ) ll ll n 0 0 0 0 e} e} a) [natio more/same 987) 987) 987) 987) l-ag l-ag (200 (200 (200 (200 995) [n 995) [n 998 n n n n usehold usehold Participa st (1978) [1 st (1978) [1 e e [2 sites; P-N] {schoo [2 sites; P-N] {schoo Adults Gleaso W Elderly Butler (1985) [6 sites; P-N] Posner (1 Children Gleaso All ho Basiotis (1 Butler (1985) [6 sites; P-N] Adults Gleaso Children Gleaso Cook (1 W Elderly Posner (1 Perez-Escami Rose (1 All ho Basiotis (1 Cook (1 Perez-Escami e impact of th more d ct examined th a t nal; D-R] tha nsume s ant imp ts co a) [natio 1) [1 county; P-N] n studie 998 Signific Participa Children Rose (1 Futrell (197 of table. Findings from at end vin a See notes Table 15— Outcome Ribofl Thiamin

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 69 Chapter 3: Food Stamp Program ] 2 less l; D-R d l; P-N] ct a a Continued— nal; P-N] nsume [nationa [nation ) ant imp 0 ts co ontinued 7a) n 2) [natio C 00 (200 Signific n ez (198 Participa {preschool} Elderly Butler (1996) [6 sites; D-R] Lop Children Gleaso Adults Dixon (2 dividuals— ] ] ] es of in less d l; D-R l; D-R l; D-R ntak 0) nal; P-N] nal; P-N] nal; P-N] nsume site; P-N] site; P-N] [nationa [nationa [nationa a (200 1 1 ) ) ) ll 0 0 0 ts co e} 0) [natio 0) [natio n 1) [1 county; P-N] 2) [natio 988) [ 988) [ l-ag 00 (200 (200 (200 n n n n n Participa {preschool} [2 sites; P-N] {schoo t impact Adults Dixon (2 Rural Butler (1996) [6 sites; D-R] Perkin (1 Perkin (1 Perez-Escami Futrell (197 Elderly Butler (1996) [6 sites; D-R] Whitfield (1982) [1 city; D-R] Wome Fraker (199 Wome Fraker (199 Children Gleaso Children Gleaso Adults Gleaso n ifica p Program on dietary i 6 6 ] ] m d l; P-N] l; P-N] No sign l; D-R l; D-R l; D-R] l; D-R] l; P-N] l; P-N] l; P-N] nsume a nal; D-R] nal; P-N] nal; P-N] e Food Sta [nationa [nationa [nationa [nationa ts co ) ) [nationa [nationa [6 sites; P-N] ation [nationa [nationa s s State; D-R] State; D-R] ) ) n 4 4 0 0 8) 8) e} a) [natio 0) [natio 0) [natio 98 98 more/same 987) 987) 987) l-ag (200 (200 995) [n 998 n n usehold usehold Participa st (1978) [1 st (1978) [1 imer (199 imer (199 e e e e {schoo Children Gregorio (1 Fraker (199 Cook (1 Posner (1 Butler (1985) [6 sites; P-N] All ho Basiotis (1 W Elderly W Children Gleaso Rose (1 Gregorio (1 Fraker (199 Adults Gleaso W Elderly W All ho Basiotis (1 e impact of th 1 6 more d l; P-N] ct examined th a a t 0) l; P-N] l; D-R] a a nal; D-R] tha nsume s [nation a (200 [6 sites; P-N] ation ation s ant imp ll ts co a) [natio 7a) n 987) studie 995) [n 995) [n 998 Signific usehold ez (198 [2 sites; P-N] Participa Rose (1 Rose (1 Elderly Lop Children Cook (1 Whitfield (1982) [1 city; D-R] All ho Children Perez-Escami Elderly Posner (1 Butler (1985) [6 sites; P-N] of table. Findings from at end

m u i c l See notes a Table 15— Outcome Minerals C Iron

70 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program less d ct a Continued— nsume ant imp ts co ontinued n C Signific Participa dividuals— ] ] ] ] ] es of in less d l; D-R l; D-R l; D-R l; D-R l; D-R l; P-N] l; P-N] ntak nal; P-N] nal; P-N] nal; P-N] nsume site; P-N] [nationa [nationa [nationa [nationa [nationa [nationa [nationa 1 State; D-R] ) ) ) ) ) 0 0 0 0 0 8) 8) ts co 0) [natio n 2) [natio 2) [natio 988) [ 00 00 (200 (200 (200 (200 (200 n n n n n n n st (1978) [1 imer (199 imer (199 Participa e e e {preschool} {preschool} {Whites} t impact Gleaso Adults Dixon (2 W Adults Gleaso Children Gleaso Children Gleaso Wome Fraker (199 Wome Perkin (1 Children Gleaso Elderly W Elderly W Adults Dixon (2 n ifica p Program on dietary i ks} 6 6 ] ] ] c m d {Bla No sign l; D-R l; D-R l; D-R l; D-R] l; D-R] l; P-N] 0) l; P-N] nsume a nal; D-R] nal; D-R] nal; P-N] te; P-N] e Food Sta [nationa [nationa [nationa si ts co [nationa [nationa a (200 ation [nationa s s ) ) ) ll n 0 0 0 8) e} e} a) [natio a) [natio 0) [natio more/same 987) 987) l-ag l-ag (200 (200 (200 995) [n 998 998 n n n n n usehold usehold (1988) [1 Participa imer (199 e [2 sites; P-N] {schoo {schoo Adults Gleaso Children Gleaso Children Gleaso Wome Fraker (199 Children Perez-Escami Wome Perkin Rose (1 All ho Basiotis (1 Rose (1 Elderly W Cook (1 All ho Basiotis (1 e impact of th more d ct examined th a t l; P-N] l; P-N] a a nal; D-R] nal; P-N] tha nsume s ation ation ant imp ts co a) [natio 0) [natio n studie 995) [n 995) [n 998 Signific Participa Children Cook (1 Children Rose (1 Cook (1 Fraker (199 of table. Findings from at end horus esium See notes Table 15— Outcome Magn Phosp Zinc

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 71 Chapter 3: Food Stamp Program 5 ] ] less l; D-R l; D-R d l; D-R] ct a Continued— nsume [nationa [nationa [nationa s ) ) ant imp 0 0 ts co ontinued n 998) C (200 (200 Signific n n usehold Participa {preschool} Adults Gleaso Children Gleaso All ho Basiotis (1 dividuals— ] ] ] ] ] 7 es of in less d l; D-R l; D-R l; D-R l; D-R l; D-R l; D-R] ntak l; P-N] l; P-N] 0) a a nal; P-N] nal; P-N] nsume [nationa [nationa [nationa [nationa [nationa ation ation [nationa a (200 [6 sites; P-N] s ) ) ) ) ) ll 0 0 0 0 0 ts co e} n 2) [natio 2) [natio 998) 987) 99) [n 99) [n l-ag 00 (200 (200 00 (200 (200 (200 n n n n n usehold Participa ilde (19 ilde (19 {schoo [2 sites; P-N] t impact Children Gleaso All individuals W Children Gleaso Children Gleaso Children Perez-Escami Adults Dixon (2 Elderly Posner (1 Gleaso All ho Basiotis (1 Adults Dixon (2 Children Gleaso All individuals W n

] ] ifica p Program on dietary i N N 5 5 ] ] ] ] - - m P P

; ; d e e t t No sign l; D-R l; D-R l; D-R l; D-R a a l; D-R] l; D-R] t t S S 0)

1 1 nsume nal; D-R] [ [

) ) 3 3 e Food Sta 0 0 [nationa [nationa [nationa [nationa ts co [nationa [nationa a (200 s s ) ) ) ) 0 0 ll n 0 0 0 0 2 2 ( ( a) [natio more/same n n 998) 998) a a (200 (200 (200 (200 s s 998 n n n n n n usehold usehold e e Participa Y Y - - y y [2 sites; P-N] e e All ho Basiotis (1 Adults Gleaso Children Rose (1 Adults Gleaso Children Gleaso Children Perez-Escami Elderly F Adults Gleaso Elderly F All ho Basiotis (1 e impact of th more d ct examined th a t tha nsume s ant imp ts co n studie Signific Participa of table. Findings from at end ducts esterol um See notes Table 15— Outcome Other dietary components Chol Fiber Sodi Food Intake Fruits and fruit juices Grain pro

72 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program ] less l; D-R d ct a Continued— nsume [nationa ) ant imp 0 ts co ontinued n C (200 Signific n Participa Adults Gleaso dividuals—

] ] N N ] ] ] ] - - P P

; ; es of in e e less t t d l; D-R l; D-R l; D-R l; D-R a a t t d meats} S S ntak 0) 0)

1 1 [ [ nal; P-N]

) ) nsume 3 3 starchy} 0 0 [nationa [nationa [nationa [nationa a (200 a (200 ) ) ) ) 0 0 ll ll 0 0 0 0 2 2 ts co ( ( e} e} n 2) [natio n n a a l-ag l-ag 00 (200 (200 (200 (200 s s n n n n n n e e Y Y - - Participa y y [2 sites; P-N] {fish an {schoo {preschool} {schoo [2 sites; P-N] { e e t impact Perez-Escami Children Gleaso Adults Dixon (2 Elderly F Elderly F Children Gleaso Children Gleaso Children Gleaso Perez-Escami n

] ifica p Program on dietary i N ] ] ] ] ] ] - m P

; d e t No sign l; D-R l; D-R l; D-R l; D-R l; D-R l; D-R a t S l; P-N] l; P-N] 0) 0) 0)

a a 1 nsume [ nal; P-N] nal; P-N]

) ll others} 3 a e Food Sta 0 [nationa [nationa [nationa [nationa [nationa [nationa ation ation ts co a (200 a (200 a (200 ) ) ) ) ) ) 0 ll ll ll n 0 0 0 0 0 0 2 ( e} more/same 2) [natio 2) [natio n 99) [n 99) [n a l-ag 00 00 (200 (200 (200 (200 (200 (200 s n n n n n n n e Participa Y - y ilde (19 ilde (19 [2 sites; P-N] {eggs} [2 sites; P-N] [2 sites; P-N] { {schoo {preschool} {preschool} e Adults Dixon (2 Gleaso Elderly F All individuals W Children Gleaso Adults Dixon (2 Gleaso All individuals W Children Perez-Escami Children Gleaso Perez-Escami Perez-Escami Children Gleaso Adults Gleaso e impact of th 5 5 more l; D-R] l; D-R] l; D-R] d ct examined th a t l; P-N] l; P-N] a a tha nsume s ation ation [nationa [nationa [nationa s s s ant imp ts co n 998) 998) 998) 99) [n 99) [n studie Signific usehold usehold usehold ilde (19 ilde (19 Participa All ho Basiotis (1 Basiotis (1 All ho All individuals W All individuals W All ho Basiotis (1 of table. Findings from rs at end a ucts les b d sug See notes Table 15— Outcome Meat, poultry, fish, and meat substitutes Milk and milk prod Vegeta Adde

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 73 Chapter 3: Food Stamp Program less wer d o ct ct a a Continued— nal; P-N] nsume ant imp ant imp ts scored l ts co n ontinued n 2) [natio C 00 Signific Signific Participa Participa Adults Dixon (2 dividuals— 8 8 ] ] ] ] ] P-N] P-N] l; P-N] es of in less wer e; e; t t d o l; D-R l; D-R l; D-R l; D-R l; D-R ona ntak Sta Sta 0) nsume 10 [nationa [nationa [nationa [nationa [nationa 03) [1 03) [1 ) ) ) ) ) 000) [nati ts scored l 0 0 0 0 0 (200 ts co n (2 a n a l; P-N] a (200 (200 (200 (200 (200 n n n n n Participa Participa [nation t impact t impact Adults Gleaso Gleaso Elderly Fey-Yensan (20 Adults Gleaso Children Bhattachary Children Bhattachary Adults Gleaso Elderly Fey-Yensan (20 Children Gleaso n n ifica ifica p Program on dietary i m d No sign No sign 0) 0) nsume ts scored e Food Sta n ts co a (200 a (200 er/same ll ll n more/same high Participa Participa [2 sites; P-N] [2 sites; P-N] Children Perez-Escami Children Perez-Escami e impact of th 9 t er u more gh i l; D-R] d tho ct ct i examined th a a t l; P-N] a tha nsume (1997) s ation c [nationa s ant imp ant imp n ts scored h adults w la n ts co n 998) l; D-R] 99) [n a studie Signific Signific usehold Participa ilde (19 [nation Participa Able-bodied dependents (ABAWDS) Kramer-LeB All individuals W All ho Basiotis (1 s of table. asure Findings from at end ting d me a y lic ages o d fats See notes Table 15— Outcome Adde Alcoh bever Sweets an desserts High-fat snack foods Outcome Summar Healthy E Index (HEI)

74 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program y rated wer ses o s s ct endation a r the elderly ed in the and satu . otein. s l r ext discu recomm not qualitatively adding up all the p ant imp cal (P-N = participant y ts scored l n d significance of ). ere om fat ch ontinued }. r re r The t odels u s C significant findings ma s “w ting,” or small sample size Signific cket pproa se Participa , and riboflavin fo calories f a 24-hou ch a t (p <0.001 r n rgy provided b {in bra ency. etary Guidelines han others. r nutrient e of three m d, direction an becau Di sea ality of the diet. fican attern of non he “vote coun u e age of ote re roup dividuals— her tha hors e ot signi s tice of subg age of ene ideration t c aut istent p ] s y s rwise n : percent ency rat ein, calcium, iron s es of in u rcent con e), and the wer o satisfied th ntified food frequ lts for the ard sitive and o re con pe l; D-R selection bias, on erred b he higher the q f t su r 1, a ntak r oid the pra nqua s po v 1 Stat re t r Unless othe no s pre a le and prot for the . ols fo r iduals wh score, s wa hapte ty o merit mo ight stand [nationa also identifies the d to a ) ci ts scored l he 0 n et e e} l samp antified food freq d in c enefit ased on of indiv entry hat cont n dummy studies p b l, which wa l-ag significant ower t (200 rura cautione onqu n note s age rs reported th ure b s s e cell some Participa The l stam rcent {schoo t impact ers are Children Gleaso n ly food how well they me d iron for the k ided. ata from a n nt for mea ifica research. A ample, national vs. 1 RDA, but wa v ings from r the pe bivariate model t FSP participatio p Program on dietary i x o of and the autho e y. Read subgroup, th ] ] r n d d legumes. s wee m sed on OLS mode y of wa , find , an s basi s n sed o y (for e e ba No sign l; D-R l; D-R r pecific s significa re not p sed on a s tage of th wa significant fo ted in that rients (protein an s on the oefficient for the stud tes we ts scored but ly to a ercen are ba ted. nsideratio del also included s were ba was e Food Sta n n p [nationa [nationa n e c t this goal. o ables, grain ) ) diet er/same r women a re cture of the bod e co e of 0 0 tima . et s r s (rural sample only), , but sco s cted nut y high umption veg o mee scop s sive pi r children (200 (200 oru orted fo n n Participa s were pre y significant s individuals’ <0.05), but th prehen ein intake as a ate, the ign and othe {preschool} otal energ ood con y and sele orted fo ndings pertain o ngs rep st studies. i re phosph (p s FSP benefits. M component t test f fruits and lts would be interpre Children Gleaso Adults Gleaso rg less likely t e e impact of th tailed impact e di y f of f ly rep com on d prot k nge s sed a ke o it sco del. stud se de stro or ene wee e atistical y’s resu earch d del. Fin s being r s becau I in that er E ts. but not statisticall and significant the publicati d he food-ba f providing a percentage of 90). Finding e gh effect of i y). Whe but no st ustment mo es in re e r rte s ct erred mo s from the sults only f articipant recall) measure c examined th a single stud t ef r sw gnificant for mean stud ngs e fo st (19 calcium, and inta r repo name, ble. SP p hour s t si negative negative se tha finding o gh no ction-adj differen s s ant imp spon with F ts scored h be the p n author’ compara r is similar to the H easure in which t ts, oils, and ng, and Po e was n d in the interest o en thou nificant findi r mean intake a r HEI (24- a al. (1998) a ate sele vitamin A, thiamin, C, and se re c phasize v ported. studie orted detailed re as ergy, sodium, and ors to d in (1987) not r f QI) Signific , Lo e k r senio vari NES-II data. NES-I data. porte (D e ct, e rol, ss rticipation was rticipation was ) rep e fo k r s. Because of Participa R = do e auth cally sig - siotis et e re r Fra at were re cipation w y, D t of total en s of NHA s of NHA rs to also asse ality Index n adapted HEI m e or FSP pa or FSP pa ond (1996 sults a ysi ysi not significant fo not significant fo d statisti tion of the bi d for Ba ne measu y derlying effe c t stud med by th P parti ein, choleste re cular result d sults th ercen Qu in (1985) and A ym as as k ) ref un y Findings from ed o r porte r anal r anal prot cient f cient f r FS s chool children, differen om re s r rticipan Cell entries show the (1990 s for A han 10 p with parti s and dee er 11 k Authors used a The Dieta nonpa Results fo Results fo For pre Difference w The coeffi Authors reporte Difference w Authors u The coeffi Notes: Nonsignificant Result Findings re Butler and Ra Fra 1 2 3 4 5 6 7 8 9 10 11 Table 15— Outcome Diet Quality Index vs. indicate a true studies methodological limitations and em coefficient fo sample). The stu different” f analysi compromised fun of less t fat, intake of

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 75 Chapter 3: Food Stamp Program

Vitamins and Minerals et al., 2000; Rose et al., 1998a), and zinc (Rose et al., 1998a; Cook et al., 1995). Rose and colleagues report- Few studies found that FSP participation was significant- ed the same result for iron in an earlier paper (1995); ly related to intake of vitamins and minerals. Moreover, that paper only looked at iron intake. in keeping with the results observed for energy and protein, the direction of the FSP effect was not consis- Findings from the more recent and methodologically tent across studies that did report significant results. rigorous study by Gleason et al. (2000) also raise doubts about FSP effects on preschool children. The The largest number of significant effects were reported only significant effect reported for preschool children by authors who focused on preschool children. Three in the Gleason et al. study was that FSP participants studies (Perez-Escamilla et al., 2000; Rose, et al., had a significantly lower intake of iron.44,45 1998a; Cook et al., 1995) reported that FSP participation increased children’s intakes of several vitamins Other Dietary Components and minerals. A handful of studies examined impacts of FSP partici- The Perez-Escamilla study, based on a small local pation on the intake of cholesterol, fiber, and/or sodi- sample, found that FSP participation was associated um. Gleason et al. (2000) found that FSP adults consumed significantly less dietary fiber than nonpartici- with increased energy-adjusted intakes of vitamin B6, folate, and iron. pant adults. Basiotis and his colleagues found that sodium intake was significantly higher in FSP house- Rose and his colleagues analyzed data from the 1989- holds than in nonparticipant households. 91 CSFII and found that FSP participation was associated with increased intakes of vitamin A, niacin, thi- Food Intake amin, iron, and zinc. The authors reported that they Six studies assessed the impact of FSP participation on investigated the possibility of selection bias in their food intake patterns on one or more population sub- results and found “no evidence” of it. No information groups. Findings from the available studies are mixed is provided, however, on how the issue was investigat- but provide little indication that the FSP has a positive ed and how the authors reached this conclusion. influence on food intake patterns. Using data from the most recent CSFII 1994-96, Gleason and his colleagues Cook et al. (1995) analyzed data from the 1986 CSFII (2000) found that receiving FSP benefits was associated low-income supplement and compared the percentage with significantly lower consumption of vegetables of FSP children and nonparticipating children with among adults and of grains among preschoolers. average intakes below 70 percent of the RDA. This study did not use multivariate techniques to control for Wilde and his colleagues (1999) used the same data as differences between the two groups; however, limita- Gleason et al. but estimated impacts for all individuals in tion of the sample to children in households under 125 FSP households rather than for specific subgroups. They percent of poverty provides at least some statistical found that FSP participation was associated with signifi- control. The authors reported significant and positive cantly greater consumption of meat (considered a benefi- FSP effects for a number of nutrients (vitamin B , 12 cial effect) as well as significantly greater intakes of folate, calcium, magnesium, and zinc). added sugar and added fat (not considered beneficial).

Confidence in the findings from these studies is dimin- Using data from an earlier wave of the CSFII (1989-91), ished by the small overlap in the significant effects Basiotis et al. (1998) found that the weekly value of FSP reported. All three studies examined intakes of vitamin benefits was significantly and positively related to con- A, vitamin B , vitamin C, folate, niacin, riboflavin, thi- 6 sumption of vegetables, milk and milk products, and amin, calcium, iron, and zinc. Of these, conclusions meat. Other studies that examined FSP impacts on intake about the impact of the FSP were consistent across all of specific types of food found no significant effects. three studies only for vitamin C and riboflavin. In both cases, the conclusion was that the FSP had no effect. For all of the other vitamins and minerals, one or two of the 44However, the percentage of FSP and non-FSP preschool children studies—but never all three—reported a significant FSP with iron intakes equivalent to 70 percent of the RDA was not significantly effect. The only nutrients for which there was any over- different. 45Gleason et al. (2000) reported a significant FSP effect for folate intake lap in significant effects were folate (Perez-Escamilla among school-age children, but intakes among preschool children were not et al., 2000; Cook et al., 1995), iron (Perez-Escamilla significantly different.

76 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Summary Measures although some of these studies included longitudinal as well as cross-sectional data. Several authors examined impact of the FSP on overall diet quality, using the Healthy Eating Index (HEI). The The following sections summarize findings for each HEI, developed by USDA’s Center for Nutrition outcome. Drawing firm conclusions about FSP impacts, Policy and Promotion (CNPP), is a summary measure with the possible exception of the impact on food of overall diet quality (Kennedy et al., 1995). The security, is not possible. The number of studies avail- index is comprised of 10 component scores that are able for any given outcome and population subgroup is weighted equally in the total score. Five of the compo- limited, and each study has important limitations. nent scores are food-based and evaluate food consumption compared with the Food Guide Pyramid rec- Food Security ommendations. Four component scores are nutrient- based and assess compliance with recommendations The relationship between FSP participation and food for maximum daily intake of fat, saturated fat, choles- security is a complex one. Food insecurity is likely to terol, and sodium. The 10th component score assesses lead households to seek food assistance, and receipt of the level of variety in the diet.46 Gleason et al. (2000) food stamp benefits may subsequently improve the also examined FSP impacts on an HEI-like summary household’s food security. This situation makes esti- measure known as the Dietary Quality Index (DQI). mates of FSP impacts on food security particularly vulnerable to problems of selection bias and reverse Findings from the available studies are mixed and, causality. giving precedence to the Gleason et al. (2000) study, provide little evidence that FSP participation influ- This difficulty is apparent in conflicting findings ences overall dietary quality. Dixon (2002) found that reported in the literature. Most participant vs. nonpar- HEI scores for FSP adults were significantly lower ticipant studies found that FSP participants were more than HEI scores for nonparticipant adults. Dixon did likely to be food insecure than nonparticipants. not limit her sample to low-income individuals, how- (Jensen, 2002; Cohen et al., 1999; Alaimo et al., 1998; ever, and her model controlled for relatively few meas- Hamilton et al., 1997; Cristofar and Basiotis, 1992; ured characteristics. Kisker and Devaney, 1988). On the other hand, Rose et al. (1998b), using a dose- Other Nutrition and response approach, found that food insufficiency was Health Outcomes inversely related to the size of the food stamp benefit and the relationship was stronger than the relationship The literature search identified a relatively limited num- between food insufficiency and other incomes. A com- ber of studies that investigated the impact of the FSP on parable pattern was reported by Cristofar and Basiotis other nutrition- and health-related outcomes. (Note that (1992) in a model that included all households. (Food studies that examined shopping patterns—such as, types stamp benefits did not have a significant effect in a of stores used and food expenditure shares—have been separate model that was limited to households with excluded from this review because of their tenuous preschool children). relationship to nutritional status.) Characteristics of these studies are summarized in table 16. Three of the cashout studies (Alabama “pure,” Alabama ASSETS, and San Diego) also considered Outcomes examined in this research include food food security. In the Alabama ASSETS demonstration, security (14 studies), birthweight (2 studies), weight members of the cashout group were significantly more and/or height (6 studies), nutritional biochemistries (3 likely to have skipped a meal due to lack of food or studies), and general measures of nutrition and/or money to buy food (Davis and Werner, 1993). health status (2 studies). (Some studies looked at multiple outcomes). The research on food security includes Two recent studies that used sophisticated techniques participant vs. nonparticipant, dose-response, and to control for selection bias help clarify the relation- cashout studies. Research on all of the other outcomes ship between FSP participation and food security. Both is limited to participant vs. nonparticipant comparisons, found that, once one controlled for selection bias, there was no evidence of significantly greater levels of food 46Results for component scores, when reported, have been summarized insecurity (or insufficiency) among FSP participants. in preceding sections of table 15. The analysis completed by Gundersen and Oliveira

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 77 Chapter 3: Food Stamp Program it n n b od ation ation h obits obits f sion ed pro s) s Continued— gressio gressio alysis o of an us equ us equ te re te re order egres th 2 pr th 3 pr eo eo i i a a isons ison uare ortions ortions Analysis met Bivariate t-test Multivari model model w prop prop model w (survey weight Multivari Chi-sq Bivariate Simultan Compar Compar Simultan Logistic r n ticipatio mmy; mmy mmy mmy mmy mmy mmy mmy mmy mmy u u u u u u u u u u par on d on d on d on d on d on d on d on d on d on d efit amount Measure of Participati Participati Participati Participati Participati Participati Participati Participati Participati Participati ben on and health outcomes triti t vs. t vs. t vs. t vs. t vs. t vs. t vs. t vs. t vs. ts vs. pant pant pant pant pant pant pant pant pant pant Design n n n n n n n n n n partici partici partici partici partici partici partici partici partici partici Participa non Participa non Participa Participa Participa non Participa non non non Participa non Participa non Participa non Participa non 5 e olds olds olds olds ) - ibl Program on other nu e ,900) mple ,300) w months 2 in past men 6 were useh useh useh useh o WIC or ~ d l n ages 1- o = sa

e n d households ho ho ho ate n Stamp a ildr

h 0) ) latio lds (n 99) 8 rticip 9 (sample siz a ncome (n ncome ho ncome ncome ncome ncome ncome w 3 eho , ,452) ,228) ,733) ,285) ,358) dren ages 8 Popu 3 3 3 5 1 3 21,81 = outh ages 12-16 n Low-i (n= (n= FSP and FSP-elig Chil to 5 years who participating in had p year (n= (n= hous ( Low-i (n= (n= income c (n= years (n=1,930) Low-i Y Low-i Low-i Low-i Low-i the Food f pact o 1

Hartford, l m a ES-III SIPP -LI cs in low- S i n 2 i 8 n SPD l d 9 AN u 9 H t 199 i 1

g enta d n n Data source o iatric cli l a

9-80 NFCS 5-86 CSFII-LI 1 and 7 0 April 6-97 NFSP 5 CPS nonparticipant comparisons D 9 P 9 1988-94 N 197 FSS-CPS income areas of CT (1999) 1 experim 199 2 ped 198 1988-94 NHANES-III S 200 199 199 7) of table.

) nd udies that examined the i 999) d t 8 a 1) 8 3) 2) a d d 9 at end 992) 1 00 ( d

et al. (199 000) on an y n (200 (200 e n n n et al. (1 n a v See notes e Table 16—S Study Food security: Participant vs. Huffman an Jense Jense Gunders Oliveria (2 Bhattachary Currie (2 Perez-Escamilla et al. (2000) Cohe Alaimo et al. (1998) Hamilto Cristofar an Basiotis (1 Kisker an D

78 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program ; ast n n n n n n n od h age le ed-effects t s and ance sion Continued— gressio gressio gressio gressio gressio gressio gressio te re te re te re te re te re te re te re te 2-s egres a a a a a a a a is of vari uare test es and fix Analysis met Multivari Multivari Multivari Multivari analys Multivari squar model fixed-effects models Logistic r Multivari Multivari Multivari Chi-sq inued t n

t n u o ticipatio mmy mmy mmy mmy mmy mmy ship ship ship m nefit nefit nefit u u u u u u a

er er er par r a l d be d be d be l on d on d on d on d on d on d o d

l nt nt nt a u n n Measure of Group memb Participati Participati Participati amou dummy an Participati Participati Group memb of food stamps amou dummy an Group memb amou dummy an Participati A on on and health outcomes—Con mparis ment ment treatment triti o n n e g g s n ed c o t vs. t vs. t vs. t vs. t vs. t vs. ants to check ants to check pant pant pant pant pant pant Design p n n n n n n ison of s on on s e e r om assi om assi - e partici partici partici partici partici partici s o Rand of particip or coup Participa Participa Participa non non non Participa non Participa non Rand of particip or coup Compar and match counti D Participa non s 2 e ing 0) 988 g liv ,620 from ,900) Program on other nu ) 6 4 e 200) mple ,371) ty men, 1 s s (n= 17 13,39 r poor, poor h incom ousin t t and 1

derly births o = sa e ) n n n n (n~ 3 79 0 le) (n= olds ed h 3 th 2 i , es 12- iz latio lds wit 0 -40 (n= d ants (n 3 0 (sample siz en 19 useh = ncome w ncome el eho g wome ,386) ,568) ,143) ,920) n dren ages 5-12

Popu 2 2 1 7 d 185% of pov n Infants born to Infants born to Low-i (n= FSP participa youn hous < (n~ women w betwe in subsi (82% fema All ho ASSETS and FSP particip Low-i (n= a FSP participa ages 2 Youth ag (n= Chil the Food Stamp f ent 0) 0) 0) 97) pact o 1 m t comparisons S hout 996- out pplem ETS areas in SY SY SY CD L L L d su l stration (199 stration (199 stration (199 ego cas Data source t vs. nonparticipant comparisons 992 SIPP n n n ecticut (1 ncome ama cash 9-91 CSFII 7 PSID- 9-87 N 5-96 N 9-88 N 7 nonparticipan 198 197 199 Alab 198 demo Conn and 1 demo Alabama ASS demo 197 NLSY-chi Low-i 199 San Di monstrations de Participan 3) 1) of table. rticipant vs. (199 2) a udies that examined the i height: 98b) e t 003) l d o 3) et al. (200 at end n d/or 003) 001) nd C 992) an rner (199 e See notes Table 16—S Study Food security: Dose-response estimates Rose et al. (19 Food security: Cashout Fraker et al. (1992) Ohls et al. (199 Davis and W Birthweight: P Korenm Miller (1 Currie a Weight an Fey-Yensa Gibson (2 Jones et al. (2 Gibson (2

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 79 Chapter 3: Food Stamp Program n n n n n od h OVA s and ance gressio gressio gressio gressio gressio te AN te re te re te re te re te re a a a a a a is of vari uare test Analysis met Multivari Multivari Multivari Multivari Multivari Multivari Chi-sq analys inued t n ticipatio mmy mmy mmy mmy mmy mmy mmy u u u u u u u par on d on d on d on d on d on d on d person-years. s t en s Measure of Participati Participati Participati Participati Participati Participati Participati repre in, on and health outcomes—Con e size min A, rum iron, triti a r. Sampl t vs. t vs. t vs. t vs. t vs. t vs. t vs. pant (iron, pant (iron) pant (album pant pant pant pant Design n n n n n n n bin, se rol, vit e lo g hereafte partici partici partici partici partici partici partici Participa non Participa Participa non hemo cholest Participa non vitamin C, E, carotenoids) vitamin C, E) non Participa Participa non non Participa non ing . der s. s g liv e Program on other nu ) ic and biannually t e 200) mple rticipant comparisons 94 17 16 16 NES-I) ousin and ol derly derly sa n . y le) (n= 88, ed h es 20 es 12- es 12- es 12- ,3 5) iz latio t Supplement. urve 1 d upplement. Training Serv (sample siz ncome el ncome el ,684, NHA ,920) ,358) ,358) ,598) ow-Income Sample dren ages 0-7 Popu 1 7 1 1 6 10,54 t comparisons n 1979 and 19 n come S in subsi (82% fema (n= Low-i Adults ag Low-i (n= and (n= NHANES-II) Youth ag (n= Youth ag (n= Youth ag (n= Chil (n= the Food Stamp ment and f hild Developmen viduals. ndividuals - L - Low In : Participant vs. nonpa mploy ent y I . y urrent Population S 97) in - C pact o 1 Indi ipation. annually betwee y m s b cs c rvey tatus . nonparticipa ES-II ES-I and 996- ke pplem ed s b 8 rough E AN AN th H H y 198 d su l ant vs collect trition Examination Survey. , cs. umption Surve ogram Su u health s gram Parti i d Intake . r s Data source me Dynami ement of the C Food Inta y ecticut (1 ncome areas m ficienc on co 6-80 N 1-73 N 7 NLSY 6 and of Foo erson y nd Pro on or p Particip Dyna Suppl 1988-94 NHANES-III 197 Conn Low-i 197 198 1988-94 NHANES-III 1988-94 NHANES-III NLSY-chi 199 y each ome a to Self-Suf 3) nal Health and N Inc opulation Surve Program nwide Food C al Food Stamp P ues ng Survey of nuing Surve d Securit nel Study of In

o of t of en y y h nt P hemistries: e c i c nd nd Conti udies that examined the i t b d s: a a a I = Natio et al. (200 a a ce H n r

001) d 000) 000) n 992) an a

ASSETS = Av FSS-CPS = Fo CPS = Curre CSFII = Continui CSFII-LI = NFCS-L NFSPS = Nation NHANES = Natio NLSY = National Longitudinal Survey of Youth. PSID-CDS = Pa SIPP = Surv SPD = Surve z 7b) e Data sou Multiple observations for p 1 2 o Table 16—S Study Bhattachary Currie (2 Korenm Miller (1 Nutritional bio Dixon (2002) Bhattachary Currie (2 L (198 General measures of nutriti Fey-Yensa Gibson (2

80 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

(2001) used data from the 1991 and 1992 SIPP panels limited statistical power) because the two births in the and used a simultaneous equation model with two pro- sibling pair had to differ with respect to the outcome bits. The analysis examined reported levels of food in order to be included in the model. insufficiency using the so-called “USDA food insufficiency question” that preceded the 18-item Federal Weight and/or Height food security module, the currently accepted standard Six of the identified studies assessed the impact of for measuring household and individual food security FSP participation on weight and/or height. Two studies (Price et al., 1997; Bickel et al., 2000). Huffman and examined linear growth and/or the prevalence of Jensen (2003) expanded on the work done by underweight among children. Five studies focused on Gundersen and Oliveira, incorporating information on the prevalence of overweight or obesity among chil- labor force participation decisions and using the more dren (1 study), adolescents (2 studies), adults (1 severe outcome of food insecurity with hunger based study), and the elderly (1 study). Gibson (2001) exam- on the 18-item Federal food security module. These ined the prevalence of both underweight and over- authors also simulated the effects of changes in FSP weight among adolescents. benefits, unemployment rate, and non-labor income and found that FSP benefits were more effective in Children and Adolescents reducing levels of food insecurity with hunger than Korenman and Miller (1992) used NLSY data to exam- pure cash transfers. Future efforts to understand the ine the prevalence of stunting (defined as height-for-age impact of FSP participation on food security may ben- below the 10th percentile on NCHS growth curves) efit from a longitudinal approach that measures and wasting (defined as weight-for-height below the changes for households over time. 10th percentile) among infants and children up to age Birthweight 7. The sample included children born between 1981 and 1987 who had height and weight measured in at Two of the identified studies examined the impact of least one of the NLSY Child Supplements (1986 or FSP participation on birthweight. Currie and Cole 1988).47 Models, which did not control for selection (1991) used data from the National Longitudinal bias, were estimated to look at both short-term and Survey of Youth (NLSY) to investigate effects on long-term effects of poverty and FSP participation. In infant birthweight of women’s participation in the FSP models that controlled only for current income and and other means-tested programs during pregnancy. In FSP receipt during the year preceding the measure- addition to standard multivariate regressions, the ment, no significant FSP effect was found. authors estimated fixed-effects models, looking at birthweights of sibling pairs. Using an instrumental In a model that controlled for long-term poverty variables approach to control for self-selection, they (measured by the average income-to-needs ratio of the found no significant effect of a mother’s FSP partici- mother over the 10-year NLSY time span), a modest pation on the likelihood that her infant would weigh at but significant effect on stunting was found, with FSP least 6 pounds. participants more likely to be stunted. The authors speculated that the positive relationship between stunt- Korenman and Miller (1992) completed an analysis ing and FSP receipt may reflect aspects of long-term that used the same data as Currie and Cole and similar economic deprivation that were not adequately cap- analytic techniques. However, they estimated impacts tured in the model. A related analysis lends some cre- for “very poor” women, those with incomes between dence to this hypothesis: Children who received FSP zero and 50 percent of the poverty line, and “less poor benefits for a portion of the years they were in poverty women,” those with incomes between 50 and 100 per- were significantly less likely to be wasted than chil- cent of poverty. In addition, they did not use instru- dren with a comparable poverty history who never mental variables and they adjusted NLSY income received food stamps. measures to exclude the value of FSP income. Findings from a fixed-effects model indicated that FSP Bhattacharya and Currie (2000) used data from participation was associated with a decreased likeli- NHANES-III to examine the relationship between FSP hood of low birthweight (less than 5.5 pounds) among very poor women (p <0.10). The authors reported this 47The researchers pooled data for the 1986 and 1988 supplements, with as a statistically significant finding, noting that the the result that more than one observation was included for some sample members. They appropriately caution that this feature leads to overstated sample available for the fixed effects logit model of levels of significance because repeat measures for individual children are low birthweight (n=153) was small (and therefore had likely to be more highly correlated than measurements across children.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 81 Chapter 3: Food Stamp Program participation and obesity among youth between the reduced odds of being at risk of overweight, compared ages of 12 and 16. They compared the proportion of with those who did not participate in the FSP. This was youth who were obese, based on Body Mass Index true for females living in both food-secure and food- (BMI).48 Cutoffs were adapted from standards defined insecure households. for adults. No FSP effect was detected. All of these results are subject to selection-bias prob- Gibson (2001) used data from NLSY97 to examine the lems, an important consideration in any attempt to link relationship between FSP participation and the likeli- weight status to participation in a food assistance pro- hood that youth between the ages of 12 and 18 would gram. In addition, results of both Bhattacharya and be underweight or obese. Like Bhattacharya and Currie (2000) and Gibson (2001) should be interpreted Currie, Gibson used BMI to classify subjects and with caution because the BMI cutoffs used in their based her cutoffs on standards defined for adults. She analyses were adapted from standards developed for estimated models that examined the impact of current adults rather than from the BMI-for-age charts devel- FSP receipt and current income as well as models that oped specifically for use with children and adolescents controlled for long-term poverty. In the models that (Kuczmarski, 2000). The use of self-reported weights looked at current FSP participation, FSP participation in the Jones et al. (2003) study is a concern. It is was associated with a significant decrease in the likeli- doubtful that cross-sectional studies can adequately hood that a youth would be obese. In the model that address questions about program impacts on children’s controlled for long-term poverty, this association was weights and heights. Indeed, researchers who attempt- no longer significant. The authors did not attempt to ed to assess the impact of the WIC program on these control for selection bias because “it is difficult to outcomes concluded that a longitudinal study with come up with an appropriate instrument for Food serial measurements was essential (Puma et al., 1991). Stamp receipt.” Adults Jones et al. (2003) looked at the relationship between Gibson (2003) used panel data from the 1985-96 food security, participation in FANPs, including the FSP, waves of the NLSY to assess the relationship between and the risk of overweight among children 5-12 in low- FSP participation and obesity (BMI $ 30) among income households (<185 percent of poverty). The adults ages 20-40. Her analysis included measures of authors used data from the 1997 Panel Study of Income both current and long-term FSP participation. The Dynamics (PSID) Child Development Supplement. sample was restricted to FSP participants and nonpar- Risk of overweight was defined as BMI-for-age at or ticipating individuals residing in households that were above the 85th percentile on BMI-for-age charts income-eligible for the FSP.49 Data on height and designed specifically for use with children and adoles- weight were self-reported. cents. Weights were reported by primary caregivers, and heights were measured by field interviewers. The Ordinary least squares models were estimated with and authors indicated that approximately 86 percent of the without fixed effects. Preliminary results showed that children had been weighed within the preceding month current and long-term FSP participation was signifi- and that 16 percent of caregivers had to estimate cantly related to the prevalence of obesity among weight because they had no recent reference point. women, but not among men. For this reason, the detailed analysis focused exclusively on women. Four The analysis compared the risk of overweight among different fixed effects models were estimated with children living in food-secure and food-insecure slightly different specifications. Results were largely households, while controlling for participation in a consistent across models and indicated that, among number of FANPs as well as other relevant character- low-income women, current participation in the FSP istics. Results showed that FSP participation did not was associated with an increase in the predicted proba- affect the likelihood that males would be overweight, bility of current obesity of 2 percentage points (a 9- regardless of whether they lived in food-secure or percent increase). Participation in the FSP in each of food-insecure households. Among females, however, the previous 5 years, compared with no participation those who participated in the FSP had a significantly

49The income cutoff for nonparticipants was defined as a family 48Body Mass Index (BMI) is the accepted standard for classifying adi- income-to-needs ratio of no more than 2, relative to defined income eligi- posity (or fatness) in adults (Barlow and Dietz, 1998). Since 2000, BMI- bility criteria. This cutoff ensured that the panel included individuals who for-age has also been recommended as a screening tool for children over crossed in and out of poverty and FSP eligibility (and perhaps FSP partici- the age of 2 (Kuczmarski et al., 2000). pation), but remained near-poor when ineligible.

82 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program over that period, was associated with an increase in the the FSP on a number of different nutritional bio- predicted probability of current obesity of 4.5 percent- chemistries. Bhattacharya and Currie focused on age points, or roughly 21 percent. youths ages 12-16 and examined the prevalence of anemia (based on low levels of hemoglobin or hemat- To test the sensitivity of her results, Gibson reestimat- ocrit), as well as the prevalence of high serum choles- ed all of the models using two different samples. She terol and low serum levels of vitamins A, C, and E, also estimated models that included controls for among FSP participants and nonparticipants. No sig- change in FSP eligibility and marital status in the pre- nificant differences were detected. vious calendar year as well as the timing of recent pregnancies-events that might trigger FSP participa- Dixon’s analysis focused on adults 20 and older. She tion. Finally, she examined the impact of current and compared the percentage of individuals with low long-term participation in AFDC (as an alternative serum levels of albumin, hemoglobin, iron, vitamin C, indicator of “social program participation”). No vitamin E, and carotenoids. She reported significant detailed data were presented, but the author reported differences between FSP participants and nonpartici- that estimates for all alternative models were similar to pants for albumin, vitamin C, and carotenoids. As the main analysis in both magnitude and significance. noted previously, however, Dixon did not limit her sample to low-income individuals and her model con- Although carefully designed and implemented, Gibson’s trolled for relatively few measured characteristics. analysis remains open to problems of selection bias and reverse causality. The fact that the analysis did not General Measures of Nutrition include information on food security status (because the or Health Status data are not available in the NLSY) is also a concern. Two of the identified studies assessed the impact of Other research has found a significant and positive FSP participation on general measures of nutrition or association between food insecurity and the prevalence health status. In her 2001 analysis of NLSY97 data, of overweight (see, for example, Townsend et al., 2001). described in a preceding section, Gibson examined A number of theories have been proposed to explain self-reported health status and the prevalence of chron- the apparently paradoxical relationship between food ic disease (as reported by parents or other primary insecurity and overweight (see Gibson, 2003; Townsend caregivers) among youths ages 12-18. Results showed et al.), but none has been thoroughly tested. that FSP participation was not significantly related to Elderly either outcome. Fey-Yensan et al. (2003) studied a small group of low- Fey-Yensan et al. (2003) examined self-reported gen- income elderly individuals in Connecticut. They eral health status, self-reported functional status, and reported that a greater percentage of FSP participants nutritional risk in a small group of low-income elderly than nonparticipants had BMIs $27. The analysis was individuals in Connecticut. Nutritional risk was meas- based on simple chi-square comparisons, however, and ured using the Nutrition Screening Initiative (NSI) data on height and weight were self-reported. Checklist.50 The authors found no significant differences between groups in general health status or func- Nutritional Biochemistries tional status. They did find, however, that FSP partici- Lopez and Habicht (1987b) examined a variety of pants had a significantly greater mean score on the measures of iron status among low-income elderly NSI checklist (signifying a greater level of nutritional individuals in NHANES-I and NHANES-II. risk) than either income-eligible or higher income non- Differences between FSP participants and nonpartici- participants. The authors also reported that FSP partic- pants were not statistically significant. Moreover, dif- ipants were more likely than nonparticipants to report ferences were inconsistent in direction, in some cases having fewer than two meals per day or not having suggesting that elderly FSP participants had better iron enough money to buy food. As noted above, however, status than nonparticipants (total iron binding capacity, this study used simple chi-square analyses. Therefore, free erythrocyte protoporphyrin), and in other cases findings are suggestive only. suggesting the opposite effect (hemoglobin, hematocrit, transferrin saturation, and serum iron). 50The NSI is a national collaborative effort of professional organizations committed to identifying and treating nutritional problems among the elderly. Leading sponsors include the American Academy of Family Bhattacharya and Currie (2000) and Dixon (2002) both Physicians, the American Dietetic Association, and the National Council on used data from NHANES-III to assess the impact of Aging. See www.aafp.org/nsi.xml.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 83 Chapter 3: Food Stamp Program

Summary This study also found that FSP adults ate significantly fewer servings of vegetables and less dietary fiber than The FSP provides benefits earmarked for at-home food nonparticipating adults. consumption to low-income households of all types. A substantial body of literature establishes firmly that, Studies that looked at the impact of the FSP on food while the greater part of food stamp benefits given to security have reported conflicting results. Some found households are used to free up resources to spend on that FSP participants were more likely than other low- things other than food, FSP benefits do cause house- income households to experience food insecurity. holds to spend more on food than they otherwise Other studies reported an inverse relationship-that FSP would. Moreover, the San Diego cashout demonstra- participants were less likely than nonparticipants to be tion established firmly that the use of earmarked food food insecure. The relationship between FANP partici- stamp benefits leads to a greater increase in expendi- pation and food security is a complex one and is par- tures for at-home food than would occur if households ticularly vulnerable to problems of selection bias and received the same benefit amount as unconstrained reverse causality. Food insecurity is likely to lead cash supplements. households to seek food assistance, and receipt of food stamp benefits may subsequently improve the house- It seems likely that the FSP increases the availability of hold’s food security. food energy and protein at the household level. Both of these effects were documented in a number of dif- Two recent studies that used sophisticated techniques ferent studies, including the San Diego cashout study. to attempt to control for selection bias suggest that, The FSP may also increase the availability of a num- once selection bias is controlled for, FSP participants ber of vitamins and minerals; however, the evidence in are no more likely to suffer from food insecurity (or this area is weaker. The strongest study that reported insufficiency) than nonparticipants. Moreover, one of significant effects on household availability of vita- the studies suggested that FSP benefits are more effec- mins and minerals used data that were collected in the tive in reducing levels of food insecurity with hunger 1970s, before elimination of the purchase requirement. than pure cash transfers. The San Diego cashout study found that FSP coupon households had greater availability of a number of Relatively little research has considered FSP impacts vitamins and minerals than cash households, but the on other nutrition- and health-related outcomes. differences were not statistically significant. Moreover, the number of studies available for any given outcome and population subgroup is limited, and The research shows little evidence that the FSP consis- each study has important limitations. tently affects the dietary intakes of individuals. There are scattered indications that FSP participation may The pattern of extant research suggests some paths for improve vitamin and mineral intakes of young chil- future research. There seems little need to document dren, but these findings were not replicated in the most further the relationship between food stamp benefits and recent and well-conducted analysis. Moreover, limita- at-home food expenditures. However, given the increas- tions in measurement techniques and nutrition stan- ing role that foods consumed away from home play in dards used in existing research make it impossible to the diets of most Americans (Lin et al., 1999), a more adequately address the critical research question of detailed examination of the impacts of the FSP on whether the prevalence of inadequate nutrient intakes expenditures for away-from-home food may be useful. differs for FSP participants and nonparticipants. In general, the impact of the FSP on nutrient availability Only a few studies looked at the impact of FSP partici- at the household level is of less interest than the impact pation on the intake of carbohydrates, fat, saturated on individual intakes. However, household availability fat, cholesterol, sodium, or fiber or on patterns of food is a more stable measure than individual intake and, intake. For the most part, these studies found little evi- therefore, has the potential for providing valuable dence of FSP impacts. Gleason et al. (2000), the information about the impact of the FSP. Future strongest study completed to date, found that pre- inquiries in this area should examine impacts associat- school FSP participants ate significantly fewer served with food use both at home and away from home. ings of grains and grain products than comparably aged nonparticipants and were significantly less likely Updated and improved studies of FSP impacts on indi- to meet the Dietary Guidelines recommendation of vidual dietary intakes are also needed because so many less than 10 percent of total energy from saturated fat. of the previous studies are dated, inconclusive, and

84 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program used dietary assessment methods that are not consis- food security as well as other variables that may be tent with currently recommended practices (see IOM, associated with weight status and should include care- 2001). Improved assessment of dietary intakes will ful attempts to control for self-selection. increase the likelihood that studies can detect small but meaningful FSP impacts. In addition, ongoing efforts to expand nutrition education in the FSP should be continued and evaluated. If Given the increasing problem of overweight and obesi- the FSP is to influence dietary intakes of individual ty in the United States, additional research on the rela- participants and, thus associated outcomes, such as tionship between FSP participation and patterns of bodyweight and other aspects of nutritional status, the overweight and obesity is desirable. Ideally, height and program must provide effective nutrition education to weight data should be measured rather than self- participants or find ways to connect FSP participants reported. Such research should include measures of with nutrition education activities sponsored by other programs and agencies.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 85 Chapter 3: Food Stamp Program

References Bhattacharya, J., and J. Currie. 2000. Youths at Nutritional Risk: Malnourished or Misnourished? Akin, J.S., D.K. Guilkey, B.M. Popkin, et al. 1987. Working paper 7686. Cambridge, MA: National “Determinants of Nutrient Intake of the Elderly,” Bureau of Economic Research. Journal of Applied Gerontology 6(3):227-58. Bishop, J.A., J.P. Formby, and L.A. Zeager. 2000. Akin, J.S., D.K. Guilkey, B.M. Popkin, et al. 1985. “The Effect of Food Stamp Cashout on “The Impact of Federal Transfer Programs on the Undernutrition,” Economics Letters 67:75-85. Nutrient Intake of Elderly Individuals,” Journal of Human Resources 20:383-404. Bishop, J.A., J.P. Formby, and L.A. Zeager. 1992. “Nutrition and Nonparticipation in the United States Alaimo, K., R.R. Briefel, E.A. Frongillo, Jr., et al. Food Stamp Program,” Applied Economics 1998. “Food Insufficiency Exists in the United States: 24(N9):945-49. Results from the Third National Health and Nutrition Examination Survey (NHANES-III),” American Blaylock, J.R. 1991. “The Impact of Equivalence Journal of Public Health 88(3):419-26. Scales on the Analysis of Income and Food Spending Distributions,” Western Journal of Agricultural Allen, J.E., and K.E. Gadson. 1983. Nutrient Economics 16(1):11-20. Consumption Patterns of Low-Income Households. TB-1685. USDA, Economic Research Service. Breunig, R., I. Dasgupta, C. Gundersen, et al. 2001. Explaining the Food Stamp Cash-Out Puzzle. FANRR- Barlow, S., and W. Dietz. 1998. “Obesity Evaluation 12. USDA, Economic Research Service. and Treatment: Expert Committee Recommendations,” Journal of Pediatrics 103(2):e29. Brown, M., S.R. Johnson, and R.L. Rizek. 1982. Food Stamps and Expenditure Patterns: A Statistical Bartlett, S., and M. Hart. 1987. Food Stamp Recipients’ Analysis. University of Missouri-Columbia. Report Patterns of Benefit Redemption. Cambridge, MA: Abt submitted to U.S. Department of Agriculture, Food Associates Inc. and Nutrition Service.

Basiotis, P., M. Brown, S.R. Johnson, et al. 1983. Butler, J.S., J.C. Ohls, and B. Posner. 1985. “The “Nutrient Availability, Food Costs, and Food Stamps,” Effect of the Food Stamp Program on the Nutrient American Journal of Agricultural Economics 65:685-93. Intake of the Eligible Elderly,” Journal of Human Resources 20(3):405-420. Basiotis, P., S. Johnson, K. Morgan, et al. 1987. “Food Stamps, Food Costs, Nutrient Availability and Nutrient Butler, J.S., and J.E. Raymond. 1996. “The Effect of Intake,” Journal of Policy Modeling 9:383-404. the Food Stamp Program on Nutrient Intake,” Economic Inquiry 34:781-98. Basiotis, P., C. Kramer-LeBlanc, and E. Kennedy. 1998. “Maintaining Nutrition Security and Diet Quality: Chavas, J.P., and M.L. Yeung. 1982. “Effects of the the Role of the Food Stamp Program and WIC,” Family Food Stamp Program on Food Consumption in the Economics and Nutrition Review 11(1-2):4-16. Southern United States,” Southern Journal of Agricultural Economics 14(1):131-39. Beebout, H., E. Cavin, B. Devaney, et al. 1985. Evaluation of the Nutrition Assistance Program in Chen, J.S. 1983. Simultaneous Equations Models with Puerto Rico: Volume II, Effects on Food Expenditures Qualitative Dependent Variables: A Food Stamp and Diet Quality. Washington, DC: Mathematica Program Participation and Food Cost Analysis. Policy Research, Inc. Unpublished doctoral dissertation from University of Missouri. Benus, J., J. Kmenta, and H. Shapiro. 1976. “The Dynamics of Household Budget Allocation to Food Cohen, B., J. Ohls, M. Andrews, et al. 1999. Food Expenditures,” The Review of Economics and Statistics Stamp Participants’ Food Security and Nutrient 58:129-38. Availability: Final Report. Princeton, NJ: Mathematica Policy Research, Inc.

86 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Cohen, B.E., and N. Young. 1993. “Impacts of the Dixon, B.L. 2002. “Differences in Dietary Intakes and Washington State Food Stamp Cashout Demonstration Serum Nutrients Between Adults from Families Who on Household Expenditures and Food Use,” in Received Foods Stamps and Adults from Families N. Fasciano, D. Hall, and H. Beebout (eds.), New Who Did Not Receive Food Stamps, NHANES-III, Directions in Food Stamp Policy Research. Papers 1988-1994,” FASEB Journal 16(4):A234. Abstract of Presented at the Food and Nutrition Service Research paper presented at annual meeting of the Professional Conference, Washington, DC, June 25, 1993, Research Scientists on Experimental Biology, April pp. 83-100. 2002. Slides provided by author.

Cole, N. 1997. Patterns of Food Stamp and Cash Fey-Yensan, N., C. English, H. Pacheco, et al. 2003. Welfare Benefit Redemption. USDA, Food and "Elderly Food Stamp Participants Are Different From Consumer Service. Eligible Nonparticipants by Level of Nutrition Risk but Not Nutrient Intake,” Journal of the American Cook, J.T., L.P. Sherman, and J.L. Brown. 1995. Dietetic Association 103(1):103-7. Impact of Food Stamps on the Dietary Adequacy of Poor Children. Medford, MA: Center on Hunger, Figlio, D., C. Gundersen, and J. Ziliak. 2000. “The Poverty and Nutrition Policy, Tufts University School Effects of the Macroeconomy and Welfare Reform on of Nutrition. Food Stamp Caseloads,” American Journal of Agricultural Economics 82:635-41. Cristofar, S., and P. Basiotis. 1992. “Dietary Intakes and Selected Characteristics of Women Ages 19-50 Fraker, T.M. 1990. The Effects of Food Stamps on Years and Their Children Ages 1-5 Years by Reported Food Consumption: A Review of the Literature. Perception of Food Sufficiency,” Journal of Nutrition USDA, Food and Nutrition Service. Education 24:53-8. Fraker, T.M., S.K. Long, and C.E. Post. 1990. Cunnyngham, K. 2002. Trends in Food Stamp Program Analyses of the 1985 Continuing Survey of Food Participation Rates: 1994-2000. USDA, Food and Intakes by Individuals. Volume I, Estimating Usual Nutrition Service. Dietary Intake, Assessing Dietary Adequacy, and Estimating Program Effects: Applications of Three Currie, J., and N. Cole. 1991. Does Participation in Advanced Methodologies Using FNS’s Four-Day Transfer Programs During Pregnancy Improve Birth Analysis File. USDA, Food and Nutrition Service. Weight? Los Angeles, CA: University of California at Los Angeles. Fraker, T.M., A.P. Martini, J.C. Ohls, et. al. 1992. The Evaluation of the Alabama Food Stamp Cash-out Davis, E.E., and A. Werner. 1993. The Effects of Food Demonstration: Volume 1, Recipient Impacts. USDA, Stamp Cash-out on Participants and Food Retailers in Food and Nutrition Service. the Alabama ASSETS Demonstration. Cambridge, MA: Abt Associates Inc. Frazao, E. 1999. “Chapter 1: High Costs of Poor Eating Patterns in the United States,” in E. Frazao (ed.), Devaney, B., and T. Fraker. 1989. “The Effect of Food America’s Eating Habits: Changes and Consequences. Stamps on Food Expenditures: An Assessment of AIB-750. USDA, Economic Research Service. Findings from the Nationwide Food Consumption Survey,” American Journal of Agricultural Economics Friedman, M. 1957. A Theory of the Consumption 71(1):99-104. Function. Princeton, NJ: Princeton University Press.

Devaney, B., and R. Moffitt. 1991. “Dietary Effects of Futrell, M.F., L.T. Kilgore, and F. Windam. 1975. the Food Stamp Program,” American Journal of “Nutritional Status of Black Preschool Children in Agricultural Economics 73(1):202-11. Mississippi,” Journal of the American Dietetic Association 66:22-27.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 87 Chapter 3: Food Stamp Program

Gibson, D. 2003. “Food Stamp Program Participation Johnson, S.R., J.A. Burt, and K.J. Morgan. 1981. “The is Positively Related to Obesity in Low-Income Food Stamp Program: Participation, Food Cost, and Women,” Journal of Nutrition 133:2225-31. Diet Quality of Low-income Households,” Food Technology 35(10):58-70. Gibson, D. 2001. “Poverty, Food Stamp Program Participation and Health: Estimates from the Jensen, H. 2002. “Food Insecurity and the Food Stamp NLSY97,” in R.T. Michael (ed.), Social Awareness: Program,” American Journal of Agricultural Adolescent Behavior as Adulthood Approaches. New Economics 84(5):1215-28. York, NY: Russell Sage Foundation. Jones, S.J., L. Jahns, B. Laraia, and B. Haughton. Gleason, P., A. Rangarajan, and C. Olson. 2000. 2003. “Lower Risk of Overweight in School-age Food Dietary Intake and Dietary Attitudes Among Food Insecure Girls who Participate in Food Assistance: Stamp Participants and Other Low-Income Results from the Panel Study of Income Dynamics Individuals. USDA, Food and Nutrition Service. Child Development Supplement,” Archives of Pediatrics & Adolescent Medicine 157(8):780-84. Gregorio, D.I., and J.R. Marshall. 1984. “Fine Tuning Well-Being: Food Stamp Use and Nutritional Kennedy, E.T., J. Ohls, S. Carlson, et al. 1995. “The Adequacy of Children’s Diets,” Social Science Healthy Eating Index: Design and Applications,” Quarterly 65:1137-46. Journal of the American Dietetic Association 95(10):1103-09. Gundersen, C., and V. Oliveira. 2001. “The Food Stamp Program and Food Insufficiency,” American Kisker, E.E., and B. Devaney. 1988. The Food Choices Journal of Agricultural Economics 83(4):875-87. of Low-Income Households: Final Report. Washington, DC: Mathematica Policy Research, Inc. Hama, M.Y., and W.S. Chern. 1988. “Food Expenditure and Nutrient Availability in Elderly Korenman, S., and J. Miller. 1992. Food Stamp Households,” Journal of Consumer Affairs 22(1):3-19. Program Participation and Maternal and Child Health. Report submitted to USDA, Food and Hamilton, W.L., J.T. Cook, W.W. Thompson, et al. 1997. Nutrition Service. Household Food Security in the United States in 1995: Summary Report of the Food Security Measurement Kornfeld, R. 2002. Explaining Recent Trends in Food Project. Cambridge, MA: Abt Associates Inc. Stamp Program Caseloads: Final Report. E-FAN-02- 008. USDA, Economic Research Service. Huffman, S.K., and H. Jensen. 2003. Do Food Assis- tance Programs Improve Household Food Security? Kott, P.S. 1990. “The Effects of Food Stamps on Food Recent Evidence from the United States. Working Expenditures: Comment,” American Journal of paper 03-WP-335. Ames, IA: Iowa State University, Agricultural Economics 72:731. Center for Agricultural and Rural Development. Kramer-LeBlanc, C., P. Basiotis, and E. Kennedy. Hymans, S.H., and H.T. Shapiro. 1976. “The 1997. “Maintaining Food and Nutrition Security in the Allocation of Household Income to Food Consumption,” United States with Welfare Reform,” American Journal of Econometrics 4:167-88. Journal of Agricultural Economics 79(5):1600-07.

Institute of Medicine, Food and Nutrition Board. 2001. Kuczmarski, R., C. Ogden, L. Guo, et al. 2002. 2000 Dietary Reference Intakes: Application in Dietary CDC Growth Charts for the United States: Methods Assessment. Washington, DC: National Academy Press. and Development. Vital and Health Statistics Series 11, No. 246. Centers for Disease Control and Prevention, Jacobsen, J., N. Rodriguez-Planas, L. Puffer, et al. 2001. National Center for Health Statistics. The Consequences of Welfare Reform and Economic Change for the Food Stamp Program-Illustrations Lane, S. 1978. “Food Distribution and Food Stamp from Microsimulation: Final Report. E-FAN-01-003. Program Effects on Food Consumption and Nutritional USDA, Economic Research Service. Achievement of Low Income Persons in Kern County, California,” American Journal of Agricultural Economics 60(1):108-16.

88 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 3: Food Stamp Program

Levedahl, J.W. 1995. “A Theoretical and Empirical Perkin, J., L.A. Crandall, and S.F. McCann. 1988. Evaluation of the Functional Forms Used to Estimate “Ethnicity and Food Stamp Program Participation: the Food Expenditure Equation of Food Stamp Effect upon Dietary Intakes of Low-income Mothers Recipients,” American Journal of Agricultural Served by a North Florida Family Practice Center,” Economics 77:960-68. Journal of the American Dietetic Association 88:1081- 86. Levedahl, J.W. 1991. The Effect of Food Stamps and Income on Household Food Expenditures. TB-1794. Posner, B.M., J.C. Ohls, and J.C. Morgan. 1987. USDA, Economic Research Service. “Impact of Food Stamps and Other Variables on Nutrient Intake in the Elderly,” Journal of Nutrition Lin, B.-H., J. Guthrie, and E. Frazao. 1999. “Chapter 12: for the Elderly 6(3):3-16. Nutrient Contribution of Food Away from Home,” in E. Frazao (ed.), America’s Eating Habits: Changes and Prato, A.A., and J.N. Bagali. 1976. “Nutrition and Conse-quences. AIB-750. USDA, Economic Research Nonnutrition Components of Demand for Food Items,” Service. American Journal of Agricultural Economics 58:563-67.

Lopez, L.M., and J.P. Habicht. 1987a. “Food Stamps Price, D.W. 1983. Effects of Socioeconomic Variables and the Energy Status of the U.S. Elderly Poor,” and Food Stamp Participation on the Consumption of Journal of the American Dietetic Association Selected Food Groups. Research Bulletin No. XB 87(8):1020-24. 0932. Agricultural Research Center, Washington State University. Lopez, L.M., and J.P. Habicht. 1987b. “Food Stamps and the Iron Status of the U.S. Elderly Poor,” Journal Putnam, J., and S. Gerrior. 1999. “Chapter 7: Trends in of the American Dietetic Association 87(5):598-603. the U.S. Food Supply, 1980-97,” in E. Frazao (ed.) America’s Eating Habits: Changes and Consequences. Moffitt, R. 1989. “Estimating the Value of an In-kind AIB-750. USDA, Economic Research Service. Transfer: The Case of Food Stamps,” Econometrica 57(2):385-409. Ranney, C.K., and J.E. Kushman. 1987. “Cash Equivalence, Welfare Stigma, and Food Stamps,” Neenan, P.H., and C.G. Davis. 1977. Impact of the Southern Economics Journal 53(4):1011-27. Food Stamp Program and Expanded Food and Nutrition Education Programs on Food Expenditures Rose, D., J.P. Habicht, and B. Devaney. 1998a. and Nutrient Intake of Low-Income Rural Florida “Household Participation in the Food Stamp and WIC Households. Gainesville, FL: Florida Agricultural Programs Increases the Nutrient Intakes of Preschool Experiment Station, Project AS-01629. Children,” Journal of Nutrition 128(3):548-55.

Ohls J.C., and H. Beebout. 1993. “Chapter One: The Rose, D., C. Gundersen, and V. Oliveira. 1998b. Socio- Context,” in J.C. Ohls and H. Beebout (eds.) The Food Economic Determinants of Food Insecurity in the United Stamp Program: Design Tradeoffs, Policy, and States. Evidence from the SIPP and CSFII Datasets. Impacts. Washington, DC: Mathematica Policy TB-1869. USDA, Economic Research Service. Research, Inc., pp. 1-20. Rose, D., D. Smallwood, and J. Blaylock. 1995. Ohls, J.C., T.M. Fraker, A.P. Martini, et al. 1992. The “Socio-economic Factors Associated with the Iron Effects of Cash-out on Food Use by Food Stamp Intake of Preschoolers in the United States,” Nutrition Program Participants in San Diego. USDA, Food and Research 15(9):1297-1309. Nutrition Service. Rosso, R. 2003. Characteristics of Food Stamp Perez-Escamilla, R., A.M. Ferris, L. Drake, et al. 2000. Households: Fiscal Year 2001. USDA, Food and “Food Stamps are Associated with Food Security and Nutrition Service. Dietary Intake of Inner-City Preschoolers from Hartford, Connecticut,” Journal of Nutrition 130:2711-17. Salathe, L.E. 1980. “The Food Stamp Program and Low-income Households’ Food Purchases,” Agricultural Economic Research 32(4):33-41.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 89 Chapter 3: Food Stamp Program

Scearce, W.K., and R.B. Jensen. 1979. “Food Stamp U.S. Department of Health and Human Services. 2000. Program Effects on Availability of Food Nutrients for Healthy People 2010: Understanding and Improving Low Income Families in the Southern Region of the Health, 2nd Edition. United States,” Southern Journal of Agricultural Economics 11(2):113-20. Wallace, G., and R. Blank. 1999. “What Goes Up Must Come Down? Explaining Recent Changes in Public Senauer, B., and N. Young. 1986. “The Impact of Food Assistance Caseloads,” in S. Danziger (ed.), Economic Stamps on Food Expenditures: Rejection of the Conditions and Welfare Reform. Kalamazoo, MI: Traditional Model,” American Journal of Agricultural Upjohn Institute. Economics 68(1):37-43. Weimer, J.P. 1998. Factors Affecting Nutrient Intake of Smallwood, D.M., and J.R. Blaylock. 1985. “Analysis the Elderly. AER-769. USDA., Economic Research of Food Stamp Program Participation and Food Service. Expenditures,” Western Journal of Agricultural Economics 10(1):41-54. West, D.A. 1984. Effects of the Food Stamp Program on Food Expenditures. Research Bulletin No. XB Southworth, H.M. 1945. “The Economics of Public 0922. Agricultural Research Center, Washington State Measures to Subsidize Food Consumption,” Journal of University. Farm Economics 27:38-66. West, D.A., and D.W. Price. 1976. “The Effects of Townsend, M.S., J. Peerson, B. Love, et al. 2000. Income, Assets, Food Programs, and Household Size “Food Insecurity is Positively Related to Overweight on Food Consumption,” American Journal of in Women,” Journal of Nutrition 131:1738-45. Agricultural Economics 58(1):725-30.

Tuttle, C. 2002. Characteristics of Food Stamp West, D.A., D.W. Price, and D.Z. Price. 1978. Households: Fiscal Year 2001 (Advance report). “Impacts of the Food Stamp Program on Value of USDA, Food and Nutrition Service. Food Consumed and Nutrient Intake among Washington Households with 8-12 Year Old Children,” U.S. Department of Agriculture. 2004. “Veneman Western Journal of Agricultural Economics 3:131-44. Announces Full Implementation of Food Stamp Program Electronic Benefits Transfer System,” USDA Whitfield, R.A. 1982. “A Nutritional Analysis of the News Release 0251.04, June 22, 2004. Food Stamp Program,” American Journal of Public Health 72(8):793-99. U.S. Department of Agriculture, Food and Nutrition Service. 2003a. Program data. Available: http://www. Wilde, P. 2001. “Strong Economy and Welfare Reform fns.usda.gov/pd. Accessed April 2003. Contribute to Drop in Food Stamp Rolls,” FoodReview 24(1):2-7. USDA, Economic Research Service. U.S. Department of Agriculture, Food and Nutrition Service. 2003b. “Food Stamp Program Nutrition Wilde, P., P. Cook, C. Gundersen, et al. 2000a. The Education Fact Sheet.” Available: http://www.fns.usda. Decline in Food Stamp Program Participation in the gov/fsp/menu/admin/nutritioned/fsheet.htm. Accessed 1990s. FANRR-7. USDA, Economic Research April 2003. Service.

U.S. Department of Agriculture, Food and Nutrition Wilde, P., S. Hofferth, S. Stanhope, et al. 2000b. “Pre- Service. 2001. The Decline in Food Stamp 1997 Trends in Welfare and Food Assistance in a Participation: A Report to Congress. National Sample of Families,” American Journal of Agricultural Economics 82(3):642-48. U.S. Department of Agriculture, Food and Nutrition Service. 2000. “Food and Nutrition Service (FNS) Wilde, P., P.E. McNamara, and C.K. Ranney. 1999. Strategic Plan 2000 to 2005.” Available: http://www. “The Effects of Income and Food Programs on Dietary fns.usda.gov/oane/MENU/gpra/FNSStrategicplan.htm. Quality: a Seemingly Unrelated Regression Analysis Accessed April 2002. with Error Components,” American Journal of Agricultural Economics 81:201-213.

90 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4 WIC Program

The Special Supplemental Nutrition Program for Women, Program growth was particularly rapid during the first Infants, and Children (WIC) was established to provide decade of operation. Between FY 1975 and FY 1985, “supplemental nutritious food as an adjunct to good health WIC participation increased from 344,000 participants care during critical times of growth and development, in per month to more than 3.1 million. On average, par- order to prevent the occurrence of health problems and ticipation increased about 26 percent per year improve health status...” (P.L. 95-627).51 The WIC pro- (USDA/FNS, 2003a). During the next decade, the program targets five specific groups: pregnant women, gram continued to grow each year but at a notably infants, children up to their fifth birthday, breastfeeding slower pace. Total monthly participation increased women (up to 1 year after an infant’s birth), and non- from about 3.1 million in FY 1985 to 6.9 million in breastfeeding postpartum women (up to 6 months after an FY 1995. The annual increase during this period aver- infant’s birth). In addition to belonging to one of these tar- aged about 8 percent. Since the late 1990s, WIC par- get groups, WIC participants must be low-income and ticipation has stabilized. Participation actually declined have one or more documented nutritional risks. by 1-2 percent in 3 consecutive years between FY 1998 and FY 2000, but has increased modestly (2-3 WIC offers a combination of services, including sup- percent per year) since then. plemental foods that have been specifically selected to supply nutrients potentially lacking in participants’ Much of WIC’s growth over the years has been fueled diets, nutrition education, and referrals to health care by favorable Congressional funding, which has been and social services. WIC services do not fluctuate by influenced at least partially by research suggesting that household income. All participants have access to the WIC participation during pregnancy increases infant same basic benefits. The types and amounts of supple- birthweight and decreases Medicaid costs. In the early- mental food provided to each participant are based on to mid-1990s, program growth was also fueled by participant category, age (for infants), and individual infant formula rebate programs, which became manda- needs and preferences. tory in 1989 (P.L. 101-147). Under the rebate programs, each State awards a competitively bid contract WIC is not an entitlement program, so the number of to one infant formula manufacturer. For the exclusive participants served by the program may be affected by contract on WIC infant formula, manufacturers agree Federal funding levels. In FY 2002, WIC served 7.5 to provide rebates to WIC State agencies for each can million participants per month at an estimated total of formula purchased by WIC participants. The funds cost of $4.3 billion (U.S. Department of Agriculture received through the rebate system are used to contain (USDA), Food and Nutrition Service (FNS), 2003a). overall costs and to support provision of program benefits to additional participants. In FY 2002, the WIC Program Overview program recognized $1.5 billion in rebate savings (USDA/FNS, 2003b). A major impetus for the WIC program was the 1969 White House Conference on Food, Nutrition, and Program Administration Health, which reported nutritional deficiencies among FNS and its seven regional offices provide cash grants low-income pregnant women and young children. WIC to State WIC agencies, issue regulations, and monitor began as a 2-year pilot program in 1972 and was author- compliance with these regulations. State WIC agencies ized as a permanent program in 1975 (P.L. 94-105). In operate in each of the 50 States, as well as in the the intervening years, WIC has grown substantially District of Columbia, Puerto Rico, Guam, American and has become a key component of the nutrition safe- Samoa, and the American Virgin Islands. Thirty-three ty net provided for low-income Americans. Indian Tribal Organizations also serve as State WIC agencies (USDA/FNS, 2003b). State WIC agencies 51WIC was formerly known as the Special Supplemental Food Program contract with local WIC agencies to provide WIC for Women, Infants, and Children. The program name was changed under benefits to participants, monitor compliance with reg- the Healthy Meals for Healthy Americans Act of 1994 (P.L. 103-448) to emphasize that WIC is a targeted supplemental nutrition program rather ulations, and provide technical assistance to local than an income supplement program. agency staff.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 91 Chapter 4: WIC Program

Funds allocated to local WIC agencies are used to pro- Since the mid-1980s, several legislative actions have vide supplemental foods to WIC participants and to expanded Medicaid income eligibility for pregnant pay administrative costs, including the costs of certify- women, infants, and children. As a result, some States ing applicants as eligible and providing nutrition edu- have adopted Medicaid income-eligibility limits that cation. Each of the roughly 2,200 local WIC agencies exceed the WIC maximum of 185 percent of poverty. In operates one or more service delivery sites where par- October 2002, 17 States had Medicaid eligibility stan- ticipants go to receive WIC services (Bartlett et al., dards that exceeded the WIC cutoff (National 2002). Most of the local agencies are State, county, or Governor’s Association, 2003). In most cases, the local health departments. Other organizations, howev- expanded income-eligibility cutoff is 200 percent of er, such as hospitals, State- or locally sponsored mater- poverty and is limited to pregnant women and/or infants. nal and child health programs, and community action agencies, also provide WIC services. In addition to meeting eligibility requirements associated with residency, participant category, and income, Participant Eligibility each WIC participant must be at nutritional risk, as documented by a competent professional authority (a WIC eligibility is based on four factors: State resi- physician, nutritionist, nurse, or other health profes- dence, categorical eligibility, income eligibility, and sional). Before 1999, State agencies established their nutritional risk. Unless they are part of a migrant farm own nutritional risk criteria following broad guidelines worker family, WIC participants must be residents of in Federal regulations. This autonomy meant that the the State or other jurisdiction (U.S. territory or Indian criteria used to define nutritional risk and, consequent- reservation) supplying the WIC benefits. ly, program eligibility, varied across State agencies. Participants must also belong to one of five categori- This variability raised concerns about equity. To cally eligible groups—women during pregnancy and up address these concerns, FNS asked the Institute of to 6 weeks after delivery, breastfeeding women (who Medicine (IOM) to review the scientific basis for the can participate for up to a year after giving birth), non- criteria being used to define nutritional risk and to rec- breastfeeding postpartum women (who can participate ommend about appropriate criteria for future use for up to 6 months after giving birth or other termina- (IOM, 1996). The IOM report formed the basis for a tion of pregnancy), infants (0-12 months), and children standardized list of nutritional risk criteria to be used up to the age of 5. In April 2002, 50 percent of all in all WIC programs nationwide. States are still free to WIC participants were children and 26 percent were define the specific criteria used to determine program infants. The remaining 24 percent were women—11 eligibility, but, since April 1999, criteria must be percent pregnant women, 8 percent postpartum non- selected from the approved list. breastfeeding women, and 6 percent breastfeeding As noted previously, WIC is not an entitlement program. women (Bartlett et al., 2003; Kresge, 2003). The program must operate within annual funding levels Income eligibility for the WIC program is defined by established by Congress. The number of participants each State agency. The cutoff may not be more than served each year depends on available funding and the 185 percent or less than 100 percent of the Office of cost of running the program. To deal with the possibility Management and Budget’s (OMB) poverty income that local programs may not be able to serve all eligible guidelines. As of April 2000, all State agencies used an people, WIC uses a priority system to allocate avail- income eligibility cutoff of 185 percent of poverty able caseload slots to eligible applicants. The priority (Bartlett et al., 2002). Program regulations allow local system is designed to ensure that available services go WIC agencies to determine that participants are to those most in need. In general, pregnant women, adjunctively income-eligible for WIC if they or certain breastfeeding women, and infants are given higher pri- family members participate in Medicaid, Temporary ority than children and nonbreastfeeding postpartum Assistance for Needy Families (TANF), or the Food women. In addition, applicants with nutritional risks Stamp Program (FSP). Since October 1998, applicants that are based on hematologic measures, anthropomet- not certified under adjunctive income-eligibility provi- ric measures, or medical conditions are given higher priority than applicants with nutritional risks based on sions must present documentation of income at certifi- 52 cation (P.L. 105-336). Before this regulation went into dietary patterns or other characteristics. effect, some States allowed applicants to self-report income without documentation. 52See 7 Code of Federal Regulations (CFR), 246.7.

92 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

The relative importance of the priority system has 2003). To achieve this goal, the program provides a declined over time as increasing funds have allowed combination of services, including supplemental foods, the program to serve many lower-priority individuals. nutrition education, and referral to health and social Between 1988 and 1997, favorable Congressional services. Participants are generally certified to receive funding and cost-containment measures (especially benefits for 6-month periods and must be recertified to formula rebates) fueled an overall increase of 106 per- continue receiving benefits. Exceptions to this rule cent in WIC participation. Participation increased more include pregnant women (who are certified for the substantially for children than for higher-priority duration of the pregnancy and up to 6 weeks postpar- groups (128 percent vs. 110 percent for women and 70 tum), infants (who are generally certified up to 1 year percent of infants). The reason for the disparity was of age), and nonbreastfeeding postpartum women that a large percentage of eligible women and infants (whose eligibility expires at 6 months postpartum). were already participating because of their higher priority (Oliveira et al., 2002). Supplemental Foods The supplemental foods provided by WIC are good Today, the WIC program serves almost half of all infants sources of many nutrients, including those potentially in the U.S. and about a quarter of the children ages 1-4 lacking in the diets of low-income pregnant women years (Hirschman, 2004). The question of how many and children—protein, iron, calcium, and vitamins A eligible participants go unserved has been the subject and C. Foods available in WIC food packages include of much debate. Historically, FNS has estimated the milk, eggs, cheese, dried beans and peas, peanut but- number of individuals eligible to participate in WIC in ter, full-strength (100 percent) fruit or vegetable juices order to predict WIC caseloads. FNS’s estimates have high in vitamin C, and breakfast cereals high in iron been questioned in recent years, however, because esti- and low in sugar. Food packages for infants are limited mated coverage rates for some participant categories to iron-fortified infant formula, infant cereals, and, for have exceeded 100 percent. Program advocates argued infants 4 months and older, 100 percent fruit or veg- that FNS underestimates the number of eligible indi- etable juices high in vitamin C. Breastfeeding women viduals, while others, including members of Congress, whose infants do not receive WIC formula may also raised concerns that the program is serving ineligible receive carrots and canned tuna. individuals. In response to these concerns, FNS completed a number of studies to identify problems with Federal regulations specify minimum nutritional the existing estimation methodology and potential requirements for all WIC foods (USDA/FNS, 2003c). solutions (see, for example, Gordon et al., 1999 and State WIC agencies are not required to authorize every 1997). As a result of these efforts, a new methodology available food that meets minimum nutritional require- was introduced for estimating the number of WIC eli- ments. States may limit authorization to specific gibles at the State level (Gordon et al., 1999). brands and types of food based on cost, distribution within the State, participant acceptance, and/or admin- Before revising the methodology used at the National istrative feasibility. level, FNS asked the Committee on National Statistics of the National Research Council to convene a panel WIC food packages are meant to supplement partici- of experts to study the existing methodology and make pants’ diets and are not expected to fully satisfy daily recommendations for improvement. The panel con- nutritional needs. The type and quantity of foods pro- cluded that the existing methodology substantially vided to individual participants vary by participant cat- underestimates the number of individuals eligible to egory. Federal regulations define maximum monthly participate in WIC (Ver Ploeg and Betson, 2003). The allotments for different types of participants primary reason for the underestimation is that the (USDA/FNS, 2003c). Maximum monthly allotments methodology does not adequately measure monthly must be made available to participants if medically or income and adjunctive eligibility. The panel proposed nutritionally warranted. However, WIC staff may tailor two alternative approaches to estimating WIC eligibility. the content of food packages (within maximum allot- At the time this report went to press, FNS was in the ments) to meet individual needs and preferences. process of implementing the panel’s recommendations. Most WIC participants receive vouchers or checks to Program Benefits use in purchasing supplemental foods at authorized retail WIC was designed to counteract the negative effects outlets. In a limited number of geographic areas, foods of poverty on prenatal and pediatric health (Kresge, are delivered to participants’ homes or participants

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 93 Chapter 4: WIC Program pick up foods at warehouses. In recent years, several State and local WIC agencies have broad autonomy to States have conducted pilot tests on the use of elec- develop plans and procedures for providing nutrition tronic benefits transfer (EBT) systems in disbursing education to WIC participants. Consequently, WIC WIC benefits. At least one State has implemented EBT nutrition education is quite diverse and may vary in statewide and another State is considering a statewide both quantity and quality from one site to the next. A EBT system. variety of methods may be used to provide nutrition education. For example, participants may be counseled In mid-2003, FNS launched an initiative to revise one-on-one, may attend classes, or may view videos, existing food packages based on current nutrition rec- filmstrips, or slide presentations on a range of nutri- ommendations, updated information about the dietary tion- or health-related topics. Providers are encouraged patterns and nutritional needs of low income women, to ensure that nutrition education messages take into infants and children, and new products in the market- account participants’ educational levels, nutritional place (Federal Register, 2003). Following a period of needs, household situations, and cultural preferences. public comment, USDA asked the IOM to convene a panel of independent experts to review available sci- Although local WIC agencies are required to offer ence and public comments and to develop recommen- nutrition education, participants are free to decline dations for revising WIC food packages. A preliminary these services without affecting receipt of other pro- report was released in mid-2004 (IOM, 2004) and the gram benefits. To maximize participation, local final report is expected in 2005 (Okita, 2004). agency staff tend to schedule nutrition education activities to coincide with issuance of WIC vouchers Nutrition Education (Fox et al., 1998). Because the food package does not meet participants’ Referrals to Health and Social Services total nutrient needs, nutrition education is seen as an essential part of the WIC Program. Nutrition education Local WIC agencies are expected to serve as a link provides a mechanism for teaching WIC participants between participants and the health care system and to about recommended eating patterns and for encourag- promote routine use of preventive health care services. ing them to adopt positive food-related attitudes and Local WIC staff are also encouraged to provide refer- behaviors. Program regulations define two broad goals rals, as needed, to appropriate social services, such as for WIC nutrition education: the FSP, Medicaid, TANF, and other programs relevant to the participants’ needs. The degree to which local • To stress the relationship between proper nutrition WIC agencies actually facilitate linkages to health and and good health, with special emphasis on the nutri- social services varies depending on the adequacy of tional needs of the program’s target populations. the health and social service infrastructure at the State and local levels and the extent to which participants • To assist individuals at nutritional risk in achieving a are already linked into health and social service net- positive change in food habits, resulting in improved works before coming to WIC (Fox et al., 1998). nutritional status and the prevention of nutrition- related problems (7 CFR, 246.11). Research Overview In practice, WIC nutrition education addresses many other topics, such as breastfeeding promotion; the need The WIC program has been studied widely. Indeed, it to avoid cigarettes, alcohol, illicit drugs, and over-the- is the most studied of the Federal FANPs with regard counter medications during pregnancy; and the impor- to impacts on nutrition- and health-related outcomes. tance of childhood immunizations. The available body of research is impressive in size and, in many circles, is seen as solidly convincing that State WIC agencies are required to earmark at least WIC has positive impacts, particularly on birth out- one-sixth of annual administrative funds for nutrition comes. The truth is, however, that much of the avail- education. Local WIC agencies are required to offer all able research is clouded by the overarching problem adult participants and caretakers of infant and child of selection bias. In addition, the complexity of the participants at least two nutrition education contacts health outcomes that have been studied has presented during each certification period. For participants with unique challenges to WIC researchers, further compro- certifications that extend beyond 6 months, nutrition mising their ability to obtain clear estimates of pro- education must be offered quarterly. gram impact.

94 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Over the years, USDA has made a considerable invest- and a specially created WIC-Medicaid database that ment in trying to elucidate the impact of WIC on par- included data on Medicaid expenditures, maternal ticipants’ nutrition and health status. The first national WIC participation during pregnancy, and birth out- evaluation of WIC was completed when the program comes for live births in five States in 1987-88 was still very young (Edozien et al., 1979). The so- (Devaney et al., 1990/91; Devaney, 1992; Devaney called Medical Evaluation of WIC included more than and Schirm, 1993). These secondary analyses have 50,000 WIC participants in 14 States and examined focused almost exclusively on impacts on birth out- impacts on birth outcomes, child growth, anemia, and comes, including savings in Medicaid costs. other measures of nutritional status. Study authors reported positive impacts, but the study has been wide- In addition to USDA-sponsored research, many inde- ly criticized for, among other things, poor response pendent researchers have looked at WIC impacts using rates on followup measures and dissimilarities between secondary analyses of existing databases, as well as participant and nonparticipant groups. In addition, the primary data collected on State or local samples. The study’s dose-response design, which compared newly remainder of this chapter summarizes findings from all enrolling participants (nonparticipants) with partici- of this research. pants who had been in the program for some time (participants), has come to be regarded as a poor The discussion is organized around WIC participant design for studying birth outcomes. categories. Impacts of prenatal WIC participation are discussed first. The bulk of this research focuses on In the early 1980s, USDA sponsored the National WIC impacts on birth outcomes, with a much smaller body Evaluation (NWE) (Rush et al.,1986) which consisted of work examining impacts on pregnant women them- of four substudies, including an historical study of selves. Research that examined the impact of WIC par- birth outcomes (Rush et al.,1988a); a longitudinal ticipation on the initiation and/or duration of breast- study of pregnant women (Rush et al.,1988d); a cross- feeding is also included in this section. The rationale is sectional study of infants and children (Rush et al., that the decision to initiate breastfeeding is generally 1988c); and a study of food expenditures (Rush et made before an infant leaves the hospital, making the al.,1988b). Although the NWE is generally regarded as prenatal period a key point for intervention. Although a carefully implemented study and remains the largest decisions about breastfeeding duration are generally and most comprehensive study of WIC ever complet- made during the postpartum period, for ease of discus- ed, it also had problems with noncomparability sion, all research related to breastfeeding outcomes are between participant and nonparticipant groups, as well discussed in the same section. as with crossovers between groups. The second section summarizes research that assessed In the late 1980s, USDA undertook a feasibility assess- impacts of WIC participation on infants and children. ment and design effort aimed at developing and fielding The third section describes studies that have assessed a study that would produce reliable estimates of the impacts on postpartum women (both nonbreastfeeding impact of WIC on infants and children. Outcomes to and breastfeeding). The final section summarizes find- be examined included dietary intake, anemia, physical ings from four studies that examined impacts on all and cognitive growth, and use of health care services. types of WIC participants, without differentiating par- Unfortunately, the so-called WIC Child Impact Study ticipant groups, or on household-level outcomes. was canceled in 1992, at the request of Congress, before the full evaluation could be fielded. Results Selection Bias from a limited field test provide some information Because use of randomized experiments is considered about potential impacts on young infants (6 months unethical by many policymakers and program admin- old) but fall far short of providing valid impact esti- istrators, only one study (Metcoff et al., 1985) used mates (Burstein et al.,1991). In addition, the field test random assignment to study the impact of the WIC suffered from some of the same problems with non- program (see chapter 2 for an explanation of the ran- comparability and crossovers that affected the NWE. domized experiment). Random assignment was feasible for these authors because, at the time the study USDA’s most recent efforts to assess WIC impacts was conducted, the demand for WIC participation at have relied on secondary analyses of extant databases, the study site exceeded the available funding. All most notably the 1988 National Maternal and Infant other studies of WIC impacts have used quasi-experi- Health Survey (NMIHS) (Gordon and Nelson, 1995) mental designs.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 95 Chapter 4: WIC Program

Selection bias, as discussed in chapter 2, is driven by capita State-level WIC food expenditures (a proxy for the fact that women who participate in WIC or who the availability of WIC services); an indicator of enroll their infants or children in WIC may differ in whether the family had income from wages (as an unmeasured ways from women who are eligible but do indicator of the level of contact with public assistance not participate. These differences may influence the agencies); and an indicator of WIC participation dur- outcomes being studied. This influence could run in ing previous pregnancies. either direction, resulting in overestimation or underestimation of the true effect of the program. For exam- Ultimately, Gordon and Nelson deemed their efforts to ple, women who participate in WIC or enroll their control for selection bias to be unsuccessful. After sev- children in WIC may be more health-conscious and eral different estimation procedures and model specifi- motivated than women who do not participate, or may cations yielded implausible and highly unstable be more knowledgeable about and connected to the results, they concluded the following: health care system. These women and their offspring might have better outcomes than nonparticipants, even It is possible that the selection-bias-correction models of the effects of WIC on birth outcomes produce unstable and in the absence of the program. In this case, estimates implausible results because the factors affecting WIC partic- of WIC impacts would be overstated. On the other ipation and birthweight are very nearly identical, since WIC hand, because the WIC program specifically targets targets low-income women at risk for poor pregnancy out- individuals who are at nutritional risk, WIC partici- comes. In this case, modeling the participation decision is pants may be more likely, a priori, to have poor out- not likely to be a useful approach to controlling for selection bias. comes than otherwise comparable individuals who do not enroll in the program. In this case, estimates of Brien and Swann (1999) analyzed data from the WIC impacts would be understated. NMIHS with the explicit goal of developing strategies to deal with selection bias. To minimize potential bias, The problem of selection bias was largely ignored in they restricted their sample to non-Hispanic Blacks the earliest WIC research. The first study to attempt to and non-Hispanic Whites and carried out separate control from selection bias was the WIC-Medicaid analyses for each group. The authors used a two-stage Study, which estimated the impact of prenatal WIC estimation procedure, similar to the basic approach participation on a number of birth outcomes (Devaney used by Gordon and Nelson (1995). To model the par- et al. (1990/91)). Study authors estimated a number of ticipation decision, the authors used a variety of State- different selection-bias-adjustment models but ulti- level characteristics that served as proxies for the mately rejected all of them because they produced availability and “generosity” of WIC and other welfare implausible findings and were extremely sensitive to programs. These characteristics included relative ease minor changes in specification and to estimation pro- of the State’s WIC income certification policies, pres- cedures. Devaney and Schirm (1993) reported compa- ence of brand-name purchase restrictions for WIC rable experiences in a subsequent analysis of the same foods, presence of adjunctive income eligibility for dataset. Researchers attributed the problems encoun- AFDC participants, value of the first trimester hemo- tered in attempting to control for selection bias to the globin level used to define nutritional risk, number of limited number of variables available in the adminis- WIC clinics per 1,000 low-income persons, number of trative (Medicaid and WIC) and birth certificate data WIC clinics per 100 square miles, AFDC guarantee for included in the WIC-Medicaid database. a family of four, and average Medicaid expenditure for Gordon and Nelson (1995) used the NMIHS, a nation- a family of four. ally representative dataset that includes information on Like previous researchers, Brien and Swan estimated the characteristics of women who gave birth in 1988 several models with different combinations of instru- and their offspring, to study the effects of WIC. With ments. They, too, found that results were very sensitive access to a much richer data set, Gordon and Nelson to model specification. In some cases, results showed a were able to control for many more covariates in their negative association between WIC participation and basic model, including income and use of cigarettes, birth outcomes, although the differences were not sta- alcohol, marijuana, and cocaine. They also had more tistically significant. Moreover, the sensitivity to options for variables (instruments) to include in selec- model specification varied substantially by race, sug- tion-bias-adjustment models. They estimated several gesting that the instruments used did a better job of models of the effect of WIC on birthweight, using predicting WIC participation among Blacks than various combinations of the following variables: per among Whites.

96 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Brien and Swan also estimated a fixed-effects model, Birth Outcomes: Research Overview using a sample of women who had at least one birth The literature search identified 38 studies that exam- before the 1988 NMIHS birth. This approach assumes ined impacts of prenatal WIC participation on a vari- that critical unmeasured differences between WIC par- ety of birth outcomes.54 The outcomes most frequently ticipants and non-WIC participants are mother-specific studied were mean birthweight and likelihood of low and do not vary over time. The analysis examined birthweight (defined as an infant weighing less than whether differences in WIC participation status for the 2,500 gm, or 5.5 pounds (lb)). Other birth outcomes two births affected outcomes.53 The fixed-effects included mean gestational age (length of gestation at model yielded results that were generally smaller in time of delivery), and likelihood of very low birth- magnitude and stronger in statistical significance than weight (less than 1,500 gm, or 3.3 lb), premature birth the two-stage model. For Whites, findings for the inci- (generally defined as birth before 37 weeks gestation), dence of low birthweight were completely different (a intrauterine growth retardation (IUGR) or being small- negative but statistically insignificant effect) than find- for-gestational age, neonatal mortality, and infant mor- ings for the two-stage model. tality. Several studies also examined the impact of pre- As the preceding discussion illustrates, the problem of natal WIC participation on Medicaid costs associated selection bias has proven especially thorny in research with delivery and newborn care (up to 30-60 days on birth outcomes. Some researchers who have inves- after birth). tigated WIC impacts on other outcomes or for other Selected characteristics of these studies are summa- WIC participant groups have reported success in con- rized in table 17. The 38 identified studies can be trolling for selection bias. This has not been a univer- divided into four groups based on scope/generalizabili- sal experience, however, and many of these researchers ty and general methodology. The two national USDA- have also struggled with limited candidates for identi- sponsored WIC evaluations, although substantially dif- fying variables and with models that produce inconsis- ferent in design, make up Group I. The strongest of the tent, implausible, or unstable results. two, the NWE, includes two different components— the Longitudinal Study of Pregnant Women (Rush et Impacts of WIC al., 1988d) and the Historical Study of Pregnancy Prenatal Participation Outcomes (Rush et al., 1988a). Although the NWE is the most recent national evaluation of the WIC pro- The prenatal component of the WIC program is, by gram, it is based on data collected in 1982 and 1983 far, the most studied part of the program. The vast (Rush et al. 1988d) and historical data from the mid- majority of the research in this area focuses on impacts 70s through 1980 (Rush 1988a), and is therefore on birth outcomes. Substantially less research has been quite dated. done on the impact of prenatal WIC participation on the initiation of breastfeeding. Even less research has Group II includes nine studies that used national sur- examined the impact of prenatal WIC participation on vey data, almost always the NMIHS, to examine WIC the women themselves (for example, on women’s impacts on birth outcomes. Although some of the dietary intake and/or nutritional status). A small num- research that used the NMIHS was completed recently, ber of studies have examined the relationship between all of it is based on births in 1988. One study in this prenatal WIC participation and child development out- group (Kowaleski-Jones and Duncan, 2002) used data comes. Because several of these studies look at WIC from the 1990-96 waves of the National Longitudinal participation during infancy or childhood, in addition Survey of Youth (NLSY). to prenatally, these studies are discussed later in this chapter-in the section that deals with impacts on Group III, the largest group, includes 15 studies that infants and children. linked State-level files of WIC participant information with other State-level files, generally vital statistics

53The assumption that key unmeasured maternal characteristics do not vary over time is a generous one. There is no guarantee that maternal 54Several very early unpublished papers and reports included in a review effects on a pregnancy—for example, the mother’s general health, use of prepared by Rush and colleagues (1986) for the NWE are not included cigarettes and alcohol, weight gain, and diet—are constant over time, and because they could not be located. Given the age of the data and the the NMIHS had relatively limited information on characteristics associated descriptions included in Rush’s summary, it is doubtful that these documents with earlier pregnancies. Moreover, there is no guarantee that women who would add anything to the present discussion. Most, if not all, of these had two births are representative of all prenatal WIC participants studies appear to have centered on cross-tabulations that were subjected to (Besharov and Germanis, 2001). few, if any, statistical controls.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 97 Chapter 4: WIC Program n n n n od h Continued— gressio gressio gressio gressio ts te re te re te re te re a a a a cos e r Analysis met Multivari Multivari Multivari Multivari x e d health ca f + ind o mmy mmy 6 n h u u g s ciate 6 months) n h i t l C l g I o on d on d n r e n l ants wit e g Measure of participation y ong-term ( n l i y w C penetratio r I e a participation Participati with short- (< and l participation participants vs. particip months) W N Participati v W s s er p alysi t vs. ts, t vs. pant pant mes, including asso n n n o WIC g ate grou Design partici partici etration ov Participa non Participa before vs. after, separ Participa non time to birth outcomes relatin program pen Trends an e l n birth outc C b 2 I e i re ) that ding o inics nant C v a n ts g i e I e atal i n W l - un e n en - rths i o preg e latio ast 22 pants 3 statio ho w n alth cl 4) tative m ted e tals n i d o ants wh a gate dat is) ally ng pr ipation o n c nic with l a n eton b Popu ,000) ,935) atally i ic he partici ite, Black, or 1 3 12,81 (sample siz C participan C d rtic h I I n publ or hosp (n= receivi were at le weeks g (n= care in surro non Postpartum W particip particip pren (n~ singl women w Hispa W N/A (Aggre analys sample of W Nation represe a W C pa in 5 on ery 3-84) 1 llecti 198 es in livery, tions natal WI in 14 i deliv ( o t n y every 3 d DC l de ics l S n pre f H I M C sites and 5 ce after I Data source n N lment, oximate 3-76) 2-80) atal cli 8 8 pact o 9 m Primary data c 1 and o months unti (197 appr in 19 WIC sites States. Data were collected at time of WIC enrol Vital statistics records for 1,392 cou 19 States an (197 Record abstrac 174 W pren surveys onal and and fant , , , , w w n e e e g g g a a nd i l l t, very low t t, and a a amined the i h h h e birth e birth ght, ght, ght, is of nati nal a n n g g g od of lo i i i od of lo od of o o i i e ex t t t iho w Outcome(s) a a iho iho natal a t t h s s t r i e e Birthwei Birthwei g Birthwei g gestatio birthwe b prematur neo mortality rates birthwe likel likel prematur fetal mortality rate Likel of table. udies tha ) ) t E E at end 03) II: Secondary analys I: National evaluations en et al. 8a) (NW 8d) (NW 9) See notes Table 17—S Study Group Rush et al. (198 Rush et al. (198 Edozi (197 Group Finch (20

98 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program n od eity h ction- acks and Continued— ltan t models nants of e s model; l s model ls l u gressio s e is termi te re te ral se e a a ustmen e nalys dj s ate mode Analysis met e t i h (1) Multivariate regression Multivari bias-a (2) Fixed-effect separ Logit a estimated for B (2) Fixed-effect (1) Multivariate regression, including attempt to control for sim and sev W birth outcom regression analysis to identify d Multivari d 1 for ing first mmy mmy mmy ever r d n status ancy u u u ut u n o n ciate g on d on d on d ted d ted a id a a d witho Measure of participation Participati Medica with an Participation dummy Participati for each pre trimester (2) Participati particip (1) Participation dummies: 1 for particip Participati t vs. t vs. t vs. t vs. t vs. pant pants pant pant pant mes, including asso n n n n n o Design partici partici partici partici partici Participa non Participa non Participa non Participa non Participa non g 0 0 ) n birth outc C C 2 I I 4 re ) k n en en r to blin 45) 2 e e n e W W e e e me- h at 1,984) l l dr dr - - 1 4 g h , , n 199 n 199 t b b n n i i - 8 8 n= o o al ris g g ,254 n g th at least latio od = i i i ho w n n blin l l h prio anic anic 6 o n n i e e ( n twee twee r d d - - sp sp e e u n n C and i i C and n e e e nic wom 996, wit 996 ( I I l d a a e Popu ,796) ,778) m m b 8 (n= i 7 7 453 si n m o o (sample siz C and inco C C rticipation o r g I I I c c i o l o n n (n= women (n= Hispa e pairs) born b and 1 same peri (n= w (2) NLSY chil born b and 1 least 1 other si at nutrition b 1 live birt 198 i non-H women w (n= i non-H women w pairs of births) (2) W W (1) W (1) NLSY chil W W C pa 1 natal WI SY S S pre L f H H I I M M Data source N N 8 NMIHS 8 NMIHS 8 8 0-96 N 8 8 pact o 9 9 m 1 198 1 198 199 th and and fant very , , w w 4 w ality n d o e n ght a nd i t and t amined the i h h e birth e birth ne gr eemi wei on, nued ght, ght g g od of lo od of i i od of lo e ex atal mort t iho iho w Outcome(s) iho natal a h t r i birthwe retardati prematur heavy pr low birth b Likel rate Likel intrauteri Birthwei Neon Birthwei likel prematur neo mortality rates of table. er k udies tha v t 9) at end n d Par nd unca n (199 n an 0) 7) 2) 8) See notes Table 17—S health care costs—Conti Study Kowaleski-Jones and D Hoga (200 Brien a Swan Moss and Car Frisbie et al. (199 (200 (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 99 Chapter 4: WIC Program e ls . , ent. s n n ght n n a i ons of b od oups - h n AFDC o is. Continued— i ls for LBW inati r Whites, as t alternativ ate mode s cks ght and ceipt of alysis gressio e b gressio gressio gressio c a ntrol for 5, 6 e ral l aly as adjustm e nd/o e te re te re te re te re s a a separ a a eity. ght an ent. Separate and re ites. n-bi g id a d h vs. normal wei n on com ate mod e i Analysis met ogit an d d d u l c n Multivari Multivari Multivari Multivari Attempted, but rejected, selectio i models to co simultan for Blacks and well as sev adjustm models for Bl and W vs. normal wei VLBW Separ for each of 4 subgr base income Medica inclu and l Birthwei r e d mmy mmy mmy men ecific u u u C per n p ciate I e m on d on d on d o ant wo w d in W Measure of e participation l le b i g i l enrol State-specific numb of pregn 1,000 State-s e Participati Participati Participati g t vs. t vs. t vs. pant pant pant mes, including asso n n n o Design egate data partici partici partici Cost- effectiveness study usin aggr Participa non Participa non Participa non birth record files n birth outc C C 2 is s I I ) ents s nts e sis an e n W W n me- pi - - e n n unti o o m latio ho n n o 99) reside w d d id reci s with ite analy d not n n e e 57 co 0+ h l a a Popu ,170) ,905) atal care b i 6 3 295,5 (sample siz C C and inco C rticipation o g I I I i l 50,00 Data for 677 counti for W and 3 W with 5,000+ Blacks for Black analy Medica who di participate in high- risk obstetrical program (n= e African Americ women w received some pren (n= (n= W W C pa y r hed with Medicaid and/or a c u id, n 1 a J ta for the natal WI a n nd of 0 n hs in 0 e iles mat s i e e e S us d pre w C, Medica t f H I I e er 20 l statistics b M ounti W Data source N a d d i 7 Cens 6 and th 8 NMIHS 8 r 8 pact o o l 9 m U.S. Decemb 199 197 Linke F records for birt and vita large c 1 198 WIC participation f and fant very , , w lity, w w ality n e g ght a al , nd i l t t, very low t, very low t and t, and a amined the i h h h h h e birth 7 wei nued ght, n g g g g g od of lo od of lo i i i i i od of lo onat o i e ex t atal mort ho t i iho w Outcome(s) a iho natal morta natal a t h t s r i e Likel rate Neon Birthwei g birthwe neo postne mortality, infant mortality b birthwe birthwe low birth prematur neo mortality rates birthwe likel Likel of table. 5) udies tha t 04) 99 at end 95) State-level studies using nd III: ton (1 (19 g n 8) See notes Table 17—S health care costs—Conti Study Covin Gordon a Nelso Joyce et al. (198 Group Roth et al. (20

100 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program g 12 , 10 . , 8 ent. in d n n n n n eity eity. od eity h l Continued— ltan ltan includ ls for ltan ysis, a rejecte gressio u u gressio gressio gressio e gressio u dels to al 11 as adjustm te re te re te re te re te re alysis alysis, a eity a a a a n-bi obit an ate mod ative mo Analysis met Multivari attempts to control for simultan Attempted but selectio Probit an Multivari including attempt to control for sim and pr including attempt to control for sim Probit an Multivari Multivari Multivari Blacks and non-Blacks Separ including sever altern control for sim d 9 C mmy mmy mmy mmy mmy : I ” n atal onse: ed by 30 u u u u u e ciate p on d on d on d on d on d on statio on W pons ge of e n se-res Measure of o participation gth of pren C “exposure I Participati Dose res Len W Participati Participati dummy: Enroll weeks g Participati Participati and d Percenta gestatio Participati t vs. t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant pant mes, including asso n n n n n n n o Design partici partici partici partici partici partici partici Participa non Participa non Participa non Participa non Participa non Participa non Participa non n birth outc C C C C C C C ) 2 I I I I I I I ) led led on 4 ents ents ents ents ents ents e e e n W W W W W W W 1 ol ol pi pi pi pi pi pi et ------n n n n n n n irths e o o o o o o o ngl latio 19,6 n n n n n n n 58 ) 2) 58 ) 5) 0) 58 ) e enr e enr d d d d d d d id reci id reci id reci id reci id reci id reci n n n n n n n natal car natal car ad liv a a a a a a a eton b Popu 111,9 53,78 111,9 42,96 21,90 111,9 (sample siz C C C C C C C rticipation o I I I I I I I W Medica (n= W W women with full- term births (n= W W W Medica Medica who wer Medica Medica with live si births (n= who wer (n= in pre and h singl (n= in pre (n= W Medica (n= C pa e id, id, id, g r 1 a h files for files May New c mber natal WI record, an 1992 d birth d d n s n North i h e ig a ce d e e l lin pre rths in a i C, Medica C, Medica C, Medica C an f t I I I I i p d deat s W W W W rth recor rth recor Data source i i o d d d d ina 7-88) 7-88) 7-88) h 7 births in 2 and D 3 d pact o n m FNS WIC/Medicaid (198 record files for births in Mich 199 Linke Carol Linke and b Linke and b for 1988 b Linke FNS WIC/Medicaid (198 birth an 199 North Caro 199 a files for births i Jersey betw FNS WIC/Medicaid (198 y and r fant very very , , lity, w w , , s’ e w n e v g f ght ght ght ght a , o nd i l t t, very low t and t and ei ei a amined the i h h h h e birth d w w wei wei nued ght, ght, n g g g g o od of lo od of lo od of od of i l stay, i i i od of od of lo o o fant mortality, i a e ex t h n h of infant t i iho iho iho iho w Outcome(s) a l iho iho natal morta natal a t h e t s r k i i e L low birth birthwe Medicaid costs hospit low birth birthwe Medicaid costs birthwe Medicaid costs Likel low birth Likel Likel neo mortality neo and i low birth Birthwei g b lengt Birthwei likel Likel likel prematur Medicaid costs of table. ) udies tha ) t 2 3 9 3) 9 at end d 1 ( 000) a et al. y er et al. er and li e a us (200 n 8) 3) 0/91) a v See notes e Table 17—S health care costs—Conti Study Gregory an deJes Buesch Horton (2 Ahluw (199 Buesch (199 Devaney and Schirm (199 D Devaney et al. (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 101 Chapter 4: WIC Program n n n e od h Continued— gressio gressio gressio covarianc te re te re te re a a a Analysis met Multivari Analysis of Multivari Multivari d hers mmy mmy mmy C on sity” I c u onse: onse: u u ngth n ciate u s sed e of e on d on d on d a vel W ancy e se resp se resp le b ed for le Measure of o o participation nditur emed vo ar valu C food costs I Participati Doll rede and d Participati Program “inte W adjust of pregn and d variab county-l expe Participati s er e ps ntrol ne) alysi t vs. t vs. t vs. pant pant mes, including asso n n n grou id, o WIC g nce Design ed on th partici etration ov partici Participa matched co time to birth outcomes Participa non relatin pen Trends an within 3 Participa non basis of insura coverage (Medica private, no defin n birth outc o C C h 2 t I I t ) s i 13 ents ) aring ) s en e n W W w s pi h - - r t ed n births i C and y n n r I i t dent a i o o l b latio i p n n 94) leto b ovid n i 1 d d s, all of id reci C wom r county o s I 1 n n t e e n 4 a a e Popu , ,546) atal care l o 9 8 132,9 g p (sample siz C and other C C rticipation o I I I s n = i e n Data for 75 (of 100) counti W which pr W pren services for all county resi (rather than sh r Medica anoth non-W with sing ( (n= Matched W S W women (n= C pa nd a na, id, i 0-85) 1 s for d atistics, ty-level 1988 t rths f natal WI and e n n ucture spital h record Carol Missouri a u s for births etration e c s (198 2 bi i eat pre rd fil e rk State in the i h d ho C, birth C, Medica C, birth and f o I I I o p g vital s a ge fil r W W W gate co Data source in g d d d nditur d o 2 births in m pact o e m Linke birth record, hospital care, and d inclu d in Missour expe data for North Aggre files for 198 service infrastr characteristics, program pen Linke death rec 198 Linke record, an dischar in New Y last 6 months o and , and , , , , lity lity caid w w w e e e i g g g ght a a a , , l l t t t, very low t, very low t, and a a amined the i h h h h h e birth e birth wei nued ght, ght, ght, n n nal- g g g g g od of i i i i i od of lo od of lo od of lo o o i i e e ex t t t iho w w Outcome(s) a a iho iho iho natal morta natal morta t t h h t t s s r r i i e e Birthwei neo rate, and Med b costs Birthwei Birthwei likel g g low birth birthwe birthwe prematur Medicaid costs prematur small-for- gestatio birthwe b neo likel likel Likel of table. ) udies tha 6) e t 8 at end 198 er 0) 7) See notes Table 17—S health care costs—Conti Study New York Stat (199 Simpson ( Stockbau (198 Schramm (198

102 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program , n n e. e od h es for Continued— s 16 ite, and arisons gressio h aly mp covarianc covarianc o te re te regressio p. a a ate an Analysis met ite, non-W h Multivari Multivari including attempt to control for simultaneity Bivariate c Analysis of W Separ Analysis of total grou n o i t d a p i c mmy mmy mmy mmy mmy nancy i e of t onse: u u onse: onse: u u u ngth C and r C I ciate p I a p W f on d on d on d on d on d o d ancy ar valu e n se resp se resp se-res ll ed for le Measure of C o m o o o o nt of preg participation i I ons t e a e W r C food costs I d u n e Participati Participati o Participati Participati Participati Months on W W adjust of pregn and d perce and d and d D and d r coup C I ntrol 14 non- t vs. t vs. t vs. t vs. ts vs. pant pant pant pant mes, including asso n n n n n n-W o Design ps: p of nonpartici partici partici partici C births C births; I I Participa non Participa matched co Participa non Participa non grou (1) all no births; (2) random sample of W (3) matched grou W Participa 3 different non n birth outc C C 2 ton I I ’s ) l for 15 ) liv- nts ents n- o e a ed” ) l en e a n ; 7 e W W e s pi g - - 1 de r n births n births c C i i C and n i n pit I 1 I a , o o n latio p W 0 ho d n n a leto leto 9 erserv p 6 2 d d id reci C wom ants wh s = ive sin I i 2 3 n n C and no n I 1 7 a a ( H Popu ,628) , ,707) , - 7 4 4 6 h n (sample siz C not reported) C C rticipation o t I I I = = r o i n n W All W ( non-W with sing N WIC HealthStart particip had a l Matched W Medica b primary hos ered at the ar (n= the “und (n= W sample for non- W women w with sing Missouri res ( C pa pi- 988 id, 1 ram for ta s for birth, and na hos natal WI g a rds een 1 e ital care 1989 nuary caid Missouri a w Jersey’s i o i ates for sp d o une pre rd fil rths in i i C, birth, C, birth, and C, Medica f I o I I en Ja dize l records, d in Ne W W W Start pro Data source eath rec 996 d d d ents betw le nant Med 0 births in 8 and J pact o m Standar collected for women enrol Linke Health preg recipi birth, and h Medica records for 1980 births in Missour and 1 death certific births in 1 Ind 198 and d tal betwe for 1978 b Massachusetts Linke Linke death rec 198 d , , lity sta- ihoo onal- neo- ght, w w w w e el g ei a lik , l t t t, and t, and estati a amined the i h h h h e birth nued ght, ght, ght, ght, ght, ge n g g g g i i i i od of lo od of lo od of lo od of lo o i e e irth, and ex t t w w Outcome(s) a iho iho iho iho natal morta t h h t s t r r i i e Birthwei Medicaid costs infant mortality rate b birthwe Birthwei likel Birthwei g likel Birthwei tional age, of low birthw prematur small-for-g b natal mortality rate Birthwei age b likel neo rate birthwe likel of table. r State and local studies udies tha 5) e t h nd 3) at end er an a IV: Ot 6) 6) See notes Table 17—S health care costs—Conti Study Stockbau (198 Schramm (198 Kotelchuck, et al. (1984) Group Reichm Teitler (200 Brown et al. (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 103 Chapter 4: WIC Program n n od h Continued— ance s gressio gressio vari te re te re a a Analysis met Bivariate t-test Multivari Analysis of Multivari d mmy mmy mmy cy WIC : er’s ckups on acts, nths i u u u n e ciate nt gna on d on d on d pons d, nutriti ucher p on moth Measure of participation le g pre ation co d Dose res educ and vo Number of mo enrol Participati base participation in durin Participati Participati ed ts, t vs. t vs. pant pant mes, including asso n n n ent o omiz Design partici partici Participa before vs. after Participa non Participa non experim Rand n birth outc C 2 o I d ) t t bers o with ancy 17 e size s and caid- en en e n W n s d - gh g e- a ar e n C t nts and I b o ents c t t; roughly i latio ies n y h h W d ast 1 c erag l g g icted d e i i ants wh ent num infa n 17 ye r l a d n us pre e t e a p < a a ers up to 20 a Popu ,907) nant wom nant wom n w es vs. small or e ibl n 1 217) 410) 519) g r (sample siz C and Medi C rticipation o d e I I e e r r n W elig a measurem toddl months of age at least 2 hei had at le previo (n= selected at mid- p W Income-eligible preg P (n= (n= preg particip were on pre birthwe equiv w have av babi large bab (n= C pa on ) atal 4 1 ) llecti ics in 6 tal in natal WI clinics in 1 county 9 o pren on ties cal 83-8 n s C pi n I d 7-8 o pre artment in 3 WIC clinics ealth o f a (19 h r on-W 1 hos on at a o dep n Data source b h blic h 9-81) 0-81) ama cou h artment cli g C records in C and medi i pact o I I e m (197 Texas (198 Primary data c Primary data collecti clinic i (198 Oklahom W healt in pu dep Alab W records in and 4 n the same Bost n amined the i nued ght ght ght ght ex t Outcome(s) Birthwei Birthwei Birthwei Birthwei of table. udies tha t er et at end ng 4) ndi 1) 5) 5) ins et al. See notes al. (198 Table 17—S health care costs—Conti Study Mays-Scott (199 Coll (198 Metcoff et al. (198 Heime

104 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program d) n n od h means Continued— f arisons s (no gressio gressio o on mp o te re te re a a isons oporti Analysis met Multivari and pr statistical tests reporte Bivariate c Multivari Compar d hers ed mmy mmy mmy c hs v u nths onse: onse: u u u ciate on d on d on d , mont d rs recei se resp se resp e Measure of C o o participation I W n N/A Participati o Participati Number of vo receive and d Number of mo vouch and d Participati ntrol t vs. t vs. t vs. t vs. pant pant pant mes, including asso n n n n o Design partici partici partici Participa non Participa Participa matched co non Participa non - i n birth outc 2 c 30 re ) t at i al cy of t e e en e n n r - n me- me- 18, 19 an a e C for at C and I p I d enat ng m irs) n-WIC n gna eeks latio 18 ho w W o egn o wer a o ivi cai w n i e d C pairs no I al pr n e e e l l a g pre Popu ,297) nant wom b b ibl i i 418 p 1 114) (sample siz C and inco C and inco C rticipation o g g I I I i i l l W on Med least 16 w durin (n= W e elig women w W (n= preg (n= care (n=101) e non-W time of recruitment and rec identic pants wh weeks pr Matched W C pa n- on ble 1 C land, 2 I llecti facilities eas rted) y natal WI tts in which 1982) r sites and data from in o h c areas of o cal cal W rds in 4 e s C o I e alt d 2 phi on- pre hic a f e an C he s in Mar C site and 1 no h WIC was ogra I I faciliti alysis of e Data source grap edy et al., h abl 9-80) 3-78) 3-78) C was not availa C and medi C and medi pact o I I I m Primary data c records in WIC sites and n W healt WIC site in Florida (Dates not rep at 1 W Medicaid rec counti W (197 in whic avail of Massachus (197 4 geo records in W W non-W in 4 ge Massachusetts (197 (Rean Kenn , , w w e e d g g n a a a - l l t t a a amined the i h h nued ght, ght, ght n n g g i id costs i od of lo od of lo on o o care i i e e ex t t h t w w Outcome(s) a a iho iho t t h h s t s t r r i e i e Medica utilizati b Birthwei healt Birthwei likel Birthwei g birth, and fetal death rate b likel small-for- g of table. udies tha t at end 3) nd al. edy, edy a e (198 3) See notes Table 17—S health care costs—Conti Study Kenn Bailey et (198 Paig Kenn et al. (1982) Kotelchuck (1984)

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 105 Chapter 4: WIC Program r ring C 3, n s. du k s od of s, all h at non-WI gressio nd 40 wee who enrolled afte uring first 7 te re participant fetal growth s a complication rolled at 21-37 al care s, a ) d be twice that k ific y or Analysis met c s t c prenat n gestational age (29, 3 x-spe Multivari be Medicaid emies ma 8 months; (2 eks, 36 wee riables. ere not in WIC. an timing of WIC enrollment. e of se n to based o on, and low = en group. r birth defe va d w 2 we fo vy pre rts, ve who received stati mmy her th and w el with WIC participant u cipant ks, 3 ciate during first coho ks ge were stational ag re noted on d y. ester) rat 28 wee asic mod our of the fi or women f Measure of m on. . or ge participation thers e nonparti e stud -20 wee separate l ces we ality rate for hea tri ati n f b nd st rtum, a l gible for WIC or kno in th i ree Participati eren a mort hed on ational age: (1) v resholds: or 2 c a st l age, (2) b at ks ge by mea whose mo est (1 s p ort th ation th ationa ere mat care and g 30 wee tion or later st bjects = enrolled at 12 st ts, re income-eli group, (2) th e mes, including asso of women in the n ncy weight. ation was su o significant diff tional age nt esta ate grou Design weight rce ainder w cipant ks g who we (Authors rep Participa before vs. after, separ by length of g . cipation and ge s if the inform delivery. No The rem rthweights of nd pre-pregna ancy, enrolled in WIC postpa ation and (2) by s cipants i e. station, medium eight at gesta cts on birth u d e nonparti st a about 75 pe n birth outc gn 2 7 weeks. alysi ) fined ed b cy, r e e n s ge me- come. eks ge k ncy an d on initiation of prenatal rts d latio pants s of impa ved birthw r stational ag compa ester WIC parti e l t coho ed base m Popu ,514) uring their pre atus. tus, and in b nalyse ore 12 we re 32 wee hors partici ring pregnan and marital stat (obse ailable for only i for ge 2 t on included in th enrolled in WIC at 36 wee st (sample siz C and inco v rticipation o t a g fo u I ed for in the an rict ren t-tri cipants. i ring pregna s limited to nonparti on of less than 3 l cy, ho (n= non W e stati re a rcen ing du stati rital sta parti k C pa , the a r controll egnan oup wa n r our diffe ks ge r , A ity, ma P listed for WIC d on g in WIC (1) be e data we 1 , s for f mple was rest simultaneity in s this p ncy, smo it- howeve gh = enrolled bef par d in y s ) and t ht that controlled a a ded women w , of ter for WIC sa natal WI ari initiatio 4 generally d n complications du h; more and ge e ss than 85 pe 7 u s ea clu r d utcom r egn d Infant lle ed hi birth o but rol for r 77) ducation, and marital ere w pre C of le e model f comp ect (MIC) in 971- ce, age e wa y ho enrolle n e Data source ction, an C h rolled in WIC. conside I om sampl for birthweig se. n g -age bias, l with addition of control for firs W e pact o as l es that a ed on ra births this pr l nt group w r e hose w f owth ratio lled in WIC after 33 wee ce, parity, e C-se model that in come. Birth o children’s growt odel r significantly g e of 2,500 gm o t m up. Incom women enro before (1 after (1974- o Medical records for rand Maternity an Care Proj A , v s er o s s rol variable. t r nro cipation to cont out atch c on e, ra s a well as separat wa c mode eight s stational by cont si w parti s but ne ison g r ) ba w nfant Health Survey. icipate in WIC w y sample indicat rol for ge area on ag pact of WIC t men were m participant the nonparticipa ucation, gravidity, number of ucation, numb rt efined as fetal g ple varie on birth defe women who e compa mated alternative amined the i h l age as a els: (1 d d sic model a ud Also estimated m of C-section ed as birth s nued ght, ants. g ent wo i od of lo ighborhood s im m e ex e t ce, e ce, e w Outcome(s) iho nths. el, esti h articip t y wa rdation d r i me n estationa The rate b Birthwei likel a timated ba al Maternal and I s sample; sam caid = FNS’ WIC/Medicaid databa preemie defin th reta in s men who did pa applicable. within catch on age, ra on age, ra 80 percent of asic mod alternative mod els included (1) s, e dels defined WIC w s of stud rate model to cont nd (3) g re included in the g first 6 mo vy o udies tha elivery. ered nonp ht infants). wo r t cu s: id ilities hed hed hed ea description of the st s we c c c urin sepa s), a ce s r k e for imately r con x station. N/A = Not re fac (3) d cy and d a ks FNS WIC/Medi NLSY = National Longitudinal Survey of Youth. NMIHS = Nation In addition to b Alternative mo Estimated two Pairs mat Pairs mat Pairs mat Included The main fo WIC-eligible women included in Appro Data sou Unless the Maximum analysis Intrauterine g Used three alternative definitions of WIC For all outcome Authors also examined impact Alternative mod Exposu Notes: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 income level and 37 wee normal birthweig distribution). H months; weeks ge Table 17—S health care costs—Conti Study Silverman (1982) 36 wee health c pregnan

106 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program files and Medicaid files, to study birth outcomes gestational age, influence of the comparison group among WIC participants and nonparticipants. With used, and use and adequacy of prenatal care. three exceptions (Devaney et al., 1990/91; Devaney, 1992; Devaney and Schirm, 1993), all of the studies in Simultaneity of WIC Participation and Gestational Group III are based on data from one State. The three Age. Women who deliver early have less chance of excepted studies used the FNS WIC-Medicaid data- enrolling in WIC. Women who go to term have a base. This database was assembled by FNS to address greater chance of enrolling. Consequently, both the a congressional mandate to determine “savings in decision to participate in WIC and the length of WIC Medicaid costs for newborns and their mothers during participation are inexorably linked with gestational the first 60 days after birth from participating in the age, an important predictor of most birth outcomes. WIC program during pregnancy” (Devaney et al., This simultaneity means that assessments of WIC 1990). A secondary objective for the database was to impact that rely solely on a binary indicator of partici- examine effects of participation on birthweight and pation are likely to overstate the impact of the pro- gestational age. The FNS WIC-Medicaid database gram. Moreover, because the duration of WIC partici- includes WIC participation, birth certificate, and pation is also simultaneous with gestational age, a tra- Medicaid claims data for five States (Florida, ditional dose-response approach—estimating WIC Minnesota, North Carolina, South Carolina, and impacts based on number of months of WIC participa- Texas). For the first four States, the database includes tion—although employed in several studies summa- data for all births in 1987. For Texas, the database rized in table 17, is not a satisfactory solution to the includes data for all births during the first 6 months problem. of 1988. Gordon and Nelson (1995) studied several approaches Most of the research in Group III is based on data col- to addressing the relationship between the timing of lected in the 1980s. However, four studies are based WIC enrollment and gestational age (pregnancy dura- on more recent data. Roth et al. (2004) analyzed data tion). These included omitting very late enrollees for Medicaid births in Florida from 1996-2000;55 (enrolled after the eighth month) from the WIC group, Gregory and deJesus (2003) analyzed births in New including gestational age as an independent variable in Jersey for an 18-month period in 1992-93; Buescher the regression, and defining several cohorts of WIC and Horton (2000) used 1997 data from North participants based on gestational age (pregnancy dura- Carolina (this study is an update of a previous study tion) at the time of WIC enrollment. All of these conducted in 1988 (Buescher et al., 1993)); and approaches decreased estimated impacts to varying Ahluwalia et al. (1998) used data for 1992 births in degrees. Gordon and Nelson ultimately concluded that Michigan. each of the approaches to controlling for simultaneity systematically underestimated the impact of WIC Finally, Group IV includes 11 State or local studies because they effectively eliminated any effect WIC that examined WIC impacts among pregnant women might have on extending gestation. The authors sug- receiving care in particular programs, hospitals, or gested that results from analyses using a binary indica- clinics. All but one of these studies (Reichman and tor of WIC participation (participant vs. nonpartici- Teitler, 2003) used data that were collected in the pant) and those comparing various cohorts of WIC 1970s and 1980s. Reichman and Teitler used data that participants (in an effort to control for simultaneity) were collected between 1988 and 1996. probably bound the magnitude of the true effect.

Methodological Considerations Influence of the Comparison Group Used. Research Before reviewing findings from the studies presented has consistently shown that specific types of women in table 17, it is important to understand three are more likely than other women to participate in methodological considerations that, in addition to WIC. Characteristics associated with increased likeli- selection bias, affect interpretation of research on hood of WIC participation include younger age, lower birth outcomes: simultaneity of WIC participation and income, lower educational levels, being unmarried, and being African American. Several early studies of the impact of WIC on birth outcomes attempted to 55 This study was first presented in 2000, with a subset of the data (Roth control for these differences by creating matched pairs et al., 2000). Just before this report went to press, the author provided an update that includes data for the full 5-year period (1996-2000) (Roth et of participants and nonparticipants (Kennedy and al., 2004). A manuscript is currently in preparation. Kotelchuck, 1984; Kotelchuck et al., 1984;

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 107 Chapter 4: WIC Program

Stockbauer, 1987). Matching was limited, however, to inappropriate because prenatal care and WIC participa- variables that were available on birth certificates, most tion may be simultaneously chosen.56 often maternal age, race, parity, education, and marital status. Researchers were unable to control for other Birth Outcomes: Research Results important variables, particularly income and key char- Table 18 summarizes findings from the available acteristics related to pregnancy, and generally did not research by outcome. Results for each study are reported do so in their analyses (for example, analyses for using the primary author’s name. In the interest of pro- Kennedy and Kotelchuck (1984) and Kotelchuck viding a comprehensive picture of the body of research, (1984) were limited to chi-squares and t-tests). Thus, both significant and nonsignificant results are reported the comparability of treatment and comparison groups in table 18 and in all other “findings” tables included in in these studies is still open to question, despite the this report. As noted in chapter 1, a consistent pattern fact that the groups were “matched.” of nonsignificant findings may indicate a true underlying effect, even though no single study’s results would In interpreting findings from these studies, it is impor- be interpreted in that way. Readers are cautioned, how- tant to realize that, to the extent that comparison- ever, to avoid the practice of “vote counting” or group women were higher income or less at-risk than adding up all the studies with particular results. WIC women, the true impact of the WIC program (at Because of differences in research design and other the time these studies were conducted) may have been considerations, as discussed in the text, findings from underestimated. some studies merit more consideration than others. In 1985, Schramm studied the impact of WIC on For the first two outcomes in the table (mean birthweight Medicaid costs for newborns in Missouri. By limiting and mean gestational age), a higher value is associated the analysis to Medicaid recipients, all of whom were with a positive WIC impact. For the remaining out- income-eligible for WIC, Schramm created a ready- comes (for example, the likelihood of low birth- made comparison group and minimized (but did not weight), a lower value is associated with a positive eliminate) the potential influence of noncomparable WIC impact. The column headings in table 18 vary incomes. The approach used by Schramm has been accordingly, so that significant positive WIC effects adapted and used by many other researchers, most are always shown in the far left column of the table. notably in the USDA-sponsored WIC-Medicaid studies (Devaney et al., 1990/91; Devaney, 1992; Devaney and As noted in the preceding discussion, all of the avail- Schirm, 1993) (see Group III in table 17). In interpret- able studies have limitations that require that their ing results of these Medicaid-based studies, it is impor- findings be caveated. Thus, no single study provides a tant to recognize that they are limited to the lowest definitive answer on WIC’s effectiveness, but the body income WIC participants. At the time these studies were of research provides suggestive evidence. As table 18 conducted, WIC eligibility was defined as 185 percent illustrates clearly, the majority of studies reported posi- of poverty, while Medicaid eligibility was generally set tive differences that favor WIC participants. In most at 130 percent of poverty or lower. Because lower cases, differences were statistically significant. income women are at higher risk of poor birth outcomes, these studies probably overstated the impact of WIC. In 1992, the General Accounting Office (GAO) completed a meta analysis of existing WIC studies that Use and Adequacy of Prenatal Care. Receiving ade- yielded estimates of cost savings attributable to WIC quate prenatal care is expected to independently affect (GAO, 1992). The meta analysis included 17 studies most birth outcomes. Consequently, most recent of WIC impacts on rates of low birthweight that were research has controlled for the adequacy of prenatal care in order to estimate the independent effect of 56Some early studies included prenatal care use and/or adequacy as sep- WIC—that is, the impact of WIC over and above the arate outcome measures. While most of these studies found positive associ- impact of receiving adequate prenatal care. However, ations between WIC participation and measures of prenatal care, these esti- because encouraging prenatal care and potentially pro- mates have largely been discounted because cross-sectional studies can not disentangle the direction of the effect. Higher rates and quality of prenatal viding a link to such care is a major focus of the WIC care among WIC participants may result from either WIC referring women program, including adequacy of prenatal care as a to prenatal care or prenatal care providers referring enrolled women to covariate effectively understates the full impact of the WIC. Because of this limitation and the fact that prenatal care is now almost universally used as a covariate rather than an outcome, results of WIC program. Moreover, Currie (1995) argues that analyses that looked at the impacts of WIC on prenatal care are not includ- including prenatal care in multivariate models may be ed in this summary.

108 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program ct a wer Continued— ts lo n ant imp Signific Participa , s s} s} hite hite l] tcome wer l] {W l] {W C} a a nal] nal] I ts lo W n [nationa 87) ation ation hite} 79) d) [natio d) [natio er (19 999) [n 999) [n 988 988 Participa en (19 ion on birth ou 3 months on t {< [1 State] {W t impact Brien (1 Brien (1 Rush (1 Edozi Rush (1 Stockbau n ifica 8 } e 6 s] No sign e /sam al WIC participa l] {Blacks t a gher 5 counti ) [1 county] site] 86) 6 2 [1 site] ation 84) [1 State] 1 5) [1 State] ts hi 8 hite} prena n 85) [ kers} kers} of er (19 o o 996) [ n (19 ck (19 999) [n u Participa ins (19 [1 State] {W {nonsm {nonsm Stockbau Brien (1 Brown (1 Coll Bailey (1983) [2 sites] Silverma Metcoff (1985) Schramm (198 Kotelch e impact } 2 7 1 l] l] l] l] ct 3 examined th a a a t l] {Blacks sts 0) [1 State] 0) [1 State] gher 4

a nal] nal] C} ite} ] 4) I h s tha ate] ate] ate] W d ts hi W 91) [5 States] 01) [5 States] s o n 3) [1 State] / / [nationa [nationa 86) [1 State] ) ) ) 87) [1 State] [nation 87) 86) [nation ation 84) [1 State] 6) [1 State] 0) [1 State] ant imp e (199 e (199 o 0 0 4 4 2 h 00 9 9 8 8 8 r 79) 79) er (198 a) [natio a) [natio 9 9 9 3) o 1 1 1 995) 995) b l] ( ( ( ng er (19 er (19 er (19 er (19 th care co

ck (19 98 h a studie l 999) [n 988 988 y (19 y y y (19 y an (2 u er (200 g Signific i Participa e e d d d ndi en (19 en (19 e e e e months on n n n n

n n n 3 [4 areas in 1 St {3+ [4 areas in 1 St [4 areas in 1 St [1 site] {smokers} [ [2 sites] {smokers} [nation [1 State] {Blacks} [1 State] {non- e e e K Kotelch Devan Edozi Edozi Gordon (1 Mays-Scott (1991) [1 site] Devan New York Stat Metcoff (1985) Heime Bailey (1 K Reichm New York Stat K Rush (1 Kowaleski-Jones (2002) Buesch Brien (1 Rush (1 Stockbau Gordon (1 Stockbau Stockbau Schramm (198 Stockbau of table. t h sociated hea nal g i Findings from at end statio rthwe i e See notes Table 18— including as Outcome Mean b Mean g age

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 109 Chapter 4: WIC Program l] a ct a Continued— gher

12,000 } d > i ts hi e a n 5) [nation

ant imp c i l 99 b com u n p

o Signific ton (1 Participa ual i n g

d n {ann a Covin , s s} hite tcome gher l] {W a nal] nal] ts hi (2000) n k ation d) [natio a) [natio l] d Par a 988 988 999) {n Participa n an ion on birth ou t [nation t impact Hoga Brien (1 Rush (1 Rush (1 n ifica 8 } 15 ] No sign e same t a al WIC participa l] {Blacks t t a nal] nal] S wer/

1 ate] [ ) [1 county] Blacks} site]

[1 State] [1 State] 87) [1 State] ) 86) 87) ) ) 2 ation 1 ts lo 4 2 8 8 hite} hite} prena n 8 9 03) 03) a) [natio d) [natio 9 9 1 1 198 of ( ( er (19 er (19 er (19 acks} 996) [

n (19 l 999) [n 988 988 y y d e e n Participa n a v n [4 areas in 1 St [1 State] {W {non-B [1 State] {W [1 State] {non- {Whites} e e Silverma Bailey (1983) [2 sites] K Stockbau D Gregory (20 Stockbau Gregory (20 Brien (1 Rush (1 Brown (1 Stockbau Rush (1 Simpson ( d e impact e 11 11 4

l] l] 10, 10, ] } a a s 12 14 d l] l] e ct 13 t examined th a a a t a l] wer 9 9 sts—Continu 0) [1 State] 0) [1 State] t a ite} S

h 4 tha ts lo ate] up note [ (2000) W ) [1 State]

91) [5 States] n s 3) [1 State] / 5) [nation ) 5) [nation 87) 87) 86) [nation [nation ) k ation 84) [1 State] 8 6) [1 State] 5) [1 State] ro 0) [1 State] 1) [1 State] 3) [1 State] 3) [1 State] ant imp e (199 e (199 0 2 2 9 g 00 9 8 99 99 9 03) 03) 9 9 1 1 995) 995) l] d Par ( ( 03) [n er (19 er (19 er (19 th care co

ck (19 a studie a (19 l y y (19 y an (2 u er (200 er (200 er (199 er (199 li Signific Participa ton (1 ton (1 e d e a g g n an e n n a v n [4 areas in 1 St [1 State] {Blacks} [1 State] {Blacks} [1 State] {Blacks} {except sub [nation [1 State] {Blacks} [1 State] {non- e e Buesch Stockbau New York Stat Roth (2004) [1 State] Roth (2004) [1 State] Reichm Buesch Finch (20 Gregory (20 Ahluw Buesch Devan Kotelch K Gregory (20 Hoga Buesch D Covin New York Stat Gordon (1 Gordon (1 Covin Stockbau Schramm (198 Schramm (198 Stockbau

of table. y r w sociated hea e v

Findings from f ght at end

o t

h d wei g o od of lo i o e h i iho w l h e t 2,500 gm) 1,500 gm) See notes r k i i Table 18— including as Outcome Likel b (< L low birth (<

110 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program ct a Continued— gher ts hi n 86) 87) ant imp hites} hites} er (19 er (19 Signific Participa [1 State] {W [1 State] {W Stockbau Stockbau , 8 } s s} hite

] e t tcome gher a l] {W l] {Blacks t a a S

1 ts hi [

n ) ation ation 3 9 9 1 (

999) [n 999) [n y Participa e n ion on birth ou a t v e t impact Brien (1 Brien (1 D n ifica 8 8 } s} s} e 6 hit hite 17 14 No sign 18 19 s] l] same e a al WIC participa l] {W l] {Blacks t nal] a a nal] nal] nal] nal] {W wer/ ate] 91) [1 State] [1 State] / ) 87) [1 State] 87) [nation natio ation ation 84) [1 State] 6) [1 State] ts lo 0 4 prena n 9 8 03) a) [natio d) [natio d) [natio 9 3) [4 counti 97) [ 1 995) 88) [natio of ( er (19 er (19 acks}

ck (19 l 999) [n 999) [n 988 988 988 y y (19 u e d e Participa e (198 n n [4 areas in 1 St [1 State] {Blacks} {non-B e Brien (1 Schramm (198 Rush (1 Frisbie (19 Rush (1 Stockbau K Gordon (1 Joyce (19 Stockbau Rush (1 Brien (1 Devan Gregory (20 Paig Kotelch d e impact e ks} 7

] 19 s l] e ct t examined th a a t a wer l] sts—Continu 0) [1 State] 1) t nal] a nal] ites} S nal] {Blac

h 4 tha ts lo [ W [1 State]

91) [5 States] 91) [4 States] n s ) [1 State] [1 State] / / 87) [1 State] 86) [1 State] [nation 86) ) natio 84) [1 State] 6) [1 State] 5) [1 State] 3) [1 State] ant imp e (199 e (199 4 0 0 3 8 9 9 9 03) 03) a) [natio 9 97) [ 8) [nation 1 16 995) 88) [natio ( er (19 er (19 er (19 th care co

ck (19 studie l 988 y (19 y (19 y u er (199 Signific Participa e e e edy (19 n a v [1 State] {non- {Blacks} [1 State] e Joyce (19 Stockbau Kenn Devan Moss (199 D Gregory (20 Gordon (1 Gregory (20 Buesch New York Stat Devan Kotelch New York Stat Frisbie (19 Stockbau Stockbau Schramm (198 Schramm (198 Rush (1 of table. sociated hea h all- ality y 1 Findings from l at end eks ealt e birth ne on/sm ancy, od of od of id/h irth atal mort iho iho oximate 36-37 we See notes Table 18— including as Outcome Mean Medica care costs Likel prematur (< gestation) Likel intrauteri growth retardati for-gestational- age b Neon (birth through early inf appr month)

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 111 Chapter 4: WIC Program t y s for ific el that c s to size 742 in ct re not xt spe a riables tha adding up gher noted, s mod The te ts hi were lower than sample n r all (28 vs. odels we s ant imp e v nting,” or herwise sm e m ain only to a v cou significant findings ma very ; howe Signific Participa s and a fixed effect “vote ngs pert alternati tion than others. el did not control for va tcome hat limited WIC participant a s for WIC infant ple was for r findi

od st . tice of e. drugs. Unless ot s ou imester c attern of non c r e sam a t t s. r i h Where conside W - e first istent p n and model t s o n

d nt. uthors. The m con t more participant oup for some n fants in this orts, r gm. a caid hospital co 1 State). a te o avoid the p

y ction of differen e of alcohol and sts, s ne during th 21 s coh 1, a meri e d t gher re s r l] o s cial g y a ed b . Medi significa ra s nd u therwise noted, findings t weight in s ts hi re co y 1 city or hapte off b n t. ation studie rts. nal-age

of statu cautione . ing, a s ) k s preferr e al vs. i d in c a ese wa time of birth. or both Whit cohor were ple t-tes among non-whi determine di some r statio m Unless o w results b k coho e e smo rs are 999) [n ek ry-low-birth e ks. Participa ge note r he basi om r s r p d sim ve

-wee y d health ca care, one of th v wee reports ple, nation a t impact e y. Reade Brien (1 h n

ciate e v exam , findings f r act and . a ion bias, which gnificant impact gnificant impact f ble only up to the s ifica nalysis use research. A ented, could not s and 32 number of h 36- and 40 per also baseline models. s ct k si e r model for 28-we of prenatal o t or

participation on t y of r three of the fou kets}. fo ation y e pa cy r sele l c t pre d in that wa availa rt. Th e

sitive, sitive, si k study (fo ] 14 wever, a No sign i l ant, in l] e

re identical and for anoth t s a wer a coho s primary {in bra t e nal] l were no S

s ts lo t eek e of the 1 positive WIC imp e. ta or the full sample found po found po e, models fo ntrolled for data were s for [ sample. Th her conside n n ies. t site]

ison we rolled by 30 wee nd significant e r [nation ) eks gestation). a ot signific da be interprete 1 co e and adequa s ot p bgroup cture of the bod 3 ud s, including asso i s sul 37-w scop e groups; ho 9 c i c t 1) [natio r total 9 cau r t re el that defined WIC st 1 onal ag 995) a c sitive a ( who en compa ran well as f 996) [

p e, but n sive pi s fo

s showed s 988 y Participa v s would (< 37 we come stational ag C because stati e sted. I on d po esign and n yse W r one a (

v aris ee comparisons ee comparisons prehen ree insu e ge model that ate, and the 000), a ath rate be ever, to those ch d del. e r r i anal here refle ng situation, u r aturely he stronge significant for s. rts an s negati s Rush (1 Gordon (1 Brown (1 D a s not te y’s result identifies the su rt rth out me for a mod com livi em on d on comp all th eem r s mo r , how om t 996-2 a sa rolled for ge coho e wa r rolled for ge stud ct e wa timates fo ut of th ut of th p c tically ported sed cipant c s (1 s n two-sta a ek s the orn p omen. tinued Differen

cohort ed, but two han neonatal de the publicati cell entry also ] C infants in fixed-effe l fants b f providing 4) are b ndings re based o ed, but e ed. Two o ed. Two o act. s t erences in rese s model. s t not stati n s s s r nd 32-we same w cy. he differen e e ct ct zes findings f gh no single t r t examined bi n bu tional-length coho a cally significant t si a - a wer —Con he a ti t ks, in s model that cont s oup u for WI c k (198 hou rathe S r name, oup u oup u oup u y

r r r cluding age, income, obability of heavy 99) a pha s r 4 tha ts lo gest ve annua [ m s for t statis en t pregna

n s [1 State] ulation, the ) v at p <0.10. cance of ant imp 3 cause of diff 9 ily the direction) was 03) r four 9 author’ parison g eath rate spital sta r o significant imp gnificant in model that cont ificant for 28 nt in fixed-effe 1 dels that limited WIC parti fo as not d Kotelchu parison g parison g parison g 20 ach of fi rence d in the interest o han one model, fi ( significant in n n C participant d genous, in

e ns and e s l signifi studie

onths of udy pop Swann (19 effect, e . y r ho ation statu Signific Participa com cessar re t d m st nd e y an com com com sig ks w sults. Be senio not si e gnificant, for bla gnificant for p fetal n porte c i n pregnancie re a s m endo tically fou r significa ed for icip v s tantially greater among i rlying s o t {Blacks} shorte e re s e f r r statistica and si Gregory (20 D e and and p sub e

betwee s pact in model t s tiv cantly i impact on ng non-bla s epending on i significant in mo sults a m positive but d significant diffe s sitive but not sitive d for Brien and d for Kenned t estimated mo epending on epending on epending on s e was s not stati s favorable to WI sitive t han the entire h particular true unde dological limitatio re fect (but not ne but the

idered to be rent. nce l po po small. WIC par s as cipated in first 6 s, a Findings from e s no that c onse analyse r i s po porte porte e wa e wa e amo t her t sults d etho d signif con c c c of ef s . was was i ncy t t Cell entries show the a t ho parti s h first year

studies mple) o sses m a mpact wa Differen Differen Impact Mixed re No significant i Reporte N Impac Finding reflect Differen Impact was po Difference w Size of impact Mixed results d Dose-resp Mixed results d Authors reporte Mixed results d Impact wa Significant difference not Notes: Nonsignificant For Findings re Findings re 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15I 16 17 18 19 20 21 Table 18— by prenatal Outcome Infant mortality (later infa throug of life) subgroup rat be indicative of a all the studies wit discu qualitatively diffe the authors significance estimated differe non-Whites a those w full s non-WIC infant

112 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program deemed to be adequate in sample size and design. All then, (2) none of the reviewed studies was generalizable of these studies are included in tables 17 and 18.57 By to the entire WIC population, and (3) GAO researchers statistically combining the results of these studies, relied most heavily on findings from the WIC-Medicaid GAO researchers estimated WIC’s effect on reducing Study, which was largely limited to the very lowest the incidence of low birthweight as well as the inci- income WIC participants (GAO, 1992).60 USDA officials dence of very low birthweight.58 They then used this also stressed that the report did not adequately caveat its information to estimate the number of infants born in findings in recognition of the selection-bias problem. 1990 who would have been born with low birthweights if their mothers had not received WIC bene- Since the GAO meta analysis was completed, 13 addi- fits. Finally, cost savings attributable to WIC were tional studies have examined WIC’s impact on birth- determined by combining the estimate of averted low weight and/or Medicaid costs using techniques that were birthweight and very low birthweight infants with comparable to or better than those used in the studies information on the excess costs associated with caring reviewed by GAO. These include studies that involved for these infants. Cost estimates included short-term national datasets (Finch, 2003; Kowaleski-Jones and hospital costs, expected long-term disability costs, and Duncan, 2002: Hogan and Park, 2000; Brien and Swann, expected special education costs. A substantial propor- 1999; Covington, 1995; Gordon and Nelson, 1995), as tion of total costs were attributable to medical care well as studies that focused on one State (Roth et al., costs in the first year of life. 2004; Gregory and deJesus, 2003; Reichman and Teitler, 2003; Buescher and Horton, 2000; Ahluwalia et al., The GAO researchers concluded that prenatal WIC 1998). Two other studies used data from the WIC- participation reduced the incidence of low birthweight Medicaid Study (Devaney, 1992) and data from one by 25 percent (estimates from the studies examined hospital (Brown et al., 1996). With the exception of ranged from 10 percent to 43 percent) and the incidence Brown et al. (1996), all of these studies reported a sig- of very low birthweight by 44 percent (study estimates nificant WIC impact overall or for at least one subgroup. ranged from 21 to 53 percent). When these estimates Moreover, the studies by Kowaleski-Jones and Duncan were applied to 1990 births and associated costs, result (2002) and Brien and Swan (1999) included controls indicated that providing WIC services to mothers who for selection bias that the authors deemed successful. delivered babies in 1990 would ultimately save more than $1 billion in costs for Federal, State, local, and Taken as a whole, the available body of research pro- private payers. Savings to the Federal Government vides strong, suggestive evidence that WIC has a posi- were estimated at $337 million. These findings are the tive impact on mean birthweight, the incidence of low source of an oft-cited claim that “every dollar invested birthweight, and several other key birth outcomes, and in [prenatal] WIC saves $3.50 in other costs.”59 that these positive effects lead to savings in Medicaid costs. Even recognizing the pervasive self-selection In commenting on the GAO report, USDA officials problem and the fact that virtually all studies have other raised appropriate concerns that GAO’s conclusions over- limitations that limit generalizability, the consistency of stated the impact of WIC because (1) the reviewed stud- the results across studies is noteworthy. This is especially ies used data collected between 1982 and 1988, but both true when one considers that the bulk of the literature Medicaid and WIC had changed substantially since is comprised of relatively large, well-conducted studies, includes both national samples and State-level data that essentially amount to point-in-time censuses, and 57 In the GAO meta analysis, each of the five States studied by Devaney includes data from a number of different time periods. et al. (1990/91) in the WIC-Medicaid Study were considered as separate studies. Other studies included in the meta analysis were Silverman (1982), Kennedy et al. (1982), Kennedy and Kotelchuck (1984), Bailey et al. Other reviewers have reached similar conclusions (1983), Metcoff et al. (1985), Stockbauer (1986, 1987), Schramm (1985, (Rossi, 1998; Currie, 1995). Currie (1995) offers the 1986), the NWE (Rush et al., 1986, 1988a, 1988d), and Buescher et al. following observation: (1993). (The GAO report used a 1991 version of the work Buescher and his colleagues published in 1993). 58Estimates related to the incidence of very low birthweight are based Without knowing more about the selection mechanism on data from 5 of the 17 studies that provided separate estimates for inci- underlying participation in the program, it is difficult to dence of low and very low birthweight. In estimating reductions in very low birthweight attributable to WIC and the associated cost savings, the authors applied results from these studies to the other 12 studies. 60The WIC-Medicaid Study estimated, with the appropriate caveats, that 59The $3.50 savings (calculated in 1990 dollars and assuming a 2-percent every dollar spent on prenatal WIC participation generated more than discount rate) accrues over 18 years. Savings in the first year of life were $1.00 in Medicaid savings. This analysis considered only Medicaid expen- estimated at $2.89 per Federal dollar spent on prenatal WIC participation. ditures during the first 60 days after birth.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 113 Chapter 4: WIC Program

assess the probable direction of the bias. However, the fac- to Medicaid participants in North Carolina. At the time tors governing selection into the WIC Program are likely to of the first study in 1988, the Medicaid income-eligi- vary considerably over time and across sites. “...Hence, the bility cutoff was 100 percent of poverty, and a total of fact that the estimated effects are remarkably consistent across samples drawn from different states and at different 21,900 Medicaid births were included in the study times suggests that the positive results are not entirely driv- (Buescher et al., 1993). At the time of the second study en by the selection of women who are likely to have good in 1997, the Medicaid cutoff for pregnant women was outcomes into the program.” (p. 100). 185 percent of poverty, and the number of Medicaid births was almost double, at roughly 43,000 (Buescher Thus, the evidence that WIC participation during preg- and Horton, 2000). Although both studies found that nancy positively influences birth-related outcomes is WIC decreased the likelihood of low birthweight and fairly convincing. Beyond that, however, little else is very low birthweight, the magnitude of the differences clear. Because of the design characteristics that con- between WIC participants and nonparticipants was tribute to inherent underestimation or overestimation smaller in 1997 than it had been in 1988 (odds ratios of WIC impacts and the wide range of reported esti- of 1.36 vs. 1.45 for low birthweight and 1.90 vs. 2.15 mates, it is difficult to characterize the relative size of for very low birthweight). WIC’s impact—for example, the estimated reduction in the prevalence of low birthweight infants—with any Initiation and Duration of Breastfeeding: confidence. Moreover, subgroup analyses by some Research Overview researchers suggest that WIC impacts may be stronger among Blacks and other minorities than among Whites Impacts on breastfeeding are discussed in this section (Gregory and deJesus, 2003; Brien and Swann, 1999; because, as mentioned previously, any impact WIC Stockbauer, 1986, 1987) and among those at the low- may have on the decision to breastfeed is clearly tied est income levels (Finch, 2004; GAO, 1992). to nutrition education and/or breastfeeding promotion services provided to the mother during pregnancy. In addition, many important changes have taken place (Impacts on breastfeeding duration and other infant since the data used in most of this research were col- feeding practices may be influenced by WIC services lected. These changes may influence the extent to provided after birth.) which findings from previous research apply to today’s WIC program. The most noteworthy changes include The literature search identified few studies that the following: assessed the impact of WIC on breastfeeding behaviors. Many identified studies examined the impact of • A substantially higher level of program penetration specific breastfeeding promotion strategies/programs in most areas of the United States than was present on WIC participants. However, such studies do not in the mid-to-late 1980s (that is, most eligible prena- address the impact of the WIC program per se. That is, tal applicants are able to enroll and waiting lists tend they provide no information on what breastfeeding ini- to be the exception rather than the rule). tiation and duration rates would look like in the absence of the WIC program. • More generous Medicaid income-eligibility criteria for pregnant women (including some that exceed the Official WIC policy has always encouraged breast- WIC cutoff of 185 percent of poverty), which infer feeding. Both programmatic and research interest in automatic income-eligibility for WIC. this topic grew in the late 1980s and early 1990s, however, when national survey data indicated that • The use of standardized nutritional risk criteria. breastfeeding rates were declining nationwide (as the WIC program was growing) and that the rate of breast- Welfare reform legislation, which did not affect WIC feeding among WIC participants was less than the directly, may also have affected the circumstances of national average and less than the rate for low-income both WIC participants and nonparticipants. Any of nonparticipants. these changes may influence both the presence and size of WIC impacts as well as variation in impacts Many investigators have examined predictors of across subgroups. breastfeeding behaviors. Results have been very consistent and have demonstrated that women who are Two studies by Buescher and his colleagues illustrate African American, less educated, low-income, and how the prenatal WIC population in one State has younger are less likely to breastfeed than other changed over time. Both of these studies were limited women. These demographic characteristics are also

114 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program associated with an increased likelihood of WIC partici- Over the years, low-income women have exhibited the pation, so it is not surprising that studies that have lowest response rate and have therefore been oversam- included WIC participation among the list of potential pled. Weights used in analyzing survey data are specif- breastfeeding predictors have almost invariably found ically designed to account for differences in response a negative association or no association between WIC rates and coverage of various population subgroups participation and breastfeeding. (GAO, 1993; Ryan et al., 1991).

These negative statistics have prompted substantial Finally, two State and local studies examined WIC commentary and questions over the years, particularly: impacts on breastfeeding and infant feeding practices Does the formula provided by WIC act as a disincentive (Group III). Burstein and her colleagues (1991) report- to breastfeeding? and Does the WIC program devote ed preliminary impact estimates from the field test of adequate resources to breastfeeding promotion? the WIC Child Impact Study. A much smaller, local Obtaining reliable answers to these questions is com- study looked at the impact of multiple spells of partici- plicated by substantial selection bias that makes it more pation on breastfeeding rates among Hmong and likely that researchers will find a negative association Vietnamese WIC participants in northern California between WIC participation and breastfeeding. As just (Tuttle and Dewey, 1994). noted, the demographic characteristics of women who are least likely to breastfeed closely parallel the char- Initiation and Duration of Breastfeeding: acteristics of women who are most likely to participate Research Results in WIC. In addition, it is reasonable to assume that In the NWE, the Longitudinal Study of Women found women who have decided to formula-feed may be more that WIC participants were both less likely to plan to likely to participate in WIC than women who have breastfeed (breastfeeding intention) and less likely to elected to breastfeed in order to obtain the free formu- initiate breastfeeding in the hospital than income-eligi- la. The incentive to participate may be substantially ble nonparticipants (Rush et al., 1988d) (table 20). reduced for women who have decided to breastfeed. However, study investigators discounted the finding about breastfeeding initiation because they believed it The literature review identified nine studies that was influenced by a substantial amount of missing attempted to estimate the impact of WIC participation data in the hospital records that provided data for the on breastfeeding behaviors (table 19). Studies that analysis. used only t-tests or correlation coefficients to examine this relationship, without controlling for measured dif- A study completed by Ryan and his colleagues in ferences between groups, are not included. As just 1991, using RLMS data for 1984 and 1989, reported noted, these studies are virtually guaranteed to find a that breastfeeding rates, and extended breastfeeding in negative association or no association between WIC particular (6 months or more), declined disproportion- participation and breastfeeding because of the demo- ately among WIC participants during this period. Even graphic characteristics of WIC participants. after controlling for measured differences between groups, nonparticipants were 1.5 times more likely Two components of the NWE examined breastfeeding than WIC participants to initiate breastfeeding in the in a fairly limited way (Rush et al., 1988d; 1988c) hospital. This study contributed substantially to the (Group I). Five studies used national survey data to debate about the role of the WIC program in promot- study the impact of WIC on breastfeeding (Group II). ing breastfeeding. Two of these studies used the NMIHS, one study used the NLSY, and two studies, including one study con- The reliability of these findings was called into question ducted by the GAO in response to a congressional because of concerns about the adequacy of the single request, used the Ross Laboratory Mother’s Survey survey item used to classify WIC participants and non- (RLMS). The RLMS, in various forms, has been ongo- participants and lack of attention to the issue of selection ing for more than 40 years and is used to document bias (Tognetti et al., 1991). The survey item used to national trends in infant feeding. The RLMS includes a identify WIC participants asked whether the mother or mail survey of a large nationally representative sample the target infant participated in WIC at any time since of mothers of 6-month-old infants. The sample repre- the infant’s birth. This composite question did not allow sents 70-82 percent of all new mothers in the United differentiation of women who participated in WIC pre- States (Ryan, et al., 1991). Response rates have gener- natally (and therefore had the opportunity to be exposed ally been lower than desired for scientific surveys. to WIC breastfeeding promotion advice and activities)

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 115 Chapter 4: WIC Program n n g od h Continued— udin s model l gressio gressio as te re te re a a on, inc n bi Analysis met Multivari attempt to control for selectio (3) Fixed-effect regressi Multivari (1) (2) Multivariate

, y l l C

I a f mmy mmy t o u u a W

n e o e t g r n a on d on d i p

n n g o o n

Measure of i i participation t d d p e u e l s c c a n n Participati Participati i b i Participation dummy t vs. t vs. t vs. pant pant pant n n n Design partici partici partici Participa non Participa non Participa non

) 0 9 5 of 2 3 ) en en ., e ded blic dren nant d r n and ts 9 l ring n nal born n e e g , n o h at 1,282) dr dr n io 3 di in n 199 n 198 clu 91 = n n= preg et al n ics o latio dre d chil ( rn du low) ng pu h 517) tative

ncom paris o ngitu e other s n twee twee di l ally e e clin a en 19 995, wit 995 ( om samp t h i Popu ,370) atal care i p receiv ng b 8d be 5 (n= on breastfeeding p 2 970) (sample siz C participan s I m o a (1) NLSY chil healt h pren surroun grou sample of W NLSY chil and 1 born b Nation represe and com (2) Low-i Rand infants an in the lo study of wome (see Rus betwe the same per (n= of women i (n= and 1 least on sibli (3) NLSY chil born b 198 199 on on 1 llecti llecti WIC progr o o ics ics SY n n the L f C sites and 55 C sites and 55 I I Data source 4 W 4 W atal cli atal cli 3-84) 3-84) 9-95 N pact o m Primary data c 198 in 17 pren (198 Primary data c in 17 pren (198 surveys onal d ing ing ing nd nd amined the i is of nati on an on on a on a on on ex t Outcome(s) Breastfeed initiati Breastfeed intenti durati Breastfeed initiati initiati durati of table. udies tha ) t E al. at end II: Secondary analys I: National evaluations 8d) (NW 2) See notes Table 19—S Study Group Rush et al. (1988c) (NWE) Rush et al. (198 Group Chatterji et (200

116 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program ent n n n n od h Continued— sion with gressio gressio gressio gressio as adjustm gres te re te re te re te re a a a a n-bi Analysis met Multivari Multivari Multivari 3-stage re selectio Multivari d e t a p i mmy mmy mmy mmy s : c i u u u u e mmy t e r u a p

on d on d on d on d pons y l s vice d u Measure of d participation o i v e r Participati and a Participati Participati Participati in WIC Dose res Number of tim p

o s o e m- e c l i ant b v to i d g t vs. t vs. t vs. with ts wh pants pants pant pant pant i l a n n n d n

e l ntinued rticip ants wh d - stpartu t receive a e e o Design o v i Co m partici partici partici partici partici e o c c e n Participa r i advice and did n non to breastfeed compare particip Participa non Participa non and p only p Prenata participants vs. non Participa non

3 g e l b 2 an ic i ) C s th r c ose for ts i n I g 428) n edin cided who e i n i a 9 l

d d was l eri e s - t 79,

l ispa mothers e n l 198 latio pants rth nts wi data e i a 34) tative m

ese W t unde e d n o ants wh d o n ally c prior to the ons of e on-H nde o n gest chil d Popu ,170) ,089) i es. Analys p ut infant fe partici 4 and

ite women u 6 4 122) 120,3 s l (sample siz C participan d h I c e n n am on breastfeeding— R Nation (n= non (n= 198 sample of of 6-month-o babi i (n= represe and n a W were n abo plans infant’s b youn less than 1 ye Hmong and Vietnam particip (n= respo complet questi interest (n= W Mexican-Am

S on M 1

n llecti R i ornia

f ve births o i and 9 s 8 d MS 9 o the WIC progr L 1 o f

n Cal h d r n o Data source a b C clinics

I h 8 NMIHS li 9-92 R 8 NMIHS 4 g i 8 pact o e 9 m 198 Primary data c in W 198 n 198 1 1 norther community ing ing ing ing ing nd nd amined the i udies on on on on a on a on on ex t Outcome(s) Breastfeed Breastfeed intenti Breastfeed initiati Breastfeed initiati initiati durati durati initiati Breastfeed of table. y e udies tha 91) t w at end State and local st III: ar et al. 5) 2) 4) See notes Table 19—S Study Balcaz (199 GAO (1993) Schwartz et al. (199 Ryan et al. (19 Group Tuttle and De (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 117 Chapter 4: WIC Program , as n od s, all h ction bi gressio e sample. te re participant a Analysis met Multivari including attempt to control for sel be Medicaid of the initial survey n to w rcent mmy u on d Measure of d only 34 pe participation rise gible for WIC or kno Participati sample comp t vs. pant re income eli n ntinued Design Co partici who we Participa non cipants 2 ) e hs 807) l mplete data, analysis e n e l b i t (n= g latio i l h e g i - e om samp C and I Popu m o (sample siz c s limited to nonparti n am on breastfeeding— Rand of W i infants (6 mont old) stratified by birthwe oup wa r on

on g ) 1 s er excluding cases with inco 1 llecti 9 ari - Aft o 0 d North 9 9 the WIC progr comp f 1 (

a an a Data source n i l o r y 50 percent. pact o l es that a a . m up. Primary data c in Florid C y o r imate x pro ison g ’s Surve r r nfant Health Survey. y sample indicat ing was ap nd compa amined the i ud s Mothe rvey on a on ex t Outcome(s) r su boratorie Breastfeed initiati durati al Maternal and I se rate fo re included in the udies tha t s: description of the st we ce s r all respon r NLSY = National Longitudinal Survey of Youth. NMIHS = Nation RLMS = Ross La 1) Data sou Unless the Ove 1 2 3 Table 19—S Study Burstein et al. (199 income level

118 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program en y ific y c l] ses a s ct e betwe 3 7 spe 3,5 a wer c l] l] l] a a 1 a ll} ts lo a n ) [nation men who ext discu ) ation ation adding up all the ) roportionatel ant imp 5 differen 5 2 9 0 0 The t ain only to a l] {over 2 ice} s disp ( significant findings ma a

991) [n 991) [n i ting,” or j Signific Participa ar (199 r e s wa t t rd ngs pert a [nation {no adv h ed virtually no han others. ing. However, wo Balcaz GAO (1993) [nation C Schwartz (19 Ryan (1 Ryan (1 findi veal attern of non “vote coun s re spital reco Where tice of ideration t c istent p s s con articipant l] 1 State). re con with formula feed a 3, 4 1, a wer r oid the pra element on ho WIC p v nal] nal] a y ts lo 1 city or n hapte merit mo ) [nation 6 d to a ) ) [2 States] 2 ombination 2 9 tum-onl d) [natio d) [natio al vs. d in c r 00 991 xt). in c studies r 988 998 e} Participa cautione stfeeding e relevant dat note s (see te some ively o dels. s ple, nation {advic stfeeding. t impact ers are t that th Burstein (1 Chatterji (2 Rush (1 Rush (1 Schwartz (19 n stfeed. c xclu vored exam natal and postpa r ifica research. A usted mo ings from her e suspe y. Read o brea ram on brea y of thors fa rs kets}. g of pre wa , find u c s a l] n study (fo No sign a stfeed, eit gher e autho {in bra nal] nal] ted in that brea . Th omparison e of the ts hi s ed with initiation of brea o t which the nsideratio n selection-bias-adj ) [nation re likely to plan t c natio natio at 2 bgroup cture of the bod co s. C 9 scop r soci s ice} ion and sive pi 998c) [ 998c) [ Participa onparticipant alf of the subje prehen ate, and the equation model, icipants to plan t {no adv ign and othe ositively a le-equat - st studies. s rt p lts would be interpre Schwartz (19 Rush (1 Rush (1 e identifies the su e impact of the WIC pro st h com ere significantly mo on d sing nge old. s vs. n a s ge. r stro n nonpa he single y’s resu eed w earch d nt for s stf ll entry also 2 participant the publicati is 6 month f providing l] ed in WIC was ct model. al discha ssing for almo l] sed on t a es in re consiste o brea ct s from the t c examined th single stud a effe t - gher ospit e ba ere ticipat r less likely tha name, vice s were mi tha finding ts hi gh no . differen [nationa s ) [nation s n community] ulation, the ce ) 1 ant imp 2 5 9 (1991) w e data author’ iously par ice} of prenatal WIC r s v gnificantly ding when infant d in the interest o al. (2002) a en thou al. ficant in fixed ding before h si phasize v eastfeed on studie udy pop receiving ad e} r s cau Signific ar (199 Participa st senio stfee porte ct, e stfee erji et t be were s. Because of c had pre orted {with adv {advic e re r cally signi compari Balcaz Schwartz (19 Tuttle (1994) [1 spe ly su on of brea is for sults a ratio of brea i statisti s woman d for Burstein et d for Chatt derlying effe han the entire s re cular result IC and rep s en who did not b a un rted Findings from g om porte porte e high her t r ing in d d Cell entries show the all, WIC participants with parti for w r on Ove Number of time Limited to initiat Results a Result repo Difference w Based on odd Notes: Nonsignificant Findings re Findings re 1 2 3 4 5 6 7 Table 20— Outcome Intention to breastfee Breastfeed initiati Duration of breastfee subgroup rat indicate a true studies methodological limitations and em participated in W skipped the two groups.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 119 Chapter 4: WIC Program from those who participated only after the birth of the than income-eligible nonparticipants. Neither WIC child. The combination of prenatal and postpartum participation nor receipt of breastfeeding advice had a participants and, potentially, infant-only participants, significant impact on the duration of breastfeeding. may have diluted the apparent WIC effect. Balcazar et al. (1995) used NMIHS data to assess pre- GAO (1993) used RLMS data for 1992 to address con- dictors of breastfeeding intentions and found a similar gressional questions about the effectiveness of WIC’s relationship between receiving advice/encouragement current breastfeeding promotion efforts.61 GAO’s analy- to breastfeed and reported breastfeeding intentions. sis included a multivariate regression to investigate the While the relationship between WIC participation and relationship between prenatal WIC participation and breastfeeding intentions was negative overall, the rela- initiation of breastfeeding in the hospital. Results tionship was positive among women who reported showed that, after controlling for differences in meas- receiving breastfeeding advice/encouragement from ured characteristics (including education, income, race, WIC staff. In addition, receiving advice/encourage- and a variety of other characteristics known to be asso- ment to formula feed was negatively associated with ciated with breastfeeding rates), prenatal WIC partici- breastfeeding intentions. pants were just as likely as postpartum-only participants to initiate breastfeeding. Moreover, prenatal WIC par- An obvious concern about both of these studies is ticipants were significantly less likely than nonpartici- whether self-reported data about receiving advice are pants to initiate breastfeeding. Study authors cautioned biased in any way. For example, women who breastfed that the analysis did not control for selection bias and could have been more apt to report having gotten advice that unmeasured characteristics, whether related to the to do so. Or, WIC staff could have provided breastfeed- woman herself or her interaction with the WIC pro- ing advice/encouragement to women who indicated an gram, may have contributed to the observed differ- interest in breastfeeding. To address this issue, Schwartz ences between WIC participants and nonparticipants. and his colleagues estimated an alternative, two-stage equation that omitted the breastfeeding advice variable. In 1992, Schwartz et al. used data from the NMIHS to The alternative model yielded results that were substan- examine the impact of WIC on breastfeeding. They tially different from the results (reported above) for the estimated three equations jointly and simultaneously to three-stage model. In the two-stage model, the coeffi- control for self-selection and to model the decision to cient for WIC participation, which was strongly and initiate breastfeeding and, for those who breastfed, the significantly negative in the initial three-stage model, duration of breastfeeding. The analysis looked at the was positive and not statistically significant, suggesting combined influence of participating in WIC and receiv- that WIC participation had no impact on breastfeeding ing advice and encouragement from WIC staff to breast- initiation. The fact that the two models produced such feed. In the joint model, the coefficient for WIC partic- divergent results is somewhat troubling. Given the ipation was significant and negative and the coefficient potential problems with the reliability of data on for receiving breastfeeding advice was significant and breastfeeding advice, one must question the authors’ positive. The interpretation is that the impact of WIC on uncaveated preference for the three-stage model. breastfeeding was mediated by whether the woman was encouraged by WIC staff to breastfeed her infant. After The most recent study of WIC’s impact on breastfeed- controlling for socioeconomic differences, prenatal WIC ing was completed in 2002 by Chatterji and col- participants who reported having received advice/ leagues. They used data from the NLSY to examine encouragement to breastfeed were more likely to initi- breastfeeding initiation and duration among children ate breastfeeding than either participants who did not born between 1989 and 1995. WIC participation was receive such advice or income-eligible nonparticipants. defined based on the mother’s participation during the 62 In contrast, WIC participants who did not report year of the child’s birth. No information was avail- receiving such advice/encouragement to breastfeed able on whether WIC participants received advice or were significantly less likely to initiate breastfeeding encouragement to breastfeed.

61Although recognizing the potential problem of nonresponse bias in the 62The authors also completed parallel analyses using a variable that RLMS data, GAO researchers pointed out that survey weights were specif- defined WIC participation based on participation during pregnancy or at the ically designed to deal with this issue and that estimates of national breast- time of birth. Results of these analyses were reportedly “very similar” but feeding rates derived from the RLMS were consistent with those of key were not presented. In addition, for children born in 1994, WIC participation government-sponsored national survey efforts, including the NMIHS and was proxied based on WIC participation during the year that preceded the the National Survey of Family Growth (NSFG). child’s birth because data on WIC participation were not available for 1994.

120 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

The authors used two different approaches to control between WIC participants and nonparticipants were not for selection bias—a two-stage model and a fixed- statistically significant. Results of the two-stage models effects model that used data for sibling pairs. To model yielded no significant findings, although coefficients for the participation decision in the two-stage model, the WIC participation were consistently negative. The fixed- authors used variables that represented State-level effects model found that WIC participation had a signifi- WIC and Medicaid policies. Information on State-level cant, negative effect on breastfeeding duration (mean WIC policies were obtained from the biennial WIC number of weeks breastfed). Participant and Program Characteristics (WIC PC) Studies, so assigning values to individual sample Although the authors say that their instruments per- members was somewhat imprecise. Values for children formed fairly well, they ultimately rejected the selec- born in years for which WIC PC data were not avail- tion-adjusted results—which found no significant WIC able (1991, 1993, and 1995) were assigned based on effect—in favor of the baseline regression results- WIC PC data for the following year. In addition to which found a negative WIC effect. The rationale for Medicaid income eligibility cutoffs, State-level factors this decision was that Hausman tests suggested that considered in the model included links between WIC WIC participation was not endogenous. This conclu- and Medicaid, TANF, and FSP, WIC policies about sion is open to question, given that the Hausman test income documentation, and the presence of nutrition- depends heavily on the availability of good instru- based restrictions on WIC food packages. ments (Carlson and Senauer, 2003). Moreover, the authors clearly stated that their hypothesis was that The authors describe several other variables that were “despite the important efforts the WIC program has considered but ultimately excluded from the model made to increase breastfeeding during the 1990s, WIC because they did not pass the test of over-identifying participation is still associated with lower rates of restrictions or “were very poor predictors of WIC par- breastfeeding because of the valuable infant formula ticipation.” These variables included monthly (as available to participants.” opposed to less frequent) voucher issuance, nutritional risk criteria, nonnutrition-based food package restric- Viewed in concert, the available studies provide no tions (for example, restrictions related to package size firm ground for making causal inferences about the or brand), and costs of WIC food packages. The fact impact of WIC on breastfeeding initiation or duration. that these variables were excluded from the final Statistics do show, however, that breastfeeding rates model suggests that the selection-adjustment model, among WIC participants have been increasing. The like those used in research on birth outcomes, was RMLS shows a 69-percent increase between 1990 and very sensitive to changes in specification. 2000 in the percentage of WIC mothers who initiate breastfeeding and a 145-percent increase in the per- The authors reported results for a standard regression centage who were still breastfeeding at 6 months (baseline model), the selection-adjusted model, and the (Oliveira, 2003). This increase cannot be attributed to fixed-effects model. For baseline and selection-adjust- the WIC program because breastfeeding rates have ed models, impacts were estimated for the full sample been climbing for the population overall. However, as well as for a low-income sample. Outcomes includ- since the late 1980s, USDA has specifically targeted ed breastfeeding initiation and whether breastfeeding promotion of breastfeeding in the WIC program lasted for 16 weeks. For the fixed-effects model, the (USDA/FNS, 2003a). For example, in 1989, P.L. 101- dependent variable was the number of weeks the child 147 required that USDA develop standards for breast- was breastfed, including zeros for nonbreastfed feeding promotion and support and targeted $8 million infants. The fixed-effects model included 970 children for State-level efforts in this area. In 1992, P.L. 102- who had one or more siblings in the sample; however, 342 required that USDA establish a national breast- only about 15 percent of these children lived in a fami- feeding promotion program. That same year, USDA ly where WIC participation varied across siblings. instituted an enhanced food package for women who exclusively breastfeed. The enhanced package has Results of baseline regressions showed a significant, additional amounts of juice, cheese, and legumes and negative association between WIC participation and includes carrots and canned tuna. In 1994, P.L. 103- breastfeeding initiation in both the full sample and the 448 increased the amount of money each State was low-income sample. Coefficients for breastfeeding dura- required to devote to breastfeeding promotion and tion (whether infant was breastfed for at least 16 weeks) required that all States collect data on the incidence were also negative for both samples, but differences and duration of breastfeeding among WIC participants.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 121 Chapter 4: WIC Program

Finally, in 1998, P.L. 105-336 authorized the use of evaluations (Group I), two studies that were based on State administrative funds for the purchase or rental of secondary analysis of data from NHANES-III (Group breast pumps. II), one large State study (Group III), and five small, local studies (Group IV). USDA has also implemented several breastfeeding promotion demonstrations and has disseminated find- Nutrition and health characteristics examined in these ings and recommendations to State and local WIC studies include dietary intake (six studies), nutritional agencies.63 Evaluations of several of these demonstra- biochemistries—most often iron status or the preva- tions have found that breastfeeding promotion efforts lence of anemia (five studies), and weight gain during during pregnancy can positively effect the initiation of pregnancy (four studies). Like much of the research on breastfeeding among low-income women and that sup- WIC impacts, most of these studies are quite dated. At port during the postpartum period can positively influ- least three of the studies (Kennedy and Gershoff, ence breastfeeding duration. It is beyond the scope of 1982; Endres et al., 1981; Edozien et al., 1979) are this review to summarize these initiatives. However, based on data collected in the 1970s. One study pub- the interested reader is referred to Weimer (1998), lished in 1983 (Bailey et al.) did not report the dates Bronner et al. (1994), and Sanders et al. (1990). that data were collected. Only three studies (Roth et al., 2004; Mardis and Anand, 2000; Kramer-LeBlanc In 1996, USDA entered into a cooperative agreement et al., 1999) are based on data that were collected after with Best Start Social Marketing to develop and 1984. Roth et al. (2004)—the most recent study— implement a national breastfeeding promotion cam- focused primarily on impacts on birth outcomes; of the paign. The program was officially launched in August outcomes discussed in this section, only weight gain 1997. In 2003, the program was expanded to include during pregnancy was included. training programs for WIC staff in implementing and managing peer counselor programs. Both Mardis and Anand (2000) and Kramer-LeBlanc et al. (1999) examined dietary intakes and used bivari- Clearly, WIC’s focus on breastfeeding promotion has ate t-tests to assess differences between participants increased substantially since the NMIHS, which prob- and nonparticipants. Thus, while these studies are use- ably provided the best data, was conducted. While it ful in understanding observed differences between makes sense for USDA to focus research efforts on dietary intakes of WIC participants and nonpartici- identifying effective breastfeeding promotion strategies, pants, based on relatively recent data, they do not pro- it would also be useful to obtain updated impact analy- vide valid estimates of WIC impacts. Both studies ses. Additional work with the NLSY data may provide used the same dataset (NHANES-III) and the same some insights, and the Panel Study of Income Dynamics samples. Mardis and Anand’s analysis focused on food Child Development Supplement (PSID-CDS), which group intakes, while Kramer-LeBlanc’s analysis was implemented in 1997, may also be a useful data looked at nutrient intakes. source. This longitudinal study collects information on both breastfeeding and WIC participation (Logan et None of the studies that examined the relationship al., 2002). between prenatal WIC participation and the nutrition and health characteristics of pregnant women attempted to Nutrition and Health Characteristics of control for selection bias. However, one of the local Pregnant Women: Research Overview studies (Metcoff et al., 1985) is the only study known Ten of the identified studies looked at the impact of to have used a randomized design to study WIC impacts. prenatal WIC participation on the nutrition or health The authors were able to use a randomized design characteristics of pregnant women themselves (table because, at the time data were collected—early in the 21).64 These studies include both of the national WIC WIC program’s history—the need for WIC services in the area under study exceeded available resources.

63It is beyond the scope of this review to summarize these initiatives, however, the interested reader is referred to Weimer, 1998, Bronner et al., Nutrition and Health Characteristics of 1994, and Sanders et al., 1990. Pregnant Women: Research Results 64Fraker, Long, and Post (1990) attempted to examine the impact of WIC participation on all types of women (pregnant, breastfeeding, and Dietary Intake postpartum combined) using the 1985 CSFII data. However, because of the very small sample of WIC participants (64), the authors recommended that With the exception of the descriptive analyses completed results be considered investigatory only. by Mardis and Anand (2000) and Kramer-LeBlanc et

122 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program n n od h Continued— s s gressio gressio te re te re a a Analysis met Multivari Bivariate t-test Bivariate t-test Multivari omen Newly mmy mmy mmy : pants i u u u e nt w th of nts with a g on d on d on d n pons e g partic Measure of participation lin participation vs. participa varying l Dose res enrol Participati Participati Participati s p l a ts, ts, t vs. t vs. t vs. pant pant pant n n n n n outcomes of pregn mistries: ate grou e Design partici partici partici bioch Participa (2) Dietary intake: Participa before vs. after, separ before vs. after, same women (1) Nutrition Participa non Participa non Participa non

) 3 n and health 2 7 ) nant r n ts 4 3 en ublic n e g , n ) n n n o 3 5 e e in = 8 m m preg n 8 ics o latio led i , ( ng p o o tative

paris 2 s w w n di l ~

ally clin a nrol ant wom n e e t l l ( h i Popu atal care i p receiv

b b p i i 242) 242) (sample siz C C participan s g g I I i i o l l am on nutritio Pregn who e W (n= e (n= e pren surroun grou sample of W Nation represe healt and com h WIC and income- WIC and income-

e r ain o on on

C 1 I were were llecti llecti hs States W ES-III ES-III months WIC progr t tractions nd ag

a a o o 4 proxi- n o t 1 AN AN ics n n i n the H H

t f ery C sites and 55 ery, and onc I rd abs n o e liv e Data source m l lment, ap 4 W atal cli 3-84). Dat 3-76). Dat atal care a l o r pact o n m 1988-94 N 1988-94 N Primary data c and rec in 17 pren (198 collected at time of e after deliv Primary data c (197 collected at time of WIC enrol mately every 3 until d in 19 sites i pren gestation at about 8 mo survey data onal ght ancy i , n e e e e

, of nati n n i e of e of amined the i b

, , preg o l a a enc enc g i ex t ht gai nancy w o Outcome(s) m e m n e gain preval a preg Dietary intak Dietary intake, h Dietary intak Dietary intak preval anemi weig of table. c udies tha ) t n E la 0) at end II: Secondary analysis I: National evaluations en et al. d (200 8d) (NW 9) See notes Table 21—S Study Group Rush et al. (198 Edozi (197 Group Mardis and Anan Kramer-LeB et al. (1999)

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 123 Chapter 4: WIC Program n n od h Continued— ance s gressio gressio te re te re of vari a a Analysis met Continued Multivari Bivariate t-test Multivari Analysis omen— mmy mmy mmy mmy u u u u nt w a on d on d on d on d Measure of participation Participati Participati Participati Participati ed t vs. t vs. t vs. pant pant pant n n n outcomes of pregn ent omiz Design partici partici partici Participa non Participa non Participa non experim Rand birth record files

n and health C C 2 o I I d - ) t t at bers i al ents

e t 0 size en en e n W W r s me- d pi - - an a e- a e n n t p b enat ng re 3 o o c

t; roughly n i latio e ies n n y h egn 99) o d ivi

c erag g n icted e d d e i id reci ent num

n r l d not al pr d n n e a p a l a a

Popu nant wom nant wom n on nutritio

es vs. small or b e i 295,5 410) 519) g r (sample siz C and inco C C g I I I e i e m r l a selected at mid- p W W Income-eligible preg W care (n=101) e (n= time of recruitment and rec identic weeks pr Medica preg who di participate in high- risk obstetrical program (n= cipants w on pre birthwe equiv w have av babi large bab (n=

y r hed with Medicaid and/or a c u on on on id, a n 1 a J ic in 1

llecti llecti llecti rted) Florida ics in 6 n WIC progr nd of 0 n hs in o o o ties o n n ahom 0 e n iles mat e e al cli the w ealth C, Medica t f I e er 20 l statistics b C site i l in Okl C site and 1

I W I a Data source a d blic h 0-81) 3-84) ama cou d i artment cli 6 and th r pact o o l m Linke F records for birt and vita Primary data c Decemb (198 (198 Primary data c Primary data c in pu non-W dep 199 Alab hospit at a prenat at 1 W (Dates not rep l WIC participation f ona i ght ght i i amined the i nal mistries mistries ancy we ancy we ex e e t Outcome(s) Pregn gain gain Pregn Variety of nutrit bioch Dietary intake, nutritio bioch of table. r State and local studies udies tha e t 04) h at end State-level studies using al. III: IV: Ot 3) 5) 5) ins et al. See notes (198 Table 21—S Study Group Roth et al. (20 Group Coll (198 Metcoff et al. (198 Bailey et

124 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program eir h n od s, all h s gressio te re participant or WIC during t a d f Analysis met Continued Bivariate t-test Multivari be Medicaid olled in WIC. n to were wait-liste r w n omen— ed mmy : v u ver e C e nt w I a on d pons rs recei Measure of e articipant group participation ood but ne rh gible for WIC or kno bo Participati Dose res Number of W vouch he nonp , r s p ame neigh ts, ts vs. pants re income eli n n outcomes of pregn nd afte in s ate grou Design partici ilities who we fac Participa before vs. after, separ Participa non before a are gible women included in t cipants n and health 2 ) alth c o 4 d

e g n - n n e C i C l I I l m o latio W o r 766)

n w d ants wh ants an

e n

e l a y Popu table 5). WIC-eli nant W l on nutritio

b i t non-WIC he 232) w (sample siz C g I i e s limited to nonparti m l a were on the program for 6 months or more (n= particip N preg particip W e (n= are a c . y s approach. oup wa C r I on g 1 s birthweight (see analysi d prenatal is facilities WIC progr ari e for sites and nant W o 22 h c areas of cal v n C C on I ei and lls alt c phi the comp f ure s of preg C he s in Illi ants in ogra I e Data source 8-79) 3-78) C and medi pact o I es that a ch mea trition Examination Surve men who re m up. Dietary reca (197 counti records in W W particip non-W in 4 ge Massachusetts (197 sample o wo r r ea sing on impact of WI were ison g ried fo cu r ls d va n ve a e

e y sample indicat n udy fo i tum, or size al Health and Nu r compa amined the i b ud st a r o l tp g ex o t Outcome(s) m e hematocrit l Dietary intak H ANES = Nation pants in large aximum; sample H ed in WIC pos re included in the ) udies tha N t rtici description of the st we ce: s nd r , enroll cy edy a 1) Data sou Unless the Approximate m Subset of pa 1 2 3 4 pregnan Table 21—S Study Kenn Endres et al. (198 Gershoff (1982 income level

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 125 Chapter 4: WIC Program al. (1999), all of the studies that have assessed the observed differences in the dietary intakes of prenatal impact of WIC participation on the dietary intake of WIC participants and nonparticipants. It provides pregnant women are quite old. Indeed, aside from information on whether pregnant WIC participants these two studies, the most recent study is the NWE consumed more or less energy and nutrients than preg- (Rush et al.,1988d), which used data collected in nant nonparticipants, but this information cannot be used 1983-84. Findings from such dated studies are subject to conclude that WIC participants were more or less to concerns about changes in the program and its par- likely than nonparticipants to have adequate intakes. ticipant groups over time, as discussed in the preceding section on birth outcomes. Finally, as noted in chapter 2, the estimation of food and nutrient intake is an elaborate process that is sub- In addition, a compelling argument can be made that ject to significant measurement error. This error may impacts on diet-related outcomes are even more sensi- make detecting differences between participant and tive to temporal considerations than impacts on other nonparticipant groups difficult. outcomes. For example, the American food supply has changed dramatically since the early 1980s, with Although subject to the above limitations, as well as to important implications for observed dietary intakes. potential selection bias, evidence from early studies Americans are eating substantially more grains than paints a reasonably consistent picture of WIC’s they were two decades ago, particularly refined grains, impacts on the dietary intakes of pregnant women as well as record-high amounts of caloric sweeteners (table 22). The evidence suggests that WIC participa- and some dairy products and near-record amounts of tion increased intakes of food energy and most of the added fats (Putnam and Gerrior, 1999). Over time, nutrients examined, including four of the five nutrients myriad new products have come onto the market and traditionally targeted by the program—protein, vitamin food enrichment policies and standards have changed. C, iron, and calcium. Evidence for vitamin A, the fifth In addition, food purchasing behaviors may have been WIC nutrient, is less consistent, but vitamin A intake is influenced by including, for example, more food eaten especially difficult to estimate because the distribution away from home, smaller households, more two-earner is so skewed (vitamin A is concentrated in large and single-parent households, and increased ethnic and amounts in relatively few foods). The early evidence racial diversity (Putnam and Gerrior, 1999). These fac- also suggests that WIC increased intakes of vitamin tors make the recent studies by Mardis and Anand B6, which the program has targeted in recent years. (2000) and Kramer-LeBlanc et al. (1999), although The NWE (Rush et al., 1988d) also found that preg- strictly descriptive, important for understanding poten- nant WIC participants consumed significantly more fat tial WIC impacts in the current environment. than nonparticipants. However, if intake is translated into percentage contribution to energy intake (using All of the available research on dietary intakes of reported means for fat and energy), both groups con- pregnant women is also subject to the limitations that sumed about 37 percent of energy from fat. affect most of the available research on diet-related impacts of FANPs, as discussed in chapter 2. All of the NWE authors (Rush et al., 1988d) pointed out that the available studies used intake data for a single day and, relative magnitude of the incremental intakes observed therefore, provide weak estimates of individuals’ usual among pregnant WIC participants were plausible in dietary intake. In addition, in assessing intakes of food that they were comparable to the levels of supplemen- energy, vitamins, and minerals, researchers generally tation achieved in smaller, intensively controlled clini- compared mean intakes of participants and nonpartici- cal trials. Moreover, a thorough analysis of the sources pants relative to the Recommended Dietary of nutrients in women’s diets completed for the NWE Allowances (RDAs), or compared the proportion of confirmed that differences in the diets of WIC partici- individuals in each group with intakes below a defined pants and nonparticipants were attributable to con- cutoff, using a “more is better” approach in interpret- sumption of WIC foods. ing findings. None of the studies used the approach recently recommended by the IOM, which calls for Other authors also found a relationship between use of data on usual intake and comparisons to defined observed differences in nutrient intake and the types of Estimated Average Requirements (EARs) (IOM, 2001). food provided in WIC food packages. Endres et al. (1981) found that pregnant WIC participants con- Consequently, the available research provides an sumed milk, juice, and fortified cereals more often imperfect picture of the substantive significance of than pregnant nonparticipants (statistical significance

126 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program less d ct a Continued— nsume ant imp ts co n Signific Participa women t n a less d s of pregn nal] nal] nsume (1999) (1999) (1999) (1999) (1999) (1999) (1999) (1999) (1999) natio c c c natio c c c c c c n n n n n n n n n ts co la la la la la la la la la n 00) [ 00) [ l] l] l] l] l] l] l] l] l] etary intake a a a a a a a a a i Participa [nation [nation [nation [nation [nation [nation [nation [nation [nation t impact Mardis (20 Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Mardis (20 n ifica ram on the d g d No sign nsume nal] co (1999) (1999) c c ts [1 State] n n n the WIC pro la la d) [natio 3) [2 sites] 2 more/same l] l] 981) 98 a a 988 Participa [nation [nation Kramer-LeB Endres (1 Rush (1 Bailey (1 Kramer-LeB e impact of l] l] ct examined th t a nal] nal] nal] nal] nal] nal] nal] tha s [nationa [nationa [1 State] [1 State] [1 State] [1 State] [1 State] [1 State] [1 State] ant imp 79) 79) d) [natio d) [natio d) [natio d) [natio d) [natio d) [natio d) [natio 981) 981) 981) 981) 981) 981) 981) studie 988 988 988 988 988 988 988 Signific en (19 en (19 Participants consumed more Rush (1 Endres (1 Rush (1 Rush (1 Edozi Endres (1 Rush (1 Endres (1 Rush (1 Rush (1 Edozi Endres (1 Bailey (1983) [2 sites] Endres (1 Endres (1 Endres (1 Rush (1 of table. Findings from tes at end a 1 6 12 ydr 1 h energy and macronutrients See notes Table 22— Outcome Food Energy Protein Fat Saturated fat Carbo Vitamins Vitamin A Vitamin B Vitamin B Vitamin C Vitamin D Vitamin E

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 127 Chapter 4: WIC Program less d ct a Continued— nsume ant imp ts co Continued n Signific Participa women— t n a less d s of pregn nal] nsume (1999) (1999) (1999) (1999) (1999) (1999) (1999) (1999) (1999) (1999) natio c c c c c c c c c c n n n n n n n n n n ts co la la la la la la la la la la n 00) [ l] l] l] l] l] l] l] l] l] l] etary intake a a a a a a a a a a i Participa [nation [nation [nation [nation [nation [nation [nation [nation [nation [nation t impact Kramer-LeB Kramer-LeB Mardis (20 Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB Kramer-LeB n ifica ram on the d g d No sign nsume WIC pro co (1999) c ts n n la 3) [2 sites] more/same l] 98 a Participa [nation Kramer-LeB Bailey (1 e impact of the l] l] l] l] l] ct examined th t a nal] nal] nal] nal] nal] nal] nal] tha s [nationa [nationa [nationa [nationa [nationa [1 State] [1 State] [1 State] [1 State] [1 State] [1 State] [1 State] [1 State] ant imp 79) 79) 79) 79) 79) d) [natio d) [natio d) [natio d) [natio d) [natio d) [natio d) [natio 981) 981) 981) 981) 981) 981) 981) 981) studie 988 988 988 988 988 988 988 Signific en (19 en (19 en (19 en (19 en (19 Participants consumed more Rush (1 Rush (1 Rush (1 Endres (1 Edozi Edozi Endres (1 Bailey (1983) [1 site] Endres (1 Endres (1 Edozi Rush (1 Rush (1 Endres (1 Rush (1 Rush (1 Endres (1 Edozi Edozi Endres (1 Endres (1 of table. Findings from at end vin horus 1 m a esium n u esterol See notes Table 22— Outcome Folate Niaci Ribofl Thiamin Minerals Other dietary components Calci Iron Magn Phosp Zinc Chol Fiber

128 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program s ho y re s sult ific less s we ses c d s ct vitamin A a for s o a spe nsume and women w th participant ext discu s adding up all the ant imp significant. so included re nparticipant me-eligible ts co o Continued n The t cally significant findings ma port al s. rtain only t and no ting,” or Signific and inc s statisti (significant, wi significant impact ngs pe Participa ant han others. to women— rticipants s not s t findi nparticipant attern of non “vote coun rticip o n r non-WIC participant a n pa e wa refer c tice of ideration t c istent p s s betwee ate). Where ption of selenium hat text s con s of WIC pa less ce d re con ke he differen x nce 1 St s of pregn eline intake, fo r higher income n 1, a s r oid the pra ut t were analyzed separately). Re r a v nal] y fo the e table in t r b ed inta s, b s city o nsume r ud l differe (1999) hapte merit mo natio c d to a n ts co la tradict l vs. 1 n d in c es. With compa usted fo n 00) [ l] studies of the st etary intake rs a pants), al participant i cautione se note xt co r C s rtici cy, adj Te some cou Participa [nation also presented data t impact ers are Kramer-LeB Mardis (20 n er the ovided. um, and caroten r v hors example, nationa et. Both autho e pregnan research. A ifica s ings from ficant. s o y. Read wer among WI ram on the d (for not p h aut y of g kets}. wa , find d c s lo signi s e data udy id, seleni n No sign c during lat ing more than nonpa rol group m {in bra s also nsume WIC pro ted in that timates are henic a of the st co intakes n the sam , intake wa s (1999) nsideratio ot ke wa m c ts t to cont s consu bgroup cture of the bod co n ta n r la scope more/same r mean l] a sive pi solute g Participa tter compared medians. Bot acin, but point e prehen te, and the of 24-hou 999) are based o [nation ign and othe ke in ab st studies. s n hat vitamin A in sium, retinol, pant lts would be interpre Kramer-LeB da ed from treatmen e impact of the e identifies the su t, with participant ta he la s o com ov nge al. (1 r in a can paris cipants. stro r, pota . Fo s and t y’s resu earch d s al, except t ll entry also ot signifi vitamin A, and ni intake he publication f providing r coppe -LeBlanc et d on com y n nonparti ed mean es in re s from the women who m r ct es (n se c examined th rgy, single stud t a pa re identic amer r ame, t e ba (181 r r ene tha d data fo finding h we gh no differen s dy s te c ulation, the ce ant imp nd caroten author’s n effect fo repor ge of total energ 000) and K d in the interest o en thou (1998d) a nsuming less tha phasize v studie ta nts) a (2 udy pop o the stu ods, whi he former com Signific nior co also T st s porte ct, e rcen s. Because of rticipa he se sh et al. e re d no WIC Participants consumed more r entry int a pe s ticipant s a par et al. (1999) e reporte sults a d for Ru ) derlying effe han the entire re cular result han nonpa pants at n NHANES-III. un Findings from s i intake from WIC fo ydrat rs porte rtici her t a rboh with parti d sug um uming less t s Edozien (1979 For ca Note: Cell entries show t Nonsignificant Findings for Mardis and Anand Kramer-LeBlanc Findings re 1 2 Table 22— Outcome Sodi Adde subgroup rat indicate a true studies methodological limitations and em nonparticipant con nonsignificant, with were WIC pa for analysis of and niacin.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 129 Chapter 4: WIC Program not reported) and consumed greater total quantities of Given the increasing prevalence of pregnancy-associ- milk. Bailey et al. (1983) found that pregnant WIC ated obesity (Lederman et al., 2002) and the potential participants ate fortified cereals, a major source of role of the WIC program in curtailing this problem, it iron, more often than pregnant nonparticipants. is important to obtain valid estimates of WIC’s impacts on women’s dietary intakes based on more up- Results from early research do not permit an assess- to-date information. ment of the potential impact of WIC on intake of folic acid. All of the available studies were completed Nutritional Biochemistries before the recent widespread fortification of cereals Five studies examined the impact of WIC participation and grain products with folic acid and before the on nutritional biochemistries of pregnant women. The increased attention to folic acid supplementation dur- most commonly examined outcomes were hemoglo- ing pregnancy. Inadequate intake of folic acid has been bin/hematocrit and the prevalence of anemia. The two associated with neural tube defects (Centers for national WIC studies looked at hemoglobin levels and Disease Control and Prevention (CDC), 1992). reported conflicting results. The NWE (Rush et al., 1988d), which had a comparatively stronger research Findings from the recent Kramer-LeBlanc et al. (1999) design, compared final hemoglobin measurements of analysis of data from NHANES-III stand in stark con- pregnant women who were and were not participating trast to the patterns described above. In that analysis, in WIC. The analysis controlled for length of gestation the only nutrient for which a significant difference was and a number of other covariates and found no signifi- detected in median intakes of pregnant WIC partici- cant difference between the two groups. Edozien et al. pants and income-eligible nonparticipants was seleni- (1979) compared hemoglobin levels for newly um. A comparison of the nutrient intakes of WIC par- enrolling pregnant women and women who had been ticipants and the maximum nutrient contribution of the in WIC for less than 3 months and more than 3 WIC food package for pregnant women suggested that months, adjusting for a number of covariates. The WIC participants may not have redeemed all of their authors reported significant differences for both com- vouchers or consumed all of the food provided. parisons (women who had been enrolled in WIC for As noted previously, the Kramer-LeBlanc et al. analysis either length of time had significantly lower levels of was strictly descriptive and does not constitute a valid anemia than newly enrolling pregnant women). This assessment of WIC impacts. Moreover, the analysis may finding was not internally consistent with other meas- have been hampered by small sample sizes (only 71 ures of iron status included in the study, however, so it WIC participants). Nonetheless, the fact that findings must be interpreted with caution. from this study show virtually no overlap with findings Kennedy and Gershoff (1982) used multivariate from earlier studies raises a question about changes in regression techniques to compare final hemoglobin the intakes of pregnant women over time. Consequently, levels (generally measured at 34 weeks gestation or positive findings from earlier studies cannot be assumed later) among WIC participants and nonparticipants, to apply to today’s prenatal WIC participants. using the number of WIC vouchers received as the Only one study (Mardis and Anand, 2000) assessed independent variable. The authors reported that WIC intakes of prenatal WIC participants and nonpartici- participation had a positive, significant effect on final pants in relation to consumption patterns recommend- hemoglobin levels. ed in the Dietary Guidelines for Americans.65 This Using a small sample of women in their 30th week of analysis found no significant differences in intakes of pregnancy (43 participants and 58 controls), Bailey et total fat, saturated fat, cholesterol, or sodium. al. (1983) looked at biochemical indicators of iron, Moreover, with the exception of cholesterol, intakes of folate, and vitamin B status. The authors found no both participants and nonparticipants exceeded recom- 6 significant difference between WIC participants and mended levels. With regard to food intake, no signifi- nonparticipants in mean hematocrit levels. They did cant differences were detected between WIC partici- find a positive, significant difference for transferrin pants and nonparticipants in consumption of grains, saturation (a measure of iron status) and a significant, vegetables, fruits, milk, or meats and beans. negative difference for serum folacin (a measure of

65Kramer-LeBlanc et al. (1999) also report data for intake of total fat, folate status). The authors cautioned, however, that saturated fat, cholesterol, and sodium, but it is the same data reported in serum folate is very sensitive to short-term dietary Mardis and Anand (2000). intake (foods consumed shortly before the measure is

130 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program taken) and is therefore not a good indicator of long- women who are underweight at the start of pregnancy term nutriture or tissue stores of folate. The study also gain somewhat more weight than the average woman examined red blood cell folacin, a better measure of and women who are overweight at the start of preg- tissue stores, and found no significant difference nancy gain somewhat less weight than the average between WIC participants and nonparticipants. (IOM, 1990). In most recent studies of WIC impacts, weight gain, if assessed at all, has been used as a Finally, Metcoff and his colleagues (1985) examined covariate in analyses examining impacts on infant 16 different nutritional biochemistries, assessing birthweight rather than as a main outcome. change between mid- and late pregnancy. After controlling for baseline values, the week of gestation at which the first measurements were taken, and the Impacts of WIC Participation interval between measurements, the authors found no on Infants and Children significant differences between pregnant women who Although infants and children make up more than three- were randomly assigned to WIC and non-WIC groups. quarters of the total WIC population, very little research The relative paucity of research and the disparity in has been done on these participant groups until recently. design and analytic techniques used in the studies that In the late 1980s and early 1990s, FNS undertook a 5- have been completed make it impossible to draw any year effort to design and field-test a longitudinal study firm conclusions about the impact of WIC participa- of the short- and long-term impacts of WIC on infants tion on the nutritional biochemistries of pregnant and children. The University of North Carolina (UNC) women. The relationship may, indeed, be difficult to and the Research Triangle Institute (RTI) completed a elucidate. As Rush et al. (1988d) pointed out, assess- feasibility study in 1989 (Kotch et al., 1989) and a pro- ment of hemoglobin concentration, arguably the most posed a matched comparison group design. FNS had straightforward and widely used measure of nutritional some concerns about the feasibility of creating adequate- status among other population groups, is complicated ly matched groups using State vital statistics data, how- during pregnancy by numerous physiologic processes ever, and decided to conduct a field test of the proposed that are not completely understood. Rush and his col- design and to develop and test an alternative design. leagues contended that adequate assessment of iron In 1989, Abt Associates Inc. and the Johns Hopkins status during pregnancy requires the collection of sev- University completed a field test of two alternative eral, more complex hematologic indices (transferrin research designs—the original quasi-experimental saturation and serum iron) that are not readily avail- design proposed by the UNC/RTI team as well as a able in most WIC or medical records. modified experimental design (Puma et al., 1991). Weight Gain During Pregnancy Researchers used experiences from the field test, including preliminary estimates of program impacts, to Both of the national WIC evaluations (Edozien et propose a design for a national evaluation of the al.,1979; Rush et al.,1988d) examined weight gain dur- impact of WIC on infants and children. FNS was in ing pregnancy, which is known to be associated with the process of reviewing proposals submitted by adequate birthweight. Edozien et al. reported a positive research organizations interested in implementing the impact, but Rush and his colleagues found no effect. A project when Congress canceled the project. study completed in 1985 by Collins et al., as well as a stronger and more recent study by Roth et al. (2004) Today, we still do not have solid answers to many of also found no effect. the questions the WIC Child Impact Study would have addressed. But a number of recent studies have begun Assessing the impact of WIC on weight gain during to fill this critical information gap. pregnancy may be subject to considerable measurement error. In order to gauge total weight gain, pre-pregnancy Research Overview weight must be known, and in many cases, this is self- The literature search identified 41 studies that exam- reported by the woman. If pre-pregnancy weights are not ined the relationship between WIC participation and reliable, it is impossible to determine accurately how nutrition and health outcomes of infants and children. much weight was gained during pregnancy and to assess Characteristics of these studies are summarized in the relative adequacy of the weight gain. Widely accept- table 23. The two national WIC evaluations are repre- ed recommendations published by the IOM specify sented (Group I). Group II includes 16 studies that used ranges of pregnancy weight gain and recommend that

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 131 Chapter 4: WIC Program nd n n od h Continued— statistical s gressio gressio o of means a te re te re a a ison Analysis met ortions (n Multivari Bivariate t-test prop tests reported) Multivari Compar

children , y l l C

I a f mmy mmy mmy t o u u u a W

n e o s and e t g t r n a on d on d on d i p

n n g o o n

Measure of i i participation t d d p e u e l s c c a n n N/A Participati i Participati b i Participati t vs. t vs. ts, t vs. pant pant pant n n n n outcomes of infan Design partici partici partici Participa non Participa non Participa before vs. after Participa non n and health C of 2 47 I hs ) - e t dren dren men d l 88d) n e n ,006) W me- e e e 3 3 r h ) study t d l

l + i latio ges 6 d chil d chil n i 0 h

ina 0 c 4 (n= 4 of wo 24 mon d

0 e 17) e , e om samp en a l d Popu ,022) ,370) 6 itud b on nutritio i u 3 2 1 l (sample siz C and inco C infants an C and non- g I I I = i c m l n n a ( ages 2- childr infants an (n= W Rand infants an ages 0- i W long of women (se Rush et al. (19 in tabl (n= e ages 1- W t n ke n on on a 1 llecti llecti in 14 WIC progr 6 and 11 (1983) cipatio o o ics n the C enrollme f C sites and 55 I I n after take n Data source gai 4 W atal cli 3-76) 2 FITS, usual int pact o m Primary data c and a months of parti (197 in 19 WIC sites States. Data collected at time of W Primary data c in 17 pren 1988-94 NHANES-III, usual i 200 surveys d onal th ad ht, ght, ntal i al e t, hea e, arm e e and ig e e g h in tion ealth avior, amined the i n a is of nati tive he nal mistries, n ex e h t ht, variety of ht, heig ht, and h Outcome(s) d iro eral h circumferenc care, beh vocabulary, and memory measures, h weig circumferenc circumferenc skinfold thickness, immuniz status, use of preve Dietary intake, bloo Dietary intake, infant feed Dietary intak practices, he Dietary intake, weig weig nutritio bioch gen status, and de healt of table. udies tha t at end II: Secondary analys I: National evaluations nd Fox en et al. 9) 4) 4) See notes Table 23—S Study Group Rush et al. (1988c) (NWE) Edozi (197 Group Cole a (200 Ponza et al. (200

132 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program ; n n od ations h did not Continued— 4 t equ ection for s gressio gressio u as obit te re te re a a n bi ated b ent corr Analysis met Multivari Multivari implem selectio investig Bivariate t-test Ordered pr

t n: d o

en en n

n d , children—Continued

a n

d C a I e n

ior WIC mmy, mmy, mmy mmy l never , me me and r cipatio a u u u u b the past, W

e e i

l l n co co g C C and n b b C childr C childr i s and d p I I i i but l n n I I t o g g e

e on d on d on d on d i i

W l l y ,

W l ted i ted e e ibl C parti e n n s I a a - - n-W n-W l o u Measure of e e b lity an lig

participation i r o ed by i ed by i i infan g m m e i ibi v l o o v e e c c r e n n n and e Participati particip particip elig participation: I divid divid never o Participati Participati prior W p with no with no i Participati n i t vs. t vs. t vs. t vs. pant pant pant pant n n n n outcomes of Design partici partici partici partici Participa non Participa Participa non non Participa non

n and health C C C f 2 me I I I ) o of ) o

72) n me- e were n W W W 9 me- - - - e % ,7 8 r n n n 5 1 6 d o o o d in l 8 i = latio ges ges 2 inc 130% n n n 2) 0) 1 h n

n lle nutritio - ( c 5 who d d d

0 n y n n e e months months months 3 t en a en a l l a a a r Popu 5 5 ps: < 0

b b dren ages

on e i i 21,21 15,50 (sample siz C and inco C C- and inco C C v d g g I I I I I i i m o l l n a 24-6 (1) WIC sample: W (2) Full sample: W e e school, i 19-3 24-3 (n= (n= not enro poverty (n= a ages 2- childr childr grou p W W W Chil 1 ES-III WIC progr 998 CSFII

AN S the H I nd 1 f

N N

S I 1 Data source 0 N -

4-96 a 9 0 0 9 pact o 0 9 m 1988-94 199 2 1

s s u u t t a a t t s s

e n n o o i i t t ealth amined the i a a z z i i n n ex t u u Outcome(s) eral h m m m m Physician-reported gen status Dietary intak I I of table. udies tha ) t 3 0 at end 0 d 2 (

r et al. e n an n u a-Riz et al. 4) 3) 1) a n See notes e Table 23—S Study Sieg (200 Luma (200 Shefer et al. (200 Carlso S

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 133 Chapter 4: WIC Program ; n n od h Continued— s model sions s gressio gressio gres te re te re a a Analysis met ntile re Multivari (1) Multivariate regression Bivariate t-test Multivari (2) Fixed-effect qua children—Continued mmy mmy mmy u u u s and t on d on d on d Measure of participation infan Participati Participation dummy Participati Participati t vs. t vs. t vs. t vs. pant pant pant pant n n n n outcomes of Design partici partici partici partici Participa non Participa non Participa non Participa non 5 g n and health 2 ) n- n-

84) d d blin n n n e n ,652) ,509) e me- me- meat ,9 e e e g 5 2 h r r r t 1

d d d l l l g i i i latio od blin onths) n h h h i nutritio c c c r 4 (n= 5 (n= 1,302

u C and no C and no e e e 5 I I en (n= en with l l l d Popu

b b b on i i i 453 si n (sample siz C and inco C infants an C infants an C and inco C and inco r g g g I I I I I i i i m l l l o a (1) W W NHANES-III = 2,979 (12-59 m SIPP = (1-4 years) CCDP = 2,067 (2 years) e childr W pairs) same peri (n= e least 1 other si ages 1- b (2) W childr W e ages 2- W W 1 waves WIC progr 998 CSFII 6 0-9 the CDP nd 1 f Data source 3-95 SIPP 5-97 C 4-96 CSFII 4-96 a pact o m 1988-94 NHANES-III 199 199 NLSY, 199 199 199 cal, th take cial l d al ysi e r in h a ent ent tion amined the i a e tive he nal, an nal mistries, status, dental , use of n ex e opm d sug h h t Outcome(s) cognitiv devel emotio care, and p preve Dietary intake, height, weight, nutritio bioch immuniz status, genera healt healt Motor skills, so skills, and temperam Adde Dietary intak of table. udies tha ) t ga- 2 0 at end n d Sie 2) unca 0) 0) See notes Table 23—S Study Kranz an Riz (200 Variyam (20 Burstein et al. (200 Kowaleski-Jones and D (200

134 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program ; 6 n n n n n

od ) h t h did not Continued— stment for g means t i s bias gressio gressio gressio gressio gressio u ju e w ke) as

on- a d te re te re te re te re te re n a a a a a ison of ent n bi ated b a ent ad

t Analysis met h g i e h Multivari Multivari Multivari Bivariate t-test Compar (dietary int ( Multivari implem selectio investig Multivari with selecti adjustm children—Continued y mmy mmy mmy mmy : l : WIC; child recall on u u u u e e r capita C and I e s and d in ich t h on d on d on d on d pons pons ld p olle on of 4 ed W Measure of participation eho e of month infan C benefit I Participati Participati Participati Dose res hous Valu W also tested for combin Participati Dose res FSP participati Proporti was enr days on w t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant n n n n n n outcomes of Design partici partici partici partici partici partici Participa non Participa non Participa non Participa non Participa non Participa non

s C

n and health I d C C ) l 2 I 59 4 4 I ) - - - 5 o ere - W 88 ) s n n e n

4 W W h me- me- e e n 4 ,4 e ar g g and r r i

= s 3 nts and nts and d d in in l l d u n i i latio ges 2 ges 2 ges 1 ges 1 not not ( d d e o

h h lds wh t e e nutritio c c h 4 4 in a infa infa

- d in FSP p e e e e e n also 1 i en a en a en a en a l l l eho

Popu ,636) c b b b i s on ibl ibl i i i 6 180) 499) 800) t (sample siz C and non- C and inco C and inco C and non- r e g g g I I I I i i i m a g l l l a months (n= elig childr who wer breastfee childr reside e e (n= hous at least 1 other perso p ages 1- elig (n= childr months-4 ye (n= who wer breastfee e childr (n= a WIC and income- WIC and income- W W W W 1 ES-III ES-III WIC progr AN AN the H H f Data source 5 CSFII 9-91 CSFII 4-96 CSFII 9-91 CSFII pact o m 1988-94 N 198 1988-91 N 198 199 198

t h g i e e e e e w

d amined the i n a

, ex t t Outcome(s) h g i e Dietary intak Dietary intak Dietary intak Dietary intak Dietary intake, h Iron intake of table. c udies tha 98) 95) t n la ntrol at end nd en e Co 0) 0) 5) See notes et al. (1999) Table 23—S Study Oliveira a Gunders (200 Kramer-LeB Rose et al. (19 Centers for Diseas Rose et al. (19 Fraker et al. (199 (199

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 135 Chapter 4: WIC Program ge ; as- n n n n ards od h did not Continued— ing 2-sta ction-bi t s obit dels al haz gressio gressio gressio gressio e u ud pr on as d te re te re te re te re a a a a ent mo g to control for n bi ated b ent sel 8 is, incl dere oporti in Analysis met Multivari Bivariate z-test Multivari Multivari and or analys selectio model and pr models Multivari adjustm implem investig children—Continued mmy mmy mmy fined as : dium, on per nths any u u u e de s and pati t n onse: on d on d on d p pons tio tive WIC 7 a a Measure of participation ow emed infan e, high, me C vouchers I Participati (any partici year) Dose-res Number of mo W rede Participati Participati Dose res Cumul particip non and l t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant n n n n n outcomes of Design partici partici partici partici partici Participa non Participa non Participa non Participa non Participa non

n and health C C C C ) 2 I I I I hs ic ) 77 its 3 ents ents ents nths t dren f n e n 8 W W W W me- pi pi pi - - - - e 1 r eci d n n n n = d enef ly e o o o o l n i latio d chil v ( n n n n i 5) 5) 5-21,2

nutritio -59 mo e c 5 59 mon 3 4 d d d d id reci id reci id reci id b

2 - c n n n n uous e 2 e l a a a a

r Popu

b s on i 49,79 49,79 16,33 o (sample siz C and inco C C C C e g I I I I I i h m g l a W W W Medica ages 0- (n= infants an Medica ages 0- ages 1- (n= w Medica contin W W e a Medica ages 1 (n= for 4 age-sp cohorts) 0 n 92 1 C I North e in 19 2 WIC progr files for birth, birth, urce files birth birth n in ked data ked data ked data ked data id, WIC, id, Midwest na in n 9 a a 9 li a C, and d W 9 I o ing ing ing ing l lin l lin l lin l lin 1 rn i lin ois from 199 96

the 94) n bor f o ud ud n ud ud ource Reso l l l l i ina ina ina ina

a id, W id, an a Illin h 19 (19 birthday Data source n n en b i n th l 2-96) 3-97) 2-97) 2. Data bas gitud gitud gitud gitud o r pact o a m CSFII data for record, Medic (199 base, inc Lon AFDC/TANF, FSP, and WIC files for all childre born i North Caro C and Are throug for childre (199 record, Medic childr Medica records for children born in North Car base, inc Lon regio (199 Medica Lon base, inc Area Res includes data through the 5 base, inc 199 Lon ted ices al t a v to ve and costs l d n ti e e n nd al s, and e, re e amined the i ce of are-re a n ncy) , failur on a nci care ser a ex h t Outcome(s) Number of d Prevale Dental-c use of preve healt thrive, nutritional deficie anemi Health c utilizati Dietary intake visits per year Medicaid costs use of dent services (preventiv restorative, an emerge of table. ) udies tha 0 4a) 4b) t d 0 0 0 at end an Secondary analysis of State-level files n er et al. III: 3) See notes Table 23—S Study Group Lee et al. (20 Lee et al. (20 Buesch (200 Lee et al. (20 Partingto Nitzke (1999)

136 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program od h Continued— s s s ession; uare i-sq alysi alysi an an ar regr uare test Analysis met ular ch ang (2) Multivariate regression Trends Trends Chi-sq (1) Line children—Continued mmy mmy u u s and t on d on d Measure of participation infan N/A Participati N/A Participati - s vs. 5 e als p t, and ts, d by e pant h asure ce ce n outcomes of ar for ar: g n n i ng visit l sampl l an enc pecific Design der, partici Prevale type of screeni estimates for each State in year interv overal age, race/ ethnicity, gen birthwe Prevale vs. followu measures (2) Participant non by age estimates for each ye overal (1) Overall and age-s preval estimates for each ye Initial me Participa before vs. after n and health 2 59 hs hs hs ) 00 - dren dren dren t t t ) d or e il il il n 9 y 6- 0 l als 5 3) 7 , latio ges 0 10 9 0-19,5 nutritio 9 60 mon 59 mon 59 mon 48,00 4 72,98 en a = Popu ,692) oximate n on ( 5 12,00 (sample siz C infants an

I ) m 1 a Infants and ch ages 6- ( Infants and ch ages 6- month interv (n= (5,500- records per State per year) months with 3 more WIC visits at appr ages 6- records per year) Infants and ch (n= childr (2) (n= W by

, a essee 1 n SS and a 94) -84) ided t WIC progr n C C Mexico, - cky, 1980s- N a, dNSS ost data v 5 I I st data n o 1 7 t o 8 g ntu d Tenn M n the

i Ped f , , New a, Utah, and h d ee Pe a o s ants in n 90s) (m d a ed by W ed by W Data source a ase (19 a i 5-85) (Mo 4-76) W s

i C programs) d u pact o I n o m datab birth records for WIC particip Tenness programs) (197 L provid programs) (2) Linke mid-19 Oklahom Vermont (early (1) PedNSS data for Arizona, Ke Oregon, an provid PedNSS data for Color PedNSS data for Vermont (19 (most data pro W WIC records in PedNSS data for Arizon Kentucky, Tennessee, a (197 ght, i

n t i amined the i h b ce of ce of ce of g o n n n i l a a a e g ex o t w Outcome(s)

m d e n Prevale anemi Prevale anemi Prevale anemi H hematocrit, he a of table. ) udies tha t 7 at end 1) 7) 8) See notes Table 23—S Study Sherry et al. (200 Sherry et al. (199 Yip et al. (198 USDA/FNS (197

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 137 Chapter 4: WIC Program n od h Continued— s dds ratios bed gressio o d scri e te re a juste uare test d Analysis met Multivari Age-a Chi-sq Not well d children—Continued ss e who te e who mmy, mmy mmy mmy e s a e a s u u u u C subjects s and C I I t on d on d on d on d to tho n nd tho n-W t particip se of acc t perceiv Measure of participation o u o ed i infan d for W Participati Participati Participati beca did n issues a did n nee divid Participati with no ts, t vs. t vs. t vs. pant pant pant n n n n outcomes of Design partici partici partici Participa before vs. after Participa non Participa non Participa non

11 n and health C 2 59 4 I ) ps - - 2 ge; cted d e on 53) 23) nder) e n o W 1 e me- - ou ,0 ,9 n 7 5 nts o d ge latio ges 6 ges 2 tions at n

a nutritio -to-dat en wh d infa p n e e an omly sel en a en a a ger than Popu

al size gr on ibl 270) 150) (sample siz C and inco C C infants an I I I m a months (n= W childr W of childr were u Rand immuniz (n= sample (matched on ag (n= 12 months of a equ childr W elig youn months (n= is, nic on on o pol 1 of Los ck, and l centers llecti llecti a a o WIC progr not 998- a DC, s for 1 o o nea nter in Mt. e dic s) in the gton, f n les (1 antly Hispa e l record d) care ce Data source come ar les (dates h n 7-99) omin 1) pact o m Medica Ange reporte low-i Medical records for 3 WIC sites in Chicag (199 pred healt Vernon, NY (interviews and record abstractio Primary data c at urban me in Washin Baltimore, Min Boston, Little R Los Ang 200 Primary data c atus atus ived t t e d s s ht, n n amined the i ce of ld foo r-perc n e status, and a ex h eho t ht, weig Outcome(s) Heig Prevale Immunizatio Immunizatio anemi caregiv healt security hous of table. r State and local studies udies tha 04) e 02) t 0 h al. at end 98) IV: Ot een et 0) See notes Table 23—S Study Group Black et al. (20 Kahn et al. (2 Shah (200 James (19

138 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program , as n od h Continued— s ance ction bi gressio e te re of vari a uare test uare test Analysis met Multivari including attempt to control for sel Analysis Chi-sq Chi-sq children—Continued mmy mmy mmy mmy u u u u s and t on d on d on d on d Measure of participation infan Participati Participati Participati Participati , r s p ts vs. ts, t vs. t vs. pants pant pant n n n n outcomes of nd afte ate grou Design partici partici partici Participa non before a Participa before vs. after, separ Participa non Participa non

) 7 n and health 25 of 2 0 m ) = ) e , nths en 8 ed and o l n n e were n 2 me- me- me- = d C e e dr 5 I r r n der ( = nts (6 d d

l l t n i i latio d) ( h

h icity (n h nutritio -23 mo ed with of W g c c n 5 4 who oup) ; match i C chil infa

6 - I a e e e e e, gen 1 om samp l l

Popu ,225) w nos b b s on ibl i i 2 h (sample siz C and inco C and inco C and inco e t g g I I I r i i m i l l g a on ag and eth diag anemi non-W ages 1- each gr W sample e Subset of ran ages 1 (n~ Rand W elig W months ol stratified by b e a on on 74

) 1 1 ted) llecti llecti 9 Los r clinic in and 6- WIC progr - 973- esota o o n l o p areas of 1 0 a e d North 9 uth nn 9 the f owu 1 ( olis (1

a an a incom nd yo center i ap Data source n 977) i les; initi h l o r pact o a m Medical records for 1 and 1 Minne healt Medical records for 1 child a Primary data c month foll measures Ange in Florid C Primary data c in low- (dates not rep county in Mi ght, ght, ad i i d e e

, ,

n n n t i i i amined the i h b b b g o o o i l l l e g g g ex t ht, and h w o o o Outcome(s)

m m d m e e n e Serum ferritin, Dietary intake, h Hemoglobin circumferenc Dietary intake, h hematocrit, an h hematocrit, he weig hematocrit, he a of table. udies tha 85) t 86) at end nd 1) 6) See notes Table 23—S Study Burstein et al. (199 Brown a Tieman (19 Smith et al. (198 Miller et al. (19

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 139 Chapter 4: WIC Program nd of n n n od h -month Continued— statistical 3 s gressio gressio gressio o of means a te re te re te re oup ( ded” measures a a a ison ad Analysis met ortions (n Multivari Bivariate t-test Multivari Multivari prop tests reported) “value- by age gr intervals) Compar children—Continued mmy mmy mmy mmy mmy u u u u u s and t on d on d on d on d on d Measure of participation infan Participati Participati Participati Participati Participati h s or p ed” t vs. ts, t vs. t vs. t vs. d pant, pant ntrol ntrol n n n n n d outcomes of d growt ate grou Design ng co ng co partici partici Participa non Participa before vs. after, separ Participa sibli Participa sibli (“value a Participa vs. actual growth) non expecte

t n and health 2 36 lers id h ) - ) s le irs irs ho d and and a g n n e i n 3 h % of me- me- g g ib e 8 in in 5 nts ages nts and w lig ents

C pa C pa = latio ges 9 od d tod I I ted” i ted” i d n d after age d after age nths w nutritio a a ( Medic n 8; 1 10; 1

id-e infa infa n a lle lle s

e e t en a 21 sibl 19 sibl h Popu ,907) t h ng W ng W on ibl ibl 138) 1 n g (sample siz C and inco C and inco C prenatally C prenatally C- and i I I I I I o m e a W m elig for at least 75 study peri (n= childr W were o 0-11 mo elig W ages 6- “particip Sibli 1 enro 1 (n= pairs) W ages 8- “particip Sibli 1 enro 1 (n= pairs) infants an up to 20 mont with at least 2 h Medica W measurem (n=

n on on a ble ods

1 IC n i

land, 2 llecti llecti rho d y in which r WIC progr tractions tractions -W rural ies in o o o es not es not e rds in 4 l lth center b l n, CT, o d 2 o r the ount ve igh f n a ties in rd abs rd abs e nd afte n

e an o o s in Mar na (dat na (dat d) d) h WIC was ity hea n e Data source on of WIC ia ia e abl r 9-80) 4-79) d l C was not availa C and 4 non pact o i I I h m Medical records for initiati before a in New H inner-c (197 Medicaid rec counti W in whic avail reporte Louis Primary data c in 3 rural c and rec Medical records in 3 W c (197 Primary data c and rec reporte clinics in the same Boston ne in 3 cou Louis ctual and ht e l l ig es e iora amined the i id costs on care ehav ex h t ht, and a Outcome(s) 12 Medica utilizati healt Hemoglobin Hemoglobin, height, weig variety of intel and b measures IQ scores and school grad gain Expected w of table. udies tha e 5) t n o er et 98 at end e 3) S ng - z 4) ndi e u e (198 q gham (1 z See notes a Table 23—S Study V et al. (1985) Hicks and Lan Heime al. (198 Paig Hicks et al. (1982)

140 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program e at e. c -96 C od s, all h ded the d s current WI ate insuran of the 1994 v prece participant Medium = 34-66 rtion nd/or ed journal article Analysis met oblem er a r without pri Bivariate t-test ar p be Medicaid , only that po rticul peer-review n to icipants. han 66 percent, w rt ed with form children—Continued r cted a pa a mmy om the r u condu comp s and t r WIC pa s on d y wa High = more t le was limited to infants dropped f Measure of participation d. ud infan if diagnosis of re rates fo mp and were sa ero gible for WIC or kno z age Participati eeme oup s. s r ata we s. ver me the st birthday the ti e non-WIC these d nt effect d been red ts, s coded a re income-eli n outcomes of mparison g ver, cond nd th Design se hers ha mmunization co ain co cause, at who we c Participa before vs. after first and r income, a s. participation wa cipants 1994 be in more significa -height. Howe WIC vou r IC were m n and health or 2 heir ) alysi ble 6 up oxy fo e emia n activities and i r la sulted ow

n a p latio i itially s b nutritio o l se of an ad foll g ed a u Popu s ed only data f o t and weight-fo enrollment, WIC nt age in which on s 37) at model re (sample siz m ed for in the an participated in W s u s limited to nonparti m rre e sessed between t a (n= measure avai who h h Infants ages 0- months in certified for WIC beca s ver cu r WIC e wa c recall. U f WIC immunization controll s fo oup wa r le children. Th on g 1 half was a s he first insuran ed data on heigh reason e 1 through WIC progr ari 1 clinic generally intensity o be s -eligib n only t Children who ne health om ag d life and the r f comp e wa es.

. use amination Survey. y might e x Data source ams. m viduals. . 6-77) o m c ogr t thors months f pact o er in 1981 includ ship betwee r stud es that a t of private u Indi ipation. y o au m of up. Incom on. WIC records in in Fayette Co, KY (197 nt outcom c .

o h r re s b c eceip dy. a s unde ffe e umpti

r s o and Nutrition E ison g f r

bject, but s for di velopment Pr con gram Parti d Intake ssed in the first year of s e i dlers Stu r r su a y sample indicat sse l outcome d by Heimending e del of the relation on percentage in v Foo collected. R ull sample of WIC and income

compa varie amined the i b ra ud group e nd Pro less. s a with missing data o r z l i d ve nd Tod ition Surveillance Syste Child De s ex r

oup s not calls p t mplete r e Outcome(s) l p ome a at se sion for f for foo s m a Hemog ird National Health Inc hensive nfants a s ng Survey of

; re of Th e l y he sample wa ded two re income wa e fact th p applicable. e a multivariate mo n regre = 33 percent o re included in the antial problems udies tha n coded m t eding I nt. t w a s s: s description of the st we ce s r m sub ral dissertation co u rol for th s. m had bee i x se of CCDP = Comp CSFII = Continui FITS = Fe NHANES-III = NIS = National Immunization Survey. NLSY = National Longitudinal Survey of Youth. PedNSS = Pediatric Nut SIPP = Surv a nt, and Lo 9) M Information on A docto iler et al. Data sou Unless the Definition of comparison g Also estimated Roughly half of t Authors also ra WIC participation defined based To cont CSFII data inclu e Note: N/A = Not 1 2 3 4 5 6 7 8 9 10 11 12 Table 23—S Study W income level participant perce of WIC enrollme data set becau (197

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 141 Chapter 4: WIC Program data from national surveys. Group III includes nine Research Results studies that relied on State-level databases or, in one Dietary Intake: Children case, a regional database. These include WIC/Medicaid databases similar to those used in the previously summa- Table 24 summarizes findings from the 16 studies that rized research on birth outcomes, State-level files from examined the impact of WIC participation on the dietary the Pediatric Nutrition Surveillance System (PedNSS),66 intakes of children.67 The table is divided into six sec- and regional data from the Continuing Survey of Food tions: food energy and macronutrients, vitamins, min- Intakes by Individuals. Fourteen of the identified studies erals, other food components, food group servings, and are other types of State and local studies (Group IV). summary measures of dietary quality. The text follows this general organization but discusses findings for WIC research on infants and children is notably more vitamins and minerals and for food group intakes and recent than the previously summarized research on summary measures in combined sections. birth outcomes, breastfeeding, and the nutrition and health characteristics of pregnant women. Indeed, as Most of the studies completed to date are subject to the shown in table 23, there have been several very recent methodological limitations that affect most existing contributions to this literature. Of the 41 identified FANP research on dietary intakes, as discussed in chap- studies, 10 are based on data collected primarily or ter 2 and in the preceding section on impacts among exclusively in the early to mid-1990s, 10 are based on pregnant women. However, two very recent studies data collected in the mid- to late 1990s, and 3 used (Cole and Fox, 2004; Ponza et al., 2004) avoided these data that were collected exclusively in 2000 or later or limitations. Both used the approach recommended by the had data collection periods that started late in the IOM (2001) to estimate usual intakes of WIC partici- 1990s and extended beyond 2000. The relative recency pants and nonparticipants and, for nutrients with estab- of these studies is particularly important because of the lished EARs, used the recommended EAR-cutpoint increase in child participation experienced during the method to estimate the percentage of children whose early 1990s. Studies based on data collected after this usual intakes were adequate. Cole and Fox (2004) time are more likely to be generalizable to the current explored the nutrient intakes of children ages 1-4, using population of WIC children and are less subject to bias data from NHANES-III.68 Ponza and his colleagues used associated with restricted program access. data from the Feeding and Infants and Toddlers Study (FITS), which included a national sample of infants and Although some studies included both infants (under 12 toddlers ages 4-24 months. Compared with national dis- months of age) and children (1-4 years), the available tributions, the FITS sample had slightly higher incomes, research is heavily slanted toward children. Children a smaller percentage of Hispanics, and a lower rate of were included in all but 4 of the identified studies, and WIC participation (Devaney et al., 2004; Ponza et al., 22 of the identified studies focused exclusively on 2004). Although neither of these studies was intended to children. Given that children make up 50 percent of provide valid estimates of the impact of WIC on dietary the WIC population overall, this emphasis is not inap- intakes—Cole and Fox (2004) used bivariate t-tests to propriate (Bartlett et al., 2003). assess differences between groups, and Ponza et al. (2004) did not test for statistical significance—findings A number of different outcomes have been examined, are useful in providing an up-to-date perspective on the with the most common being dietary intake (17 studies), relative likelihood of adequate nutrient intakes among growth (12 studies), and anemia/iron status (16 studies). WIC participants and nonparticipants. In addition, four studies looked at general health status, as perceived by caregivers or assessed by physicians, six In reviewing findings, greatest weight is given to the studies focused on immunization status, and eight stud- study by Oliveira and Gundersen (2000), who ana- ies examined health or dental care use and/or costs. lyzed data from the 1994-96 CSFII and employed a Finally, five studies looked at developmental outcomes, unique strategy to control for selection bias. To get and two studies assessed impacts on household food suf- around the fact that the CSFII does not include vari- ficienty/security. Findings for each of these outcomes are ables that provide suitable controls for selection bias, summarized, in turn, in the sections that follow.

67A total of 17 studies examined dietary intakes, but the study by 66PedNSS is a program-based nutrition monitoring system coordinated Burstein et al. (1991) was limited to infants. by the CDC. It includes State-reported data from programs that serve low- 68Because NHANES-III collected a second day of dietary recall data for income infants and children. A majority of PedNSS data comes from WIC only 5 percent of respondents, the authors obtained the estimates of intraindi- programs. vidual variation needed to estimate usual intakes from the 1994-96 CSFII data.

142 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program l] - C less a l] - I d ct a Continued— nal] - I nsume y} ) [nation [nationa 4 natio ant imp s 0 t vert ts co 79) o n 988c) [ Signific en (19 a-Riz (20 130% p Participa {< Sieg Rush (1 Edozi and/or infan less l] - C ans} d a l] - C l] - C l] - C s of children l] - B l] - B l] - C l] - C nal] - C a a a a

} nsume y (1999) (1999) ) [nation county] - C t c c r 4 1 natio ) [nationa ) [nationa ation ation an-Americ n n e 0 ts co 0) [nationa v la la n o 000 000 00 p l] - C {4 years} l] - I {4-11 months} + 4) [nation 4) [nation

986) [ 95) [n 95) [n etary intake a a i 988c) [ % 00 00 0 3 a-Riz (20 1 Participa < { [nation {1 year} [nation C {1-4 years} t impact Cole (2 Burstein (2 Sieg Kramer-LeB CDC (19 Cole (2 {Blacks, Mexic Oliveira (2 Kramer-LeB CDC (19 Rush (1 Burstein (2 Brown (1 n ifica ram on the d I g C C C d l] - C l] - C l] - C on] - C a a a No sign l] - C l] - C l] - C l] - C l] - C l] - C l] - B l] - C l] - C nsume a a a nal] - C nal] - C nal] - I nal] - I+ nal] - I+ nal] - I+ a a a WIC pro co erty} erty} (1999) (1999) (1999) (1999) ) [nation ) [nation ) [nation c c c c ts 4 4 4 ation ation ation natio natio natio natio ) [nationa ) [nationa ation 99) [1 regi n n n n n 0 0 0 0) [nationa la la la la 0) [natio 0) [natio 000 000 more/same % pov % pov 00 (19 l] - I+C l] - I+C l] - I {2-3 months} + l] - I {2-3 months} 4) [nation 4) [nation 004) [ 004) [ n 95) [n a a a a 988c) [ 998) [n 998) [n 998) [n 988c) [ 00 00 185 185 Participa a-Riz (20 a-Riz (20 a-Riz (20 {Whites} [nation {130- C {1-3 years} [nation {3, 4 years} [nation [nation {130- Rush (1 CDC (19 Sieg Kramer-LeB Burstein (1991) [2 States] - Fraker (199 Rose (1 Rush (1 Ponza (2 Rose (1 Cole (2 Ponza (2 Rose (1 Burstein (2 Oliveira (2 Cole (2 Sieg Kramer-LeB Sieg Fraker (199 Burstein (2 Kramer-LeB Partingto Kramer-LeB e impact of the I on] - C l] - C ct examined th a t l] - C l] - C a a tha (1999) s [nationa c ation ant imp 99) [1 regi n la 79) (19 l] - I {4-11 months} 4) [nation n a studie 998) [n 00 Signific en (19 [nation {2 years} Participants consumed more Rose (1 Partingto Burstein (1991) [2 States] - Cole (2 Edozi Kramer-LeB of table. 1 Findings from at end ergy energy and macronutrients See notes Table 24— Outcome Food Food en Protein Fat Saturated fat

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 143 Chapter 4: WIC Program d e u less n d i ct t a n Continued— Co nsume (1999) c ant imp s— n t ts co la n l] - I {4-11 months} a Signific Participa [nation Kramer-LeB and/or infan less l] - C d a s of children nal] - C nal] - I nal] - I nal] - C nsume erty} (1999) (1999) (1999) (1999) (1999) ) [nation county] - C c c c c c 4 natio 1 natio natio n n n n n 0 ts co la la la la la 0) [natio n % pov l] - I {2-3 months} l] - C {4 years} l] - I {4-11 months} l] - I {2-11 months} + l] - C {4 years} 986) [ 004) [ etary intake a a a a a i 988c) [ 988c) [ 185 a-Riz (20 Participa [nation [nation {130- [nation C {1-3 years} [nation [nation t impact Brown (1 Rush (1 Kramer-LeB Fraker (199 Ponza (2 Kramer-LeB Sieg Kramer-LeB Rush (1 Kramer-LeB Kramer-LeB n ifica ram on the d I I g C C d No sign l] - C l] - C l] -C l] - C l] - C l] - C C pro l] - C nsume a a nal] - C nal] - I nal] - I nal] - I+ nal] - I+ nal] - C nal] - C a co (1999) (1999) (1999) (1999) (1999) county] - C c c c c c ts ation ation natio 1 natio natio natio natio natio ) [nationa ) [nationa n n n n n n 0) [nationa 0) [nationa la la la la la 0) [natio 000 000 more/same 00 00 l] - C l] - I {2-3 months} + l] - I {2-3 months} l] - I {2-11 months} + l] - C {4 years} 4) [nation 004) [ 986) [ 004) [ a a a a a 988c) [ 998) [n 998) [n 988c) [ 988c) [ 988c) [ 00 Participa [nation C {1-3 years} [nation C {1-4 years} [nation [nation [nation Burstein (1991) [2 States] - Rush (1 Burstein (2 Oliveira (2 Rose (1 Kramer-LeB Ponza (2 Kramer-LeB Cole (2 Ponza (2 Rush (1 Burstein (2 Oliveira (2 Kramer-LeB Rush (1 Kramer-LeB Rush (1 Kramer-LeB Rose (1 Burstein (1991) [2 States] - Fraker (199 Brown (1 e impact of the WI C C C l] - C a l] - I+ l] - I+ l] - C ct examined th a t l] - C a nal] -C nal] - I+ nal] - C tha y} (1999) ) [nation s [nationa [nationa c 4 ation natio natio ant imp n 0 0) [nationa vert la 79) 79) 0) [natio o 00 l] - I {4-11 months} + a studie 998) [n 988c) [ 988c) [ Signific en (19 en (19 a-Riz (20 130% p [nation C {1-3 years} {< Participants consumed more Sieg Rush (1 Fraker (199 Rush (1 Oliveira (2 Rose (1 Edozi Edozi Kramer-LeB of table. Findings from tes at end a 6 12 ydr h See notes Table 24— Outcome Carbo Vitamins Vitamin A Vitamin B Vitamin B Vitamin C

144 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program d e u less n l] - I l] - I d i ct t a n Continued— nal] - I Co nsume (1999) (1999) (1990 [nationa [nationa c c c natio ant imp s— n n n t ts co la la la 79) 79) n l] - I {4-11 months) l] - I {4-11 months} l] - I {4-11 months} a a a 988c) [ Signific en (19 en (19 Participa [nation [nation [nation Edozi Kramer-LeB Kramer-LeB Edozi Rush (1 Kramer-LeB and/or infan ans} less d s of children l] - B nal] - I a nsume (1999) (1999) (1999) county] - C c c c natio 1 ation n n n ts co la la la n l] - C l] - C {4 years} l] - C {4 years} 986) [ 95) [n etary intake a a a i 988c) [ hites, Mexican-Americ Participa [nation [nation {W [nation t impact Kramer-LeB Kramer-LeB Rush (1 CDC (19 Brown (1 Kramer-LeB n ifica ram on the d 3 g C d l] - C a No sign l] - C l] - C C pro l] - B l] - C nsume a nal] - C nal] - C nal] - I nal] - I nal] - I+ nal] - C nal] - C a a (1999) (1999) (1999) (1999) (1999) (1999) ) [nation c c c c c c ts co 4 ation natio natio natio natio natio ) [2 States] – I ation n n n n n n n 0 0) [nationa la la la la la la 0) [natio 0) [natio 991 more/same 00 4 l] - I {2-3 months} l] - I {2-3 months} + l] - I {2-3 months} + l] - I {2-3 months} + l] - I {2-11 months} + l] - I {4-11 months} + 4) [nation 004) [ 95) [n a a a a a a 988c) [ 988c) [ 998) [n 988c) [ 988c) [ 00 Participa a-Riz (20 [nation [nation C {1-3 years} [nation {1, 3, 4 years} [nation C {1-4 years} C {1-4 years} [nation {Blacks} [nation C {1-4 years} C {1-3 years} Kramer-LeB Rush (1 Kramer-LeB Cole (2 Ponza (2 Kramer-LeB Rush (1 Sieg Kramer-LeB Rush (1 Oliveira (2 Fraker (199 Rush (1 Burstein (1 CDC (19 Kramer-LeB Rose (1 Kramer-LeB Fraker (199 e impact of the WI C C C 2 l] - C l] - C l] - I+ l] - I+ l] - I+ l] - C l] - C l] - C ct examined th a t l] - C l] - C l] - C l] - C l] - C l] - C a a a a a nal] - C nal] - C a tha (1999) (1999) s [nationa [nationa [nationa [nationa [nationa c c ation ation ation ation ation natio natio ) [nationa ) [nationa ant imp n n 0) [nationa la la 79) 79) 79) 79) 79) 000 000 00 l] - I {2-3 months} l] - I {4-11 months} 4) [nation a a studie 998) [n 998) [n 998) [n 998) [n 998) [n 988c) [ 988c) [ 00 Signific en (19 en (19 en (19 en (19 en (19 [nation {2 years} [nation Participants consumed more Rose (1 Kramer-LeB Cole (2 Rose (1 Kramer-LeB Burstein (2 Rush (1 Edozi Variyam (2002) [national] - C Edozi Edozi Burstein (2 Rose (1 Rose (1 Edozi Rush (1 Edozi Oliveira (2 Rose (1 of table. Findings from at end vin m a u n See notes Table 24— Outcome Vitamin E Folate Niaci Ribofl Thiamin Minerals Calci

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 145 Chapter 4: WIC Program d e C u less n l] - I d i ct t a n Continued— l] - C nal] - I nal] - I+ a Co nsume (1999) (1999) (1999) [nationa c c c natio natio ant imp s— n n n t ts co la la la 79) n l] - I {4-11 months} l] - I {4-11 months} l] - I {4-11 months} 4) [nation a a a 988c) [ 988c) [ 00 Signific en (19 Participa [nation [nation [nation {4 years} Rush (1 Edozi Kramer-LeB Kramer-LeB Rush (1 Kramer-LeB Cole (2 and/or infan less l] - C d a l] - C s of children l] - C a nsume erty} (1999) (1999) (1999) (1999) (1999) ) [nation c c c c c 4 ation n n n n n 0 ts co 0) [nationa la la la la la n % pov 00 l] - I+C l] - I+C l] - C {4 years} l] - C {4 years} etary intake a a a a i l] - I {2-3 months} + 998) [n 185 a a-Riz (20 Participa [nation [nation [nation C {4 years} {130- [nation t impact Kramer-LeB Kramer-LeB Oliveira (2 Kramer-LeB [nation Kramer-LeB Rose (1 Sieg Kramer-LeB n ifica ram on the d g C d l] - C a No sign l] - C l] - C l] - C C pro l] - C l] - C l] - C l] - C nsume a nal] - C nal] - C nal] - I+ nal] - C a a a a

} co y (1999) (1999) (1999) (1999) ) [nation t c c c c r ts 4 ation natio natio ) [nationa ) [nationa n n n n e n 0 v la la la la 0) [natio 0) [natio o 000 000 more/same p l] - C {1-3 years} l] - I {2-3 months} + l] - I {2-3 months} + l] - I {2-3 months} + 4) [nation 4) [nation 4) [nation 4) [nation

004) [ a a a a 998) [n 988c) [ % 00 00 00 00 0 3 Participa a-Riz (20 1 < [nation [nation C {1-3 years} {3 years} [nation C {4 years} {3, 4 years} [nation C {1-3 years} { Kramer-LeB Kramer-LeB Cole (2 Cole (2 Rose (1 Sieg Ponza (2 Fraker (199 Cole (2 Fraker (199 Cole (2 Burstein (2 Burstein (2 Kramer-LeB Kramer-LeB Rush (1 e impact of the WI C 5 I C l] - C a l] - C l] - C l] - I+ l] - C l] - C ct examined th a t l] - C l] - C l] - C l] - C l] - C l] - C a a a a nal] - I+ a a tha (1999) ) [nation county] - C s [nationa [nationa [nationa [nationa c ation 4 ation ation ation 1 natio ) ) ant imp n 0 2 2 0) [nationa 0 0 la 79) 79) 00 l] - I {4-11 months} + 4) [nation 4) [nation 986) [ a studie 998) [n 995) [n 998) [n 998) [n 988c) [ 00 00 Signific en (19 en (19 a-Riz (20 {2 years} C {1-3 years} {2 years} [nation Participants consumed more Rose (1 Cole (2 Edozi Cole (2 Burstein (1991) [2 States] - Rush (1 Rose (1 Sieg Brown (1 Oliveira (2 Kramer-LeB Rose (1 Rose (1 Edozi Variyam (20 Variyam (20 of table. Findings from at end horus esium esterol See notes Table 24— Outcome Iron Magn Phosp Zinc Other dietary components Chol Fiber

146 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program — d d e e u u l] - C less n n a i d i t ct t n a n o C nal] - C Co nsume (1999) ) [nation c 4 ant imp s— n 0 t ts co la n l] - I {4-11 months} 02) [natio a Signific a-Riz (20 Participa [nation Sieg Kranz (20 Kramer-LeB and/or infan less d on] - C on] - C on] - C on] - C s of children l] - C l] - C a a nsume (1999) (1999) c c 99) [1 regi 99) [1 regi 99) [1 regi 99) [1 regi n n ts co la la n (19 (19 (19 (19 l] - I {2-3 months} + l] - I { 2-11 months} + 4) [nation 4) [nation n n n n etary intake a a i 00 00 Participa {4 years} [nation {total vegs} C {1-4 years} [nation C {1-3 years} t impact Partingto Cole (2 Cole (2 Burstein (2000) [10 sites] - C Partingto Kramer-LeB Partingto Burstein (2000) [10 sites] - C Kramer-LeB Partingto n ifica ram on the d g d l] - C on] - C a No sign l] - C C pro l] - C l] - C l] - C l] - C l] - C nsume a a a a a co (1999) ) [nation c ts 4 ) [10 sites] - C ) [nationa 99) [1 regi n n 0 la 000 000 more/same (19 l] - C {4 years} 4) [nation 4) [nation 4) [nation 4) [nation 4) [nation n a 00 00 00 00 00 Participa a-Riz (20 {3 years} {2, 4 years} [nation Cole (2 Burstein (2 Cole (2 Sieg Cole (2 Burstein (2 Cole (2 Partingto Kramer-LeB Cole (2 e impact of the WI l] - C on] - C on] - C a ct examined th a t l] - C l] - C a a tha ) [nation s 4 6 ) [10 sites] - C ant imp 99) [1 regi 99) [1 regi 0 s} g 000 (19 (19 4) [nation 4) [nation n n studie 00 00 Signific . a-Riz (20 e {2 years} {“other”ve Participants consumed more Partingto Burstein (2000) [10 sites] - C Cole (2 Burstein (2 {3 years} Partingto Sieg Cole (2 rvings e

at end of tabl Findings from r p s a g n les u les b s b

grou d um e d See notes d Table 24— Outcome Sodi A Food Dairy Fruit Fruit juice Fruit and vegeta Grains Meat/bea Vegeta

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 147 Chapter 4: WIC Program ts nd d y - e {in u wer n ses ticipan i o ss). r s ct ed 24 t (B = both r a group n Co selenium, and d that WIC ext discu A and C, iron, a adding up all the riables model ant imp evalence of at compa s— sugars (pa r er, ts scored l t n consumed le The t s foun ter p copp significant findings ma ental va ting,” or trients, findings are Signific m grea cipants so identifies the sub analyses th Participa ean intake rotenes, parti otes. oup for added EARs (vitamins ( r han others. s for these nu m s had a nt and/or infan attern of non “vote coun participant group(s) involved: sult rty g For s. ve re tice of rticipant ideration t c istent p s s (2001). amin D, total ca estimate con escribed in footn e), and the ation model and instru wer with established vit 30% of po o re con d rs was significa and reflect is that examined lation, the cell entry al s ants, pa IOM qu a s of children 1, a e direction. pu for r oid the pra . s v , are the 1 Stat rt point s WIC r sug y y osit y po r the <1 ngle-e epo hapte nparticip ty o fo r nutrient merit mo d to a o opp dded ci s ts scored l n . Fo entary analys 11 month not r re stud d in c - th the si 1) m sions, if an . studies rence ve to n usual intake rticipated in etary intake s 4 enti mmended b s s i cautione t bo s. as in the ut did note (200 pa co s w M some Participa ver s re O ffect dequate t impact ers are hat, relati vegetable n y children, b nt infants age r by the I he RDA. Supple C participant ifica quantile regre research. A ample, national vs. 1 cautioned tha e ings from t cates t x d significant diffe i f arch nce of a s o y. Read he method m p rather than the o rs ram on the d st bserved e e difference in intake of a ded y of g rticipa wa , find eale s ing t v r for WI grou children who ne y th s n re” ind y (for e er/sa No sign u other s re gh i C pro commen l] - C served in the a ific sub 77 percent lower prevale c oup, onl ted in that the stud r s re and nonpa w of either infant umed mo del. The autho analysi nsideratio 9. pants with oes, and s od spe d a al intakes cture of the bod co atter three, the o u e of rty g tat a r s con ve s ha ticipant s ts scored h partici ted mo intakes 4) [nation ks-only scop s n y sive pi c 00 For the l % of po energ ed using meth with intakes belo prehen model. Variations ob ate, the n WIC par ium in gm, mean was lowe etween groups. ign and othe 75 quantile only. rrent WIC 185 rate sna in st studies. c mes, white po s s “participant Participa of adequate u b pertain only to lts would be interpre s Cole (2 at participant , e e impact of the WI on s cu e icipants. s com 30- e on d c ion-bias-adju epa ce nge egu u rt c ssi s a s th ngs betwee e 1 Th stro total cal gre alen y’s resu v earch d r ison of select r s ces For th ” mean pre er intakes, estimat l the publicati oup differen feren OLS re f providing the r , as well l less e of differen a l intakes. A gnificant at all quantiles except 0. gnificant at the 0. gh sed on compa i s Where findi al intakes. c u es in re d d s from the ct si si u s c on examined th among WIC pa nts. more). single stud ota a s t u percentage of infant d en-g lly t e ba r c ume name, ater s nding s kilocalories. Fo sed on tha base e for t umed fi gh no r differen s s s significan vegetable ant imp e ba ssed the re gre r ts scored h statistically statistica con n author’ s 004) refle r sse 02) are as a er 1,000 d in the interest o c) a en thou al. (1991) a (2 of adequate u and I = infants). phasize v a studie cipants con ants an non-WIC infa ported significant dif a rk-green Signific z (2004) a e w e w ce z significant betwe c c re senio porte ct, e ticip both studies estimate he statistical r s. Because of alen Participa yam (20 v sh (1988 e re d no o, intakes we s, which calcium p r and “parti calcium th s s s, differen s, differen 04) and Pon reporte sults a analysi d for Burstein et d for Siega-Ri d for Ru d for Vari g is for ) derlying effe 004, only)), to the pre re cular result sible results. d less and iron (pa did not test t he first tw asure un r t Findings from porte porte porte porte ual intake za, 2 children, C = me s s deep-yellow and da Cole (20 y Cell entries show the }. with parti s consume d relative ced implau umed less) s Edozien (1979 In quantile analysi Based on main Reported findin In quantile analysi Exclude Notes: Nonsignificant Data for Ponza (2004) Findings re Findings re Kramer-LeBlanc (1999) also Findings re Findings re 1 2 3 4 5 6 Table 24— Outcome Summar Total HEI score infants and bracket indicate a true studies methodological limitations and em protein (Pon reporte adequate u con potassium. Fo produ hour intakes. infants

148 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program as reported by previous researchers (Fraker, 1990), Only Cole and Fox (2004) reported a significant dif- Oliveira and Gundersen limited their analysis sample ference (based on bivariate t-tests), and it was limited to WIC participants and low-income nonparticipants to 2-year-olds. who lived in households where at least one other member was on the WIC program. The rationale for Results for protein are equivocal. Oliveira and this restriction was that it effectively controlled for key Gundersen (2000) found no significant difference in sources of selection bias, including lack of awareness intakes of participants and nonparticipants. However, of the WIC program and resistance to participation the earlier study by Rose, Habicht, and Devaney because of stigma or other reasons. The authors (1998) found that WIC participants consumed signifi- acknowledge that two important sources of potential cantly more protein than nonparticipants. It is possible bias remain, both of which are associated with that the effect on protein intake may be small (and rationing rather than self-selection. The income-eligi- therefore not detected by Oliveira and Gundersen) and ble nonparticipant group may have included (1) chil- limited to the lowest income participants. dren who were not actually eligible for WIC because they did not have a certified nutritional risk and (2) Relatively few studies have examined intakes of fat, children who were fully eligible but could not partici- saturated fat, and carbohydrates, and Oliveira and pate because the local WIC program had no available Gundersen (2000) is not among them. Consequently, slots. Both of these sources of bias would tend to the best available data in this area comes from a study underestimate program impacts. by Siega-Riz et al. (2004). Siega-Riz and her colleagues used the 1994-96, 1998 CSFII dataset to assess intakes A downside to the approach used by Oliveira and of children who were not enrolled in school and who had Gundersen is that it severely limited sample sizes. The household incomes below 185 percent of poverty. The final analysis sample included only 180 children, where- authors did not attempt to control for selection bias, as the full 1994-96 CSFII database included 1,206 acknowledging the limitations of the CSFII data. To children who were either enrolled in WIC or were provide a somewhat greater level of control for unmea- income-eligible. The small sample size means that the sured differences between groups, they completed sep- analysis was likely able to detect only large differ- arate analyses for children with incomes below 130 ences. This limitation, in combination with the remain- percent of poverty and children with incomes between ing sources of selection bias, means that the study pro- 130 percent and 185 percent of poverty. They also vides a fairly conservative estimate of WIC’s effects. included a sizable number of covariates in their models, including variables that controlled for mother’s age, Another limitation to the Oliveira and Gundersen television viewing, use of dietary supplements, presence study is that it examined a relatively limited set of of dietary restrictions, and enrollment in child care. nutrients. For most nutrients not examined by Oliveira and Gundersen, the strongest available evidence comes The analysis revealed no significant differences in from a study completed by Rose, Habicht, and intakes of fat, saturated fat, or carbohydrates (expressed Devaney (1998), who used data from the 1989-91 as a percentage of total energy intake) between WIC round of the CSFII. This study may overstate WIC participants and nonparticipants with household impacts, however, because the authors did not control incomes between 130 and 185 percent of poverty. for selection bias (they report that they “found no evi- However, among lower income children—those resid- dence of it”) and limited their sample to children in ing in households with incomes below 130 percent of FSP-eligible households (household incomes below poverty—WIC participants consumed significantly 130 percent of poverty). This sample represents the less fat and significantly more carbohydrates than non- lower end of the income distribution of WIC partici- participants. These suggestive findings would be more pants and children in this group may benefit from convincing if they were replicated in the restricted WIC’s supplemental food benefit more than higher sample analyzed by Oliveira and Gundersen. income children would. Vitamins and Minerals. Giving precedence to Oliveira Food Energy and Macronutrients. The evidence sug- and Gundersen, there is strong evidence that WIC par- gests that WIC participation does not affect children’s ticipation increases children’s intakes of vitamin B6, intakes of food energy. Although WIC participants tend folate, and iron. The evidence that WIC increases chil- to have greater energy intakes than nonparticipants, dren’s iron intake is particularly strong. Almost all of these differences tend not to be statistically significant. the identified studies assessed iron intake, and all but

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 149 Chapter 4: WIC Program one of the studies that used multivariate analysis tech- inadequate usual intakes of riboflavin, thiamin, or niques found significantly greater intakes among WIC magnesium, 3 percent had inadequate usual intakes of children. Consistent results for Oliveira and Gundersen niacin, and 58 percent had inadequate usual intakes of (2000) and Rose, Devaney, and Habicht (1998) also vitamin E. The authors urged caution in interpreting strongly suggest that WIC participation does not sig- the finding for vitamin E, given that clinical data from nificantly affect children's intakes of vitamin A, vita- NHANES-III do not indicate problems with vitamin E min C, or calcium. status. They suggested that the high prevalence of apparently inadequate vitamin E intakes may be asso- For other vitamins and minerals, evidence of a signifi- ciated with the difficulty of assessing the types and cant WIC impact is less clear. Rose, Habicht, and amounts of fats and oils used in cooking and/or with Devaney (1998) reported a significant impact on chil- variability in food composition databases. dren’s intake of zinc, while Oliveira and Gundersen found no such effect. Rose, Habicht, and Devaney also Data from Cole and Fox (2004), Devaney et al. reported significant impacts for several nutrients that (2004), and Ponza et al. (2004) suggest that the preva- were not included in the Oliveira and Gundersen study, lence of inadequate nutrient intakes among very young including vitamin E, niacin, riboflavin, thiamin, and children is low, and that today’s WIC children are magnesium. In all cases, mean intakes were greater for doing as well nutritionally as their nonparticipating WIC participants than for nonparticipants. These find- counterparts. However, the fact that the descriptive ings suggest a WIC impact among the lowest income analyses completed by Cole and Fox (2004a) and children but would be more convincing if they were Ponza et al. (2004) did not reveal meaningful differ- replicated in the restricted sample used by Oliveira and ences in the prevalence of nutrient inadequacy among Gundersen. WIC and non WIC children does not necessarily mean that the WIC program has no impact on children’s As noted previously, increased nutrient intake by par- diets. It may be, for example, that WIC is responsible ticipants does not necessarily mean that participants for bringing intakes of participating children up to the are more likely than nonparticipants to have adequate level of other children. The question of WIC impacts diets. Recent data on usual nutrient intakes of age- cannot be assessed even at a basic level without multi- eligible children indicate that the vast majority of both variate analysis techniques that, at a minimum, control WIC and non-WIC children have nutritionally ade- for measured differences between the two groups. quate diets. Cole and Fox (2004) found that virtually all children ages 1-4, regardless of WIC participation Other Dietary Components. Information on the impact status, had adequate usual intakes of iron and zinc. of WIC on other dietary components, including choles- Ponza et al. (2004) reported similar findings for iron terol, sodium, fiber, and added sugars, is very limited. for children ages 1 and 2.69 The majority of studies that looked at these components were descriptive studies that assessed differences Neither Cole and Fox nor Ponza et al. assessed intakes between groups with bivariate t-tests or did not assess of vitamin B6 or folate (the other two nutrients found statistical significance. to be significant in Oliveira's and Gundersen's analysis). However, findings from the main FITS analysis, There is no convincing evidence that WIC participa- which did not differentiate children by WIC participation influences children’s intakes of cholesterol, sodi- tion status, showed that less than 1 percent of all 1- um, or fiber. There is suggestive evidence, however, and 2-year-olds had inadequate usual intakes of vita- that WIC participation decreases children’s consump- min B6, and only 2 percent had inadequate usual tion of added sugar. Using data from the most recent intakes of folate (Devaney et al., 2004). The main CSFII, Siega-Riz et al. (2004) and Kranz and Siega- FITS analysis also provides information on the other Riz (2002) both found that WIC children consumed nutrients for which Rose, Habicht, and Devaney significantly less added sugar than non-WIC children. (1998) reported a significant WIC impact. FITS found In the Siega-Riz et al. study, this difference was that less than 1 percent of children ages 1 and 2 had assessed based on the percentage of food energy provided by added sugar and was significant for two different income samples (<130 percent of poverty and 69 As discussed later in this chapter, the adequacy of children's iron 130-185 percent of poverty). In the Kranz and Siega- intakes is consistent with declining levels of anemia in this population and may reflect an indirect effect of the WIC program on the availability and Riz study, the outcome measure was teaspoons of use of iron-fortified breakfast cereals. sugar per 100 kilocalories and the difference was also

150 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program observed for two different income groups (<130 supplementary analyses that used mean intakes found, percent of poverty and ≤185 percent of poverty). like Rush et al., that WIC infants consumed signifi- While suggestive of a positive WIC effect, the evi- cantly less calcium than non-WIC infants. dence would be more convincing if it were replicated in the restricted sample used by Oliveira and For the NWE, Rush and his colleagues completed a Gundersen (2000). detailed analysis of the sources of nutrients in infants' diets and found that the greater iron intakes and lower Food Group Intake and Summary Measures of calcium, magnesium, and phosphorus intakes noted for Dietary Quality. Data on the impact of WIC participa- WIC infants were related. All of these findings were tion on children’s food intake or on overall dietary associated with an increased use of cows' milk among quality are also very limited. Most of the studies that non-WIC infants. Because the American Academy of looked at these outcomes used simple bivariate t-tests. Pediatrics recommends that cow’s milk not be fed to And, as table 24 clearly illustrates, the overlap in sig- infants younger than 12 months, the lower intakes of nificant findings across studies is small. The available calcium, magnesium, and phosphorus among WIC data are too limited to support even tentative conclu- infants were not interpreted as negative impacts. sions about WIC impacts in these areas. Burstein and her colleagues found a similar pattern. Specifically, they found that, among nonbreastfed Dietary Intake: Infants infants, WIC infants were more likely to receive for- Five of the identified studies reported separate esti- mula and non-WIC infants were more likely to receive mates of WIC's impact on the dietary intake of WIC cow’s milk. Moreover, among formula-fed infants, and non-WIC infants (table 23). This includes both WIC infants were more likely to receive iron-fortified national WIC evaluations (Rush et al., 1988c; Edozien formula and non-WIC infants were more likely to et al., 1979), the field test of the WIC Child Impact receive formula that was not fortified with iron. Study (Burstein et al., 1991), and the more recent descriptive studies completed by Kramer LeBlanc et Recent descriptive studies provide some evidence that al. (1999) and Ponza et al. (2004). Cole and Fox differences between WIC infants and non-WIC infants (2004) looked at reported infant feeding patterns of in the use of cow’s milk may persist today. For exam- WIC and non WIC infants but did not examine dietary ple, Kramer-LeBlanc and her colleagues (1999) found intake per se. that, among infants ages 4-11 months, WIC participants consumed significantly less protein, calcium,

Of the available studies, the strongest are the field test magnesium, riboflavin, vitamin B12, and sodium. All of the WIC Child Impact Study (Burstein et al., 1991) of these nutrients occur in greater concentrations in and the NWE (Rush et al., 1988c), although both have cow’s milk than in iron-fortified infant formula. In methodological limitations. As shown in table 24, both addition, Cole and Fox (2004) analyzed the infant the NEW and the field test of the WIC Child Impact feeding inventory in NHANES-III and found that Study found that WIC infants had significantly higher WIC participants were significantly less likely than intakes of iron than non-WIC infants. Ponza et al.'s nonparticipants to be fed cow’s milk before 12 (2004) recent assessment of usual nutrient intakes months of age. found that WIC infants ages 7-11 months had greater mean usual intakes of iron than did nonparticipant In an analysis of 24-hour intakes, Ponza et al. (2004) infants and, more importantly, that the prevalence of found no significant difference between WIC infants adequate usual iron intakes was greater for WIC and non-WIC infants in the percentage consuming infants than for non-WIC infants (99 percent vs. 90 cow’s milk. In addition, findings from an inventory of percent). The statistical significance of these differ- feeding practices that assessed whether an infant had ences was not tested. ever been fed cow’s milk found no difference between WIC and non-WIC infants ages 7-11 months. Reported The NWE also found that WIC infants consumed sig- feeding of cow’s milk was rare among younger infants nificantly less calcium, magnesium, and phosphorus (4-6 months). In this age group, however, significantly than non-WIC infants. Burstein and her colleagues more WIC infants than non-WIC infants had been fed (1991) reported no impact on calcium in their main cow’s milk at some point. These results should be analysis, which assessed the percentage of infants con- interpreted with caution because the comparison group suming less than 77 percent of the RDA; however, used in Ponza et al.'s analysis included all income levels, which may obscure differences between WIC

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 151 Chapter 4: WIC Program participants and income-eligible nonparticipants, who the percentage consuming ready-to-eat cereal was constitute a more appropriate comparison group. 77 percent lower for WIC participants than for nonparticipants. Burstein and her colleagues (1991) also found that WIC participation was associated with more appropri- Growth ate introduction of solid foods. WIC infant feeding A total of 12 studies attempted to measure the impact guidelines, which are based on recommendations of of WIC on the growth of infants and children (table the American Academy of Pediatrics and other expert 23). Findings from these studies are summarized in groups, recommend that no solids be introduced until table 25. Note that the far-left column of the table, infants are at least 4 months of age. Indeed, the WIC labeled “Participants higher,” includes findings that food package for infants younger than 4 months is lim- can be considered both positive and negative. For ited to iron-fortified formula (USDA/FNS, 2003c). example, greater height or length-for-age among WIC Burstein and her colleagues found that nonparticipant participants would generally be considered a positive infants were significantly more likely than WIC finding, while a greater prevalence of overweight infants to be fed solid foods before 4 months of age. would be considered a negative finding.

It is not clear whether this finding still holds for Many of the earliest efforts to assess WIC’s impact on today’s WIC infants. Based on the infant feeding children’s growth were hampered by technical difficul- inventory in NHANES-III, Cole and Fox (2004) found ties such as missing or inaccurate data in medical no difference between WIC participants and nonpartic- records or WIC files (Heimendinger et al., 1984; ipants in the percentage of infants or children who USDA/FNS, 1978) and problems with equipment cali- were fed solid foods before 4 months of age. Similarly, bration (Burstein et al., 1991). Self-selection issues Ponza and his colleagues (2004) found no differences have also affected this research. In the NWE, Rush between WIC participants and nonparticipants in the and his colleagues (1988d) reported differential mean ages at which infant cereal and pureed baby recruitment of children with abnormal growth (over- foods were introduced. These data may be less reliable weight, underweight, or stunted) into WIC, in keeping than the data from the Burstein et al. study, however, with the program’s focus on individuals with identifi- because they are based on a more extended recall peri- 70 able nutritional risks. This pattern of self selection is od. In addition, as noted previously, the all-income likely the reason for the significantly greater preva- comparison group used by Ponza and his colleagues lence of underweight and growth retardation among may obscure differences between WIC participants and WIC children reported by Cole and Fox (2004) and income-eligible nonparticipants. Burstein et al. (2000) in their more recent descriptive analyses of NHANES-III data. Kramer-LeBlanc et al. (1999) found that carbohydrate and fiber intakes among infants 4-11 months were sig- In the 1991 field test of the WIC Child Impact Study, nificantly lower for WIC participants than for income- Burstein and her colleagues (1991) explicitly attempt- eligible nonparticipants and suggested that this pattern ed to control for selection bias. In their final report, may be associated with earlier introduction and greater however, they present results from both the single- consumption of cereal among non-WIC infants. Data equation models and the instrumental-variables models from Ponza et al., suggest that the difference in cereal because there was some concern about the perform- consumption may be concentrated among older infants ance of both models. As shown in table 25, the instru- and, therefore, not associated with better adherence to mental-variables model found that WIC and a signifi- infant feeding guidelines, per se. Ponza and his col- cant negative effect on infants’ length-for-age. The sin- leagues found no difference between WIC participants gle-equation model found that WIC participation had a and nonparticipants in consumption of either infant significant and negative effect on head circumfrence. cereal or ready-to-eat cereal among infants ages 4-6 months. Among infants ages 7-11 months, however, Heimendinger et al. (1984) attempted to compensate for problems of self selection as well as the potential for regression to the mean in a longitudinal data set by determining the expected rate of growth and compar- 70 The Burstein, et al. (1991) study was limited to 6-month-old infants, ing “value added” measures for WIC and non WIC so caregivers reported on relatively recent feeding practices. The NHANES-III infant feeding histories analyzed by Cole and Fox (2004) children who had at least two weight and height meas- included infants up to 12 months. urements. She demonstrated a positive WIC effect on

152 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program 1 I C due C due I I ct a wer Continued— nal] - C ts lo h no W h no W n ate] - B t natio ant imp d} - I ed wit ed wit 988c) [ Signific Participa es 000) [1S {compar {compar to no nee to access problems} - I Black (2004) [6 cities] Black (2004) [6 cities] Burstein (1991) [2 States] - Rush (1 Lee (2 ed I ed} - I l] - C wer l] - B l] - C nal] - I nal] - C d developmental outcom a a ts lo cans} e to no ne i an n cities] compar natio natio ) [2 States] - ) [nationa ation C du I -Amer 991 000 4) [nation n 04) [6 95) [n 988c) [ 988c) [ Participa 00 with no W {Mexica t impact Burstein (1 Rush (1 Cole (2 Rush (1 Burstein (2 CDC (19 Black (20 n nutrition, health, ifica other e I I 4 5 No sign /sam l] - B l} - C nal] - I nal] - B nal] - B a a gher ounty] - C ounty] - C counties] - C natio natio c c natio ) [2 States] - program on ation ation ts hi C n I 991 2) [3 4) {n 986) [ 986) [ 95) [n 988c) [ 988c) [ 988c) [ 00 Participa {Blacks, Whites} Brown (1 Rush (1 Burstein (1991) [2 States] - Rush (1 Burstein (1 Hicks (198 Rush (1 Black (2004) [6 cities] - I CDC (19 Cole (2 Brown (1

the impact of W - 2

] s e t l] - C l] - B l] - B l] - B l] C due a I ct t ) a 5 S gher

l] - C 8 4 a 1) [3 areas [ 9

1 ) 1 month of ( ts hi 8

h no W n n [nationa [nationa [nationa [nationa cities] counties] - C site] - C 7 e ) [nationa ant imp 9 n 1 o 79) 79) er (198 79) 79) (

e 000 2) [3 ed wit 4) [nation S S ng 04) [6 d withi - N 995) [1 z 00 Signific lle F Participa ndi e / en (19 en (19 en (19 en (19 u A q D z to access problems} - I birth} - I {compar {enro in 1 State] - B [1 site] - B S a dren Black (20 Edozi Hicks (198 Edozi Burstein (2 Edozi Miller (1 Heime Smith (1986) [1 site] - B V Cole (2 Weiler (1979) [1 site] - I U Edozi of table. it/ or chil r / ht d ig e 3 da- e at end ure 3 g gth ht ce/ ce/ ce of ce of matoc bin or ight es of iron status n n n n e ig od of od of lo g ht/len iho iho erwe ight e See notes Table 25—Findings from studies that examined of infants an Outcome Growth Expected w gain Heig W Prevale likel und Prevale failure to thrive Prevale growth retar tion/stuntin Head circumferenc Prevale likel overwe Measur Mean h hemo other meas

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 153 Chapter 4: WIC Program

6 -

] s e t 7 l] - B a ct t ) a wer 5 S Continued—

l] - B l] - C 8 nal] - B nal] - B 8 4 a a [ 9

ts} 1 ) ts lo i ( 8 year}

n [nationa onal] – B 7 e ) [2 States] – I ant imp 9 n 1 d 1 o 79) (

e 991 01) [natio 97) [natio a) [1 State] - C 4) [nation 4) [nation S S - N z 00 00 Signific 87) [nati Participa F es 004 e / en (19 u A q D z {emergency vis {infants an [1 site] - B S a Sherry (20 Lee (2 Cole (2 Sherry (19 Burstein (1 U Cole (2 Yip (19 V Edozi ed l] - C wer l] - B l] - C l] - C d developmental outcom a a a ts lo e to no an n years} cities] {compar counties] - C sites] - B ) [nationa C du d 3 I 000 2) [3 4) [nation 4) [nation 4) [nation 04) [6 002) [3 Participa 00 00 00 d} - I {2-4 years} with no W nee {infants an t impact Cole (2 Black (20 Cole (2 Cole (2 Kahn (2 Hicks (198 Burstein (2 n nutrition, health, ifica other e No sign l] - C l] - C l] - C /sam l] - B l] - C l] - C a a a gher ation ) [nationa ) [nationa ) [nationa program on 0) [1 city] - C ts hi State] - B C n I ears} 00 3) [1 State] - I 000 000 000 4) [nation 4) [nation 988) [n 00 00 000) [1 een (2 e (198 Participa {MMR} {past year} {1 and 2 y Burstein (2 Burstein (2000) [10 sites] - C Shah Cole (2 Burstein (2 Rush (1 Burstein (2 Lee (2 Paig Cole (2 James (1998) [1 site] - C the impact of W C due l] - C I ct a l] -C gher l] - C l] - C l] - C {ever} nal] - C nal] - C a a a a ts hi h no W n ation [nationa 3) [1 State] - C ant imp State] - B 3) [natio 00) 01) [natio a) [1 State] - C ed wit 4) [nation 4) [nation 4) [nation 988) [n er (200 00 00 00 Signific (200 Participa n (20 000) [1 004 n tion/health/development to access problems} - I {compar {all others} {4 years} dren—Continued Black (2004) [6 cities] Buesch Lee (2 Burstein (2000) [10 sites] - C Rush (1 Lee (2 Carlso Cole (2 Luma Shefer (20 Cole (2 Cole (2 of table. ia ed on or chil t r / es of nutri m d e n at end lth ce/ ce of i r-repor n n e on of od of ncy an care or care a l h iho See notes Table 25—Findings from studies that examined of infants an Outcome Prevale likel anemi Prevale deficie Other measur Health status - caregiv Health status - physician assessed Immunizatio status Dental hea status Utilizati healt denta services

154 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program s C y I st {in tantial ses -year- s s ct (B = both a roup wer s. i s with no W ts lo Medicaid co } - I d nd a sub n d anemia was ext discu oduced adding up all the ant imp r 8 erence for3 the subg een initial and first nee w The t when b) [1 State] - B ent} 1} l] - C {difficult iate analy significant findings ma s. Diff a bet e-related r ar ting,” or s Signific Participa es 004 IC compare so identifies multiv [nation due to no {W {ages 0- ntal-ca eight-for-height a gnificance ic han others. djusted model p temperam Kowaleski-Jones (2000) Black (2004) [6 cities] Lee (2 f w participant ry al de difference f si y attern of non “vote coun n participant group(s) involved: a speci r-old - ction-a tice of yea ideration t c istent p - s s con e), and the ell just short o d the sele rticipation. re con ht-for-age or low 1- and 2 1, a wer ulation, the cell ent s r oid the pra nce f re largely limited to el an P pa v 1 Stat well as the WIC e d developmental outcom r nd a ts lo likelihood of having an y pop n h we hapte ty o merit mo c S as d to a ci ined s] - C stud m d in c e low length/heig a b) [1 State] - B {age 2} . The differ prenatal FS l] - C {motor and studies quation mod x for SE months) a ntire a ges, whi Participa cautione note n 004 s ntrol idered some cha s s that e single-e p to 12 [nation social skills} [all sampl e anemia rolled for r t impact ers are han the e con Kowaleski-Jones (2000) Burstein (2000) Lee (2 n nutrition, health, analysi ifica research. A ample, national vs. 1 lly significant. e levels. ings from were so cont measu x l both the rather t y. Read nd did not co es. C r tica ho other I y of ) to s ese apparent wa up , find aselin parate s ring infancy (u n y (for e No sign stati s du ems} - I gher subgro sts. Se r, that th al. (1987 s model that a ted in that ge. with no W cautioned that the stud ts hi turned to b nd children w n mean z-sco was not probl d nsideratio p et n ect weve r-a rs e program on o cture of the bod co s a caid co r re c e of articipation. es i eff specific C r - c I 3) [1 State] - I C p b) [1 State] - B {age 3} ren scop sive pi dental care cost limited to WIC children a luded, h Participa 004 c IC compare e (198 on a fixed ber of infant con ps in length-fo ia defined by Yi -related Medi prehen p differences. ate, the s or diffe due to access {W ign and othe m r tion; thereafte odel. The autho e st studies. s rs was not to have u r but the differen ertain only to a lts would be interpre Lee (2 Paig Black (2004) [6 cities] o e a ased h p com on d crite nge -grou that a ted m . s rather than WI aut s een grou ). stro participa cipants 2 in the n ) are b y’s resu earch d ) ysis s s the impact of W ement between r defined cutoff e betw c the publicati f providing ow n (2000 ds anal en using the onths of y significant obin. es in re a height. The Where findings ct s from the n nonparti c c height(cm wh single stud a oglobin percentile, a bel s for total dental-c l] - C r gher 10. a name, s the selection-adju the tren ve o d Dun finding initial measu result ts hi hemogl r tantial decrease gh no x (100/ r differen s s n ) ation significance of 3) [1 State] - C ant imp rs in significant gnificant differen author’ t (kg r d sub ased on si er mean hem h as orted fo ngs at p <0. d in the interest o en thou ) reflect and I = infants). atocrit o but not statisticall phasize v rtions abo d high weight for 988) [n re more likely th atistical er (200 e b eig Signific only during first 6 m e rep e w r Participa r ski-Jones an senio c (w porte y foun opo ct, e r s we ble to erro (2004b he st s. Because of of hem e re stud s r parent dren—Continued slightly low direction, Hicks (1982) [1 site] - C Hicks (1985) [1 site] - C Buesch Rush (1 conside et’s Index l not test t sitive impact sults a arison of p . d significant findi s had d for Lee d for Kowale was ap s or chil derlying effe participant the basi re cular result t / umber d l ere attributa e -group differen d u c o NS (1978) un w s o F porte porte comp Quete f e

r mparable in children, C = ental

d l e l Cell entries show the o }. b that WIC with parti i s co s h s opm se in the n e u s a l u Single-equation model found no The differen Based on Based on WIC participant The between A significant, po Authors reporte p Notes: Nonsignificant Findings re Findings re Findings for Burstein (1991) a Paige (1983) did The USDA/ 1 2 3 4 5 6 7 8 o m Table 25—Findings from studies that examined of infants an Outcome Health care/dental care costs H security Devel outcomes infants and bracket indicate a true studies methodological limitations and em showed olds wa i increa followup visits defined solely on

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 155 Chapter 4: WIC Program the rate of weight gain among infants and toddlers, but For both WIC infants and infants who were not partic- had to abandon the analysis of impacts on linear ipating in WIC because their caregivers did not per- growth because of problems with missing data. ceive a need for WIC services, z-scores for weight-for- age and length-for-age indicated that children were The strongest and most recent data on WIC’s impacts growing normally. In contrast, z-scores for both of on the growth of infants and children come from studies these measures were below normal for infants who by Black et al. (2003) and Lee et al. (2000). Black et were not participating in WIC because of access prob- al. reported data from the Children’s Sentinel Nutrition lems, indicating that these infants were both under- Assessment Project (C-SNAP). C-SNAP included low- weight and short, relative to national norms. income infants whose caregivers were recruited at six urban medical centers. WIC participation status was Multivariate analyses revealed that WIC infants were not a criterion for enrollment into the study and was significantly longer and significantly less underweight not known at the time of recruitment. The study did than infants who did not participate in WIC because of not collect information on household income, so the access problems. Infants who did not participate in authors used the absence of private health insurance as WIC because their caregivers did not perceive a need a proxy to identify a comparison group of income-eli- for WIC services were comparable to WIC infants in gible nonparticipants. The authors did not control for weight-for-age, but were significantly longer than WIC selection bias but collected information on reasons for infants. There were no differences between WIC par- nonparticipation. They used these data to divide ticipants and either group of nonparticipants in the infants who were not participating in WIC into two prevalence of overweight. The unadjusted prevalence groups: those who did not participate because their for all three groups of infants exceeded the 5 percent caregivers had problems accessing the program (64 that would be expected based on national norms (7-9 percent of the total) and those who did not participate percent). because their caregivers did not perceive a need for WIC services (36 percent). Reported access problems The authors concluded that WIC participation has a included being on a waiting last, scheduling difficul- positive impact on infant growth. While acknowledging ties, missed appointments, lack of time to pick up the potential for selection bias, the authors suggest that vouchers, relocation, and lack of transportation. There findings from the comparison of WIC participants with were noteworthy differences in the sociodemographic the slightly more advantaged comparison group characteristics of the three groups. Caregivers who (infants whose caregivers did not perceive a need for reported no need for WIC services were more likely WIC services) strengthen this conclusion. Because of than either caregivers of WIC participants or care- differences in sociodemographic characteristics as well givers who reported WIC access problems to be as a difference in household food security (WIC partic- employed, married, and White and were less likely to ipants were more likely to be food insecure), one be receiving subsidized housing, TANF, or food might expect this group of nonparticipants to do better stamps. Caregivers who did not participate in WIC than WIC participants. However, both groups had because of reported access problems had lower rates of comparable weight-for-age and, although nonpartici- employment than WIC participants as well as lower pants were significantly longer, both groups were rates of participation in other assistance programs. growing normally. The authors also point out that infants who did not participate in WIC because of The study examined z-scores for weight-for-age, length- access problems had a higher rate of breastfeeding ini- for-age, and weight-for-length, using age- and gender- tiation than WIC participants (62 percent vs. 54 per- specific norms. Because data were collected in medical cent). Because healthy breastfed infants generally settings, information was collected on hydration status, grow more rapidly than nonbreastfed infants during which can dramatically affect an infant’s weight. The the first two months of life, the unadjusted z-scores authors used multivariate models to estimate differences observed between these two groups are contrary to between WIC participants and each of the nonparticipant what one would expect to see (WIC infants were groups. In addition to sociodemographic characteristics, growing normally, but infants in the WIC-access-prob- models included variables to control for maternal depres- lem group had weight-for-age and length-for-age z- sion, birthweight, and breastfeeding status. scores that were below national norms).

156 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Lee and his colleagues (2000) assessed the impact of however, that studies like these can provide definitive WIC participation on the prevalence of failure to answers to questions about WIC’s impact on the growth thrive. Failure to thrive is a general diagnosis that can of infants and children. The researchers involved in have many causes. The sentinal finding, however, is a designing and implementing the field test of USDA’s failure to gain weight and to grow as expected. The planned WIC Child Impact Study concluded that the authors used a longitudinal, linked database that only way WIC’s impacts on child growth and develop- included, for all children born in Illinois between 1990 ment—indeed, WIC’s impact on virtually any outcome and 1996, birth record, WIC, and Medicaid data for beyond dietary intake—can be reliably assessed is 1990 though 1998.71 The analysis was limited to chil- through a longitudinal study that includes serial meas- dren who were born between 1990 and 1994 and were urements, repeated at regular intervals, for both WIC continuously enrolled in Medicaid from birth until participants and nonparticipants (Puma et al., 1991). their fourth birthday. To control for the fact that children diagnosed with failure to thrive may be more Anemia/Iron Status likely to participate in WIC—because it is an accepted A number of different methods were used in the 16 nutritional risk—the authors included in the WIC par- studies that examined the impact of WIC on children’s ticipant group only those children whose WIC enroll- iron status, including typical participant vs. nonpartici- ment predated their diagnosis. Children whose enroll- pant designs that made cross-sectional comparisons of ment in WIC occurred after their diagnosis were con- single point in time measurements, longitudinal or fol- sidered nonparticipants. Results showed that Medicaid lowup studies in which initial measures for the same children who had ever participated in WIC were sig- individuals were compared at a later time point (designs nificantly less likely than those who never participated referred to as “participants, before vs. after” and “par- to be diagnosed with failure to thrive (based on pri- ticipant vs. nonparticipant, before and after” in chapter mary diagnosis in Medicaid claims data). 2), and time series analyses that used aggregate data to compare the prevalence of anemia in a particular pop- In recent years, increasing attention has been paid to ulation over time. Although each of these designs has problems at the opposite end of the growth spectrum— weaknesses, as described in chapter 2, the consistency the problem of overweight among children, including of findings is compelling. The majority of studies that very young children. A 1995 CDC report provides sug- examined the relationship between WIC participation gestive evidence that WIC participation is not associat- and iron status/anemia found that WIC participation ed with the prevalence of overweight among children. was associated with an increase in mean levels of Using NHANES-III data, CDC analysts examined hemoglobin or hematocrit and/or a decrease in the children’s weight-for-height, relative to national stan- prevalence of anemia (table 25). In most cases, these dards, by age and race/ethnicity. They found that dif- differences were statistically significant. ferences between WIC and income-eligible nonparticipant children were neither consistent nor statistically The most convincing evidence comes from analyses significant (CDC, 1995). White and Black WIC chil- done by Yip and his colleagues at the CDC using dren tended to weigh more than low-income nonpartic- PedNSS data (Yip et al., 1987). The CDC researchers ipants of the same race, and Mexican-American WIC looked at the prevalence of anemia in infants and chil- children tended to weigh less than their non-WIC dren ages 6-60 months between 1975 to 1985, a period counterparts, but none of the differences were statisti- of substantial growth in the WIC program. They docu- cally significant. Multivariate analysis (not further mented a steady decline in the prevalence of anemia, described) did not change the pattern or significance of from 7.8 percent in 1975 to 2.9 percent in 1985. Using the findings. detailed data from one State, the authors demonstrated that the socioeconomic status of the population had Data from Black et al. (2004) and Lee et al. (2000) remained stable over this period. The authors also provide suggestive evidence that WIC may have a pos- compared initial and followup measures of hemoglo- itive effect on growth in infants and children, and the bin or hematocrit (taken roughly 6 months apart) for 1995 CDC study suggests that WIC participation is not approximately 73,000 WIC children. The analysis associated with excess weight in children. It is doubtful, revealed decreased levels of anemia at followup.

Another CDC analysis reported on trends between 71The database also included information on FSP and TANF/AFDC participation, which the authors used to examine patterns of program partici- 1980 and 1991 (Yip et al., 1992). During this period, pation during welfare reform in Illinois (1990-98). the prevalence of anemia decreased by more than 5

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 157 Chapter 4: WIC Program percent for most age and race/ethnicity specific sub- General Health Status groups. Other measures of childhood health monitored A total of four studies examined the health status of WIC in PedNSS, including the prevalence of low birth- and non-WIC infants and/or children using caregiver weight, low height-for-age, low weight-for-height, and reports and/or physician reports (table 24). The strongest high weight-for-height (overweight), generally data are provided by recent studies by Black et al. remained stable. Comparable findings have been (2004) and Carlson and Senauer (2003). Both studies reported by Sherry et al. (1997; 2001). suggest a positive WIC impact, as shown in table 25, The CDC analyses suggest that WIC has a direct but the potential for selection bias remains a concern. impact on the prevalence of childhood anemia as well Black and her colleagues (2004) assessed differences a probable indirect effect. WIC requires use of iron- in the perceptions of caregivers about the health status fortified infant formulas and includes iron-fortified of urban low-income infants enrolled in the previously breakfast cereals in its food packages. Because more described C-SNAP study. Caregivers were asked to than half of all formula sold in the United States, as rate their infants’ health status as excellent, very good, well as a large proportion of breakfast cereals, are pur- good, fair, or poor. The question and response options chased with WIC vouchers (Batten et al., 1990), manu- were taken directly from the NHANES-III caregivers facturers have consciously focused on bringing to mar- interview. Results showed that, relative to WIC ket iron-fortified products that are allowable in WIC infants, infants who were not enrolled in WIC because food packages. These foods have assumed a leading of access problems were significantly more likely to position in their respective markets and have therefore be rated as having fair or poor health. There was no been increasingly consumed by both WIC and nonsignificant difference in the caregiver-rated health sta- WIC children. As a result, the WIC program may have tus of WIC infants and infants who were not enrolled contributed to the observed improvement in the preva- in WIC because their caregivers did not perceive a lence of anemia in the general population of low- need for WIC services. income U.S. children. Carlson and Senauer (2003) assessed WIC impacts on The declining prevalence of iron deficiency may have physician-assessed health status using NHANES-III diminished the predictive value of anemia as a screen data. Physicians rated children’s health status (excel- for iron deficiency. When the prevalence of iron defi- lent, very good, good, fair, or poor) after completing a ciency is high, anemia is a good predictor of iron defi- physical examination. Because so few children were ciency. However, when the prevalence is low, the reported to be in poor health, the poor and fair cate- majority of anemia (low hemoglobin or hematocrit gories were combined for the analysis. The authors levels) is due to other causes (U.S. Department of used ordered probit models to estimate the likelihood Health and Human Services (HHS), 2000; Sherry et of a child being in excellent health. Two analysis sam- al., 2001). In young children, a likely cause is infec- ples were used. The “WIC sample” included children tion and inflammation associated with viral illness between the ages of 24 and 60 months who either par- and, to a lesser extent, hereditary anemias (HHS, ticipated in WIC or were income-eligible based on 2000; Bogen, 2002). This may be the reason that Kahn household income or Medicaid participation. WIC par- et al. (2002), in tracking the prevalence of anemia ticipation was defined at the household rather than among infants and children in three Illinois clinics individual level. The “full sample” included all children between 1997 and 1999, found a substantial amount of between the ages of 24 and 60 months who had com- crossover between anemic and nonanemic groups over plete data. Results for both the full sample and the WIC time (Bogen, 2002). Future efforts to examine WIC’s sample indicated that WIC participation had a signifi- impact on the iron status of infants and children should cant, positive effect on the likelihood that a child would assess the prevalence of iron deficiency and/or iron- be in excellent health. The size of the effect (increased deficiency anemia rather than simple anemia. A recent probability of being in excellent health) ranged from descriptive analysis of NHANES-III data found that 4.6 to 11.4 percentage points, depending on the model WIC children were significantly less likely than used. Effects were consistently higher in the “WIC income-eligible nonparticipant children to be iron defi- sample,” indicating that the impact of WIC on health cient (Cole and Fox, 2004). status was most pronounced for poorer children.

158 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

The previously described study by Lee et al. (2000) Shaheen and her associates (2000) studied immunization looked at the prevalence of “nutritional deficiencies” coverage among low-income children in Los Angeles. among Medicaid children in Illinois who did and did They completed a household enumeration of children not participate in WIC. The measure, based on primary ages 24-47 months in 30 randomly selected clusters diagnosis in Medicaid claims data, is not well described, (block groups) in and around downtown. A random but is said to “include conditions such as malnutrition sample of 300 children was selected and their parents and vitamin deficiencies.” The authors found that chil- or guardians were interviewed in person. Information dren who had ever participated in WIC were signifi- on immunization status was abstracted from home cantly less likely than children who had never partici- immunization cards, if available (81 percent of all pated in WIC to be diagnosed with these conditions cases), or obtained from health care providers. WIC (finding is not included in table 25). As previously participation was positively but not significantly asso- mentioned, the analysis explicitly excluded from the ciated with completing immunization series on time. WIC participant group children whose diagnosis pre- Among children who had not received their first round dated their WIC enrollment. of immunizations at the recommended age, the combination of WIC participation and having a home immu- Immunization Status nization card was significantly and positively associat- The literature search identified six studies that examined ed with being fully immunized by 24 months of age. the immunization status of WIC participants relative to Sample sizes for the latter analysis were very small nonparticipants (table 24). Results are summarized in and may have obscured any independent WIC effect. table 25. Although this research suggests that WIC has a positive impact on children’s receipt of complete and Shefer and her colleagues (2001) analyzed data from timely immunizations, the findings are particularly the 1999 National Immunization Survey (NIS). The vulnerable to selection bias. Mothers who are motivat- NIS has been conducted by the CDC since 1994 to ed to enroll their child in WIC may also be motivated estimate vaccination coverage rates for U.S. children to ensure that the child is properly immunized. ages 19-35 months. Children were divided into four groups based on WIC participation status and income: The NWE (Rush et al.,1988c) found that children who currently on WIC, previously on WIC, never on WIC were enrolled in WIC after the first year of life were but income-eligible, and never on WIC and not significantly more likely than children in the control income-eligible. Bivariate comparisons showed that, group to have an immunization card and to have among children with household incomes at or below received a measles vaccination (given after 15 months 100 percent of poverty, children who had ever partici- of age). In addition, children whose mothers participat- pated in WIC were significantly more likely to have ed in WIC prenatally and were enrolled in WIC during up-to-date immunizations at 24 months than children early infancy were more likely to have received diph- who had never participated in the program (71 percent theria-pertussis-tetanus (DPT) immunization and, to a vs. 56 percent). Moreover, across all income groups, lesser extent, polio vaccines. children who currently participated in WIC were more likely than previous WIC participants to have up-to- James (1998) studied the immunization status of 150 date immunizations at 24 months (75 percent vs. 69 children receiving care at one health care center in New percent). However, current WIC participants were less York State. She randomly selected a sample of WIC well immunized than higher income children who had children who were up to date on immunizations at 12 never been on WIC (75 percent vs. 83 percent). months of age and matched them on age and gender (only) with subjects from a randomly selected group of Luman and her colleagues (2003) expanded on Shefer’s non-WIC children who were receiving care at the same work, analyzing data from the 2000 NIS to identify health care center. She then documented immunization maternal characteristics associated with children’s 72 status at 24 months of age and used chi-square tests to immunization. Children not currently participating in assess differences between the two groups. No signifi- WIC were divided into three nonparticipant categories cant differences were found. This result is not surpris- comparable to those defined by Shefer et al. (2001). A ing in view of the fact that all of the study children were clearly in “immunization aware” households (all 72Luman et al. (2003) characterized WIC participation as a “maternal children had up-to-date immunizations at 1 year of characteristic,” but the 2000 NIS instrument actually collected information age) and all had a primary source of health care. on the child’s participation (Abt Associates Inc., 2002).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 159 Chapter 4: WIC Program multivariate regression that controlled for child’s age eight factors that had a positive, significant effect on showed that children who were currently participating immunization rates in Georgia’s public health clinics. in WIC were significantly more likely than any of the nonparticipant groups to be fully immunized. Children Use and Costs of Health Care Services who had never participated in WIC but were income- The NWE examined use of preventive health care eligible were least likely to be fully immunized. services among WIC and non-WIC infants and children and found no significant difference between the The positive WIC impact suggested by this research, if two groups (Rush et al., 1988c). More recently, three real, may be influenced by an ongoing collaboration studies examined the relationship between children’s between USDA and the CDC to use the WIC program WIC participation and the use of health care services as a means to improve immunization rates among the (Lee et al, 2000; Buescher et al., 2003) and dental care Nation’s low-income children. Since the early 1990s, a services (Lee et al., 2004a) using datasets similar to variety of strategies has been used to promote timely the WIC-Medicaid databases used to assess WIC’s and complete immunizations among WIC participants, impact on birthweight (table 24). All three studies including the following (Shefer et al., 2001): reported that WIC participation had a significant, positive effect on the use of health care/dental care servic- • Assessment and referral—May occur at each WIC es (table 25). These results suggest a positive WIC visit or at each 6-month recertification visit and may impact; however, only the study that looked at dental include a computerized tracking system. care services controlled for selection bias. Thus, the • Escort programs—Children who need immuniza- two studies that assessed use of health care services tions are escorted to locations where immunizations are vulnerable to potential selection bias—it is possi- are given. ble that children who have health problems or who use more health care services may be more likely to be • Voucher incentive programs—Restrict access to referred to WIC. In addition, results of all three studies WIC vouchers until a child is fully immunized (for have limited generalizability because they used example, require that caregivers come into the WIC datasets for a single State (one study used data from clinic every month rather than every 2 months to Illinois and two used data from North Carolina) and pick up WIC vouchers). were limited to WIC participants enrolled in Medicaid.

• Outreach and tracking programs—May involve mail, As described previously, Lee et al. (2000) used a lon- telephone, or home visit reminders for underimmu- gitudinal database that included birth record, WIC, and nized children. Medicaid data for all children born in Illinois between 1990 and 1996. They used a proportional hazards Randomized trials have demonstrated that some of these model to estimate the effect of WIC participation on strategies can dramatically increase immunization cov- the timing of children’s first screening in the Early erage among WIC participants (Birkhead et al., 1995; Periodic Screening, Diagnosis, and Treatment Hutchins et al., 1999). In addition, Shefer et al. (2001) (EPSDT) program.73 Their analysis included all chil- modeled the relationship between WIC immunization dren new to the Medicaid program between 1991 and activities and immunization rates among WIC children. 1997. The dependent variable was the time between Using data from the 1999 NIS and data from an annual entry into the Medicaid program and receipt of the survey of WIC directors and State immunization pro- first EPSDT screening. WIC participation was coded gram directors, Shefer and her colleagues found that as a time-varying covariate. Results showed that WIC WIC participants in States with high-intensity immu- participation had a significant, positive effect on the nization activities (50 percent or more of WIC children likelihood of receiving EPSDT screening. enrolled in sites that implemented an immunization intervention at every visit) had significantly higher Buescher et al. (2003) and Lee et al. (2004a) used a rates of up-to-date immunization at 24 months than database that included linked birth certificate, Medicaid, WIC participants in States with low-intensity immu- and WIC records for all children born in North Carolina nization activities (less than 50 percent of WIC chil- in 1992. Buescher and his colleagues assessed the dren enrolled in sites that used an immunization intervention and the intervention was implemented only at 73EPSDT is a comprehensive, prevention-oriented child health program recertification visits). Finally, Dietz et al. (2000) found that State Medicaid agencies are required to provide for all Medicaid recip- that a WIC voucher incentive program was one of ients under the age of 21. See www.cms.hhs.gov/medicaid/epsdt.

160 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program impact of WIC participation on the use and cost of other common childhood illnesses (asthma, gastroenteri- health care services. Lee and her associates focused on tis, and allergy) varied by age cohort and/or level of WIC impacts on the use of dental care services. They WIC participation. In most cases, however, high WIC also completed a separate analysis that assessed participation was associated with a significantly greater impacts on dental care costs (Lee et al., 2004b). likelihood of diagnosis for these conditions. A high level of WIC participation was also associated with a signifi- Buescher et al. (2003) studied children ages 1-4 who cantly greater likelihood of being hospitalized. had been continuously enrolled in Medicaid for at least 1 year of life. Separate analyses were completed for four Given these patterns, it is not surprising that WIC par- different age cohorts based on completed year of age. ticipation was associated with increased Medicaid A cumulative measure of WIC participation was defined expenditures. WIC participation was consistently asso- based on the percentage of months from age 1 through ciated with greater expenditures for outpatient services the current age in which WIC food vouchers had been (all four age cohorts and all three levels of WIC partic- redeemed. Three levels of WIC participation were identi- ipation). In addition, medium and high levels of WIC fied: high = more than 66 percent; medium = 34-66 participation were generally associated with signifi- percent; low = 33 percent or less. Because the analysis cantly greater Medicaid costs overall and for individual looked at cumulative use of health care services and types of medical care (EPSDT, well-child care, physi- cumulative Medicaid costs, this definition of WIC par- cian services, drugs, and dental care). The authors con- ticipation was deemed more appropriate than a defini- cluded that “the bottom line is that children enrolled in tion based on current participation. Medicaid who participate in WIC are linked to the health care system and are much more likely to receive Tobit regression was used to analyze Medicaid cost data both preventive and curative care, whereas Medicaid- because of the large number of zero values. Logistic enrolled children who do not participate in WIC sim- regression was used to estimate the odds of having a ply are not as connected to the health care system.” well-child visit, being hospitalized, having an emergency room visit, and being diagnosed/treated for a common Lee et al. (2004a; 2004b) studied the impact of chil- childhood illness. An array of variables was used to dren’s WIC participation on the use and cost of dental control for sociodemographic characteristics of the child care services. Their analysis of dental care use con- and his/her mother, including prenatal WIC participation trolled for selection bias by using a two-stage model for the mother and participation in WIC as an infant for that incorporated State-level WIC data to predict WIC the child. In addition, the authors included a variable to participation (Lee et al., 2004a). The model included designate whether EPSDT services were received in a three variables that the authors found to be correlated public health department. This variable helped control with WIC participation but not with the use of dental for differences in costs associated with public vs. private care services, including the number of WIC clinics per health care providers, as well as for co-location of WIC county, the number of full-time WIC workers per clinics with child health care services. The authors county, and WIC hours of operation. attempted to control for selection bias but, after examining several multi-stage models, were unable to iden- The analysis looked at the annual use of dental care tify good predictors of WIC participation that were not services as well as the types of services used (diagnostic/ also associated with the outcome measures. preventive, restorative, emergency) by children ages 1-4. A categorical rather than continuous outcome measure Results showed that WIC participation had a significant, was used to represent annual use of dental care servic- positive effect on the likelihood that children would es (no visits, one visit, two or more visits) because the receive any well-child care and on the likelihood that recommended number of visits is two per year and children would receive at least one EPSDT visit per year, because visits in excess of two per year may be more as recommended. This effect was noted for all four age indicative of severe dental disease than of access to cohorts and, with one exception, for all three levels of dental care services. WIC participation was measured WIC participation. Comparable results were noted for based on the number of months when any WIC food the likelihood of visiting an emergency room or of being vouchers were redeemed during each year of life. In diagnosed with otitis media or upper respiratory infec- addition to controls for sociodemographic characteris- tion. Medium to high levels of WIC participation were tics, the model controlled for length of Medicaid positively associated with the likelihood of being diag- enrollment as well as the relative availability of dental nosed with a lower respiratory infection. Results for practitioners (ratio of dentists per population).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 161 Chapter 4: WIC Program

WIC participation was associated with a significant Cognitive Development and Behavior increase in the use of dental care services. Children Five of the identified studies assessed WIC impacts on who participated in WIC for either 6 months or a full measures of cognitive development or behavior (table year were significantly more likely than children who 24). The NWE assessed children’s attention and cogni- did not participate in WIC to have both one dental tion using three standardized tests known to be respon- visit per year and two or more dental visits per year. sive to changes in early life: the Infant Behavior Children who participated in WIC were also signifi- Inventory (IBI), the Peabody Picture Vocabulary Test cantly more likely to have used preventive and restora- (PPVT), and the McCarthy Scales of Infant Develop- tive dental health services and significantly less likely ment (Rush et al., 1988c). The results indicated that to have had an emergency dental visit. The latter find- infants and children whose mothers had participated in ing was significant at the p <0.10 level. The authors WIC prenatally had significantly higher receptive suggest that WIC participation provides children with vocabulary scores than infants and children whose moth- a “better connection to the health care system that can ers had not participated in WIC (table 25). In addition, lead to care that is more planned and less urgent.” children who enrolled in WIC after the first year of Lee and her colleagues (2004b) also assessed the impact life had significantly better digit memory (counting of WIC participation on Medicaid costs for dental care, backward and forward) than children in the control but used a slightly different approach. They did not group. Rush and his colleagues appropriately cau- control for selection bias because they had to use a two- tioned that small sample sizes and substantial differ- stage model to first estimate the probability of a child ences in the socioeconomic status of WIC and control having any dental care expenditures (many children had group members limit the strength of these associations. no dental care expenditures). In addition, they estimat- Hicks, Langham, and Takenaka (1982) studied 21 sibling ed separate models for infants and for cohorts of chil- pairs in Louisiana. One of the siblings (the younger) had dren ages 1, 2, and 3. They excluded 4-year-olds from received WIC benefits both “prenatally” (during the the analysis to minimize the potential for simultaneous third trimester), through the first year of life, and on determination. Finally, they defined WIC participation into childhood, while the other (older) sibling received as a dichotomous variable, reflecting whether any of WIC benefits only in childhood—that is, after 1 year the child’s WIC food vouchers had been redeemed in of age. The children were studied when the younger each particular year of life. Age-specific models con- children were about 6 years old and the older children trolled for WIC participation at other ages. were about 8. The “early supplementers” had partici- Results showed that, among infants and children ages pated in WIC an average of 56.1 months (counting the 1 and 2, WIC participation was associated with a sig- prenatal period), while WIC participation for the “late nificantly greater likelihood that a child would have supplementers” averaged 30.8 months. some Medicaid charges for dental care services. This Investigators used a battery of measures to assess behav- relationship was not observed among 3-year-olds. In ior and cognitive performance and also obtained school addition, among infants and 1-year-olds, WIC partici- grades for reading, writing, and arithmetic. Strong, pation was associated with significantly lower dental positive, and statistically significant differences, favoring care costs (19-20 percent less). No significant differ- the child with greater WIC exposure, were found for ences were detected in total dental care costs for children most of the cognitive and behavioral measures assessed, ages 2 and 3. Investigation of the sources of dental including IQ, attention span, visual-motor synthesis, care costs showed that WIC participants were signifi- and school grade point average. The authors attributed cantly less likely than nonparticipants to have received these differences to a superior nutrition environment dental care services in a hospital setting (as opposed to for the “early supplementers” during critical periods of a primary care setting). This difference in dental care brain development in the last trimester of pregnancy source accounts for at least some of the lower cost and the first 6 months of life. A followup study com- observed for WIC participants despite a higher preva- pleted 32 months later reported that the findings for IQ lence of dental care charges overall. The authors con- scores and school grades, the only two measures repli- cluded that observed difference between WIC partici- cated, still held (Hicks and Langham, 1985). pants and nonparticipants may be the result of nonparticipants having more dental problems and/or a greater The Hicks studies have been heavily criticized by others tendency to use emergency room services rather than who have studied the effects of nutrition supplementa- standard outpatient/primary care services. tion on cognitive development (Pollitt and Lorimor,

162 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

1983). One of the greatest concerns is the magnitude of positive WIC effect based on results from the fixed- the differences reported—which exceed many obtained effects model that was significant at the p <0.10 level. from controlled trials of nutrition supplementation in pop- The direction of WIC coefficients in the models that ulations with high prevalence of malnutrition and infec- estimated impacts on motor and social skills was also tious diseases. Other problems include the small sam- sensitive to specification, and none of the models ple size and the potential that the “late supplementers” found a significant WIC effect, even at p <0.10. differ in important ways from the “early supplementers.” Food Security A stronger and more recent study by Kowaleski-Jones Only one of the identified studies used the 18-item and Duncan (2000) examined the impact of prenatal USDA food security module to assess the impact of WIC participation on temperament and the develop- WIC participation on household food security. In the ment of motor and social skills. The authors used data previously described C-SNAP study, Black et al. (2004) from the NLSY, focusing on children born to NLSY assessed household food security among low-income participants between 1990 and 1996 (collection of data infants who did and did not participate in WIC. The on WIC participation began in 1990). authors found, relative to WIC participants, a significantly lower rate of household food insecurity (or a Temperament was measured with a composite “difficult higher level of food security) among infants who were temperament” index, which included subscales for pre- not participating in WIC because their caregivers did dictability, fearfulness, positive affect, and friendliness— not perceive a need for WIC services. There was no factors thought to be precursors of personality develop- significant difference in household food insecurity ment and social adjustment. Motor and social skills between WIC participants and nonparticipants who did were assessed based on an established mother-reported not participate because of access problems. scale that measures motor, social, and cognitive development of young children from birth though age 3. Impacts of WIC Participation on The analysis used the earliest available measures for Nonbreastfeeding Postpartum each child. Measurements for about half of the chil- Woman and Breastfeeding Women dren were collected during their first year of life. Measures for the other half were collected between their The literature search identified two studies that assessed first and second birthdays. Analytic models controlled WIC impacts on nonbreastfeeding postpartum WIC for the child’s age at the time of measurement. participants and only one study that looked at the impact of WIC participation on breastfeeding participants. In The authors used both standard regression models and addition, the previously described study by Kramer- fixed-effects models, based on sibling pairs, to estimate LeBlanc et al. (1999) compared dietary intakes of WIC impacts. Three different specifications were used breastfeeding and nonbreastfeeding postpartum WIC for both models: one that controlled for the child’s participants to income-eligible nonparticipants. The sociodemographic characteristics, one that added an analysis of breastfeeding women was hampered by array of prenatal and maternal characteristics (many of small sample sizes. The analysis of nonbreastfeeding which dropped out in the fixed-effects model), and one postpartum women showed that WIC participants con- that also controlled for prenatal participation in the FSP. sumed significantly more calcium, riboflavin, and retinol than nonparticipants. For the difficult temperament index, the direction of the coefficient for prenatal WIC participation varied between Nonbreastfeeding Postpartum Women the standard regression models (positive coefficient, Pehrsson et al. (2001) assessed iron status among non- indicating that WIC participation was associated with breastfeeding postpartum women. The study took an increased likelihood of having difficult tempera- place during a time when some counties in Maryland ment) and the fixed-effects models (negative coeffi- were certifying only high-risk nonbreastfeeding post- cient, indicating that WIC participation was associated partum women because of funding shortages. The par- with a decreased likelihood of having a difficult tem- ticipant group included 57 low-risk WIC participants perament). The only model in which WIC participation who were recruited from WIC sites in Baltimore City, was significant at the p <0.05 level or better was the which was enrolling all eligible postpartum applicants. simplest regression model, which did not control for The nonparticipant group included 53 WIC-eligible prenatal and maternal characteristics or for FSP partic- women who were recruited from WIC sites in counties ipation. However, the authors reported a significant, that were enrolling only high-risk women. Subjects

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 163 Chapter 4: WIC Program were matched on race and age. All were older than 19, factors that can influence iron status and recommend had delivered a full-term infant the preceding month, that future studies collect an expansive set of variables were free of major health problems, and qualified as that will allow for a more sophisticated analysis. These low-risk nonbreastfeeding postpartum participants. include biochemical indicators of infection and inflammation, menstrual history, use of oral contraceptives Women were recruited into the study within 30 days of and reestablishment of menses, and medical documen- delivery and were followed at 2, 4, and 6 months post- tation on the presence of infection, inflammation, and partum. Major outcome variables included four measures conditions that can affect blood-related measures (for of iron status (hemoglobin, transferrin receptor, ferritin, example, sickle-cell anemia and thalassemia). and ratio of transferrin receptor to ferritin). Dietary iron intake was estimated using a self-administered In the other identified study of nonbreastfeeding post- food frequency questionnaire designed for respondents partum women, Caan et al. (1987) studied the benefits with low levels of literacy. of WIC participation during the interpregnancy interval, looking at women’s nutritional status at the start Although mean hemoglobin concentrations were com- of the second pregnancy and birth outcomes of that parable at baseline, the mean for WIC participants pregnancy. The study involved 47 local WIC agencies increased over time. At 6 months postpartum, the dif- in California that had different policies for enrolling ference was statistically significant, after controlling nonlactating postpartum women between 1981 and for FSP participation, smoking status, use of iron sup- 1983. Because of funding shortages during that period, plements, and interpregnancy interval (months elapsed some local agencies were not able to serve these lower between immediately preceding pregnancy and prior priority participants. pregnancy). Moreover, at 6 months postpartum, significantly more nonparticipants than participants were In 1983, researchers recruited newly enrolling pregnant anemic based on the CDC-recommended cutoff for women in both types of local agencies (agencies that had hemoglobin. No significant differences between WIC and had not served nonbreastfeeding postpartum women participants and nonparticipants were detected for any between 1981 and 1983). To be eligible to participate of the other measures of iron status or for dietary in the study, women had to meet the following criteria: intake of iron. (1) given birth to another infant between 1981 and 1983; (2) participated in WIC during the previous pregnancy; The authors concluded that nonbreastfeeding postpar- and (3) did not breastfeed the first infant. Women were tum WIC participants who experienced 6 uninterrupted divided into two groups based on their postpartum par- months of participation were significantly less likely to ticipation after the previous pregnancy: an extended become anemic than comparable women who did not feeding group, which included women who had received participate in WIC during the postpartum period. postpartum WIC benefits for 5-7 months, and a limited Because they observed no difference between groups feeding group, which included women who received in means for the other measures of iron status, in the postpartum benefits for 0-2 months. Both groups had a percentage of women classified as iron deficient based comparable interpregnancy period (the time elapsed on these measures, or in dietary iron intake, the between first and second pregnancies) of 3 years or less. authors concluded that observed differences in hemoglobin concentrations and in the prevalence of anemia The final analysis sample included 642 women (307 may not be associated with improvements in iron sta- limited feeding and 335 extended feeding). The study tus among participants. Rather, these differences may found a positive, significant impact of extended WIC be attributable to increased rates of other nutritional participation after the first pregnancy on both birth- deficiencies, compromised health care, and infection weight and birth length of the second infant. The odds or inflammation among nonparticipants. ratio of having a low-birthweight infant approached significance, but, because low birthweight is a rare This study may have underestimated WIC’s effect on event, small sample sizes hampered the analysis (24 the iron status of postpartum women because most of infants, 5.1 percent of all infants in the limited feeding the women, including all but two of the nonpartici- group and 3.2 percent of all infants in the extended pants, participated in WIC during pregnancy and feeding group). because all subjects were relatively iron replete when they entered the postpartum period. In addition, these Positive effects from WIC participation were also authors, like Rush et al. (1988d), discussed the myriad reported for maternal outcomes. Women in the extended

164 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program feeding group who had been obese at the start of the middle class non-WIC women had higher mean first pregnancy were 50 percent less likely than compa- intakes of energy and nutrients at the time of the first rable women in the limited feeding group to be obese at interview. WIC participants significantly increased the start of the second pregnancy. Although not a statisti- their intakes over the 2-month study period, while cally significant trend, women in the extended feeding nonparticipants’ intakes decreased, resulting in compa- group who had been underweight at the onset of the rable intakes for the two groups. While the results are first pregnancy tended to weigh more at the onset of intriguing, little can be concluded from this dated, the second pregnancy. Finally, mean hemoglobin lev- poorly designed study. els (measured at the time of enrollment for the second pregnancy) were significantly higher in the extended feeding group; however, women were no more likely Impacts of WIC Participation on WIC than those in the limited feeding group to be anemic. Households or Undifferentiated WIC Participants Caan and her colleagues point out that evidence from other studies suggests that physiologic and metabolic Five of the identified studies looked at the impact of adjustments associated with pregnancy proceed more WIC participation on all types of WIC participants normally in women whose nutritional status is good at (without differentiating women, infants, and/or chil- the beginning and very early stages of pregnancy. dren) or on household-level outcomes. Two studies Thus, the authors assert, among women with short looked at dietary quality, one study examined house- interpregnancy intervals, WIC participation during the hold food use, and two studies estimated impacts on postpartum period may have an even stronger positive household food expenditures. Both of the latter studies impact on birth outcomes than prenatal participation used data from the NWE. alone, especially for women who might not enroll in Dietary Quality WIC until the second trimester of pregnancy. The authors appropriately acknowledge that their study has Basiotis, LeBlanc, and Kennedy (1998) used data from several limitations, including exclusion of a subgroup the 1989-91 CSFII to look at dietary quality among of recruited women because of problems with missing low-income households eligible to participate in the FSP data and the potential bias associated with the fact that (income less than 130 percent of poverty) and assessed women in each study group were drawn from mutually the impact of participation in the FSP (entire house- exclusive sets of local agencies. hold) and WIC (any household member). Dependent measures used in the study were the Healthy Eating While further research is needed to replicate or expand Index (HEI) and its component scores. the work done by Pehrsson et al. and Caan et al., these studies are important in that they provide the only The authors found that having one or more household source of information on potential WIC impacts members participate in WIC was associated with a very among nonbreastfeeding postpartum women, the pro- strong positive impact on dietary quality. Specifically, gram’s lowest priority group. If the hypotheses out- overall HEI scores for households with one or more lined by these researchers, particularly those of Caan WIC participants were 23 points higher than scores for et al., are valid, there may be reason to rethink the low other households, a substantial effect given that the priority assigned to this participant group. mean score for all of the low-income households included in the sample was 62 points. WIC participa- Breastfeeding Women tion was also associated with significant increases in In addition to the descriptive study by Kramer- all component scores except those associated with veg- LeBlanc et al. (1999), the only study that focused etable consumption and intake of saturated fat. specifically on breastfeeding women examined nutri- Wilde et al. (2000) used more recent CSFII data ent intakes in a very small convenience sample of (1994-96) to estimate the impact of WIC participation breastfeeding WIC participants (n=11) and an even on dietary quality. This study used data on Food Group smaller sample of middle-class breastfeeders not par- Pyramid servings to examine impacts on intake of ticipating in WIC (n=5) (Argeanas and Harrill, 1979). meats, fruits, vegetables, grains, dairy, added sugars, Researchers collected 24-hour dietary recalls from and added fats. The only significant effect noted for each subject at approximately 6 weeks postpartum and WIC was a positive one for intake of added sugars again 2 months later. The authors reported that the (fewer teaspoons). Results did not differentiate

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 165 Chapter 4: WIC Program between women participants and child participants. in family income and food expenditures between WIC The authors suggest that this effect may result from and non-WIC households that actually became worse participants substituting WIC-supplied fruit juices and after statistical controls for several sociodemographic cereals for higher-sugar soft drinks and cereals. characteristics; disparate results for data collected using different methods (diary vs. recall); and a devalu- While these studies are important, in that they are the ing of the WIC food package by WIC participants.75 only ones to examine WIC impacts on dietary quality, expanding the focus to impacts on particular types of Arcia, Crouch, and Kulka (1990) reanalyzed the NWE WIC participants would be useful. Although the num- data and estimated empirical models of WIC impacts ber of pregnant and postpartum women is likely to be on monthly food expenditures; the degree to which small in the CSFII database, the combined CSFII WIC benefits substituted for foods that would have 1994-96, 1998 CSFII database includes a substantial been purchased anyway; and the degree to which WIC number of age-eligible children and a reasonable num- benefits were shared with unintended family members. ber of infants. These researchers concluded that participation in the WIC program by a pregnant woman has a more signif- Household Food Use icant impact on the type of foods households purchase Taren et al. (1990) studied food use among low- than on how much is spent. Participation by a child, on income households. The sample included families par- the other hand, was reported to have a positive effect ticipating in food cooperatives for low-income families on both food expenditures and food purchases. in Hillsborough County, FL, as well as families partic- It is difficult to draw conclusions about WIC impacts ipating in the local Expanded Food and Nutrition on household food expenditures given the two diver- Education Program (EFNEP). Data were collected on gent conclusions drawn from the same data set. the number of servings of 27 different foods used in However, Rush and his colleagues make a compelling the household the previous week. A serving was argument that limitations of both the sample drawn for defined as the preparation and offering of a particular this substudy and the data collected make it difficult to food, without attention to portion size or multiple have confidence in the findings from any analysis helpings during the same meal. (Rush et al., 1988b). Multiple regression analysis was used to identify factors associated with the number of family food servings used Summary per week. Participation in WIC had a positive, significant impact on the number of servings of food used The preceding discussion clearly illustrates that an per week. The report does not provide details about the extensive amount of research has been conducted on food list used in the study or about relative impacts for the WIC program. At the same time, it demonstrates that different categories of food, making it impossible to coverage of the five different participant groups is very determine whether the list gave substantially more uneven in the existing research and that important gaps weight to WIC foods than to other foods or the relative remain in information about potential program impacts. contribution of WIC foods to the overall total. Research is most extensive in the area of birth out- Household Food Expenditures comes. Although concerns about self selection persist, the sheer volume of studies that have reported signifi- The NWE included a substudy to examine the impact cant impacts—in different subgroups of WIC partici- of WIC participation on household food expenditures 74 pants, at different points in time, and using different (Rush et al., 1988b). The researchers concluded, research designs and analysis methods—suggests that however, that the study could provide little useful WIC does have a positive impact on birthweight, as information because of three key problems: disparities well as a number of other birth-related outcomes, and

74Fraker, Long, and Post (1990) attempted to study the impact of WIC and FSP participation on household food expenditures using data from the 75WIC vouchers and checks specify particular types and amounts of 1985 CSFII. The authors concluded that they could not reliably estimate the food that can be purchased but do not include information on currency impact of WIC, however, because of small sample sizes, the apparent com- value (like food stamps do). Therefore, food expenditures estimated by plexity of the relationship between WIC participation and food expenditures, simple recall are more prone to underestimation (devaluing) than those and a lack of relevant variables for use in modeling. There was evidence estimated by the diary method where every item is reported and valued that the effect of the WIC program on household expenditures may vary, individually. Moreover, WIC participants may be less aware of the actual depending on the number and type of WIC participants in the household. cost of WIC food items than their nonparticipant counterparts.

166 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program significantly lowers birth-related Medicaid costs. Research on WIC impacts on pregnant women (other Because of the design characteristics that contribute to than impacts on birth outcomes) is scarce and relative- inherent underestimation or overestimation of WIC ly dated. Even less is known about impacts on breast- impacts and the wide range of reported estimates, char- feeding and nonbreastfeeding postpartum women. acterizing the relative size of WIC’s impact with any Exploration of impacts on postpartum participants confidence is difficult (for example, the estimated reduc- seems especially important to pursue. The limited tion in the prevalence of low-birthweight infants). available research suggests that postpartum WIC partici- Moreover, subgroup analyses completed by some pation may be associated with improved birth outcomes researchers suggest that WIC impacts are likely to be in the subsequent pregnancy and with improved nutri- greatest among Blacks and the lowest income women, tion, health, and/or weight status for the women. If groups with the highest incidence of low birthweight. these relationships are confirmed through more definitive research, there may be reason to rethink the lower In addition, many important changes have taken place priority assigned to nonbreastfeeding postpartum since most of the available research was conducted. women. In view of the ongoing obesity epidemic, the These changes may influence the extent to which find- potential for WIC to play a role in addressing pregnan- ings from previous research apply to the WIC program cy-related weight retention, which seems to be especially as it operates today. Some of the most noteworthy prevalent among minority women (Gore et al., 2003; changes include: a substantially higher level of program Abrams et al., 2000), seems particularly important. penetration in most areas of the United States than was present in the mid- to late 1980s when most of the There is no solid evidence about the impact of WIC on research was completed (that is, most eligible prenatal the initiation and duration of breastfeeding. Moreover, applicants are able to enroll in the program); more most of the studies that are available were completed generous Medicaid income-eligibility criteria for preg- prior to a considerable expansion of breastfeeding pro- nant women (including some that exceed the WIC cut- motion efforts in the WIC program. Some early studies off of 185 percent of poverty), which infers automatic suggested that WIC participants may follow recom- income-eligibility for WIC; and the use of standard- mended infant feeding guidelines more closely than ized nutritional risk criteria. Furthermore, welfare reform nonparticipants, delaying introduction of cow’s milk legislation, which did not affect WIC directly, may until infants are 1 year and delaying the introduction have affected the circumstances of both WIC partici- of solid foods until 4-6 months. However, recent pants and nonparticipants. Any of these changes may descriptive studies raise doubts about whether these influence both the presence and size of WIC impacts, differences persist today. An updated study of WIC’s as well as variations in impacts across subgroups. impacts on infant feeding practices would fill these important information gaps. Attention to selection-bias In an ideal world, USDA would be able to periodically issues will be especially critical for such a study. complete studies like the WIC/Medicaid study at the national level. The seeds for such an undertaking have Finally, although recent research has begun to fill an been planted. The 2003 revisions to the U.S. Standard information gap that existed for many years, the basis Certificate of Live Birth include the addition of an is small for drawing definitive conclusions about item to collect information about WIC participation WIC’s short- and long-term impacts on infants and during pregnancy.76 Some States already collect this children, the majority participant groups. The evidence information and the hope is that all States will do so is fairly strong that WIC improves children’s iron sta- by 2009 (Sondik, 2003). Until data are available tus. Recent studies suggest that WIC participation pos- nationwide, an updated WIC/Medicaid study that uses itively affects children’s use of health care services, data from States that do include WIC information on immunization status, and overall health, however, birth certificates is an attractive option. potential selection bias remains a concern. Little is known about the impact of WIC on children’s long- term health and development. Moreover, while evidence is convincing that WIC participation increases children’s intakes of selected nutrients, the influence of these increases on the extent to which WIC children consume adequate diets is not clear. An updated ver- 76The item reads, “Did mother get WIC food for herself during this pregnancy?” Revised birth certificate available at www.cdc.gov/nchs/data/ sion of the WIC Child Impact Study planned some dvs/birth11-03final-ACC.pdf. Accessed June 2004. years ago seems overdue.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 167 Chapter 4: WIC Program

References Besharov, D., and P. Germanis. 2001. Rethinking WIC. Washington, DC: The AEI Press. Abrams, B., S. Altman, and K. Pickett. 2000. “Pregnancy Weight Gain: Still Controversial,” American Birkhead G.S., C.W. LeBaron, P. Parsons, et al. 1995. Journal of Clinical Nutrition 71(5):1233s-41s. “The Immunization of Children Enrolled in the Special Supplemental Food Program for Women, Infants, and Abt Associates Inc. 2002. National Immunization Children (WIC): The Impact of Different Strategies,” Survey. 2000 Public Use Data File: Documentation, Journal of the American Medical Association Code Book and Frequencies. Report prepared for the 274(4):312-16. Centers for Disease Control and Prevention. Available at: http;//www.cdc.gov/nis/pdfs/niscbk00.pdf. Black, M., D. Cutts, D. Frank, et al. 2004. “Special Accessed June 2003. Supplemental Nutrition Program for Women, Infants, and Children Participation and Infants’ Growth and Ahluwalia, I.B., V.K. Hogan, L. Grummer-Strawn, et Health: A Multisite Surveillance Study,” Pediatrics al. 1998. “The Effect of WIC Participation on Small- 114:169-76. for-gestational-age births: Michigan, 1992,” American Journal of Public Health 88(9):1374-77. Bogen, D.L. 2002. “Anemia Screening in the Special Supplemental Nutrition Program for Women, Infants, Arcia, G.J., L.A. Crouch, and R.A. Kulka. 1990. “Impact and Children: Time for Change?” Pediatrics 156:969- of the WIC Program on Food Expenditures,” American 70. Journal of Agricultural Economics 72:218-26. Brien, M.J., and C.A. Swann. 1999. Prenatal WIC Argeanas, S., and I. Harrill. 1979. “Nutrient Intake of Participation and Infant Health: Selection and Lactating Women Participating in the Colorado WIC Maternal Fixed Effects. Unpublished manuscript. Program,” Nutrition Reports International 20(6):805-10. Bronner, Y.L., D.M. Paige, S.M. Gross, et al. 1994. Bailey, L.B., M.S. O’Farrell-Ray, C.S. Mahan, et al. Nutrition Education for Pregnant Women and 1983. “Vitamin B , Iron, and Folacin Status of 6 Caretakers of Infants: A Review of Research. USDA, Pregnant Women,” Nutrition Research 3:783-93. Food and Nutrition Service.

Balcazar, H., C.M. Trier, and J.A. Cobas. 1995. “What Brown, H.L., K. Watkins, and H. Hiett. 1996. “The Predicts Breastfeeding Intention in Mexican-American Impact of the Women, Infants, and Children Food and Nonhispanic White Women? Evidence from a Supplement Program on Birth Outcomes,” American National Survey,” Birth 22:74-80. Journal of Obstetrics and Gynecology 174:1279-83.

Bartlett, S., R. Olvera, E. Bobronikov, et al. Brown, J.E., and P. Tieman. 1986. “Effect of Income 2003. Study of WIC Participant and Program and WIC on the Dietary Intake of Preschoolers: Characteristics: 2002. USDA, Food and Results of a Preliminary Study,” Journal of the Nutrition Service. American Dietetic Association 86:1189-91.

Bartlett, S., R. Olvera, N. Gill, et al. 2002. WIC Buescher, P.A., and S. Horton. 2000. Prenatal WIC Participant and Program Characteristics: 2000. Participation in Relation to Low Birth Weight and USDA, Food and Nutrition Service. Medicaid Infant Costs in North Carolina - a 1997 Update. Study No. 122. Raleigh, NC: North Carolina Basiotis, P., C. Kramer-LeBlanc, and E. Kennedy. 1998. Department of Health and Human Services, Center for “Maintaining Nutrition Security and Diet Quality: the Health Information and Statistics. Role of the Food Stamp Program and WIC,” Family Economics and Nutrition Review 11(1-2):4-16. Buescher, P.A., S. Horton, B. Devaney, et al. 2003. “Child Participation in WIC: Medicaid Costs and Use Batten, S., J. Hirschman, and D. Thomas. 1990. of Healthcare Services,” American Journal of Public “Impact of the Special Supplemental Food Program Health 93(1):145-50. on Infants,” Pediatrics 117:S101-09.

168 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Buescher, P.A., L.C. Larson, M.D. Nelson, et al. 1993. Covington, M.T. 1995. Protective Factors in the “Prenatal WIC Participation Can Reduce Low Birth Content of Prenatal Care Services that Promote Weight and Newborn Medical Costs: a Cost-benefit Normal Birth Weight Deliveries Among African- Analysis of WIC Participation in North Carolina,” American Women. Unpublished doctoral dissertation, Journal of the American Dietetic Association University of North Carolina at Chapel Hill. 93(2):163-66. Currie, J. 1995. “WIC and School Nutrition Burstein, N., M.K. Fox, J.B. Hiller, et al. 2000. WIC Programs,” in J. Currie (ed.), Welfare and the Well- General Analysis Project: Profile of WIC Children. Being of Children. Chur, Switzerland: Harwood USDA, Food and Nutrition Service. Academic Publishers, pp.92-103.

Burstein, N., M.K. Fox, and M.J. Puma. 1991. Study of Devaney, B. 1992. Very Low Birthweight Among the Impact of WIC on the Growth and Development of Medicaid Newborns in Five States: The Effects of Children: Field Test. Final Report, Volume II: Prenatal WIC Participation. USDA, Food and Preliminary Impact Estimates. Cambridge, MA: Abt Nutrition Service. Associates Inc. Devaney, B., L. Bilheimer, and J. Schore. 1990. The Caan, B., D.M. Horgen, S. Margen, et al. 1987. Savings in Medicaid Costs for Newborns and Their “Benefits Associated with WIC Supplemental Feeding Mothers from Prenatal WIC Participation in the WIC During the Interpregnancy Interval,” American Program: Volume 1. USDA, Food and Nutrition Service. Journal of Clinical Nutrition 45:29-41. Devaney, B., L. Bilheimer, and J. Schore. 1991. Carlson, A., and B. Senauer. 2003. “The Impact of the The Savings in Medicaid Costs for Newborns and Special Supplemental Nutrition Program for Women, Their Mothers from Prenatal WIC Participation in Infants, and Children on Child Health,” American the WIC Program: Volume 2. USDA, Food and Journal of Agricultural Economics 85(2):479-91. Nutrition Service.

Centers for Disease Control and Prevention. 1995. Devaney, B., and A. Schirm. 1993. Infant Mortality “Nutritional Status of Children Participating in the Among Medicaid Newborns in Five States: The Effects Special Supplemental Nutrition Program for Women, of Prenatal WIC Participation. USDA, Food and Infants, and Children B United States, 1988-1991,” Nutrition Service. Morbidity and Mortality Weekly 45(3): 65-69. Devaney, B., P. Ziegler, S. Pac, et al. 2004. “Nutrient Centers for Disease Control and Prevention. 1992. Intakes of Infants and Toddlers,” Journal of the Ameri- “Recommendations for the Use of Folic Acid to can Dietetic Association 104(1, Supplement):S14-S21. Reduce the Number of Cases of Spina Bifida and Other Neural Tube Defects,” Morbidity and Mortality Dietz, V.J., A.L. Baughman, E.F. Dini, et al. 2000. Weekly Report 41(RR-14):1-7. “Vaccination Practices, Policies, and Management Factors Associated with High Vaccination Coverage Chatterji, P., K. Bonuk, S. Dhawan, et al. 2002. WIC Levels in Georgia Public Clinics,” Archives of Participation and the Initiation and Duration of Pediatric Adolescent Medicine 154:184-89. Breastfeeding. Discussion Paper No. 1246-02. Madison, WI: Institute for Research on Poverty. Edozien J., M.D. Boyd, B. Switzer, et al. 1979. “Medical Evaluation of the Special Supplemental Food Cole, N., and M.K. Fox. 2004. Nutrition and Health Program for Women, Infants, and Children,” American Characteristics of Low-Income Populations: Journal of Clinical Nutrition 32:677-92. Volume II, WIC Participants and Nonparticipants. E-FAN-04-014-2. USDA, Economic Research Service. Endres, J., M. Sawicki, and J.A. Casper. 1981. “Dietary Assessment of Pregnant Women in a Supplemental Collins, T.R., S.T. DeMellier, J.D. Leeper, et al. 1985. Food Program,” Journal of the American Dietetic “Supplemental Food Program: Effects on Health and Association 79:121-26. Pregnancy Outcome,” Southern Medical Journal 78(5):551-55.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 169 Chapter 4: WIC Program

Federal Register. 2003. “Special Supplemental Heimendinger, J., N. Laird, J. Austin, et al. 1984. “The Nutrition Program for Women, Infants, and Children Effects of the WIC Program on the Growth of (WIC): Revisions to the WIC Food Packages.” Infants,” American Journal of Clinical Nutrition 40: 68(178):53903-10. 1250-57.

Finch, B.K. 2003. “Socioeconomic Gradients and Low Hicks, L.E., and R.A. Langham. 1985. “Cognitive Birth-Weight: Empirical and Policy Considerations,” Measure Stability in Siblings Following Early Health Services Research 38(6, Part II):1819-41. Nutritional Supplementation,” Public Health Report 100(6):656-62. Fox, M.K., N. Burstein, J. Golay, et al. 1998. WIC Nutrition Education Assessment Study: Final Report. Hicks, L.E., R.A. Langham, and J. Takenaka. 1982. USDA, Food and Nutrition Service. “Cognitive and Health Measures Following Early Nutritional Supplementation: a Sibling Study,” Fraker, T.M., S.K. Long, and C.E. Post. 1990. Analysis American Journal of Public Health 72(10):1110-18. of the 1985 Continuing Survey of Food Intakes by Individuals. Volume I, Estimating Usual Dietary Hirschman, J. 2004. Presentation to Expert Committee Intake, Assessing Dietary Adequacy, and Estimating for Review of Food Practices. Institute of Medicine. Program Effects: Applications of Three Advanced February 26, 2004. Available: http://www.iom.edu/ Methodologies Using FNS’s Four-Day Analysis File. object.file/master/19/293/0.pdf. Accessed June 2004. USDA, Food and Nutrition Service. Hogan, D.P., and J.M. Park. 2000. “Family Factors and Frisbie, W.P., M. Biegler, P. deTurk, et al. 1997. “Racial Social Support in the Developmental Outcomes of and Ethnic Differences in Determinants of Intrauterine Very Low Birthweight Children,” Clinics in Growth Retardation and Other Compromised Birth Perinatology 27(2):433-58. Outcomes,” American Journal of Public Health 87(12):1977-83. Hutchins, S.S., J. Rosenthal, P. Eason, et al. 1999. “Effectiveness and Cost-Effectiveness of Linking the Gordon, A., R. Kliman, J. Ohls, et al. 1999. Estimating Special Supplemental Program for Women, Infants, the Number of People Eligible for WIC and the Full- and Children (WIC) and Immunization Activities,” Funding Participation Rate: A Review of the Issues. Journal of Public Health Policy 20(4):408-26. USDA, Food and Nutrition Service. Institute of Medicine, Food and Nutrition Board. Gordon, A., K. Lewis, and L. Radbill. 1997. Income 2001. Dietary Reference Intakes: Application in Variability Among Families with Pregnant Women, Dietary Assessment. Washington, DC: National Infants, or Young Children. USDA, Food and Academy Press. Consumer Service. Institute of Medicine. 2004. Proposed Criteria for Gordon, A., and L. Nelson. 1995. Characteristics and Selecting the WIC Food Packages. Washington, DC: Outcomes of WIC Participants and Nonparticipants: National Academy Press. Analysis of the 1988 National Maternal and Infant Health Survey. USDA, Food and Nutrition Service. Institute of Medicine. 1996. WIC Nutrition Risk Criteria: A Scientific Assessment, pp.1-22 and 335-52. Gore, S.A., D.M. Brown, and D.S. West. 2003. “The Washington, DC: National Academy Press. Role of Postpartum Weight Retention in Obesity Among Women: A Review of the Evidence,” Annals of Institute of Medicine. 1990. Nutrition During Behavioral Medicine 26(2):149-59. Pregnancy, Part I: Weight Gain. Washington, DC: National Academy Press. Gregory, P., and M. deJesus. 2003. “Racial Differences in Birth Outcomes and Costs in Relation to Prenatal James, J.M. 1998. Immunization and its Implication on WIC Participation,” New Jersey Medicine 100(3): WIC and NonWIC Participants. Unpublished Master’s 29-36. thesis. New York Medical College.

170 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Joyce, T., H. Corman, and M. Grossman . 1988. “A Kresge, J. 2003. WIC Participant and Program Cost-effectiveness Analysis of Strategies to Reduce Characteristics: PC2002, Executive Summary. USDA, Infant Mortality,” Medical Care 26(4):348-60. Food and Nutrition Service.

Kahn, J.L., H.J. Binns, T. Chen, et al. 2002. Lederman, S.A., G. Alfasi, and R.J. Deckelbaum. “Persistence and Emergence of Anemia in Children 2002. “Pregnancy-associated Obesity in Black Women During Participation in the Special Supplemental in New York City,” Maternal and Child Health Nutrition Program for Women, Infants, and Children,” Journal 6(1):37-42. Pediatrics 156:1028-32. Lee, B.J., L. Mackey-Bilaver, and R.M. Goerge. 2000. Kennedy, E.T., and S. Gershoff. 1982. “Effect of WIC The Patterns of Food Stamp and WIC Participation Supplemental Feeding on Hemoglobin and Hematocrit and Their Effects on Health and Low-Income of Prenatal Patients,” Journal of the American Dietetic Children. JCPR Working Paper No. 129. Chicago, IL: Association 80:227-30. Northwestern University/University of Chicago Joint Center for Poverty Research. Kennedy, E.T., S. Gershoff, R. Reed, et al. 1982. “Evaluation of the Effect of WIC Supplemental Lee, J.Y., R.G. Rozier, E.C. Norton, et al. 2004a. Feeding on Birth Weight,” Journal of the American “Effects of WIC Child Participation on Use of Oral Dietetic Association 80(3):220-27. Health Services,” American Journal of Public Health 94:772-77. Kennedy, E.T., and M. Kotelchuck. 1984. “The Effect of WIC Supplemental Feeding on Birth Weight: A Lee, J.Y., R.C. Rozier, E.C. Norton, et al. 2004b. “The Case-control Analysis,” American Journal of Clinical Effects of the Women, Infants, and Children’s Nutrition 40:579-85. Supplemental Food Program on Dentally-Related Medicaid Expenditures,” Journal of Public Health Kotch, J.B., M. McCann, and D. Shanklin. 1989. Dentistry 64:76-81. Assessing the Impact of the WIC Program on Infants and Children: Final Design Report. Chapel Hill, NC: Logan, C., M.K. Fox, and B.-H. Lin. 2002. Effects of University of North Carolina. Food Assistance and Nutrition Programs on Nutrition and Health, Volume 2: Data Sources. FANRR-19-2. Kotelchuck, M., J.B. Schwartz, M.T. Anderka, et al. USDA, Economic Research Service. 1984. “WIC Participation and Pregnancy Outcomes: Massachusetts Statewide Evaluation Project,” Luman, E.T., M.M. McCauley, A. Shefer, et al. 2003. American Journal of Public Health 74(10):1086-92. “Maternal Characteristics Associated with Vaccination of Young Children,” Pediatrics 111(5):1215-18. Kowaleski-Jones, L., and G.J. Duncan. 2000. Effects of Participation in the WIC Food Assistance Program on Mardis, A., and R. Anand. 2000. A Look at the Diet of Children’s Health and Development: Evidence from Pregnant Women. Nutrition Insights No. 17. USDA, NLSY Children. Discussion Paper No. 1207-00. Center for Nutrition Policy and Promotion. Madison, WI: Institute for Research on Poverty. Mays-Scott, C.L. 1991. Adolescent Pregnancy and Kowaleski-Jones, L., and G. Duncan. 2002. “Effects of Infant Outcome: The Influence of Physiological and Participation in the WIC Program on Birthweight: Evi- Socioeconomic Determinants on Pregnancy Outcome. dence from the National Longitudinal Survey of Youth,” Unpublished doctoral dissertation, Texas Women’s American Journal of Public Health 92(5):799-804. University.

Kramer-LeBlanc, C., A. Mardis, S. Gerrior, et al. 1999. Metcoff, J., P. Costiloe, W.M. Crosby, et al. 1985. Review of the Nutritional Status of WIC Participants. “Effect of Food Supplementation (WIC) During USDA, Center for Nutrition Policy and Promotion. Pregnancy on Birth Weight,” American Journal of Clinical Nutrition 41:933-47. Kranz, S., and S.M. Siega-Riz. 2002. “Sociodemo- graphic Determinants of Added Sugar Intake in Miller, V. S. Swaney, and A. Deinard. 1985. “Impact Preschoolers 2 to 5 Years Old,” Journal of Pediatrics of the WIC Program on the Iron Status of Infants,” 140:667-72. Pediatrics 75(1):100-5.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 171 Chapter 4: WIC Program

Moss, N.E., and K. Carver. 1998. “The Effect of WIC Ponza, M., B. Devaney, P. Ziegler, et al. 2004. “Nutrient and Medicaid on Infant Mortality in the United States,” Intakes and Food Choices of Infants and Toddlers American Journal of Public Health 88(9):1354-61. Participating in WIC,” Journal of the American Dietetic Association 104(1, Supplement):S71-S79. National Governor’s Association. 2003. MCH Update 2002: State Health Coverage for Low-Income Pregnant Puma, M.J., J. DiPietro, J. Rosenthal, et al. 1991. Study Women, Children, and Parents. Issue Brief. Washington, of the Impact of WIC on the Growth and Development DC: NGA Center for Best Practices, June 9, 2003. of Children: Field Test, Volume 1: Feasibility Assessment. Cambridge, MA: Abt Associates Inc. New York State Department of Health, Bureau of Nutrition. 1990. The New York State WIC Evaluation: Putnam, J., and S. Gerrior. 1999. “Chapter 7: Trends in The Association Between Prenatal WIC Participation the U.S. Food Supply, 1980-97,” in E. Frazao (ed.), and Birth Outcomes. Albany, NY. America’s Eating Habits: Changes and Consequences. AIB-750. USDA, Economic Research Service. Okita, J.R. 2004. “IOM Project: Review of the WIC Food Packages.” Presentation made to the National Reichman, N.E., and J.O. Teitler. 2003. “Effects of WIC Association Leadership Conference. March 14, Psychosocial Risk Factors and Prenatal Interventions 2004. Available: http://www.iom.edu/ object.file/mason Birthweight: Evidence from New Jersey’s Health ter/19/309/0.pdf. Accessed June 2004. Start Program,” Perspectives on Sexual and Reproductive Health 35(3):130-7. Oliveira, V. 2003. WIC and Breastfeeding Rates. FANRR-34-2. USDA, Economic Research Service. Rose D., J.P. Habicht, and B. Devaney. 1998. “Household Participation in the Food Stamp and WIC Oliveira, V., and C. Gundersen. 2000. WIC and the Programs Increases the Nutrient Intakes of Preschool Nutrient Intake of Children. FANRR-5. USDA, Children,” Journal of Nutrition 128(3):548-55. Economic Research Service. Rose, D., D. Smallwood, and J. Blaylock. 1995. Oliveira, V., E. Racine, J. Olmstead, et al. 2002. The “Socio-economic Factors Associated with the Iron WIC Program: Background, Trends, and Issues. Intake of Preschoolers in the United States,” Nutrition FANRR-27. USDA, Economic Research Service. Research 15(9):1297-1309.

Paige, D.M. 1983. Medical Assistance Costs and Rossi, P.H. 1998. Feeding the Poor: Assessing Federal Utilization Patterns in WIC Enrollees. Baltimore, MD: Food Aid. Washington, DC: The AEI Press, pp. 44-65. Johns Hopkins University, School of Hygiene and Public Health and the Maryland Department of Health Roth, J., J. Yang, S. Wu, et al. 2004. Evaluating the and Mental Hygiene. Impact of WIC Participation on Pregnancy Outcomes: Florida, 1996-2000. Presented at the 16th Annual Partington, S., and S. Nitzke. 1999. “Intake of Food Conference of the Southeast Evaluation Association, Guide Pyramid Servings: A Comparison of WIC January 29-30, 2004. (Initially included in review Children in Wisconsin and Children from 1994 based on preliminary 2000 presentation at American CSFII,” Journal of Nutrition Education 31:38-42. Academy of Pediatrics Meeting (Roth et al., 2000)).

Pehrsson, P.R., P.B. Moser-Veillon, L. Sims, et al. 2001. Roth, J., S. Wu, R. Carter, et al. 2000. “Impact of “Postpartum Iron Stats in Nonlactating Participants Medicaid Managed Care and WIC Participation on and Nonparticipants in the Special Supplemental Infant Birth Outcomes,” Pediatric Research 47(9): Nutrition Program for Women, Infants, and Children,” 1309. (Meeting abstract). American Journal of Clinical Nutrition 73:86-92. Rush D., J.M. Alvir, D.A. Kenny, et al. 1988a. “The Pollitt, E., and R. Lorimor. 1983. “Effects of WIC on National WIC Evaluation: Evaluation of the Special Cognitive Development,” American Journal of Public Supplemental Food Program for Women, Infants and Health 73(6):698-700. Children, III: Historical Study of Pregnancy Outcomes,” American Journal of Clinical Nutrition 48:412-28.

172 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 4: WIC Program

Rush, D., M.R. Kurzon, W.B. Seaver, et al. 1988b. Shefer, A.M., E.T. Luman, B.H. Lyons, et al. 2001. “The National WIC Evaluation: Evaluation of the “Vaccination Status of Children in the Women, Infants, Special Supplemental Food Program for Women, and Children (WIC) Program: Are We Doing Enough Infants and Children, VII: Study of Food to Improve Coverage?” American Journal of Expenditures,” American Journal of Clinical Preventive Medicine 20(1, Supplement 1):47-54. Nutrition 48:512-19. Sherry, B., D. Bister, and R. Yip. 1997. “Continuation Rush D., J. Leighton, N.L. Sloan, et al. 1988c. “The of Decline in Prevalence of Anemia in Low-income National WIC Evaluation: Evaluation of the Special Children: the Vermont Experience,” Archives of Supplemental Food Program for Women, Infants and Pediatrics and Adolescent Medicine 151:928-30. Children, VI: Study of Infants and Children,” American Journal of Clinical Nutrition 48:484-511. Sherry, B., Z. Mei, and R. Yip (2001). “Continuation of the Decline in Prevalence of Anemia in Low- Rush, D., W.B. Seaver, D.G. Horvitz, et al. 1986. The Income Infants and Children in Five States,” National WIC Evaluation: An Evaluation of the Pediatrics 107(4):677-82. Special Supplemental Food Program for Women, Infants and Children, Volumes I-III. USDA, Food and Sherry, B., M. Zuogo, and R. Yip (2001). “Continuation Nutrition Service. of the Decline in Prevalence of Anemia in Low-Income Infants and Children in Five States,” Pediatrics Rush D., N.L. Sloan, J. Leighton, et al. 1988d. “The 107(4):677-82. National WIC Evaluation: Evaluation of the Special Supplemental Food Program for Women, Infants and Siega-Riz, S.M., S. Kranz, D. Blanchette, et al. 2004. Children, V: Longitudinal Study of Pregnant Women,” “The Effect of Participation in the WIC Program on American Journal of Clinical Nutrition 48:439-83. Preschoolers’ Diets,” Journal of Pediatrics 144:229-34.

Ryan, A.S., D.R. Rush, F.W. Krieger, et al. 1991. Silverman, P.R. 1982. The Effect of a Local Prenatal “Recent Declines in Breastfeeding in the United Nutrition Supplementation Program (WIC) on the States: 1984 Through 1989,” Pediatrics 88(4):719-27. Birth Weight of High Risk Infants. Unpublished doctoral dissertation. University of Pittsburgh. Sanders, A., T. Romashko, H. Fleischman, et al. 1990. WIC Breastfeeding Promotion Study and Simpson, K.N. 1988. Analyzing the Influences of Demonstration: Phase IV Report. USDA, Food and Selected Public Prevention Programs on Low Nutrition Service. Birthweight in North Carolina Counties. Unpublished doctoral dissertation, University of North Carolina at Schramm, W.F. 1985. “WIC Prenatal Participation Chapel Hill. and its Relationship to Newborn Medicaid Costs in Missouri: a Cost/benefit Analysis,” American Journal Smith, A., G. Branch, S. Henry, et al. 1986. of Public Health 75(8):851-57. “Effectiveness of a Nutrition Program for Mothers and Their Anemic Children,” Journal of the American Schramm, W. F. 1986. “Prenatal Participation in WIC Dietetic Association 86(8):1039-42. Related to Medicaid Costs for Missouri Newborns: 1982 Update,” Public Health Report 101(6):607-15. Sondik, E.J. 2003. Letter to State vital statistics regis- trars. Available: http://www.cdc.gov/nchs/ Schwartz, J.B., D.K. Guilkey, J.S. Akin, et al. 1992. vital_certs_rev.htm. Accessed June 2004. The WIC Breastfeeding Report: The Relationship of WIC Program Participation to the Initiation and Stockbauer, J.W. 1986. “Evaluation of the Missouri Duration of Breastfeeding. USDA, Food and WIC Program: Prenatal Components,” Journal of the Nutrition Service. American Dietetic Association 86(1):61-67.

Shaheen, M.A., R.R. Frerichs, N. Alexopoulos, et al. Stockbauer, J.W. 1987. “WIC Prenatal Participation 2000. “Immunization Coverage Among Predominantly and its Relation to Pregnancy Outcomes in Missouri: Hispanic Children, Aged 2-3 Years, in Central Los A Second Look,” American Journal of Public Health Angeles,” Annals of Epidemiology 10:160-68. 77(7):813-18.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 173 Chapter 4: WIC Program

Taren, D.L., W. Clark, M. Chernesky, et al. 1990. U.S. General Accounting Office. 1992. Early “Weekly Food Servings and Participation in Social Intervention: Federal Investments Like WIC Can Programs among Low-income Families,” American Produce Savings. GAO/HRD-92-18. Washington, DC: Journal of Public Health 80(11):1376-78. US Government Printing Office.

Tognetti, J., J.D. Hirschman, and J.E. McLaughlin. U.S. General Accounting Office. 1993. Breastfeeding: 1991. “Decline in Breastfeeding?” Pediatrics 88(4): WIC’s Efforts to Promote Breastfeeding Have 873-74. Increased. GAO/HRD-94-13. Washington, DC: U.S. Government Printing Office. Tuttle, C.R. and K.G. Dewey. 1994. “Determinants of Infant Feeding Choices Among Southeast Asian Variyam, J. 2002. “WIC Participation and the Nutrient Immigrants in Northern California,” Journal of the Intake of Preschoolers.” Unpublished paper. American Dietetic Association 94(3):282-86. Vazquez-Seoane, P., R. Windom, and H. A. Pearson. U.S. Department of Agriculture, Food and Nutrition 1985. “Disappearance of Iron-deficiency Anemia in a Service. 1978. CDC Analysis of Nutritional Indices for High-risk Infant Population Given Supplemental Iron,” Selected WIC Participants. New England Journal of Medicine 313(19):1239-40.

U.S. Department of Agriculture, Food and Nutrition Ver Ploeg, M. and D. Betson. 2003. Estimating Service. 2003a. “Legislative History of Breastfeeding Eligibility and Participation for the WIC Program: Promotion Requirements in WIC.” Available: Final Report. Washington, DC: National Academy Press. http://www.fns.usda.gov/wic/ breast-feeding/ bflegishistory.htm. Accessed August 2003. Weiler, P.G., P. Stalker, S.W. Jennings, et al. 1979. “Anemia as a Criterion for Evaluation of a Special U.S. Department of Agriculture, Food and Nutrition Supplemental Food Program for Women, Infants and Service. 2003b. Program data. Available: http://www. Children,” Pediatrics 63(4):584-90. fns.usda.gov/pd. Accessed August 2003. Weimer, J. 1998. Breastfeeding Promotion Research: U.S. Department of Agriculture, Food and Nutrition The ES/WIC Nutrition Education Initiative and Service. 2003c. “WIC Program, Benefits and Services: Economic Considerations. AIB-744. USDA, Economic WIC Food Package.” Available: http://www.fns.usda. Research Service. gov/wic/benefitsandservices/ foodpckg.HTM. Accessed August 2003. Wilde, P.E., P.E. McNamara, and C.K. Ranney. 2000. The Effect on Dietary Quality of Participation in the U.S. Department of Agriculture, Food and Nutrition Food Stamp and WIC Programs. FANRR-9. USDA, Service. 2003d. “Women, Infants, and Children: Economic Research Service. Frequently Asked Questions About WIC.” Available: http://www.fns.usda.gov/WIC/FAQs/faq.htm#8. Yip, R., N. Binkin, L. Fleshood, et al. 1987. Accessed November 2003. “Declining Prevalence of Anemia Among Low-income Children in the United States,” Journal of the U.S. Department of Health and Human Services. 2000. American Medical Association 258(12):1619-23. Healthy People 2010: Understanding and Improving Health, 2nd edition. Yip, R., I. Parvanta, K. Scanlon, et al. 1992. “Pediatric Nutrition Surveillance System: United States, 1980- 91.” Morbidity and Mortality Weekly Report 41(7):1-24.

174 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5 National School Lunch Program

The National School Lunch Program (NSLP) is the who are not certified for meal benefits, elementary oldest and second-largest food and nutrition assistance school students are more likely to participate than sec- program (FANP) in the U.S. Department of Agriculture ondary school students, and males are more likely to (USDA) nutrition safety net. Targeted specifically to participate than females (Fox et al., 2001; Gleason, school-age children, the NSLP is the cornerstone of the 1996; Maurer, 1984; Akin, 1983a). largely school-based child nutrition programs, which also include the School Breakfast Program (SBP), the Since 1998, when the NSLP was expanded to include Child and Adult Care Food Program (CACFP), the after-school snacks, this component of the program has Summer Food Service Program (SFSP), and the been growing steadily. Between FY 2000 and FY Special Milk Program (SMP). 2002, the number of after-school snacks provided through the NSLP increased from 70 million to 123 Schools that participate in the NSLP receive Federal million (USDA/FNS, 2003a).78 reimbursement for each program meal served to students. USDA does not reimburse schools for adult The NSLP is administered by the Food and Nutrition meals, second meals, and a la carte items (including Service (FNS) and its regional offices. At the State extra servings of components of program meals). level, the program is administered by State agencies, Since 1998, the program has also covered snacks most often departments of education. State agencies served to children in after-school programs. Any child oversee Federal reimbursements, provide technical in a participating school is eligible to receive NSLP assistance, and monitor program performance. At the meals; in FY 2002, more than 28 million children par- local level, the program is operated by school food ticipated on an average school day. The program authorities (SFAs). Most SFAs are individual school served more than 4.7 billion lunches and 123 million districts; however, regional school unions and residen- after-school snacks, at a cost of $6.9 billion (USDA, tial childcare institutions also serve as SFAs. Food and Nutrition Service (FNS), 2003a). Federal Subsidies

Program Overview Participating SFAs receive two types of Federal assistance: cash reimbursements and commodities. Cash The NSLP was established in 1946 to “safeguard the reimbursements are based on the number of lunches health and well-being of the Nation’s children and to and after-school snacks served, established reimburse- encourage the domestic consumption of nutritious ment rates, and the poverty level of participating stu- agricultural commodities and other foods....”77 A major dents. A cash subsidy is provided for every program impetus for the program was the prevalence of nutrition- lunch and snack served. Additional cash subsidies are related health problems identified during the screening provided for meals and snacks served to children who of young men for military service in World War II. qualify for free or reduced-price meal benefits. Currently, students eligible for free lunches and snacks Today, almost 99 percent of public schools and 83 per- are those from families with incomes at or below 130 cent of all public and private schools combined partici- percent of the Federal poverty level. Students from pate in the NSLP. Nationally, the program is available families with incomes between 130 and 185 percent of to about 92 percent of all students (Burghardt et al., poverty are eligible to receive reduced-price lunches 1993; Burghardt and Devaney, 1995). On an average and snacks.79 school day, about 60 percent of children in schools that offer the NSLP participate in the program (Fox et al., 2001). Participation varies with household income, 78To be eligible to receive reimbursement for after-school snacks, school age, and gender. For example, studies have shown that districts must participate in the NSLP and must sponsor or operate an after- school program that provides children with regularly scheduled educational students certified to receive free or reduced-price or enrichment activities in a supervised environment (USDA/FNS, 2003b). lunches are more likely to participate than students 79Federal regulations allow schools that operate in high-poverty areas (areas where 50 percent or more of school-age children are eligible for free or reduced-price meals) to receive the “free” reimbursement rate for all 77National School Lunch Act of 1946, Public Law 79-396. after-school snacks, regardless of the students’ family incomes.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 175 Basic cash reimbursement rates for the 2002-03 school The components of the traditional NSLP meal year were $2.14 for free lunches, $1.74 for reduced- pattern are: price lunches, and $0.20 for lunches served to children who purchased meals at the full price (referred to as • Meat or meat alternate: 1 serving per meal “paid meals”).80 Snacks were reimbursed at rates of $0.58, $0.29, and $0.05, respectively. • Vegetables, fruits, and/or full-strength juices: 2 or more servings per meal Children eligible for reduced-price lunches cannot be charged more than $0.40 per lunch. SFAs set their • Grains/breads: 1 or more servings per meal/8 serv- own prices for full-price/paid lunches, but must ings per week operate their school meal service program on a non- • Milk: 1 serving per meal. profit basis (USDA/FNS, 2003c). Of the 4.7 billion lunches served in FY 2002, 48 percent were served Over the years, research has shown that, with few excep- to children eligible for free meals and 9 percent tions, the meals offered in the NSLP provide students the to children eligible for reduced-price meals 81 opportunity to satisfy one-third of their daily needs for (USDA/FNS, 2003a). food energy and an array of essential vitamins and minerals (Burghardt et al., 1993; Wellisch et al., 1983). Schools receive agricultural commodities on an entitlement basis and may also receive bonus com- In the early 1990s, however, USDA’s first School modities. Entitlement commodity assistance is based Nutrition Dietary Assessment Study (SNDA-I) exam- on the number of reimbursable lunches served the ined the nutrient content of school lunches in compari- previous school year. For the 2002-03 school year, son with recommendations included in the Dietary the cash value of entitlement commodities was Guidelines for Americans (USDA and U.S. Department $0.1525 per meal (USDA/FNS, 2003c). Schools may of Health and Human Services, 1990) and the National elect to receive cash in lieu of commodity foods. In Research Council’s Diet and Health report (National addition to entitlement commodities, schools may Research Council, 1989). SNDA-I found that, in com- request bonus commodities—commodities that parison with these guidelines, NSLP meals were high become available through agricultural surplus—in in fat, saturated fat, and sodium, and low in carbohy- amounts that can be used without waste. The types drates (Burghardt et al., 1993). At the time the SNDA- and amounts of bonus commodities available vary I data were collected (the 1991-92 school year), from year to year depending on purchasing decisions schools were not required to offer meals that were made by USDA. consistent with these guidelines.

Nutrition Standards The School Meals Initiative To be eligible for Federal subsidies, NSLP meals must for Healthy Children meet defined nutrition standards. Program regulations In response to the SNDA-I findings, USDA made a have long stipulated that lunches should provide one- commitment to implement the Dietary Guidelines in third of the children’s Recommended Dietary the NSLP. The embodiment of this commitment is the Allowances (RDAs). To ensure that these standards are School Meals Initiative for Healthy Children (SMI). The met, program regulations have historically included SMI, launched in 1995, is designed to improve the food-based menu planning guidelines. These guide- nutritional quality of school meals by providing schools lines, originally known as the “Type A meal pattern,” with educational and technical resources that can be define specific types of food to be offered as well as used to (1) assist foodservice personnel in preparing minimum acceptable portion sizes. nutritious and appealing meals and (2) encourage children to eat more healthful meals. Key components of the SMI include revised nutrition standards for school 80Reimbursement rates for both lunches and snacks are higher in Hawaii meals, a major restructuring of menu planning require- and Alaska. In addition, lunch reimbursement rates are higher for schools ments, and a broad-based nutrition education program. that operate in high-poverty areas (60 percent or more of students eligible for free or reduced-price meals). The nutrition standards established under the SMI 81Information on the percentage of after-school snacks served to children eligible for free or reduced-price meal benefits is not available in pub- maintain the longstanding goal of providing one-third licly available summaries of administrative data. of students’ daily needs for food energy and nutrients. In

176 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program addition, the standards include goals for fat and saturated Related Program Changes fat content that are consistent with the Dietary The SMI has been supported by several parallel initia- Guidelines recommendations.82 The Healthy Meals for tives. For example, considerable effort has been devoted Healthy Americans Act (P.L. 103-448) formally required to improving the nutrient profile of the commodity that school meals be consistent with the Dietary foods provided to NSLP schools (Buzby and Guthrie, Guidelines and that schools begin complying with SMI 2002). In addition, under the Nutrition Title of the nutrition standards in the 1996-97 school year unless a 2002 Farm Act, USDA received $6 million for a pilot waiver was granted by the relevant State agency. The program to provide fresh and dried fruits and fresh regulatory requirement that school meals be consistent vegetables to children in elementary and secondary with the Dietary Guidelines has been incorporated into schools. The pilot program, which was implemented in the FNS strategic plan. The current goal is for all schools the 2002-03 school year, was very well received to satisfy these standards by 2005 (USDA/FNS, 2000a). (Buzby et al., 2003) and was expanded under the Child Under SMI and the Healthy Meals for Healthy Nutrition and WIC Reauthorization Act of 2004 (P.L. Americans Act, menu planning requirements were 108-265). restructured to offer schools several alternatives to the Most recently, policymakers have begun to focus on traditional food-based NSLP meal pattern. These the “school nutrition environment” (Ralston et al., include a computer-based menu planning approach 2003; American School Food Service Association known as Nutrient Standard Menu Planning (NSMP). (ASFSA), 2003; USDA/FNS, 2000b). A school’s NSMP focuses on the nutrient content of meals rather nutrition environment includes the nutritional quality than the specific types of food offered. School districts of reimbursable school meals, the availability and may implement NSMP on their own or may contract nutritional quality of competitive (non-NSLP) foods, with an outside agency. An enhanced food-based meal meal scheduling, physical characteristics of the cafete- pattern was also developed. This is similar to the tradi- ria, nutrition education and marketing activities, and tional pattern but requires more servings of breads and the school’s commitment to nutrition and physical grain products over the course of a week and larger activity. servings of fruits and vegetables. School districts may also use any other reasonable approach to menu plan- Major attention has been focused on the issue of ning, subject to State agency guidelines. Implementation “competitive foods” in school meal programs—foods of new menu planning systems has proved to be a other than those included in NSLP and SBP meals lengthy process. In the 1999-2000 school year, only 63 (USDA/FNCS, 2001). In 1998-99, USDA’s second percent of all SFAs reported that they had “fully imple- School Nutrition Dietary Assessment Study (SNDA-II) mented” the menu planning system of their choice found that more than half of all elementary schools had (Abraham et al., 2002). Eighty-five percent indicated a la carte programs that offered items other than milk, that their plans were at least three-quarters implemented. juice, and desserts (Fox et al., 2001). The same was true of roughly 90 percent of middle schools and high The nutrition education component of the SMI is the schools. Many schools had a la carte programs that Team Nutrition Initiative (TN) (see chapter 16). TN pro- made it possible for students to purchase complete vides technical assistance, educational resources, and meals on an a la carte basis. SNDA-II also demonstrat- training to school foodservice personnel, children, pared that revenue from a la carte sales was inversely ents, teachers, and school administrators. TN uses related to rates of student participation in the NSLP. behavior-oriented strategies to (1) assist school foodservice personnel in preparing and serving meals that The CDC-sponsored School Health Policies and meet the SMI nutrition standards without sacrificing Programs Study (SHPPS) found that more than a quar- taste or attractiveness, (2) promote healthful eating ter of elementary schools, 62 percent of middle/junior habits and regular physical activity among both children high schools, and 95 percent of senior high schools and parents, and (3) build a support base among school had vending machines (Wechsler et al., 2001). A sub- administrators and other school and community partners stantial number of schools also had school stores, can- for healthy patterns of eating and physical activity teens, or snack bars available to students during meal (USDA/FNS, 2002). time. Some school districts have entered into potentially lucrative “pouring rights” contracts that may lead to 82Goals for sodium and cholesterol content are not included in SMI nutrition standards. However, schools are encouraged to monitor levels of increased availability and marketing of soft drinks (Lin these dietary components. and Ralston, 2003; Nestle, 2000).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 177 Chapter 5: National School Lunch Program

Although not part of the reimbursable NSLP meal, com- Consequently, even these relatively recent data may petitive foods contribute to students’ in-school dietary not provide an accurate picture of the nutrient content intake. Currently, there are no Federal nutrition-related of meals currently offered in the NSLP. standards governing competitive foods, and research has shown that foods offered through these alternative It is important to keep the changing nature of the NSLP sources tend to be high in fat, sodium, and/or sugar in mind when reviewing the summary of NSLP research (French et al., 2003; Kubik et al., 2003; Zive et al., 2002; that follows. The existing research provides information USDA/FNCS, 2001; Wechsler et al., 2001; Glengdahl on previous and potential impacts of the NSLP; howev- and Seaborn, 1999). Concerns about the negative impact er, new research is essential to understanding the impact of competitive foods has prompted calls for action at of the NSLP as it operates today (Guthrie, 2003). local, State, and Federal levels to limit their availability and/or to establish nutrition standards for them. Research Overview Implications for Interpreting The literature search identified a total of 26 studies Available Research that examined the impact of the NSLP on nutrition- The vast majority of the research reviewed in this chap- and health-related outcomes of participating children. ter is based on data that were collected before the SMI Among these are two USDA-sponsored national evalu- was launched or in the very early stages of its imple- ations that included student-level outcomes: the mentation. This includes all of the studies that were National Evaluation of School Nutrition Programs national in scope, those with the strongest designs and (NESNP) conducted in 1980-81 and the SNDA-I analysis methods, and all of the studies that looked at study, conducted in 1991-92. A third USDA-sponsored impacts on students’ dietary intakes over 24 hours. national evaluation, the SNDA-II study, conducted in the 1998-99 school year, is not included in this sum- Given the nature and extent of the changes associated mary because it did not collect student-level data. with the SMI—changes that specifically targeted the nutrient content of school lunches and students’ con- Most studies examined impacts on dietary intake at sumption of healthful lunches—it is important that lunch and/or over 24 hours. A smaller number consid- results of the available research be interpreted in the ered impacts of NSLP participation on other measures proper context. Existing research provides a compre- of nutrition and health status or on household food hensive picture of past and potential impacts of the expenditures. One study also looked at impacts on NSLP; however, because of the major changes that school attendance and cognitive performance. have been implemented under SMI and related ongo- The majority of the available research is quite dated. ing changes, it cannot be assumed that these findings Fifteen of the 26 studies used data that were collected apply to today’s NSLP. during or prior to the early 1980s. And, as noted, Indeed, there is evidence that the nutrient content of almost all studies used data that were collected prior to meals offered in the NSLP has changed since the imple- implementation of the SMI. mentation of the SMI. The SNDA-II study found that, Measures of Participation relative to lunches offered in 1991-92 (as reported in SNDA-I), lunches offered in 1998-99 were significantly Measures of program participation varied across studies lower in total fat, saturated fat, and sodium (although, (and sometimes within studies, depending on the out- on average, lunches continued to exceed Dietary come being evaluated). Most studies equated the pur- Guidelines and NRC recommendations for those nutri- chase or consumption of a school lunch on the day(s) ents). Moreover, SNDA-II demonstrated that reductions of dietary assessment with program participation. This in fat and saturated fat content could be achieved with- entails some risk that children who usually eat a school out sacrificing overall nutrient content. That is, although lunch did not do so on the day of the survey, or that lower in fat, NSLP lunches continued to meet the goal some who ate the school lunch on that day usually do of providing one-third of the RDAs for key nutrients. not participate. However, NESNP researchers evaluated the extent of this problem and concluded that defin- The SNDA-II data were collected relatively early in ing participation on the basis of one day’s behavior the implementation of the SMI and, since that time, gave an accurate picture of participation, at least with efforts to implement the SMI nutrition standards have the large sample available in that study and with data continued at the Federal, State, and local levels. collected on all 5 weekdays (Wellisch et al., 1983).

178 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program

Several studies defined program participation on the with intakes below a defined cutoff. Again, none used basis of usual NSLP participation practices, such as the approach recommended by the IOM, which calls whether a student usually ate a school lunch a mini- for use of data on usual intake in conjunction with mum number of times per week or the proportion of defined Estimated Average Requirements (EARs) potential lunches eaten during the study period. In (IOM, 2001). evaluating the impact of the NSLP on students’ linear growth, the NESNP used historical information on stu- Consequently, the available research presents an dents’ participation from grade 1 through the current imperfect picture of the substantive significance of dif- grade, and computed an average weekly NSLP partici- ferences observed in the dietary intakes of NSLP par- pation rate for each student (Wellisch et al., 1983). ticipants and nonparticipants. It provides information on whether NSLP participants consumed more or less Definition of nonparticipant comparison groups also energy and nutrients than nonparticipants. However, differed across studies. Most researchers used nonpar- this information cannot be used to draw conclusions ticipants in the same schools as NSLP participants. about whether NSLP participants were more or less Hoagland (1980) distinguished between students who likely than nonparticipants to have adequate intakes. did and did not have the NSLP available to them. The NESNP oversampled schools that did not offer the In addition, research has shown that assessing the NSLP. However, investigators ultimately concluded dietary intakes of children presents unique challenges that schools not offering the NSLP did not constitute (Medlin and Skinner, 1988; Baranowski and Simons- an appropriate comparison group, so they used nonpar- Morton, 1991) and that recall-based data collection ticipants in NSLP schools instead. methodologies do not necessarily work well with young children (Baxter, 2000 and 2003). As a result, The following two sections summarize major findings recall-based measures of the dietary intakes of NSLP from existing research on the impact of the NSLP on participants and nonparticipants are subject to nutrition- and health-related outcomes. The first sec- increased measurement error. tion summarizes studies that assessed impacts on students’ dietary intake. The second section discusses Research Overview studies that examined impacts on other nutrition and The literature search identified 19 studies that exam- health outcomes, including weight and height, nutri- ined the impact of NSLP participation on students’ tional biochemistries, household food expenditures, dietary intakes. Characteristics of these studies are and school performance. summarized in table 26. Most studies examined impacts on intake of food energy and nutrients. Impacts on Dietary Intake However, three studies (Devaney et al., 1993; Gleason and Suitor, 2001; Rainville, 2001) looked at impacts Estimates of NSLP impacts on dietary intake are sub- on food consumption as well as energy and nutrient ject to the limitations discussed in chapter 2. Most intake, and four studies (Cullen et al., 2000; Melnick studies used data for a single day or meal, and there- et al.,1998; Wolfe and Campbell, 1993; Yperman and fore provide weak estimates of individuals’ usual Vermeersh, 1979) looked only at impacts on food con- intake. Some studies used multiple days of data or sumption. Five studies focused exclusively on impacts other means (such as a food frequency checklist) to on dietary intake at lunch, seven studies looked at both better capture usual intake. However, none of the lunch and 24-hour intakes, and seven studies focused available studies used the approach to estimating usual exclusively on 24-hour intake. intake that was recently recommended by the Institute of Medicine (IOM, 2001).83 The available research can be divided into three groups. Group I includes the two national evaluations that exam- Similarly, in assessing intakes of food energy, vita- ined student-level outcomes: SNDA-I (1991-92 school mins, and minerals, researchers generally compared year) and NESNP (1980-81 school year). Group II mean intakes of participants and nonparticipants or includes five studies that are based on secondary compared the proportion of individuals in each group analysis of data from national cross-sectional surveys. Two of these studies (Gleason and Suitor, 2001 and 83Gleason and Suitor (2001) used these methods to describe dietary 2003) used data collected between 1994 and 1996. The intakes of U.S. children. However, in assessing differences in intakes of NSLP participants and nonparticipants, they used regression-adjusted mean other three studies in this group are based on data from intakes that were based on 1 or 2 days of data. the late 1970s or early 1990s. Group III consists of 11

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 179 Chapter 5: National School Lunch Program e n n n n od d h e t s model rients) s t t Continued— s for full s (foods) u bias- gressio gressio gressio w-incom j o d l a on- - te re te re te re n a a a ison of o ent (nu i s s Analysis met e r g e means Multivari with fixed-effec to control for selectio bias Multivari with selecti Multivari Compar r Bivariate t-test sample and sample adjustm Bivariate t-test h on h on h on h on h on Measure of participation y intakes Ate NSLP lunc recall day Ate NSLP lunc recall day Ate NSLP lunc recall day Ate NSLP lunc recall day Ate NSLP lunc recall day t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant ents’ dietar n n n n n Design partici partici partici partici partici Participa non Participa Participa Participa Participa non non non non s 1 e ) ) e 4 n a 1 l day 6 , nts nts nts in nts in nts in Program on stud latio 1 18 with 18 with = n ( ,350) ,556) ,866) ,556) es 1-12 es 1-12 es 1-12 esce esce esce esce esce dren and dren and dren and dren and dren and 3 6 1 6 a Popu t (sample siz a Chil adol Chil adol grad (n~ Chil adol grad (n= Chil adol ages 6- 2 days of intak d Chil adol ages 6- or 2 schoo of intake dat (n= grad (n= e e v v lls lls uti uti al School Lunch hour hour hour on onsec onsec method e 24- e 24- e 24- our reca our reca Data collection Singl Singl recall recall recall 24-h 24-h Singl 2 nonc 2 nonc the Nati f P 1 N d pact o ools tative tative m ic an ic surveys n n ally ally ubl ubl 0-81) 1-92) 4-96 CSFII 4-96 CSFII 0-81 NES onal Data source Nation represe sample of students from 329 p Nation represe sample of students from 276 p schools (198 199 199 private sch (199 198 er at at is of nati e e ov take nd nd nd nd nd ours ours ours ours ours n urs Outcome(s) Nutrient i Nutrient intake at lunch a over 24 h Food intak lunch Nutrient intake at lunch a over 24 h Nutrient intake at lunch a over 24 h Nutrient intake at lunch a over 24 h Food intak lunch and 24 ho at lunch a over 24 h of table. udies that examined the i t 7) at end 03) 01) and and II: Secondary analys I: National evaluations n n 3) llisch et al. e See notes Table 26—S Study Group Devaney et al. (1993) (SNDA-I) W (198 (NESNP) Group Gleaso Suitor (20 Gleaso Suitor (20 Fraker (198

180 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program n laysis od h ance ance s Continued— s and ession; gressio vari vari usted an te re a g regr adj Analysis met Analysis of Bivariate t-test chi-square test Chow tests Multivari Switchin of covariance Analysis of Gender- inued t (vs. r of r of data h on h on h on h on nch y er of er of u a be be unc unc unc unc mb mb u u h) tion d a Measure of participation y intakes—Con observ sack lunc Ate school l recall day Ate school l days ate school lunch to n days of dietary days ate school lunch to n days ate any l Ratio of num Ratio of num Ate school l Ate school l recall day recall day 3 , 2 4 2 2 t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant ents’ dietar n n n n n n Design partici partici partici partici partici partici Participa non Participa Participa non non Participa Participa Participa non non non ) 5 5 e n n n n nts nts nts Program on stud latio 18 18 21 ,397) ,797) ,554) ,554) ,155) es 2-4 es 2 and es 2 and esce esce esce dren i dren i dren i dren and dren and dren and 1 1 1 1 3 570) Popu (sample siz grad (n= Chil grad grad (n= (n= Chil Chil Chil adol ages 6- (n= Chil adol ages 6- (n= Chil adol ages 6- (n= rd rd n tion e) e) School Lunch plus plus o o o a ll ll al rec rec hour d d ntitativ ntitativ ste method a e 24- our reca our reca qua qua Data collection 24-h 24-h and w Visual observ of food selecti Single 24-hour recall (non Singl Single 24-hour recall (non 2-day foo 2-day foo recall the Nation f ty 8) 1 w 0 1 99 pact o ools rk Ci Ne o d m n n led nts in g New rge samples a astern an (1 in l 7-88) 9-90) omly ic an 7-78 NFCS 7-78 NFCS 1-74 Data source Michig York State, exclud York City (198 schools i southe schools i 197 197 197 NHANES-I Students in 1 All stude rand selected classrooms in 25 samp publ private sch in New Y Students in 5 (198 at studies with e e e ours ours ours ours Outcome(s) Nutrient intake over 24 h Nutrient intake over 24 h Nutrient intake over 24 h Nutrient intake at lunch Food intak lunch Food intak over 24 h Food intak at lunch of table. udies that examined the i ate and local t at end ll IIIA: St e e nd ill la 0) 1) 8) 3) 3a) 3b) lfe and o See notes Table 26—S Study Akin et al. (198 Akin et al. (198 Hoag Group Rainv (200 Melnick et al. (199 W Campb (199 (198

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 181 Chapter 5: National School Lunch Program nd n uls od h sts means means ance ance a an-Ke Continued— e gressio vari ewm te re of vari a ison of ison of d) d) Analysis met e test rang Compar (type of statistical test not reporte Analysis Analysis of Multivari Student-N Compar (no statistical t reporte Unmatched t-test inued t re o h) on ool y: y h (vs. h (vs. h (vs. d on s h) on ys ys study ne eek, week, week a al hi nc se d unc da da a unc ing n d d on h or o que i t unch or a bag l at least ly ate sch v ng mac d r Measure of participation e y intakes—Con s b offered dur perio lunch) on o Participati dummies b lunch 3 times/week 0-1 time per w 2-3 times per 4-5 times per Ate NSLP lunc home l snack bar l food recor Took 70% or m of school me Usual Ate NSLP lunc brown usual fre Ate NSLP lunc sack lunc vendi food recor 6 2 7 t vs. ts, t vs. t vs. t vs. t vs. pant pant pant pant pant ents’ dietar n n n n n n Design partici partici partici partici partici non Participa before vs. after Participa non non Participa Participa non non Participa Participa ) ) 6 8 e n 282 28) n n n nts nts in Program on stud 7 latio -16 0 ,495) es 1-4 es 5 and e 5 (n= es 7 and esce esce dren i dren i dren ages dren i dren and dren and 8 233) 512) 254) Popu (sample siz (n= grad grad (n= Chil Chil Chil 8-12 (n= Chil adol ages 1 (n= grad (n= Chil adol grad Chil end rd e n tion k School Lunch v o daily i o a t al s rec d hour calls, d utive ste method a e 24- Data collection 3-day foo Singl 3 nonconsecu Single 24-hour recall 24-hour re including 1 wee day recall and w of food selecti Visual observ food recor 5 consec the Nation f t 5 o e 0 ) n n i i es r 1 l l e n 71 ty o o i d pact o o o 1 le h h on m n ican- nts in nts in grad c c districts ons i w York i s s d) d eers small samples en from 1 e e l l 2) 1-73) 9) ama d d d d ashingt Data source i i State, Blacks and Mex Americans w W in 1 district rural Ne State (1970- Students in 1 m selected classrooms in 3 schools i district in Alab Students in 1 m All stude Students in schools/ in 8 reg All stude selecte Sample of childr States, plus volunt (197 oversamp (197 Texas (dates n reporte Salt Lake C (198 at studies with e nd ours ours ours Outcome(s) lunch Nutrient intake over 24 h Nutrient intake at lunch a over 24 h Food intak Nutrient intake at lunch Nutrient intake at lunch Nutrient intake over 24 h of table. udies that examined the i ate and local t at end t of e r n a f l on, IIIB: St e 8) 2) 0) 1) 4) W en et al. d See notes n Table 26—S Study Price et al. (197 Emmons et al. (197 U.S. Departme Health, Educati a (HEW) (10- State Nutrition Survey) Group Cull (200 Ho et al. (199 Perry et al. (198

182 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program ct n stri od h of gressio cond di te re a nalysis the se e utritionally adequate, Analysis met stics: n 2-way a varianc Multivari ri inued e t aract ut the analysis of ate 5 s h on n y a h o r to data o unc Measure of participation y intakes—Con d on baseline ch n the analysis, b se Number of d school l days pri Ate NSLP lunc recall day collection ely i at s ba roup separ t vs. t vs. pant pant ents’ dietar n n ered t subg id Design s partici partici con Participa non Participa non differen r were ) e n ricts d for fou 104) n Program on stud rte latio (n= 1 repo escents in es 10 es 1-3 dren i 307) e two dist Popu (sample siz grad (n= Chil Adol grad and 1 daily. red. Th Results were re offe y School Lunch te NSLP al enc hour st we . y fa method h program. udents a e 24- c eak r income. st - Data collection Food frequ Singl checklist recall ee lun the Nation r f d free b ination Surve m 1 viduals. nd not low . pact o l in 60 percent of ms. y unch an lic Indi h), a ction of a f y s; c m n ra nts in 2 hoo nia s b omly s. prog e in 2 n pub h free l nd lunch meals. student Data source ption Surve Rand selected students in 1 urba All stude classrooms per grad schools i Califor high sc Kansas and Nutrition Exa d Intake ast a m u s her lunch er than participant Foo where bot and after introdu eakf gible for free lun r i e nd ours ours s rath rs. of b s he with non c s before come (el skippe ide Food Con ng Survey of cond district Outcome(s) Nutrient intake at lunch a over 24 h Food intak over 24 h nch was lun iate NLSP and ot a se d intake skippers s y, low-in udies that examined the i contribution t s: pare for lu 0) ce d rate r n and nd CSFII = Continui NHANES-I = First National Health NFCS = Nationw n (198 9) Data sou Did not different Included lunch Accounte Study included Study com Unit of analysi 1 2 3 4 5 6 7 Table 26—S Study Howe a Vade Yperma Vermeersch (197 did not sepa nutritionally need

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 183 Chapter 5: National School Lunch Program

State and local studies. Six of these studies (Group III- to many relevant variables and their findings, including A) included relatively large numbers of children (more differences between selection-adjusted and unadjusted than 500) from multiple sites—schools, SFAs, or results, make intuitive sense. Others reviewing the same States. The remaining five studies (Group III-B) had literature (see, for example, Rossi, 1998 and Devaney substantially smaller samples and generally weaker et al., 1997) have reached the same conclusion. designs. With the exception of studies by Rainville (2001) and Cullen et al. (2000), all of the Group III A more recent study by Gleason and Suitor (2003) studies are based on data collected before the imple- used data from the 1994-96 wave of the CSFII. This mentation of the SMI. Six are based on data from the study improved upon the SNDA-I analysis by using a mid-1980s or earlier. fixed-effects model to control for selection bias. The analysis included 1,614 children who (1) attended The strongest evidence about the impact of the NSLP schools where the NSLP was offered and (2) had 2 on the dietary intake of participating students comes days of intake data, at least one of which was a school from the SNDA-I study (Devaney et al., 1993) and an day. The fixed-effects model was estimated in a analysis by Gleason and Suitor (2003) that used data paired-differences form, where differences between the from the 1994-96 Continuing Survey of Food Intakes 2 days of intake data were regressed on corresponding by Individuals (CSFII). SNDA-I is the most recent, differences in student characteristics, including NSLP comprehensive, and state-of-the-art study designed participation status. Thus, the estimation of NSLP specifically to study the NSLP. It included nationally impacts was based on variation in NSLP participation representative samples of public and private schools status of specific individual students rather than on and of students attending those schools. Information, variation in participation status of different groups of including a single 24-hour recall, was collected for students. This ensured that the estimate was not influ- 3,350 school-age children in 329 schools. enced by unmeasured differences that may have existed between different groups of students. SNDA-I used a participant vs. nonparticipant design. However, Devaney and her colleagues used an instru- The analysis included both students who reported mental variables approach to control for selection bias. intake for 2 school days and those who reported intake The authors confirmed the robustness of their results for 1 school day and 1 non-school day. To control for using a variety of specifications. The model used to the possibility that students’ intakes varied on school estimate impacts on dietary intake at lunch controlled and non-school days for reasons other than the NSLP, for the price charged for a full-price lunch; student sta- the model included a dummy variable that indicated tus with regard to free and reduced-price meal bene- whether the intake day was a school day. The model fits; interaction terms for the price of a full-price lunch also attempted to control for potential unobserved dif- and benefit eligibility categories; availability of offer- ferences that may have had varying influences on chil- vs.-serve;84 ability to leave school for lunch; availabil- dren’s consumption behaviors on the 2 days. For ity of low-, moderate-, and high-fat lunches;, and serv- example, it included the day of the week, the number ing capacity of the lunch room. In estimating impacts of hours of television watched on the intake day, two on 24-hour nutrient intake, researchers adjusted for variables that indicated frequency of exercise, and self-selection into the NSLP but not into the SBP variables that indicated whether reported intakes were because they concluded, based on exploratory analy- heavier or lighter than usual. ses, that there was no selection bias in breakfast intakes. An earlier study by Gleason and Suitor (2001) also used the 1994-96 CSFII. However, that study did not Selection-bias adjustments are not without problems and attempt to control for selection bias. The authors raised frequently produce implausible results (see discussion appropriate concerns about likely selection bias and in chapter 4). SNDA-I analysts, however, had access cautioned that estimates of differences between NSLP participants and nonparticipants observed in that analysis should not be interpreted as valid estimates of 84Offer vs. serve (OVS) is a NSLP policy that allows students to refuse NSLP impacts. some of the foods offered to them in reimbursable school lunches. At the time the SNDA-I data were collected, OVS was mandatory for secondary Although SNDA-I (Devaney et al., 1993) and the most schools and was optional, at the discretion of local authorities, for elementary schools. Under OVS, students could refuse two of the three meal components recent study by Gleason and Suitor (2003) provide the in the traditional food-based meal pattern that was in effect at the time. strongest available data on NSLP impacts, both studies

184 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program are based on data collected prior to the SMI. The liter- and Suitor (2003) are based on fixed-effects models. ature search identified only two studies that compared For the most part, findings from the two studies are dietary intakes of NSLP participants and nonpartici- consistent. Where findings diverge, Gleason and pants using data collected sometime after the SMI was Suitor’s results are considered stronger because of the implemented (Rainville, 2001; Cullen et al., 2000).85 improved methods used to control for selection bias. Rainville looked at both food and nutrient consumption at lunch, comparing intakes of students who ate SNDA-I researchers stressed the importance of look- NSLP lunches and students who ate lunches from ing separately at NSLP effects by both age and gender. home. The study by Cullen et al. looked only at con- They pointed out, for example, that lunch options are sumption of fruits and vegetables at lunch, comparing usually more varied for older students and that these contributions of NSLP lunches and snack-bar lunches. students typically make their own decisions about Both of these studies were local in scope and both what to eat for lunch. Younger students, on the other have substantial methodological limitations relative to hand, generally have fewer options and decisions SNDA-I and Gleason and Suitor (2003), particularly about their lunches are often made by parents. with regard to generalizability and selection bias. Moreover, research has shown that adolescent females However, when viewed in concert with findings from are more likely than males or younger children to con- SNDA-II, these more recent studies provide suggestive sume diets low in nutrients relative to the RDAs. evidence of post-SMI impacts of the NSLP. In SNDA-I, selection-bias adjustments made little dif- Impacts on Intake of Food Energy ference in conclusions about NSLP effects on younger and Nutrients at Lunch children, but substantially affected the conclusions about older students, particularly females. SNDA-I Nine studies examined the impact of NSLP participation conducted subgroup analysis by age and gender (6- to on students’ intake of food energy and nutrients at lunch. 10-year-olds, 11- to 18-year-old males, 11- to 18-year- Results of these studies are summarized in table 27. old females) and by income (low-income and non-low- The table is divided into four sections: food energy income (income greater than 185 percent of poverty)). and macronutrients, vitamins, minerals, and other In table 27, results of SNDA-I subgroup analyses are dietary components. The text follows this general reported when estimates for one or more subgroups organization, but combines findings for vitamins and differed from results of the overall analysis and when minerals in one section. the result of one of the analyses—the overall analysis or the subgroup analysis—revealed a statistically sig- In the interest of providing a comprehensive picture of nificant difference. the body of research, both significant and nonsignificant results are reported in table 27 and in all other Food Energy and Macronutrients “findings” tables. As noted in chapter 1, a consistent pattern of nonsignificant findings may indicate a true Findings from SNDA-I (Devaney et al., 1993) and underlying effect, even though no single study’s results Gleason and Suitor (2003) suggest that, prior to the would be interpreted in that way. Readers are cau- implementation of the SMI, NSLP participants and tioned, however, to avoid the practice of “vote count- nonparticipants consumed roughly equivalent amounts ing,” or adding up all the studies with particular of food energy at lunch. (Note that results are reported results. Because of differences in research design and in table 27 using only the senior author’s name and other considerations, findings from some studies merit that SNDA-I results are reported as Devaney (1993).) more consideration than others. The text discusses Neither study found a significant difference in the methodological limitations and emphasizes findings energy intakes of NSLP participants and nonpartici- from the strongest studies. In this case, emphasis is pants at lunch. Interestingly, however, both sets of given to findings from SNDA-I (Devaney et al., 1993) researchers found that impact estimates that were not and from the most recent Gleason and Suitor (2003) adjusted for selection bias showed that NSLP partici- study, for the reasons discussed previously. All find- pants consumed significantly more food energy than ings reported for SDNA-I are based on selection-bias- nonparticipants (data not shown). Devaney and her adjusted models, and all findings reported for Gleason colleagues attributed the difference between the two results to differences in unobserved characteristics that may affect participation, such as differences in 85Studies that examined the nutrient content of NSLP and non-NSLP meals as offered or served, but did not assess food and/or nutrient intakes appetite, food preferences, and food energy needs. These of NSLP participants and nonparticipants, were not included in this review. factors are controlled for in the selection-bias-adjusted

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 185 Chapter 5: National School Lunch Program } } d ] less l d l] l] d a ct n a o Continued— i nch nal] nal] t a n nsume ups note up note [ [nationa [nationa ) ) ) ro ro ant imp } 3 1 3 g g 0 9 ts co es at lu 7) [natio 7) [natio 9 n 1 ( (200 (200 y Signific n n e n a y intak v Participa {except sub {males, 5-1 {except sub e Fraker (198 Fraker (198 Gleaso Gleaso D dietar nts’ less ] ] l l d a a n n o o i i nal] nal] nal] t t a a nsume n n [ [ } [10 schools] [10 schools] [10 schools] ) ) 0 ) ) ) 3 3 1 1 1 21} 21} ts co 9 9 7) [natio 7) [natio 7) [natio n 9 9 4) [3 schools] 1 1 ( ( y y h Program on stude e (200 e (200 e (200 e e ill ill ill n n a a v v Participa {males, 15- {11-18; females} {males, 15- {females, 5-1 e e t impact Rainv Fraker (198 Fraker (198 Fraker (198 Perry (198 Rainv Rainv D D n ifica d } } } d d ] ] l l d l] No sign a a n n tional School Lunc } o o i i 1 nsume nal] nal] nal] t t e a a m co } n n ups note ups note up note [ [ o 1 e Na [10 schools] [nationa c ) ) ts 2 ) ) ro ro ro n 3 3 n i 1 3 g g g - 9 9 7) [natio 7) [natio 7) [natio 9 9 w more/same 1 1 o l ( ( (200 ; y y e (200 n 8 91) [1 school] e e 1 ill n n Participa - a a 1 v v 1 { {except sub {except sub {females, 15- {except sub e e Fraker (198 Fraker (198 D Ho (19 Perry (1984) [1 site] D Rainv Gleaso Howe (1980) [1 site] Fraker (198 e impact of th } } d ] ] ] l l l d l] l] l] l] l] l] l] l] l] a a a ct examined th n n n l] a t o o o i i i 2 nal] nal] t t t a a a } n n n tha ups note up note [ [ [ me} 4 s [10 schools] [nationa [nationa [nationa [nationa [nationa [nationa [nationa o ) ) ) 1 ) ) ) ) ) ) ) ) ) [nationa ) [nationa ro ro ant imp 3 3 3 1 1 1 1 3 3 3 1 g g 9 9 9 7) [natio 7) [natio 9 9 9 983 983 0) [1 schoo 1 1 1 w-inc ( ( ( studie 98 (200 (200 (200 (200 (200 (200 (200 y y y e (200 Signific n n n n n n n 91) [1 school] 91) [1 school] 91) [1 school] 91) [1 school] e e e ill n n n a a a llisch (1 llisch (1 v v v e e {6-10; lo {except sub {except sub {females, 11- Participants consumed more e e e Gleaso Gleaso Ho (19 Ho (19 Ho (19 Fraker (198 W Howe (1 Gleaso Ho (19 W Fraker (198 D Gleaso Rainv D Gleaso D Gleaso Gleaso of table. s Findings from e t at end a r d ergy y h energy and macronutrients o b r See notes a Table 27— Outcome Food Food en Protein C Fat Saturated fat

186 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program } ] less l d l] d a ct n a o Continued— i 2 t a n nsume up note [ [10 schools] ) ) ) [nationa ro ant imp 3 1 g 9 ts co 9 n 983 1 ( y e (200 Signific 91) [1 school] e ill n a llisch (1 v Participa e {except sub e Rainv Ho (19 D W less ] ] ] ] l l l l d l] l] l] l] l] a a a a n n n n o o o o i i i i t t t t a a a a nsume n n n n [ [ [ [ [nationa [nationa [nationa [nationa [nationa me} ) ) ) ) ) ) ) ) ) 3 3 3 3 3 1 3 3 3 co ts co 9 9 9 9 n n 9 9 9 9 1 1 1 1 ( ( ( ( (200 (200 (200 (200 (200 } y y y y h Program on n n n n n low-i 8 e e e e 1 n n n n - a a a a 1 v v v v Participa 1 {non- { e e e e t impact Gleaso D Gleaso D Gleaso Gleaso D D Gleaso n ifica d } ] ] ] ] l l l l d l] l] l] l] No sign a a a a n n n n l] tional School Lunc o o o o i i i i 1 nsume t t t t a a a a } } ; ; n n n n up note [ [ [ [ e e 8 8 [nationa [nationa [nationa [nationa me} ) ) ) ) ts co 1 1 ) ) ) ) ro m m 3 3 3 3 n 3 1 3 1 g o o co 9 9 9 9 c c n 9 9 9 9 n n 0) [1 schoo more/same 4) [3 schools] i i 1 1 1 1 - - ( ( ( ( 98 (200 (200 (200 (200 w w y y y y n n n n low-i o o 91) [1 school] e e e e l l - - n n n n Participa n n a a a a o o v v v v {non- {females, 11- n {females, 11- n {except sub impact of the Na e e e e Ho (19 Gleaso D D D Gleaso Perry (198 Gleaso Gleaso Howe (1 D e } } } d d ] ] ] l l l d l] l] l] l] l] l] l] l] l] l] l] a a a ct examined th n n n l] l] l] a t o o o Continued i i i t t t a a a n n n tha ups note ups note up note [ [ [ s [10 schools] [10 schools] [10 schools] [10 schools] [10 schools] [10 schools] [10 schools] [nationa [nationa [nationa [nationa [nationa [nationa [nationa ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) [nationa ) [nationa ) [nationa ) [nationa ro ro ro ant imp 3 3 3 1 1 1 1 1 1 1 1 1 1 3 1 1 3 g g g 9 9 9 9 9 9 983 983 983 983 0) [1 schoo 0) [1 schoo 0) [1 schoo 4) [3 schools] 4) [3 schools] 1 1 1 ( ( ( studie 98 98 98 (200 (200 (200 (200 (200 (200 (200 y y y e (200 e (200 e (200 e (200 e (200 e (200 e (200 Signific n n n n n n n e e e ill ill ill ill ill ill ill n n n a a a llisch (1 llisch (1 llisch (1 llisch (1 v v v e e e e {except sub {except sub {except sub Participants consumed more e e e Gleaso Rainv Howe (1 Gleaso Gleaso W Perry (198 Howe (1 W Perry (198 Howe (1 Rainv D Gleaso Gleaso Rainv Gleaso Gleaso Rainv Rainv Rainv W W D D Rainv y intakes at lunch— of table. Findings from at end 6 12 vin a ents’ dietar n See notes Table 27— stud Outcome Vitamins Vitamin A Vitamin B Vitamin B Vitamin C Vitamin D Vitamin E Folate Niaci Ribofl

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 187 Chapter 5: National School Lunch Program less l] d ct a Continued— nsume ) [nationa ant imp ts co n 983 Signific llisch (1 Participa e {older} W less ] ] l l d a a n n o o i i t t a a } nsume ; n n [ [ e 8 ) ) 1 m 3 3 o ts co 9 9 c n 9 9 n i 1 1 - ( ( w y y h Program on o e e l - n n n a a o v v Participa {females, 11- n e e t impact D D n 3 ifica } me o d } ] ] ] ] ] l l l l l d l] l] No sign a a a a a n n n n n w-inc tional School Lunc o o o o o i i i i i 1 nsume t t t t t a a a a a co } n n n n n up note [ [ [ [ [ on-lo 8 w -income} w -income} [nationa [nationa ) ) ) ) ) ts 1 ) ) ro o o 3 3 3 3 3 n 3 3 g 9 9 9 9 9 18; n 9 9 9 9 9 more/same 4) [3 schools] 1 1 1 1 1 ( ( ( ( ( (200 (200 ; non-l ; non-l y y y y y n n 8 8 91) [1 school] e e e e e n n n n n Participa a a a a a v v v v v {11-1 {11-1 {females, 11- {6-10; 11- {except sub impact of the Na e e e e e Ho (19 D D D D Gleaso Perry (198 D Gleaso e 3 } } } } d d d ] ] ] ] ] l l l l d l] l] l] l l] l] l] l] l] l] l] l] l] l] l] a a a a ct a examined th n n n n l] l] l] a n t o o o o Continued o i i i i i t t t t t a a a a a n n n n tha n ups note ups note ups note up note [ [ [ [ [ s [10 schools] [10 schools] [10 schools] [10 schools] [nationa [nationa [nationa [nationa [nationa [nationa [nationa [nationa [nationa ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) [nationa ) [nationa ) [nationa ) [nationa ) [nationa ro ro ro ro ant imp 3 3 3 3 1 1 1 1 3 1 1 1 3 3 1 1 3 1 g g g g 9 9 9 9 0 9 9 9 9 0 983 983 983 983 983 0) [1 schoo 0) [1 schoo 0) [1 schoo 4) [3 schools] 4) [3 schools] 1 1 1 1 2 ( ( ( ( ( studie 98 98 98 (200 (200 (200 (200 (200 (200 (200 (200 (200 y y y y ger} e (200 e (200 e (200 e (200 Signific n n n n n n n n n n e e e e o ill ill ill ill n n n n s a a a a a llisch (1 llisch (1 llisch (1 llisch (1 llisch (1 v v v v e e e e e e {except sub {except sub {except sub {except sub {youn l Participants consumed more e e e e Gleaso Gleaso G Gleaso W Perry (198 Rainv Gleaso W Perry (198 Howe (1 W Gleaso Rainv Howe (1 W D Gleaso D Rainv W Howe (1 D Gleaso Rainv D Gleaso Gleaso y intakes at lunch— of table. Findings from at end horus m esium ents’ dietar u i c l See notes a Table 27— stud Outcome Thiamin Minerals C Iron Magn Phosp Zinc

188 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program y less ses l] l] d s ct a ific nal] c ne of the nsume more spe } [10 schools] [nationa [nationa ext discu 0 ) adding up all the a ) ) ant imp 1 3 1 ts co 7) [natio n The t significant findings ma at least o (200 (200 ting,” or e (200 Signific n n ts. ill and tain only to e l ed significantly Participa p {females, 5-1 per m m han others. u Fraker (198 Gleaso Gleaso Rainv studen a s e s attern of non “vote coun l l con a r e v tice of o ideration t c istent p r s s o f s } rticipants Where findings con t l less d ] ong low-incom l nt. u d re con s . a m t e n 1, a n r fica r oid the pra o i v nal] t m nly a a schools). o NSLP pa nsume } n 3 ups note [ 8 hapte merit mo ) d to a 1 ro 3 g ts co 9 7) [natio al vs. n 9 d in c ted that 1 cally significa tistically signi ( studies differed fr s significant o y cautione h Program on sta note e ult s re atisti n s indica some a v Participa we s st ple, nation s not {females, 11- {except sub s e t impact ers are D Fraker (198 n when res exam articipant r ly ifica research. A ings from n but wa n and wa fference i y. Read s. d on y of kets}. ctio ctio wa , find d } c s d ] ] n l l study (fo , the d ente l] l] No sign and nonp r a a s e n n v tional School Lunc o o {in bra i i 1 nsume nal] t t a a ted in that co } e of the n n are pre ups note [ [ 1 nsideratio ; howe [nationa [nationa rticipants ) ) ts 2 s ) ) ro es bgroup cture of the bod 3 3 co n 3 3 g s scop r 9 9 was in same dire was in same dire 7) [natio 9 9 bias-adjusted model more/same aly - 1 1 ( ( een pa nalyse (200 (200 sive pi y y n n 91) [1 school] e e n n Participa a a e betw v v c prehen ate, and the selection {except sub {females, 15- bgroup an ign and othe impact of the Na e e ended lunch ended lunch st studies. s u m lts would be interpre v v Fraker (198 Gleaso D D Gleaso Ho (19 s e identifies the su e subgroup a ) com on d nge a stro 1993 y’s resu earch d s ll entry also et al., the publicati y ] f providing een NSLP and een NSLP and l l] l] l] l] observed in all ll findings are fro es in re a ct s from the ne c examined th s n single stud a a t o Continued i 2 t ct. A name, a s rn wa n tha finding ction-bias-adjusted differen [ gh no ome} differen s [10 schools] [10 schools] [10 schools] s [nationa [nationa [nationa [nationa ulation, the ce ) ) ) ) ) ) ) ) ant imp 3 1 1 1 1 3 1 1 9 cant effe 9 author’ e, sele r SNDA-I (Dev 1 Difference betw Difference betw w -inc d in the interest o en thou ( phasize v This patte studie (200 (200 (200 (200 udy pop y of e (200 e (200 e (200 Signific n n n n signifi 91) [1 school] 91) [1 school] st e s s. senio ill ill ill n porte ct, e ult a s. Because of v ck lunch. ck lunch. {6-10; lo e re Participants consumed more e tically erall sampl r res , s sa sa v Gleaso Gleaso Gleaso Ho (19 Rainv Rainv Gleaso Rainv D Ho (19 y intakes at lunch— vs. vs. ability sults a P P derlying effe nonparticipant han the entire re cular result d a stati e n un Findings from rs r NSL r NSL her t a report Cell entries show the with parti s ents’ dietar d sug esterol um Results fo Results fo In main analysis for o Notes: Nonsignificant To maintain read 1 2 3 Table 27— stud Outcome Other dietary components Adde Chol Fiber Sodi subgroup rat indicate a true studies methodological limitations and em analyse magnesium tha

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 189 Chapter 5: National School Lunch Program results. This may explain the significant differences in and saturated fat, on average, than lunches offered in energy intake reported by other researchers whose secondary schools (the statistical significance of these analyses did not account for selection bias. differences was not tested). In addition, the schools included in Rainville’s study, and the lunches they The available studies are largely consistent in finding offered, may not be representative of lunches offered that NSLP participants consumed significantly more pro- nationwide. The two volunteer school districts that par- tein at lunch than nonparticipants. SNDA-I did not find ticipated in Rainville’s study were relatively affluent— this effect in the overall analysis. However, subgroup with 18 percent and 25 percent of students approved analyses revealed a significant NSLP impact among 6- for free and reduced-price meals, respectively—and to 10-year-olds and among low-income children. the reimbursable lunches offered in these districts provided even less fat (29.4 percent of total food energy) SNDA-I and Gleason and Suitor (2003) both found than required under the SMI (no more than 30 percent). that lunches consumed by NSLP participants prior to By comparison, SNDA-II found that lunches offered in the SMI were significantly lower in carbohydrates, as elementary schools provided an average of 33.5 per- a percentage of total food energy, than lunches con- cent of total food energy from fat. sumed by nonparticipants. SNDA-I subgroup analyses revealed that this pattern did not hold for females ages Vitamins and Minerals 11-18. Gleason and Suitor (2003) found that the differ- Data from SNDA-I (Devaney et al., 1993) and ence in carbohydrate consumption was due to Gleason and Suitor (2003) suggest that prior to imple- decreased consumption of added sugars among NSLP mentation of the SMI, NSLP participants had signifi- participants. Consumption of other forms of carbohy- cantly greater lunch intakes of vitamin B , riboflavin, drates was essentially equivalent for the two groups. 12 calcium, magnesium, phosphorus, and zinc than non- Findings from SNDA-I and Gleason and Suitor (2003) participants. Subgroup analyses conducted by Devaney are also consistent with regard to intakes of total fat and and her colleagues revealed substantial variation in saturated fat. The data indicate that, prior to implementa- these results across subgroups. Most often, significant tion of the SMI, lunches consumed by NSLP participants differences were concentrated among 6- to 10-year- provided significantly more fat and saturated fat, as a olds and low-income students. percentage of total energy intake, than lunches con- Findings from SNDA-I and Gleason and Suitor differ for sumed by nonparticipants. SNDA-I subgroup analyses vitamins A and C. SNDA-I found that NSLP participants revealed that the difference in intake of total fat was consumed significantly more vitamin A and signifi- concentrated among 6- to 10-year-old and non-low- cantly less vitamin C at lunch than nonparticipants.86 income students. Gleason and Suitor observed comparable trends in Findings from Rainville (2001), the only study in this intake, but found that differences between NSLP par- group based on data collected after implementation of ticipants and nonparticipants were not statistically sig- the SMI, paint a notably different picture. Rainville nificant. As noted above, results from Gleason and found no significant differences in the mean carbohy- Suitor are considered stronger because of the improved drates, fat, and saturated fat content of NSLP lunches approach to selection-bias adjustment in their analysis. and sack lunches consumed by elementary school stu- For all other vitamins and minerals, neither SNDA-I dents. These results suggest that the carbohydrate con- nor Gleason and Suitor (2003) found significant differ- tent of NSLP lunches has increased since the imple- ences between lunch intakes of NSLP participants and mentation of the SMI, while the fat and saturated fat nonparticipants. It is likely that the significant effects content has decreased. This is consistent with the trend reported in other studies are at least partially attributa- observed in SNDA-II (Fox et al., 2001). ble to selection bias. Both SNDA-I researchers and However, Rainville’s analysis did not adjust for selection Gleason and Suitor (2003) found significant effects for bias, and factors other than selection bias may have con- thiamin, vitamin B6, folate, and iron in regression tributed to their more positive findings. For example, models that did not adjust for selection bias (data not Rainville included only 2nd, 3rd, and 4th graders, while both SNDA-I and Gleason and Suitor (2003) included 86Although participants consumed significantly less vitamin C at lunch students in grades 1 through 12. SNDA-II found that than nonparticipants, intakes of both groups far exceeded the one-third lunches offered in elementary schools were lower in fat RDA standard defined for NSLP meals.

190 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program shown); however, these effects disappeared in models NSLP participants consumed significantly more cho- that controlled for selection bias. lesterol than nonparticipants.89

Every study that examined intakes of riboflavin, calci- Gleason and Suitor (2003) also studied fiber intake. um, and phosphorus found that NSLP participants con- They found that lunches consumed by NSLP participants sumed significantly larger amounts of these nutrients contributed significantly more fiber than those con- at lunch than nonparticipants (although SNDA-I found sumed by nonparticipants. that results varied for some subgroups of children). It is generally accepted that this pattern is due to Impacts on Total Daily Intake of increased consumption of milk, a concentrated source Food Energy and Nutrients of all these nutrients, by NSLP students (Lin and To have a meaningful influence on students’ nutrition Ralston; 2003, Devaney et al., 1993; Radzikowski and or health status, NSLP impacts on dietary intake must Gale, 1984). (Impacts on food consumption patterns be sustained over the course of the day. It is possible are discussed in more detail in a subsequent section.) for effects on lunch intakes to be offset by other meals and snacks consumed throughout the day. Therefore, Moreover, analyses completed by both SNDA-I and for a more complete appreciation of how the NSLP NESNP (Wellisch et al., 1983) researchers suggested affects students’ dietary intake, it is important to exam- that differences in the vitamin and mineral intakes of ine the program’s effect on the total diet. In the avail- NSLP participants and nonparticipants at lunch are due able literature, this was generally assessed as impacts to the types of food consumed, rather than the quanti- on 24-hour intake. ties. Both SNDA-I and NESNP examined the nutrient density of lunches and found it to be higher in lunches Fourteen studies examined the impact of NSLP partici- eaten by NSLP participants than those eaten by non- 87 pation on 24-hour intake of food energy and nutrients. participants. Although only the NESNP results were (Seven of these studies also assessed lunch intake and tested for statistical significance, both groups of inves- were included in the preceding section.) Findings are tigators concluded that the NSLP increased intakes of summarized in table 28 and discussed below. All the selected nutrients by providing lunches that were studies that assessed impacts on 24-hour intake were more dense in those nutrients, rather than by provid- completed before implementation of the SMI. In addi- ing more food. tion, SNDA-I did not include subgroup analyses for the 24-hour intake data. Consequently, little is known Results of the SNDA-II study, which found that reduc- about differential impacts on 24-hour intake for vari- tions in the fat and saturated fat content of NSLP meals ous subgroups of students. A few of the early NSLP were achieved without reducing vitamin and mineral studies did look at impacts among selected subgroups content, suggest that impacts on intake of key vitamins (Akin, 1983a and b; Hoagland, 1980; U.S. Department and minerals are likely to persist in post-SMI meals. In of Health Education and Welfare, 1972; Emmons et fact, SNDA-II found that the average vitamin and min- al., 1972). However, findings from these studies are eral content of the lower fat lunches offered in the quite dated, most of the studies used simple bivariate 1998-99 school year was significantly greater than the comparisons, and none attempted to control for selec- vitamin and mineral content of higher fat lunches tion bias. offered in 1992-93 (SNDA-I) (Fox et al., 2001).88 Food Energy and Macronutrients Other Dietary Components With emphasis given to findings from SNDA-I Both SNDA-I (Devaney et al., 1993) and Gleason and (Devaney et al., 1993) and Gleason and Suitor (2003), Suitor (2003) found that pre-SMI lunches consumed the available data indicate that before implementation by NSLP participants and nonparticipants provided of the SMI, NSLP participants and nonparticipants comparable amounts of cholesterol and sodium. consumed similar amounts of food energy and protein However, the SNDA-I subgroup analysis revealed that over 24 hours. among 6- to10-year-olds and low-income students, SNDA-I and Gleason and Suitor (2003) both found that, in comparison with nonparticipants, 24-hour 87Nutrient density measures nutrient intake relative to energy intake: % RDA for nutrient ‘X’ % RDA for energy. 88SNDA-II looked only at vitamin A, vitamin C, calcium, and iron—the 89Overall mean cholesterol intakes at lunch were less than one-third of nutrients that are specifically addressed in SMI standards. the recommended daily maximum of 300 milligrams (NRC, 1989b).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 191 Chapter 5: National School Lunch Program } ours d ] less l l] l] d a ct n a 24 h o Continued— i nal] nal] t a n nsume ups note [ } [nationa [nationa ) 0 ) ) ro ant imp 3 1 3 g 9 ts co es over 7) [natio 7) [natio 9 n 1 ( (200 (200 y Signific n n e n a y intak v Participa {females, 5-1 {except sub e Fraker (198 Gleaso Gleaso Fraker (198 D dietar nts’ } 1 ] ] less l l d ] l d a a a n n n l] o o i i o t t i nal] nal] nal] t a a a n n nsume [ [ n ups note [ week} , 15-21} ) ) ) 0 0 State] 0 ro 3 8 8 g 21} ts co 9 9 9 7) [natio 7) [natio 7) [natio n 9 1 1 0) [1 schoo ( ( 1 ( 78) [1 d 98 d y h Program on stude n n e a a l l n g g a a a v Participa {2-3 times per {except sub {females, 5-1 {males, 15- o o e t impact H Howe (1 Price (19 D Fraker (198 Fraker (198 H Fraker (198 n ifica d 1 ] ] } l l ] l d l] a a No sign a n n n l] o o i i tional School Lunc o ate} ate} t t i nsume nal] nal] nal] nal] t a a a n n co } [ [ n equ equ up note [ 1 ) ) e Na [nationa ) ts 2 0 0 ) ro } 3 n 8 8 3 g 4 21} 9 ly ad ly ad 9 9 7) [natio 7) [natio 7) [natio 7) [natio 9 1 1 0) [1 schoo more/same ( ( 1 ( nal nal d 98 d (200 y n n n e a a l l n Participa g g a a a v {females, 15- {except sub {nutritio {males, 5-1 {males, 15- {nutritio o o e Fraker (198 H H Fraker (198 Emmons (1972) [1 district] Gleaso Fraker (198 Fraker (198 D Howe (1 Emmons (1972) [1 district] e impact of th } } } d ] ] l l d d l] l] l] l] l] l] l] l] l] l] a a ct examined th n n a t l] o o i i a nal] nal] t t a a } n n tha ups note up note up note [ [ 4 s [nationa [nationa [nationa [nationa [nationa [nationa [nationa ation [nationa ) ) 1 0 States] 0 States] State] ) ) ) ) ) ) ) ) [nationa ) [nationa ro ro ro ant imp 3 3 1 1 1 3 3 1 3 g g g 9 9 7) [natio 7) [natio 9 9 983 983 1 1 ( ( 78) [1 studie (200 (200 (200 (200 (200 (200 (200 83a) [n 83a, b) y y Signific n n n n n n n e e (1972) [1 (1972) [1 n n a a llisch (1 llisch (1 v v e e {females, 11- {except sub {except sub {except sub Participants consumed more e e Gleaso Gleaso Fraker (198 W Akin (19 Emmons (1972) [1 district] HEW Gleaso Fraker (198 W Akin (19 Gleaso Price (19 Gleaso D Emmons (1972) [1 district] D HEW Gleaso Gleaso of table. Findings from te at end a ergy energy and macronutrients See notes Table 28— Outcome Food Food en Protein Carbohydr Fat Saturated fat

192 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program ] less l d a ct n a o Continued— ate} i t a n equ nsume [ ) ant imp 3 9 ly ad ts co 9 n 1 ( nal y Signific e n a v Participa {nutritio e D Emmons (1972) [1 district] ] less } l ] ] ] l l l d d l] l] a l] a a a n n n n l] l] l] o i o o o ate} t i i i t t t a a a a n nsume [ n n n equ up note [ [ [ ) [nationa [nationa ) ) ) ) 0 ) ) [nationa ro 3 3 3 8 1 1 g ts co 9 9 9 ly ad 9 n 9 9 9 1 983 0) [1 schoo 0) [1 schoo 0) [1 schoo ( 1 1 1 ( ( ( nal 98 98 98 d (200 (200 y y y h Program on n n n e e e a l n n n g a a a llisch (1 a v v v Participa e {nutritio {except sub o e e e t impact H Howe (1 Gleaso Emmons (1972) [1 district] W Howe (1 D D D Howe (1 Gleaso Emmons (1972) [1 district] n ifica d } ] } l d ] } l d l] l] l] l] l] l] l] l] l] a No sign a n n o l] l] l] l] i tional School Lunc o ate; t i a a a a nsume t a a n co [ n equ ups note up note [ ) w income [nationa [nationa [nationa [nationa [nationa [nationa [nationa [nationa [nationa ation ation ation ation ) ts ) 0 ) ) ) ) ) ) ) ) ro ro 3 n 8 1 1 1 3 3 3 3 3 3 g g 9 ly ad 9 9 1 more/same ( 1 ( nal d (200 (200 (200 (200 (200 (200 (200 (200 (200 years} years} 83b) [n 83a) [n 83a) [n 83b) [n ears; lo y come} n n n n n n n n n n 8 8 e n a l n Participa g a a v low-i {12-1 {12-1 {except sub {except sub {nutritio {6-11y impact of the Na o e d Akin (19 Gleaso Gleaso Gleaso Akin (19 Gleaso Gleaso Gleaso Gleaso Gleaso Akin (19 Emmons (1972) [1 district] Emmons (1972) [1 district] H D Gleaso Akin (19 e e } ontinu C } 1 ] l d come l] l] l] l] a l] l] l] l] n ct n rs— examined th a o t l] l] l] l] l] l] i t a a a a a a a n [ tha edy; edy} } n-low i ups note ) [nationa s e [nationa [nationa [nationa ation ation ation ation ation ation [nationa ) 0 States] 0 States] 0 States] 0 State] ) ) ) ) [nationa ) [nationa ) [nationa ro ant imp 3 m 8 1 3 1 g 9 o ly ne ly ne 9 c 1 983 983 983 ( n ars; no ars} ars} i nal nal 78) [1 - studie d (200 (200 (200 83a) [n 83a) [n 83b) [n 83b) [n 83a) [n 83a) [n 83a, b) over 24 hou y (19 w Signific n n n n s e o a l l (1972) [1 (1972) [1 (1972) [1 e t g llisch (1 llisch (1 llisch (1 o a e e e {except sub {6-11 ye {6-11 ye {6-11 ye {nutritio {nutritio n Participants consumed more o W Akin (19 H W Akin (19 Akin (19 Gleaso Akin (19 Gleaso Akin (19 Emmons (1972) [1 district] HEW Akin (19 Emmons (1972) [1 district] HEW W Gleaso Akin (19 Price (19 HEW Devan y intak of table. Findings from at end 6 12 ents’ dietar n See notes Table 28— stud Outcome Vitamins Vitamin A Vitamin B Vitamin B Vitamin C Vitamin E Folate Niaci

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 193 Chapter 5: National School Lunch Program less d ct a Continued— nsume week} State] ant imp ts co n 78) [1 Signific Participa {2-3 times per Price (19 ] ] less l l ] ] l l d a a a a n n n n o o i i o o ate} ate} ate} t t i i t t a a a a n n nsume [ [ n n equ equ equ [ [ ) ) ) ) 0 0 3 3 8 8 ts co 9 9 ly ad ly ad ly ad 9 9 n 9 9 1 1 ( ( 1 1 ( ( nal nal nal d d y y h Program on n n e e a a l l n n g g a a a a v v Participa {nutritio {nutritio {nutritio o o e e t impact H Emmons (1972) [1 district] H Emmons (1972) [1 district] D D Emmons (1972) [1 district] n ifica d } } ] ] l l d d ] ] ] l l l l] l] l] a a l] l] No sign a a a n n n n n l] l] l] o o l] i i tional School Lunc o o o ate} t t i i i a nsume t t t a a a a a n n co [ [ n n n equ ups note ups note [ [ [ ) ) [nationa [nationa [nationa ation ) ) ) ts ) 0 0 ) ) ) [nationa ) [nationa ro ro 3 3 3 } n 8 8 1 3 3 g g 9 9 9 ly ad 9 9 9 9 9 1 1 983 983 0) [1 schoo 0) [1 schoo 0) [1 schoo more/same ( ( 1 1 1 ( ( ( nal 98 98 98 d d (200 (200 (200 years} 83a) [n y y y ncome n n n n n 8 e e e a a l l n n n Participa g g a a a llisch (1 llisch (1 a a v v v e e {low-i {except sub {12-1 {nutritio {except sub impact of the Na o o e e e d H H Gleaso Gleaso W Howe (1 D Howe (1 D Emmons (1972) [1 district] Emmons (1972) [1 district] Gleaso W Akin (19 Howe (1 Emmons (1972) [1 district] D Emmons (1972) [1 district] e e ontinu C } ] d l l] l] l] l] l] l] l] l] l] l] ct a rs— examined th l] a n t l] l] l] o a a a i t a tha edy; edy} edy} } eek} n up note [ s e [nationa [nationa [nationa [nationa [nationa [nationa ation ation ation [nationa 0 States] 0 States] 0 States] 0 States] State] State] State] ) ) ) ) ) ) ) ) [nationa ) [nationa ) [nationa ro ant imp m 3 3 3 1 1 1 1 g o ly ne ly ne ly ne 0 c 0 983 983 983 0) [1 schoo n ars} 2 i nal nal nal 78) [1 78) [1 78) [1 ( - studie 98 (200 (200 (200 (200 (200 (200 83a) [n 83a) [n 83a) [n 83a, b) over 24 hou w Signific n n n n n n n s o o l (1972) [1 (1972) [1 (1972) [1 (1972) [1 e s t a llisch (1 llisch (1 llisch (1 o e e e e {0-1 time per w {nutritio n {6-11 ye {except sub {nutritio {nutritio l Participants consumed more Gleaso Gleaso G Gleaso Price (19 Gleaso Gleaso Price (19 Emmons (1972) [1 district] HEW W Akin (19 W Akin (19 Emmons (1972) [1 district] Emmons (1972) [1 district] HEW HEW Gleaso W Akin (19 Howe (1 Price (19 Emmons (1972) [1 district] HEW Akin (19 y intak of table. Findings from at end vin m a esium ents’ dietar u i c l See notes a Table 28— stud Outcome Ribofl Thiamin Minerals C Iron Magn

194 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program d r y sse less -hou ses l] l] sse d s ct a y a ific c s and 24 nsume spe . Stud [nationa [nationa ext discu ke s adding up all the a ) ) ant imp a e 3 1 c ts co n The t e int significant findings ma m (200 (200 ting,” or Signific n n differen tain only to Participa per -hour han others. Gleaso Gleaso 4 es in at-ho c attern of non “vote coun an 2 tice of istent p differen ideration t c s s con } Where findings less d a d or smaller th re con oked at oid the pra rent v rs lo nal] schools). nsume 3 ups note hapter 1, diffe merit mo d to a ro g ts co 7) [natio al vs. n d in C week. studies gram. Autho cautione h Program on er note s h pro some c Participa ple, nation {except sub s were either not t impact ers are Fraker (198 n exam 4-5 times p r ifica research. A P ings from of free lun e intake y. Read y of kets}. wa -21} , find d c 1 5 ction s ] ] l l ] ] ] ] t-hom n l l l l study (fo l] l] l] a a No sign a a a a n n n n n n o o i i tional School Lunc o o o o {in bra t t i i i i nsume nal] t t t t a a a a a a n n ted in that co [ [ e of the n n n n rticipated in NSL del. [ [ [ [ nsideratio ) ) [nationa [nationa [nationa after introdu ) ) ) ) ts ferent but a 0 0 ) ) ) ; females, 1 bgroup cture of the bod 3 3 3 3 co n 8 8 1 3 3 scop 0 r 9 9 9 9 9 9 7) [natio 9 9 9 9 1 1 more/same ted mo re dif ( ( 1 1 1 1 ually pa e and s ( ( ( ( r s d d (200 (200 (200 sive pi y y y y n n n n n e e e e s we a a l l n n n n Participa g g a a a a s befo a a v v v v prehen n who u ate, and the .10. {males, 5-1 ign and othe impact of the Na o o e e e e st studies. s 0 ake d lts would be interpre Gleaso Fraker (198 H Gleaso D H Gleaso D D D r intake e identifies the su e d e com on d nge a hou childre - r stro o y’s resu selection-bias-adju earch d s e 24 ison of int r r ontinu ll entry also C the publicati 8) are f . f providing sed on l] l] l] l] l] l] l l] e whe significant at p es in re compa s ct s from the a rs— c examined th s single stud rted a n t l] e ba o a on i t name, al. (197 a s repo tha finding n sed gh no [ orted a differen s a s [nationa [nationa [nationa [nationa [nationa [nationa ation ulation, the ce State] ) ) ) ) ) ) ) ) [nationa ant imp 3 3 1 1 1 3 1 0 author’ 983 0 r r Price et sted/not al., 1993) ar 2 d in the interest o ere rep en thou 78) [1 ( 2) are b fo phasize v studie (200 (200 (200 (200 (200 (200 83a) [n ble 26). udy pop over 24 hou te significant are tho Signific n n n n n n n ) w st s y et o senio s porte e ct, e s (197 not sults s a llisch (1 s. Because of ane e e v l e re Participants consumed more r rted a s (see ta Gleaso Gleaso G Gleaso Gleaso Gleaso Akin (19 Price (19 W Gleaso y intak -I (De A noted, re sults a et al. (1983b difference mons et al. es repo derlying effe han the entire re cular result c subgroup un Findings from r rs her t a Cell entries show the Differen with parti horus m ents’ dietar d sug esterol u i d Significance of Notes: Nonsignificant Findings for SND Findings for Akin Unless otherwise Findings for Em 1 o Table 28— stud Outcome Phosp Zinc Other dietary components Adde Chol Fiber S subgroup rat indicate a true studies methodological limitations and em intakes. impacts in fou

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 195 Chapter 5: National School Lunch Program intakes of NSLP participants were lower in carbohy- eating occasions other than lunch. Gleason and drates and higher in total fat and saturated fat as a per- Suitor’s model was based on differences within indi- centage of total energy intake. Gleason and Suitor vidual students rather than between groups of students, (2003) also found that 24-hour intakes of NSLP partic- and the model included covariates that controlled for ipants were significantly lower in added sugars than several potentially relevant variables, including Body the intakes of nonparticipants. All these findings are Mass Index (BMI),90 frequency of exercise, hours of consistent with findings from the analysis of lunch television watched, and whether reported intakes were intakes, indicating that pre-SMI impacts on intakes of heavier or lighter than usual. carbohydrate, fat, and saturated fat persisted over the course of the day. Other Dietary Components Both SNDA-I (Devaney et al., 1993) and Gleason and Vitamins and Minerals Suitor (2003) found that NSLP participation did not Findings from SNDA-I (Devaney et al., 1993) and affect students’ 24-hour intakes of cholesterol or sodi- Gleason and Suitor (2003) are divergent for 24-hour um. This is consistent with findings from their respec- intakes of most vitamins and minerals. SNDA-I found tive analyses of lunch intakes (using results for the that most of the increases in vitamin and mineral overall SNDA-I sample). intakes observed at lunch diminished over the course of the day. In SNDA-I, the only significant NSLP In addition, Gleason and Suitor (2003) found that impacts that persisted over 24 hours were an increase NSLP participants consumed significantly more fiber in vitamin A intake and a decrease in vitamin C intake. over 24 hours than nonparticipants. This is consistent (Overall mean vitamin C intakes of both groups were with the finding from the analysis of lunch intakes and more than 250 percent of the RDA.) NSLP partici- suggests that the NSLP’s impact on fiber intake per- sists over the course of the day. pants’ 24-hour intakes of vitamin B12, calcium, phosphorus, magnesium, and zinc continued to be greater than those of nonparticipants, but the differences were Impacts on Food Consumption Patterns not statistically significant. Examining the food consumption patterns of NSLP participants and nonparticipants can prove helpful in In contrast, Gleason and Suitor (2003) found that all of understanding the effects the NSLP has on students’ the impacts on vitamin and mineral intakes observed at nutrient intake. Several researchers looked at food lunch persisted over 24 hours. Specifically, they found consumption patterns, using a number of approaches. that, relative to nonparticipants, NSLP participants had SNDA-I researchers examined the percentage of stu- significantly greater 24-hour intakes of vitamin B12, dents consuming specific foods and food groups at riboflavin, calcium, magnesium, phosphorus, and zinc. lunch (Devaney et al., 1993). Simple weighted tabula- In keeping with findings from their analysis of lunch tions were reported, without adjustment for observed intakes, Gleason and Suitor (2003) found no signifi- differences in characteristics of the two groups or for cant impact on 24-hour intakes of vitamins A or C. selection bias. In their first analysis of the CSFII data, Gleason and Suitor (2001) computed the number of As noted, findings from Gleason and Suitor (2003) are Food Guide Pyramid servings consumed by each child considered stronger than findings from SNDA-I. and compared regression-adjusted means. Their analy- Indeed, Devaney and colleagues cautioned that sis looked at both lunch and 24-hour consumption. SNDA-I’s estimates of NSLP impacts over 24 hours were less precise than their estimates of NSLP impacts Cullen et al. (2000) also looked at Food Guide Pyramid at lunch. This is true because estimates of 24-hour servings, but their analysis was limited to lunch and to impacts are influenced by differences in unmeasured fruits and vegetables. Rainville (2001) and Wolfe and characteristics and measurement error associated with Campbell (1993) compared cumulative counts of food other eating occasions, in addition to differences in items within Food Guide Pyramid groups (expressed unmeasured characteristics and measurement error as categorical variables). Finally, Melnick et al. (1998) associated with lunch. The same is true of Gleason and and Yperman and Vermeersh (1979) used index scores Suitor’s (2003) estimates of NSLP impacts over 24 to reflect 24-hour food consumption. Melnick and his hours, of course; however, the fixed-effects model esti- colleagues computed a Food Guide Pyramid score and mated by Gleason and Suitor (2003) did a better job than the SNDA-I model of controlling for unmeasured 90BMI is the accepted standard for defining overweight and obesity. characteristics that may have affected consumption at BMI is equal to [weight in kilograms] ) [height in meters]2.

196 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program a 5-A-Day score for each student and also tabulated Gleason and Suitor (2001) reported that NSLP impacts the number of servings of fats, oils, and sweets con- on consumption of milk, vegetables, and meat were sumed. Yperman and Vermeersch constructed a meas- maintained over 24 hours. However, the decreased ure similar to the Food Guide Pyramid score, using consumption of grain products at lunch noted among data from a food frequency checklist. NSLP participants did not persist over 24 hours, nor did the decreased consumption of sweetened beverages. Because none of the studies that examined impacts on food consumption controlled for selection bias, con- Melnick et al. (1998) found that NSLP participants clusions about impacts on these outcomes are more consumed fewer servings of fats, sweets, and oils over tentative than those about impacts on intake of energy 24 hours than nonparticipants. These researchers also and nutrients. Results of the available studies, summa- found that NSLP participants scored higher than non- rized in table 29, are largely consistent. Only the stud- participants on a composite measure that evaluated ies by Rainville and Cullen are based on data that were servings from the Food Guide Pyramid (5th graders), collected sometime after the implementation of the SMI. as well as on a 5-A-Day score that looked specifically at fruit and vegetable consumption. Food Consumption at Lunch The available data suggest that NSLP participants con- The study by Yperman and Vermeersch (1979), which sumed more milk and vegetables at lunch and fewer found that the NSLP had a significant negative impact sweets and snack foods than nonparticipants. Findings on students’ “dietary complexity,” stands in stark con- for other food groups are equivocal. SNDA-I found trast to the positive or neutral findings of other studies. that a significantly greater proportion of NSLP partici- This result can be heavily discounted, however, because pants than nonparticipants consumed grain products at the study is dated and methods used to collect and ana- lunch. In contrast, Gleason and Suitor (2001) found lyze food group data do not meet current standards. that, on average, NSLP participants consumed significantly fewer servings of grains at lunch than nonpartic- Impacts on Other Nutrition- and ipants. In both cases, between-group differences were Health-Related Outcomes relatively small. The literature search identified 10 studies that looked The Gleason and Suitor (2001) finding deserves more at impacts of the NSLP beyond food and nutrient weight than the SNDA-I finding because the former intake. Table 30 describes these studies, three of which analysis looked at the actual number of servings con- are also included in the preceding section on dietary sumed (rather than the percentage of children eating at intake (Wellisch et al., 1983; Hoagland et al., 1980; least one item within the food group) and adjusted for and Emmons et al., 1972). Six studies looked at chil- differences in observed characteristics of students. dren’s weight and/or height. Four studies looked at Rainville (2001) reported results similar to Gleason and biochemical measures, specifically iron status or Suitor and found that the increase in the number of grain serum cholesterol levels, and three looked at impacts items consumed by nonparticipants was attributable to a on household food expenditures. Findings from these high prevalence of sandwiches in lunches from home. studies are summarized in table 31. One study (Gretzen and Vermeersch, 1980) also looked at school Gleason and Suitor (2001) found no difference attendance and cognitive performance. That study, between NSLP participants and nonparticipants in con- which is dated and has serious limitations as a test of sumption of fruits and juices at lunch. However, most the NSLP, found no effects of participation in a free of the other studies reported that NSLP participants school lunch program on any of these measures.91 consumed more fruit and juices than nonparticipants. Weight and Height Food Consumption Over 24 Hours Few studies have looked at the relationship between Data on food consumption patterns of NSLP partici- NSLP participation and children’s weight status or lin- pants and nonparticipants over 24 hours are more lim- ear growth, and none of them offers definitive evidence. ited. (SNDA-I (Devaney et al., 1993), Rainville The NESNP (Wellisch et al., 1983) measured students’ (2001), and Cullen (2000) did not evaluate 24-hour height, weight, and triceps skinfold (a measure of body consumption.) The available data suggest that some NSLP impacts on food consumption at lunch main- 91The study analyzed 8 years of school records in an attempt to determine tained over 24 hours, while others faded. the impact of free school lunch, alone and in combination with Head Start.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 197 Chapter 5: National School Lunch Program s n r ] less l l] l] e d a ls] ls] ct n o o a o Continued— i t a n nsume [ [10 schools] [10 schools] [10 schools] [nationa [nationa ) ) ) ) ) ) ant imp 3 1 1 1 1 1 9 ption patt ts co 9 n 3) [51 scho 3) [51 scho 1 ( (200 (200 y e (200 e (200 e (200 Signific n n e ill ill ill n a lfe (199 lfe (199 v Participa o o e Rainv W Gleaso D Gleaso Rainv Rainv W nts’ food consum less ] l d a n o i t a nsume n [ ) 3 ts co 9 n 9 1 ( y h Program on stude e n a v Participa e t impact D n ifica d l] No sign ls] tional School Lunc nsume co schoo e Na [nationa ts 5 ) n 1 more/same 000) [ (200 n Participa en (2 Cull Gleaso e impact of th 1 1 ] ] ] ] l] l l l l l] l] l] a a a a ls] ls] ls] ct ls] examined th n n n n o o o a t o o o o i i i i t t t t a a a a n n n n tha [ [ [ [ schoo [nationa s [10 schools] [10 schools] [10 schools] [nationa [nationa [nationa ) ) ) ) ) ) ) ) 5 ) ) ) ant imp 3 3 3 3 3 1 1 1 1 1 1 9 9 9 9 9 9 9 9 9 3) [51 scho 3) [51 scho 3) [51 scho 1 1 1 1 ( ( ( ( 000) [ studie (200 (200 (200 y (19 y y y y e (200 e (200 e (200 Signific n n n e e e e e ill ill ill n n n n a a a a en (2 lfe (199 lfe (199 lfe (199 v v v v o o o Participants consumed more e e e e Gleaso W W W Rainv Gleaso Gleaso Rainv D D Rainv D Cull D Devan of table. ption s m ed ts, Findings from ng lk at end e 2 nsu ducts les at b eeten ages ucts See notes Table 29— Measure Lunch co Fruits and fruit juices Grain pro Meat, poultry, fish, and me substitutes Milk and mi prod Vegeta Fats, oils, and salad dressi Snack foods Sugar, swe and sw bever

198 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program less d ls] ct o a Continued— nsume ant imp 8) [25 schools] ts co n 3) [51 scho 99 Signific lfe (199 Participa o Melnick (1 W nts’ less d l] l] nsume [nationa [nationa ) ) 1 1 ts co n (200 (200 h Program on stude n n Participa t impact Gleaso Gleaso n ifica d l] No sign tional School Lunc nsume co [nationa ts ) n 1 more/same (200 n act of the Na Participa Gleaso e imp l] l] l] ls] ls] ls] ls] ct examined th o o o o a t tinued tha s [10 schools] [nationa [nationa [nationa on ) ) ) ) ant imp 1 1 1 1 C 3) [51 scho 3) [51 scho 3) [51 scho 3) [51 scho studie (200 (200 (200 e (200 Signific n n n ill lfe (199 lfe (199 lfe (199 lfe (199 o o o o Participants consumed more Gleaso W Rainv W Gleaso Gleaso W W of table. ption patterns— m Findings from u r of lk at end s e rsity/ ducts e les at ned b ils ages ucts See notes Table 29— food con Measure Food div total numb food items 24 hours Fruit and fruit juices Grain pro Meat, poultry, fish, and me substitutes Milk and mi prod Vegeta Fats, sweets, and o Snack foods Sweete bever

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 199 Chapter 5: National School Lunch Program y a and wer s ses o s ct a n

ific r c e d n spe percentage of u ext discu

ures for sod e adding up all the a based o ant imp 9) [2 schools] s e r ts scored l all difference e r e n h The t

d significant findings ma e n (197 ting,” or d Signific sed on th r o tain only to c Participa e r ( separate mea per

are ba ood diversity, p han others. pbell (1993) a Yperma u d f o attern of non “vote coun r g

cluded r e h t tice of nts’ ideration t c o istent p s s n a

s es and in Where findings con a g wer ney et al. (1993)

r Wolfe and Cam o re con e ra consumption an c eva u 1, a ve r oid the pra a v s meat schools). sults fo 3 s for D hapte in merit mo d to a s ts scored l n eetened be al vs. d in c or tomato Result rence studies . s up, and re d cautione h Program on stude at sw note s atoe h gro some Participa c ple, nation nsume t impact ers are n in ea exam r than pot looked only r ifica e research. A ings from exception of diffe rvings co y. Read se h items y of kets}. wa c , find c s les oth n b study (fo No sign son et al. of lun tional School Lunc {in bra s number of ted in that e of the p. Glea d vegeta n ts scored ount ed that, with the nsideratio n c rt an er/same e bgroup cture of the bod co scop 8) [25 schools] r on mea high 99 ruits”) sive pi mulativ s as a grou Participa u act of the Na grade} ion but repo nd rage pt e based under “f prehen ate, and the r {2 ign and othe ve m st studies. a s lts would be interpre Melnick (1 u e identifies the su e imp s com on d ased on c nge r selection bias. a con ed here stro ened be y’s resu earch d our ) are b s cord ll entry also er the publicati d sweet adjusted fo p (re f providing gh ed. or 24-h i es in re f s from the ct rt hours. c examined th ille (2001 d for Cullen (2000) single stud a t v s, an name, repo are not tinued nding s er 24 one grou tha fi v gh no differen s ent data sweet s on ulation, the ce or Rain s ant imp f C ts scored h 8) [25 schools] 8) [25 schools] n author’ al. (2001) an r ults bles as 99 99 al., 1993) d in the interest o ot pre en thou sugar, er of items phasize ta v es studie udy pop ersisted o Signific st y et senio porte umb ct, e grade} son et ed at th up. R k s. Because of Participa ane v {5 e re r Glea and vege Melnick (1 Melnick (1 umption p s s sed on n e -I (De r A ts for sults a bell (1993) did n u ba ption patterns— s derlying effe han the entire s re cular result h con sul a e c m r un e Findings from u her t m s e

). y et al. (1993) loo y s onsuming food gro looked at fruits Cell entries show the r xity Score c with parti a e rical sco m m Study Devane Notes: Nonsignificant Food group re Findings for SND Wolfe and Camp 1 2 u hildren Table 29— food con Measure S 5-A-Day Index Score Food Guid Pyramid Index Score Dietary Compl subgroup rat indicate a true studies methodological limitations and em c catego observed in lun vegetable fruit drinks/flavored drinks.

200 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program n n n n od h means ance; on Continued— gressio gressio gressio gressio of means d) d) essi te re te re te re te re of vari a a a a ison ison of ate t-tests Analysis met ar regr Multivari Multivari Analysis Multivari Line bivari Multivari (type of statistical test not reporte Compar (type of statistical test not reporte Compar d d gh re of h” riod o h at least h on ffered c grade 1 g free ation ek e -term in g rt that “chil rt that chil unc unc h in d l lun o o rticip arly throu eals o a unc e lon Measure of participation gul n receiv g study p e 8 on and health outcomes Parent rep Bega “participates” school l and re Ate school l Averag grad 3 times per we recall day Parent rep eats schoo weekly p Ate school l Took 70% or m school m durin Not reporte triti l) o (iron, before otein) 5 t vs. t vs. t vs. t vs. t vs. t vs. ts, before vs. ts, before pant, pant pant pant pant pant Design n n n n n n n n rol, pr 4 e partici partici partici partici partici partici Program on other nu Participa non Participa non Participa after (cholester cholest non Participa Participa non Participa Participa non vs. after and after Participa non d h 772) wit = ) e ,556) n ,155) 6 3 mes es 6-21 332) des 1, 2, des 1-4 des 2 an = scents o School Lunch le latio al d 42) gra gra gra n 7 n n n nts in nts in ado nts ag a ld inc -15 (n= Popu n 1 ,797) (sample siz e 1 r eho ,155) es 1-8 (n= es 1-12 (n esce esce esce dren i dren i dren/ dren i dren ages 5-12 dren and dren and d l 3 844) i h 185% of poverty (n d hous grad Chil Chil adol Chil ages 1 Chil Chil 5 (n= Chil grad and 6 (n= (n= Chil adol (n= adol C the Nation f n e n d n I pact o 1 - S m hool i hools i on rict in York City E schools i State c c esentativ SFA in e, ools ograms s N 1 pr A 1 n H pari ent nt 3 N nts in selecte ic sch nts in 2 g New w York tion pr of students from 4 n in ally re nia Data source 7 opm ubl leme - na es in 1 dist 7-88) 0-81) 0-71) 7 PSID, Child 1 7 9 199 All stude Students in 5 1 Students in 4 s India programs i Califor Students in 1 s grad rural Ne (197 New York Stat exclud (198 Baltimore, MD Devel Supp Nation All stude interve and 2 com sample 276 p (198 2) 2 3) of table. hemistries ) 198 c 94) height udies that examined the i 0 980) t 003) 1) 8 9 00 at end 1 2) d/or ( nd d n a l iah a e (197 g lfe et al. (19 llisch et al. ( a o e See notes o Table 30—S Study Weight an Jones et al. (2 W W (NESNP) Gretzen and Vermeersch (1 Emmons et al. (197 Paig Nutritional bio Kand Peterson (2 H

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 201 Chapter 5: National School Lunch Program . n n n od h ls for means cond bias e gressio gressio gressio of means inued t 6 on- d) d) te re te re te re the se a a a ison ison of ent ate mod Analysis met ritionally adequate, Multivari with selecti adjustm Compar (type of statistical test not reporte (type of statistical test not reporte Multivari Multivari Separ Blacks, Whites, Mexican- Americans. Compar eristics: nut ut the analysis of t c 4 ara re per riod o ffered NSLP member s ng a NSLP at e old llar d ekly ek eh eals o e duri tes in n the analysis, b a Measure of baseline ch s (do participation e g study p e of free school on and health outcomes—Con ely i at ed on s typical we least onc Took 70% or m Any hous particip Not reporte Valu lunch school m durin Current we participation month) triti separ s ba ered group id s sub before , con n) were . t vs. ts vs. t vs. t vs. ts, before vs. pants pant pant pant Design n n n n n ricts our different tures r f partici partici partici partici Program on other nu pendi after (iron) Participa non Participa Participa non and after (iro Participa non Participa non rted fo e two dist Lunch es es red. Th ce. d SBP in ex ) e n Results repo rman des 1, 2, des 1-4 re offe latio al School 78) 56) 42) gra gra ,7 ,5 7 n n st we . nts in grad nts in grad and 5 6 y fa plement. Popu (sample siz uded NSLP an h program. l demic perfo c esce esce eak dren i dren i dren ages 8-12 dren and dren r 844) 992) a n income. - c inc s. s. 1-12 (n= Chil Chil adol 1-12 (n= and 6 (n= (n= Chil (n= Chil adol Chil ee lu the Nation ure; r f s and a d free b ination Surve d e m n nd not low evelopment Sup n pact o 1 ms. 2-73) ture mea unch an ndance, h), a ction of a f c m ra hools i and lunch meal rict in P State e c t gions i esentativ ools hools/ r s N pendi Nutrition Program e pr prog x h free l d (197 kfa e hool atte le 3 on State; s ea ic sch nts in 2 selecte r w York of students in a and Nutrition Exa namics, Child D ally re y Data source School ubl D es in 1 dist 0-81) 0-71) 0-81 NES her lunch shingt week where bot and after introdu gible for free lun a i ons of b res Students in sc All stude Students in 4 s grad rural Ne (197 Baltimore, MD W Blacks and Mexican- Americans w Nation districts in 8 re oversamp sample 276 p (198 198 come u ame s s before come (el not contributi al Evaluation of cond district 2) ure 3) ined physical fitness, sc iate NLSP and ot a se d intake m 198 (1976) y, low-in parate udies that examined the i a od expendit t x s: o se pare ce 2) r 1) ipation meas NESNP = Nation NHANES-I = First National Health PSID = Panel Study of In e (197 g (199 llisch et al. ( st and Price Data sou Study also e Study included Study com Did not different Partic e e 1 2 3 4 5 6 Table 30—S Study Emmons et al. (197 Paig Household f Lon W (NESNP) W district did not nutritionally need

202 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program s 6 1,2 e ] l ct m a ls] 3 a n o wer Continued— o i nal} t a ts lo n females} [ n e ) ant imp natio 3 th outc l 8 9 2) [4 schoo 4) [1 State] 1 secur ( 003) [

n Signific h Participa c s e (197 i l l lfe (199 e o {food-i {grade 1} W Paig W Jones (2 ion and hea t tri nd 6 ls] ls] wer nal} ts lo n males} males a e 6} e 6} [4 schools] e natio 0) re fe 2) [4 schoo 2) [4 schoo u secur 003) [ 2, grad 1, grad n Participa h Program on other nu e e e (197 e (197 {males} {grad {grad food-sec {food-i t impact Paig Gretzen (198 Paig Jones (2 n ifica e 1 ] l No sign l] a /sam ls] ls] 7 6 n tional School Lunc o i nal} t gher les} a edy; a n [ e Na

[4 schools] [4 schools] [4 schools] ) ) [nationa natio ts hi 3 0) 0) 0) n re m ly ne 8 u 983 9 2) [4 schoo 2) [4 schoo 4) [1 State] me} 1 nal ( o 003) [

h c s e (197 e (197 i Participa l llisch (1 lfe (199 l e o e {females} {grade 1} {grade 2} {food-sec low inc {nutritio W Gretzen (198 Emmons (1972) [1 district] Paig Paig Gretzen (198 Jones (2 W Emmons (1972) [2 districts] Gretzen (198 W e impact of th l] l] ct examined th a t gher

} tha ts hi r

n e ) [nationa ) [nationa ) ant imp d l 3 o 8 { 9 983 983

4) [1 State] 4) [1 State] ] l 1 ( a studie

5 n Signific h Participa o c i t s i l a l lfe (199 llisch (1 llisch (1 lfe (199 n e e e o o [ {older} {older} W W W W W

of table. t 4 x ight h e e g i Findings from e at end dy fat h d h

r ht/ ight o f ig

t t bility of bility of h h ht g g erwe i i e e See notes Table 31— Outcome Weight an Heig Proba und W W Body Mass Ind Percent bo Proba overwe overfatness

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 203 Chapter 5: National School Lunch Program y ses s ct a wer ific c ts lo spe n ext discu adding up all the a ) ) ant imp 1 1 0 0 0 0 The t 2 2 significant findings ma ( (

ting,” or Signific Participa h h a a i i tain only to d d n n per a a han others. K K attern of non “vote coun found. tice of ideration t c istent p s s

Where findings es were con ] ] c l l a a re con n n ren ls] 1, a wer o o r i i oid the pra t t v a a schools). n n ts lo [ [ 6

n hapte ) ) merit mo d to a 0 0 8 8 9 9 al vs. d in c 1 1 2) [4 schoo ( (

studies d d Participa cautione h Program on other n n note No significant diffe s a a l l e (197 g g e. some c a a ple, nation o o t impact ers are H H Paig n exam r ifica research. A ings from erforman y. Read y of e kets}. 8 8 wa , find ] c

s ] s l t n study (fo No sign a c i ademic p n /sam r weight.” ls] t o i tional School Lunc s {in bra t i a d

gher n over ted in that [ e of the 2

[ nsideratio )

, and ac ntile in NHANES I and II. (Wolfe et al., 1994). ) 0 ts hi bgroup cture of the bod 2 co sk of 8 scop rce n r 7 nce 9 9 1 2) [4 schoo 1 ( area

(

d sive pi r. s n n a l o pact of the Na e (197 Participa g m a ition for “at ri all). prehen ate, and the ign and othe . ea > 90th pe or arm fat o younge m r st studies. g BMI. s ) lts would be interpre Emmons (1972) [2 districts] Paig H E e identifies the su sm school attenda e im in nd com defin s on d nge a centile too s stro y’s resu per earch d CDS’s s et al., 1983 cal fitness, ll entry also arts. the publicati (sample . f providing ales age 10 a ile and arm fat a NCHS t hysi ch ot significant u th es in re p s from the ct inued c examined th single stud rted a t t gher centile name, owth r nding s sults n repo percen tha Con th ts hi fi r and fem per gh no differen s s n ulation, the ce not ant imp skinfold (Wellisch CDC g author’ s r n sted/ p younge d in the interest o NCHS also examined en thou BMI >90 e th phasize v rcentile; re studie ) : udy pop te c s nd Signific Participa st senio pe d triceps fatfold > 75 porte th ct, e not ure s s (1980 s. Because of s of tri e re r percentile o th

85 l = t for height <25 t for age an sults a two mea ales age 11 a urement hemistries o th at area <10 difference r s derlying effe health outcomes— c han the entire l re cular result w e o d t un r s Findings from e e t her t l n nd Vermeersch i d using s o b e l ite gro h Cell entries show the o l c o sse with parti

g h zen a on status c o m u L m r Included only m Based on weigh Based on arm f Based on mea Based on weigh Based on BMI > Asse Significance of Notes: Nonsignificant Gret 1 2 3 4 5 6 7 8 e e D Table 31— nutrition an Outcome Nutritional bio H Hematocrit Compos and ir variable S L subgroup rat indicate a true studies methodological limitations and em

204 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program fat). Participation was defined on the basis of chil- (CDC, 2003). As used by Jones et al. (2003), it also dren’s average weekly participation from first grade includes children who are considered to be overweight through the current school year. The analysis, which (BMI at or above the 95th percentile) (CDC, 2003). did not control for selection bias, found that older par- Weights were reported by primary caregivers and ticipants weighed more and had greater mean triceps heights were measured by field interviewers. The skinfold measurements than comparably aged nonpar- authors indicate that approximately 86 percent of the ticipants. These findings are difficult to interpret, how- children had been weighed within the preceding month ever, because the authors did not provide information and that 16 percent of caregivers had to estimate on whether program participants were closer to age- weight because they had no recent reference point. standardized norms than nonparticipants (Rush, 1984). Thus, it is not clear whether the findings suggest a The analysis assessed the likelihood of being at risk of health benefit or risk. If, for example, children tended becoming overweight among children living in food- to be underweight for their age or stature, greater secure and food-insecure households, while controlling weight and fatness among participants could be con- for participation in a number of FANPs and other rele- sidered a benefit. On the other hand, if children tended vant characteristics. Results showed that NSLP partici- to be overweight for age or stature, these findings pation did not affect the likelihood that males would would not be considered beneficial. be at risk of becoming overweight, regardless of whether they lived in food-secure or food-insecure A more recent study by Wolfe et al. (1994) obtained households. The likelihood of being at risk of becom- similar results and reported them in a more easily inter- ing overweight was also unaffected by NSLP partici- preted manner. The authors assessed the prevalence of pation status among females in food-secure house- overweight in elementary school children in New York holds. Among females in food-insecure households, State, using BMI and measures of triceps skinfold and however, those who participated in the NSLP had 71- arm fat area. Data were compared with national refer- percent reduced odds of being at risk of becoming ence data for 1974 and 1980. NSLP participants were overweight, compared with those who did not partici- defined on the basis of a parent report that the child “eats pate. The authors offer no explanation for the apparent school lunch.” The authors concluded that overweight protective effect of NSLP participation among food- was a problem among elementary school students in insecure females and suggest that more research is New York State, and that students who ate the school needed to understand the relationship between income, lunch tended to be fatter than those who did not. food security, FANP participation, and weight status.

Wolfe and her colleagues made no attempt to assess or It is doubtful that cross-sectional studies can adequately control for selection bias, a critical consideration in address questions about program impacts on children’s estimating the impact of a feeding program on weight weight and height. Indeed, researchers who attempted status. Thus, these results indicate that NSLP partici- to assess the impact of the WIC program on these out- pants in New York State were more overweight than comes concluded that a longitudinal study with serial nonparticipants. They do not, however, indicate that measurements was essential (Puma et al., 1991). these differences are the result of NSLP participation. It is possible, for example, that overweight children Nutritional Biochemistries chose to participate in the NSLP more often than Four of the studies identified in the literature search nonoverweight children. examined impacts of the NSLP on nutritional biochemistries. Researchers examined measures of iron A recent study by Jones et al. (2003) looked at the rela- status (hemoglobin and/or hematocrit) and/or choles- tionship between food security, participation in FANPs, terol levels. Only Hoagland (1980) used a national including the NSLP, and the risk of overweight. The dataset in assessing these outcomes. The three smaller, authors used data from the 1997 Panel Study of local studies used the “participants, before vs. after” Income Dynamics (PSID) Child Development design (essentially a longitudinal design with a single Supplement to examine the risk of overweight among followup measurement), which is preferable to the children ages 5-12 in low-income households (income “participant vs. nonparticipant” design for assessing <185 percent of poverty). Risk of overweight was impacts on biological variables. With the exception of defined as BMI at or above the 85th percentile. This the recent study by Kandiah and Peterson (2001), stud- cutoff is routinely used to identify children who are ies are based on data from the 1970s. None of the considered to be “at risk” of becoming overweight studies reported a significant NSLP effect.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 205 Chapter 5: National School Lunch Program

Iron will be an addition to total household food expenditures. However, its value may be partly offset if the Analyzing children’s hemoglobin and hematocrit val- household reduces some food expenditures because of ues from NHANES-I, Hoagland (1980) found no sig- the availability of the subsidized lunch—if, for exam- nificant differences between NSLP participants and ple, money that would have been spent to purchase nonparticipants. Working with a sample of children lunch for the student is applied to nonfood uses. from four schools in Baltimore, Paige (1972) found no effects on hematocrit levels or on a composite variable Two of the three studies that examined impacts on that reflected both growth status and iron status. household food expenditures were based on data collect- Emmons and colleagues (1972) found so few students ed for the NESNP. Wellisch and her colleagues (1983) with low iron status that they did not test the signifi- reported a dollar-for-dollar increase in the value of food cance of differences between groups. available to participating households as a result of partic- Cholesterol ipation in the NSLP. Long (1991) reanalyzed the NESNP data, using only the sample of students who Hoagland (1980) also used NHANES-I data to assess attended schools that offered the NSLP and adjusting for children’s cholesterol levels. He found no significant selection bias, and obtained somewhat different results. difference between NSLP participants and nonpartici- She found that the overall impact of NSLP participa- pants. In the same analysis, Hoagland attempted to tion was to increase household food expenditures, but look for differences in biochemical indicators of pro- she estimated that each additional NSLP benefit dollar tein-calorie malnutrition. Finding no abnormal levels reduced other food expenditures by about $0.61, for a of serum albumin in the sample, however, he conclud- net addition of $0.39 to the value of food expenditures ed that these measures were not useful for assessing on behalf of the household. Long’s results were compa- NSLP health impacts. rable to those of West and Price (1976), who evaluated the impact of free school lunches on household food Kandiah and Peterson (2001) examined total choles- expenditures for a sample of children ages 8-12 in terol and LDL cholesterol levels in a group of 30 chil- Washington State. dren and adolescents ages 11-15. The sample was limited to students who ate NSLP meals at least three West and Price (1976) and Wellisch et al. (1983) both times per week. Baseline levels were compared with found somewhat larger supplementation effects for followup levels measured 4 months later. Results Black than White households. Supplementation was also showed that both total cholesterol and LDL cholesterol somewhat greater for Hispanic households, but the effect levels decreased significantly over the 4-month period. was not statistically significant. Long’s reanalysis of the This was true for students who ate NSLP lunches as NESNP data did not include subgroup analyses. well as those who ate both breakfast and lunch. Rather than attribute these changes to a positive impact of the NSLP, the authors concluded that changes in choles- Summary terol levels over time were due to hormonal fluctua- The body of research reviewed in this chapter indi- tions associated with puberty. No information was pro- cates that, prior to implementation of the SMI, the vided on the protocol used in collecting students’ NSLP had a significant impact on the dietary intake of blood samples (for example, whether students were school-age children and adolescents. There is strong fasting when bloods were drawn and for what period evidence that the program increased children’s intakes of time) or on the reliability of the measures obtained. of selected vitamins and minerals at lunch (vitamin B , riboflavin, calcium, phosphorus, magnesium, and Household Food Expenditures 12 zinc), and the strongest available evidence suggests Assessment of household food expenditures can pro- that these effects persisted over 24 hours. Because of vide information on the extent to which receipt of limitations in the dietary assessment methodologies NSLP meal benefits increases the value of food avail- used, it is not possible to determine whether NSLP able to families.92 Potentially, the NSLP meal benefit participants were more likely than nonparticipants to have adequate intakes of these vitamins and minerals. 92Food expenditure data have also been used to evaluate the success of the NSLP in meeting its second objective: encouraging the consumption of There is also convincing evidence that, prior to the domestic agricultural products. The program is considered efficient in meeting its agricultural support goals if most of the NSLP subsidy is spent on food and SMI, NSLP participants consumed less carbohydrate little is diverted to nonfood expenditures (Radzikowski and Gale, 1984). and more fat and saturated fat (as a percentage of total

206 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program food energy) than nonparticipants, both at lunch and impacts on vitamin and mineral intakes. However, evi- over 24 hours. The strongest available evidence sug- dence from SNDA-II suggests that changes in the gests that the difference in carbohydrate intake was macronutrient profile of school lunches has been due to decreased consumption of added sugars among achieved without compromising overall vitamin and NSLP participants. mineral content. In fact, lunches offered in 1998-99 provided significantly greater amounts of vitamin A, Finally, the available evidence indicates that, prior to vitamin C, calcium, and iron than lunches offered in the SMI, NSLP participation had no significant effect 1991-92. on intake of sodium or cholesterol. NSLP participation was associated, however, with a significantly greater With the exception of impacts on household food intake of dietary fiber, both at lunch and over 24 hours. expenditures, the existing evidence is too limited to support conclusions about whether NSLP participation Evidence from the SNDA-II study demonstrates that, affects other nutrition- and health-related outcomes. even in the early stages of the SMI, schools had made There is limited, but reasonably strong, evidence that significant progress in decreasing the fat and saturated NSLP participation increases total household food fat content of school lunches and in increasing the car- expenditures. However, the available data are quite bohydrate content. Since SNDA-II data were collected dated (the most recent were collected in the early (the 1998-99 school year), efforts have continued at 1980s). the Federal, State, and local levels to make school lunches consistent with SMI standards for these nutri- Clearly, there is a critical need for an updated study of ents. In addition, there has been increased emphasis on the NSLP and its impacts on children. To provide a nutrition education for school-age children to promote comprehensive picture of how the NSLP influences acceptance and consumption of healthier school meals. children’s food and nutrient intakes, future studies will need to differentiate between the multiple sources of Consequently, the current impact of the NSLP on stu- foods and beverages available at school (reimbursable dents’ intakes of total fat, saturated fat, and carbohy- meals, a la carte purchases, vending machines, snack drates is unknown and can only be answered with new bars, etc.). research. The same can be said of the program’s

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 207 Chapter 5: National School Lunch Program

References Centers for Disease Control and Prevention. 2003. “CDC Growth Chart Training Modules.” Available: Abraham, S., M. Chattopadhyay, M. Montgomery, et al. http://www.cdc.gov/nccdphp/dnpa. Accessed May 2002. The School Meals Initiative Implementation Study: 2003. Third Year Report. USDA, Food and Nutrition Service. Cullen, K.W., J. Egan, T. Baranowski, et al. 2000. Akin, J.S., D.K. Guilkey, P.S. Haines, et al. 1983a. “Effect of a la Carte and Snack Bar Foods at School “Impact of the School Lunch Program on Nutrient on Children’s Lunchtime Intake of Fruits and Intakes of School Children,” School Food Service Vegetables,” Journal of the American Dietetic Research Review 7(1):13-18. Association 100(12):1482-86.

Akin, J.S., D.K. Guilkey, and B.M. Popkin. 1983b. Devaney, B.L., M.R. Ellwood, and J.M. Love. 1997. “The School Lunch Program and Nutrient Intake: a “Programs That Mitigate the Effects of Poverty on Switching Regression Analysis,” American Journal Children,” Future Child 7(2):88-112. of Agricultural Economics 65(3):477-85. Devaney, B.L., A.R. Gordon, and J.A. Burghardt. 1993. American School Food Service Association. 2003. The School Nutrition Dietary Assessment Study: Dietary “Legislative Update: Focus on Child Nutrition Intakes of Program Participants and Nonparticipants. Continues.” Available: http://www.asfsa.org/newsroom/ USDA, Food and Nutrition Service. sfsnews/legupdate0803.asp. Accessed August 2003. Emmons, L., M. Hayes, and D.L. Call. 1972. “A Study Baranowski, T., and B.G. Simons-Morton. 1991. of School Feeding Programs. I. Economic Elgibility “Dietary and Physical Activity Assessment in School- and Nutritional Need. II. Effects on Children with age Children: Measurement Issues,” Journal of School Different Economic and Nutritional Needs,” Journal of Health 61:195-97. the American Dietetic Association 61(9):262-75.

Baxter, S.D., W.O. Thompson, A.F. Smith, et al. 2003. Fox, M.K., M.K. Crepinsek, P. Connor, et al. 2001. “Reverse Versus Forward Order Reporting and the Second School Nutrition Dietary Assessment Study Accuracy of Fourth-graders’ Recalls of School Breakfast (SNDA-II): Final Report. USDA, Food and Nutrition and School Lunch,” Journal of the American Dietetic Service. Association 100(8):911-18. Fraker, T. 1987. Final Report: The Sodium and Macro- Baxter, S.D., W.O. Thompson, and H.C. Davis. 2000. nutrient Content of USDA School Lunches. USDA, “Prompting Methods Affect the Accuracy of Children's Food and Nutrition Service. School Lunch Recalls,” Journal of the American Dietetic Association 100(8):911-18. French, S.A., M. Story, J.A. Fulkerson, and A.F. Gerlach. 2003. “Food Environment in Secondary Burghardt, J., A. Gordon, N. Chapman, et al. 1993. The Schools: a la Carte, Vending Machines, and Food School Nutrition Dietary Assessment Study: School Food Policies and Prac-tices,” American Journal of Public Service, Meals Offered, and Dietary Intakes. USDA, Health 93(7):1161-67. Food and Nutrition Service. Gleason, P., and C. Suitor. 2003. “Eating at School: Burghardt, J.A. and B.L. Devaney (eds.) 1995. How the National School Lunch Program Affects “Background of the School Nutrition Dietary Children’s Diets,” American Journal of Agricultural Assessment Study,” American Journal of Clinical Economics 85(4):1047-61. Nutrition 61(1, Supplement):178S-181S. Gleason, P., and C. Suitor. 2001. Children’s Diets in Buzby, J.C., and J.F. Guthrie. 2002. Plate Waste in the Mid-1990’s: Dietary Intake and Its Relationship School Nutrition Programs: Final Report to Congress. with School Meal Participation. USDA, Food and E-FAN-02-009. USDA, Economic Research Service. Nutrition Service.

Buzby, J.C., J. Guthrie, and L.S. Kantor. 2003. Gleason, P.M. 1996. Student Participation in the School Evaluation of the USDA Fruit and Vegetable Pilot Nutrition Programs: An Econometric and Simulation Program: Report to Congress. E-FAN-03-006. USDA, Model. USDA, Food and Nutrition Service. Economic Research Service.

208 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 5: National School Lunch Program

Glengdahl, M.C., and C.D. Seaborn. 1999. “Comparison Kubik, M.Y., L.A. Lytle, P.J. Perry, and M. Story. of Cost and Selected Nutrients of the National School 2003. “The Association of the School Food Service Hot Lunch, a la Carte, and Combination of School Environment with Dietary Behaviors of Young Lunch/a la Carte Menus,” Journal of the American Adolescents,” American Journal of Public Health Dietetic Association 99(Supplement):A-49. Abstract of 93(7):1168-73. paper presented at annual meeting of the American Dietetic Association, October 1991. Lin, B.-H., and K. Ralston. 2003. Food Assistance Research Brief—Competitive Foods: Soft Drinks vs. Milk. Gretzen, D., and J.A. Vermeersch. 1980. “Health FANRR-34-7. USDA, Economic Research Service. Status and School Achievement of Children from Head Start and Free School Lunch Programs,” Public Health Long, S.K. 1991. “Do the School Nutrition Programs Reports 95(4):362-68. Supplement Household Food Expenditures?” Journal of Human Resources 26(4):654-78. Guthrie, J. 2003. Food Assistance Research Brief— Do Healthy School Meals Cost More? FANRR-34-6. Maurer, K. 1984. “The National Evaluation of School USDA, Economic Research Service. Nutrition Programs: Factors Affecting Student Participation,” American Journal of Clinical Nutrition Ho, C.S., R.A. Gould, L.N. Jensen, et al. 1991. 40:425-47. “Evaluation of the Nutrient Content of School, Sack and Vending Lunch of Junior High Students,” School Medlin, C., and J.D. Skinner. 1988. “Individual Food Service Research Review 15(2):85-90. Dietary Intake Methodology: a 50-year Review of Progress,” Journal of the American Dietetic Hoagland, G.W. 1980. Feeding Children: Federal Association 88:1250-57. Child Nutrition Policies in the 1980s. Washington, DC: U.S. Government Printing Office. Melnik, T.A., S.J. Rhoades, K.R. Wales, et al. 1998. “Food Consumption Patterns of Elementary School Howe, S.M., and A.G. Vaden. 1980. “Factors Children in New York City,” Journal of the American Differentiating Participants and Nonparticipants of the Dietetic Association 98(2):159-64. National School Lunch Program. I. Nutrient Intake of High School Students,” Journal of the American National Research Council. 1989. Diet and Health: Dietetic Association 76(5):451-58. Implications for Reducing Chronic Disease Risk. Washington, DC: National Academy Press. Institute of Medicine, Food and Nutrition Board. 2001. Dietary Reference Intakes: Application in Dietary Nestle, M. 2000. “Soft Drink Pouring Rights,” Public Assessment. Washington, DC: National Academy Health Reports 115:308-17. Press. Paige, D. 1972. “The School Feeding Program: An Jones, S.J., L. Jahns, B. Laraia, and B. Haughton. Underachiever,” Journal of School Health 42(7):392-95. 2003. “Lower Risk of Overweight in School-aged Food Insecure Girls who Participate in Food Perry, L.H., E.C. Shannon, K. Stitt, et al. 1984. “Student Assistance: Results from the Panel Study of Income Lunch Practices: a Comparison of Cost and Dietary Dynamics Child Development Supplement,” Archives Adequacy of School Lunch and Brown Bag Lunches,” of Pediatrics & Adolescent Medicine 157(8):780-84. School Food Service Research Review 8(2):114-18.

Kandiah, J., and C. Peterson. 2001. “National School Price, D.W., D.A. West, G.E. Scheier, et al. 1978. Breakfast and Lunch Meals: Blood Lipid Levels in “Food Delivery Programs and Other Factors Affecting Prepubescents,” FASEB Journal 15(4):A273. Abstract Nutrient Intake of Children,” American Journal of of paper presented at the annual meeting of the Agricultural Economics 60(4):609-18. Federation of American Societies for Experimental Puma, M.J., J. DiPietro, J. Rosenthal, D. Connell, et al. Biology, March 2001. 1991. Study of the Impact of WIC on the Growth and Development of Children: Field Test, Volume 1: Feas- ibility Assessment. Cambridge, MA: Abt Associates Inc.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 209 Chapter 5: National School Lunch Program

Radzikowski, J., and S. Gale. 1984. “The National U.S. Department of Agriculture, Food and Nutrition Evaluation of School Nutrition Programs: Conclusions,” Service. 2000b. Changing the Scene: Improving the American Journal of Clinical Nutrition 40:454-61. School Nutrition Environment. USDA, Food and Nutrition Service. (A tool kit for schools, available Rainville, A.J. 2001. “Nutritional Quality of through http://www.fns.usda.gov/tn/Healthy/chang- Reimbursable School Lunches Compared to Lunches ing.html). Brought from Home in Elementary Schools in Two Southeastern Michigan Districts,” Journal of Child U.S. Department of Agriculture and U.S. Department Nutrition & Management 25(1):13-18. of Health and Human Services. 1990. Nutrition and Your Health: Dietary Guidelines for Americans, 3rd Ralston, K., J. Buzby, and J. Guthrie. 2003. Food Assis- edition. Home and Garden Bulletin No. 232. tance Research Brief—A Healthy School Meal Environ- ment. FANRR-34-5. USDA, Economic Research Service. U.S. Department of Health, Education and Welfare. 1972. Ten-Nutrition Survey: 1968-1970. Highlights. Rossi, P.H. 1998. “The School Meals Programs,” in Feeding the Poor: Assessing Federal Food Aid. Wechsler, H., N., Brener, S. Kuester, and C. Miller. Washington, DC: The AEI Press, pp. 66-80. 2001. “Food Service and Foods and Beverages Available at School: Results from the School Health Rush, D. 1984. “The National Evaluation of School Policies and Programs Study 2000,” Journal of School Nutrition Programs: Editor’s Technical Notes,” Health 71(7):313-23. American Journal of Clinical Nutrition 40:462-64. Wellisch, J.B., S.D. Hanes, L.A. Jordon, et al. 1983. U.S. Department of Agriculture, Food, Nutrition, and The National Evaluation of School Nutrition Consumer Services. 2001. Foods Sold in Competition Programs: Final Report. Volumes 1 and 2. Santa with USDA School Meal Programs: A Report to Monica, CA: Systems Development Corporation. Congress. Available: http://www.fns.usda.gov/ cnd/lunch/CompetitiveFoods/report_congress.htm. West, D.A., and D.W. Price. 1976. “The Effects of Income, Assets, Food Programs, and Household Size U.S. Department of Agriculture, Food and Nutrition on Food Consumption,” American Journal of Service. 2003a. Program data. Available: http://www. Agricultural Economics 58(1):725-30. fns.usda.gov/pd. Accessed April 2003. Wolfe, W.S., and C.C. Campbell. 1993. “Food pattern, U.S. Department of Agriculture, Food and Nutrition diet quality, and related characteristics of schoolchild- Service. 2003b. “Afterschool Snacks in the NSLP: Basic ren in New York State,” Journal of the American Questions and Answers.” Available: http://www.fns.usda. Dietetic Association 93(11):1280-84. gov/cnd/Afterschool/NSLP_QA.htm. Accessed July 2003. Wolfe, W.S., C.C. Campbell, E.A. Frongillo, et al. U.S. Department of Agriculture, Food and Nutrition 1994. “Overweight Schoolchildren in New York State: Service. 2003c. “Program Fact Sheet: National School Prevalence and Characteristics,” American Journal of Lunch Program.” Available: http://www.fnsusda.gov/ Public Health 84(5):807-13. cnd/Lunch/default.htm. Accessed June 2003. Yperman, A.M., and J.A. Vermeersch. 1979. “Factors U.S. Department of Agriculture, Food and Nutrition Associated with Children’s Food Habits,” Journal of Service. 2002. “Team Nutrition Policy Statement.” Nutrition Education 11(2):72-76. Available: http://www.fns.usda.gov/tn/grants/TN_ PolicyStatement.pdf. Accessed April 2002. Zive, M.M., J.P. Elder, J.J. Prochaska, et al. 2002. “Sources of Dietary Fat in Middle Schools,” U.S. Department of Agriculture, Food and Nutrition Preventive Medicine 35(4):376-82. Service. 2000a. “Food and Nutrition Service (FNS) Strategic Plan 2000 to 2005.” Available: http://www. fns.usda.gov/oane/MENU/gpra/FNSStrategicplan.htm. Accessed April 2002.

210 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6 School Breakfast Program

The School Breakfast Program (SBP) began as a pilot number of States” on a competitive basis. The grants, program in 1966. The intent was to provide breakfast at which were targeted toward school districts that served school to children from poor areas who may not have large proportions of low-income children, were funded eaten breakfast at home, and to children in rural areas at a level of $3 million in 1990. The funds were to be who ate an early breakfast, did chores, and then arrived used to help cover nonrecurring costs associated with at school hungry after traveling long distances (Devaney initiating the SBP.94 Since 1989, the size of the SBP and Stuart, 1998). The program was modeled after the has more than doubled, increasing from 3.8 million National School Lunch Program (NSLP), which had breakfasts per day in FY 1989 to 8.1 million breakfasts been in existence for some 20 years when the SBP was per day in FY 2002 (USDA/FNS, 2003a). established. The combination of the NSLP and SBP was intended to provide “a coordinated and comprehensive The SBP operates in essentially the same manner as the child food service [program] in schools” (P.L. 89-842). NSLP (see chapter 5). The program is administered by FNS at the Federal level and by school food authorities Schools that participate in the SBP provide breakfasts (SFAs) at the local level. SFAs receive cash reimburse- to children, regardless of household income. Schools ments for each meal served (commodities are tied to the receive Federal reimbursements for each meal served, NSLP). For the 2002-03 school year, the basic subsidies with higher reimbursements for meals served free of were $1.17 for free breakfasts, $0.87 for reduced-price charge or at a reduced price to children from low- breakfasts, and $0.22 for breakfasts served to children income families. In FY 2002, more than 8 million chil- who purchased meals at the full price (referred to as dren participated in the SBP on an average school day. “paid meals”).95 Children eligible for reduced-price Approximately 1.4 billion meals were served, at a total breakfasts cannot be charged more than $0.30 per Federal cost of $1.6 billion (U.S. Department of breakfast. SFAs set their own prices for full-price/paid Agriculture (USDA), Food and Nutrition Service breakfasts, but must operate their school meal service (FNS), 2003a). program on a nonprofit basis (USDA/FNS, 2003b). Of the 1.4 billion breakfasts served in FY 2002, 83 percent were served to children who received their meals Program Overview free or at a reduced price (USDA/FNS, 2003a). The SBP was authorized by the Child Nutrition Act of 1966 (P.L. 89-842).93 Greater Federal subsidies were Program Use offered for schools identified as having a “severe need” In comparison with the NSLP, the SBP is available to as a way of encouraging participation by schools in low- fewer children and student participation rates are lower. income areas (which tended to have higher operating The SBP is offered in approximately 78 percent of the costs). Congress authorized the SBP as a permanent pro- schools and institutions that offer the NSLP (USDA/ gram in 1975. While the program continued to provide FNS, 2003b and USDA/FNS, 2003c). Using data from greater subsidies to schools in areas of severe need, the the first School Nutrition Dietary Assessment Study authorizing legislation declared that the SBP was target- (SNDA-I), Rossi (1998) found that in schools where the ed to “all schools where it is needed to provide adequate SBP was available, only 78 percent of children who nutrition for all children in attendance” (P.L. 94-105). were eligible for free or reduced-price breakfasts were certified to receive meal subsidies. And of those certi- In 1989, the Child Nutrition Act was amended with the fied, only 37 percent participated in the breakfast pro- specific intention of expanding the availability of the gram. The combined effect was that at the time the SBP in the Nation’s schools. The Secretary of Agriculture was required to award startup grants, administered through State agencies, to “a substantial 94Changes made by the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA; Public Law 104-193) eliminated this grant program. 95Reimbursement rates are higher for Hawaii and Alaska. In addition, 93Much of the text in the program overview section also appears in schools that operate in high-poverty areas may qualify for “severe-need” another report prepared by Abt Associates Inc. (McLaughlin et al., 2002). reimbursement. In the 2002-03 school year, severe-need schools could receive A preliminary draft of this chapter was used in preparing that report. up to an additional $0.23 for free and reduced-price breakfasts.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 211 Chapter 6: School Breakfast Program

SNDA-I data were collected (1991-92 school year), period, breakfasts and lunches are offered to all stu- only 29 percent of children eligible for free and dents free of charge and schools do not have to con- reduced-price breakfast were eating school breakfasts. duct free and reduced-price certifications. State agencies may grant subsequent 4-year extensions if there Findings from more recent studies are similar. The sec- has been no substantial change in the income level of a ond School Nutrition Dietary Assessment Study school’s target population (USDA/FNS, 2003d). (SNDA-II), completed in the 1998-99 school year, School districts are responsible for costs in excess of found that 22 percent of children in SBP schools par- Federal reimbursements. In the 1999-2000 school year, ticipated in the program on an average day (compared an estimated 3,154 schools (3.8 percent of all schools) with 60 percent for the NSLP) (Fox et al., 2001). used either Provision 2 or 3 (Abraham et al., 2002). Students approved for free meals participated in the SBP at a higher rate (39 percent) than students Some States require that all schools, or schools with a approved for reduced-price meals (20 percent) or stu- specific proportion of low-income students, participate dents who purchased full-price meals (8 percent). in the SBP. According to the Food Research and Action Participation was greatest in elementary schools (26 Center (FRAC), 37 of the 50 States had their own leg- percent) and lowest in high schools (11 percent). islative requirements related to the SBP in the 2002-03 school year, and/or provided funding for school break- A USDA-sponsored study found that a major factor fasts (Food Research and Action Center, 2003). affecting application and participation decisions related to Twenty-five States had laws mandating that specific the NSLP and SBP was the perceived stigma of receiving schools participate in the SBP, and 22 States provided free or reduced-price school meals (Glantz et al., 1994). some type of funding.97 Three States (Illinois, This was found to be more of an issue for the SBP and Maryland, and Massachusetts) provided State funding for secondary school students than for the NSLP and ele- for so-called “universal-free” school breakfast in cer- mentary school students. Study findings suggested that tain schools. In these schools, breakfasts are provided parents and older students believed that receiving free or free to all children regardless of household income. In reduced-price meals labeled students as poor and set addition, North Carolina provided funding for univer- them apart. While program regulations require school dis- sal-free school breakfasts for kindergarten students.98 tricts to ensure that children approved for free and reduced-price meals are not overtly identified, the percep- The idea of providing universal-free school breakfasts tion was that simply eating a school breakfast carries a became increasingly popular in the 1990s. Several States stigma, regardless of one’s income status. and school districts implemented demonstrations to test the feasibility and impact of such programs. Early Several other factors have been identified as potential results indicated that universal-free breakfasts substan- barriers to SBP participation. These include scheduling tially increased participation. Program evaluators also (when breakfast is served relative to the official start reported positive effects on tardiness, absentee rates, of the school day), meal prices, competing a la carte academic achievement, and related outcomes. offerings, bus/transportation issues, lack of time to eat, However, because most of the demonstrations were lack of space, and student preferences for other foods small, used nonexperimental designs, and had other (Reddan et al., 2002; Rosales and Jankowski, 2002; design and/or data limitations, these findings were Project Bread, 2000). considered tentative (McLaughlin et al., 2002).

Offering a free breakfast to all school children, regard- To obtain a more scientifically sound assessment of less of family income, is viewed as a promising vehi- the potential impacts of providing universal-free cle for increasing participation in the SBP. Under school breakfasts, Congress authorized the School existing Federal regulations, schools may eliminate the Breakfast Program Pilot Project (SBPP) in 1998 (P. L. burden of determining eligibility for meal benefits and 105-336). This 3-year demonstration project, adminis- provide all meals free of charge. Under Provisions 2 tered by FNS, includes a comprehensive evaluation of and 3 of the National School Lunch Act (which govern both the implementation and impact of a universal-free both the NSLP and the SBP), schools are reimbursed at established rates for a 4-year period.96 During this 97Counts are not mutually exclusive. Some States provide no funding and/or have no mandates. 96Schools may operate under Provisions 2 or 3 for one or both meal pro- 98Minnesota also provided universal-free breakfast funding from 1999 to grams. Currently, more schools operate under these provisions for the SBP 2002. However, the statute that granted the funding was repealed by the State than for the NSLP (USDA, FNS, Office of Analysis and Evaluation, 2004). legislature in 2003.

212 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program school breakfast program. The project began in the The SNDA-I study demonstrated that, prior to the 2000-01 school year and ended at the end of the 2002- SMI, breakfasts offered in the SBP were consistent 03 school year. Results from the first year of imple- with SMI nutrition standards for key nutrients, but mentation, including information on impacts on a vari- were low in energy relative to defined Recommended ety of student outcomes, were published in late 2002 Energy Allowances (REAs), high in fat and saturated (McLaughlin et al., 2002). A final report covering all 3 fat, relative to Dietary Guidelines recommendations, years of the pilot is expected in 2004. and high in sodium relative to the National Research Council’s (NRC) recommendation (Burghardt et al., Nutrition Standards 1993). Data from SNDA-II, collected in the 1998-99 To be eligible for Federal subsidies, SBP meals must school year (early in SMI implementation), showed meet defined nutrition standards. As described in improvement in the nutritional profile of SBP meals. chapter 5, USDA launched the School Meals Initiative Breakfasts offered in 1998-99 continued to meet stan- (SMI) in 1995 to improve the nutritional quality of dards for key nutrients, but were significantly lower in school meals. Prior to the SMI, schools that participat- total fat, saturated fat, and sodium than pre-SMI break- ed in the SBP were required to follow a meal pattern fasts (Fox et al., 2001). Indeed, the average nutrient that specified the types and amounts of foods and bev- profile of breakfasts offered in the1998-99 school year erages to be offered to students of different ages. The was consistent with SMI nutrition standards for both basic meal pattern, which was modeled after the NSLP total fat and saturated fat. Breakfasts offered in ele- meal pattern, includes: mentary schools were also consistent with the NRC’s recommendation for sodium, and breakfasts offered in • Milk: 1 serving per meal secondary schools all but met this standard (601 milligrams (mg) sodium, on average, compared with a • Fruit, juice, or vegetables: 1 serving per meal standard of 600 mg). On average, breakfasts continued to fall short of the benchmark for energy content.101 • Meat or meat alternate and bread or bread alternate: 2 servings total per meal.99 In the years since the SNDA-II data were collected, efforts to implement the SMI nutrition standards have Under the SMI, new nutrient-based standards were continued at the Federal, State, and local levels. established for SBP meals. SMI nutrition standards Consequently, even this relatively recent data may not specify that breakfasts must provide, over the course provide an accurate picture of the nutrient content of of a week, an average of 25 percent of students’ daily meals currently offered in the SBP.102 requirements for energy (calories) and key nutrients (calcium, iron, protein, and vitamins A and C). The existing literature on SBP impacts needs to be Breakfasts must also be consistent with the Dietary considered cautiously because program operations Guidelines for Americans recommendations for intake changed substantially after most of the available re- of fat and saturated fat.100 search was completed. The SMI and related initiatives may have affected the meals offered to students and The Healthy Meals for Healthy Americans Act (P.L. students’ consumption of those meals. In addition, the 103-448) formally required that school meals be con- concerted efforts made in recent years to increase par- sistent with the Dietary Guidelines and that schools ticipation in the SBP may have led to changes in the begin complying with SMI nutrition standards in the characteristics of the children being served by the pro- 1996-97 school year unless a waiver was granted by gram. This, in turn, may lead to changes in program the relevant State agency. The regulatory requirement that school meals be consistent with the Dietary 101For secondary school breakfasts, the difference between the mean ener- Guidelines has been incorporated into the FNS strategy content of pre-SMI and post-SMI breakfasts was statistically significant. gic plan. The current goal is that all schools will satis- 102The more recent Evaluation of the SBPP (McLaughlin et al., 2002) assessed the nutrient content of SBP meals in elementary schools partici- fy these standards by 2005 (USDA/FNS, 2000). pating in the SBPP demonstration in the 2000-01 school year. However, data from that study are not directly comparable to data from SNDA-I. The SBPP analysis was based on the meals actually selected by students (weighted nutrient analysis), while the SNDA-I and SNDA-II results dis- 99One serving from each category or two servings from one of the two cussed above were based on meals offered to students (unweighted nutrient categories. analysis). SNDA-II included both weighted and unweighted analyses. A 100Goals for sodium and cholesterol content are not included in SMI comparison of weighted analysis results from SNDA-II and the Evaluation nutrition standards; however, schools are encouraged to monitor levels of of the SBPP suggests that the fat and saturated fat content of SBP meals in these dietary components. elementary schools has continued to decline.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 213 Chapter 6: School Breakfast Program impacts. For these reasons, new research is essential to secondary schools and a nationally representative sam- understanding the nutrition- and health-related impacts ple of students attending those schools. SDNA-I of the SBP as it operates today (Guthrie, 2003). researchers included statistical controls for selection bias in their analysis. SNDA-I data are dated, however, because they were collected before the SMI and before Research Overview recent initiatives to increase SBP participation. This review, like the other reviews in this report, focuses on research that examined the impact of a federally Data from the SBPP are more recent—collected in sponsored food and nutrition assistance program—in this spring 2001—but they are not nationally representative case the SBP—on health- and nutrition-related out- and are based on data from six school districts that comes. A related body of research focuses on the gen- volunteered to participate in a universal-free breakfast eral impacts of eating breakfast rather than the specific demonstration. The SBPP used a randomized experi- impacts of participating in the SBP (eating an SBP mental design; however, the evaluation was designed breakfast). Much of this research was conducted in to assess the impact of a universal-free breakfast pro- controlled environments or in developing countries, and gram rather than the impact of the SBP, per se. The is not reviewed here. The interested reader can find main analyses completed for the first-year SBPP report summaries of these and related studies in two other compared the entire treatment group (students in reports (Jacobson et al., 2001; Briefel et al., 1999). schools where universal-free breakfast was available) with the entire control group (students in schools Relevant SBP research can be divided into two cate- where the standard SBP was available). Results of gories: (1) studies that looked at impacts on students’ these analyses provide no information on the question dietary intakes and (2) studies that looked at impacts that is central to understanding the impact of the SBP: on academic performance and related outcomes such Do the dietary (or other) outcomes of students who as attendance, tardiness, and behavior: A few studies participate in the SBP differ from those of students (see table 32) also examined impacts on height and/or who do not participate in the program? weight or nutritional biochemistries. (None found significant effects.) The evaluation of the SBPP is the However, SBPP researchers completed a separate only study to look at all of these outcomes concurrent- analysis that does provide some insight on this issue. ly. The following sections describe each body of A statistical procedure was used to estimate impacts on research and summarize key findings. students who actually participated in the universal- free school breakfast program. In the analysis of Impacts on Students’ Dietary Intakes dietary intakes, universal-free breakfast participation was defined as consumption of a school breakfast on The literature search identified 14 studies that attempt- the day dietary intake data were collected. The Bloom ed to estimate SBP impacts on children’s dietary intake correction (Bloom, 1984) was used to adjust the esti- (table 32). This includes two national evaluations that mate of the average impact on the entire treatment included student-level measures—SNDA-I and group, based on the difference between the proportion NESNP (the National Evaluation of School Nutrition of students in treatment and control schools who ate Programs)—as well as a reanalysis of the SNDA-I data breakfast on a typical school day. Results provide (Group I). (The third national evaluation conducted by unbiased estimates of the impact of participating in USDA—SNDA-II—did not collect student-level data). universal-free school breakfast.103 These findings are Also included are four studies based on secondary suggestive of the impact of participating in the regular analysis of data from national surveys (Group II), five SBP some 6 years after the SMI was launched.104 State and local studies (Group III), and two studies of universal-free breakfast demonstrations (Group IV). A recent study by Gleason and Suitor (2001) also deserves comment. This study used data from the The strongest available data on SBP impacts in this area come from the SNDA-I study (Gordon et al., 103 1995 and Devaney and Stuart, 1998) and from the For more information, see McLaughlin et al. (2002), chapter 4 and appendices C and F. first-year report of the evaluation of the SBPP 104The characteristics of meals provided in universal-free breakfast pro- (McLaughlin, 2002). SNDA-I is the most recent, com- grams are likely to be comparable to those provided in the regular SBP prehensive, and state-of-the-art study designed specifi- (see McLaughlin et al., 2002). However, the characteristics and consumption behaviors of students who choose to participate in universal-free cally to study the SBP. It included a nationally repre- school breakfast and students who choose to participate in the regular SBP sentative sample of public and private elementary and may not be comparable.

214 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program n n n n n od d h e t rients) s t Continued— s (foods) u bias bias gressio gressio gressio gressio gressio j d a on- on- - te re te re te re te re te re n a a a a a ison of o ent (nu ent i s s Analysis met e r g e means Multivari Multivari Bivariate t-test Multivari with selecti Compar r Multivari adjustm with selecti adjustm Multivari h on fast fast fast fast fast fast LP LP lunc ll day ll day ll day ll day ll day Measure of participation Ate SBP break on reca Ate SBP break on reca Ate SBP break on reca Ate SBP break on reca Ate SBP break on reca Ate SBP break and NS recall day (nonparticipants ate NS lunch only) y intakes t vs. t vs. t vs. t vs. t vs. t vs. pant pant pant pant pant pant n n n n n n s’ dietar t Design partici partici partici partici partici partici non non Participa non non non Participa Participa Participa Participa non Participa d) ) e n 18) ple ges ,1 am on studen nts in nts in nts in nts in 2 latio reporte 18 t 1-21 ncome en a ,966) ,966) ,693) ,809) ,180) es 1-12 es 1-12 es 1-12 esce esce esce esce dren ages dren and dren and dren and dren and 2 2 2 2 2 Popu (sample siz Chil 5-10 (n= Chil adol grad (n= 6-18 (sam size no Chil adol and 1 (n= grad (n= Chil adol SBP schools ages 6- (n= Chil adol grad (n= Low-i childr e e v v lls lls uti uti hour hour hour hour hool Breakfast Progr onsec onsec method e 24- e 24- e 24- e 24- our reca our reca Data collection Singl Singl recall Singl recall recall 24-h recall 24-h Singl 2 nonc 2 nonc the Sc f P 1 N d d pact o ools ools tative tative tative m ic an ic an ic surveys n n n ally ally ally ubl ubl ubl 4-96 CSFII 4-96 CSFII 0-81 NES onal Data source private sch private sch sample of students from 329 p sample of students from 329 p sample of students from 276 p schools Nation represe 199 Nation represe Nation represe 198 199 d at at is of nati e e e ours ours ours od er 24 er 24 er 24 er 24 2 iho Outcome(s) breakfast breakfast an over 24 h of eating breakfast Nutrient intake over 24 h Nutrient intake at breakfast and ov hours Likel Nutrient intake at breakfast and ov hours Food intak Nutrient intake at breakfast and ov hours Nutrient intake at breakfast and ov hours Food intak Food intak over 24 h of table. udies that examined the i t 9) at end 01) and II: Secondary analys I: National evaluations n 5) 3) 9) llisch et al. e See notes Table 32—S Study Group Devaney and Stuart (1998) (SNDA-I) Gordon et al. (199 (SNDA-I) W (198 (NESNP) Group Gleaso Suitor (20 Basiotis et al. (199 Devaney and Fraker (198

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 215 Chapter 6: School Breakfast Program n od h means ance ance ance ance Continued— gressio vari te re of vari of vari of vari a ison of d) Analysis met Analysis Multivari Compar (type of statistical test not reporte Analysis of Analysis Analysis d o i re r kfast kfast kfast o ool inued e on in s study a a a t p e e e al l ati a ng ing t Con n e rticip ll day ll day ll day a m ly ate sch i d r Measure of participation e p x breakfast 4-5 times/week Ate school br Usual offered dur perio on reca Took 70% or m of school me 60% p Ate school br Ate school br SBP on days in school duri e on reca on reca y intakes— 4 5 t vs. t vs. t vs. ts, t vs. t vs. pant pant pant pant pant n n n n n n s’ dietar t Design partici partici partici partici partici non non non Participa Participa Participa before vs. after non non Participa Participa Participa 0 0 ) 1 1 e 6 n 28) n n am on studen nts 7 latio 3 21 es 3-6 es 1-4 esce dren ages dren i dren i dren age dren age dren and 555) 844) 393) 393) 412) Popu (sample siz Chil 8-12 (n= grad (n= Chil Chil grad (n= Chil (n= (n= (n= Chil adol ages 6- Chil d n e e k v , e Breakfast Progr e lls uti w hour hour hour hour hour 1 g onsec method n i e 24- e 24- e 24- e 24- e 24- d our reca u l Data collection c n day Singl Singl Singl 3 nonc 24-h i Singl recall recall Singl recall recall recall the School f t t e 5 5 r 1 e n ear ear d pact o 4-8 4-8 w 88) 88) le on e m ican- nts in 2 districts ons i n, CA ls in i 987- 987- lusa H lusa H 0-71) 1-73) 0-71) 1-74 ashingt Data source and 1 and 1 Study (198 Study (198 State; Blacks and Mex Americans w W HANES-I 197 All stude school districts in rural N Boga Boga 2 schoo Compto (197 Students in schools/ in 8 reg oversamp (197 York State (197 2 udies ours ours ours ours er 24 2 Outcome(s) Nutrient intake over 24 h Nutrient intake at breakfast and ov hours Nutrient intake over 24 h Nutrient intake over 24 h Nutrient intake at breakfast Nutrient intake over 24 h of table. udies that examined the i t at end State and local st III: nd la 0) 2) 9) 8) 3a) 3b) See notes Table 32—S Study Hoag (198 Group Nicklas et al. (199 Nicklas et al. (199 Emmons et al. (197 Hunt et al. (197 Price et al. (197

216 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program al tot e n up rat od baseline h bgro were a sepa bed. gressio sed on scri e te re There a al-free breakfast ants (su es) ps ba Analysis met , are from a r e ne unit. v Not well d Multivari with Bloom correction to assess impact on univers particip analys idered o s port, howe ee call inued t program. fast different subgrou t con s re ur s al-fr kfa Con were mple. a ll day d for fo Measure of participation 9 d no brea nd 3-5) in the s Ate SBP break on reca Ate univers breakfast on re day reporte discussed in thi K-2 a that ha uded y intakes— l sults ol Re re inc ). ed t vs. pant n ent we tions (e.g., rol scho s’ dietar ts t omiz Design district partici call. pan ast. i trol group. Findings non Participa Rand experim ol to cont configura t (one s f the re kfa ) ool breakf e al scho n t grade n entire con e SBP partic sch d brea n n he day o re am on studen w income. latio an r the th h t on periment c ,290) es 3-6 es 2-6 s x dren i dren i s fo not lo 225) 4 Popu umed a s. s (sample siz kfa grad (n= Chil grad (n= Chil con nts in e ools with differe t effect e outcome e SBP. Only nd free lun ) a t s, sch with sample to th ual s Breakfast Progr rict se a ll, with ing stud f r k cess a c . e y purpo r ple (us e a b method ren in the compa gnificant findings. und no significan av e ment group our reca l si g ed in universal-free brea Data collection me (eligible for free lunch), child n i s. unch (one dist treat S recall 24-h second recall for subsam intake) the School and fo inco e l f icipat - y ly 20 rt ination Surve did not h l m st program, On o curit 1 l duals. o o who y. se s, RI, th d pact o l i sampling/matching ut reported no d d nts tually pa ction of fre or the entire o b ce, RI m n n needy, low c ary stu n 8 For s f of Indivi d fo tude ee breakfa ght, Nutrition Program r s ho a a f come Data source s 2-6. s of the ts w with schools i Provide matched w pairs of scho units in 6 sch 70 matche Element schools i Central Fal districts and Nutrition Exa d Intake School nutritionally focu n and/or wei Foo pants s on BMI an ared out and after introdu i 2,7 studen d ate, ntrol schools. c at ct n s in grade e co o s ours iversal-free breakfast er 24 t the mai ct of introducing udent n o on height s before s and 70 al Evaluation of ng Survey of Outcome(s) pact breakfast an over 24 h Nutrient intake at breakfast and ov hours Food intak Nutrient intake at breakfast the effe t was n ared SBP parti m examined impa d intake s ies of u sed on st ain analysis comp utritionally adequ udies that examined the i mated impact d t ti cu s: kfa omp s pare c ce ent schools r y also y’s m y y fo in et ristics: n tud 2) IV: Stu ghl s that e te CSFII = Continui NHANES-I = First National Health NESNP = Nation c 6) Data sou Also examined i The s Study com Study examined School brea The stud The stud The stud 1 2 3 4 5 6 7 8 9 Table 32—S Study Group McLau al. (200 Cook et al. (199 chara of 73 treatm analysi

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 217 Chapter 6: School Breakfast Program

1994-96 wave of the Continuing Survey of Food no significant impact on the likelihood of breakfast Intakes by Individuals (CSFII) to study food and nutri- consumption, regardless of the definition used. Among ent intakes of SBP participants and nonparticipants. students from low-income households, however, avail- Although well done and based on more recent data ability of the SBP was associated with a significantly than SNDA-I, this study is not as strong as either greater likelihood that students would eat a more sub- SNDA-I or the evaluation of the SBPP, for at least two stantial breakfast, (a breakfast that satisfied definition reasons. First, the CSFII data are generalizable to the 2 or 3). At the same time, availability of the SBP made U.S. population as a whole, but not to schoolchildren it significantly less likely that low-income students specifically. Second, Gleason and Suitor did not would consume a nominal breakfast (a breakfast that attempt to control for selection bias, presumably provided 10 percent or less of the REA).105 These because of the lack of relevant variables in the CSFII results, summarized in table 33, suggest that, at the dataset. Indeed, the authors caution that, because of time the SNDA-I data were collected, the primary likely selection bias, the estimates presented in their objective of the SBP was being met. That is, low- report should not be interpreted as estimates of SBP income students were more likely to eat breakfast if impacts. the SBP was available in their school.106

Impacts on the Likelihood That Impacts on Dietary Intake Students Will Eat Breakfast Table 34 summarizes results of studies that compared The overarching goal of the SBP is to provide break- dietary intakes of SBP participants and nonparticipants fast to children who might otherwise not eat before at breakfast. (As noted previously, the evaluation of starting the school day. The original analysis of the the SBPP (McLaughlin et al., 2002) actually compared SNDA-I data (Gordon et al., 1995) found that the like- intakes of participants and nonparticipants in schools lihood that a child would eat breakfast before school where universal-free breakfasts were available). Table began was not significantly different for children in 35 provides comparable data for intakes over 24 hours. schools that did and did not offer the SBP. About 12 Both tables are divided into five sections: food energy percent of the children in each type of school ate no breakfast. This analysis was flawed, however, because 105The results differed slightly for elementary and secondary school stu- it defined children who ate breakfast as those who dents. For those in secondary school, a significantly greater likelihood of breakfast consumption was observed only for the most stringent breakfast consumed at least 50 calories between the time of definition (2 food groups and more than 10 percent of the REA). waking and 45 minutes after the start of school, a 106The Evaluation of the SBPP (McLaughlin et al., 2002) assessed the threshold that could include extremely small snacks. impact of a universal-free breakfast program on the likelihood that students would eat breakfast. These data are not included in this review because As an example, an average-size sandwich cookie pro- they have limited applicability to the regular SBP, where free breakfasts are vides approximately 50 calories. available only to students who are certified to receive that benefit.

A reanalysis of the SNDA-I data, completed by Table 33—Low-income students' breakfast Devaney and Stuart (1998), considered three different consumption patterns by SBP availability definitions of “breakfast.” Each definition was based Type of breakfast consumed on foods consumed between waking and 45 minutes Any food Food from after the start of school and included foods consumed or beverage: two food at home and at school. The three definitions were: groups ≤10% >10% plus (1) Consumption of any food or beverage SBP availablility None REA REA >10% REA (except water). Percent (2) Consumption of food or beverages that con- SBP available 12.5 13.0* 6.1* 67.4* SBP not available 13.3 22.8 8.6 54.8 tributed more than 10 percent of the REA. Notes: *SBP vs. non-SBP difference is statistically significant (p<0.01). (3) Consumption of food or beverages from at least REA = Recommended Energy Allowance. two of five major food groups PLUS more than Results reported are for elementary students. For secondary 10 percent of the REA. students, a significantly greater likelihood of breakfast consumption was observed only for the most stringent definition (two food groups and >10 percent of REA). Results of this analysis indicated that, for the student Source: Devaney and Stuart (1998), reanalysis of data from population as a whole, the availability of the SBP had SNDA-I (Gordon et al., 1995).

218 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program ] less l l] l] d a l] ct n a a o Continued— i st t a fa n nsume [ [nation ) ) [nationa ) [nationa ant imp 9 8 ts co 9 break n 983 983 1 995) ( y Signific e s at n a llisch (1 llisch (1 v Participa e e {11 to 21} e W D W Gordon (1 y intake ] ] ] ] less ] ] l l d l] l] a a n n dents’ dietar o o i i t t a a nsume n n [ [ [nationa [nationa ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 9 9 1 1 ts co 8 8 n 9 9 1 1 in (2 in (2 in (2 in (2 ( ( (200 (200 y y gram on stu n n ghl ghl ghl ghl e e n n a a v v Participa Pro {5 to 10} e e t t impact McLau Nicklas (1993a) [1 city] Gleaso McLau McLau Gleaso D McLau D n fas ifica ] ] ] ] ] d ] l l] l] l] l] No sign a l] l] l] l] l] l] n a a a a a a o i nsume t a School Break co n [ [nationa [nationa [nationa [nationa [nation [nation [nation [nation [nation [nation ) ts ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 9 n 1 1 1 1 8 9 more/same 1 995) 995) 995) 995) 995) 995) in (2 in (2 in (2 in (2 in (2 ( (200 (200 (200 (200 y n n n n ghl ghl ghl ghl ghl e n Participa a v {11 to 21} e D McLau Gleaso Gordon (1 Gleaso Gordon (1 Gordon (1 McLau Gordon (1 McLau McLau Gleaso McLau Gordon (1 Gleaso Gordon (1 e impact of the ] l l] l] l] l] a l] l] ct examined th n a a a t o i t a n tha edy} [ s [nationa [nationa [nationa [nation [nation ) city] city] city] city] city] ) ) ) ) [nationa ant imp 9 } 1 1 1 8 ly ne 9 983 1 995) 995) ( nal studie (200 (200 (200 996) [1 996) [1 996) [1 996) [1 996) [1 (1993a) [1 city] y ncome Signific n n n e n a llisch (1 v e {5 to 10} {low-i {nutritio Participants consumed more e Cook (1 Cook (1 Gleaso Gleaso Nicklas Nicklas (1993a) [1 city] Emmons (1972) [2 districts] Cook (1 Gordon (1 Gordon (1 Nicklas (1993a) [1 city] D Nicklas (1993a) [1 city] W Cook (1 Emmons (1972) [2 districts] Cook (1 Gleaso of table. Findings from tes at end a 6 12 ergy energy and macronutrients See notes Table 34— Outcome Food Food en Protein Carbohydr Fat Saturated fat Vitamins Vitamin A Vitamin B Vitamin B Vitamin C

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 219 Chapter 6: School Breakfast Program ] ] l l a a ] less l l] d a ct n a o Continued— i t 3) [nation 3) [nation a n nsume [ 198 198 ) ) [nationa ant imp 9 8 ts co 9 n 983 1 ( y Signific e n a llisch (1 llisch et al. ( llisch et al. ( v Participa e e e e W W D W less d l] l] l] l] l] a a dents’ nsume [nationa [nationa [nationa [nation [nation ) ) ) 1 1 1 ts co n 995) 995) (200 (200 (200 gram on stu n n n Participa Pro t t impact Gleaso Gleaso Gleaso Gordon (1 Gordon (1 n fas ifica ] ] ] ] ] ] d l] l] No sign l] l] a a nsume co [nationa [nationa [nation [nation ts ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts n 1 1 more/same 995) 995) in (2 in (2 in (2 in (2 in (2 in (2 (200 (200 n n ghl ghl ghl ghl ghl ghl act of the School Break Participa McLau Gordon (1 Gleaso McLau McLau Gordon (1 Gleaso McLau McLau McLau e imp ] ] ] l l l] l] l] l] l] a a l] l] l] ct ed examined th n n a a a a t o o i i t t a a tinu n n tha [ [ s [nationa [nationa [nationa [nation [nation [nation ) ) on ) ) ) ) [nationa ) [nationa ant imp 002) [6 districts 9 9 } 1 1 1 C 8 8 9 9 983 983 1 1 995) 995) 995) in (2 mes} ( ( 79) [2 schools] st— studie (200 (200 (200 o 996) [ 1 city] 996) [ 1 city] y y ncome Signific n n n ghl e e fa n n a a llisch (1 llisch (1 v v eak e e {all inc {low-i Participants consumed more e e Hunt (19 Gleaso Gleaso McLau Cook (1 W Emmons (1972) [2 districts] W Gordon (1 Gleaso D Cook (1 Gordon (1 D Gordon (1 Emmons (1972) [2 districts] of table. es at br Findings from at end vin y intak m a esium u n See notes Table 34— dietar Outcome Vitamin E Folate Niaci Ribofl Thiamin Minerals Calci Iron Magn

220 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program ] ] less l l] d a ct n a o Continued— i t a n nsume [ [nationa ) ) ant imp 002) [6 districts 9 1 8 ts co 9 n 1 in (2 ( (200 y Signific n ghl e n a v Participa e D McLau Gleaso ] ] less d l] dents’ nsume [nationa ) 002) [6 districts 002) [6 districts 1 ts co ins} n in (2 in (2 (200 gram on stu e gra n ghl ghl Participa Pro {whol t t impact McLau McLau Gleaso n fas ifica ] ] ] d l] l] l] l] No sign l] l] l] a a a nsume co [nationa [nationa [nationa [nationa [nation [nation [nation ts ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts n 1 1 1 1 ns} more/same 995) 995) 995) in (2 in (2 in (2 (200 (200 (200 (200 n n n n ghl ghl ghl act of the School Break Participa {total grai McLau Gordon (1 Gordon (1 Gleaso Nicklas (1993a) [1 city] Gordon (1 Gleaso Gleaso McLau McLau Gleaso e imp ] ] ] 1 l] l] l] l] l] l] l] l] l] l] ct ed examined th a a a a a t tinu tha ins} s [1 city] [nationa [nationa [nationa [nationa [nationa [nation [nation [nation [nation on ) city] city] city] ) ) ) ) ) ) [nationa ant imp 002) [6 districts 002) [6 districts 002) [6 districts a 1 1 1 1 1 C le gra 983 995) 995) 995) 995) 993 in (2 in (2 in (2 st— studie (200 (200 (200 (200 (200 (1 996) [1 996) [1 996) [1 Signific n n n n n who ghl ghl ghl fa llisch (1 eak e {non- Participants consumed more Nicklas (1 McLau McLau Gleaso Nicklas (1993a) [1 city] Cook (1 Gleaso W Cook (1 Gleaso Gleaso Cook (1 Gordon (1 Gordon (1 Gordon (1 Gleaso Gordon McLau of table. rvings es at br e Findings from rs at end p s a y intak horus grou d sug esterol um See notes Table 34— dietar Outcome Phosp Zinc Other dietary components Chol Fiber Sodi Adde Food Dairy Fruits Grains

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 221 Chapter 6: School Breakfast Program ] y d st ific c are less ses sse d s akfa ct spe a sse y a nsume timates. ome-bre significant s ext discu adding up all the s ant imp 002) [6 districts del. e stud d h ts co n The t Th ain only to a in (2 significant findings ma s) an orted a rt point e ting,” or Signific sion mo ghl ced. s rep s ngs pert Participa han others. McLau findi rence s introdu attern of non “vote coun did not repo ndard regre h wa c Where s. Diffe tice of f sta ideration t c istent p rol schools (nutrient s s con less icipants, but d cont program 1 State). d re con e only lun rt r t s 1, a dents’ a r oid the pra npa v atche kf no ical to results o ct whe nsume 1 city or hapte merit mo d to a and distri ts co al vs. n d in c ants h and brea ally ident studies cipants in m c p i n the cautione gram on stu note s some essenti Participa free lun ple, nation Pro nonparti t but not i t impact ers are n fas exam r ifica research. A een SBP partic ings from nts and nch and y. Read ] results were y of kets}. wa , find d c s rticipa n study (fo re introduced, l] l] No sign l] a st pa cing free lu ote that {in bra n nsume ted in that co e of the ors trodu nsideratio o difference betw [nationa [nationa [nation ts d lunch we ) ) n 002) [6 districts bgroup cture of the bod co n 1 1 scop e breakfa r t an fter in ere. more/same s 995) in (2 found (200 (200 sive pi kfa y n n ghl act of the School Break Participa orted h prehen ate, and the universal-fre ign and othe ted models. Auth efore and a st studies. s oth brea s lts would be interpre McLau Gleaso Gleaso Gordon (1 e identifies the su e imp en are rep s b com min C. The on d h b nge a c a ke t -adju i s v stro d y’s resu earch d s betwe n s a ct in whi ll entry also m u the publicati i ison of inta f providing arison ection-bia c r l es in re l] ct s from the a ed c examined th pa sel a c single stud a t f for distri m significant findings o on name, tinu s s on comp tha finding e n co gh no k differen s s s). sed [nation ulation, the ce on a ant imp t C n i author’ d r 995) e are ba d in the interest o s significant e based based o en thou s ls (food r phasize v ) a st— studie ble 33). Only udy pop (1 s Signific e w st fa are s senio porte s ct, e 96) a a s. Because of eak o e re Participants consumed more s r l s (see ta Gordon a ) 3 8 et al. (19 sults a k 9 mons (1972) rdon et al. (1995 derlying effe 1 han the entire versal-free schoo post difference re cular result es at br ( subgroup . e/ l un r Findings from r a her t t e les Cell entries show the h b c with parti here p y intak s i l l umers in uni e s Notes: Nonsignificant Findings for Go Findings for Coo W Findings for Em Table 34— dietar Outcome Meat Vegeta subgroup rat indicate a true studies methodological limitations and em con those w impacts in fou

222 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program } } 0 1 1 - 5 { less ] l d s ct l] {11-2 a r a n u Continued— o i t a nsume n [

[nationa ) ant imp 9 8 ts co 89) er 24 ho 9 n 1 (

79) [2 schools] y y (19 Signific s ov e e n a v Participa e Nicklas (1993b) [1 city] Hunt (19 Devan Nicklas (1993b) [1 city] D y intake } 0 ] ] ] ] ] ] ] 1 - 5 {

less ] ] l l d l] l] l] l] l] a a l] l] n n a a dents’ dietar o o i i t t a a nsume n n [ [

[nationa [nationa [nationa [nationa [nationa [nation [nation ) ) ) ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 9 9 1 1 1 1 1 ts co 8 8 n 9 9 1 1 995) 995) in (2 in (2 in (2 in (2 in (2 in (2 in (2 ( (

79) [2 schools] 79) [2 schools] (200 (200 (200 (200 (200 y y gram on stu n n n n n ghl ghl ghl ghl ghl ghl ghl e e n n a a v v Participa Pro e e t t impact McLau McLau McLau McLau McLau Gleaso D Hunt (19 Gleaso Gordon (1 McLau D Gleaso Gleaso McLau Hunt (19 Gordon (1 Gleaso n fas

} ifica 1 2 ] ] - 1

1 ] { l ] l] l] a No sign l l] l] l] l] l] l] n a a a a a a o n i t o i nsume t a a n School Break co [ n

[ )

[nationa [nationa [nation [nation [nation [nation [nation ts [nationa ) 0 ) ) 002) [6 districts 002) [6 districts n 9 8 1 1 8 9 9 1 more/same 999) ( 995) 995) 995) 995) 995) 1 in (2 in (2

(

79) [2 schools] 79) [2 schools] 79) [2 schools] d (200 (200 y n n n ghl ghl e a l n Participa g a a v o e Hunt (19 McLau Hunt (19 Gleaso Gordon (1 Gleaso Gordon (1 Gordon (1 Gordon (1 Gordon (1 Hunt (19 Basiotis (1 McLau D H e impact of the

] ] l] l l a l] l] a a l] l] l] ct n n examined th a a a o o t i i t t a a n n [ [ tha

) [nation ) ) s [nationa [nationa [nation [nation [nationa 0 State] 0 0 ) ) ant imp 8 8 8 1 1 9 9 1 1 999) ( ( 995) 995)

75) [1 studie d d (200 (200 Signific nd (19 n n n n a a la l l g g a a Participants consumed more o o Gleaso Gleaso Gordon (1 Gordon (1 Nicklas (1993b) [1 city] H Hoag Price (19 Nicklas (1993b) [1 city] H Nicklas (1993b) [1 city] Basiotis (1 of table. s Findings from e t at end a r 6 12 d ergy y h energy and macronutrients o b r See notes a Table 35— Outcome Food Food en Protein C Fat Saturated fat Vitamins Vitamin A Vitamin B Vitamin B Vitamin C

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 223 Chapter 6: School Breakfast Program } 0 1 - 5 { less ] l d ct a a n Continued— o i t a nsume n [

) ant imp 9 8 ts co 9 n 1 (

y Signific e n a v Participa e D

} 1 2 ] ] ] ] ] ] - 1

1 ] { less l ] d l] l] l] l] a l n a o n i t o i t a a n nsume [ n

[ )

[nationa [nationa [nationa [nationa ) 0 ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 9 8 1 1 1 1 ts co 8 9 n 9 1 ( 1 in (2 in (2 in (2 in (2 in (2 in (2

(

79) [2 schools] 79) [2 schools] 79) [2 schools] 79) [2 schools] d (200 (200 (200 (200 y gram on n n n n n ghl ghl ghl ghl ghl ghl e a l n g a a v Participa Pro o e t t impact McLau McLau Gleaso Hunt (19 McLau McLau Hunt (19 Gleaso D McLau Gleaso Gleaso McLau Hunt (19 Hunt (19 H n fas ifica ] ]

] ]

l l ] l l] l] l] a a No sign a l] l] l] l] n n n a a a a o o i i o t t i nsume t a a a n n co [ [ n

[ ) )

[nationa [nationa [nationa [nation [nation [nation [nation ) ts ) 0 0 ) ) 002) [6 districts 002) [6 districts 9 n 1 8 8 1 1 8 9 9 9 1 1 more/same ( ( 1 995) 995) 995) 995) in (2 in (2

(

79) [2 schools] d d (200 (200 (200 y n n n n n ghl ghl e a a l l n Participa g g a a a v o o e d H Hunt (19 Gordon (1 Gordon (1 Gordon (1 D Gleaso Gleaso McLau McLau H Gleaso Gordon (1 e impact of the School Break e ontinu

]

l l] l]

] ] l l a a l] a l] l] l] a l] l] l] l] ct a n rs— examined th n a a a a a n o t i o t o i i t t a a a n [ n tha n

[ [ ) ) [nation ) [nation

[nationa [nation [nation [nation [nation ) State] State] 0 0 ) 0 State] State] ) ) [nationa ) [nationa ) [nationa ant imp 9 8 8 1 8 1 8 9 0 9 1 0 983 983 983 ( 1 995) 995) 995) 995) 2

( 75) [1 75) [1 75) [1 75) [1 ( (1

studie d

(200 over 24 hou y Signific n nd (19 n nd (19 n s e a o l la la n e s g a a llisch (1 llisch (1 a v e e e l Participants consumed more o e Price (19 Gleaso H G Gordon (1 Gordon (1 Price (19 W Price (19 Gordon (1 Hoag Price (19 Wellisch Gordon (1 D W Hoag y intak of table. Findings from at end

vin horus m

a esium ents’ dietar u n i i c c l a See notes a i Table 35— stud Outcome Vitamin E Folate N Ribofl Thiamin Minerals C Iron Magn Phosp

224 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program ]

] less l l] d l] a ct n a o Continued— i t a n nsume [

[nationa ) [nationa ) ant imp 002) [6 districts 9 1 8 ts co 9 n 999) 1 in (2 (

79) [2 schools] (200 y Signific n ghl e n a v Participa e D Gleaso McLau Hunt (19 Basiotis (1 ] ] ] ] ] ] less d l] l] l] l] nsume [nationa [nationa [nationa [nationa ) ) 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 1 1 ts co n 999) 999) in (2 in (2 in (2 in (2 in (2 in (2 79) [1 school] 79) [2 schools] 79) [2 schools] (200 (200 gram on n n ghl ghl ghl ghl ghl ghl Participa Pro t t impact Hunt (19 McLau Gleaso McLau McLau Hunt (19 Gleaso Basiotis (1 McLau McLau McLau Basiotis (1 Hunt (19 n fas ifica ] ] ] d l] l] l] l] l] l] No sign l] l] l] l] l] a a a nsume co [nationa [nationa [nationa [nationa [nationa [nationa [nation [nation [nation ts [nationa [nationa ) ) ) ) ) ) 002) [6 districts 002) [6 districts 002) [6 districts n 1 1 1 1 1 1 more/same 999) 999) 995) 995) 995) in (2 in (2 in (2 79) [2 schools] (200 (200 (200 (200 (200 (200 n n n n n n ghl ghl ghl Participa d Basiotis (1 Nicklas (1993b) [1 city] Gordon (1 Gleaso McLau Gordon (1 Gordon (1 Basiotis (1 Gleaso Gleaso Gleaso Gleaso Gleaso McLau McLau Hunt (19 e impact of the School Break e ontinu C l] l] l] l] ct rs— examined th a t tha s [1 city] [1 city] [nationa [nationa [nationa [nationa ) ) ) ) ant imp b b 1 1 999) 999) 993 993 studie (200 (200 over 24 hou Signific n n s e Participants consumed more Nicklas (1993b) [1 city] Nicklas (1 Nicklas (1 Gleaso Basiotis (1 Basiotis (1 Gleaso y intak of table. rvings e Findings from rs at end p s a les b grou ents’ dietar d sug esterol um See notes Table 35— stud Outcome Potassium Zinc Other dietary components Chol Fiber Sodi Adde Food Dairy Fruits Grains Meat Vegeta Soda

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 225 Chapter 6: School Breakfast Program rt y p u ide wer ses v o o r s ct g a b u s

c i f i ut did not repo ext discu c adding up all the did not pro ant imp e ts scored l p s n

The t a s, but o significant findings ma t

ting,” or Signific y l n o effect

Participa t n i a t han others. r e significant effects b p attern of non “vote coun s g significan d no n i d tice of n ideration t c istent p i oun f s s

e ted no r r e o con hey f wer h T o re con W

rep . ) y 1, a r oid the pra s l v o o h c hapte merit mo s d to a

cin. The ts scored l 3 n . nia s d in c and vitamin C. v

l studies a cautione n gram on note o i s t a some Participa n

Pro , t e t impact ers are l n p m fas a in, thiamin, iron, x ifica research. A c ings from e

y. Read r o f y of , nia ( wa 6 , find

s y d n No sign u t s

e . } h scored t s

t ted in that f vitamin A, thiamin, riboflavin, and e ts o

k nsideratio n c e er/same a p cture of the bod co r o r b c

s, iron, s n

i high ru { vitamin A, B e

h sive pi p t

u Participa d o r n g a

b , phosho prehen nesium, u e ign and othe t s st studies. s

a d lts would be interpre e e impact of the School Break e d e h com t nge n lcium,

a o s i t e i a f stro i c t y’s resu i earch d otein, mag l n r s b e ontinu u d i p C

er o e protein, ca s rgy, p f providing l h gh t y, i a

l] es in re

, ct s from the rs— c examined th rg y e single stud a r t t m n a e

n s of ene l l tha finding s e ’ gh no differen s r s c s of ene

[nationa o ant imp e h t ts scored h h t u

n take , a 999) n n ed intake r d in the interest o en thou o phasize i o v studie i t ss over 24 hou n Signific a l s e u porte e s ct, e

p asse e o s. Because of sessed i Participa h p s t

e re

r y Basiotis (1 w d o s u t h e y intak s s r

) also a e sults a s u 983) also r i e s derlying effe t i re cular result r n a t 978 e un n e

Findings from s. s. e e

l m l h

e y n r C a : with parti a h s t ents’ dietar

m e r t e o m h N Nonsignificant Wellisch et al. (1 Price et al. (1 t u a Table 35— stud Outcome S Total HEI r indicate a true studies methodological limitations and em point estimate point estimate

226 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program and macronutrients, vitamins, minerals, other dietary the day (table 35). Of the studies that examined both components, and food group servings. The text follows breakfast and 24-hour intakes, only the Devaney and this general organization, but discusses findings for Fraker (1989) reanalysis of NESNP data found that the vitamins and minerals in one section. As in all such SBP increment in food energy was not maintained over tables included in this report, results for each study are 24 hours. reported using the primary author’s name. In the interest of providing a comprehensive picture of the body With regard to other macronutrients, SNDA-I found of research, both significant and nonsignificant results that SBP participants consumed significantly less car- are reported in tables 35 and 36 and in all other “find- bohydrates at breakfast than nonparticipants and, ings” tables. As noted in chapter 1, a consistent pattern although the differences were not significant, tended to of nonsignificant findings may indicate a true underly- consume more fat and saturated fat, both at breakfast ing effect, even though no single study’s results would and over 24 hours. be interpreted in that way. Readers are cautioned, however, to avoid the practice of “vote counting,” or The evaluation of the SBPP (McLaughlin et al., 2002), adding up all the studies with particular results. the only post-SMI study identified, found no signifi- Because of differences in research design and other cant differences in energy and macronutrient intakes of considerations, findings from some studies merit more universal-free breakfast participants and nonpartici- consideration than others. The text discusses method- pants, either at breakfast or over 24 hours. Moreover, ological limitations and emphasizes findings from the the general trend was the reverse of the trend observed strongest studies. in SNDA-I. That is, on average, point estimates for the percentage of calories from fat and saturated fat were In this case, emphasis is given to findings from lower for universal-free breakfast participants than SNDA-I (Gordon et al., 1995) for the reasons cited nonparticipants. previously. Findings from the evaluation of the SBPP (McLaughlin et al., 2002) are considered to provide These results imply a change in the nutrient profile of some insight into potential changes in program impact SBP meals over time. The suggested trend—that SBP over time. To provide additional context for these meals are lower in energy and protein and lower in fat observations, data from the SBPP evaluation are con- and saturated fat (as a percentage of total energy) than sidered in light of data from the SNDA-II study (Fox they were at the time the SNDA-I data were collect- et al., 2001). ed—is consistent with findings from SNDA-II (Fox et al., 2001). SNDA-II compared the nutrient content of Findings reported for SNDA-I are based on results of SBP breakfasts offered in 1998-99 with SBP break- regression models that controlled for selection bias using fasts offered in 1991-92 (SNDA-I). an instrumental variables approach. The models used are analogous to those used in assessing NSLP impacts Vitamins and Minerals (see chapter 5). However, Gordon and her colleagues Among studies completed prior to the SMI, there is a found few substantive differences between results of virtual consensus that the SBP increased students’ models that did and did not attempt to control for intakes of three minerals—calcium, phosphorus, and selection into the SBP and noted that statistical tests magnesium—both at breakfast and, when assessed, over rejected the presence of selection bias. They appropri- the full day. There is also a consistent finding that the ately caveat this comment with the observation that the SBP increased riboflavin intake at breakfast, but this available identifying variables were not strong predic- effect generally did not persist over the full day. All of tors of SBP participation. In estimating impacts on 24- these nutrients (calcium, phosphorus, magnesium, and hour nutrient intake, models adjusted for self-selection riboflavin) occur in concentrated amounts in milk. into the NSLP but not the SBP. Findings from the Evaluation of the SBPP (McLaughlin Energy and Macronutrients et al., 2002) are somewhat consistent with this picture, Most studies completed prior to the implementation of but suggest that the current impact of school breakfast the SMI, including the SNDA-I study (Gordon et al., on mineral intake is smaller than previously estimated 1995), found that SBP participants consumed more food and that none of the impacts persist over 24 hours. In energy and protein at breakfast than nonparticipants the SBPP evaluation, universal-free breakfast partici- (table 34) and that this boost persisted over the course of pants were found to consume significantly more calcium and phosphorus at breakfast than nonparticipants,

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 227 Chapter 6: School Breakfast Program but neither of these differences persisted over 24 were reported and the statistical significance of differ- hours. Differences for magnesium and riboflavin were ences between groups was assessed using bivariate t- not statistically significant for either time point. In tests. McLaughlin et al. (2002) and Gleason and Suitor addition, the SBPP evaluation estimated usual daily (2001) computed the number of Food Guide Pyramid (24-hour) intakes and assessed the impact of universal- servings consumed by each child and assessed differ- free breakfast on the likelihood that students had ade- ences between groups using multivariate regressions. quate intakes. No significant differences were found Both analyses looked at breakfast consumption as well between students who participated in universal-free as consumption over 24 hours. breakfast and those who did not. Basiotis and his associates (1999) used data from the Data from SNDA-II provide a potential explanation 1994-96 CSFII to compare scores for food-based com- for the apparent change in impact over time. SNDA-II ponents of the Healthy Eating Index (HEI). These scores found that SBP breakfasts offered in 1998-99 provided are based on comparisons of Food Guide Pyramid 5-6 percent less calcium than breakfasts offered at the servings to age-specific recommendations. The paper time SNDA-I data were collected, although breakfasts presents results of bivariate t-tests but reports that offered at both points in time more than satisfied the results of multivariate analyses were consistent. program standard of providing one-fourth of children’s daily calcium needs (Fox et al., 2001). This pattern Findings from McLaughlin et al. (2002) provide the was observed for both elementary and secondary strongest suggestive evidence of current SBP impacts. schools. SNDA-II did not assess magnesium, phospho- These data indicate that universal-free breakfast partic- rus, or riboflavin content. ipants consumed significantly more servings of fruit and dairy products at breakfast than nonparticipants, and Other Dietary Components significantly fewer servings of meats and meat substi- SNDA-I (Gordon et al., 1995) found that SBP partici- tutes. However, data on 24-hour intakes indicate that pants consumed more cholesterol and sodium than all of these effects dissipated over the course of the day. nonparticipants (negative trends), both at breakfast and Impacts on School Performance and over 24 hours. However, none of the differences were Cognitive/Behavioral Outcomes statistically significant. The most recent (and expanding) focus of the relevant The SBPP evaluation (McLaughlin et al., 2002) found SBP literature considers impacts of eating school break- that universal-free breakfast participants consumed fast on a variety of cognitive and behavioral outcomes significantly less cholesterol than nonparticipants, both related to school performance. Characteristics of eight at breakfast and over 24 hours. In addition, mean sodi- studies identified through the literature review are um intakes were lower for universal-free breakfast par- summarized in table 36. (As noted previously, research ticipants; however the difference was not statistically conducted outside the United States or in controlled significant. The SBPP evaluation also assessed fiber environments has not been included in this review.) With intake and intake of added sugars. There was no sig- one exception (Meyers, 1989), these studies were done nificant difference between universal-free breakfast to evaluate universal-free breakfast programs rather participants and nonparticipants for either measure. than the actual SBP. Consequently, findings from these studies provide, at best, suggestive evidence of poten- The apparent shift in program impacts over time tial SBP impacts. Because the SBP does not offer implied by the SBPP data is consistent with data breakfasts free of charge to all students, impacts from SNDA-II. SNDA-II found that SBP breakfasts observed in demonstrations of universal-free breakfast offered in 1998-99 were significantly lower in cho- cannot be assumed to apply to the regular SBP. lesterol and sodium than breakfasts offered in 1991- 92 (Fox et al., 2001). Studies in this group used one of two approaches to defining a comparison group. The approach used most Food Intake often was to compare schools that offered universal- A few researchers have examined SBP impacts on free breakfast (treatment schools) with matched food consumption patterns. SNDA-I researchers schools that offered the regular SBP (control schools). (Gordon et al., 1995) examined the percentage of stu- The criteria used to match schools and the relative dents that consumed one or more foods from specific comparability of the schools ultimately selected varied food groups at breakfast. Simple weighted tabulations across studies. Some studies used a pre/post design,

228 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program n up od h bgro ance utcomes Continued— sion gressio vari te re egres a al-free breakfast ants (su is) Analysis met Logistic r Multivari with Bloom correction to asses impact on univers particip analys Analysis of al/cognitive o y of ee the havior a e 3 al-fr ent 3 her n n ing) e al-free al-free al-free ed i ed i Measure of participation univers SBP school Enroll Ate univers breakfast on d measurem (short-term cognitiv function Cumulative participation in univers breakfast over year (all ot measures) univers SBP school Enroll mance and b , r ed t vs. ts, pants pant n n nd afte ent ate omiz Design ps, plus partici partici non Rand experim Participa before vs. after, separ grou participants vs. non Participa before a ) 5 e n n ic hools grams on school perfor n n 7) latio nce dren for dren i en i ,290) es 2-6 es 3 and dren i not stated) 4 43,06 Popu (sample siz fast pro All chil attenda measures; childr grad (n= for academ measures (n= grad Chil sample sc (n= All chil reak and and and ee b d test d test d test dize dize dize method l records l records l records Data collection scores Schoo standar scores Schoo standar scores standar Schoo universal-fr f l 5- o l o n o 2 99 pact o d ls in 02) 01) m o ools i ota 8-20 9-20 0) Data source Minnes (199 Baltimore (1 pairs of scho units in 6 sch 200 455 sch 70 matche 48 scho districts (199

ic t, t, t n n n alth e, e e e ing, m nce, adem eme eme e , and 1 lin itive v h emic emic e avior i h Outcomes c Attendanc acad achiev healt Cogn function attenda tardiness, beh acad achiev student he status Attendanc and ac discip a of table. udies that examined the i t at end in et 2) ghl 3) 1a) See notes Table 36—S Study Peterson et al. (200 McLau al. (200 Murphy et al. (200

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 229 Chapter 6: School Breakfast Program e r y of n parate od h se ance; ance sion bed gressio vari vari scri from a e te re egres a school unit. The cipation on the da ate t-tests Analysis met er, are one v s Multivari Analysis of Not well d Analysis of bivari Logistic r of parti SBP. idered a s e basis ear g 1 g 1 con th in in l y this report, howe o in n n were ing 1 eek eek al-free al-free ncy of eating ncy of eating ed i ed i g scho ssed Measure of participation t implement the o nd 3-5) Ate SBP on 3 of 5 days dur selected week durin univers SBP school Enroll univers SBP school Freque breakfast dur Enroll index w Freque breakfast dur index w s discu did n K-2 a , r omes were estimated on Finding c t t vs. ts, ts, t vs. ts, pants pant pant u n n n n n nd afte o

that did and tions (e.g., 76). m ate group. r Design ps, plus ols e partici partici partici t e year. - t r s (n= o scho non Participa before vs. after, separ grou participants vs. non before vs. after Participa before a Participa non Participa before vs. after Participa h configura s

n o

) d) s t e n c t grade n comes in a n ls acher rating he entire control hools p o grams on n n n ents m latio out I 4 rticipation over th

. dren i t pa s and te pro ,023) es Pre-K-6 es 3-6 es 3-8 t a dren i dren i dren i s from t not stated) not reporte 1 133) Popu kf (sample siz fas grad (n= grad (n= sample sc (n= All chil Chil Varied by outcome for both scho Chil and stud grad (n= Chil y compared ools with differe outcome inued reak ud ata (n=85) cumulative t s. s, sch with ws ws , and , and and and ct he st ee b t and ie ie sis of se Con n d test d test d test st. T l-fr a purpo dize dize dize method l records l records l records l records t, teacher und no effe ment group n ion in universal-free brea ed for interview d d on the ba Data collection e breakfa tcomes— pat treat vari scores scores scores, pare School records student interv Schoo standar Schoo standar standar Schoo student interv pare Schoo univers s f and fo timate y s curit re e ools. s, RI, th pact o l se universal-fre i sampling/matching ia d sch or the entire ls in ls in ls in 00) 00) ce, RI m n n n es we entary o o o l in n lph For y of s f d) Sample size rol foo nce, MA de ents’ actual partici 5-87) 4) 8-20 7-20 al/cognitive ou stud and cont Data source come s 2-6. outcom stud (dates not reporte Boston, MA (199 Lawre (198 Maryland (199 schools i Provide (199 1 schoo Baltimore; 2 schools i Phila 30 scho 16 scho 55 scho matched w All elem schools i Central Fal s not a a s of BMI ared out

s and 70 amined the i t, t t t imum sample). ct ger term ased on ) w x n s in grade n n n e, e, e, e, e, ical b e e e d behavior ex s t ing m m m og (ma school nal eme e e e (1989 udent on lon v v v emic emic emic emic s e e e i i i h h h Outcomes c c c psychol measures, acad a Attendanc tardiness, acad achiev emotio function Attendanc Attendanc tardiness Attendanc tardiness, acad a Attendanc tardiness, acad a . study ded data l r examined impa sed on st o treatment ain analysis comp d impact udies tha rmance an mated impact c t ti et a cu o s f rs y’s m y also y fo al of 73 ement an 5 s that e r 0) 8) 6) 1b) 9) The stud The stud The stud For school-re The Meye 1 2 3 4 5 Table 36—S school per Study Murphy et al. (200 Murphy et al. (200 Murphy et al. (199 Cook et al. (199 Meyers et al. (198 were a tot analysi measu

230 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program where data collected before the implementation of uni- reported a significant positive effect. Similarly, all five versal-free breakfast was compared with data collected of the studies that assessed impacts on tardiness found after implementation. significant reductions in tardiness at universal-free schools. The stronger evaluation of the SBPP, which In this research, impacts were generally measured on used a randomized design and estimated impacts based the basis of group membership rather than on individ- on cumulative program participation over the course of ual behavior. As discussed in the preceding description the intervention year, found no significant differences of the SBPP evaluation, impact analyses generally in attendance or tardiness.107 compared the entire treatment group (students in schools where universal-free breakfast was available) Academic Achievement with the entire control group (students in schools All of the studies in this group considered the impact where the standard SBP was available). This is a much of offering universal-free breakfasts on academic less precise definition of participants and nonpartici- achievement. Most studies used standardized test scores pants than is used in the research that examined SBP to assess impacts, although a few used student grades. impacts on students’ dietary intake and limits the confidence one can place in the findings, relative to poten- As table 37 clearly illustrates, results of the SBPP tial impacts of the regular SBP. evaluation stand in stark contrast to results of most of the other studies. As noted earlier in this chapter, As noted previously, however, the evaluation of the USDA sponsored the evaluation of the SBPP to pro- SBPP included a separate analysis that compared uni- vide a scientifically sound study of this issue. Virtually versal-free breakfast participants and nonparticipants all of the other studies in this group are limited to one on the basis of actual participation in the school break- geographic area (one city or State), most had small fast program. For analyses that focused on school-per- sample sizes, and there was no consistency across formance outcomes, participation was defined on the studies in the measures used to assess achievement. basis of either same-day or cumulative participation over Moreover, all of these studies are subject to problems the implementation year, depending on the outcome. of selection bias because they used nonexperimental This more precise definition of universal-free breakfast designs. Finally, as Ponza and his colleagues (1999) participation, combined with the randomized design, point out, the analyses used in many of these studies dictates that considerably more credence be given to are open to question because they did not adequately results of the SBPP study than to the other studies. control for clustering.

In interpreting these findings, however, it is important The SBPP evaluation does not suffer from the design to note that (1) breakfast skipping was low in SBP and measurement weakness that limit the other studies schools; most children ate something for breakfast in this group. As such, it provides definitive data on either at school, home, or elsewhere and (2) findings the impact of universal-free breakfast participation. are based on data from the first year of a 3-year The SBPP study compared gains in standardized test demonstration and may not hold across all 3 years. scores for reading and math for universal-free breakfast participants and nonparticipants (defined on the Key findings for all studies are summarized in table 37 basis of cumulative annual participation rates), and and are discussed below, by outcome. found no significant differences. Attendance and Tardiness Cognitive Functioning Attendance and/or tardiness are important school per- The SBPP evaluation also examined the impact of formance outcomes because they may serve as media- same-day participation in universal-free breakfast on tors of any effect breakfast consumption may have on three different measures of cognitive functioning: learning. If the presence of a breakfast program stimulus discrimination, digit span, and verbal fluency. encourages attendance and/or discourages tardiness, These measures assess students’ memory and retrieval the program may have a positive influence on school skills as well as attentional abilities, and all three were performance simply by increasing the amount of time expected to be sensitive to the immediate effects, if students spend at school.

Five of the seven studies that looked at the effect of 107Data on tardiness were not consistently available at the student and/or universal-free school breakfast on attendance rates school level.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 231 Chapter 6: School Breakfast Program ] y nal ses s ct rse a ific ositio c ng} ts wo spe i n ext discu adding up all the a ant imp 002) [6 districts 3) [1 State] ad 00 ated opp The t in (2 significant findings ma ting,” or Signific Participa ghl grade re tain only to rd {teacher-r scale} {3 per han others. McLau Peterson (2 attern of non “vote coun tice of istent p ideration t c s s ] ] ] ] con Where findings ow a re con rse oid the pra v ffects. schools). ts wo 3 hapter 1, n merit mo d to a 002) [6 districts 002) [6 districts 002) [6 districts 002) [6 districts 3) [1 State] 00 al vs. d in C nificant e in (2 in (2 in (2 in (2 studies sig cautione Participa note ghl ghl ghl ghl ty to focus and foll programs on i s some st ple, nation {abil instructions} {other scales} fa t impact ers are McLau Peterson (2 McLau McLau McLau n exam r and found no ifica research. A ings from y. Read break ] ] rity e y of kets}. wa e , find c s e n study (fo secu No sign l-fr a {in bra grad th tter/same ted in that

e of the } ; 5 nsideratio [1 State] [1 State] a h ) [1 State] ) [1 State] ) ) t ts be 002) [6 districts 002) [6 districts 3) [1 State] bgroup cture of the bod co b b b b a univers n scop ng, writing} r d

00 l a 001 001 001 001 in (2 in (2 u icipa sive pi d i ghl ghl v i grade mat Part d rd n i prehen ate, and the {office referrals} {3 math, readi { ign and othe impact of st studies. s nitive functioning and food lts would be interpre Murphy (2 McLau McLau Murphy (2 Peterson (2 Murphy (2 Murphy (2 e identifies the su e com on d tcomes nge cog a stro y’s resu earch d term - s ll entry also short the publicati f providing es in re ct s from the c on examined th single stud a t s better name, al/cognitive ou s ores} tha finding gh no c [1 city] [1 State] [1 State] differen s s [1 city] [1 city] [1 city] [3 schools] [3 schools] [3 schools] [3 schools] 1 city ulation, the ce ) ) ) [1 State] [1 city] [1 city] ) [1 city] city] city] ant imp d impact a b b b e s ns} author’ sse r 001 001 001 000) 000) 001 000) 998) 998) 998) 998) 000) d in the interest o en thou se phasize v nsio l-wid studie udy pop 996) [1 996) [1 s Signific Participants st d behavior senio porte ct, e s. Because of also a {schoo {suspe e re r Murphy (2 Murphy (2 Cook (1 Meyers (1989) Cook (1 Murphy (2 Murphy (1 Meyers (1989) Murphy (2 Murphy (1 Murphy (2 Meyers (1989) Murphy (2 Murphy (2001a) [1 city] Murphy (2 Murphy (2 Murphy (1 Murphy (1 (2002) sults a derlying effe han the entire rmance an re cular result o un

f Findings from t orted her t n e e Cell entries show the ality m with parti nal e ior/ status v h e i h Notes: Nonsignificant McLaughlin et al. c Table 37— school per Outcome Attendanc Tardiness Academic a Behav emotio function Nurse referrals/rep healt subgroup rat indicate a true studies methodological limitations and em

232 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program any, of breakfast consumption. The analysis revealed saturated fat, as a percentage of total energy intake, as only very minor differences between groups, and none well as intakes of cholesterol and sodium, were greater of the differences was statistically significant. for SBP participants than nonparticipants. Data from the post-SMI SBPP evaluation suggest that, currently, Other Outcomes there are few significant differences in the nutritional Studies of universal-free school breakfast have also quality of breakfasts consumed by SBP participants examined measures of student behavior and health. and those consumed by nonparticipants, and that dif- The evaluation of the SBPP found a significant and ferences that are observed dissipate over the course of negative effect of universal-free breakfast participation the day. While not definitive, the patterns observed in (defined on the basis of cumulative participation rates the SBPP data are consistent with the most recent over the demonstration year) on teacher-rated behav- national study of the nutritional characteristics of SBP ioral opposition, but no effects on a variety of other meals (SNDA-II). behavioral measures. Although data from the SBPP and SNDA-II studies are The evaluation of the SBPP also examined impacts of useful, the true impact of the post-SMI SBP on stu- universal-free breakfast participation on student health dents’ dietary intakes is unknown. As discussed in status, Body Mass Index (BMI) (a measure of weight detail in chapter 5, there is a critical need for an updat- status), and food security status. No significant effects ed study of both the NSLP and the SBP and the pro- were reported. grams’ impacts on children. Data from several State and local studies of universal- Summary free school breakfast demonstrations reported that the availability of a universal-free breakfast program had a The available research suggests that low-income students positive impact on attendance, tardiness, academic are more likely to consume a substantial breakfast when achievement, and/or related outcomes. However, the the SBP is available to them. Pre-SMI research indicated methodologically superior evaluation of the SBPP that SBP participants had significantly higher intakes found no such effects. The only significant impact of nutrients provided by milk (calcium, phosphorus, reported in the first-year report of the SBPP evaluation magnesium, and riboflavin) at breakfast and/or over 24 was an increase in oppositional behavior among long- hours. There was also strong evidence that SBP partic- term participants in unversal free breakfast. The pro- ipants consumed significantly more food energy and ject’s final report, expected in 2004, will confirm protein at breakfast than nonparticipants, as well as less whether these results held over all 3 years of the carbohydrates. In addition, although differences were demonstration. not statistically significant, mean intakes of fat and

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 233 Chapter 6: School Breakfast Program

References Glantz, F., R. Berg, D. Porcari, et al. 1994. School Lunch and Breakfast Nonparticipants Study. USDA, Abraham, S., M. Chattopadhyay, M. Montgomery, et al. Food and Nutrition Service. 2002. The School Meals Initiative Implementation Study: Third Year Report. USDA, Food and Nutrition Service. Gleason, P., and C. Suitor. 2001. Children’s Diets in the Mid-1990’s: Dietary Intake and Its Relationship Basiotis, P., M. Lino, and R.S. Anand. 1999. Eating with School Meal Participation. USDA, Food and Breakfast Greatly Improves Schoolchildren’s Diet Nutrition Service. Quality. Nutrition Insights No. 15. USDA, Center for Nutrition Polity and Promotion. Gordon, A.R., B.L. Devaney, and J.A. Burghardt. 1995. “Dietary Effects of the National School Lunch Program Bloom, H. 1984. “Accounting for No-Shows in and the School Breakfast Program,” American Journal Experimental Evaluation Designs,” Evaluation Review of Clinical Nutrition 61(1, Supplement):221S-31S. 8(2):225-46. Guthrie, J. 2003. Food Assistance Research Brief— Briefel, R., et al. 1999. Universal-free School Do Healthy School Meals Cost More? FANRR-34-6. Breakfast Program Evaluation Design Project: Review USDA, Economic Research Service. of Literature on Breakfast and Learning. USDA, Food and Nutrition Service. Hoagland, G.W. 1980. Feeding Children: Federal Child Nutrition Policies in the 1980s. Washington, Burghardt, J., A. Gordon, N. Chapman, et al. 1993. DC: U.S. Government Printing Office. The School Nutrition Dietary Assessment Study: School Food Service, Meals Offered, and Dietary Hunt, I.F., H.M. Lieberman, A.H. Coulson, et al. 1979. Intakes. USDA, Food and Nutrition Service. “Effect of a Breakfast Program on the Nutrient Intake of Black Children,” Ecology of Food and Nutrition Cook, J.T., O. Punam, and G.L. Kelly 1996. 8(1):21-28. Evaluation of a Universally-Free School Breakfast Program Demonstration Project: Central Falls, Rhode Jacobson, J. et al. 2001. Designs for Measuring How Island. Medford, MA: Tufts University, Center on the School Breakfast Program Affects Learning. Hunger, Poverty, and Nutrition Policy. E-FAN-01-013. USDA, Economic Research Service.

Devaney, B., and T. Fraker. 1989. “The Dietary McLaughlin, J., L. Bernstein, M.K. Crepinsek, et al. Impacts of the School Breakfast Program,” American 2002. Evaluation of the School Breakfast Program Journal of Agricultural Economics 71(4):932-48. Pilot Project: Findings from the First Year of Implementation. USDA, Food and Nutrition Service. Devaney, B., and E. Stuart. 1998. “Eating Breakfast: Effects of the School Breakfast Program,” Family Meyers, A.F., A.E. Sampson, M. Weitzman, et al. Economics and Nutrition Review 1998:60-62. 1989. “School Breakfast Program and School Performance,” American Journal of Disabled Children Emmons, L., M. Hayes, and D.L. Call. 1972. “A Study 143:1234-39. of School Feeding Programs. I. Economic Elgibility and Nutritional Need. II. Effects on Children with Murphy, J.M., M. Pagano, and S.J. Bishop. 2001a. Different Economic and Nutritional Needs,” Journal of Impact of a Universally-Free, In-Classroom School the American Dietetic Association 61(9):262-75. Breakfast Program on Performance: Results from the Abell Foundation’s Baltimore Breakfast Challenge Food Research and Action Center. 2003. School Evaluation. Boston, MA: Massachusetts General Breakfast Scorecard: 2002. Washington, DC. Hospital.

Fox, M.K., M.K. Crepinsek, P. Connor, et al. 2001. Murphy, J.M., and M. Pagano. 2001b. Effects of a The Second School Nutrition Dietary Assessment Study Universally Free, In-Classroom School Breakfast (SNDA-II): Final Report. USDA, Food and Nutrition Program: Final Report from the Third Year of the Service. Maryland Meals for Achievement Evaluation. Boston, MA: Massachusetts General Hospital.

234 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 6: School Breakfast Program

Murphy, J.M., M. Pagano, K. Patton, et al. 2000. Reddan, J., K. Wahlstrom, and M. Reicks. 2002. Boston Public Schools Universal Breakfast Program: “Children’s Perceived Benefits and Barriers in Final Evaluation Report. Boston, MA: Massachusetts Relation to Eating Breakfast in Schools With or General Hospital. Without Universal School Breakfast,” Journal of Nutrition Education and Behavior 34(1):47-52. Murphy, J.M., M.E. Pagano, J. Nachmani, et al. 1998. “The Relationship of School Breakfast to Psychosocial Rosales, W., and J. Janowski. 2002. The State of and Academic Functioning: Cross-sectional and Breakfast in Wisconsin. Milwaukee, WI: Hunger Task Longitudinal Observations in an Inner-city School Force of Milwaukee. Sample,” Archives of Pediatrics and Adolescent Medicine 152(9):899-907. Rossi, P.H. 1998. Feeding the Poor: Assessing Federal Food Aid. Washington, D.C.: The AEI Press. Nicklas, T.A., B. Weihang, and G. Berenson. 1993a. “Nutrient Contribution of the Breakfast Meal U.S. Department of Agriculture, Food and Nutrition Classified by Source in 10-year-old Children: Home Service. 2003a. Program data. Available: http://www. versus School,” School Food Service Research Review fns.usda.gov/pd. Accessed April 2003. 17(2):125-32. U.S. Department of Agriculture, Food and Nutrition Nicklas, T.A., B. Weihang, L. Webber, et al. 1993b. Service. 2003c. “School Breakfast Program: Fact Sheet.” “Breakfast Consumption Affects Adequacy of Total Available: http://www.fns.usda.gov/cnd/breakfast/ Daily Intake in Children,” Journal of the American aboutBFast/bfastfacts.htm. Accessed August 2003. Dietetic Association 93(8):886-91. U.S. Department of Agriculture, Food and Nutrition Peterson, K., M. Davison, K. Wahlstrom, et al. 2003. Service. 2003c. “Program Fact Sheet: National School Fast Break to Learning School Breakfast Program: A Lunch Program.” Available: http://www.fns.usda.gov/ Report of the Second Year Results, 2001-2002. cnd/Lunch/default.htm. Accessed June 2003. Minneapolis, MN: University of Minnesota. U.S. Department of Agriculture, Food and Nutrition Ponza, M., R. Briefel, W. Corson, et al. 1999. Service. 2003d. “Provision 1, 2, & 3 Fact Sheet.” Universal-free School Breakfast Program Evaluation Available: http://www.fns.usda.gov/cnd/governance/ Design Report: Final Evaluation Design. USDA, Food prov-1-2-3/prov1_2_3Factsheet.htm. Accessed August and Nutrition Service. 2003.

Price, D.W., D.A. West, G.E. Scheier, et al. 1978. U.S. Department of Agriculture, Food and Nutrition “Food Delivery Programs and Other Factors Affecting Service. 2000. “Food and Nutrition Service (FNS) Nutrient Intake of Children,” American Journal of Strategic Plan 2000 to 2005.” Available: http://www. Agricultural Economics 60(4):609-18. fns.usda.gov/oane/MENU/gpra/FNSStrategicplan.htm. Accessed April 2002. Project Bread. 2000. The Massachusetts Child Hunger Initiative: An Action Plan To End Child Hunger in Wellisch, J.B., S.D. Hanes, L.A. Jordon, et al. 1983. Massachusetts. Boston, MA. The National Evaluation of School Nutrition Programs: Final Report. Volumes 1 and 2. Santa Monica, CA: Systems Development Corporation.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 235 Chapter 7 Child and Adult Care Food Program

The Child and Adult Care Food Program (CACFP) To be eligible for Federal reimbursement, providers provides Federal funds for meals and snacks served to must serve meals and snacks that meet established children and adults in licensed, nonresidential day care meal pattern requirements. These requirements are facilities, including family and group child care modeled on the food-based menu planning guidelines homes, some child care centers, Head Start programs, used in the National School Lunch Program (NSLP) after-school care programs, and adult day care and the School Breakfast Program (SBP) (see chapters centers.108 Since July 1999, the program has also 5 and 6). The meal patterns specify foods (meal com- served preschool children who reside in homeless shel- ponents) to be offered at each meal and snack as well ters. Federal assistance reimburses care providers at as minimum portion sizes. For children, minimum por- fixed rates for each meal and snack served. tion sizes vary by age. Currently, CACFP meals and snacks are not required to meet specific nutrient-based The limited amount of research on the CACFP is standards such as those implemented in the mid-1990s almost entirely descriptive, focusing on the character- for the NSLP and SBP. Child care centers and homes istics of participating institutions and the children and may receive reimbursement for two meals and one adults they serve. Several studies, including four snack or two snacks and one meal per child per day. nationally representative studies sponsored by the U.S. Homeless shelters may receive reimbursement for Department of Agriculture (USDA), have documented three meals per child per day. the nutrient content of meals and snacks offered to participants, but only one of these studies examined Child Care Component the nutrient content of meals and snacks offered in The CACFP began as a pilot program in 1968, known nonparticipating institutions. as the Special Food Service Program for Children Some studies have assessed the nutrient contribution (SFSPFC). The SFSPFC was established under Section of CACFP meals and snacks to participants’ overall 17 of the National School Lunch Act (42 U.S.C. 1766). diets. However, there has been no research on the Participation was initially limited to center-based child impact of the program on participants’ nutrition and care in areas with poor economic conditions. health status, relative to nonparticipants. Beginning in 1976, family child care homes were also eligible to participate, provided that they met State licensing requirements, where these were imposed, or Program Overview obtained approval from a State or local agency. Homes had to be sponsored by a nonprofit organization that The CACFP is, in reality, two separate programs. One assumed responsibility for ensuring compliance with component serves children in child care centers, fami- Federal and State regulations and that acted as a con- ly child care homes, after-school care programs, and duit for meal reimbursements. These rules govern par- homeless shelters, and the other component serves ticipation of family child care homes to this day. adults attending adult day care centers. Program regulations allow these components to be administered by The CACFP became a permanent program in 1978. At two separate State agencies. the time, the program was focused exclusively on children and was called the Child Care Food Program The child care component of the program is substantially (CCFP). The program was not renamed the Child and larger than the adult care component. In December 2001, Adult Care Food Program (CACFP) until 1987, when the program served an average of 2.6 million children the adult day care component of the program was and 74,000 adults per day (USDA, Food and Nutrition added. (The adult care component of the program is Service (FNS), 2002a). In FY 2002, the $1.9 billion reim- discussed in a subsequent section of this chapter.) bursed to participating institutions supported the provision of 1.7 billion meals and snacks to children and 44.6 Initially, the system used in the CCFP to reimburse both million meals and snacks to adults (USDA/FNS, 2003). centers and homes was modeled after the system used in the NSLP. Three categories of reimbursement were 108Program regulations refer to snacks as “supplements.” established, based on family income, and a means test

236 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program was used to determine the family incomes of individual This growth brought with it a change in the profile of children. The largest reimbursement was provided for children being served by the CCFP. In early 1980, pro- meals served to children with family incomes of 125 gram administrative data showed that most of the chil- percent or less of the Federal poverty level (“free” dren served in participating homes were from low- meals); a lesser reimbursement was provided for meals income families—only 32 percent of these children served to children whose family incomes ranged from were from families with incomes above 195 percent of 125 to 195 percent of poverty (“reduced-price” meals); the poverty level. By January 1982, however, most of and the lowest reimbursement was provided for meals the children served in participating homes were from served to children whose family incomes exceeded 195 middle-income families—62 percent of children in percent of the poverty guideline (“full price” meals).109 participating homes were from families with incomes above 195 percent of the poverty level (Glantz et al., Applying the means test in family child care settings 1983). By 1995, with over 190,000 homes participat- was perceived to limit participation. Providers coming in the program, more than 75 percent of the chil- plained that the means test was overly burdensome and dren in participating homes were from families with too invasive for their relationship with the few families incomes above 185 percent of the poverty level (the for whom they provided child care. In addition, spon- revised threshold for eligibility for reduced-price sors claimed that meal reimbursements were insuffi- meals established in 1982) (Glantz et al., 1997). cient to cover their administrative costs and allow for adequate reimbursement to the homes.110 As a conse- Program Changes To Improve quence, very few homes participated in the program— Benefit Targeting fewer than 12,000 by December 1978, approximately Since the mid- to late 1990s, several changes have 2 years after homes were eligible to participate. been implemented to better target the benefits provid- The 1978 Child Nutrition Amendments (P.L. 95-627) ed through the child care component of the CACFP incorporated wide-ranging changes to the CCFP with and to expand program coverage to meet the needs of the purpose of expanding participation, particularly low-income children receiving care in other settings. among family child care homes. Most significantly, the The most dramatic change was implemented in 1996 1978 Amendments eliminated the means test for as part of the Personal Responsibility and Work homes. The three-level reimbursement structure was Opportunity Reconciliation Act (PRWORA) (P.L. 104- replaced with a single reimbursement rate for all par- 193). PRWORA changed the reimbursement structure ticipants, at a level slightly below the free-meal reim- for family child care homes to target benefits more bursement rate for child care centers. In addition, the specifically to homes serving low-income children. amendments separated the reimbursement of sponsors’ The new rate structure for family child care homes administrative costs from the meal reimbursement for took effect July 1, 1997. family child care homes. Other changes included alter- Under the new reimbursement structure, family child native procedures for approving homes and startup and care homes located in low-income areas have reimburse- expansion funds for family child care sponsors. ment rates that are similar to the rates that existed for The 1978 Amendments provided financial incentives all family child care homes before PRWORA. A low- for homes serving middle-income children to partici- income area is defined as either an area where at least pate in the CCFP and for sponsoring agencies to half of the children live in families with incomes below recruit such homes into the program. Following the 185 percent of the poverty level or an area served by implementation of these amendments in May 1980, the an elementary school in which at least half of the family child care component of the program began a enrolled children are eligible for free or reduced-price period of tremendous growth. Between June 1980 and school meals. Homes where the provider’s own March 1981, the number of participating homes income is below 185 percent of the poverty guideline increased by 40 percent—from 17,000 to 43,000. have the same reimbursement structure as homes located in low-income areas. Homes meeting one of these criteria are referred to as tier I homes. 109Effective January 1982, the income eligibility for free meals was increased from 125 to 130 percent of poverty and the threshold for All other homes are reimbursed at substantially lower reduced-price meals was reduced from 195 to 185 percent of poverty. rates. This latter group of homes, referred to as tier II 110Meal reimbursements generated by participating homes were paid directly to the sponsoring agency. The sponsor was permitted to deduct homes, includes those that are neither located in a low- administrative costs before passing reimbursement on to providers. income area nor operated by a low-income provider.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 237 Chapter 7: Child and Adult Care Food Program

Tier II homes can receive the higher tier I reimburse- Table 38—Meal reimbursement rates for homes ment rates for meals and snacks served to children and centers, July 1, 2002-June 30, 20031 from families with incomes below 185 percent of Family child Child or adult poverty, if family income is documented. FY 2003 care homes care centers reimbursement rates are shown in table 38. Tier I Tier II Meal rate rate Paid Reduced Free As noted previously, family child care homes can par- Dollars ticipate in the CACFP only if they are sponsored by a recognized sponsoring agency. Sponsors are responsi- Breakfast 0.98 0.37 0.22 0.87 1.17 ble for determining that homes meet the CACFP eligi- Lunch/supper 1.80 1.09 .20 1.74 2.14 Supplement (snack) .53 .14 .05 .29 .58 bility criteria, for providing training and other support 1 to family child care providers, and for monitoring homes Meal reimbursement rates are higher in Alaska and Hawaii. to ensure that they comply with applicable Federal and State regulations. Sponsors receive and verify the Table 39—Monthly administrative cost homes’ claims for CACFP reimbursement, forward the reimbursement rates for sponsors of family child claims to USDA for payment, receive the reimburse- care homes, July 1, 2002-June 30, 20031 ments from USDA, and distribute the meal reimburse- Number of homes Rate per home ments to the homes. Sponsors receive Federal reim- Dollars bursement for the costs of providing these administrative services that are the lesser of (1) actual costs, (2) Initial 50 (homes 1-50) 84 the budget amount approved by their State CACFP Next 150 (homes 51-200) 64 Next 800 (homes 201-1,000) 50 office, (3) 30 percent of total program funds (funds All additional (homes 1,001 and over) 44 and administrative reimbursements), or (4) the sum of 1Administrative cost reimbursement rates are higher in Alaska the number of homes sponsored times the administra- and Hawaii. tive cost reimbursement rates shown in table 39.

The legislative changes enacted under PRWORA do component of the CACFP to include after-school care not affect sponsors’ administrative payment levels, but programs and homeless shelters. To be eligible for par- do add new responsibilities. Sponsors have been given ticipation, after-school programs must be located in primary responsibility for classifying providers as tier geographic areas where 50 percent or more of the chil- I or tier II. In addition, for tier II homes seeking reim- dren enrolled in school are eligible for free or reduced- bursement at the tier I level for individual children, price meals in the NSLP. They must also provide regu- sponsors are responsible for administering the income lar, structured activities for children, including educa- test. Parents send income verification forms directly to tional and enrichment activities (USDA/FNS, 2002b). sponsors, who then determine whether the household Snacks are served free of charge, and providers are income is below 185 percent of the poverty guideline. reimbursed at the free snack rate for all snacks provid- Providers are notified of the number of children ed. Reimbursement is limited to one snack per child approved for the higher reimbursement rates but not per day on school days, weekends, or holidays during the names of the children approved. the school year.

A congressionally mandated study of the effect of tier- P.L. 105-336 also added homeless shelters to the list of ing found that the legislative change achieved desired institutions eligible to participate in the CACFP. The objectives: The number of low-income children served participation of homeless shelters grew out of a in CACFP homes grew by 80 percent between 1995 and demonstration project (the Child Nutrition Homeless 1999, and the number of meal reimbursements for low- Demonstration) that was authorized by P.L. 101-147 (the income children doubled (Hamilton et al., 2001). Child Nutrition and WIC Reauthorization Act of 1989). Moreover, tiering had no adverse effect on either the The purpose of the demonstration was to determine the number or nutritional characteristics of meals and snacks best means of providing year-round food assistance to offered by tier II providers (Crepinsek et al., 2002). homeless preschool children residing in emergency shelters (Macro International, 1991). Sites selected for Other Recent Program Changes the demonstration provided free meals and snacks to In 1998, the Child Nutrition Reauthorization Act (P.L. children, following CACFP meal pattern guidelines, 105-336) expanded institutional eligibility for the child and received standard CACFP reimbursements.

238 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program

The demonstration ran for 4 years (FY 1990 through The adult component of the CACFP operates in essen- FY 1994) and increased from one sponsor and four tially the same manner as the child care center compo- shelters serving approximately 22,000 meals (Macro nent of the program. Adult day care centers are reim- International, 1991) to 59 sponsors and 81 shelters bursed for meals and snacks served to participating serving more than 700,000 meals and snacks (Fox and adults, using the same income-eligibility criteria and Cutler, 1996). Because the demonstration showed the reimbursement rates as participating child care centers. feasibility of providing USDA-reimbursed meals and Moreover, meals and snacks served in adult day care snacks in a variety of homeless shelters, the Homeless centers must meet CACFP meal pattern requirements Child Nutrition Program was established as a perma- to qualify for Federal reimbursement. nent program in 1994 (P.L. 103-448). The program was incorporated into the CACFP in July 1999. Review of Research on the Child In 2000, the Agricultural Risk Protection Act (P.L. Care Component of the CACFP 106-224) expanded institutional eligibility for the To date, no research has examined the impact of the CACFP to include some for-profit child centers. child care component of the CACFP on participants’ Eligibility was extended to centers where 25 percent or nutrient intake or other nutrition- and health-related more of enrolled children (or 25 percent of licensed outcomes. Ten descriptive studies of the child care capacity, whichever is less) are eligible for free and component of the CACFP were identified. Characteris- reduced-price meals.111 Initially, the timeframe for this tics of these studies are summarized in table 40. temporary provision was December 21, 2000, to September 30, 2001. In 2001, under P.L. 107-76, the Seven studies, four of which were national in scope, timeframe was extended through September 2002. In examined the nutrient content of meals and snacks 2003, the FY 2003 appropriations bill extended the offered in child care centers and/or homes participat- timeframe through September 2003 (Garnett, 2003). ing in the CACFP. Six studies assessed children’s nutrient intake from CACFP meals. Four of these stud- Finally, in 2000 and 2001, after-school care programs ies looked at nutrient intake both in and out of care, so in seven States (Delaware, Michigan, Missouri, New were able to describe the contribution of CACFP York, Oregon, Pennsylvania, and Illinois) were author- meals and snacks to children’s overall diets. ized to provide supper to participating children (USDA/FNS, 2002c). Only one study, the first national study of the program, examined meals and snacks offered to nonparticipating Adult Day Care Component children—that is, to children receiving care in child In 1987, as a means of increasing support for elderly care centers that did not participate in the CACFP feeding programs, P.L. 100-175 amended the Older (Glantz and O’Neill-Fox, 1982). This study, which was Americans Act to mandate that the CCFP be expanded completed when the program was still focused exclu- to allow eligible adult day care centers to participate. sively on children, is described next. Centers that provide day care services to persons age 60 or older or to functionally impaired persons 18 and older Evaluation of the Child Care Food Program are eligible to participate in the program. Eligible centers The 1978 Child Nutrition Amendments (P.L. 95-627) have the option of participating in the CACFP or in the directed USDA to study meal quality in day care centers Nutrition Services Incentive Program (discussed in and homes that participated in the CCFP. The study chapter 10), but cannot receive reimbursement under examined the nutrient content and nutrient density of both programs for the same meal. Participation in the meals and snacks offered in participating and nonpar- adult component of the CACFP has increased steadily ticipating child care centers as well as the quality and over time, after a period of rapid growth in the early variety of foods offered (Glantz and O’Neill-Fox, years of operation. Since 1988, the number of meals 1982). The study also described meals and snacks served to adults has increased from 2 million to offered in participating homes, but did not include a approximately 45 million (USDA/FNS, 2003). sample of nonparticipant homes.112

111For-profit centers that either have tax-exempt status or receive Title XX funding for 25 percent or more of enrolled children have long been eligible for CACFP participation. The provision in P.L. 106-224 did not affect 112The study was unable to identify a sample of eligible nonparticipating eligibility of so-called “Title XX centers.” homes.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 239 Chapter 7: Child and Adult Care Food Program d— ue ecutive ecutive d r the o Contin h ncons ncons t e m forms for a 5-day forms for a 5-day forms for 3 10- forms for a 10- forms for a 10- n and/o u u u u u ent for 2 o i t m c e l l o tions for 2 no tions for 2 no ed men ed men ed men ed men ed men c with par e days a a d a ll t erv od erv a utiv ister ister ister ister ister rio D n n n n n e iods iod iod ood Progra y peri d d onsec our reca a In-care obs days nonc In-care obs days 24-h Self-admi Self-administered menu forms for a 3-day p perio Self-admi Self-admi perio Self-admi day per day per Self-admi Self-administered menu forms for a 20-d day per Care F 2 h t s 2 6 2 1 r a o s N n a n i K r les: le of 37 le of 33 le of 54 le of 1,96 le of 17 n e i t nters p p p p p p n s r e e c t nters n 1 e e sam e sam e sam e sam e sam e sam e Child and Adult e l f c p o of 40 ce d m e l e a t p ting ce g centers homes S c a g m mple e p l n i a a mes mes esentativ esentativ esentativ esentativ esentativ esentativ atin e nters s s pr pr pr pr pr pr e pati y i c l d ho d ho n rticip ts m offered in th nd ce e ence s a npartic i i o n n ally re ally re ally re ally re ally re ally re d ina e n v a cks r n 100 p 64 no 60 partic from CACFP meals and snacks o 6 C 4 Nation centers an Nation homes a tier II homes Nation Nation centers an Nation in Texas centers Nation Conve Carol City area eals and snacks offered participan nd sna m s s s s s a of m A A A A A es es es D D D D D als a lin lin lin R R R R R o o o o o t t t t t ide ide ide to RDAs to RDAs u u u to RDAs, to RDAs ericans ericans e e e e e e e v v v v v i i i i i t t t t t tiv tiv e for a a a a a a a l l l l l tive tive ) e e e e e a a lin s r r r r r ent of me ed ed by progr ( t s s s s s e r take relative to u u u u u ety Dietary G Dietary G Dietary G u n n n n n um s e e e e e enus rel enus rel Nutrient content Children’s nutrient intake s a ines for Am ines for Am m m m m m m m e t con f f f f f f f ry Guide M h the h the h the o o o o o o o intake rel intake rel rient in idel idel a y of menus t t t t t t nd vari t n n n n n n n ent ent e e e e e e e t t t t t t t e wit e wit e wit Diet n n n n n n n ensit cks con d o o o o o o o ality a a c c c c c c c ncy of foods serv e nutrien t t t t t t t etary Gu etary Gu ianc ianc ianc n n n n n n n the CACFP: the CACFP: e e e e e e e Di Di i i i i i i i r r r r r r r t t t t t t t u u u u u u u In-care nutri N N In-care nutri and Americans Out-of-care nut RDAs an Compl for Americans Nutrient d Food qu and for Americans Compl for Americans Compl Freque N N N N N eals and sn ) xamined th mponent of mponent of ) 2 8 003 9 1 ( ) x 2 o 0 F 0 - l ) ) l 2 of table. i ( 3 9 udies that e ) e ribution of m Burstein (2 . 9 8 t l t 7) 7 N 9 9 e child care co e child care co a ’ 9 ) 1 1 ) h h nd t 9 3 ( ( O at end 2 e 1 8 . . l l ( 9 d 9 k a a . 9 t con n l 1 e t t 1 ( a a s ( e e r t nsek a n z i t e e y y y e p k n d e e m l l x a e i i a u See notes l r o r r r t o Table 40—S nutrien S Studies of t C F B D B D G Studies of t Crepi Fox et al. (199

240 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program e e e our utive d the s for o d d 24-h h onsec nsecutiv nsecutiv nsecutiv t e m forms for a 5-day recor n u ent for 5 ent for 5 ons an d and/or o i t m c e l l o tions for 3 co tions for 5 co tions for 5 co tions for 10 c servati ed men ys ys c ned foo with par with par a a a days a b o a ll ll t tai erv erv erv erv a n ister D n utive utive da utive da ood Progra d our reca our reca In-care obs days Parent-mai 3 consec days consec In-care obs days 24-h In-care obs Self-admi recall consec In-care obs days 24-h perio In-care meal Care F nus trong” le of 85 ers in er in s nters in nters in t t 2 p eaker me d 94 e e sam ed l e Child and Adult p of 12 ce of 20 ce of 4 cen of 1 cen m a tinu S mple mple mple mple on ults a a a a centers an 6 centers with “ esentativ C pr ad g a d 6 centers with w ty ty i i ts— tin ence s ence s ence s ence s offered in th y care i i i i a a n n n ally re n cks Conve central Texas, menus, an Conve adult d particip Nation Boston are Conve Kansas C Kansas C Conve sna participan m a als and to RDAs, to RDAs to RDAs to RDAs to RDAs Guide d Food d tive tive tive tive ) , an a a a a s ed ed by progr ( ent of me es e t r take relative to in u um s s a , and Foo e t con M intake rel intake rel intake rel intake relative intake rel rient in ent ent ent ent cks con nutrien a ncy of foods serv Dietary Guidel CACFP Freque In-care nutri In-care nutri In-care nutri In-care nutri Guide Pyramid Pyramid Out-of-care nut RDAs, Dietary Guidelines Out-of-care nutrient intake relative to RDAs In-care nutrient Out-of-care nutrient intake relative to RDAs Out-of-care nutrient intake relative to RDAs eals and sn amined the ex t ribution of m 93) udies tha 993) t t e adult care component of h 91) 92) t con y d u t Table 40—S nutrien S Briley et al. (19 Drake (19 Drake (19 Glantz et al. (1983) Studies of t Ponza et al. (1

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 241 Chapter 7: Child and Adult Care Food Program

Design centers. Nonparticipating centers tended to offer a morning snack rather than a complete breakfast. The study used a quasi-experimental design to evaluate the effects of CCFP participation on the quality of Nutrient Density. Nutrient density scores were used to meals offered to children in child care. Data were col- assess the overall quality of the meals and snacks lected from a nationally representative sample of 100 offered.114 With the exception of the score for vitamin C, participating child care centers, 60 eligible nonpartici- nutrient density scores for meals and snacks offered in pating centers, and 62 participating family child care participating centers and homes were significantly higher homes. A one-day site visit was completed in each than scores for meals and snacks offered in nonparticipat- child care center and home. The preparation and serving centers. That is, participating centers and homes ice of one lunch and one snack were observed. Data on served meals and snacks that provided significantly the portions of food served to children were obtained greater amounts of nutrients, relative to food energy. using a “plate game,” a technique that obtained actual gram weights of the portions served to or taken by a Food Quality and Variety. Three daily menus were sample of six children. These data were used to impute scored using an index that reflected the quality of portion sizes for an analysis of three daily menus from foods served (for example, fresh fruits or vegetables a randomly selected week. vs. canned fruits or vegetables), as well as the variety of foods served within major food groups. Mean The main limitation of this design is the potential for scores on this index were significantly greater for par- selection bias; child care centers and homes that were ticipating centers and homes than for nonparticipating more concerned with nutrition (and therefore more centers. Differences were noted for both the quality likely to offer nutritious meals and snacks) may have and variety components of the index. been more likely to choose to participate in the CCFP than other child care providers. The analysis did not Limitations attempt to control for selection bias or for measured differences that may have existed between participat- Findings reported by Glantz and O’Neill-Fox (1982) ing and nonparticipating providers. Results of the should be treated with some caution because the data study are based on simple comparisons of the nutri- are now considerably out of date. Many market and tional characteristics of meals and snacks offered by legislative changes may have affected characteristics participating and nonparticipating providers. Data of foodservice programs in both participating and non- were tabulated separately for participating homes and participating centers and homes. Moreover, as just participating centers. noted, the study did not attempt to deal with potential selection bias or adjust for any differences in measured Findings characteristics that may have existed between the participant and nonparticipant groups. This allows for the Three measures were used to assess the nutritional possibility that the differences observed between par- characteristics of meals and snacks offered by partici- ticipating centers and homes and nonparticipating cen- pating and nonparticipating providers: nutrient content; ters were a reflection of a greater interest in or focus nutrient density; and food quality and variety. On each on nutrition among CCFP providers than among other of these measures, participating centers and homes providers. scored significantly higher than nonparticipating centers. Other Studies of the Child Care Component of the CACFP Nutrient Content. Compared with nonparticipating centers, both participating centers and homes offered All of the other studies identified for this review meals and snacks that provided a significantly greater examined the nutrient content of meals and snacks proportion of the Recommended Dietary Allowances offered to or consumed by participating children but (RDAs) for food energy and for all nutrients examined did not assess program impact—that is, the studies did except vitamins A and C.113 At least part of this differ- not include comparisons to nonparticipating institu- ence was due t CCFP centers and homes serving tions or children. breakfast much more frequently than nonparticipating

113Meals and snacks offered in participating centers and homes were 114Nutrient density scores were calculated for each nutrient examined using also higher in vitamins A and C. These differences, however, were not sta- the following equation: percent RDA for nutrient/percent RDA for food ener- tistically significant. gy. A score of less than 1 indicated a suboptimal ratio of nutrients to calories.

242 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program

CACFP regulations and guidance materials provide Nutrient Content of CACFP Meals and only broad standards for meals and snacks offered Snacks Offered to Participating Children under the program. In the absence of specific nutrient- Seven studies described the nutrient content of meals based standards, most of these descriptive studies used and snacks offered to participating children (table 40).115 the recommendations of the American Dietetic During periods when USDA allowed providers to be Association (ADA) (1994, 1999) as a benchmark for reimbursed for up to two meals and two snacks per assessing the nutrient content of CACFP meals and day for each child in care, most studies found that, on snacks. The ADA recommends that children in care for average, the combinations of meals and snacks com- 8 or more hours per day receive food that provides at monly offered to children provided them with the least one-half to two-thirds of their daily needs for opportunity to obtain at least 50 percent of their daily energy and nutrients (based on age-appropriate RDAs). energy and nutrient needs (Fox et al., 1997; Drake, In addition, the ADA recommends that meals and 1992; Glantz and O’Neill-Fox, 1982).116 Briley et al. snacks be consistent with the Dietary Guidelines for (1993) reported similar results but found that meals Americans. and snacks offered to children in full-time care fell short of the ADA’s recommendation for food energy, Findings from the identified studies must be interpreted iron, and niacin, a finding that may have been an arti- in light of shifting program policies regarding the max- fact of the way the data were analyzed.117 imum number of meals and snacks eligible for reimbursement. Before 1981, participating centers and homes Two studies conducted during a time when CACFP could be reimbursed for up to two meals and two snacks reimbursements were limited to two meals and one per day for each child in care. The previously described snack per child per day (Briley et al., 1989; Domer, evaluation of the CCFP (Glantz and O’Neill-Fox, 1982) 1983) found that the full complement of meals and collected data when this policy was in effect. snacks offered in participating centers provided less than 50 percent of the RDAs for food energy and sev- The 1981 Omnibus Budget Reconciliation Act (P.L. eral nutrients. 97-35) limited reimbursements to a maximum of two meals and one snack per child per day. Following this The most recent study of CACFP meals, based on data change, there was a marked reduction in the number of collected in 1999 and limited to tier II family child care child care centers that offered a morning snack (Glantz homes, found that, on average, the mean nutrient content et al., 1988). Three of the identified studies are based of the most common combinations of meals and snacks on data collected while this policy was in place (Briley offered (breakfast, lunch, and one snack and breakfast, et al., 1989; Domer, 1983; and Glantz et al., 1983). lunch, and two snacks) satisfied the ADA guidelines for full-time care (Crepinsek et al., 2002). Indeed, The policy governing maximum reimbursements was changed again, in 1988, when Congress allowed child care providers to be reimbursed for an additional meal 115Most of the available studies provide separate results for different or snack for children in care 8 or more hours per day meals (breakfast and lunch) as well as for snacks. This discussion is limited to findings related to the full complement of meals and snacks offered, (P.L. 100-435). Thus, findings from five of the most relative to the ADA recommendations. recent studies (Crepinsek and Burstein, 2004 (which 116Glantz and O’Neill-Fox (1982) reported that, on average, the total diets used data collected as part of the study reported on by (all meals and snacks combined) offered in participating child care centers contributed only 48 percent of the RDA for iron (which is just below the Fox et al., 1997); Fox et al., 1997; Briley et al., 1993; ADA recommendation of 50 percent for children in care 8 or more hours per and Drake, 1991 and 1992) reflect a program that day). However, methodology used in this study was different from that of all allowed providers to be reimbursed for up to two other studies examined here. In computing the mean percentage of the RDAs, meals and two snacks per child per day. levels exceeding 100 percent for any individual provider were truncated to 100 percent. While this approach had the intended effect of minimizing the effects of excessively high levels on the mean values, it also understated the Under the 1996 PRWORA reforms, the so-called true mean values. One can probably safely assume that, in the absence of this “fourth-meal provision” was eliminated again. Today, truncation, the true mean percentage of the RDA for iron would have exceed- CACFP providers can be reimbursed for a maximum ed the 50 percent benchmark established by the ADA. 117The study analyzed the nutrient content of the average menu offered in of one meal and two snacks or two meals and one snack each center but did not directly measure quantities of food offered or served. per child per day, regardless of how long the child is in Rather, the analysis assumed that centers served the amounts specified in pro- care. The most recent study of CACFP meals and gram regulations, which probably resulted in an underestimation of the energy and nutrient content of the meals offered. Fox et al. (1997) found that portion snacks (Crepinsek et al., 2002) collected data in 1999, sizes actually served to or taken by children are generally equivalent to or after the PRWORA change had been implemented. greater than the minimum portion sizes specified in the CACFP regulations.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 243 Chapter 7: Child and Adult Care Food Program researchers demonstrated that the mean nutrient content relative to energy contributions. While the two most of meals and snacks offered in 1999 did not differ sig- common combinations of meals and snacks provided nificantly from the mean nutrient content of meals and 61-71 percent of children’s daily energy needs, on snacks offered by similar providers in 1995 (based on average, they provided 68-75 percent of the recom- data from Fox et al., 1997).118 This was true despite the mended daily maximum of sodium. elimination of the “fourth meal” provision and reduction in reimbursement rates for tier II providers by PRWO- The most recent study of CACFP meals and snacks, RA. Findings may differ for other types of providers completed in 2002, examined meals and snacks offered (who were affected by the change in the maximum to children age 2 and older (Crepinsek et al., 2002). number of meals and snacks eligible for reimburse- Results were similar to those reported by Fox et al. in ment but not the change in reimbursement rate). 1997. For the two most common combinations of meals and snacks offered (the same combinations assessed by As noted, the ADA recommends that meals and snacks Fox et al., 1997), mean saturated fat content exceeded offered in the CACFP be consistent with the Dietary the Dietary Guidelines recommendation and mean Guidelines. Under current program regulations, CACFP sodium content was high, relative to energy content. As providers are not required to meet these standards. noted, however, this study included only tier II homes, Indeed, the applicability of the Dietary Guidelines to so findings may be different for other types of providers. the diets of children between the ages of 2 and 5 has been somewhat controversial over the years. Nutrient Content of CACFP Meals and Snacks Consumed by Participating Children The 1995 edition of the Dietary Guidelines stated that The nutrient profile of meals and snacks actually con- recommendations for total fat and saturated fat did not sumed by participating children may differ from the apply specifically to children between the ages of 2 nutrient profile of meals and snacks offered by and 5. Rather, the recommendation was that “after [2 providers. For example, children may decline one or years], children should gradually adopt a diet that, by more of the foods offered, children may select portions about 5 years of age, contains no more than 30 percent that differ from that of the average portion, or children of calories from fat” (USDA and U.S. Department of may waste (not consume) some of the food they take. Health and Human Services (HHS), 1995). Thus, to gain a full understanding of the contributions of CACFP meals and snacks to children’s energy and The most recent edition of the Dietary Guidelines, nutrient needs, one must examine the nutrient content released in 2000, takes a firmer stand on this issue and of CACFP meals and snacks actually consumed by states specifically that advice about intake of total fat, children. saturated fat, and cholesterol “applies to children who are 2 years of age and older” (USDA/HHS, 2000). As summarized in table 40, six studies examined the nutrient content of the meals and snacks consumed by The two largest and most recent studies of CACFP children while in care. Four of these studies have lim- meals and snacks compared the nutrient content of the ited generalizability because they used small conven- combinations of meals and snacks most commonly ience samples (Briley et al., 1993; Drake, 1992 and offered by CACFP providers with Dietary Guidelines 1991; and Glantz et al., 1983). and associated recommendations for total fat, saturated fat, cholesterol, and sodium. Fox and colleagues (1997) The most recent comprehensive study of the CACFP, the limited their analysis to meals and snacks offered to Early Childhood and Child Care Study (Fox et al., children age 5 and older; at the time, the 1995 edition 1997), included meal observations in a nationally repre- of the Dietary Guidelines was in effect. Results showed sentative sample of 372 participating child care centers that the two most common combinations of meals and and homes. Observations were completed for 1,347 chil- snacks offered to children in this age group met or dren between the ages of 1 and 10. Children generally approximated the Dietary Guidelines recommendation selected portions of food that were equivalent to or for fat but exceeded the recommendation for saturated greater than the minimum portion sizes specified in the fat. In addition, meals and snacks were high in sodium, CACFP meal pattern requirements and generally consumed between 70 and 80 percent of the portions taken.

118One significant difference was detected for the combination of breakfast, lunch, and two snacks. The mean energy content of this combination Among children in care 8 or more hours per day, of meals and snacks was greater in 1999 than 1995. CACFP meals and snacks provided about 50 percent

244 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program of daily needs for energy and iron.119 Intake of calci- greatly distorted. Key findings from this analysis, um from CACFP meals and snacks approximated, on which was not limited to children in full-time care, are average, three-quarters of the RDA. Average intakes of summarized here. protein, vitamin A, and vitamin C exceeded 100 percent of the RDA. For toddlers and preschoolers, CACFP meals and snacks contributed 36-47 percent of daily energy needs Comparisons of intakes to Dietary Guidelines recom- and 45 to more than 100 percent of the RDA for key mendations were limited to 5-year-olds because of the nutrients. CACFP meals and snacks made smaller con- 1995 Dietary Guidelines restriction and limited sam- tributions to the intakes of school-age children (ages 6- ples of older children in full-time care. Findings 10) because they spend fewer hours in care (typically showed that CACFP meals and snacks consumed by 3 hours per day). For food energy and iron, the these children provided more than the recommended CACFP contribution for 6-10-year-olds was less for amounts of both total fat and saturated fat. In addition, children who received care in centers than for children intake of sodium from CACFP meals and snacks was who received care in family child care homes. The high, relative to energy intake. authors attribute this difference to the fact that children receiving care in homes are likely to be offered break- The Early Childhood and Child Care Study also fast and lunch, whereas, children receiving care in cen- attempted to describe the relative contribution of ters are likely to be offered snacks (as reported by Fox CACFP meals and snacks to children’s 24-hour et al., 1997). intakes. The analysis was ultimately abandoned, however, because of low response rates. The study CACFP meals and snacks did not contribute dispro- methodology called for two 24-hour recalls on non- portionately to children’s daily intake of fat, saturated consecutive days made up of in-care observations for fat, or sodium, although total daily intakes of children foods consumed in care, as well as telephone inter- ages 3-10 exceeded recommendations for intake of all views with parents for foods consumed at home. these nutrients. CACFP meals and snacks provided Because of difficulties in reaching parents to complete more than the recommended level of saturated fat, as a the interview about at-home consumption within 48 percentage of total food energy. hours of the in-care observations, complete data were obtained for a relatively small percentage of the sam- Children’s 24-hour intakes of food energy, protein, ple (roughly 40 percent). vitamins A and C, calcium, iron, and zinc met or exceeded the RDAs. The one exception was energy Crepinsek and Burstein (2004) recently analyzed the intake among children ages 6-10 in child care centers. 24-hour recall data from the Early Childhood and For this group of children, mean daily intake of food Child Care Study. A nonresponse analysis revealed energy was equivalent to 87 percent of the RDA. that, although children with complete and incomplete dietary recall information were similar on a number of Review of Research on the Adult key variables, children with complete information were more likely than those with incomplete informa- Care Component of the CACFP tion to be from households with incomes equal to or To date, only one study has examined the adult day care greater than 185 percent of poverty. At the same time, component of the CACFP (Ponza et al., 1993). Although children with complete and incomplete information this descriptive study of the meals and snacks served by were quite similar with regard to the meals and snacks participating adult day care centers compared the char- they were observed to eat in care. Moreover, mean acteristics of participating and nonparticipating cen- intakes of food energy and key nutrients at each eating ters, it did not collect menu or dietary intake informa- occasion differed little between the two groups. The tion from nonparticipating centers. The study collected authors concluded that, with the use of sampling menu information for a 1-week period from a national- weights to correct for discrepancies between the two ly representative sample of 85 adult day care centers groups to the extent possible, the description of participating in the CACFP as well as information on CACFP contributions to total nutrient intakes based foods consumed over a single 24-hour period by a ran- solely on respondents with complete information is not dom sample of 942 adults attending these centers.

119This analysis was limited to children under the age of 6 because older The study did not analyze the nutrient content of the children were seldom in care 8 or more hours per day. meals and snacks served in participating centers. The

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 245 Chapter 7: Child and Adult Care Food Program analysis of menus was limited to tabulations of the Other available research is less outdated but is even types and frequencies of foods offered. The study more limited than the 1982 study in the sense that it described dietary intakes of CACFP participants and includes no representation of the conditions that would assessed the contribution of program meals to total exist in the absence of CACFP. Moreover, the most daily intake. The authors examined the percentage of recent comprehensive descriptive study of the child the RDA for food energy and key nutrients consumed care component of the program (Fox et al., 1997) is during the day as part of CACFP reimbursable meals based on data that were collected when providers were and snacks, as well as the percentage of total daily eligible to receive reimbursement for an additional intake supplied by CACFP meals and snacks.120 The meal or snack for children in care 8 or more hours per study also compared the composition of CACFP meals day. Subsequent research has indicated that this change and snacks consumed by participants with the Dietary had no impact on the number or nutritional quality of Guidelines for Americans. meals and snacks served in tier II homes (Crepinsek et al., 2002), but information is lacking for other types of On average, total nutrient intake from all CACFP providers. meals and snacks supplied 42 percent of the RDA for food energy and between 52 percent (iron) and 83 per- Similarly, the one study that has been completed on cent (vitamin C) of the RDAs for key nutrients. Taken the adult component of the program was entirely together, CACFP reimbursable meals and snacks con- descriptive in nature and offers no comparison of par- sumed by participating adults contributed about one- ticipants’ nutrient intakes to those that would exist in half of their total daily intake. the absence of the CACFP.

The study also found that CACFP meals and snacks Thus, the existing literature leaves unanswered the consumed by participants were not consistent with the fundamental question of how the CACFP affects the Dietary Guidelines for Americans. On average, the nutrient intake and other nutrition- and health-related percentage of food energy derived from fat (33 per- outcomes of participating children and adults. The cent) and saturated fat (11 percent) exceeded the rec- need for an answer to this question will become ommended levels of no more than 30 percent and less increasingly important as child care use patterns con- than 10 percent, respectively. tinue to evolve and the population continues to age.

Addressing the question of program impact will almost Summary certainly require new, special-purpose research. The Very little solid information exists concerning the impact national surveys that measure nutrient intake in detail, of the CACFP on the nutrition and health outcomes of such as the National Health and Nutrition Examination participating children and adults. Only one study has Survey (NHANES) and the Continuing Survey of attempted to compare CACFP conditions with the con- Food Intake of Individuals (CSFII), do not provide ditions that would exist in the absence of the program. reliable indicators of institutional participation in the That study (Glantz and O’Neill-Fox, 1982) provides CACFP and do not even ask about meals that might be some evidence that the program improves the quality consumed in family child care homes. Thus, identify- of meals and snacks served to children in participating ing a valid sample of CACFP and non-CACFP partici- child care centers and homes, but the evidence has pants from these existing databases is unlikely. substantial limitations. First, it is quite dated. Since the In addition to collecting primary data, future studies of time that this study was completed, program regula- the CACFP should do the following: tions, program participation, and national patterns of child care use have changed significantly. Second, the • Use dietary assessment methods that measure usual study looked at only the characteristics of meals energy and nutrient intake of participating children offered and did not examine children’s actual intakes and/or adults, children and adults attending nonpar- from CACFP meals and snacks. Finally, the study ticipating day care programs, and, perhaps, children design is potentially vulnerable to selection bias. and adults who are not in care.

• Look at meals and snacks offered by participating 120Program regulations limit CACFP reimbursements to a maximum of two meals and one snack per day. Some adults in participating centers con- and nonparticipating programs and the foods actual- sume an additional snack while in care that is not reimbursable. ly consumed by attending children and adults.

246 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program

• Examine the impact of the CACFP on the total special attention will need to be paid to the methodolo- dietary intake of participants over the full day, gy used to collect complete dietary recall information including food consumed in care and food consumed for children.121 outside of care.

Given the problems encountered in the Early 121This problem has been encountered in other studies that used a com- Childhood and Child Care Study—low response rates parable methodology. For example, the first School Nutrition Dietary because of difficulties in obtaining information from Assessment Study (SNDA-I) (Burghardt et al., 1993, chapter five) achieved response rates of 30-35 percent for students in grades 1 and 2 parents about food consumed at home to couple with where parent interviews were used to collect information on out-of-school information obtained during in-care observations— food consumption for a single 24-hour period.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 247 Chapter 7: Child and Adult Care Food Program

References Drake, M.A. 1991. “Anthropometry, Biochemical Iron Indexes, and Energy Nutrient Intake of Preschool American Dietetic Association. 1999. “Position of the Children: Comparison of Intake at Daycare Center and American Dietetic Association: Nutrition Standards for at Home,” Journal of the American Dietetic Child Care Programs,” Journal of the American Association 91:1587. Dietetic Association 99(8):981-88. Fox, M.K, and M.J. Cutler. 1996. The Child Nutrition American Dietetic Association. 1994. “Position of the Homeless Demonstration Project: Year 3 Report. American Dietetic Association: Nutrition Standards for Cambridge, MA: Abt Associates Inc. Child Care Programs,” Journal of the American Dietetic Association 94:323-28. Fox, M.K., F.B. Glantz, L. Geitz, et al. 1997. Early Childhood and Child Care Study: Nutritional Briley, M.E., A.C. Buller, C.R. Roberts-Gray, et al. 1989. Assessment of the CACFP. Volume II. Final Report. “What Is on the Menu at the Child Care Center?” USDA, Food and Consumer Service. Journal of the American Dietetic Association 89(6):771-74. Garnett, S. 2003. “Child and Adult Care Food Program (CACFP): Update on For-Profit Center Eligibility.” Briley, M.E., C.R. Roberts-Gray, S.T. Jastrow, and J. Available: http://www.fns.usda.gov/cnd/care/ Vickers. 1999. “Dietary Intake at The Child-Care Regs&Policy/ForProfits/extension.htm. Accessed Center and Away: Are Parents and Care Providers April 2003. Working as Partners or at Cross-Purposes?” Journal of the American Dietetic Association 99:950-54. Glantz, F., D.T. Rodda, M.J. Cutler, et al. 1997. Early Childhood and Child Care Study: Profile of Briley, M.E., C.R. Roberts-Gray, and S. Rowe. 1993. Participants in the CACFP. Volume I. Final Report. “What Can Children Learn from the Menu at the Child USDA, Food and Consumer Service. Care Center?” Journal of Community Health 18(6):363-73. Glantz, F., J.A. Layer, and M. Battaglia. 1988. Study of the Child Care Food Program. Cambridge, MA: Abt Burghardt, J., A. Gordon, N. Chapman, et al. 1993. Associates Inc. The School Nutrition Dietary Assessment Study: School Food Service, Meals Offered, and Dietary Glantz, F., N. Goodrich, D. Wagner, et al. 1983. Intakes. USDA, Food and Nutrition Service. Evaluation of the Child Care Food Program: Results of the Child Impact Study Telephone Survey and Pilot Crepinsek, M.K, and N. Burstein. 2004. Maternal Study. Cambridge, MA: Abt Associates Inc. Employment and Children’s Nutrition: Volumes I and II. E-FAN-04-006. USDA, Economic Research Service. Glantz, F., and M.K. O’Neill-Fox. 1982. Evaluation of the Child Care Food Program: Final Report on the Crepinsek, M.K., N.R. Burstein, E.B. Lee, et al. 2002. Congressionally Mandated Studies. Volume I. Meals Offered by Tier 2 CACFP Family Child Care Cambridge, MA: Abt Associates Inc. Providers—Effects of Lower Meal Reimbursements: A Report to Congress on the Family Child Care Homes Hamilton, W., N. Burstein, and M.K. Crepinsek. 2001. Legislative Changes Study. E-FAN-02-006. USDA, Reimbursement Tiering in the CACFP: Summary Economic Research Service. Report of the Family Child Care Homes Legislative Changes Study. FANRR-22. USDA, Economic Domer, J.A. 1983. “Nutrition in a Private Day Care Research Service. Center,” Journal of the American Dietetic Association 82:290-93. Macro International. 1991. Study of the Child Nutrition Homeless Demonstration: Final Report - Year 1. Drake, M.A. 1992. “Menu Evaluation, Nutrient Intake USDA, Food and Nutrition Service. of Young Children, and Nutrition Knowledge of Menu Planners in Child Care Centers in Missouri,” Journal Ponza, M., J. Burghardt, R. Cohen, et al. 1993. of Nutritional Education 24:145-48. National Study of the Adult Component of the Child and Adult Care Food Program (CACFP). Final Report. Volumes I and II. USDA, Food and Nutrition Service.

248 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 7: Child and Adult Care Food Program

U.S. Department of Agriculture, Food and Nutrition U.S. Department of Agriculture, Food and Nutrition Service. 2003. Program data. Available http://www. Service. 2002c. “Child and Adult Care Food Program: fns.usda.gov/pd. Accessed April 2003. Facts About Reimbursement for Suppers Available to Afterschool Care Programs.” Available: http://www. U.S. Department of Agriculture, Food and Nutrition fns.usda.gov/cnd/care/Regs&Policy/snacks/supper- Service. 2003a. “Facts about the Child and Adult Care faqs.htm. Accessed March 2002. Food Program.” Available: http://www.fns.usda.gov/ cnd/care/CACFP/cacfpfaqs.htm. Accessed March 2003. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 2000. Nutrition and U.S. Department of Agriculture, Food and Nutrition Your Health: Dietary Guidelines for Americans, 5th Service. 2002b. “Child Nutrition Programs: Afterschool edition. Home and Garden Bulletin No. 232. Snacks Fact Sheet.” Available http://www.fns.usda. gov/cnd/governance/AfterschoolFactSheet.htm. U.S. Department of Agriculture and U.S. Department Accessed March 2002. of Health and Human Services. 1995. Nutrition and Your Health: Dietary Guidelines for Americans, 4th edition. Home and Garden Bulletin No. 232.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 249 Chapter 8 Summer Food Service Program

The Summer Food Service Program (SFSP) provides (USDA), Food and Nutrition Service (FNS), 2002a). funds to eligible organizations to serve nutritious Open sites are those located in areas where at least 50 meals and snacks, free of charge, to children at percent of the children are from households with approved feeding sites. The program operates mainly incomes at or below 185 percent of poverty (making during the summer when schools are not in session them eligible for free or reduced-price meals in the and the National School Lunch Program (NSLP) and NSLP or SBP). Open sites are required to be open to School Breakfast Program (SBP) are not available. provide food to all children in the neighborhood, Organizations eligible to sponsor feeding sites include regardless of their enrollment in site-sponsored activi- public or private nonprofit schools; local government ties. Enrolled sites are those in which 50 percent of the agencies; nonprofit community organizations, such as children enrolled in a program or activity offered at YMCAs, Boys and Girls Clubs, churches, National the site are eligible to receive free or reduced-price Youth Sports Programs (NYSP), and residential meals, based on individual applications. Camp sites camps.122 Because the SFSP is an entitlement pro- are residential summer camps. Camp sites receive gram, no eligible sponsor may be denied funding. reimbursement only for meals served to children whose documented household income makes them eli- Research on the SFSP has been entirely descriptive in gible for free or reduced-price meals. nature. Much of it has focused on describing program operations and characteristics of sponsoring organiza- Children up to age 18 or older, if they participate in a tions. Two national studies of the program (Gordon program for mentally or physically handicapped indi- and Briefel, 2003; Ohls et al., 1988) assessed the nutri- viduals, are eligible to receive meals. Lunch is, by far, ent content of meals served in the SFSP but did not the most frequently served meal in the SFSP. However, examine the contribution of SFSP meals to students’ sponsors may also offer breakfast, supper, and/or snacks. daily nutrient intakes or make any comparison to nutri- Most children receive one or two meals per day. ent intakes of nonparticipants. Residential camps and sites that serve migrant children may serve (and be reimbursed for) up to three meals. To receive Federal reimbursement, SFSP sites must Program Overview serve meals and snacks that meet defined meal pat- The SFSP was created to ensure that low-income terns, similar to those used in the NSLP (see chapter 5) children would have access to nutritionally balanced and the SBP (see chapter 6). meals when school is not in session. The program was created in 1968 as a 3-year pilot project and was Sponsors receive two types of reimbursement for permanently authorized as an entitlement program in each meal served, and reimbursement rates vary by FY 1975. type of meal. The largest reimbursement is for operating or foodservice costs ($2.35 per lunch or supper In most States, the SFSP is administered by State edu- served in the summer of 2003). Sponsors also receive cation agencies, the entities that oversee the NSLP and an additional per meal reimbursement to cover admin- SBP. Locally, the SFSP is operated by approved spon- istrative costs ($0.2475 per lunch or supper in self- soring organizations, which include school districts, preparation or rural sites and $0.2050 per lunch or State or local government agencies, churches, private supper in all other sites). Funds received through these nonprofit residential camps, and community organiza- two reimbursement streams are not fungible and spon- tions. Sponsors provide free meals at one or more sors must monitor their costs very closely to ensure feeding sites. that reimbursements fully cover their costs (USDA/FNS, 2002b). Feeding sites may be either “open sites,” “enrolled sites,” or camps (U.S. Department of Agriculture In FY 2002, the SFSP operated in approximately 30,000 feeding sites nationwide and served about 122NYSPs are Federally funded sports camps for low-income children. 122 million meals (USDA/FNS, 2003). During Programs are administered by colleges and universities. peak operation in July 2002, the program served

250 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 8: Summer Food Service Program approximately 1.9 million children per day.123 The 30 percent (Food Research and Action Center 124 total FY 2002 Federal cost for the SFSP was $263 (FRAC), 1993). million. The 1996 Personal Responsibility and Work Program History Opportunity Reconciliation Act (PRWORA) included many amendments to SFSP operations. Although most During its first year of operation (FY 1975), the SFSP amendments were designed to increase the efficiency served meals to an average of 1.79 million children of program administration, PRWORA also included each day. Over the next 2 years, the program grew to language that removed program expansion as a stated serve more than 2.8 million children per day. Program goal, reduced per meal cash subsidies by roughly 10 growth was sharply curtailed in 1981, however, when percent, and eliminated the subsidy for a fourth meal the 1981 Omnibus Reconciliation Act (OBRA) elimi- that had previously been provided to some sponsors. nated private nonprofit sponsors other than schools The GAO studied the effects of these changes and and residential camps. This action was taken because a reported no deleterious effect on the number of partici- 1977 report issued by the U.S. General Accounting pating sponsors and children (Robinson, 1998). State- Office (GAO) described extensive program abuses by level administrators did report, however, that the pro- these sponsors (Ohls et al., 1988). In addition, the gram may have been affected in other ways—for OBRA legislation restricted use of foodservice man- example, a reduction in the number of feeding sites, agement companies and other vendors, expanded pro- meaning children may have had to travel further to get gram monitoring and administration, and tightened eli- to a site, and/or in the number of food items provided gibility requirements. Prior to OBRA, the criteria used to participating children (Robinson, 1998). to define area eligibility for feeding sites were 30 percent of children from low-income households. OBRA In the years since PRWORA, concerns about the num- increased this threshold to the 50-percent standard that ber of low-income children who go without Federal is currently in use. meal benefits during the summer have continued to escalate. In describing the problem, Under Secretary of After implementation of the OBRA reforms, SFSP par- Agriculture Eric M. Bost pointed out that the 2 million ticipation decreased substantially. In 1985, the pro- SFSP meals served per day in FY 2000 represented gram served 1.5 million children per day, roughly half only about 12 percent of the free and reduced-price as many as had been served in 1977. The precipitous meals served each day during the regular school year decrease in participation led to renewed concerns, par- through the NSLP (Bost, 2000). Bost deemed this level ticularly among child welfare advocacy groups, that of SFSP participation, which reached “only a fraction low-income children were going without needed nour- of eligible children,” to be “unreasonably low.”125 ishment during the summer. Advocates concerned about rural hunger raised concerns that rural areas had In keeping with this overarching concern, the most particularly low participation rates and greater barriers recent legislative changes in the SFSP have focused on to participation by sponsor organizations (Shotland increasing program penetration by attracting more pro- and Loonin, 1988). gram sponsors, particularly school districts. In 2000, USDA implemented several changes designed to elim- In the mid- to late 1980s, several pieces of legislation inate and streamline paperwork requirements for spon- were passed with the aim of increasing children’s sors. In addition, in late 2000, P.L. 106-554 (The access to and participation in the SFSP. In 1989, the Consolidated Appropriations Act) authorized a special OBRA restriction on private nonprofit organizations was reversed and these organizations were again 124The Food and Nutrition Service (FNS) studied private, nonprofit eligible to serve as program sponsors. This change sponsors after this change was implemented and concluded that the change resulted in an increase in the number of feeding was successful in leading to an overall increase in program participation as sites available and, consequently, the number of well as to increased coverage of rural areas. The study also found that pri- children served. Between 1989—when the change vate, nonprofit sponsors administered the program as effectively as public sponsors (Decker et al., 1990). went into effect—and 1993, the number of 125There are several reasons that SFSP participation is lower than NSLP children served by the SFSP increased by about participation. One is that open SFSP sites must be located in low-income neighborhoods, whereas the NSLP is available everywhere. Another is that attendance at SFSP sites is voluntary, while children must attend school 123An additional 1.6 million children per day received summer meals during the year (Gordon and Briefel, 2003). In addition, systems that trans- through the NSLP as part of summer school programs or year-round schools port students to schools during the normal school year are generally not (based on reported NSLP participation for July 2002 (USDA/FNS, 2003)). operational during the summer.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 251 Chapter 8: Summer Food Service Program pilot project to increase the number of children partici- school districts must be approved by their State agencies pating in the SFSP in Puerto Rico and 13 States with and must qualify as either an open site or an enrolled low SFSP participation rates (Garnett, 2001; FRAC, site under SFSP regulations (see previous discussion). 2001).126 The pilot project was initially authorized to operate from FY 2001 through FY 2003 and was Tasse and Ohls (2003) studied early reaction to and extended by Congress through March 31, 2004. It sim- effects of seamless waivers. They reported that about plified recordkeeping and reporting requirements and 540 school districts, or about 14 percent of all SFSP provided sites with the maximum per-meal reimburse- sponsors, operated the program with a seamless waiver ment for both operating (foodservice) and administra- in summer 2002. Although school district response to tive cost reimbursements. Moreover, pilot sites were the waiver was generally positive, early evidence indi- allowed to use funds from each reimbursement stream cated that the waiver had a limited impact on the num- to cover excess costs associated with the other reim- ber of children receiving summer meals. In summer bursement stream. (As described previously, under 2002, only about 21 percent of the sponsors using the current program regulations, reimbursements for oper- waiver were new to the SFSP, and not all of these new ating and administrative costs are strictly separate.) sponsors entered the program because of the waiver. Moreover, average daily participation rates were sub- Analysis by FRAC (FRAC, 2003) and USDA’s Food stantially lower for new sponsors than for seamless and Nutrition Service (Singh and Endahl, 2004) indi- waiver sponsors as a whole (531 children per day vs. cates that States participating in the pilot successfully 972 children per day). Tasse and Ohls (2003) estimat- increased SFSP participation. Singh and Endahl found ed that on a typical day in summer 2002, about 50,000 that, in all 14 States combined, the number of SFSP children received meals who would not have done so sponsors increased by 18 percent between July 2000 without the seamless waiver. A decision about the ulti- and July 2003, and average daily participation mate success of the waiver will require information increased by 43 percent. Impacts varied substantially about impacts during summer 2003 and summer 2004. across States, however, and based on July 2003 data, many pilot States continued to have low SFSP partici- Other actions taken by USDA to increase SFSP spon- pation relative to other States. Evaluation of the pilot sorship include providing State agencies with the flexi- impacts was complicated by other SFSP initiatives bility to approve deviations in the length of time (described below) that were implemented during the between meal services and/or the duration of meal same period. service, when existing requirements pose a barrier to participation, and to consider closed, enrolled sites that Before the start of SFSP activities for summer 2002, provide services exclusively to the “Upward Bound” USDA implemented several regulatory changes designed program as categorically eligible for the SFSP. to facilitate program participation at the sponsor level, (Income-eligibility thresholds used for “Upward thereby increasing the number of children reached Bound” are identical to those used in the SFSP.) through the SFSP. The most significant change was the Finally, USDA developed a Web-based geographic nationwide implementation of “seamless summer information tool to help State agencies and other inter- waivers” for school districts that operate the NSLP ested organizations identify areas that are underserved (USDA/FNS, 2002c). The waivers, which will run by the SFSP (Gordon and Briefel, 2003).127 through FY 2004, allow school districts to offer the SFSP without having to deal with previously required paperwork and administrative requirements. School Research Review districts operate the SFSP in essentially the same manner Research on the SFSP has focused on issues related to as they operate the NSLP. All meals served at waiver program participation and operations rather than on sites are claimed as NSLP meals and are reimbursed at impacts. The most recent study of the SFSP, which was the NSLP free meal rate, which is slightly lower than the based on data collected in summer 2001, was completed SFSP rate. However, program administrators do not have in 2003 (Gordon and Briefel, 2003). The objectives of to deal with the administrative burden associated with the study were to provide information about the charac- operating two different programs. To receive a waiver, teristics of the SFSP and its operations at the State, sponsor, and site levels, to assess factors that affect participation of both sponsors and children, and to assess 126The 13 States are Alaska, Arkansas, Idaho, Indiana, Iowa, Kansas, Kentucky, Nebraska, New Hampshire, North Dakota, Oklahoma, Texas, and Wyoming. 127Available at www.ers.usda.gov/data/SFSP/.

252 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 8: Summer Food Service Program nutritional quality and plate waste of SFSP meals. The SFSP sites served meals to children in a variety of set- study did not look at the contribution of SFSP meals to tings. Most (76 percent) served meals indoors, and most children’s overall nutrient intake, or make any compar- (70 percent) used a serving line or food pickup line. isons to eligible, nonparticipating children. Nutrient analysis of randomly observed plates indicated that SFSP breakfasts, as served, provided more than one- Data were collected from all SFSP State administrators quarter of the 1989 Recommended Dietary Allowances as well as from nationally representative samples of (RDA) for protein, vitamin A, iron, calcium, and vitamin sponsors and sites; 162 feeding sites were visited in C and about 21 percent of the Recommended Energy person. In addition to in-person interviews, site visits Allowance (REA). SFSP lunches provided, on average, included structured observations of site characteristics more than one-third of the RDA for these key nutri- and operations, the types and amounts of food served ents, as well as approximately one-third of the REA. on 5 or 10 randomly selected plates, and the types and amounts of food wasted on 10 randomly selected SFSP breakfasts exceeded the Dietary Guidelines rec- plates. Lunch was always observed. If breakfast or ommendation for saturated fat content and SFSP supper were offered, one of these meals was observed lunches exceeded Dietary Guidelines recommenda- as well (snacks were not observed). tions for both total fat and saturated fat. Study authors reported that, overall, nutrient profiles of SFSP break- The study found that school districts made up 48 percent fasts and lunches were similar to those reported for of all SFSP sponsors in summer 2001 and served 51 breakfasts served in the SBP and lunches served in the percent of all SFSP meals. School districts were found NSLP (as reported in the second School Nutrition to be well-suited to serve as SFSP sponsors because Dietary Assessment Study, Fox et al., 2001). they have experience preparing and serving meals to children and have available buildings and staff. Observations of plate waste indicated that children wasted about one-third of the calories and nutrients Of all SFSP sponsors, 14 percent were government served at both breakfast and lunch.128 Waste varied agencies, generally municipal recreation or social serv- across sites and for different foods. Vitamin A at lunch ice departments. Although fewer in number, overall, was found to have the highest level of waste (53 per- government agencies were the largest sponsors, operat- cent) because of a high rate of waste for vegetables. ing 36 percent of all feeding sites and serving 31 per- Findings from the plate waste analysis were similar to cent of all SFSP meals in summer 2001. Government those reported in the previous national study of the agencies frequently used vendors to provide meals SFSP (Ohls et al., 1988), as well as in a study of plate because they lacked the facilities and/or expertise to waste in the NSLP (Reger et al., 1996). prepare meals themselves. Other Studies of the SFSP More than 8 out of 10 (83 percent) SFSP feeding sites were open sites, 14 percent were enrolled sites, and 3 Other SFSP studies have been undertaken mainly by percent were residential camps. All sites served lunch, advocacy organizations. FRAC publishes annual status 49 percent also served breakfast, 19 percent served a reports that consolidate data on SFSP program partici- snack, and only 5 percent served supper. Almost all pation by State. The reports also highlight best prac- sites (93 percent) offered some type of activities for tices, summarize new regulations, and provide other information of use to current and prospective sponsor children, including educational activities, supervised th free play, organized games or sports, arts and crafts, organizations. The 11 report in the series was pub- field trips, and swimming. lished in June 2003 and summarizes data for summer 2002 (FRAC, 2003). More than half (58 percent) of all SFSP participants in summer 2001 were elementary school children, 20 In 1995, the FRAC report also included results of a sur- percent were middle school age, 17 percent were pre- vey of 5,282 heads of households designed “to provide school age, and 5 percent were high school age. reliable information on the extent of childhood hunger in Children were racially and ethnically diverse: 39 percent were African American, 29 percent were non- 129An earlier version of the same survey was completed by Wehler et al. Hispanic White, 27 percent were Hispanic, and 5 per- (1991). 130A two-stage probability sample design was used to select census cent were Asian, American Indian, or members of block groups. All households were enumerated and screened to determine another racial/ethnic group. income and presence of children under the age of 12.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 253 Chapter 8: Summer Food Service Program the U.S.” (FRAC, 1995).129 The survey, the Community status. About 15 percent of “hungry” families partici- Childhood Hunger Identification Project (CCHIP), docu- pated in the SFSP compared with10 percent of “at mented food insufficiency among low-income families risk” families and 9 percent of “not hungry” families. with children, examined families’ attempts to cope with food insufficiency and hunger, and described conse- Summary quences of hunger for children. The survey also examined the role FANPs play in helping low-income house- The impact of the SFSP on participants’ nutrition and holds deal with food insufficiency and hunger. Low- health status has not been studied. Ongoing efforts to income households with at least one child under the age expand SFSP availability are continuing and, at least of 12 were randomly sampled in 11 different geographic in the short term, research related to the SFSP is likely locations nationwide.130 Interviews were conducted face- to focus on the effectiveness of these initiatives. to-face with the person responsible for care and feeding of the children. The recent descriptive study of the SFSP provides a solid understanding of the operations and characteris- A major finding of the study was that 71 percent of tics of the SFSP at the State, sponsor, and site levels low-income families with at least one child under the (Gordon and Briefel, 2003). The next step in evaluat- age of 12 had never heard of the SFSP. The authors ing the SFSP is to examine how eligible children who argued that low participation in the SFSP was probably do not participate in the SFSP fare during the summer. due to lack of awareness among target families. In USDA’s Food and Nutrition Service is currently under- addition, participation in the SFSP and in other child- taking a qualitative study to examine this issue. oriented FANPs was found to vary by food security Ultimately, an impact study must include detailed assessment of both SFSP participants and nonparticipants. Such a study will face several implementation challenges, including the short timeframe during 128Estimates of plate waste included only foods that were served to or which the SFSP operates (6-8 weeks), as well as ana- selected by children but not eaten (that is, some portion of the food remained lytical challenges related to selection bias. However, on the plate after children were through with their meal). Estimates did not questions about the nutrition and health impacts of the include food that might have been left in or taken from “share boxes,” designated places where children could leave food they did not want to eat or take SFSP can be addressed only with a study that looks at food left by other children. Share boxes were available at 44 percent of sites. both participants and income-eligible nonparticipants.

254 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 8: Summer Food Service Program

References Reger, C.C., E. O’Neil, T.A. Nicklas, et al. 1996. “Plate Waste of School Lunches Served to Children in Bost, E.M. 2000. “Hunger Doesn’t Take a Vacation.” a Low-Socioeconomic Elementary School in South Message from Under Secretary of Agriculture, Eric M. Louisiana,” School Food Service Research Review Bost, posted on Summer Food Service Program web- 20(supplement):13-19. site. Available: http://www.fns.usda.gov/snd/summer/ states/message.html. Accessed March 2002. Robinson, R.A. 1998. Effects of changes made to the Summer Food Service Program. Testimony before the Decker, P.T., E.E. Kisker, J. Patch, et al. 1990. An Subcommittee on Early Childhood, Youth and Evaluation of the Summer Food Service Demonstration. Families, Committee on Education and the Workforce, Final Report. Princeton, NJ: Mathematica Policy U.S. House of Representatives. Washington, DC: U.S. Research, Inc. General Accounting Office.

Food Research and Action Center. 2003. Hunger Shotland, J., and D. Loonin. 1988. Patterns of Risk: Doesn’t Take a Vacation: Summer Nutrition Status The Nutritional Status of the Rural Poor. Washington, Report, June 2003. Washington, DC. DC: Public Voice for Food and Health Policy.

Food Research and Action Center. 2001. New Pilot Singh, A., and J. Endahl. 2004. Evaluation of the 14 Makes Summer Food Participation Easier in 13 States State Summer Food Service Program Pilot Project. and Puerto Rico. Available: http://www.frac.org/ USDA, Food and Nutrition Service. html/news/sfsp_pilots.html. Accessed March 2002. Tasse, T., and J. Ohls. 2003. Reaching More Hungry Food Research and Action Center. 1995. Hunger Children: The Seamless Summer Food Waiver. Trends Doesn’t Take a Vacation: A Status Report on the in Nutrition Policy. Issue Brief No. 1. Princeton, NJ: Summer Food Service Program for Children, Third Mathematica Policy Research. edition. Washington, DC. U.S. Department of Agriculture, Food and Nutrition Food Research and Action Center. 1993. New Service. 2003. Program data. Available: http://www. Opportunities: A Status Report on the Summer Food fns.usda.gov/pd. Accessed April 2003. Service Program for Children. Washington, DC. U.S. Department of Agriculture, Food and Nutrition Fox, M.K., M.K. Crepinsek, P. Connor, et al. 2001. Service. 2002a. “Summer Food Service Program: School Nutrition Dietary Assessment Study II Frequently Asked Questions.” Available: http://www. (SNDA-II): Summary of Findings. USDA, Food fns.usda.gov/cnd/summer/about/faq.html). Accessed and Nutrition Service. March 2002.

Garnett, S.C. 2001. Implementation of the Summer U.S. Department of Agriculture, Food and Nutrition Food Service Program (SFSP) Pilot Projects Service. 2002b. “Summer Food Service Program: Authorized by the Consolidated Appropriations Act, Sponsors.” Available: http://www.fns.usda.gov/cnd/ 2001. Technical memorandum issued to FNS regions summer/sponsors/index.html). Accessed March 2002. and State agencies, January 19, 2001. U.S. Department of Agriculture, Food and Nutrition Gordon, A., and R. Briefel. 2003. Feeding Low-Income Service. 2002c. “USDA News Release: USDA Plans Children When School is Out—The Summer Food To Continue Improving Children’s Access to the Service Program: Executive Summary. FANRR-30. Summer Food Service Program.” Available: USDA, Economic Research Service. http://www.fns.usda.gov/news/releases/2002/03/0082. htm. Accessed March 2002. Ohls, J., E. Cavin, E. Kisker, et al. 1988. An Evaluation of the Summer Food Service Program: Wehler, C.A., R.I. Scott, and J.J. Anderson. 1991. A Final Report. FANRR-30. USDA, Food and Nutrition Survey of Childhood Hunger in the United States. Service. Community Childhood Hunger Identification Project. Washington, DC: Food Research and Action Center.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 255 Chapter 9 The Emergency Food Assistance Program

The Emergency Food Assistance Program (TEFAP) available commodities to States, based on poverty provides commodity foods to emergency kitchens rates, unemployment rates, and related factors. Finally, (often referred to as soup kitchens), homeless shelters, PRWORA established a requirement that $100 million and similar organizations that serve meals to the from annual Food Stamp Act appropriations be used to homeless and other needy individuals. Through food purchase commodities for TEFAP. banks and food pantries, the program also provides basic commodities to low-income households for TEFAP foods are distributed free of charge. However, preparation and consumption at home. The U.S. individuals who receive TEFAP foods for home use Department of Agriculture (USDA) purchases com- must meet eligibility criteria defined by each State. modity foods and processes, packages, and distributes The types of commodities available through TEFAP them to designated State agencies, which, in turn, dis- vary from year to year, depending on agricultural contribute them to approved local charitable organizations. ditions as well as State preferences. In FY 2001, more than 40 products were available, including canned and To date, there has been no direct evaluation of TEFAP’s dried fruits, canned vegetables, fruit juice, canned meat, effects on nutrition- and health-related outcomes. A poultry, and fish, dried egg mix, peanut butter, nonfat small number of studies have examined the characteris- dry milk, rice, pasta, and cereal (USDA, Food and tics of people who are likely to receive TEFAP foods. Nutrition Service (FNS), 2003a). In FY 2002, 611 million pounds of food were distributed through TEFAP, at a Federal cost of $435 million (USDA/FNS, 2003b). Program Overview TEFAP evolved from the Federal Surplus Relief A recently completed study of providers in the Corporation, which was established under the Emergency Food Assistance System (EFAS) found Agricultural Adjustment Act of 1933. It was reautho- that TEFAP commodities account for about 14 percent rized as the Federal Surplus Commodities Corporation of all food distributed through the EFAS (Ohls and (FSCC) under Section 32 of The Potato Control Act of Saleem-Ismail, 2002). Nationally, 55 percent of emer- 1935. From its inception, the program pursued parallel gency kitchens, 52 percent of food pantries, and 84 goals of reducing Federal food inventories and storage percent of food banks distribute TEFAP foods. costs (associated with farm price supports) and assist- ing needy households. Research Review The current program was first authorized as the The literature search identified no direct evaluations of Temporary Emergency Food Assistance Program in TEFAP’s effects on nutrition- or health-related out- 1981. In 1988, when Federal stocks of some surplus comes. The potentially relevant literature includes a foods were depleted, the Hunger Prevention Act of small number of studies that describe recipients of 1988 authorized the purchase of commodities specifi- TEFAP food either explicitly or implicitly. These stud- cally for TEFAP. These commodities are in addition to ies are summarized briefly in the sections that follow. any surplus commodities donated to TEFAP by USDA. The name associated with the acronym TEFAP was Characteristics of TEFAP Recipients changed to The Emergency Food Assistance Program TEFAP recipients are generally poor and food insecure under the 1990 Farm Act. and tend to be demographically similar to the low- income population overall (Quality Planning The Personal Responsibility and Work Opportunity Corporation and Abel, Daft, and Earley, 1987). A Reconciliation Act of 1996 (PRWORA), the initial dimension often used to differentiate TEFAP recipients welfare reform law, made several important changes in is whether they receive food for preparation and con- TEFAP. First, TEFAP was combined with the previously sumption at home or consume food available on the separate Commodity Distribution Programs for premises of prepared meal programs (Briefel et al., Charitable Institutions, Soup Kitchens, and Food Banks. 2003: Clancy et al., 1991; Bowering et al., 1991). This Second, PRWORA defined a formula for allocating distinction largely reflects the portion of recipients

256 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 9: The Emergency Food Assistance Program with a home vs. those who are homeless or whose resi- According to the EFAS survey, food pantries served dences lack facilities for storing or preparing food. 4.3 million different households, comprising 12.5 mil- Recipients who obtain TEFAP food for preparation at lion people, during a typical month in 2001. Emergency home normally get the food from food pantries or simi- kitchens served about 1.1 million people per month in lar distribution facilities, while those who obtain food 2001 (Briefel et al., 2003). Data on the characteristics through prepared meal programs normally get it at of these EFAS recipients confirm findings from earlier emergency kitchens, shelters, or other residential pro- studies and show that pantries and emergency kitchens grams, such as drug treatment or detoxification centers serve different segments of the low-income population. or battered women’s shelters. Households that used food pantries were more likely to A national study of TEFAP conducted in 1986 for include children than households that used emergency USDA by the Quality Planning Corporation (QPC) and kitchens. Almost half (45 percent) of food pantry Abel, Daft, and Earley (ADE) included a survey of households included children, compared with about 20 TEFAP recipients. At the time this study was complet- percent of emergency kitchen households (Briefel et ed, TEFAP did not include commodity distribution to al., 2003). Almost 30 percent of pantry households charitable institutions.131 Consequently, recipients were single adults living alone; more than two-thirds characterized in the study were those who met State- of these single adults were female. In contrast, more defined income-eligibility criteria and received com- than 50 percent of emergency kitchen households were modities for use at home. Such recipients are only a adults living alone, and more than 70 percent of these subset of the individuals served by TEFAP today. single adults were male. Among pantry clients, the majority racial/ethnic group was non-Hispanic White A 1988 study by Burt and Cohen provides information (49 percent), followed by non-Hispanic Black (31 per- on the characteristics of individuals who receive meals cent), Hispanic (16 percent) and other (5 percent). through emergency food kitchens, homeless shelters, Among food pantry clients, the greatest concentration and other charitable organizations. Clancy et al. (1991) of clients were non-Hispanic Black (45 percent), fol- and Bowering et al. (1991) reported results from 1990 lowed by non-Hispanic White (35 percent). surveys of New York State food pantry and emergency kitchen users, respectively. At the time these studies About a quarter of pantry clients and 40 percent of were conducted, these institutions received their USDA emergency kitchen clients were unemployed and look- commodities from the then-separate Commodity ing for work. Employment status tracked with house- Distribution Program for Charitable Institutions, Soup hold composition, with households with children being Kitchens, and Food Banks. more likely to have employed members than households without children. Education was a limiting factor The most up-to-date and complete information on the for both groups of EFAS clients, particularly those characteristics of likely TEFAP recipients comes from using food pantries. Close to 46 percent of pantry the recently completed study of the EFAS (Ohls and clients and 39 percent of emergency kitchen clients Saleem-Ismail, 2002). That study included a survey of had less than a high school education. EFAS clients served by both arms of the system—food pantries and emergency food kitchens (Briefel et al., Both groups of EFAS clients were poor. At the time they 2003). A randomly selected and nationally representa- were interviewed, 93 percent of pantry clients and 83 tive sample of 2,397 food pantry clients and 2,425 percent of emergency kitchen clients had incomes at or emergency kitchen clients were interviewed in person below 130 percent of the Federal poverty guideline. between August and November, 2001. Overall response One-third and two-fifths, respectively, had incomes rates were 70 percent and 77 percent, respectively. that were at or below 50 percent of the poverty line. In Major findings from this survey are discussed below. addition to poverty, both sets of EFAS clients experi- Findings from the two older surveys described above, enced problems with homelessness or other challenges are discussed when they provide information or a per- to daily living, such as lack of adequate food storage/ spective that is not available in the EFAS survey. cooking facilities, transportation, or working telephones. Clients of emergency kitchens were most likely to be homeless (36 percent vs. 8 percent of pantry clients). 131The Commodity Distribution Program for Charitable Institutions, Soup Kitchens, and Food Banks was a separate program until 1996 when About half of pantry clients and two-fifths of emer- PRWORA merged it into TEFAP. gency kitchen clients in the EFAS survey reported

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 257 Chapter 9: The Emergency Food Assistance Program being in “fair” or “poor” health. (This compares with between participants in the two programs was age. The one-third of the general low-income population) percentage of elderly households in TEFAP was more (Briefel et al., 2003). The earlier survey of homeless than twice that of the FSP (38 percent vs. approxi- people using emergency food assistance completed by mately 15 percent). As a potential explanation for this Burt and Cohen (1988) found that there was a relative- disparity, Levedahl et al. (1994) cite findings from two ly high prevalence of mental health problems, with 20 studies of barriers to FSP participation among the eld- percent reporting a history of mental hospitalization, erly (Ponza, 1990; Ponza and Wray, 1990). Many eld- about 20 percent reporting at least one suicide attempt, erly apparently prefer TEFAP to the FSP because and 7 percent having been diagnosed as suffering from TEFAP’s application and distribution procedures are a major psychiatric problem. On a scale measuring less complicated (Ponza 1990, as cited in Levedahl et current depression and demoralization, 49 percent had al., 1994). In addition, some elderly are resistant to high enough psychological distress scores to indicate a FSP participation because FSP coupons clearly identi- need for immediate treatment. fy users as welfare recipients (Ponza and Wray, 1990, as cited in Levedahl et al., 1994). More than half (55 percent) of the pantry-client households surveyed in 2001 visited a pantry once per month Nutrition-Related Characteristics or less often (many providers restrict the frequency of of Likely TEFAP Recipients visits) (Briefel et al., 2003). About a quarter reported Briefel and her colleagues (2003) assessed EFAS visiting two or three times per month, and about 20 clients’ food security status using the 6-item short ver- percent visited once per week or more often. Among sion of the core food security module developed by users of emergency kitchens, about 13 percent received USDA (Bickel et al., 2000). They found that about their meals at the kitchen every day. Another 43 per- three-quarters of EFAS clients were food insecure, and cent received meals 2 to 5 days per week. Clients who almost half were classified as food insecure with visited kitchens almost daily tended to rely on the hunger. This compares with a national estimate of food kitchen for multiple months; sometimes for years. insecurity of 11 percent overall and 32 percent for About three-quarters of pantry clients and more than low-income households (Nord et al., 2002). The preva- two-thirds (69 percent) of kitchen clients surveyed in lence of food insecurity varied by household composi- 2001 said they preferred to get food from a pantry or tion. The households with the greatest rates of food kitchen than “ask the Government for help” (although security were those with seniors and no children. many of these households reported relying on other Analyses showed that material hardships and food types of Federal assistance) (Briefel et al., 2003). insecurity were more severe among EFAS client More than two-thirds (69 percent) of food pantry households that used two or more forms of emergency clients and almost half (45 percent) of emergency food assistance than for EFAS client households that kitchen clients combined use of emergency food assis- combined emergency food assistance with participa- tance with participation in another food and nutrition tion in other FANPs. Nonetheless, three-quarters of assistance program (FANPs). However, although most households that combined EFAS services with FANP EFAS clients (90 percent of food pantry clients and 82 participation experienced food insecurity. percent of emergency kitchen clients) were income-eligible for the FSP, a substantial proportion of eligible Burt and Cohen (1988) examined the eating patterns of households did not participate (45 percent of pantry- homeless people who used emergency kitchens and client households and 56 percent of emergency kitchen shelters and compared them with homeless people who clients). Uncertainty about eligibility was the reason did not use these services. To obtain information on most commonly given for not participating. homeless people who did not use emergency kitchen or shelter services, the researchers interviewed home- Among food pantry clients, households that included less individuals at congregating sites—for example, seniors (adults 60 and older) but no children were more bus stations, culverts, and other homeless “encamp- likely than other types of households to use only food ments.” This sample of nonservice-using homeless pantries to obtain food assistance. The 1986 study of individuals—those who had not used any kind of shel- TEFAP recipients (analogous to today’s EFAS pantry ter or emergency kitchen for the past week—was small clients) compared the demographic profile of TEFAP and not necessarily representative. Although not gener- recipients with a profile of FSP recipients during the alizable, the information obtained provides useful same period and found that the primary difference insights about differences between the service-using

258 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 9: The Emergency Food Assistance Program and nonservice-using homeless (akin to TEFAP recipi- Summary ents and eligible nonrecipients). No research completed to date has examined the Generally, all homeless people sampled tended to eat impact of TEFAP on nutrition- and health-related out- less frequent and less adequate meals than the overall comes of program participants. The recently completed population. Homeless people who did not use either surveys of providers in the EFAS (Ohls and Saleem- emergency kitchens or shelters fared worse than those Ismail, 2002) and EFAS clients in food pantries and who did use these services. Homeless nonusers ate an emergency kitchens (Briefel et al., 2003) provide average of 1.36 meals per day compared with 1.92 researchers and policymakers with a detailed and up- meals per day for homeless service users and 3 or to-date picture of the organizational system and pro- more meals per day for the average low-income per- grams that distribute TEFAP foods and the characteris- son. Homeless nonusers were also more likely to have tics and experiences of likely recipients of TEFAP gone 1 or more days without eating during the previ- foods. These data provide a solid foundation for future ous week than homeless service users, averaging 1.35 research that may examine nutrition and health charac- days per week without eating compared with 0.66 days teristics of TEFAP recipients and, potentially, the per week for homeless service users. Homeless influence of the program on these characteristics. nonusers were more likely than homeless service users to describe their diets as fair or poor and less likely to However, any evaluation of the effects of TEFAP at have eaten foods from five core food groups. the participant level will face some formidable challenges. First, TEFAP foods compose only part of the The researchers did not assess the nutrient content of package delivered by the participating programs and, the meals actually consumed by homeless persons or because the package delivered often depends on volun- their total nutrient intake over the course of the day.132 tary contributions, its content generally fluctuates over Available information on the number of meals eaten time. Moreover, individuals tend to receive these foods per day and the number of days in which nothing was on an episodic basis, often for only a single instance. eaten strongly suggests that substantial differences Second, many recipients of TEFAP foods obtain and may exist between the food served in shelters and consume food from a wider-than-normal variety of emergency kitchens and the food actually consumed sources, ranging from supermarkets to food pantries, routinely by homeless persons. prepared and perishable food recovery programs, emergency kitchens, shelters, family, friends, panhan- dling, dumpsters, and garbage cans. Usual approaches to measuring food consumption may not be effective for some TEFAP recipients because of their unusual circumstances. Finally, in addition to the problems of 132The QED/ADE study (1987) included an assessment of the potential nutrient contribution of an average TEFAP package to the diets of regular defining the TEFAP intervention and measuring poten- program participants. The packages were found to provide significant tial outcomes, it will be difficult to construct an appro- amounts of protein and key vitamins and minerals but were high in saturated priate representation of the counterfactual—that is, fat, cholesterol, and sodium. At the time that this analysis was done, only defining and accessing an appropriate comparison/con- seven commodity foods were offered and cheese was a major component of the package. TEFAP food packages offered today have considerably more trol group of eligible individuals who do not receive variety and have been designed to be lower in fat, cholesterol, and sodium. TEFAP foods.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 259 Chapter 9: The Emergency Food Assistance Program

References Ohls, J., and F. Saleem-Ismail. 2002. The Emergency Food Assistance System—Findings From the Provider Bickel, G., M. Nord, C. Price, et al. 2000. Guide to Survey, Volume I: Executive Summary. FANRR-16-1. Measuring Household Food Security: Revised 2000. USDA, Economic Research Service. USDA, Food and Nutrition Service. Ponza, M. 1990. “The Effectiveness of USDA Food Bowering, J., K.L. Clancy, and J. Poppendieck. 1991. Assistance Programs in Meeting the Food and Nutrition “Characteristics of a Random Sample of Emergency Needs of the Low-Income Elderly,” Food Stamp Policy Food Program Users in New York: II. Soup Kitchens,” Issues: Results from Recent Research. Paper presented at American Journal of Public Health 81(7):914-17. the Food and Nutrition Service Research Conference in Washington, DC, February 22, 1990. Briefel, R., J. Jacobson, N. Clusen, et al. 2003. The Emergency Food Assistance System—Findings From Ponza, M., and L. Wray. 1990. Evaluation of the Food the Client Survey: Executive Summary. FANRR-32. Assistance Needs of the Low-Income Elderly and Their USDA, Economic Research Service. Participation in USDA Programs. USDA, Food and Nutrition Service. Burt, M.A., and B.E. Cohen. 1988. Feeding the Homeless: Does the Prepared Meals Provision Help? Quality Planning Corporation and Abel, Daft, and A Report to Congress on the Prepared Meal Provision, Earley. 1987. A Study of the Temporary Emergency Volumes I and II. Washington, DC: The Urban Food Assistance Program (TEFAP). USDA, Food and Institute. Nutrition Service.

Clancy, K.L., J. Bowering, and J. Poppendieck. 1991. U.S. Department of Agriculture, Food and Nutrition “Characteristics of a Random Sample of Emergency Service. 2003a. “Food Distribution Fact Sheet: The Food Program Users in New York: I. Food Pantries,” Emergency Food Assistance Program.” Available: American Journal of Public Health 81(7):911-14. http://www.fns.usda.gov/fdd/programs/tefap/ pfs-tefap.pdf. Accessed April 2003. Levedahl, J.W., N. Ballenger, and C. Harold. 1994. Comparing the Emergency Food Assistance Program U.S. Department of Agriculture, Food and Nutrition and the Food Stamp Program: Recipient Service. 2003b. Program data. Available: http://www. Characteristics, Market Effects, and Benefit/Cost fns.usda.gov/pd. Accessed April 2003. Ratios. AER-689. USDA, Economic Research Service.

Nord, M., M. Andrews, and S. Carlson. 2002. Household Food Security in the United States, 2001. FANRR-29. USDA, Economic Research Service.

260 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10 Nutrition Services Incentive Program

The Nutrition Services Incentive Program (NSIP), for- toward the cost of the meal, but meals are free to those merly known as the Nutrition Program for the Elderly who cannot make any contribution. (NPE), is a U.S. Department of Agriculture (USDA) program that provides cash and/or commodities to The program is available to all age-eligible individuals, agencies or organizations that sponsor Elderly regardless of household income. However, the goal is Nutrition Program (ENP) sites. The ENP, which is to target those with the greatest nutritional, social, or administered by the U.S. Department of Health and economic need, particularly low-income minorities Human Services (HHS), Administration on Aging and elders living in rural areas (Wellman et al., 2002). (AoA), is the primary vehicle for the organization and Because the program is not means-tested, the ENP is delivery of nutrition and support services to the the primary service system for elders whose incomes Nation’s elderly. The program provides meals in both may be slightly greater than the income-eligibility group and home settings. Although any person over requirements used for other programs, such as the the age of 60 is eligible to participate, local programs Food Stamp Program or the Commodity Supplemental try to target elders with the greatest nutritional and/or Food Program (Wellman et al., 2002). social needs. In recent years, the home-delivered meals component of the program has grown dramatically, The ENP is administered by a network of agencies reflecting an increase in the number of frail, home- devoted to the aging population, including State and bound elderly. Indian Tribal Organization (ITOs) units on aging, within-State area agencies on aging, and local delivery sites.134 The program has grown substantially since its Program Overview inception. In 1975, the program provided 48.5 million meals (HHS/AoA, 2002). In 1980, ENP providers The ENP was designed specifically to address prob- served 168.4 million meals, an increase of almost 250 lems of inadequate dietary intake and social isolation percent. The program continued to grow during the among the elderly. It began as a 3-year pilot program 1980s and 1990s, although at a slower pace. In FY in 1968 and was permanently authorized in 1972, as a 2001, the ENP served 253 million meals (USDA, Food new title of the Older Americans Act (OAA). In enact- and Nutrition Service (FNS), 2003).135 ing the program, Congress cited “an acute need for national policy which provides older Americans, par- Much of the increase in the ENP during the 1980s and ticularly those with low incomes, with low-cost, nutri- 1990s was in home-delivered meals. In 1978, a pro- tionally sound meals served in strategically located gram regulation that limited home-delivered meals to centers ... Besides promoting better health ... such a 10 percent of total meals was rescinded and separate program would reduce the isolation of old age, offer- authorizations were established for home-delivered ing Americans an opportunity to live their remaining and congregate meals (HHS/AoA, 2002). years in dignity” (P.L. 92-258, Section 701). Subsequently, the relative size of the home-delivered meals component of the program began to increase The ENP provides daily meals to people age 60 and steadily. In 1980, home-delivered meals represented 22 over in group settings (congregate feeding programs) percent of all ENP meals. In 1991, home-delivered and, when appropriate, at home (Meals-on-Wheels).133 meals accounted for 43 percent of all ENP meals. By Spouses of age-eligible individuals are also eligible to the end of the decade, home-delivered meals had participate, regardless of age. In addition, disabled increased to more than half (54 percent) of all ENP people who live in elderly housing facilities, people meals (HHS/AoA, 2002). This trend reflects an who accompany elderly participants to congregate feeding sites, and volunteers who assist in the meal 134Operation of the ENP by ITOs was authorized separately in 1978 service may also receive meals through the ENP. Each under Title VI of the OAA. States and ITOs continue to be authorized recipient may contribute as much as he or she wishes under separate titles of the OAA, but ENP sites in both settings operate under the same program regulations. 135USDA stopped maintaining data on the number of meals served in 133To be eligible for home-delivered meals, participants must be home- the ENP after FY 2001 because of changes in the program’s administrative bound or otherwise isolated (Wellman et al., 2002). structure, as described in subsequent sections.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 261 increased need for home-delivered meals as well as the Historically, lunch has been the focal point of the ENP, availability of increased funding. Even with this dra- and most congregate and home-delivered meal programs matic growth, many ENP sites have waiting lists for served lunch only 5 days per week. As the program has home-delivered meals (Ponza et al., 1996). matured, however, local providers have incorporated service innovations that have allowed them to better USDA’s involvement in the ENP began in 1975 when meet participants’ needs. In a 1988 survey of 450 ENP Congress authorized USDA to donate commodities to project sites, Balsam and Rogers (1991) found that many the ENP. The USDA program, known as the Nutrition projects had expanded well beyond serving only lunch, Program for the Elderly (NPE), provided commodities particularly with regard to home-delivered meals. Half to States and ITOs, which, in turn, distributed them to of the sites providing home-delivered meals offered local ENP sites. In 1977, P.L. 95-65 allowed States and meals on weekends and one in five offered supper. ITOs to elect to receive their NPE entitlement in the Comparable statistics for congregate meal sites were form of cash rather than commodities. Over time, the 17 percent and 10 percent, respectively. Other innova- predominant type of support provided by the NPE tions reported by Balsam and Rogers included contract- shifted from commodities to cash. In FY 1999, only 2 ing with restaurants and diners to provide meals, exclu- percent of the $140 million NPE appropriation was sively targeting meals served at a given site to a partic- distributed to ENP meal providers as commodities ular racial/ethnic group, and regularly scheduled visits (HHS/AoA, 2002). to congregate feeding sites by nursing home residents.

When the ENP was reauthorized in FY 2000, the name ENP sites have also developed noteworthy approaches for the USDA program was changed to the Nutrition to maximizing available Federal funding in order to Services Incentive Program (NSIP). In addition, the serve more elders and provide them with needed serv- model for administration of the program was changed ices. The most recent national evaluation of the ENP from a simple reimbursement model to an allocation “estimated that government funding investments in the model. Rather than reimbursing States and ITOs on a ENP were tripled by the program’s innovative use of per meal basis based on the number of meals served volunteers, the collection of contributions by elders to the previous fiscal year, NSIP funds are now distrib- the costs of meals, and the supplementation of Federal uted to States and ITOs based on the number of meals resources with State grants and private donations” served relative to the total number of meals served by (Balsam et al., 2000). all States and ITOs. The reason for this change was a desire to reward States and ITOs for efficient use of ENP funds can also be used for nutrition education and cash and/or commodities in providing meals to older other appropriate services (O’Shaughnessy, 1990). adults (USDA/FNS, 2002). Over time, the ENP has become an integral component of a comprehensive and coordinated system of home- In FY 2003, responsibility for the administration of the and community-based services (HCBC) (Wellman et NSIP was transferred from USDA to HHS, although al., 2002). Services provided by ENP sites include USDA continues to provide financial support and transportation, shopping assistance, health screenings, donated commodities. In FY 2002, USDA’s contribu- wellness programs, information and referral services, tion to the ENP was $152 million (USDA/FNS, 2003). and recreational and social activities.

Program Services Nutrition Standards for ENP Meals ENP providers are required to offer participants at In the early 1990s, concerns were raised about the least one “hot or other appropriate” meal per day 5 nutritional integrity of the ENP. During the OAA reau- or more days per week. Providers may elect to provide thorization hearings in 1992, several professional additional meals. Congregate meal sites must be groups involved in the ENP, including the American located in close proximity to areas with large concen- Dietetic Association and the National Association of trations of elderly residents and, to the extent possible, Nutrition and Aging Service Programs, encouraged be within walking distance of participants’ homes. Congress to incorporate minimum standards for nutri- When feasible, programs provide transportation for tion services provided under the OAA. The majority of participants who are unable to travel to the meal site the recommendations made in the hearings were ulti- on their own. Home-delivered meals can be either hot mately incorporated into law as part of the 1992 or cold. Amendments to the OAA (P.L. 102-375).

262 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program

Specifically, the 1992 Amendments stipulated that Opinion Research Corporation (ORC). The study was ENP meals must comply with the Dietary Guidelines done in two waves, with data collected in 1976-77 for Americans (DGAs) and provide the following: (reported in 1979) and in 1982 (reported in 1983). In this partially longitudinal design, 42 percent of wave I • A minimum of one-third of the Recommended participants were re-interviewed in wave II. Dietary Allowances (RDAs) if one meal per day is offered. The most recent national evaluation, the National Evaluation of the Elderly Nutrition Program, 1993-95, • A minimum of two-thirds of the RDAs if two meals is the most comprehensive evaluation of the ENP com- per day are offered. pleted to date (Ponza et al., 1996). The evaluation focused largely on dietary intake, although the social • 100 percent of the RDAs if three meals per day are support aspect of the program was also assessed. ENP offered. participants were compared with the elderly U.S. population in general, using data from the third National These standards represent a substantial change from Health and Nutrition Examination Survey (NHANES- previous practice. Before 1992, some States encouraged III), as well as with eligible nonparticipants, identified ENP sites to consider the DGAs, but neither Federal nor through Medicare beneficiary data. State guidelines required that ENP meals be consistent with the DGAs. With regard to the RDA standards, the Of the 11 local studies that attempted to measure the 1992 regulations shifted the focus from the individual impact of ENP participation on nutrition and health meal to the total meal package. Previous regulations outcomes, 8 (Group IIA in table 41) looked at congre- required that each meal supply one-third of the RDA, gate meals and 3 (Group IIB) looked at home-deliv- regardless of the type of meal or the total number of ered meals. Sample sizes for all of these studies were meals offered. The switch to standards that considered substantially smaller than those of the nationally repre- the total meal package provided more flexibility in meal sentative studies. Four studies had samples of less than planning because it allowed program planners to distrib- 100 (Gilbride et al., 1998; Steele and Bryan, 1986; ute nutrients across multiple meals as long as the total LeClerc and Thornbury, 1983; Singleton et al., 1980). combination of meals offered provided participants with Samples for the remaining seven studies ranged from the opportunity to consume specified levels of nutrients. 135 to 547.

Research Overview Identifying Nonparticipant Comparison Groups No one has studied the effectiveness of the NSIP (or All of the impact studies completed to date have used the former NPE), per se. To understand the impact of quasi-experimental designs. Most studies compared the NSIP, one has to look to research on the larger pro- program participants with a similar group of eligible gram, the ENP. The literature search identified two nonparticipants at a single point in time. Nearly all of nationally representative studies of the ENP as well as the studies defined program participants as those who 11 smaller local studies.136 Characteristics of these ate an ENP meal during the preceding 24-hour period. studies are summarized in table 41. Studies are divided into three groups. Group I includes national evalua- Researchers have used several different methods to tions, Group IIA includes local studies that focused on identify nonparticipant comparison groups and have had congregate ENP programs, and Group IIB includes varying degrees of success in establishing comparability local studies that examined the home-delivered meals between groups. Many of the local studies identified component of the ENP. nonparticipants from program waiting lists. While this approach may seem like a reasonable way to minimize The first national evaluation of the ENP was conduct- potential selection bias, it may lead to problems with the ed for the AoA by Kirschner Associates, Inc., and comparability of treatment and control groups. Not all ENP sites, particularly those that serve congregate meals, 136Studies that assessed the nutrient content of ENP meals and/or the contribution of ENP meals to the nutrient intake of participants—without have waiting lists. Sites that do have waiting lists and comparison to nonparticipants—are not included in table 41 or the summa- the individuals included on those lists may differ from ry tables presented later in this section, but have contributed to this review. sites that do not have waiting lists and the individuals These sources include Stevens et al. (1992), Vaughan and Manore (1988), Grandjean et al. (1981), Caliendo and Smith (1981), Caliendo (1980), and who participate in those sites. Moreover, individuals Caliendo and Batcher (1980). on waiting lists may be different from those receiving

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 263 Chapter 10: Nutrition Services Incentive Program ; n od h s Continued— gressio as te re a n bi ucted ucted Analysis met Multivari attempted to control for selectio Chi-square test No statistical tests cond No statistical tests cond rily y y ing a meal ll day on on a a v i e ll d ll d a a e it) t necess ed ENP o etary reca Measure of participation (did n consum Receiv on di Ate ENP meal dietary rec Ate ENP meal dietary rec ENP meals Currently rec outcomes s l e t vs. t vs. t vs. t vs. pant pant pant pant n n n n pari- ants stil d in egate sit le Design n and health partici partici partici partici Participa non Participa non and com sons to Wave I particip enrol congr Participa non Participa non ) e n 40) latio ly ly ly ly (n= ,699) ,411) ,563) 2 3 4 Popu (sample siz ENP-eligible elder (n= ENP-eligible elder (n= ENP-eligible elder (n= ENP-eligible elder ition Program on nutritio y ion ion y tive n r t t a a erson -p ola ola eta s, and n etary etary etary d rly Nutr e scree ng Initi nd i w nd is nd is method ncy, 5-d our di on our di our di our di Elde Data collection 24-h recall a intervie index 24-h recall a index 24-h recall a checklist recalls, food freque food recor level- from Nutrition Screeni 2 24-h the f ) - k 1 0 e ) s y) l 1 r d d y s d e m- m- rs e 7 g t d r 9 7 t o n n e e d t o n a r i i - Y e onl i n a pact o e s t c artici- c i , v c i ants’ usin f d w 76-7 i metro artici- ivered), t i l P m s regate meals e r e e n l r np d e N a n e o t N d ng ndom sa ndom sa p e E s egate a egate a - b a ipants in ipants in om samp ly ho n f b om sampl 2) omly e c c h h omly sel- s d t a t o t g e e i i m me-del n l t r s e Data sources e o o a c l o a e i from HCFA Medi- c file (1993-95) p and ra ple of no Parti former participants (19 rand e Rand of ENP participants (both congr h Parti h n rand selected ENP sites (both congr of particip rand (congregate and ra p (198 pants’ neighbo Residents in HUD elder facilities i p City; nonp pants from facili- t have ENP (da not reported l e e e e a amined the i udies of co n n n o i ex t i t ial r t izatio izatio u n d Outcome(s) n Dietary intak and soc contacts Dietary intak and social Dietary intak and social risk Dietary intak a of table. te and local st udies tha t at end n n n - n - o o i i l h h n n I: National evaluations IIA: Sta i i a atio atio tion of p p a 6) 8) O O 3-95) ve I (1979) ve II (1983) d d a a See notes n n Table 41—S Study Group Ponza et al. (199 (Nation Evalu the ENP— 199 Kirschner and Associates a Researc Corpor W Kirschner and Associates a Researc Corpor W Group Gilbride et al. (199

264 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program od h s and ance ance ance Continued— nalysis of s and a l vari e is of vari is of vari uare test Analysis met Analysis of Chi-sq analys Multifactoria varianc Bivariate t-test analys inued t - Con on week onths 3-5 on y wice ek 5 times da rtici- d a e per ek, but at ek lar p g last 4 m n: Ate at ENP n: Ate at ENP Measure of participation Ate ENP meal food recor patio site less than t per we least onc patio meal site 2- per we Irregular partici durin Regu Ate ENP meal at least 1 food record day Ate ENP meal times per we outcomes— t vs. t vs. t vs. t vs. pant pant pant pant n n n n Design n and health partici partici partici partici Participa non Participa non Participa non Participa non ) ) ) 5 e 0 5 n 2 3 3 1 53) = = latio n n d ( ( come y y l ly, over 7 l ly (n= n r r 185) Popu e e d (sample siz d l l ENP-eligible e (n= ENP-eligible elder years ol ENP-eligible, low-i elder ENP-eligible e ition Program on nutritio tr rd rds od ous o o ll, food ht and ds blo ht, and le, ig od rec rec e d d nous method ncy, ven t, weig us blo ht our reca h d samp Elderly Nu Data collection sample 1-day foo and ve 1-day food record, 24-h freque bloo heig tricep skinfol 3-day foo 3-day food record, veno sample, h weig the f d e s 6 1 es l t 1 nter ; ts ) pact o d) ts in ts an pants pants ts in 6 75) pants n in m n n n ea (da s (dates r nia Maine e ipants in c partici partici partici Data sources Participa ENP sites in Missouri; non from senior ce meals (19 that did not serv Participa from senior center that did not serve mea ENP sites in Missouri; non (dates not reported) nonparticipan from federally- subsidized housing units in same a not reported at 9 ENP sites in 2 norther Califor counti not reporte non Participa Parti ENP site central l e a amined the i n o i ex t i t r t nal mistries msitries u e e ht status, ht status, n d Outcome(s) n Iron intake and iron status bioch Dietary intake, weig a Dietary intake, weig nutritio bioch Dietary intak of table. udies tha t s n al. at end i d r n a a N 2) ury - c r a e 6) 3) k l j a C See notes z e Table 41—S Study Nordstrom et al. (198 Neyman et (199 C et al. (1987) L Thornb (198

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 265 Chapter 10: Nutrition Services Incentive Program n od h s and ance ance ance Continued— gressio vari te re of vari a is of vari uare test Analysis met Chi-sq analys Multivari Analysis Analysis of inued t y ing - east Con week onths on on a ek v y wice i e ll d 5 times da rtici- a d a e per ek, but at ek lar p g last 4 m n: Ate at ENP n: Ate at ENP Measure of participation patio meal site 2- per we Irregular partici patio site less than t per we least onc Regu durin ENP meals at l 2 times per we Ate ENP meal food recor Ate ENP meal dietary rec Currently rec outcomes— t vs. t vs. t vs. t vs. pant pant pant pant n n n n Design n and health partici partici partici partici Participa non Participa Participa Participa non non non ) ) ) s e 7 6 e n 4 6 5 4 derly nd = = latio n n ( ( come bou y y l l ly femal n etic el r r 97) 154) Popu e e d d (sample siz l l ENP-eligible e ENP-eligible, low-i (n= elder (n= ENP-eligible, home diab ENP-eligible e ition Program on nutritio tr of rd iew ous y v o ll, food ds ht, and le, rec etary d nts’ n inter il surve method ncy, ven nde t, weig our reca our di h d samp Elderly Nu Data collection 1-day food record, 24-h freque bloo heig tricep skinfol In-perso 24-h recall 1-day foo and ma respo physicians the f e m s s s m 6 7 6 s l 1 p- l l l d l a e ndo pact o non- rk ts in ts in ts in m pants pants pants o m n n n d g list n ants fro na; etic reci e r om samp ia e 5) 6-87) 3) partici partici partici v home-delivered meals i l Data sources e Participa from senior center that did not serve mea ENP sites in Missouri; non (197 sample of particip Participa Participa in New Y State and ra a waitin (198 from 2 senior centers that di not serve mea from senior center that did not serve mea ENP sites in souther Louis non (dates not reported) ENP sites in Missouri; non (197 Rand of diab ients of home- d l e e a amined the i udies of ty, i n o i ex t etic rsity, i t r e t mistries u iab e ht status, n d Outcome(s) n bioch Dietary intake, weig a Dietary intak Dietary intak Food secur diet div and d control of table. te and local st udies tha t al. at end IIB: Sta eton et al. 0) 0) 8) 8) See notes Table 41—S Study Kohrs et al. (198 Singl (198 Kohrs et al. (197 Group Edwards et (199

266 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program n od h s s and gressio te re a Analysis met Bivariate t-test multivari Bivariate t-test inued t ing ing r Con r v v i i e e e e Measure of participation ENP meals Currently rec 1 ENP meal p day, 5 days p week Currently rec outcomes— t vs. t vs. pant pant n n Design n and health partici partici Participa non Participa non ) ) e 8 n 4 4 54) nd nd = latio n ( bou bou y l ly (n= r Popu e d (sample siz l ENP-eligible, home elder ENP-eligible, home e ition Program on nutritio tr ep iet ht, etary etary ig e nd d method ht, and tric our di our di Elderly Nu Data collection 24-h recall a history 24-h recall, h weig skinfolds the f s e s t t 1 i d d e s s i t e e g r r a w ) 1 pact o e e rolina; d e ( v v pants pants i i m l l m s 2-83) o e e r m f d d ents of ents of i i - - a rts. s r e e l partici partici g a m m o e Data sources r o o Recip Recip h h list (198 from a waitin in North Ca non not reported from waiting l York State; non meals in N m for other p e e amined the i collection effo t h ex t g i data e w d Outcome(s) n Dietary intak a Dietary intak status primary udies tha t 86) nd g 9) All studies were 1 Table 41—S Study Ho-San (198 Steele a Bryan (19

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 267 Chapter 10: Nutrition Services Incentive Program meals because of the criteria sites use to determine Research Results who gets on the list and, once on the list, who gets served first (Ponza et al., 1996). This section summarizes findings from the available research. The discussion is organized into six sections, Neither of the national evaluations of the ENP used each of which focuses on reported effects of the ENP waiting lists to define comparison groups. The on a different outcome or group of outcomes. The out- Kirschner/ORC study (1979, 1983) drew nonpartici- comes examined include intake of food energy and pants from neighbors of participants and former partic- nutrients, nutritional biochemistries, weight status, ipants. The 1993-95 National Evaluation of the ENP socialization, food security, and nutritional risk. (Ponza et al., 1996) used Medicare beneficiary files to identify eligible nonparticipants and then contacted All of the studies that compared the nutrient content of them by telephone to screen for age, income, disability ENP with the minimum Federal requirement of one- status, and program participation. third of the RDA (per meal) found that ENP meals served to participants satisfied this standard (Ponza et Outcomes Examined al., 1996; Stevens et al., 1992; Kohrs, 1986; Kirschner/ORC, 1983; Caliendo, 1980; and Kohrs et The nutrition- and health-related outcome most often al., 1978). Thus, one can assume that participants gen- examined in this literature is dietary intake. Only a erally had access to the nutritional benefit the ENP few studies, all of which were local studies, examined was designed to deliver. nutritional biochemistries or weight status (Neyman et al., 1996; Ho-Sang, 1989; Czajka-Narins et al., 1987; Impacts on Intake of Food Nordstrom et al., 1982; Kohrs et al., 1980). The two Energy and Nutrients most recent studies (Ponza et al., 1996; Gilbride et al., 1998) included a general measure of nutritional risk. Most studies that examined dietary outcomes used a However, only Gilbride et al. (1998) compared partici- single 24-hour recall. Comparisons between participants and nonparticipants on this measure and no sta- pants and nonparticipants were based on mean intakes, tistical tests were conducted. One study examined the most often expressed as proportions of the RDAs. impact of the ENP on food security (Edwards et al., In addition to the usual problems with 24-hour recall 1993). And finally, the two national evaluations data and comparisons to RDA benchmarks (see (Kirschner/ORC, 1979, 1983; Ponza et al., 1996) chapter 2), use of the RDA in assessing intakes of eld- included assessment of social interaction. erly persons presents unique problems (Dwyer and Limitations Mayer, 1997; Ponza et al., 1996; Ponza et al., 1994; Posner, 1979). The RDAs, as they existed at the time Many of the identified studies included only simple the reviewed research was conducted, provided a sin- bivariate comparisons of participant and nonparticipant gle recommendation for all males over the age of 51 groups. Although most authors attempted to demon- and a corresponding recommendation for all females strate comparability of participant and nonparticipant over the age of 51 (National Research Council (NRC), groups on “key” variables, the lack of more sophisti- 1989). There is good evidence, however, that nutrient cated analytical controls for noncomparability substan- needs actually differ for adults over the age of 60 or tially limits the credibility of study findings. 70 (Russell and Suter, 1993). In addition, physiologic changes associated with aging, degenerative changes Most of the more recent studies (for example, Edwards related to chronic disease, and/or pharmacologic or et al., 1998; Ponza et al., 1996; Ho-Sang, 1989) used other interventions can influence nutrient absorption, multivariate regression techniques to control for differ- use, or excretion among the elderly (Ponza et al., ences in measured characteristics. However, only the 1994). Consequently, the available information on the 1993-95 National Evaluation of the ENP attempted to impact of the ENP on participants’ intake of food ener- address potential selection bias through statistical gy and nutrients must be considered even more tenta- modeling. Ponza and his colleagues (1996) estimated tive than the information available for most other food three selection-bias models but ultimately considered assistance and nutrition programs (FANPs).137 the results unreliable. They based their findings on regression-adjusted comparisons from a one-stage 137The Dietary Reference Intakes (DRIs) which have replaced the tradi- model, appropriately cautioning readers that selection tional RDAs (see chapter 2), define separate standards for adults between bias may play a role in reported results. the ages of 51 and 70 and those over the age of 70.

268 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program

Findings for all studies that examined the impact of the Findings from the remaining studies that included sta- ENP on the dietary intake of older adults are summa- tistical analyses were inconsistent, and none of the dif- rized in table 42. The table is divided into four sections: ferences between participants and nonparticipants were food energy and macronutrients, vitamins, minerals, and statistically significant. Most of these studies had very other dietary components. The table clearly illustrates small samples and relied on bivariate analyses. whether findings apply to congregate meals, home-delivered meals, or both types of meals. The text follows the The 1993-95 National Evaluation of the ENP also same general organization as table 42, but findings relat- found that ENP participants consumed a significantly ed to vitamins and minerals are discussed in one section. greater amount of protein than nonparticipants. However, the difference between participants and non- In the interest of providing a comprehensive picture of participants in the percentage of calories derived from the body of research, both significant and nonsignificant protein was not significant (participants consumed results are reported in table 42 and in all other “findings” more energy and more protein). tables. As noted in chapter 1, a consistent pattern of nonsignificant findings may indicate a true underlying Data from national surveys of food and nutrient intake effect, even though no single study’s results would be indicate that older Americans, like their younger coun- interpreted in that way. Readers are cautioned, however, terparts, typically exceed recommended intakes of both to avoid the practice of “vote counting,” or adding up total fat and saturated fat, expressed as a percentage of all the studies with particular results. Because of differ- total energy intake (Dwyer and Mayer, 1997). The ENP ences in research design and other considerations, find- does not appear to influence this situation one way or ings from some studies merit more consideration than the other, despite the previously described changes in others. The text discusses methodological limitations program regulations that incorporated the DGAs into and emphasizes findings from the strongest studies. the program’s nutrition standards. Neither of the studies that were completed after 1992 (when the DGAs In interpreting available data on the impact of the ENP were incorporated) and included statistical analyses on the dietary intake of older adults, findings from the found significant differences between ENP participants 1993-95 National Evaluation of the ENP (Ponza et al., and nonparticipants in the intake of fat or saturated fat, 1996) are given the most weight. Despite a lingering relative to energy intake (Ponza et al., 1996; Neyman potential for selection bias, this study provides the best et al., 1996). These findings were true whether ENP available information on potential nutrition- and meal(s) were consumed in congregate sites or at home. health-related impacts of the ENP. The study, which Findings from older studies (conducted before the was national in scope and is the most comprehensive 1992 policy change) are similar. study done to date, was implemented with great care and is based on relatively recent data. Most important- Vitamins and Minerals ly, the study used appropriate analytic techniques to Both of the national evaluations found that ENP partici- control for between-group differences in measured pants consumed significantly greater amounts of a wide characteristics rather than relying on unadjusted bivari- array of vitamins and minerals than nonparticipants. The ate comparisons. Study authors were also careful to earliest national evaluation (Kirschner/ORC, 1983) avoid estimating impacts on outcomes that are particu- reported that ENP participants consumed significantly larly vulnerable to selection bias, such as food security more than nonparticipants of all of the vitamins and (food-insecure individuals may seek out the ENP) and minerals examined: vitamin A, vitamin C, niacin, measures of nutritional status beyond intake of energy riboflavin, thiamin, calcium, and iron. The more recent and nutrients from ENP meals (ENP sites target the 1993-95 National Evaluation of the ENP (Ponza et al., most vulnerable elderly). 1996) reported the same pattern of findings for ENP participants who received congregate meals. ENP par- Food Energy and Macronutrients ticipants who received home-delivered meals also had In the 1993-95 National Evaluation of the ENP, Ponza higher mean intakes than nonparticipants, but some of et al. (1996) found that both congregate and home- these differences did not reach statistical significance. In delivered ENP participants had significantly higher addition, the 1993-95 National Evaluation found higher energy intakes than nonparticipants. Comparable intakes among ENP participants for a number of vita- results were reported in the other national study that mins and minerals that were not measured in the earlier looked at energy intake (Kirschner/ORC, 1983), as study: vitamins B6, B12, D, and E, folate, magnesium, well as in one local study (Kohrs, et al., 1978). phosphorus, potassium, and zinc.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 269 Chapter 10: Nutrition Services Incentive Program less d ct a Continued— nsume site] 1 ant imp ts co n 986) [ Signific kes Participa Home only Steele (1 less d ipants’ dietary inta c nal] 7) [6 sites] nsume site] site] 1 1 natio 0) [7 sites] 0) [7 sites] s (198 ts co 98 98 n n ari 986) [ 986) [ 996) [ ram on parti N eton (1 eton (1 Participa t impact Congregate only Gilbride (1998) [3 sites] Home only Ho-Sang (1989) [6 sites] Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Home only Steele (1 Czajka- LeClerc (1983) [1 site] Singl Home only Steele (1 Congregate only Ponza (1 Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Kohrs (1980) [6 sites] Singl n ifica d l] l] No sign a a nsume nal] nal] 7) [6 sites] 7) [6 sites] co site] ts 1 natio natio 9) [nation 9) [nation 0) [7 sites] 0) [7 sites] n the Elderly Nutrition Prog s (198 s (198 97 97 98 98 n n more/same ari ari 986) [ 996) [ 996) [ N N Participa eton (1 eton (1 LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Singl Both meal types Ponza (1 Congregate only Neyman (1996) [9 sites] Czajka- Kirschner (1 Congregate only Czajka- LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Singl Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Kohrs (1980) [6 sites] Home only Ponza (1 Ho-Sang (1989) [6 sites] Congregate only LeClerc (1983) [1 site] Kirschner (1 Home only Ho-Sang (1989) [6 sites] Steele (1 e impact of l] l] a a 1 ct examined th a t nal] nal] 7) [6 sites] tha s natio natio 3) [nation 3) [nation ant imp s (198 98 98 n ari 996) [ 996) [ studie N Signific Participants consumed more Kirschner (1 Congregate only Kohrs (1978) [6 sites] Both meal types Ponza (1 Kirschner (1 Congregate only Czajka- Both meal types Ponza (1 Kohrs (1978) [6 sites] of table. s Findings from e t at end a r d ergy y h energy and macronutrients o b r See notes a Table 42— Outcome Food Food en Protein C Fat

270 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program less d ct a Continued— nued ti nsume ant imp ts co n —Con Signific kes Participa {males} Congregate only Neyman (1996) [9 sites] less d ipants’ dietary inta c nal] 7) [6 sites] nsume site] 1 natio s (198 ts co n n ari 986) [ 996) [ ram on parti N Participa {males} t impact Kohrs (1980) [6 sites] Both meal types Ponza (1 Congregate only Czajka- Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Kohrs (1978) [6 sites] Home only Steele (1 Congregate only Neyman (1996) [9 sites] n ifica d l] No sign a nsume nal] 7) [6 sites] Elderly Nutrition Prog co ts natio 9) [nation 0) [7 sites] 0) [7 sites] n s (198 97 98 98 n more/same ari 996) [ N Participa eton (1 eton (1 {females} {females} Congregate only Neyman (1996) [9 sites] Ponza (1 Singl Congregate only Neyman (1996) [9 sites] Czajka- LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Kirschner (1 Home only Ho-Sang (1989) [6 sites] Congregate only Singl e impact of the l] a ct examined th a t nal] nal] nal] tha s natio natio natio 3) [nation 0) [7 sites] ant imp 98 98 996) [ 996) [ 996) [ studie Signific eton (1 Participants consumed more Kirschner (1 Congregate only Singl Both meal types Ponza (1 Home only Ponza (1 Both meal types Ponza (1 of table. Findings from at end 6 12 See notes Table 42— Outcome Saturated fat Vitamins Vitamin A Vitamin B Vitamin B

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 271 Chapter 10: Nutrition Services Incentive Program less d ct a Continued— nued ti nsume ant imp ts co n —Con Signific kes Participa less d ipants’ dietary inta c nsume site] 1 0) [7 sites] 0) [7 sites] ts co 98 98 n 80) [6 sites] {males} 986) [ ram on parti eton (1 eton (1 Participa {females} t impact Congregate only Neyman (1996) [9 sites] LeClerc (1983) [1 site] Kohrs (19 Singl Congregate only Gilbride (1998) [3 sites] Congregate only Gilbride (1998) [3 sites] Congregate only Gilbride (1998) [3 sites] LeClerc (1983) [1 site] Singl Home only Ho-Sang (1989) [6 sites] Steele (1 Congregate only Neyman (1996) [9 sites] n ifica s} d l] l] No sign a a nsume nal] nal] nal] 7) [6 sites] Elderly Nutrition Prog co site] [9 sites] {males} ts 1 ) natio natio natio 9) [nation 9) [nation 0) [7 sites] n 6 only s (198 97 97 98 n 99 more/same 80) [6 sites] {female ari 986) [ 996) [ 996) [ 996) [ N Participa eton (1 Neyman (1 Kohrs (19 Kirschner (1 Home only Ponza (1 Steele (1 Home only Ho-Sang (1989) [6 sites] Congregate only Neyman (1996) [9 sites] Czajka- Congregate only Gilbride (1998) [3 sites] Singl Congregate only Neyman (1996) [9 sites] Home only Ho-Sang (1989) [6 sites] Congregate Kohrs (1980) [6 sites] Kirschner (1 Kohrs (1978) [6 sites] Home only Ponza (1 Home only Ponza (1 e impact of the l] l] a a ct examined th a t nal] nal] nal] nal] nal] 7) [6 sites] tha s natio natio natio natio natio 3) [nation 3) [nation ant imp s (198 98 98 n ari 996) [ 996) [ 996) [ 996) [ 996) [ studie N Signific Participants consumed more Czajka- Both meal types Kirschner (1 Congregate only Ponza (1 Both meal types Ponza (1 Congregate only Ponza (1 Both meal types Ponza (1 Both meal types Kirschner (1 Congregate only Ponza (1 Kohrs (1978) [6 sites] of table. Findings from at end n See notes Table 42— Outcome Vitamin C Vitamin D Vitamin E Folate Niaci

272 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program less d ct a Continued— nued ti nsume site] 1 ) ant imp 6 only ts co n 99 —Con 986) [ Signific kes Participa [9 sites] {males} Home only Steele (1 Congregate Neyman (1 less d ipants’ dietary inta c n} 7) [6 sites] nsume patio site] site] i 1 1 0) [7 sites] s (198 ts co 98 n n 78) [6 sites] ari 986) [ 986) [ ram on parti N lar partic eton (1 Participa {regu {females} t impact Singl Congregate only Neyman (1996) [9 sites] Czajka- Kohrs (19 Congregate only Neyman (1996) [9 sites] Congregate only Neyman (1996) [9 sites] Home only Steele (1 Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Home only Steele (1 n ifica d l] l] l] No sign a a a n} nsume nal] 7) [6 sites] 7) [6 sites] Elderly Nutrition Prog co ipatio ts natio 9) [nation 9) [nation 9) [nation 0) [7 sites] 0) [7 sites] n rtic s (198 s (198 a 97 97 97 98 98 n n more/same ari ari 996) [ lar p N N Participa eton (1 eton (1 {irregu LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Kirschner (1 Home only Ponza (1 Ho-Sang (1989) [6 sites] Congregate only Czajka- LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Singl Kirschner (1 Home only Ho-Sang (1989) [6 sites] Congregate only Czajka- Congregate only LeClerc (1983) [1 site] Kohrs (1980) [6 sites] Singl Kirschner (1 Home only Ho-Sang (1989) [6 sites] e impact of the l] l] l] a a a ct examined th a t nal] nal] nal] 7) [6 sites] tha s natio natio natio 3) [nation 3) [nation 3) [nation ant imp s (198 98 98 98 n 78) [6 sites] 78) [6 sites] ari 996) [ 996) [ 996) [ studie N Signific Participants consumed more Kirschner (1 Congregate only Kohrs (19 Kirschner (1 Congregate only Czajka- Both meal types Ponza (1 Kohrs (19 Both meal types Ponza (1 Both meal types Kirschner (1 Congregate only Ponza (1 of table. Findings from at end vin m a er u i c l See notes a Table 42— Outcome Ribofl Thiamin Minerals C Copp

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 273 Chapter 10: Nutrition Services Incentive Program less d ct a Continued— nued ti nsume site] 1 ) ant imp 6 ts co n 99 —Con 986) [ Signific kes Participa [9 sites] {males} {females} Congregate only Neyman (1 Congregate only Nordstrom (1982) [6 sites] Kohrs (1978) [6 sites] Home only Steele (1 Congregate only Neyman (1996) [9 sites] less d ipants’ dietary inta c n} 7) [6 sites] nsume patio site] site] i [9 sites] {males} 1 1 ) 6 s (198 ts co n n 99 80) [6 sites] {males} ari 986) [ 986) [ ram on parti N lar partic Participa {females} {regu t impact Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Home only Steele (1 Home only Steele (1 Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Congregate only Neyman (1 Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] Czajka- Kohrs (19 Congregate only Gilbride (1998) [3 sites] Neyman (1996) [9 sites] n ifica s} d l] No sign a n} nsume nal] 7) [6 sites] Elderly Nutrition Prog co ipatio ts natio 9) [nation 0) [7 sites] 0) [7 sites] n rtic s (198 a 97 98 98 n more/same 80) [6 sites] {female ari 996) [ lar p N Participa eton (1 eton (1 {irregu Congregate only Singl Home only Ho-Sang (1989) [6 sites] LeClerc (1983) [1 site] Kohrs (19 Singl Both meal types Ponza (1 Congregate only Czajka- Kirschner (1 Home only Ho-Sang (1989) [6 sites] e impact of the l] a ct examined th a t nal] nal] nal] nal] tha s natio natio natio natio 3) [nation ant imp 98 996) [ 996) [ 996) [ 996) [ studie Signific Participants consumed more Both meal types Kirschner (1 Both meal types Ponza (1 Both meal types Ponza (1 Both meal types Ponza (1 Both meal types Ponza (1 of table. Findings from at end horus esium nium See notes Table 42— Outcome Iron Magn Phosp Potassium Sele Zinc

274 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program y less ses d s ct a nued ti nsume ext discu adding up all the ant imp (P-N = participant ts co ch }. n The t —Con s significant findings ma ting,” or Signific kes cket pproa ch a Participa r {in bra han others. sea attern of non “vote coun re roup tice of subg istent p ideration t c s s con e), and the less a d re con ipants’ dietary inta c oid the pra v 1 Stat s. nal] r 7) [6 sites] nsume hapter 1, ty o site] site] merit mo also identifies the d to a 1 1 natio ci s (198 ts co n n d in C een group entry studies ari 986) [ 986) [ 996) [ ram on parti cautione N note s e cell some Participa t impact ers are Congregate only Czajka- Kohrs (1980) [6 sites] Home only Ho-Sang (1989) [6 sites] Steele (1 Home only Steele (1 Congregate only Neyman (1996) [9 sites] Both meal types Ponza (1 n ifica research. A ample, national vs. 1 o difference betw ings from x n y. Read subgroup, th y of wa , find d was s n y (for e No sign pecific 1 s there nsume nal] ted in that Elderly Nutrition Prog co the stud ly to a ergy, nsideratio site] n significance. ts 1 natio en cture of the bod co n e of r more/same scop 986) [ 996) [ sive pi atistical Participa of st prehen ate, the ntage of food ign and othe ndings pertain o sts st studies. i s lts would be interpre Neyman (1996) [9 sites] Both meal types Home only Ho-Sang (1989) [6 sites] Steele (1 Ponza (1 Congregate only Gilbride (1998) [3 sites] e e impact of the te y f com on d erce nge a stud a p stro s e y’s resu earch d r s the publicati day). A f providing r 1979) included y). Whe es in re ct s from the c examined th a single stud t stud RC ( name, O s se tha finding gh no differen s s ner/ ant imp spon author’ r lue only (gm pe d in the interest o en thou se re phasize v va studie Signific senio porte ct, e s. Because of R = do - Participants consumed more e re r absolute (1998) nor Kirsch y, D al. ts for sults a derlying effe t stud re cular result un Findings from Gilbride et rticipan Cell entries show the with parti esterol um nonpa Significant resul Notes: Nonsignificant Neither 1 Table 42— Outcome Other dietary components Chol Fiber Sodi vs. indicate a true studies methodological limitations and em

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 275 Chapter 10: Nutrition Services Incentive Program

In general, the smaller studies did not find significant in a footnote in table 43. Data could not be included in differences between ENP participants and nonpartici- the table because authors did not report point esti- pants in vitamins and minerals. As illustrated in table mates. Two of the four studies reported significant dif- 42, significant differences that were reported included ferences between ENP participants and nonparticipants findings that were consistent with the large, national on one or more measures of iron status for specific studies (ENP participants consumed more nutrients), subgroups of the population. The pattern of findings and some that were inconsistent (ENP participants was not consistent, however, and there were more consumed fewer nutrients). “negative” than “positive” differences.

Other Dietary Components Three of the four studies evaluated serum albumin lev- The 1993-95 National Evaluation of the ENP found no els. Serum albumin is used as an indicator of malnutri- significant difference between cholesterol intakes of tion (inadequate protein intake) among the elderly. All ENP participants and nonparticipants. This finding was three studies found that mean serum albumin levels true for both congregate and home-delivered ENP par- were within the normal range and that the prevalence ticipants. Mean cholesterol intakes of all groups were of less-than-acceptable values did not differ by partici- well within the recommended range. pation status. However, analyses that compared mean serum albumin values by age and gender found some Similarly, the 1993-95 National Evaluation found no sig- statistically significant differences between partici- nificant difference between ENP participants (congregate pants and nonparticipants. Neyman et al. (1996) found or home-delivered) and nonparticipants in mean sodium that male ENP participants had significantly higher intake. However, the researchers did point out that serum albumin levels than male nonparticipants. excessive sodium intake may be a problem for some Czajka-Narins et al. (1987) found that the opposite ENP participants. The average intake of sodium among was true for females over 75 who participated in the congregate-meal participants exceeded the recommended ENP two or more times per week. maximum. Moreover, both types of ENP participants received more than one-third of the recommended daily The same three studies examined serum levels of vita- maximum for sodium through program meals. min A, a long-term measure of nutrient intake. Studies by Czajka-Narins et al. (1987) and Kohrs et al. (1980) Only one small, local study examined intake of dietary found that ENP participants had significantly higher fiber (Neyman et al., 1996). The authors found that levels of serum vitamin A, on average, than did nonparticipants in congregate ENP sites consumed less participants. Note that both of these studies reported fiber than eligible nonparticipants, but the difference that ENP participants consumed more vitamin A than was not statistically significant. nonparticipants, but the differences were not statistically significant (table 42). Kohrs and associates Impacts on Nutritional Biochemistries (1980) also found that ENP participants were significantly less likely than nonparticipants to have an Four of the small, local studies attempted to assess the abnormally low level of serum vitamin A. impact of the ENP on selected nutritional biochemistries (Neyman, et al., 1996; Czajka-Narins et al., 1987; Limited intake of vitamin A among the elderly had been Nordstrom et al., 1982; Kohrs et al., 1980). All of reported by several investigators (Kim et al., 1993; these studies were limited to congregate feeding sites. Kirschner/ORC, 1983; LeClerc and Thornbury, 1983; ). Kohrs (1982) emphasizes that “almost one-half of Findings from these studies, summarized in table 43, ENP nonparticipants are at risk for vitamin A defi- must be interpreted with caution. None of the researchers ciency” and that improvement in vitamin A status attempted to control for selection bias or used analytic appears to be one of the most important benefits of techniques to control for measured differences in charac- the ENP. teristics of participants and nonparticipants. The fact that the ENP specifically targets individuals with nutritional Kohrs et al. (1980) and Czajka-Narins et al. (1987) risks may account for the “negative” findings reported also looked at serum levels of vitamin C, which are by Czajka-Narins et al. (1987) and Neyman et al. (1996). affected by short-term (rather than long-term) dietary intake. Neither study found a significant difference in All four studies examined iron status using mean lev- mean levels of serum vitamin C. However, Kohrs et al. els of hematocrit, hemoglobin, and/or serum iron. (1980) found that ENP participants were significantly Findings for Nordstrom et al. (1982) are summarized

276 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program s u ants} ants} t a ct t a s wer Continued— l 7) [6 sites] 7) [6 sites] 7) [6 sites] a r r particip r particip ts lo a l a a n l l u ant imp u u 3 s (198 s (198 s (198 n n n trition 2 ari ari ari ants} N N N Signific Participa s of nu r {males; irreg {males; irreg {females; reg particip Czajka- Czajka- Neyman (1996) [9 sites] {females} Czajka- 2 2 2 2 2 s} s} s} hemical indicato wer 2 7) [6 sites] 7) [6 sites] 7) [6 sites] 7) [6 sites] ts lo n [9 sites] {males} [9 sites] {males} [9 sites] {males} ) ) ) 6 6 6 s (198 s (198 s (198 s (198 n n n n 99 99 99 80) [6 sites] 80) [6 sites] {female 80) [6 sites] {female 80) [6 sites] {female ari ari ari ari ram on bioc Participa N N N N {females} {females} {females} {females} t impact Neyman (1996) [9 sites] Neyman (1 Kohrs (19 Kohrs (19 Kohrs (19 Czajka- Neyman (1 Czajka- Neyman (1 Czajka- Kohrs (19 Czajka- Neyman (1996) [9 sites] n ifica } 2 2 2 e ants} ants} No sign 2 4 7) [6 sites] 7) [6 sites] 7) [6 sites] 7) [6 sites] 7) [6 sites] 7) [6 sites] gher/sam Elderly Nutrition Prog particip particip [9 sites] {males ) lar lar ts hi 6 u u s (198 s (198 s (198 s (198 s (198 s (198 n n n n n n n 99 80) [6 sites] 80) [6 sites] {males} 80) [6 sites] {males} 80) [6 sites] {males} 2 ari ari ari ari ari ari N N N N N N Participa {males; reg {males; reg {males} {males} {males} Neyman (1996) [9 sites] {females} Kohrs (19 Kohrs (19 Kohrs (19 Neyman (1996) [9 sites] Czajka- Czajka- Neyman (1996) [9 sites] {females} Czajka- Czajka- Czajka- Neyman (1996) [9 sites] Neyman (1 {males} Kohrs (19 Czajka- e impact of the } s ct examined th a t 4 gher 7) [6 sites] 7) {female lar tha ts hi u s [9 sites] n g ) ant imp 6 s (198 s (198 n n 2 99 80) [6 sites] 2 ari ari ants} studie N N Signific Participa {females} {males} {females; irre particip Neyman (1 Czajka- Kohrs (19 Neyman (1996) [9 sites] Czajka- of table. Findings from 1 at end n 1 1 n i b o l n g o m See notes e Table 43— Outcome Hematocrit H Serum iron Albumi Total protei Vitamin A Vitamin C Vitamin E Folate Zinc

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 277 Chapter 10: Nutrition Services Incentive Program y ses s ct a wer ts lo ain only to a n cell hematocrit ext discu adding up all the ant imp n pert The t normally low values. significant findings ma ting,” or Signific Participa s had ab udy findings m iron, or mea nts. han others. st attern of non “vote coun cipant rticipa 1 tice of ideration t c istent p s s nonpa r ENP parti con emoglobin, seru State). Where re con we s and or 1 1, a wer . 2 crit, h r oid the pra ) v tly fe ipant 7) [6 sites] city ts lo 1 mato n hapte merit mo [9 sites] {males} [9 sites] {males} d to a ) ) of ENP 6 6 s (198 n al vs. significan d in c on he 99 99 een partic dicator. pact 80) [6 sites] studies s, w ari ram on Participa cautione N note s bet s. s some of the in rticipation ple, nation s Neyman (1996) [9 sites] {males} t impact ers are Kohrs (19 Neyman (1 Czajka- Neyman (1 n exam nonparticipant r ct of pa difference ifica research. A n value ings from ificant positive im onparticipant y. Read y of e wa , find ed with s r n on mea study (fo No sign ., a sign ed with n r gher/sam ted in that o significant effe t, compa e of the nd no significant e based r nsideratio nding (i.e compa ts hi cture of the bod co s scop n r ound n 2 and fou ound tha noted, a s sive pi pposite fi hey f rticipant Participa us. T es and f ts}. prehen w value ate, and the ign and othe {females} {females} {females} herwise impact of the Elderly Nutrition Prog at st studies. s r for pa es the o lts would be interpre lo Neyman (1996) [9 sites] Neyman (1996) [9 sites] Neyman (1996) [9 sites] Neyman (1996) [9 sites] e e cke st tinued a com on d l valu r nge a scrib lowe on r stro de C y’s resu , earch d r s gher o th abnormally th abnorma ENP on iron i i the publicati ly and, unless ot f providing re hi howeve es in re he subgroup {in b ct s from the c examined th a single stud t we xt, gher name, s rt te tha finding ts hi gh no differen s s n values ant imp author’ r e of individuals w the impact of t e of individuals w al participants on d in the interest o en thou nutritional status— ort. Repo hether so identifies the phasize v studie l Signific me Participa ed at senio rt w porte k ct, e s of s. Because of ercentag ercentag e re r gregate ed at p ed at p r con k k (1982) loo sults a ble 3 of their rep

l t did not repo derlying effe re cular result p, the cell entry a o rol r un e Findings from e t s grou n in Ta e l tion, bu rom et al. Cell entries show the olest o with parti h C esterol ific sub entra

c c L Nordst Authors also loo As show Authors also loo Notes: Nonsignificant All findings are fo 1 2 3 4 D Table 43— biochemical indicator Outcome Chol HDL Ch L Triglycerides spe indicate a true studies methodological limitations and em con

278 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program less likely than nonparticipants to have less-than- Being too thin is not desirable, particularly among the normal levels of serum vitamin C. elderly. These findings suggest that thinner, and perhaps more frail elderly, may be self-selecting into the ENP Neyman et al., 1996 examined blood levels of vitamin (on their own volition or because they are targeted by E, folate, and zinc but reported no significant differ- the program). However, the available data are too limited ences between ENP participants and nonparticipants. to support a firm conclusion about the relationship between the ENP and the prevalence of thinness/wasting. Finally, Neyman et al. (1996), Czajka-Narins et al. It remains an interesting question for future research. (1987), and Kohrs et al. (1980), looked at serum cholesterol levels. Neyman and associates (1996) also Impacts on Socialization examined levels of HDL and LDL cholesterol and triglycerides. All of these studies reported that, As noted in the introduction to this chapter, the ENP although ENP participants tended to have lower cho- was intentionally designed to address the psychologi- lesterol levels than nonparticipants, the differences cal and sociological needs of the elderly as well as between the two groups were not significant. This their nutritional needs. The two national evaluations of finding was true for both mean cholesterol levels and the program are the only identified studies that for the prevalence of elevated cholesterol levels. attempted to systematically measure social outcomes Neyman et al. (1996) found no significant effect on of participants, relative to a group of eligible nonpar- HDL cholesterol, LDL cholesterol, or triglycerides. ticipants (Ponza et al., 1996; Kirschner/ORC, 1979, 1983). The studies employed two different measures, Impacts on Weight Status and results were divergent. Four local studies assessed the impact of the ENP on In the earliest national evaluation, Kirschner/ ORC weight status (Neyman et al., 1996; Ho-Sang, 1989; (1979, 1983) classified respondents based on isolation Czajka-Narins et al., 1987; Kohrs et al., 1980). Findings using a five-point index: (1) living alone, (2) having from these studies, like those related to nutritional bio- too few friends, (3) having no one to confide in, (4) chemistries, are subject to substantial concern about having children that do not visit, and (5) feeling lonely selection bias and must be interpreted with caution. more often. Using multiple regression techniques, the authors found that “being extremely isolated” was sig- All four studies used data on height and weight to cal- nificantly associated with use of ENP-sponsored shop- culate indices of obesity and thinness, including body ping assistance. mass index (BMI),138 ponderal index,139 and the percentage of desirable weight (table 44). Tricep skinfold The measure of socialization used in the 1993-95 thickness was used to assess fatness as well as deple- National Evaluation (Ponza et al., 1996) was the number tion of energy stores. of social contacts per month. The authors found that ENP participants had significantly more social con- Kohrs and associates (1980) found that the prevalence tacts per month than nonparticipants. As expected, the of obesity was not significantly related to frequency of data also showed that homebound participants had less participation in the ENP, despite the fact that mean ener- contact than those who attended congregate meal sites. gy intake was greater among participants (table 42). In fact, there was an association (nonsignificant) between Impacts on Food Security lower body weight (based on BMI, ponderal index, and percentage of desirable weight) and program participa- The issue of food security among ENP participants has tion. Czajka-Narins et al. (1987) found a comparable not been well researched and the relationship is a com- pattern among elders over age 75, with the association plicated one. Evidence from a national survey of the between ENP participation and lower body weight elderly (Cohen et al., 1993) indicates that elderly reaching statistical significance for males. In addition, FANP participants report higher levels of food insecu- Kohrs et al. (1980) found that, compared with female rity than those who do not participate in FANPs. Those nonparticipants, a significantly greater percentage of reporting participation in more than one FANP had the female ENP participants were thin or wasted. highest level of food insecurity. Elderly persons participating in two or more FANPs were more likely to have faced the choice between buying food and paying 138BMI = [Weight (kg)] / [Height (cm)2]. 139Ponderal index is calculated as height (in inches) divided by the cube for medications than elderly persons participating in root of weight (in pounds). only one FANP. These patterns presumably do not

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 279 Chapter 10: Nutrition Services Incentive Program y ata ses s ver, d ct a I. Howe ain only to a ext discu adding up all the ant imp pert less thin The t significant findings ma s ting,” or Signific Participants more obese/ udy findings bese. han others. st eight, weight, or BM attern of non “vote coun fied as o tice of istent p ideration t c s s classi con rticipants in h State). Where ns a re con ipants’ weight statu c or 1 oid the pra nonpa v perso 7) [6 sites] d city of New York} 1 hapter 1, e merit mo d to a w York City } s (198 less thin n al vs. d in C studies ari rticipants an ram on parti cautione N note percentage r s some Participants more obese/ ple, nation {females; Ne {males; upstat t impact ers are and/o ) Kohrs (1980) [6 sites] Home only Ho-Sang (1989) [6 sites] Ho-Sang (1989) [6 sites] Congregate only Neyman (1996) [9 sites] Congregate only Czajka- Home only Ho-Sang (1989) [6 sites] n exam r ween ENP pa exist research. A ifica ings from y y. Read bet y of wa , find the es s c } } } n study (fo s. No sign d d d s. obese/ s 7) [6 sites] 7) [6 sites] ity (where ted in that rk City} 2 e of the s te New York} } up note up note up note nt differen articipant o nsideratio e Elderly Nutrition Prog a be en group ts les ro ro ro cture of the bod co n g g g scop s (198 s (198 r we more thin n n for o s significa ari ari bet sive pi s N N Participa to norm ts}. prehen ate, and the ign and othe {females} {except sub {New York City {females; upst {males; New Y {except sub {females} {except sub were no han female nonp st studies. s fference lts would be interpre i ve e Kohrs (1980) [6 sites] Home only Ho-Sang (1989) [6 sites] Home only Ho-Sang (1989) [6 sites] Ho-Sang (1989) [6 sites] Congregate only Czajka- Congregate only Czajka- Congregate only Kohrs (1980) [6 sites] Kohrs (1980) [6 sites] e e impact of th cke r a com on d f d r nge a thin t s relati stro s y’s resu earch d s n value the publicati assified a subgroup, the f providing es in re cl subgroup {in b ct s from the the direction o c examined th e a single stud t obese/ s 7) [6 sites] 7) [6 sites] name, s tha finding ided on gh no New York differen s s v sed on mea ant imp ts les s (198 s (198 n n n e ba more thin author’ 1 1 1 r r ore likely to b pstate d in the interest o en thou ari ari m so identifies the phasize v studie l on is pro N N Signific he u senio porte ct, e Participa were s. Because of {females} {males} {females} {males} {females} e re r Congregate only Kohrs (1980) [6 sites] Congregate only Czajka- Congregate only Czajka- Kohrs (1980) [6 sites] Kohrs (1980) [6 sites] at, for t noted, findings a rts th rticipants sults a

t derlying effe re cular result and no informati p, the cell entry a h d g d un repo i Findings from e grou ndex w

Cell entries show the e l with parti reporte b eral i a ific sub r i c Female ENP pa The author s Notes: Nonsignificant Unless otherwise 1 2 e Table 44— Outcome Body mass index (BMI) Pond Percent of d Tricep skinfol thickness spe indicate a true studies methodological limitations and em are not

280 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program reflect an impact of FANP participation, but indicate The Level 1 screen (table 45) is a simple checklist that that individuals who choose to participate in FANPs can be completed largely by an elderly individual him- are more food-insecure than those who do not. self or herself, with some additional information obtained through an interview with a social service or Only one of the identified studies attempted to assess health care provider. No laboratory tests or special the impact of ENP participation on food security measurements are required. The Level 2 nutrition (Edwards et al., 1993). The study focused on a very screen encompasses a more indepth assessment by a restricted sample of elderly diabetics who were either health professional, including measurement of anthro- receiving home-delivered meals or were on a waiting pometric, biochemical, clinical, and dietary indicators list for home-delivered meals. Food-insecure individu- of nutritional status as well as an assessment of func- als were defined as those who reported that they did tional status. not have enough money to purchase the foods they needed or had some other difficulty in obtaining food. The Level 1 NSI screen is currently used in many ENP In this context, the ENP was found to have a positive programs (Dwyer and Mayer, 1997) as a means of effect on food security. Elderly diabetics who were identifying individuals who might benefit from a spe- receiving home-delivered meals were less likely than cific nutrition-related service (for example, home- comparable elders on a waiting list to be classified as delivered meals, assistance with shopping or cooking, food insecure or to go one or more days per month or nutrition education).141 Research has shown that the without food. Level 1 screen reliably identified individuals at risk for nutrition-related problems (Posner, 1993). There is Ponza and his colleagues (1996) also assessed food some concern, however, that the specificity of the security among ENP participants. Comparable data measure is less than desirable; that is, it may produce were not collected for nonparticipants, however. too many “false positives” or overestimate the preva- Instead, the authors compared findings for ENP partic- lence of significant nutritional risk (Dwyer and ipants with data for the U.S. elderly population over- Mayer, 1997). all. Food security was measured using a subset of four of the questions used in the Cohen et al. study (1993) Using the Level 1 NSI screen on a small elderly popu- (described earlier). Results indicated that, although lation in New York City, Gilbride et al. (1998) found most ENP participants reported having enough food to that the level of nutritional risk among congregate eat, they were much more likely to experience food meal participants was twice that of a group of compa- insecurity than elderly persons in the overall U.S. rable elders who did not eat at congregate meal sites. population. The authors did not assess the statistical significance of this difference. Impacts on Nutritional Risk Ponza et al. (1996) used an approximation of the Level Assessing the nutritional status of elderly individuals 1 NSI screen to assess nutritional risk among both is difficult because the factors that determine risk are congregate and home-delivered meal participants. complex and interdependent. Moreover, nutritional Overall, 64 percent of congregate and 88 percent of risk among the elderly is influenced by variables that home-delivered participants had characteristics associ- are not considered for most other age groups, includ- ated with moderate to high nutrition risk. No comparing socialization, physical functioning and mobility isons were made to nonparticipants. (frailty), and behavioral elements. To address this issue, the Nutrition Screening Initiative (NSI), a Most of the other published research related to the NSI national collaborative effort of professional organiza- and use of the NSI screen is descriptive research. tions committed to identifying and treating nutritional However, the increasing use of NSI tools in ENP sites problems among the elderly, developed a two-tiered and in other social and health care service delivery approach to screening for potential nutrition-related sites may lead to outcomes-focused research. problems.140

141Key elements of the Level 1 screen have been incorporated into a simple self-assessment tool called the DETERMINE checklist. The 140The NSI and associated nutrition screening tools are described in DETERMINE checklist is also widely used in ENP delivery sites and by detail elsewhere (Gilbride et al., 1998; Ponza et al., 1996; Posner et al., other groups and organizations working with older adults (Dwyer and 1993; Food Research and Action Center, 1987). Mayer, 1997).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 281 Chapter 10: Nutrition Services Incentive Program

Summary By all accounts, the ENP is meeting its goal of providing low-cost, nutritionally sound meals to participating eld- Since the inception of the ENP, two national evalua- ers. Program meals comply with the Dietary Guidelines tions and a number of smaller local studies have for Americans and most often far exceed the minimum attempted to assess the effectiveness of the program in of one-third of the RDA per meal as required by law. meeting its goals. All of these studies used quasi- experimental designs (participant vs. nonparticipant), The available research suggests that the ENP is provid- with nonparticipants identified in a variety of ways. ing elderly participants with more energy and nutrients Selection bias is a serious issue in all of this research. than they might otherwise consume. The two national However, only the most recent national study (Ponza evaluations report increased consumption of food energy, et al., 1996) addressed the problem systematically protein, and a broad array of vitamins and minerals. (although inconclusively). Moreover, much of the The smaller studies generally did not find significant available research used unsatisfactory analysis tech- differences, which may be due to the general absence niques, presenting simple bivariate comparisons with of analytical controls for pre-existing participant/non- no statistical controls for differences in measured char- participant differences as well as small sample sizes. acteristics of participants and nonparticipants.

Table 45—Level 1 Nutrition Screen from the Nutrition Screening Initiative If any one of the following is true, the individual may be at risk of poor nutritional status:

Body weight Has lost 5 lb or 5% of body weight in 1 month Has lost or gained 10 lb or 10% of body weight in the past 6 months BMI <211 BMI >281

Eating habits Does not have enough food to eat each day Usually eats alone Does not eat anything on one or more days per month Has a poor appetite Is on a special diet Eats vegetables 2 or fewer times daily Consumes milk or milk products once or not at all daily Consumes fruit or fruit juice once or not all daily Eats breads, cereals, pasta, rice, or grains 5 or fewer times daily Has difficulty chewing or swallowing Has more than 1 alcoholic drink per day (if woman); more than 2 drinks per day (if man) Has pain in mouth, teeth, or gums

Living environment Lives on an income of <$6,000 per year per individual in the household Lives alone Is housebound Is concerned about home security Lives in a home with inadequate heating or cooling Does not have stove and/or refrigerator Is unable or prefers not to spend money on food (<$25-$30 per person per week spent on food)

Functional status Usually or always needs assistance with any of the following:

Bathing Eating Traveling outside home Dressing Toileting Preparing food Grooming Walking or moving about Shopping for food or other necessities 1BMI = Body Mass Index = [Weight (kg)] / [Height (cm)2]. Source: American Board of Family Practice Reference Guide for Geriatric Patients, “A Dietary Assessment—Table 5..” (http://www.familypractice.com/references/guidesframe.htm). Accessed June 2003.

282 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program

While all studies of the impact of the ENP are subject that, compared with the overall elderly U.S. popula- to selection bias, studies that looked at measures other tion, ENP participants were much more likely to expe- than dietary intake (weight status, nutritional bio- rience food insecurity. chemistries, socialization, food security, and nutritional risk) are especially prone to this problem because Finally, one small study compared ENP participants the program specifically targets elders who are at and nonparticipants on a relatively simple, yet compre- nutritional or social risk. The limited information that hensive, measure of nutritional risk. The authors report is available suggests that ENP participation is not asso- that the rate of nutritional risk among congregate ENP ciated with obesity and that, in fact, thinner, more frail participants was twice that of nonparticipants. The elderly may self-select into the program. With the pos- research, however, used no statistical techniques to sible exception of serum vitamin A, which was posi- control for differences between groups or to assess the tively associated with participation in the ENP, draw- statistical significance of the observed difference. ing firm conclusions about the impact of the ENP on nutritional biochemistries is not possible. The importance of the ENP as a component of the nutrition safety net will continue to increase in coming Evidence of the ENP’s impact on reducing social isola- years as the population ages. Future research on the tion and promoting quality of life among the elderly is impacts of the ENP would benefit from a greater focus mixed. While the perceived benefit of social and support on impacts among the homebound who are most at services is quite high, only the two national evaluations risk for poor nutrition and health outcomes and repre- attempted to systematically measure social outcomes of sent an ever-increasing component of the program. In ENP participants, relative to a group of eligible non- addition, given the focus of the program on social as participants. The two studies employed different meas- well as nutritional needs, future research should include ures of socialization and reported divergent results. comprehensive assessment of the impact of the ENP on both food security and nutritional risk. Only one study examined the impact of the ENP on food security. In a very restricted sample of elderly Most importantly, future research should emphasize diabetics, the study found that ENP participants receiv- longitudinal rather than cross-sectional designs. ing home-delivered meals were less food-insecure than Although more costly, longitudinal studies would pro- nonparticipants on a waiting list for home-delivered vide a firmer foundation for studying impacts beyond meals. On the other hand, Ponza et al. (1996) found dietary intakes and for examining the influence of the ENP on seniors’ nutrition status, health status, and quality of life over time (Roe, 1989; Posner, 1979).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 283 Chapter 10: Nutrition Services Incentive Program

References Gilbride, J.A., E.J. Amella, E.B. Breines, et al. 1998. “Nutrition and Health Status Assessment of Balsam, A.L., A.F. Sullivan, B.E. Millen, and B.L. Community-residing Elderly in New York City: a Pilot Rogers. 2000. “Service Innovations in the Elderly Study,” Journal of the American Dietetic Association Nutrition Program: Two Decades of Accomplishments,” 98(5):554-58. Journal of Nutrition for the Elderly 19(4):41-48. Grandjean, A.C., L.L. Korth, G.C. Kara, et al. 1981. Balsam, A.L., and B.L. Rogers. 1991. “Serving Elders “Nutritional Status of Elderly Participants in a in Greatest Social and Economic Need: The Challenge Congregate Meals Program,” Journal of the American to the Elderly Nutrition Program,” Journal of Aging Dietetic Association 78:324-29. and Social Policy 3(2):41-55. Ho-Sang, G.M. 1989. Evaluation of the Supplemental Caliendo, M.A. 1980. “Factors Influencing the Dietary Nutrition Assistance Program for the Frail Elderly in Status of Participants in the National Nutrition New York State. Unpublished doctoral dissertation Program for the Elderly. 1. Population Characteristics from Cornell University. and Nutritional Intakes,” Journal of Nutrition for the Elderly 1(1):23-39. Kim, K.K., E.S. Yu, W.T. Liu, et al. 1993. “Nutritional Status of Chinese-, Korean-, and Japanese-American Caliendo, M.A., and M. Batcher. 1980. “Factors Elderly,” Journal of the American Dietetic Association Influencing the Dietary Status of Participants in the 93(12):1416-22. National Nutrition Program for the Elderly. 2. Relationships Between Dietary Quality, Program Kirschner Associates, Inc. and Opinion Research Participation, and Selected Variables,” Journal of Corporation. 1983. An Evaluation of the Nutrition Nutrition for the Elderly 1(1):41-53. Services for the Elderly, Volume 2: Analytic Report. Washington, DC: U.S. Department of Health and Caliendo, M.A., and J. Smith. 1981. “Preliminary Human Services, Administration on Aging. Observations on the Dietary Status of Participants in the Title III-C Meal Program,” Journal of Nutrition for Kirschner Associates, Inc. and Opinion Research the Elderly 1(3-4):21-39. Corporation. 1979. Longitudinal Evaluation of the National Nutrition Program for the Elderly: Report of Cohen, B.E., M.R. Burt, and M.M. Schulte. 1993. First-Wave Findings. Washington, DC: U.S. Hunger and Food Insecurity Among the Elderly. Department of Health, Education, and Welfare, Washington, DC: Urban Institute. Administration on Aging.

Czajka-Narins, D.M., M.B. Kohrs, J. Tsui, et al. 1987. Kohrs, M.B. 1986. “Effectiveness of Nutrition “Nutritional and Biochemical Effects of Nutrition Intervention Programs for the Elderly,” in M. L. Programs in the Elderly,” Clinics in Geriatric Hutchinson and H.N. Munro (eds.), Nutrition and Medicine 3(2):275-87. Aging. New York, NY: Academic Press.

Dwyer, J.T., and J. Mayer. 1997. Update on Risks of Kohrs, M.B. 1982. “Evaluation of Nutrition Programs Malnutrition in Older Americans: Perspectives from for the Elderly,” American Journal of Clinical New Studies. Bethesda, MD: National Aging Nutrition 36(4):812-18. Information Center. Kohrs, M.B., J. Nordstrom, E.L. Plowman, et al. 1980. Edwards, D.L., E.A. Frongillo, Jr., B. Rauschenbach, “Association of Participation in a Nutritional Program et al. 1993. “Home-delivered Meals Benefit the for the Elderly with Nutritional Status,” American Diabetic Elderly,” Journal of the American Dietetic Journal of Clinical Nutrition 33(12):2643-56. Association 93(5):585-87. Kohrs, M.B., P. O’Hanlon, and D. Eklund. 1978. “Title Food Research and Action Center. 1987. A National VII - Nutrition Program for the Elderly. I. Contribution Survey of Nutritional Risk among the Elderly. to One Day’s Dietary Intake,” Journal of the American Washington, DC: Food Research and Action Center. Dietetic Association 72(5):487-92.

284 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 10: Nutrition Services Incentive Program

LeClerc, H.L., and M.E. Thornbury. 1983. “Dietary Roe, D.A. 1989. “Nutritional Surveillance of the Intakes of Title III Meal Program Recipients and Elderly: Methods to Determine Program Impact and Nonrecipients,” Journal of the American Dietetic Unmet Need,” Nutrition Today 24(5):24-29. Association 83(5):573-77. Russell, R.M., and P.M. Suter. 1993. “Vitamin National Research Council. 1989. Recommended Requirements of Elderly People: an Update,” Dietary Allowances, 10th edition. Washington, DC: American Journal of Clinical Nutrition 58:4-14. National Academy Press. Singleton, N., M.H. Overstreet, and P.E. Schilling. Neyman, M.R., S. Zidenberg-Cherr, and R.B. 1980. “Dietary Intakes and Characteristics of Two McDonald. 1996. “Effect of Participation in Groups of Elderly Females,” Journal of Nutrition for Congregate-site Meal Programs on Nutritional Status the Elderly 1(1):77-89. of the Healthy Elderly,” Journal of the American Dietetic Association 96(5):475-83. Steele, M.F., and J.D. Bryan. 1986. “Dietary Intake of Homebound Elderly Recipients and Nonrecipients of Nordstrom, J.W., O.G. Abrahams, and M.B. Kohrs. Home-delivered Meals,” Journal of Nutrition for the 1982. “Anemia among Noninstitutionalized White Elderly 5(2):23-34. Elderly,” Nutrition Reports International 25(1):97-105. Stevens, D.A., L.E. Grivetti, and R.B. McDonald. O’Shaughnessy, C. 1990. CRS Report for Congress: 1992. “Nutrient Intake of Urban and Rural Elderly Older Americans Act Nutrition Program. Library of Receiving Home-delivered Meals,” Journal of the Congress. American Dietetic Association 92:714-718.

Ponza, M., J.C. Ohls, B.E. Millen, et al. 1996. Serving U.S. Department of Agriculture, Food and Nutrition Elders at Risk: The Older Americans Act Nutrition Service. 2003. Program data. Available: http://www. Programs, National Evaluation of the Elderly fns.usda.gov/pd. Accessed April 2003. Nutrition Program, 1993-1995, Volumes I, II, and III. U.S. Department of Health and Human Services, U.S. Department of Agriculture, Food and Nutrition Administration on Aging. Service. 2002. “Nutrition Services Incentive Program (Formerly NPE).” Available: http://www.fns.usda.gov/ Ponza, M., J.C. Ohls, and B.M. Posner. 1994. Elderly fdd/programs/nsip. Accessed March 2002. Nutrition Program Evaluation Literature Review. Princeton, NJ: Mathematica Policy Research, Inc. U.S. Department of Health and Human Services, Administration on Aging. 2002. Linking Nutrition and Posner, B.M. 1979. Nutrition and the Elderly: Policy Health: 30 Years of the Older Americans Act Nutrition Development, Program Planning, and Evaluation. Programs—Program Milestones: 1954-2002. Chapter 8: Summary of Conclusions, Intervention Design Implications, and Future Research Needs. Vaughan, L.A., and M.M. Manore. 1988. “Dietary Lexington, MA: Lexington Books, pp. 147-175. Patterns and Nutritional Status of Low Income, Free- living Elderly,” Food Nutrition News 60(5):27-30. Posner, B.M., A.M. Jette, K.W. Smith, et al. 1993. “Nutrition and Health Risks in the Elderly: The Wellman, N., L.Y. Rosenzwieg, and J.L. Lloyd. 2002. Nutrition Screening Initiative,” American Journal of “Thirty Years of the Older Americans Nutrition Public Health 83(7):972-78. Program,” Journal of the American Dietetic Association 102(3):348-50.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 285 Chapter 11 Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas

The Nutrition Assistance Program (NAP) in Puerto power. Both programs have monthly benefits that vary Rico, American Samoa, and the Northern Marianas by household size and net income, and both programs provides food and nutrition assistance to low-income are available to all applicants who meet specified eligi- individuals through block grants to territory adminis- bility criteria. trative agencies. The territories provide cash or checks to eligible participants. The NAP replaced the Food There are three major differences between the NAP Stamp Program (FSP), which operated in the territories and the FSP, however (GAO, 1992). First, NAP bene- from 1975 through 1982. fits are distributed as checks (cash) rather than food stamps (coupons). This switch in the form of the food All of the research to date on the NAP has centered on assistance benefit was motivated by the expectation of the program in Puerto Rico. Most of this research considerable savings in administrative costs. (A subse- focused on assessing the impact of replacing the FSP quent study, Beebout et al., 1985, estimated those sav- with the NAP but also provides some information ings at about $6 million annually.) Distributing bene- about the impact of the NAP per se. fits as checks was also intended to reduce fraud and theft and to eliminate the problem of food stamp trafficking. Trafficking—exchanging coupons for cash at a Program Overview reduced value—was known to be widespread in Puerto The FSP was introduced in Puerto Rico in FY 1975 and Rico and was believed to have resulted in a loss of grew rapidly. By 1977, the FSP in Puerto Rico was benefits to program participants. larger, in terms of both the percentage of the population participating and the total value of benefits issued Second, the cash benefits provided by the NAP are not each month, than any of the programs operating in the restricted. That is, NAP recipients may elect to spend 50 States (U.S. General Accounting Office (GAO), the cash they receive on something other than food. 1978). A 1983 study by the Food and Nutrition Service Food stamp coupons, on the other hand, can be (FNS), U.S. Department of Agriculture (USDA), found redeemed only for food. that about 56 percent of the Puerto Rican population The third difference between the NAP and the FSP is participated in the FSP in FY 1981. The FSP in Puerto that benefits available through the NAP are con- Rico accounted for about 8 percent of FSP participa- strained by the size of the block grant. The initial NAP tion overall and 8 percent of total Federal expendi- block grant of $825 million was $90 million (or 10 tures. Although FSP eligibility and program operation percent) less than the FY 1981 FSP allotment.142 standards were identical for the 50 States and Puerto Program administrators had to incorporate stricter eli- Rico, deductions and monthly benefits were typically gibility requirements and reduced benefit levels in lower for Puerto Rican participants. order to allocate the diminished funds. Relative to the In response to concerns about the size, expense, and participation rate at the end of the FSP, the 1984 par- management of the FSP in Puerto Rico, the 1981 ticipation rate for the NAP was down 111,000 house- Omnibus Budget Reconciliation Act (OBRA) abol- holds, a decline of about 22 percent. Weekly food ished the program and replaced it with an $825 million assistance benefit levels fell an average of $6, a 14- block grant. Puerto Rican authorities designed the percent decrease (in constant 1984 dollars). Nutrition Assistance Program (NAP) to administer the The annual block grant for the NAP in Puerto Rico block grant beginning in July 1982. The switch from was held constant at $825 million from FY 1982 the FSP to a cash delivery system was permanently authorized in September 1985. 142Puerto Rican authorities protested the reduction in the total value of The objectives of the NAP and the FSP are identical: the benefit package. Their assertions about potential deleterious effects were countered by arguments that the block grant would provide more to provide low-income households with access to a flexibility than the FSP and would result in administrative savings nutritious diet through increased food purchasing (Andrews and Pinchuk, 1984).

286 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas through FY 1986. Increases since then have averaged planning and preparation. A 7-day, aided-recall ques- 3-4 percent annually. In FY 2002, the block grant was tionnaire was used to obtain information about food $1.35 billion (USDA/FNS, 2003).143 used from household supplies. For each food item used, information was obtained on the kind of food Participation in the NAP in Puerto Rico has declined (for example, ground beef or whole milk), the form of somewhat since FY 1991 when, on average, 1.5 mil- the food (fresh, canned, or frozen), the quantity used, lion people received NAP benefits. Participation has the price paid (if appropriate), and the source (pur- been roughly level, at around 1.3 million, since FY chased, home-produced, gift, or payment). Data were 1994. The Puerto Rican population has grown steadily also collected on snacks and refreshments eaten by throughout this period, however, which means that the guests and on the number of meals eaten away from percentage of the population receiving assistance has home and associated expenditures. generally declined.144 The studies by Beebout et al. (1985) and Hama (1993) examined impacts on household food expenditures. All Research Review three studies examined impacts of the NAP on nutrient 147 All published research investigating the nutrition-related availability at the household level. These two out- impacts of the NAP has focused on the NAP in Puerto comes are related. The hypothesis is that food assis- Rico. Three such studies were identified in the litera- tance benefits lead to an increase in food expenditures, ture search.145 Study characteristics are summarized in which leads to an increase in the amount of nutrients table 46. The most widely recognized study in this available to the household. In theory, an increase in area is the study completed by Beebout et al. in 1985. nutrient availability leads to an increase in nutrient This study, as well as the more recent study by Bishop intake at the individual level; however, none of the and his colleagues (1996), focused mainly on assess- available studies of the NAP looked at nutrient intake ing the impact of replacing the FSP with the NAP but or at other nutrition- and health-related outcomes. also provides some information on impacts of the NAP itself. The third study (Hama, 1993) compared NAP Impact on Food Expenditures participants with nonparticipants in 1984, the second Both Beebout et al. (1985) and Hama (1993) estimated full year of operations under the block grant. impacts of the NAP on household food expenditures. Beebout and his colleagues reported a positive effect, All three studies used data from the 1977 Puerto Rico while Hama reported a negative effect. Theoretical and Supplement to the Nationwide Food Consumption methodological considerations limit the credibility of Survey (NFCS) and/or the 1984 Puerto Rico Hama’s finding, as discussed below. Household Food Consumption Survey (HFCS). The former survey was conducted while the FSP was still The study conducted by Beebout et al. was intended in place. Data for the latter were collected early in the principally to evaluate the impact of the NAP relative life of the NAP. to the FSP. With regard to household food expenditures, the research question was whether the change The 1977 and 1984 survey samples were both repre- from the FSP to the NAP was associated with a change sentative of the Puerto Rican population of housekeep- 146 in the amount of money households spent on food. The ing households, and the data collection methodolo- study used the 1977 NFCS data (collected when the gies were almost identical. Data were obtained from FSP was in place) and the 1984 HFCS data (collected the person identified as most responsible for meal early in the life of the NAP). Analyses attempted to separate the effect of switching from food stamp 143 The FY 2002 block grants for the Pacific Islands covered under the (coupon) benefits to cash (checks) from the effect of program (American Samoa and the Northern Marianas) were $5.3 million and $6.1 million, respectively. the tighter eligibility criteria and reduced benefit levels 144Information on participation figures for the Pacific Islands was not associated with the NAP. available. 145In 1990, Congress directed the GAO to study the NAP to determine whether NAP recipients were receiving the same nutritional benefits as other U.S. citizens receiving food assistance benefits. GAO determined that 147“Nutrient availability” reflects the nutrient content of foods used from such a study could not be completed because of time and costs constraints. the household food supply. This measure differs from nutrient intake Consequently, the GAO prepared a report that summarized available because it (1) includes food that is wasted, fed to pets, or eaten by guests research (GAO, 1992). and (2) does not include food that is obtained and eaten outside of the 146Housekeeping households are those with at least one member having household (for example, restaurant meals and foods eaten as a guest in 10 or more meals from the home food supply. other homes).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 287 Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas n- , n n e od h selectio gressio gressio n minanc o te re te re a a dels quati Analysis met ditures Multivari Multivari bias mo with 2-e Stochastic do n od expen mmy mmy ship patio u enefit u er on d on d Measure of nt participation ousehold fo Participati Participati Group memb dummy, partici dummy, and b amou t vs. with with pant out out n d d t (1977 t (1977 u u Design partici Pre-cash compare casho vs. 1984) Participa non Pre-cash compare casho vs. 1984) gram in Puerto Rico on h ) o ng ng e 95) 95) n e e l l b b i i ility ility 3,9 3,9 g g t and t and t and latio pant pant pant i i ce Pr l l ib ib n n n ) lds lds usi lds usi g e e g g e - - l e e b din i eho eho eho ,559) Popu m m g 7 eli 7 eli partici partici partici i 1 (sample siz l o o sistan e c c n n n Participa Participa i 197 criteria (n= non hous non (inclu i hous (n= Participa i 197 criteria (n= non hous As 1 Puerto Puerto Nutrition co HFCS co co 4 4 8 8 9 9 t to the t to the the f n n Data source eme eme 4 Puerto Ri 7 Puerto Ri 7 Puerto Ri pact o . . m 197 suppl NFCS and 1 Rico HFCS 198 197 suppl NFCS and 1 Rico HFCS y y t n e i ient ient d d r t o o ption Surve ption Surve u o o s s m n f f m e e u u d d d s amined the i ld nutr ld nutr l l l s o o o o o lity lity lity h h h h h ex e e e nditur nditur t abi abi abi Outcome(s) s s s u u u lability o o o H H avail House avail expe H House avail expe ide Food Con old Food Con t avai ) eh udies tha 6 s t 9 s: 9 ) ce 1 3 r ( —S 9 . l 9 nutrien a 1 46 ( p out et al. a 5) o NFCS = Nationw HFCS = Hou h m Data sou s 1 a able i T Study B H Beeb (198 and/or

288 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas

Beebout and his colleagues estimated the marginal Hama (1993). Hama used the 1984 HFCS data set propensity to spend on food (MPSF) out of food stamp (collected early in the life of the NAP) to compare NAP benefits (based on the 1977 NFCS data) and out of participants with nonparticipants. The nonparticipant NAP cash benefits (based on the 1984 HFCS data). They sample included both eligible and ineligible households. reported a positive and significant impact for NAP participation, with an MPSF of 0.21 for at-home food Hama did not produce MPSF estimates. Rather, she expenditures and 0.23 for away-from-home food expen- estimated the average difference between participant ditures. These effects translate into an estimated impact and nonparticipant households’ weekly food expendi- on weekly food expenditures of $2.39 per Adult Male tures, controlling for household income, household Equivalent (AME) per week for at-home food expen- size, and urbanization. Her conclusions were very dif- ditures and $2.61 per AME per week for total food ferent from those reported by Beebout et al. (1985). expenditures. The MPSF for NAP income was greater Hama found that NAP households spent about $5 less than the MPSF from ordinary income, but the statisti- per week on at-home food expenditures than did non- cal significance of the difference was not tested. participating households, a statistically significant effect.148 However, serious limitations in Hama’s In comparing estimated impacts for the FSP and the analysis undermine the credibility of her result. NAP, the authors found that point estimates for the MPSF out of program benefits was positive and signif- First of all, Hama’s result does not necessarily indicate icant for both programs and that differences between that NAP households spent less for food than they NAP and FSP coefficients were small and not statisti- would have spent in the absence of the benefit. Rather, cally significant. These results suggest that both the it implies that NAP households spent less for food than FSP and the NAP increased food expenditures and that the amount that would have been spent by nonpartici- the relative impact of both programs was roughly pating households with the same total income. One 149 equivalent. While there is no reason to question the likely reason for this odd finding is selection bias. basic finding—that the NAP leads to an increase in Such bias may be exacerbated by Hama’s use of the food expenditures—there is reason to question the entire sample (rather than limiting the analysis to low- broader finding—that the impact of the NAP (cash) is income households), in conjunction with an assumed equivalent to the impact of the FSP (coupons). The linear relationship between income and food expendi- study’s reliance on a pre-/post-design and use of sur- tures. If a curvilinear specification were more appro- vey samples that are separated by a 7-year interval priate, as some researchers argue, then the households makes it considerably weaker than other studies that on the extreme ends of the income distribution would have looked at the differential impact of cash and tend to have actual expenditures below their predicted coupons. In addition, some have argued that the FSP in expenditures (Moffitt, 1989). The negative coefficient Puerto Rico was essentially “cashed out” before the for the dummy variable identifying participant house- NAP was instituted (Moffitt, 1989). That is, FSP holds (all at the low end of the income range) may coupons were used as a second form of currency even have simply reflected a preponderance of negative before the changeover. residuals at the low end of the income range due to a poor fit to a straight line. Had the sample been limited As discussed in detail in chapter 3, the strongest study to low-income households, then the participation completed to date on the impact of cashing out food dummy would be independent of income and this stamps (Ohls et al., 1992) “establishes firmly that the potential source of bias would have been eliminated. coupon format of food stamps causes the FSP to increase household expenditures on food at home by a greater Impact on Household Nutrient Availability amount than would occur if the households received All three of the identified studies examined the impact the same benefit amount as cash” (Burstein et al., of the NAP on availability of food energy and selected 2004). This finding, coupled with the relative weak- nutrients at the household level. Analyses focused on ness of the NAP vs. FSP comparison in the Beebout et nutrients considered to be potentially low in the diets al. study, suggests that the positive impact of the NAP on household food expenditures may, in fact, be less than the impact that would occur under the FSP. 148Note that Hama’s results are reported in $/household/week, whereas Beebout et al. (1985) reported results in $/AME/week. 149Beebout et. al. (1985) attempted to control for selection bias in their The only other available study of the impact of the NAP models and limited the analysis sample to low-income, program-eligible on household food expenditures was completed by households.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 289 Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas of Puerto Rican households (calcium, iron, magne- each of five income quintiles. If one population had sium, vitamin A, and vitamin B6). Bishop and his col- higher mean availability at all five income quintiles, leagues also studied availability of riboflavin and then it was said to have first degree stochastic domi- niacin. The analysis methods used in the three studies nance (FSD) over the second population. T-tests were were widely divergent. also used to compare means at each of the five income quintiles. For each nutrient analyzed, the authors The models used by Beebout and his colleagues reported whether or not there was FSD (higher means (1985) assumed that NAP impacts on nutrient avail- at all income levels) and whether or not there was a ability stemmed from impacts on food expenditures. statistically significant difference in means at the low- The authors first estimated NAP impacts on food est quintile of income. The authors also carried out expenditures. Then, in a separate model, they estimate subsample analyses that compared the poorest quintile the relationship between food expenditures and avail- of participant and nonparticipant households in each ability of a particular nutrient. Next, to get the estimat- data set, using ordinary least squares regression. For ed impact of the NAP on the availability of a given purposes of this review, results of these analyses are nutrient, they multiplied the estimated NAP impact on the most relevant. food expenditures by the coefficient for the relationship between expenditures and a particular nutrient. Results of the 1984 versus 1977 analysis showed that Thus, the models assumed that at-home food expendi- the distributions of energy availability before and after tures from NAP benefits generate the same nutrient the NAP were not significantly different. Results for values as equal at-home food expenditures from ordi- nutrients varied. Some distributions improved signifi- nary income. cantly after the NAP (iron, vitamin A, and niacin), some worsened significantly (calcium and riboflavin), and Using this methodology, Beebout et al. concluded that some remained the same (magnesium and vitamin B6). the NAP reduced the percentage of participating In examining impacts by income quintiles, the authors households that failed to attain 100 percent of the noted that all of the improvements reached the lowest RDA for food energy and for five vitamins and miner- income quintile while the negative changes did not. als. Reductions between 5.0 percentage points (food energy) and 6.7 percentage points (magnesium) were Bishop and his colleagues also compared energy and estimated. No tests of statistical significance were pro- nutrient availability among NAP participants and non- vided for the impact of the NAP per se. The signifi- participants, using only the 1984 data set. They cance of differences between the FSP and the NAP restricted the sample to households in the lowest quin- was tested, and none of the differences was signifi- tile of the nutrient distribution under consideration. cantly different from zero. Among these high-risk households, NAP participation was associated with greater availability of food energy Bishop et al. (1996) used the 1977 NFCS data and the and six of the seven nutrients examined (all but calci- 1984 Puerto Rico HFCS data to determine whether um). Differences were statistically significant for iron, household nutrient availability in the Puerto Rican magnesium, and vitamin B6. population as a whole was different when the FSP was in effect than when the NAP was in effect.150 They Hama (1993) presented impacts of NAP participation compared household nutrient availability among all on household nutrient availability, estimated from island residents in 1977 vs.1984. In each data set, they reduced-form regression equations. The estimates were also compared program participants (FSP participants positive for energy, calcium, and magnesium and neg- in the 1977 dataset and NAP participants in the 1984 ative for vitamin A and vitamin B6. No statistical test data set) with nonparticipants. results were reported.

The authors compared distributions of household nutrient availability, with particular focus on households at Summary the lowest end (lowest quintile) of the income distribu- The available information on the nutrition-related tion. They used stochastic dominance methods, which impacts of the NAP (in Puerto Rico) must be consid- essentially compared household nutrient availability at ered to be both limited and dated. All three of the studies reviewed used the 1984 Puerto Rico Household 150The sample included participants and nonparticipants, and the analy- Food Consumption Survey, which was conducted just sis did not differentiate between them. 2 years after the NAP replaced the FSP in Puerto Rico.

290 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas

Because the NAP gives households extra income, it is nutrients available to households. All three of the stud- a foregone conclusion, given results of research on the ies reviewed here examined impacts on household FSP (see chapter 3), that the program will increase nutrient availability and found that the NAP increased participants’ food expenditures, on average. Not sur- availability of food energy and several vitamins and prisingly, the only study to estimate a marginal minerals. Only one study (Bishop et al., 1996) report- propensity to spend (MPSF) on food out of NAP bene- ed on the statistical significance of differences fits (Beebout et al., 1985) found a positive effect. The between NAP participants and nonparticipants, howev- estimated MPSF out of NAP benefits was greater than er, and not all of the apparently positive results were the MPSF out of ordinary income, but the statistical statistically significant. significance of the difference in coefficients was not tested. The other study that examined food expendi- Any serious understanding of current impacts of the NAP on participants’ nutrition and health status will tures did not estimate the MPSF (Hama, 1993). Results of this study imply that the MPSF out of NAP benefits clearly require new research. The existing national sur- would be lower than the MPSF out of ordinary vey of health and nutrition status, the National Health income, but this result may stem from selection bias. and Nutrition Examination Survey (NHANES), does not include Puerto Rico or the Pacific Islands. Evidence that the NAP affects household nutrient Consequently, a specialized data collection will be availability is weak but suggests that the NAP may required to address questions about the nutrition- and result in small increases in the amount of energy and health-related impacts of the NAP.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 291 Chapter 11: Nutrition Assistance Program in Puerto Rico, American Samoa, and the Northern Marianas

References Moffitt, R. 1989. “Estimating the Value of an In-Kind Transfer: The Case of Food Stamps,” Econometrica Andrews, M., and Pinchuk, R. 1984. The Puerto Rican 57 (2):385-409. Nutrition Assistance Program: A Case Study in the Dynamics of Food and Agricultural Policymaking. Ohls, J.C., T.M. Fraker, A.P. Martini, et al. 1992. Publication No. F-02216-1-85. New Brunswick, NJ: Effects of Cash-out on Food Use by Food Stamp New Jersey Agricultural Experiment Station. Program Participants in San Diego. USDA, Food and Nutrition Service. Beebout, H., E. Cavin, B. Devaney, et al. 1985. Evaluation of the Nutrition Assistance Program in U.S. Department of Agriculture, Food and Nutrition Puerto Rico: Volume II, Effects on Food Expenditures Service. 2003. Program data. Available: http://www. and Diet Quality. Washington, DC: Mathematica fns.usda.gov/pd. Accessed April 2003. Policy Research, Inc. U.S. Department of Agriculture, Food and Nutrition Bishop J.A., J.P. Formby, and L.A. Zeager. 1996. Service. 1983. Evaluation of the Puerto Rico Nutrition “Relative Undernutrition in Puerto Rico under Assistance Program. USDA, Food and Nutrition Assistance Programmes,” Applied Economics Service. 28(8):1009-17. U.S. General Accounting Office. 1992. Food Burstein, N., C. Price, P.H. Rossi, et al. 2004. Assistance: Nutritional Conditions and Program “Chapter 3, The Food Stamp Program,” in M.K. Fox, Alternatives in Puerto Rico. Report to Congressional W.L. Hamilton, and B.-H. Lin (eds.), Effects of Food Committees No. GAO/RCED-92-114. Assistance and Nutrition Programs on Nutrition and Health: Volume 3, Literature Review. FANRR-19-3. U.S. General Accounting Office. 1978. Problems USDA, Economic Research Service. Persist in the Puerto Rico Food Stamp Program, the Nation’s Largest. Report by the Comptroller General Hama, M.Y. 1993. Food and Nutrient Consumption of the United States No. CED-78-84. Patterns of Households in Puerto Rico (Nutrient Quality). Unpublished doctoral dissertation, University of Maryland.

292 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 12 Commodity Supplemental Food Program

The Commodity Supplemental Food Program (CSFP) elderly.151 Elderly participation in the CSFP began began in 1968. With the goal of improving the health with a pilot project in FY 1982. of low-income women and their infants and children, CSFP provided supplemental foods, information about Today, individuals eligible to receive free commodity good nutrition, and a link to health care. Over time, the foods through the CSFP include low-income individu- program’s focus has expanded to include the low- als who are elderly (at least 60 years old); women who income elderly, who currently make up the bulk of are pregnant, breastfeeding, or up to a year postpar- program participants. Little research has been done on tum; infants; and children who have not yet reached the CSFP, and none has been done at the national their sixth birthday. Women, infants and children are level. not eligible to participate in the CSFP if they participate in the WIC program. Eligible individuals must reside in the State or Indian reservation that adminis- Program Overview ters the CSFP. States may determine other local The impetus for creation of the CSFP stemmed largely requirements, such as nutritional risk or residence in a from response to concerns in the late 1960s about particular area. For women, infants, and children, hunger and malnutrition among vulnerable low-income income eligibility requirements are established by indi- populations. The “Hunger in America” report, released vidual States (typically 185 percent of poverty). For by the National Board of Inquiry, as well as the “Poor elderly persons, income eligibility is federally set at People’s Campaign,” led by several advocacy groups, 130 percent of poverty (USDA/Food and Nutrition were especially influential in generating the groundswell Service (FNS), 2003a). of concern that led to the program’s creation (Mahony Monrad et al., 1982). Individual States may or may not participate in the program. To initiate the CSFP or to continue it each The Supplemental Food Program, as it was initially year, States apply to USDA by submitting a plan for known, was developed jointly by the U.S. Departments program operations. Each State plan includes informa- of Agriculture (USDA) and Health, Education, and tion about all local agencies expected to administer the Welfare (the forerunner of the current U.S. Department program in the coming year, income criteria and nutri- of Health and Human Services). The program provided tional risk criteria to be used in determining eligibility, food packages, including evaporated milk, corn syrup, plans for outreach and nutrition education, procedures and “reinforced” cereals, to low-income women, for monitoring local agencies, procedures for involving infants, and preschool children. Food packages were local agencies in planning for the following year, plans distributed to participants—upon “determination [of for financial management and detection of duplicate need] by a competent medical authority”—through participation, and audit procedures (7 CFR, Part 247). health clinics, visiting nurses, and health centers that served low-income populations (Mahoney Monrad et Local agencies use a variety of outreach and education al., 1982). strategies to reach potential participants. Typically, a

Over time, other types of social service organizations 151Startup problems with the CSFP may have catalyzed the development of the WIC program. In 1970, after about a year of CSFP operations, USDA have come to serve as local CSFP agencies. In the began expressing concerns that many CSFP programs were surpassing current configuration, not all local agencies that pro- budgets and providing food to individuals who may not have been eligible. vide commodity foods also provide direct health As a result, USDA temporarily stopped expansion of existing CSFP projects, services, but all are encouraged to provide health prohibited initiation of new projects, and limited eligibility to pregnant women, postpartum women up to 12 months after delivery, and infants up information and links. In addition, with the inception to 12 months of age. At the same time, USDA began experimenting with and growth of the Special Supplemental Program an alternative pilot program that would provide pregnant and postpartum for Women, Infants, and Children (WIC) and the women with certificates redeemable in retail stores for infant formula, infant cereal, and milk. Concern within Congress about having curtailed growth of interest in issues related to aging, the the CSFP program hastened the expansion of this pilot program into what CSFP has shifted emphasis toward the low-income eventually became the WIC program (Mahony Monrad et al., 1982).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 293 client comes to a local agency, has his/her eligibility care providers may have made judgments about need confirmed, and then receives a monthly food package, in recommending some people for participation and along with nutrition education. Food packages are tai- not others. If the latter scenario is true, selection bias lored to meet individual needs. Packages contain foods could be an important factor even though the two such as infant formula and cereal, nonfat dry and evap- groups were matched on recorded characteristics. orated milk, juice, farina, oats, ready-to-eat cereal, rice, pasta, egg mix, peanut butter, dry beans or peas, canned Program exposure data, collected for participants, meat, poultry or tuna, cheese, and canned fruits and included number of food package pickups, number of vegetables (USDA/FNS, 2003a). Local agencies use a health care visits, and participation in social welfare variety of methods to provide nutrition education. programs. Outcome measures collected for women included hemoglobin and hematocrit, pregnancy The CSFP is one of the smaller of USDA’s food and weight gain, and birth outcomes. Outcome measures nutrition assistance programs. As noted, the program for children included hemoglobin and hematocrit, does not operate in all 50 States. In FY 2003, 32 immunization status, height-for-age, weight-for-age, States, the District of Columbia, and two Indian reser- and weight-for-height. Analysis methods included t- vations were authorized to operate the program tests and multivariate analysis of covariance. (USDA/FNS, 2003a). In FY 2002, 427 million individuals, the majority of whom were elderly, participated For pregnant women, impacts of CSFP were positive in the CSFP each month. The total Federal expenditure and statistically significant. Participants delivered for the program was $110 million, or less than 1 per- infants with greater gestational age, birthweight, and cent (0.3 percent) of total USDA expenditures for food birthweight adjusted for gestational age. They also had and nutrition assistance (USDA/FNS, 2003b). lower incidence of low birthweight and shorter length of stay in the hospital after birth. The association between participation and improved birth outcomes Research Review was significantly greater among high-risk pregnancy Research on the CSFP is scant. Only one study address- groups, including women with a delivery age of ing impacts of the CSFP on participants was identified younger than 16 years or women who were anemic at in the literature (Mahony Monrad et al., 1982). This the beginning of their participation, who received 1982 evaluation, completed for USDA, collected retro- inadequate prenatal care, who had five or more previ- spective administrative and medical records data from ous pregnancies, or whose weight gain during preg- two CSFP project sites in Memphis, TN, and one nancy was less than 11 pounds. The amount of food CSFP project site in Detroit, MI. All three of the sites received and the amount of prenatal care received both were large. Together, they accounted for 43 percent of had statistically significant associations with improved all CSFP participants. birth outcomes.

Participants from the selected CSFP sites were compared For children, the amount of CSFP services received with a sample of nonparticipating pregnant women and was associated with some positive outcomes, but dif- children drawn from the same local health care facili- ferences in these associations across study sites, as ties. Pregnant women in the treatment and comparison well as severe problems with sample attrition, led the groups were matched with respect to race, marital sta- authors to avoid conclusive statements about the over- tus, age, number of previous pregnancies, smoking all effects of the program on children. In two of the behavior, and prepregnancy weight. Children in treat- three sites, CSFP participation was associated with a ment and comparison groups were matched with significantly lower incidence of low weight-for-height. respect to sex, race, and birthweight. The final sample Only one study focusing on elderly CSFP participants included 842 pregnant women and 472 children. was identified but it was entirely descriptive. Koughan The authors provide no information on the processes by and Atkinson (1993) studied 104 elderly CSFP partici- which some women and children became CSFP partic- pants in New Orleans, LA. The researchers measured ipants and other, apparently equivalent, individuals at participants’ height and weight and completed a the same facilities did not. One possibility is that case- screening checklist (the DETERMINE checklist devel- loads were limited, in which case, we might expect little oped by the Nutritional Screening Initiative (NSI)) or no selection bias. Alternatively, participation may specifically designed to determine nutritional risk have been based on the individual’s request, or health among the elderly (Posner et al., 1993).

294 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 12: Commodity Supplemental Food Program

The authors reported that 80 percent of the CSFP par- Summary ticipants studied were at moderate or high nutritional risk as measured by the DETERMINE checklist. Existing literature provides very little information on Participants were an average of 69 years old, and which to base an understanding of the impact of the about three-fourths were African-American females. CSFP on participants’ nutrition and health status. Only The median body mass index (BMI) was 30, indicat- one study has attempted to estimate the impact of the ing obesity, and, of those with a BMI over 30, 50 per- CSFP. It found positive, statistically significant cent were considered to be at high nutritional risk. All impacts for pregnant women and suggestive evidence of those who had a BMI under 21, indicating that they of positive impacts for children. That study is quite were underweight for their height, were considered to dated, however, and may have been subject to selec- be at nutritional risk.152,153 tion bias. Moreover, the study population did not include the elderly, who now account for about three- quarters of program participants.

Clearly, the CSFP is a program that needs evaluation. Not only is research needed on the impacts of the program on participant nutrition and health status, but also on why States and Indian reservations choose to operate 152 The number of participants with BMI under 21 was not reported. the CSFP and how it interacts with the WIC program 153This is a more conservative cutoff for defining overweight than the cutoff recommended by the Centers for Disease Control and Prevention and the Nutrition Services Incentive Program, former- (Schoenborn et al., 2002). ly known as the Nutrition Program for the Elderly.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 295 Chapter 12: Commodity Supplemental Food Program

References Schoenborn, C.A., P.F. Adams, and P.M. Barnes. 2002. Body Weight Status of Adults: United States, 1997-98. Koughan, N., and C. Atkinson. 1993. “Nutrition Advance Data From Vital and Health Statistics No. Screening Initiative and the Louisiana Food for 330. Centers for Disease Control and Prevention, Seniors Experience,” Journal of the Louisiana State National Center for Health Statistics, September 6, Medical Society 145(10):447-49. 2002.

Mahony Monrad, D., S.H. Pelavin, R.F. Baker, et al. U.S. Department of Agriculture, Food and Nutrition 1982. Evaluation of the Commodity Supplemental Service. 2003a. “Food Distribution Fact Sheet: Food Program: Final Report—Health and Nutrition Commodity Supplemental Food Program.” Available: Impacts of Three Local Projects. USDA, Food and http://www.fns.usda.gov/fdd/programs/csfp/ Nutrition Service. pfs-csfp.03.pdf. Accessed April 2003.

Posner, B.M., A.M. Jette, K.W. Smith, et al. 1993. U.S. Department of Agriculture, Food and Nutrition “Nutrition and Health Risks in the Elderly: The Service. 2003b. Program data. Available: http://www. Nutrition Screening Initiative,” American Journal of fns.usda.gov/pd. Accessed April 2003. Public Health 83(7):972-78.

296 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 13 Food Distribution Program on Indian Reservations

The Food Distribution Program on Indian Reservations related to employment and training or time limits for (FDPIR) provides monthly supplemental food pack- able-bodied adults without dependents (ABAWDs). All ages to low-income households living on Indian reser- households residing on Indian reservations are eligible vations and to eligible American Indian households to participate in the program if they meet income and living in approved areas near reservations. Household resource standards. Households living in approved eligibility to participate in the FDPIR is based on the areas near reservations or in Oklahoma are eligible to Federal income and asset requirements used in the participate if at least one member of the household is a Food Stamp Program (FSP). member of a federally recognized tribe.155

Research literature focusing specifically on the FDPIR Households are individually certified by local offices is very sparse. The few FDPIR-specific papers and and are recertified periodically at intervals not to reports identified through the literature search describe exceed 1 year. Eligible households may choose to the role of the FDPIR in the food supply on American receive either FDPIR benefits or food stamps, but not Indian reservations. No scientific research has evaluat- both. Participating households receive a monthly food ed the impact of the program on nutrition- and health- package weighing between 50 and 75 pounds. In FY related outcomes. 1998, FDPIR food packages were updated in response to an extensive review conducted the prior year. This review was recommended by the Commodity Program Overview Improvement Council (CIC), which was established by The FDPIR was authorized under the Food Stamp Act the U.S. Department of Agriculture (USDA) to address of 1977.154 In establishing the FDPIR, Congress cited concerns about the quality of foods offered in the concerns that the FSP might not adequately meet the FDPIR. A primary concern was the high fat content of food assistance needs of low-income American Indian food packages (Dillinger et al., 1999; Smith, et al., households living on or near reservations (Usher et al., 1996, 1993; USDA, Food and Nutrition Service 1990). The primary concern was that the remote loca- (FNS), 1995). Concerns were also raised about the tion of many reservations made it difficult for American lack of fresh produce and fresh or frozen meats and Indian households to participate in the FSP. In many poultry (Dillinger et al., 1999) and about levels of instances, the distance between the reservation and the sodium and sugar (USDA/FNS, 1995). local FSP offices was substantial and/or food stores where FSP coupons could be redeemed were scarce or The updated food packages added several new prod- far away. Thus, the FDPIR was designed to provide an ucts, including low-sodium and low-fat foods and alternative to the FSP for low-income American Indian frozen, cut-up chicken. Changes were designed to households living on or near reservations. make food packages easier to use and more compatible with the preferences and nutritional needs of American Income eligibility for the FDPIR is based on federally Indians. The fat content of food packages was reduced, defined income eligibility requirements used in the FSP. relative to total energy content, and servings of vegeta- However, the FDPIR does not impose FSP requirements bles and grains were increased (USDA/FNS, 2002).

In FY 2003, more than 70 different food items were 154Earlier versions of commodity distribution programs on Indian reser- offered, including canned beef, poultry, and fish; vations were included in the 1949 and 1963 Agriculture Acts, as well as the 1973 Agriculture and Consumer Protection Act. The Federal Government canned fruits, vegetables, and juices; dried fruits; has provided limited supplies of food in various forms to American Indians dehydrated potatoes; canned soups; canned spaghetti since the time when most Indians living east of the Mississippi River were sauce; packaged macaroni and cheese and other types forcibly removed to reservations in the West and Midwest. At one point, the food distribution programs served U.S. territories in the Pacific Islands as well as Indian reservations. Most of the Pacific Island sites were phased 155In Oklahoma, which has few reservations, low-income households out during the 1980s and 1990s, as the islands converted from U.S. territo- that include at least one American Indian and reside in designated areas ries to commonwealths (U.S. Department of Agriculture (USDA), Food (including some urban areas) may participate in the FDPIR (USDA/FNS, and Nutrition Service (FNS), 2003a). 2003b).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 297 Chapter 13: Food Distribution Program on Indian Reservations of pasta; cereals, rice and other grains; cheese; egg and/or special nutrition and health challenges facing mix; peanuts; peanut butter; low-fat refried beans; and American Indians. Major themes from the available nonfat-dry and evaporated milks (USDA/FNS, 2003b). literature are briefly summarized below. Staples, such as flour, cornmeal, bakery mix, corn syrup, vegetable oil, and shortening were also offered. Characteristics of FDPIR Households Frozen ground beef and chicken and/or fresh produce In the only nationally representative study of the were also available to most programs that have facili- 156 FDPIR, Usher and his colleagues (1990) found that ties to store and handle these foods. FDPIR households were very poor. Nearly 1 in 10 FDPIR households reported having no income. More In addition to providing food, the FDPIR makes print- than one-third had gross incomes that were equivalent ed materials available to participants, such as guidance to or less than 50 percent of the 1989 Federal poverty on how to use FDPIR foods as part of a healthy diet, level. Only one in five households had incomes above commodity fact sheets that provide storage and prepa- the poverty level. ration tips, nutrition information and recipes, and a “Nutrition Facts” booklet that lists the ingredients and About half of all FDPIR households included children. nutrient composition of available commodities Almost one-quarter (23 percent) of FDPIR households (USDA/FNS, 2003b). Sponsoring agencies can also were single adults living alone. Compared with the apply for additional Federal funding to be used specifi- general population of low-income households, more cally for nutrition education. FDPIR households included one or more elderly people and fewer FDPIR households were single-parent, The FDPIR is administered at the State and local levels female-headed households. Roughly 40 percent of all by State agencies and Indian Tribal Organizations FDPIR households included an elderly person compared (ITOs). USDA provides food and administrative funding with 16 percent of low-income households in the general to the State agencies and ITOs, which are then respon- population. Single-parent, female-headed households sible for program operations, including food storage and accounted for roughly 9 percent of all FDPIR house- distribution, eligibility certification, and nutrition educa- holds compared with 47 percent of low-income house- tion. In FY 2003, the FDPIR was administered by 98 holds in general. Researchers documented a strong ITOs and 5 State agencies and provided benefits to tendency for households that were receiving Aid to approximately 243 American Indian tribes (USDA/ Families with Dependent Children (AFDC) benefits to FNS, 2003b). In FY 2002, approximately 110,000 participate in the FSP rather than the FDPIR. individuals participated in the program each month, at an annual cost of $69 million (USDA/FNS, 2003c). Usher et al. found that most FDPIR households had adequate food storage and preparation facilities. However Research Review some FDPIR households lacked at least one of five basic facilities: 20 percent did not have hot running water, 15 Research focusing specifically on the FDPIR is sparse percent had no indoor running water, 9 percent did not and there have been no impact evaluations of the pro- have a refrigerator, 6 percent did not have a stove or gram. One nationally representative study of the other cooking facility, and 7 percent had no electricity. FDPIR has been completed (Usher et al., 1990). The All of these conditions were much more frequent in the primary objectives of that study, which was based on Western Region than in other regions. Three-quarters of data collected in 1989, were to describe program oper- the households that lacked running water and 90 percent ations, describe sociodemographic characteristics of or more of the households without refrigerators or elec- FDPIR households, identify dietary needs and prefer- tricity lived in the Western Region.157 ences of low-income American Indians and examine how the FDPIR addresses those needs, and compare Importance of the FDPIR in the availability and acceptability of the FDPIR versus the Food Supply on Reservations FSP in providing food assistance. Other available liter- Many American Indian families may depend on the ature generally describes the role of the FDPIR in the monthly FDPIR food packages as their primary source food supply on Indian reservations, characteristics of the diets of specific subgroups of American Indians, 157Refers to one of FNS's seven regions. The Western Region includes Alaska, Arizona, California, Hawaii, Idaho, Nevada, Oregon, and 156Even when offered, some families are not able to use fresh or frozen Washington (and ITOs operating in those States). In 1989, about 30 percent foods because they do not have refrigerators (Ballew et al., 1997). of all local FDPIR programs were located in the Western Region.

298 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 13: Food Distribution Program on Indian Reservations of food. In its 1990 review of food assistance pro- Research has provided some evidence that the impor- grams on four Indian reservations, the General tance of the FDPIR as a component of the nutrition safety Accounting Office (GAO) noted that, for many net has increased on some reservations in recent years. Indians, the food assistance programs “constitute their Davis et al. (2002) found that FDPIR caseloads increased primary and long-term food supply because of persist- on the Northern Cheyenne reservation in Montana, while ent unemployment on the reservations” (GAO, 1990). enrollment in the FSP declined.158 The authors report that many factors contributed to this shift. One was lack of In 1993, numerous tribal officials from reservations in transportation (access to a vehicle and/or money for gas) the West and Northwest testified at a Senate hearing on to shop off the reservation, where prices are lower. In “Barriers to Participation in Food Stamp and Other addition, work requirements were seen as a disincentive Nutrition Programs of the Department of Agriculture because of high unemployment rates and a perception by People Residing on Indian Lands.” The officials that finding even a minimum wage job would result in a indicated that the American Indian residents on their loss of benefits for the household. reservations relied on the FDPIR as their primary source of food (U.S. Senate Committee on Indian Affairs and Characteristics of the Diets Senate Committee on Nutrition and Forestry, 1993). of American Indians

In his testimony at the joint hearing, Mr. John Yellow A number of reports and journal articles have assessed Bird Steele, President of the Oglala Sioux Tribal coun- the quality of the diets of American Indians, with no cil, testified that “the USDA food distribution pro- regard to presence or absence of FDPIR (although, as grams, all of them, are very much needed on Pine noted in the 1993 Senate hearings, one can safely Ridge Reservation. They are viewed not as subsis- assume that FDPIR foods play an important role in the tence. They are viewed as a primary source of food.” diets of most American Indians living on or near reser- Similar views were expressed by virtually every per- vations). Several conclusions appear repeatedly in the son who testified at this hearing, representing Indian literature. Most of the studies summarized here are tribes, reservations and trust lands, and organizations based on data that were collected before the changes in that served American Indians. FDPIR food packages. However, findings from the few more recent studies that are available are consis- Wolfe and Sanjur (1988) studied the diets and food tent with findings from earlier research. and nutrient intakes of 107 women attending food distributions on the Navajo reservation. They found that The general finding is that the high prevalence of pro- commodity foods contributed 43 percent of total ener- tein, calorie, and vitamin and/or mineral deficiencies gy intake and close to 50 percent of all nutrients exam- reported by researchers during the 1960s has been sig- ined, except vitamins A and C. Although mean nutrient nificantly reduced (Van Duzen et al., 1976). Inadequate intakes were found to be below the RDA for energy, intake of key nutrients remains a problem, especially calcium, iron, vitamin A, vitamin C, and phosphorus, for vulnerable age groups, such as children, women of the pattern of vitamin and mineral intakes was similar child-bearing age, and the elderly (Ballew et al., 1997). to that of women in the general population. Moreover, However, concerns about nutrient intakes of American the percentages of energy derived from fat, carbohy- Indians largely reflect those of the overall population. drates, and protein in the diets of these low-income For example, many are concerned that the diets of Navajo women were closer to those recommended in many American Indians living on or near reservations the Dietary Guidelines than were the percentages in are too low in variety (number and types of different the diets of women in the Nation as a whole. The foods consumed), fruits, and vegetables and too high authors concluded that: in fat (relative to food energy), highly sweetened and salted foods, and heavily sweetened drinks (Cole et al., The relative adequacy of the women’s diet, despite their 2001; deGonzague et al., 1999; Harnack et al., 1999; very low income levels, was associated with substantial use of foods provided by the Food Distribution Program. Except for vitamins A and C, commodity foods were the source of 158For the Nation as a whole, participation in the FDPIR has not approximately 50 percent of nutrient intakes. Thus, this pro- increased. Since FY 1999, average monthly participation has declined from gram appeared to make an important nutritional contribution 129,000 participants per month to 110,000 participants per month (FY to the contemporary Navajo diet. 2002). (USDA/FNS, 2003c).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 299 Chapter 13: Food Distribution Program on Indian Reservations

Story et al., 1998a, 1998b; Ballew et al., 1997; Specific Nutrition and Health Vaughan et al., 1997; Campos-Outcalt et al., 1995; Concerns Among American Indians Brown and Brenton, 1994; Jackson, 1993; Teufel and The increasing prevalence of obesity, particularly Dufour, 1990; Wolfe and Sanjur, 1988). among children, is a major health concern for the In addition, several researchers (Vaughan et al., 1997; entire U.S. population. However, the problem is partic- Calloway and Gibbs, 1976) observed preferences among ularly troubling in the American Indian population American Indians for fried foods, including fry bread, because of the high prevalence of other health prob- fried potatoes, and fried meats. These foods are typi- lems for which obesity is a serious risk factor. These cally fried in lard, commodity shortening, or butter include, but are not limited to, diabetes, coronary heart rather than vegetable oils (Wolfe and Sanjur, 1988). In disease, and hypertension. The particular histories and addition, commodity cheese has been a significant geographic and economic situations of most Indian source of fat and sodium for some groups of American reservations include numerous factors that encourage Indians (Vaughan et al., 1997; Wolfe and Sanjur, 1988). patterns of diet, food consumption, and inactivity that are highly conducive to adiposity and the onset of obe- Researchers at USDA’s Center for Nutrition Policy and sity (Story et al., 1998a; Vaughan et al., 1997; Promotion (CNPP) studied the diets of the small sub- Campos-Outcalt et al., 1995). sample of American Indians (including Alaska Natives) included in the 1994-96 Continuing Survey of Food Story et al. (1998a) note similarities between the Intakes by Individuals (CSFII). Although the sample was observed emergence of obesity and associated health small (n=107), results indicate that American Indians’ problems among American Indians and patterns that overall scores on the Healthy Eating Index (HEI) were have been observed in developing countries. As a not significantly different from the rest of the U.S. popu- result of relatively rapid shifts to high-fat diets and lation (Basiotis et al., 1999). In addition, the prevalence sedentary lifestyles, American Indians as well as several of food insecurity/food insufficiency and hunger among populations and minority groups in Africa, Asia, and American Indians was similar to that of other minority Latin America have begun to manifest an increased groups in the U.S. population (Basiotis et al., 1999). prevalence of type 2 diabetes, which is linked to obesity. Popkin (1994) describes this phenomenon as the In recent years, research has focused increasingly on “nutrition transition” that causes both under- and over- traditional foods and traditional food resources (such nutrition to occur and coexist in low-income countries. as cultivating small home gardens, harvesting wild foods, and hunting rabbits, deer, and other game) as a Brown and Brenton (1994) describe the rapid emer- means of improving the diets, food security, and/or gence of diabetes among American Indians since the self-sufficiency of American Indians (Lopez et al., 1940s. Burrows and her associates (2000) describe an 2002; Grant et al., 2000; deGonzague et al., 1999). increase of 29 percent over 7 years (from 1990 to Lopez and his colleagues (2002) recommended that 1997) in the prevalence of American Indians and FDPIR programs be allowed to purchase locally, with Alaskan Natives with diagnosed diabetes. Over the an emphasis on healthful traditional foods, up to 10 same period, the increase observed for the general percent of the foods they distribute. U.S. population was 14 percent (Burrows et al., 2000). According to the Centers for Disease Control and Research has provided some evidence that traditional Prevention (1998), the age-adjusted prevalence of diets may reduce metabolic risk factors for diabetes physician-diagnosed diabetes among American Indians and cardiovascular disease—for example, blood levels and Alaskan Natives is 2.8 times greater than the of glucose, lipids (fats), and insulin) (Murphy et al., prevalence among non-Hispanic Whites. 1995; Gittelsohn, et al., 1998; Swinburn et al., 1991; McMurry et al., 1991). In addition, a study that fol- Members of the Pima tribe are reported to have the lowed a group of Pima Indians over a 6-year period, highest known diabetes rate of any population in the found that, among women, individuals who consumed world. However, Campos-Outcalt et al. (1995) found the an “Anglo” diet were more likely to develop diabetes prevalence of diabetes among the Pasqua Yaqui tribe in than those who consumed a traditional diet or a mixed Tucson, AZ, to be as high as that of the Pima. Lopez et diet (Williams et al., 2001). al. (2002) reported similar statistics for the Tohono O’odham Nation (formerly known as the Papago

300 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 13: Food Distribution Program on Indian Reservations

Nation). One of every two Pimas over age 35 has dia- Recent work completed under the auspices of USDA’s betes, compared with 1 in 25 in the overall U.S. popu- Economic Research Service’s small grants program lation (Brown and Brenton, 1994). The rate of gesta- (Davis et al., 2002; Lopez et al., 2002; Grant et al., tional diabetes among American Indians is also among 2000) has contributed to a better understanding of the the highest in the world (Brown and Brenton, 1994). role of food assistance programs in the lives of American Indians. These exploratory studies should continue and researchers should begin to explore the Summary impact of the FDPIR (and other food and nutrition None of the literature examined for this review specifi- assistance programs) on the nutrition and health char- cally evaluated the influence of FDPIR on nutrition acteristics of FDPIR participants. and health outcomes of participants. The available literature provides largely descriptive information about A rigorous evaluation of the health- and nutrition-relat- the program and the individuals it serves. Anecdotal ed impacts of the FDPIR may be difficult to implement. evidence suggests that the FDPIR supplies a substan- The penetration of the program (as well as the alterna- tial part of the dietary intake of many American tive FSP) on Indian reservations is likely to make Indians living on or near reservations. identifying an appropriate control/comparison group difficult. Still, a better understanding of the program’s Available data on food and nutrient consumption pat- impact on participants’ lives is important because (1) terns of American Indians indicate that American Indians this population is at such high nutritional risk and (2) consume diets that are high in fat and limited in vari- their dependence on the FDPIR makes them uniquely ety. These shortcomings are not significantly different vulnerable to program effects, both positive and nega- from those observed in the population as a whole. tive. At a minimum, studies of the contribution of However, the increased prevalence of nutrition-related FDPIR foods to American Indians’ diets should be health problems among American Indians—namely updated to reflect currently available food packages. obesity, diabetes, hypertension, and related health conditions—calls for a heightened level of concern.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 301 Chapter 13: Food Distribution Program on Indian Reservations

References deGonzague, B., O. Receveur, D. Wedll, et al. 1999. “Dietary Intake and Body Mass Index of Adults in Ballew, C., L.L. White, K.F. Strauss, et al. 1997. Two Ojibwe Communities,” Journal of the American “Intake of Nutrients and Food Sources of Nutrients Dietetic Association 99(6):710-16. Among the Navajo: Findings from the Navajo Health and Nutrition Survey,” Journal of Nutrition Dillinger, T.L., S.C. Jett, M. J. Macri, and L.E. Grivetti. 27(supplement):2085s-93s. 1999. “Feast or Famine: Supplemental Food Programs and Their Impacts on Two American Indian Basiotis, P., M. Lino, and R. Anand. 1999. “The Diet Communities in California,” International Journal of Quality of American Indians: Evidence from the Food Science and Nutrition 50(3):173-87. Continuing Survey of Food Intake by Individuals,” Nutrition Insights 12(March 1999). Gittelsohn, J., M.S. Wolever, S.B. Harris, et al. 1998. “Specific Patterns of Food Consumption and Brown, A.C., and B. Brenton. 1994. “Dietary Survey Preparation are Associated with Diabetes and Obesity of Hopi Native American Elementary Students,” in a Native American Community,” Journal of Journal of the American Dietetic Association Nutrition 128:541-47. 94(5):517-22. Grant, R.C., M. Arcand, C. Plumage, and M.G White, Burrows, N.R., L.S. Geiss, M.M. Engelgau, and K.J. Jr. 2000. “Federal Food Programs, Traditional Foods, Acton. 2000. “Prevalence of Diabetes Among Native and the Gros Ventre and Assiniboine Nations of the Americans and Alaska Natives, 1990-1997: An Fort Belknap Indian Reservation,” in A. Vandeman Increasing Burden,” Diabetes Care 23(12):1786-90. (ed.), Food Assistance and Nutrition Research Small Grants Program: Executive Summaries of 1998 Calloway, D.H., and J.C. Gibbs. 1976. “Food Patterns Research Grants. FANRR-10. USDA, Economic and Food Assistance Programs in the Cocopah Com- Research Service. munity,” Ecology of Food and Nutrition 5(4):183-96. Harnack, L., M. Story, and B.H. Rock. 1999. “Diet and Campos-Outcalt, D., J. Ellis, M. Aickin, et al. 1995. Physical Activity Patterns of Lakota Indian Adults.” “Prevalence of Cardiovascular Disease Risk Factors in Journal of the American Dietetic Association a Southwestern Native American Tribe,” Public Health 99(7):829-35. Reports 110:742-48. Jackson, Y. 1993. “Height, Weight, and Body Mass Centers for Disease Control and Prevention. 1998. Index of American Indian Schoolchildren,” Journal of “Prevalence of Diagnosed Diabetes Among American the American Dietetic Association 93(10):1136-40. Indians/Alaskan Natives: United States, 1996,” Morbidity and Mortality Weekly Report 47:901-04. Lopez, D., T. Reader, and P. Buseck. 2002. Community Attitudes Toward Traditional Tohono O’odham Foods. Cole, S.M., N.I. Teufel-Shone, C.K. Ritenbaugh, et al. Sells, AZ: Tohono O’odham Community Action and 2001. “Dietary Intake and Food Patterns of Zuni Tohono O’odham Community College. Adolescents,” Journal of the American Dietetic Association 101(7):802-06. McMurry, M.P., M.T. Cerqueira, S.L. Connor, and W.E. Connor. 1991. “Changes in Lipid and Davis, J., R. Hiwalker, C. Ward, et al. 2002. “Is the Lipoprotein Levels and Body Weight in Tarahumara Food Stamp Program an Adequate Safety Net for Indians after Consumption of an Affluent Diet,” New American Indian Reservations? The Northern England Journal of Medicine 325:1704-08. Cheyenne Case,” in A. Vandeman (ed.), Food Assistance and Nutrition Research Small Grants Murphy, N.J., C.D. Schraer, M.C. Thiele, et al. 1995. Program: Executive Summaries of 2000 Research “Dietary Change and Obesity Associated with Glucose Grants. FANRR-20. USDA, Economic Research Intolerance in Alaska Natives,” Journal of the Service. American Dietetic Association 95:676-82.

302 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 13: Food Distribution Program on Indian Reservations

Popkin, B.M. 1994. “The Nutrition Transition in Low- U.S. Department of Agriculture, Food and Nutrition income Countries: An Emerging Crisis,” Nutrition Service. 2002. “Nutrition Program Facts: The Food Review 52:285-98. Distribution Program on Indian Reservations.” Available: http://www.fns.usda.gov/fdd/programs/ Smith, C.J., R.G. Nelson, S.A. Hardy et al. 1996. fdpir/pdpirfaq.htm. Accessed March 2002. “Survey of the Diet of Pima Indians Using Quantitative Food Frequency Assessment and 24-hour U.S. Department of Agriculture, Food and Nutrition Recall: The Diabetic Renal Study.” Journal of the Service. 1995. Improving USDA Commodities: 1995 American Dietetic Association 96:778-84. Tri-Agency Commodity Specification Review Report. USDA, Food and Nutrition Service. Smith, C.J., E.M. Manahan, and S.G. Pablo. 1993. “Food Habit and Cultural Changes Among the Pima U.S. General Accounting Office. 1990. Food Indians.” In Joe, J.R. and R.S. Young (Eds.) Diabetes Assistance Programs: Recipient and Expert Views on as a Disease of Civilization. Berlin, New York: Food Assistance at Four Indian Reservations: Report Mouton de Gruyter. to Congressional Requesters.

Story, M., D. Neumark-Sztainer, M.D. Resnick, et al. U.S. Senate Committee on Indian Affairs and Senate 1998a. “Psychosocial Factors and Health Behaviors Committee on Nutrition and Forestry. 1993. Barriers Associated with Inadequate Fruit and Vegetable Intake to Participation in the Food Stamp Program and among American-Indian and Alaska-Native Other Programs of the Department of Agriculture by Adolescents.” Journal of Nutrition Education People Residing on Indian Lands: Joint Hearing 30(2):100-06. Before the Committee on Indian Affairs, United States Senate, and the Committee on Agriculture, United Story, M., K.F. Strauss, E. Zephier, et al. 1998b. States One Hundred Third Congress, First Session. “Nutritional Concerns in American Indians and Alaska Natives: Transitions and Future Directions,” Journal of Usher, C.L. D.S. Shanklin, and J.B. Wildfire. 1990. the American Dietetic Association 98(2):170-76. Evaluation of the Food Distribution Program on Indian Reservations (FDPIR). Volume I. Final Report. Swinburn, B.A., V. L. Boyce, R.N. Bergman, et al. USDA, Food and Nutrition Service. 1991. “Deterioration in Carbohydrate Metabolism and Lipoprotein Changes Induced by Modern, High-fat Van Duzen, J., J.P. Carter, and R. Vander Zwagg. Diet in Pima Indians and Caucasians,” Journal of 1976. “Protein and Calorie Malnutrition among Clinical Endocrinology and Metabolism 73:156-65. Preschool Navajo Indian Children, a Follow-up.” American Journal of Clinical Nutrition 29(6):657-62. Teufel, N. I. and D. Dufour. 1990. “Patterns of Food Use and Nutrient Intake of Obese and Non-obese Vaughan, L.A., D.C. Benyshek, and J.F. Martin. 1997. Haulapai Indian Women of Arizona,” Journal of the “Food Acquisition Habits, Nutrient Intakes, and American Dietetic Association 90(9):1229-35. Anthropometric Data of Havasupai Adults,” Journal of the American Dietetic Association 97(11):1275-82. U.S. Department of Agriculture, Food and Nutrition Service. 2003a. Footnote on table “Costs of Food Williams, D.E., W.C. Knowler, C.J. Smith, et al. 2001. Distribution Programs,” Available: http://www.fns. “The Effect of Indian or Anglo Dietary Preference on usda.gov/pd/fd$sum.htm. Accessed April 2003. the Incidence of Diabetes in Pima Indians,” Diabetes Care 24(5):811-16. U.S. Department of Agriculture, Food and Nutrition Service. 2003b. “Food Distribution Programs: FAQs Wolfe, W.S., and D. Sanjur. 1988. “Contemporary Diet About FDPIR.” Available: http://www.fns.usda.gov/ and Body Weight of Navajo Women Receiving Food fdd/programs/fdpir/fdpir-faqs.htm. Accessed June 2003. Assistance: An Ethnographic and Nutritional Investigation,” Journal of the American Dietetic U.S. Department of Agriculture, Food and Nutrition Association 88:822-27. Service. 2003c. Program data. Available: http://www. fns.usda.gov/pd. Accessed April 2003.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 303 Chapter 14 WIC Farmers’ Market Nutrition Program

The WIC Farmers’ Market Nutrition Program (FMNP) the program submit to their Food and Nutrition provides low-income women and their children with Service (FNS) regional office an application that out- coupons that can be used to buy fresh fruits and veg- lines how the FMNP will operate in that State for the etables from authorized farmers and farmers’ markets. following year. Federal funds cover up to 70 percent The program primarily serves women, infants over 4 of the cost of the program, and States are required to months of age, and children up to the age of 5 who are cover at least 30 percent.161 The State’s share of funds certified to receive program benefits from the Special can come from State, local, or private sources. Supplemental Program for Women, Infants, and Children (WIC ) or are on a waiting list. States may States decide individually how they will determine choose to provide FMNP coupons to other groups of which WIC participants will also receive FMNP participants, most often low-income elderly or children coupons. Some States select coupon recipients based on over the age of 5. Costs for additional participants the State priority system for allocating WIC benefits (for must be met by State matching funds.159 example, limiting participation to pregnant and breastfeeding women). Other States distribute coupons on a Very little research has been conducted on the FMNP, first-come, first-served basis. WIC participants can and the available literature offers no firm conclusions receive farmers’ market coupons totaling $10 to $20 per about the impact of the program on nutrition-related year—in $1 or $2 denominations—at the beginning of outcomes. the fruit- and vegetable-growing season (usually in June). States may limit the FMNP to specific fruits and vegetables that are grown locally. Program Overview The FMNP is intended to encourage WIC participants Farmers’ markets or individual farmers who wish to to eat more fresh, unprepared, locally grown fruits and participate in the FMNP apply to the State agency that vegetables and to help small farmers by promoting administers the program, or in some cases, to a sub- farmers’ markets.160 The program began in 1989 as a agency that handles regulatory or other issues related pilot program in 10 States and was formally authorized to farmers’ markets. Criteria used to approve farmers’ in 1992, with mandated set-aside funding within the markets or individual farmers include membership in WIC program appropriation (Nutrition Week, 1991). the local Farmers’ Market Association, evidence of nondiscrimination, hours of market operation, market State participation occurs through an annual application location, and the amount and proportion of sales process. Each year, States interested in participating in involving fresh, unprocessed fruits and vegetables covered by the FMNP. Farmers can redeem coupons for cash either via the farmers' market manager or through 159In January 2001, USDA’s Commodity Credit Corporation (CCC) a central processing office. In some States, the FMNP launched the Senior Farmers’ Market Nutrition Program (SFMNP) as a pilot program. The SFMNP is essentially the same as the WIC version of coupons are negotiable checks that the farmer can the program but is targeted toward low-income elderly. Total costs for the deposit in a bank. Coupons are good only for the program were about $13 million in its first year of operation ($2 million of growing season in which they are issued. the initial appropriation of $15 million were carried over to FY 2002) (USDA/FNS, 2003a). In FY 2003, the SFMNP was operating in 35 States, the District of Columbia, Puerto Rico, and 3 Indian Tribal Organizations. A In FY 2003, the FMNP operated in 36 States, Guam, total of $17 million in funding was available, including the FY 2003 appro- the District of Columbia, Puerto Rico, and 5 Indian priation ($15 million) and unspent funds from FY 2002 (approximately $2 Tribal Organizations. More than 13,000 farmers and million) (USDA, FNS, 2003a). 1,911 farmers’ markets were authorized to participate 160When Congress formally approved the FMNP, USDA officials were concerned that, because farmers’ markets are also accepted in the Food in the program. In FY 2002, 2.1million participants Stamp Program (FSP), the FMNP might represent duplicate coverage. At redeemed FMNP coupons. In FY 2002, $15 million of the time, USDA favored strengthening efforts to bring farmers’ markets the Federal appropriation for the WIC program was into the FSP rather than continuing to promote the FMNP. Some Congressional representatives, on the other hand, expressed support for the FMNP because the FMNP had the additional purpose of increasing consumer awareness of farmers’ markets and thus enhancing marketing oppor- 161If approved by FNS, Indian Tribal Organizations that operate the FMNP tunities for small farmers (House Committee on Agriculture, 1992). may be approved for a lower matching rate, but not less than 10 percent.

304 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 14: WIC Farmers’ Market Nutrition Program earmarked for the FMNP. This figure increased to $20 with respect to several background characteristics: million for FY 2001 and $25 million for FY 2003 Recipients were more likely to be Black or Hispanic (U.S. Department of Agriculture (USDA)/FNS, 2003b). and to be college graduates than were WIC recipients. The analysis did not control for these differences.

Research Review Participant satisfaction with the FMNP was high. At Very little research has been conducted on the FMNP. least half of the recipients indicated that they would Three studies have attempted some form of impact continue to shop at farmers’ markets even after they analysis. All three studies focused on consumption of stopped receiving coupons, although the study did not fresh fruits and vegetables as the principal outcome of include any followup to ascertain whether this actually interest. Results were mixed, and design limitations happened. Ninety-two percent were “very” or “some- make it impossible to draw meaningful conclusions what” satisfied with the program, and 80 percent iden- about the program’s effects. tified some benefit they had derived from the program. A survey of participating farmers also found enthusias- The largest evaluation of FMNP was a USDA-funded tic support for the program. study reported in Galfond et al. (1991). This study, conducted when the program was in its demonstration An interesting theoretical exploration based on phase (the 1990 growing season), included an impact Galfond et al. is provided by Just and Weninger evaluation based on a cross-sectional sample of WIC (1997). Traditionally, Just and Weninger point out, participants drawn from several States. A telephone economists would say that a WIC participant who survey was completed with 1,503 women who chooses not to buy fresh fruits or vegetables does so received FMNP coupons during the 1990 growing sea- because she perceives the benefit she will derive from son (“recipients”), 96 women who did not receive those fruits and vegetables to be lower than the price 1990 coupons but had received them in a prior year, being asked for them. Coupons can induce her to make and 1,126 women who never received FMNP coupons the purchase because they lower the effective price to (“nonrecipients”). The women were randomly selected a level equal to or lower than the benefit she perceives. from lists of WIC participants in four States (Iowa, In this view, coupons represent a net loss to society in Massachusetts, Pennsylvania, and Vermont) and from the amount of the difference between market price of WIC participant lists from a few WIC clinics in Texas the fruit or vegetable and the price that the consumer and Washington. would be willing to pay in the absence of the coupon.

Respondents were asked to report the kinds of fruits and The FMNP experience, Just and Weninger argue, illus- vegetables they had eaten the previous day and how trates an economic effect of combining food program many servings of each they had eaten. Also, for specific coupons with nutrition information—that is, to change fruits and vegetables, respondents were asked to report the consumer’s perceptions about the benefit she is how many servings they ate in a typical week. The likely to derive from the same fruits and vegetables. In analysis, based on bivariate comparisons, found that the FMNP, the recipient receives not only coupons but recipients reported eating significantly more servings also information about the nutritional benefits of the of fruits and vegetables than nonrecipients. Recipients fruits and vegetables that a participant can buy with were also significantly more likely than nonrecipients her coupons. Galfond et al. observed that coupon to report that their fruit and vegetable consumption recipients reported greater future intentions to shop at was greater than it had been the previous year. farmers’ markets than nonrecipients. Just and Weninger reason that this difference in intentions The study had several limitations, which the authors reflects a change in recipients’ perceptions of the eco- duly noted. First, the sampling method may have led to nomic value of the fruits and vegetables they can get bias with respect to prior access to farmers’ market at the markets. This change in perceived value results because clinics offering FMNP tended to be those with in higher demand for the fruits and vegetables, result- a farmers’ market nearby. Thus, WIC participants in ing in a higher equilibrium price and inducing farmers areas with farmers’ markets could have been more likely to bring more fruits and vegetables to market. to eat fresh fruits and vegetables than even without the FMNP than WIC participants in areas without farmers’ Just and Weninger go on to estimate this economic markets. Also, the authors found that recipients differed benefit by deriving equations to describe the supply from nonrecipients at statistically significant levels and demand of fruits and vegetables and to quantify

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 305 Chapter 14: WIC Farmers’ Market Nutrition Program benefits to consumers and to society. Using data mation on participants' perceptions about the program’s reported by Galfond et al., they estimate that the net impact on their behavior (NAFMNP, 1996; Nutrition gain to society of the FMNP amounts to about 21 per- Week, 1995). The NAFMNP developed a set of ques- cent of the value of coupons redeemed. Without nutri- tions for assessing the program from the point of view tion information (that is, as an ordinary subsidy), they of farmers and recipients, and USDA distributed these estimate a net loss to society equal to 7 to 18 percent questions for States to use. States that conducted sur- of the value of coupons redeemed. veys provided the data to NAFMNP. The survey and sampling procedures apparently differed from State to The second study to estimate FMNP impacts was a State but are not described in the publications. small study reported by Anliker et al. (1992). The study, Questionnaires also differed from State to State, but conducted in 1989, during the program’s pilot phase, study organizers were able to aggregate responses for included interviews with randomly selected participants questions that were asked across States. in nine WIC programs in Connecticut. Six of the WIC programs distributed FMNP coupons. Participants Results are based on data from 24 States and 2 Indian from these programs constituted the treatment group, Tribal Organizations collected in 1995. Responses which contained 411 respondents. Participants from were obtained from 2,670 farmers (representing 33.2 the three WIC programs that did not distribute FMNP percent of participating farmers) and 24,812 recipients coupons were designated as the control group, which (representing about 3 percent of FMNP coupon recipi- contained 78 respondents. ents) who participated during the 1995 growing season. The data showed that FMNP participants general- At the time that subjects were recruited into the ly had positive impressions of the program’s impact on study—that is, before the treatment group had a their consumption of fresh produce: 71 percent of chance to use their FMNP coupons—they were asked coupon recipients said that, because of the FMNP, they how often they ate fruits and vegetables or drank ate more fresh produce than usual during the summer. juices. A followup survey, completed approximately 2 In addition, 77 percent said they planned to eat more months later, asked the same food consumption ques- fresh produce year-round, and 66 percent said they tions plus questions about use of farmers’ markets and would continue to shop at farmers’ markets even if FMNP coupons. The final analysis sample (including they did not receive additional coupons. respondents who completed both pretest and post-test surveys) include 172 FMNP participants and 44 nonparticipants. The authors report that individuals who Summary responded to the followup survey and those who did The limited available research permits no firm conclu- not differed significantly on several characteristics. sion about the impact of the FMNP on participants’ consumption of fresh produce or on any associated The analysis of program impacts used analysis of nutrition-related effects. Of the two studies that used covariance and controlled for baseline responses on quasi-experimental designs to examine FMNP impacts, frequency of fruit, vegetable, and juice consumption. one found a positive impact on participants’ consump- The authors found no statistically significant relation- tion of fresh fruits and vegetables and the other found ship between receiving FMNP coupons and the reported no significant effect. Both studies had severe method- frequency of fruit and vegetable consumption. None- ological limitations, however—likely selection bias in theless, the authors conclude that the “…Farmers’ the first case and possible selection bias combined Market Project has been generally successful in meet- with a very small sample size in the second—and ing its objectives.” This conclusion appears to be both report on a very early time period in the pro- based principally on the finding that, “more than three- gram’s history. fourths of the participants who received Farmers’ Market coupons went to the farmers’ markets and used The small dollar value of the FMNP benefit—no more their coupons to purchase fresh, locally grown produce.” than $20 per year—suggests that any impact on nutrition and health status is likely to be so small that it The third and most limited study was conducted by the would be extremely costly to measure. Research might National Association of Farmers’ Market Nutrition better be directed toward effects on participants' Programs (NAFMNP), an advocacy group in favor of awareness and use of farmers’ markets. strengthening and expanding the FMNP. The study included FMNP participants’ only and collected infor-

306 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 14: WIC Farmers’ Market Nutrition Program

References National Association of Farmers’ Market Nutrition Programs. 1996. Program Impact Report for the Anliker, J.A., M. Winne, and L.T. Drake. 1992. 1995 WIC Farmer’ Market Nutrition Program. “Evaluation of the Connecticut Farmers’ Market Washington, DC. Coupon Program,” Journal of Nutrition Education 24(4):185-91. Nutrition Week. 1995. “Farmers’ Market Program Fulfilling its Dual Mission, National Survey Galfond, G., J. Thompson, and K. Wise. 1991. Evalua- Indicates,” 25(16):1-2. tion of the Farmers Market Coupon Demonstration Project. USDA, Food and Nutrition Service. Nutrition Week. 1991. “Experiments Link Farmers to Food Program Clients.” 21(22):4-5. House Committee on Agriculture. 1992. Impact of the Farmers’ Market Nutrition Act of 1991 on Markets U.S. Department of Agriculture, Food and Nutrition and the Marketing of Fresh Fruits and Vegetables: Service. 2003l. “FNS Announces FY 2003 Senior Testimony Before the Subcommittee on Domestic Farmers’ Market Nutrition Program (SFMNP).” Marketing, Consumer and Nutrition. One Hundred Available: http://www.fns.usda.gov/WIC/SeniorFMNP/ Second Congress, Second Session. Subcommittee on SFMNPFY03.htm. Accessed April 2003. Domestic Marketing, Consumer and Nutrition. U.S. Department of Agriculture, Food and Nutrition Just, R.E., and Q. Weninger. 1997. “Economic Service. 2003b. “WIC Farmers’ Market Nutrition Evaluation of the Farmer’ Market Nutrition Program,” Program.” Available: http://www.fns.usda.gov/WIC/ American Journal of Agricultural Economics FMNP/FMNPfaqs.htm. Accessed April 2003. 79(3):902-17.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 307 Chapter 15 Special Milk Program

The Special Milk Program (SMP) operates in schools From the beginning, schools were allowed to charge and child care institutions that do not participate in children for milk, but the price could not exceed the other Federal meals programs—the National School cost to the school after the Federal subsidy. In FY Lunch Program (NSLP), the School Breakfast 1975, children eligible for free lunch under the NSLP Program (SBP), or the Child and Adult Care Food were made automatically eligible for free milk under Program (CACFP). Schools that do participate in these the SMP. This free-milk provision significantly programs may also participate in the SMP to provide increased the program’s administrative complexity milk to children in prekindergarten or kindergarten because school administrators now had to track milk programs who do not receive meals. Participating served to free-milk-eligible children separately from institutions provide milk to children and receive a milk served to other children. Federal subsidy for each half pint of milk served. Children from households with incomes at or below For 25 years, the SMP was open to all schools and 130 percent of the Federal poverty level may receive child care institutions, regardless of whether they par- milk free of charge. ticipated in other food assistance and nutrition programs (FANPs). In FY 1982, the Omnibus Budget Research on the SMP is sparse and has generally been Reconciliation Act (OBRA) limited participation to conducted as an adjunct to research on the NSLP and schools and institutions that did not participate in other SBP. Available data indicate that the SMP contributes FANPs and to private schools with annual tuition of to increased nutrient intake, particularly among chil- less than $1,500. These restrictions were eased in 1987 dren from low-income families and elementary school when the $1,500 tuition restriction on private school children. participation was lifted. In FY 1988, eligibility was reinstated for schools and institutions that participate in other FANPs. In these institutions, SMP participa- Program Overview tion is limited to half-day prekindergarten or kinder- The earliest version of the SMP began in 1940. A fed- garten students who do not receive meals. erally subsidized program distributed milk to children at 15 elementary schools in low-income areas of Since the late 1960s, when the SMP served around 3 Chicago. Children were charged 1 cent per half pint of billion half pints of milk per year, the size of the pro- milk. Those who could not pay received milk for free, gram has declined dramatically. At the same time, with the cost paid by private donations. During the other child-focused FANPs, particularly the SBP and next decade, similar programs were introduced in low- the CACFP, have grown substantially. In 1980, even income areas of other large cities. The SMP began before the OBRA changes, the SMP was providing just operating nationally in 1955 when legislation provided 1.8 billion half-pints of milk, or somewhat more than funds to the Commodity Credit Corporation (CCC) to half of the amount served in the late 1960s. By 1990, be used for school milk programs.162 In 1956, eligibili- this number had fallen to 181 million. ty for institutional participation was expanded beyond Today, relative to total Federal expenditures for food schools to include other nonprofit institutions that and nutrition assistance, the SMP is the third smallest cared for children. The SMP came under the supervi- FANP, overall, and the second smallest FANP to pro- sion of the U.S. Department of Agriculture (USDA), vide direct food assistance benefits (only the Senior Food and Nutrition Service (FNS) when it was incor- Farmers’ Market Nutrition Program and the Team porated into the Child Nutrition Act of 1966. The SMP Nutrition Initiative had lower total costs in FY 2002). was permanently authorized in 1971. In FY 2002, the SMP provided approximately 113 million half pints of milk at a cost of $16 million (USDA/FNS, 2003). Participating institutions were reimbursed for the net price of purchased milk for 162The CCC is a Government-owned and -operated entity that was created to stabilize, support, and protect farm income and prices (U.S. Department of milk served to students free of charge and $0.135 per Agriculture (USDA), Farm Service Agency (FSA), 2003). pint for milk served to paying students. In FY 2000,

308 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 15: Special Milk Program the latest year for which information is available, density.163 They reported positive impacts on INQs for almost 7,000 schools and residential child care institu- calcium and magnesium, indicating that SMP partici- tions, 1,100 summer camps, and 500 nonresidential pants consumed more of these nutrients per kilocalorie child care institutions participated in the SMP than nonparticipants. Significant negative effects were (USDA/FNS, 2002). reported for INQs for niacin, iron, and vitamin C. These results are largely consistent with the nutrient content Research Review of milk; milk is rich in calcium and magnesium and provides little to no niacin, iron, or vitamin C. Research on the SMP is extremely limited. Only two studies that assessed program impact were identified. A study by Robinson (1975) is the only other study to Both of these studies are based on data that are more address SMP impacts. This study is even older than than 20 years old, reflecting a time when the program the NESNP study, and the data are much more limited. was about 15 times as large as it is today. The only The study used data collected in March and April other relevant research identified was a study that 1975, after the free-milk provision had been imple- examined the relationship between milk consumption mented and before the SMP was restricted to schools and lactose intolerance among SMP participants. in which no other Federally funded meal service program was operating. The objectives of the study were Of the two studies that assessed SMP impacts, the most to assess the impact of the free-milk provision on the recent and comprehensive is the National Evaluation SMP, assess the impact of the SMP and its free-milk of School Nutrition Programs (NESNP), reported by provision on the NSLP, determine the sources and Wellisch et al. in 1983. This study took place after amounts of milk and food that children consumed, OBRA limited the SMP to schools that did not partici- determine which children used the SMP, determine pate in any other federally sponsored meals program. when during the day children preferred to have milk available and whether schools were meeting these Data were collected on a nationally representative preferences, and determine the extent of milk waste sample of 6,556 students (and their families) in grades associated with all USDA programs. 1-12 in 276 public schools in 90 school districts. Data collection included in-home interviews of parents Schools were selected through a two-stage sampling regarding family composition, economic status, and process. In the first stage, approximately 4,000 schools food expenditures; in-person interviews of students in were randomly selected from the Office of Education’s school, including a 24-hour dietary recall; and mail database of public and private schools. In the second surveys of State, district, and school foodservice stage, schools were stratified according to various con- administrators, with telephone followup. figurations of participation in the SMP and the NSLP. From these strata, a subsample of 768 schools was Although the study looked at a wide variety of out- selected for the study. Data were collected by means of comes, the SMP analysis focused solely on dietary a school administrator questionnaire, a foodservice intake. Participants in the SMP were compared with supervisor questionnaire, and student questionnaires children who did not participate, using multivariate administered to randomly selected classes or students regression to estimate program effects. The researchers within each school. A milk waste study was also con- cautioned that results should be interpreted with care ducted in schools that participated in the NSLP, the because of potential selection bias. SBP, or the SMP.

With that caveat, the study reported that the SMP signifi- More than 20,000 student questionnaires were collect- cantly increased students' intakes of food energy, calci- ed. Results indicated that students who attended um, riboflavin, protein, magnesium, and vitamin B6. schools that participated in the SMP drank more milk Differences were more pronounced for below-median- than students who attended schools that did not partici- income students and for elementary students. Among pate in any Federal meal service program. The author secondary students, impacts for energy were related to noted, however, that the increased milk consumption family income-the higher the family income, the larger may have resulted, at least in part, from the NSLP-90 the difference between participants and nonparticipants. percent of SMP schools also participated in the NSLP.

Wellisch and colleagues constructed an index of 163INQ = Recommended Dietary Allowance for nutrient/Recommended nutritional quality (INQ) to measure relative nutrient Energy Allowance, based on energy and nutrients consumed over 24 hours.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 309 Chapter 15: Special Milk Program

Robinson avoided drawing conclusions based on SMP- The children’s pattern of milk rejection, along with only schools on the grounds that these schools may patterns of milk rejection among adults of various have contained nonrepresentative groups of students races cited in other studies, was plotted against a dis- (in these schools, students who were eligible for free tribution of lactose-intolerance rates among African- SMP milk constituted only 3 percent of enrollment). Americans and Whites. Patterns of milk rejection Nevertheless, the report mentioned that the few free- among the SMP children appeared to follow known milk-eligible children at SMP-only schools reported patterns of lactose intolerance. The authors concluded drinking 77 percent more milk at school than corre- that some SMP participants would benefit more from sponding ineligible children. For these same children, alternative sources of protein and calcium than from away-from-school consumption was 7 percent less, the milk provided by the SMP. and overall milk consumption was 12 percent more than for ineligible children. Summary Finally, a study by Paige and Graham (1974) looked at The available information on the nutrition-related the incidence of lactose intolerance among SMP par- impacts of the SMP is very limited and is of question- ticipants. Although dated and not specifically focused able relevance to today’s program. The strongest and on the impact of the SMP, the findings may be of inter- most recent study was conducted in the early 1980s, est to researchers interested in the SMP. The authors when the program was roughly 15 times as large as it gathered data on the amount of SMP milk consumed is now. That study suggested that the SMP increased by 320 African-American children and 125 White chil- children’s intakes of food energy and several nutrients. dren in the grades 1-5. The study was conducted in two An earlier, though substantially weaker, study suggest- schools in the lowest socioeconomic census tracts in ed that the program increased children’s consumption Stamford, CT. Two separate measurements were taken. of milk. In both cases, however, researchers put sub- The analysis compared rates of milk consumption stantial and appropriate caveats around their findings. among the sampled children with race-specific rates of lactose intolerance reported in medical literature. Given the size of the SMP today, relative to other child-oriented FANPs, it is not clear that an updated The authors report that 36 percent of the African- study of the program’s impacts should be viewed as a American children drank less than 50 percent of their priority. If feasible, however, studies of the NSLP, SMP milk (a half pint) compared with 18 percent of SBP, and/or CACFP might incorporate a study on the the White children (a statistically significant difference). SMP, as was done in the NESNP.

310 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 15: Special Milk Program

References U.S. Department of Agriculture, Food and Nutrition Service. 2003. Program data. Available: http://www. Paige, D.M., and G.G. Graham. 1974. “School Milk fns.usda.gov/pd. Accessed April 2003. Programs and Negro Children: a Nutritional Dilemma,” Journal of School Health 44(1):8-10. U.S. Department of Agriculture, Food and Nutrition Service. 2002. “Special Milk Program: Fact Sheet.” Robinson, J.S. 1975. Special Milk Program Evaluation Available: http://www.fns.usda.gov/cnd/milk/about- and National School Lunch Program Survey. USDA, milk/faqs.htm. Accessed March 2002. Food and Nutrition Service. Wellisch, J.B., S.D. Hanes, L.A. Jordan, et al. 1983. U.S. Department of Agriculture, Farm Service Agency. The National Evaluation of School Nutrition 2003. “Fact Sheet: November 1999. Commodity Credit Programs: Final Report. Volumes 1 and 2. Santa Corporation.” Available: http://www.fsa.usda.gov/pas/ Monica, CA: Systems Development Corporation. publications/facts/htm/ccc99.htm. Accessed April 2003.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 311 Chapter 16 Team Nutrition Initiative and Nutrition Education and Training Program

This chapter describes two programs: the Team with educational and technical resources that could be Nutrition (TN) Initiative and the Nutrition Education used to (1) encourage children to eat healthy meals and Training (NET) Program. Both programs, which and (2) assist foodservice staff in preparing nutritious are implemented primarily in schools, differ from and appealing meals. other food assistance and nutrition programs (FANPs) in three important ways. First, the primary focus of The impetus for SMI can be traced to findings from each program is educational in nature—to promote the first School Nutrition Dietary Assessment Study healthful eating patterns. Neither program provides (SNDA-I). This national study of the NSLP and SBP food or enhances food purchasing power. Second, nei- found that NSLP meals were higher in fat, saturated ther program considers income in targeting benefits. fat, and sodium and lower in carbohydrates than rec- That is, both programs are intended to serve all chil- ommended by the Dietary Guidelines for Americans dren, unlike other programs that offer greater benefits (DGA) and the National Research Council (Burghardt to low-income children, as the National School Lunch et al., 1993; U.S. Departments of Agriculture and Program (NSLP) and School Breakfast Program (SBP) Health and Human Services, 1990; National Research do, or that are limited to children with specific nutri- Council, 1989). At the time the SNDA-I data were col- tional risks, as the Special Supplemental Program for lected, program regulations did not require that NSLP Women, Infants, and Children (WIC) does.164 Finally, meals be consistent with these recommendations, and target audiences for both TN and NET services extend meals generally satisfied the nutrition standards that beyond children to include teachers, school foodser- were in effect—providing one-third of students’ daily vice workers, parents, and community members, all of requirements for food energy and key nutrients. whom may influence children’s food choices. Since the SNDA-I study revealed that school meals After the Senior Farmers’ Market Nutrition Program, were not consistent with accepted guidelines for which began in FY 2002, TN is the youngest FANP. It healthful eating, USDA has been working on many was created in 1995 as part of the comprehensive fronts to enhance the nutritional quality of school School Meals Initiative (discussed below). The NET meals. In 1995, USDA formalized its commitment to program has been authorized for more than 25 years implementing the Dietary Guidelines in school meals but has not received funding since FY 1998. by launching SMI. Key components of SMI include changes in the nutrition standards defined for school The following sections describe the TN Initiative and meals and an expansive change in the procedures used the NET program, in turn, and then summarize rele- to plan and evaluate school menus (SMI nutrition stan- vant research for each program. The TN Initiative is dards and menu planning options are discussed in described first because it is the program that is current- detail in chapter 5). The new nutrition standards mainly active. Relatively little research has been done on tain the traditional goal of satisfying some portion of either program. students’ needs for energy and key nutrients (one-third for the NSLP and one-fourth for the SBP), but also specify goals for fat and saturated fat content that are Overview of the Team consistent with Dietary Guidelines recommendations. Nutrition Initiative To ensure that these changes in program policy and The Team Nutrition (TN) Initiative is a cornerstone of operations would be implemented successfully and USDA’s School Meals Initiative for Healthy Children accepted by program participants, USDA also included (SMI). SMI was launched in 1995, with the explicit in SMI a comprehensive plan for providing technical goal of improving school meals by providing schools assistance, educational resources, and training for school foodservice personnel as well as other stakeholders in 164The Team Nutrition Initiative also provides nutrition education materials the school meals programs—children, parents, teach- to other FANP programs, such as the WIC Program and the FSP. ers, and administrators. This plan is the TN Initiative.

312 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program

TN provides support to schools via three different home activities, community programs and events, and behavior-oriented strategies: media events and coverage. TN has established, and continues to build, relationships with national organi- • Training and technical assistance for school foodser- zations that agree to work on large, visible TN proj- vice professionals to assist them in preparing and ects. Examples include the American Culinary serving meals that meet the new nutrition standards Federation Chef and Child Foundation, which spon- without sacrificing taste or attractiveness. sors cooking and taste-testing activities in local schools, conducted by professional chefs. • Nutrition education for children and parents to build the motivation for adopting healthful eating habits In FY 2002, TN was funded at $10 million (French, and regular physical activity and the skills required 2002). This level of funding has been relatively con- to do so successfully. stant since FY 1996. Schools in all 50 States, the District of Columbia, and the U.S. territories have • Involvement of school administrators and other enrolled as TN schools. All State child nutrition agen- school and community partners in building support cies actively participate in the program and are eligible for recommended healthful eating and physical to apply for competitive grants to fund TN activities. activity patterns (USDA/FNS, 2002a).

Schools formally enroll as “Team Nutrition Schools” Research Review of the and make a commitment to do the following: Team Nutrition Initiative • Support USDA’s Team Nutrition goal and values. When TN was launched in 1995, plans were included for a pilot project that would test the effectiveness of • Demonstrate a commitment to help students meet the program in influencing children’s food choices and the Dietary Guidelines for Americans. provide information on implementation issues that could be used to guide future policies and technical • Designate a Team Nutrition School Leader who will assistance. The evaluation of this pilot project is the establish a school team. only formal evaluation of TN conducted to date.

• Distribute Team Nutrition materials to teachers, stu- Team Nutrition Pilot dents, and parents. Implementation Project • Involve teachers, students, parents, foodservice per- Seven school districts were competitively selected to sonnel, and the community in interactive and enter- serve as pilot sites for the TN Pilot Implementation taining nutrition education activities. Project (USDA/FNS, 1998). Schools were selected based on a demonstrated capacity to meet the require- • Share successful strategies and programs with other ments inherent in both TN implementation and the schools (USDA/FNS, 2004). associated evaluation. Four of the seven districts were selected to participate in an indepth outcome evalua- However, FNS has no way of tracking the extent to tion; the other three districts participated in a limited which enrolled TN schools actually engage in the process study. above activities. The TN pilot was designed to test optimal implemen- TN was explicitly designed on a theoretical frame- tation of the initiative. Participating districts were work—social learning theory—that explains how peo- required to have designated TN coordinators. In addi- ple make behavior choices. As such, the focus of the tion, staff in each district received orientation, training, program is to promote behavior change, not simply to and materials that, in regular practice, a TN school impart information (with the hope that this might lead would not receive. Although these practices may have to behavior change). TN employs six “reinforcing given the pilot sites some advantages over a typical communication channels to reach children where they school, sites also dealt with many issues that put them live, learn, and play, as well as the adults who care for in a less favorable situation than schools might typical- them and can influence their behavior” (USDA/FNS, ly encounter. These issues included a number of con- 2002b). The communication channels are foodservice straints associated with the evaluation, such as con- initiatives, classroom activities, schoolwide events, densed implementation schedules and limited read-in

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 313 Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program time, recordkeeping and other burdens associated with Because some activities could be structured to meet data collection, and restrictions on use of local media more than one requirement, school districts were outlets to avoid contamination of comparison schools. required to conduct at least five different activities. In addition, pilot schools were not immune to routine challenges faced by all schools, including competing The TN pilot was implemented and evaluated in two for limited classroom time and resistance to change on phases: spring 1996 (Phase I) and fall 1996 (Phase II). the part of school foodservice staff. On balance, results The same classrooms participated in each phase. Each from the evaluation probably provide a fairly realistic phase included a pre-post design in both treatment and picture of what TN can accomplish. comparison schools. Followup data were collected the following school year for Phase I students (who were Research Design then in fifth grade) to assess long-term retention of any Participating school districts (communities) nominated favorable impacts. at least two matched pairs of elementary schools. The evaluation assessed the impact of TN in three key Schools were matched on size (total enrollment), pro- areas: skill-based nutrition knowledge, nutrition-related portion of students eligible for free and reduced-price motivation and attitude, and food consumption behav- meals, racial/ethnic composition, extent of existing iors. Survey items and observational measures were nutrition education activities, and characteristics of the chosen to assess changes associated with specific TN foodservice program. Across all four pilot sites, 12 messages to eat more fruits, vegetables, and grains; to pairs of schools were nominated. One of the schools in eat less fat; to eat a balanced diet; and to increase the each pair was randomly assigned to the TN group and variety of foods eaten. Knowledge, attitudes, and self- the other to a comparison group, resulting in 12 treat- reported behaviors were measured using self-adminis- ment schools and 12 comparison schools.165 The com- tered questionnaires. A total of 1,509 students in Phase parison schools agreed not to implement any TN I and 1,441 students in Phase II completed both pre- and activities and to delay any non-TN nutrition educa- post-tests (response rates ranged from 86 to 91 percent). tion activities. Data on students' food choices and consumption behav- Districts selected three elementary grades in which to iors were obtained through cafeteria observations car- implement grade-specific nutrition education curricula ried out by trained field staff. More than 3,000 meals (different versions for Pre-K and K, grades 1 and 2, were assessed at each measurement point, representing and grades 3 and 5) in the TN school. The impact response rates of 79-85 percent. The evaluation also evaluation focused solely on fourth graders because of included measure of food consumption behaviors that limited resources and the belief that children in this relied on student self-reports and parents’ perceptions. age range could reliably complete study surveys, Research Findings including food frequency items. A total of 144 fourth- grade classrooms participated in the evaluation. Because the research design included multiple data points for each subject, a repeated measures approach In addition to implementing the classroom curricula, was used in the analysis. The direction and amount of pilot districts agreed to train teachers and foodservice change was assessed for treatment and comparison personnel to implement menu changes that would pro- groups, and the net difference was attributed to the mote compliance with the DGAs, and to conduct a impact of TN. Regression analysis was employed, number of different core TN activities, including, at a using a mixed models approach to control for cluster- minimum: ing of the study sample. Data were generally aggregated across districts. Data from meal observations were • Two cafeteriawide events. analyzed within district and by phase, however, • Three parent-contact activities. because of lack of comparability in menus. • Two chef activities. • One districtwide community event. Results showed that TN had small, but consistently • One districtwide media event. positive and statistically significant, impacts on two of three measures of skill-based knowledge and on three different measures of nutrition-related attitudes and motivation. For skill-based knowledge, significant and 165Because some aspects of the TN intervention specifically involve the entire school or surrounding community, randomly assigning either class- positive impacts were noted for students’ ability to (1) rooms or students to treatment and control groups was not feasible. identify healthier choices and (2) apply knowledge of

314 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program the Food Guide Pyramid. Students’ ability to apply a were largely congruent and demonstrated that TN had “balanced diet” concept also increased, relative to pretest a small but positive and statistically significant impact scores, but differences were not statistically significant. on students’ self-reported eating behaviors (all three Nutrition attitude and motivation measures included a measures). However, none of these impacts persisted general attitude score as well as separate scores for over time. perceived consequences of increased consumption of fruits, vegetables, and grains, and a “cognitive rules” The discrete components model was not successful in scale, which asked students about their willingness to identifying the most influential component(s) of the make healthier food choices and their understanding program because of change in or omission of various about what that required. The relative size of the impacts program components over the course of the demon- was small (generally an increase of less than one correct stration. The level of exposure model reinforced answer). This did not seem to be attributable to a ceiling results of previous research, indicating that the impact effect. The authors suggest that the impacts reflect the of TN varied depending on the number of channels to short implementation period used for the evaluation which a student was exposed. and speculate that greater effects could be achieved with a more protracted period of intervention. Overview of the Followup data showed that significant TN effects were Nutrition Education maintained over time, although the size of the impact and Training Program decreased for three of the five measures that were sig- The beginnings of the NET program can be traced to nificant during Phase I. Estimated impacts were equiv- the 1969 White House Conference on Food, Nutrition, alent or greater at followup, compared with Phase I, and Health (Maretzki, 1979). The White House confer- only for the general attitudes measure and for per- ence emphasized the importance of good nutrition dur- ceived consequences of increased consumption of ing childhood and the need for children, parents, and fruits, vegetables, and grains. school foodservice personnel to understand the rela- Effects on observed food selection and consumption tionship between good nutrition and health. The con- behaviors in the cafeteria were modest. The only signifi- ference stimulated interest in school-based nutrition cant effects that were noted consistently in all districts education and the possibility of collaboration between were a slight increase in the amount of grain foods eaten school foodservice staff and educational staff. and a small increase in the diversity of foods eaten The NET program was established in 1977, 8 years (the number of different food groups included and after the White House Conference, under P.L. 95 166, total number of items). Changes in the selection and the National School Lunch Act and Child Nutrition consumption of fruits, vegetables, and low-fat milk Amendments. NET was envisioned as a means of were in the expected direction, overall, but were not using the school meals programs and school cafeterias statistically significant or consistent across districts. as learning laboratories for helping children develop a Analysis of three different measures of self-reported better understanding of the principles of healthy eat- eating behaviors showed that TN had small but statisti- ing. The intent of the program was to teach children cally significant effects on students’ self-reported the value of a nutritionally sound diet, develop nutri- behaviors. The specific behaviors examined were use tion education curricula and materials, and train teach- of low-fat foods, consumption of fruits and vegetables, ers and school foodservice personnel (Maretzki, 1979). and dietary variety (the number of food groups includ- Major goals of the program, as stated in the enabling ed in meals and snacks eaten the previous day). Three legislation, include the following: different multivariate models were used to assess TN’s impact on self-reported eating behaviors: a uniform • The instruction of students, preschool through grade treatment model (treatment defined as a binary vari- 12, in the nutritional value of foods and the relation- able), a discrete component model (treats TN treatment ships between food and health. as several discrete components and estimates effects for each), and a level of exposure model (treats TN • The training of school foodservice personnel in treatment as a continuous variable, ranging from zero nutrition, foodservice management, and the use of to six, based on the number of channels to which the the school cafeteria as an environment for learning student was exposed). Results for all three models about food and nutrition.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 315 Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program

• The in-service education of teachers and other through FY 1990 and was accompanied by a decrease in school staff in nutrition education and in the use of the number of students, educators, and school foodservice the cafeteria as a learning laboratory. personnel served by the program (Kalina et al., 1989).

• The identification, development, and dissemination In 1989, growing public concern over children’s nutri- of nutrition education resources and curricula. tional well-being and specific concerns about the nutritional quality of school meals contributed to a resur- The program is administered at the State level. FNS gence of interest in NET, particularly as a means for awards NET grants to States, and States appoint a providing training for foodservice personnel (Nelsen, State Coordinator to administer the funds. The 1992). P.L. 101-147 (November 10, 1989) reauthorized Coordinator must assess the State’s nutrition education NET for 5 years (FY 1990-94). During this interval, and training needs, develop a State plan for meeting NET authorization levels grew from $10 million to those needs, and implement the program according to $25 million. In 1994, P.L. 103-448 authorized the NET the plan. NET resources may be used to develop cur- program permanently, with annual funding of $10 mil- ricula and materials, implement nutrition education lion for FY 1996 and each year thereafter. Funding programs for children, and conduct in-service training was increased to $7.5 million in FY 1991 and then to for foodservice and classroom personnel. $10 million in FY 1992. Annual funding continued to be approximately $10 million through FY 1996. States have considerable autonomy in allocating NET program funds. States have used their NET funds in In August 1996, P.L. 104-193 changed NET funding vastly different ways that reflect not only the results of from mandatory to discretionary for FY 1999-2002. their needs assessments, but also the status of their However, since 1996, NET has received funding only school-based nutrition education, training, and resources once, in FY 1998 ($3.75 million).166 The curtailment at the time NET began. Some States have spent signif- of NET funding that began in FY 1997 coincided with icant resources in curriculum development, while others the beginning of TN, which has been funded at about adopted or adapted existing materials and focused on $10 million annually since FY 1996. dissemination. Still others encouraged local school districts to develop projects that suited their own needs. Research Review of the States are required to submit annual reports that pro- Nutrition Education and vide information on program dissemination, including Training Program the number of individuals who participated in NET program activities and the number of NET publica- The NET program was developed at a time when most tions that were distributed. States must also describe nutrition education programs were based, expressly or their key accomplishments and outcomes. Reported implicitly, on the KABINS model: the assumption that outcomes are descriptive in nature, such as the number an increase in Knowledge will affect Attitudes, which and type of workshops that were held, rather than in turn will affect Behavior and ultimately Nutritional measures of program impacts. Status (Contento et al., 1992). Today, nutrition educators realize that promoting behavior change, particular- Program Funding ly among children, is a more complicated process. These understandings have contributed to the theory- NET has had a roller-coaster funding history. Funding based underpinnings of the TN Initiative, which makes for FY 1978 and FY 1979 was authorized at $0.50 per ample use of social learning theory and social market- child enrolled in schools and institutions participating ing to more directly target behavior change. in the NSLP—roughly $26.2 million per year. A minimum level of $75,000 was established for individual Given the underlying assumptions of the NET program State grants. model, it is not surprising that most studies of NET focused exclusively on impact on nutrition knowledge. This initial level of funding was not maintained for long Research has shown that change in knowledge is easi- (USDA/FNS, 2002c). By FY 1981, only 3 years after the ly achieved, even by short-term programs (Contento et program started, funding had been reduced to $15 million, a 42-percent decrease. In FY 1982, funding was further decreased to $5 million, only about 19 percent of 166In FY 1997, NET operated with $3.75 million that was reprogrammed initial funding. This level of funding was maintained from TN funds.

316 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program al., 1992). In the context of this review, research that superior gains in nutrition knowledge. In addition, assessed knowledge gain or change in attitude, without NET participants in grades 1 through 3 were also some assessment of eating behavior, was considered found to be more willing than nonparticipants to try insufficient. This research is well summarized else- new foods in the school cafeteria and more likely to where (Contento, 1992; Lytle, 1994). The following have made improvements in food preferences (based sections describe the limited available research on the on self-report). NET participants in grades 4 through 6 effects of NET interventions that measured impacts on were more willing than nonparticipants to try previ- eating behaviors. Impacts on nutrition knowledge ously rejected foods. Results showed no consistent and/or nutrition-related attitudes assessed in this effects on nutrition-related attitudes, self-reported eat- research are also described. ing behaviors, or plate waste.

National Nutrition Education and In Georgia, the State chose to follow a decentralized Training Program Evaluation model, in which schools were free to use different nutrition education curricula and materials. The evaluation The only national study of NET was completed during included 1,400 students in grades 1 through 8. Results the very early stages of the program, between 1979 and showed that NET had strong positive effects on nutri- 1980 (St. Pierre and Rezmovic, 1982). At that point, it tion knowledge but limited effects on attitudes and self- was plausible to expect program impacts in only a few reported eating behaviors (St. Pierre and Glotzer, 1981). States that had been able to begin implementation almost immediately after funds became available. Moreover, The authors appropriately point out several factors that because of the diversity of States’ goals, only State- limit the generalizability of their results. A major limi- specific impact evaluations were deemed appropriate. tation is the quasi-experimental design, with the incumbent potential of nonequivalent NET and non- Consequently, impact assessment in the National NET groups (on factors other than pretest measures). Nutrition Education and Training Program Evaluation Another major limitation is the duration and content of focused on program activities and outcomes in two the intervention. The authors questioned whether it States: Georgia and Nebraska (St. Pierre and Rezmovic, was appropriate to expect changes in behavior from an 1982). (The study also collected descriptive data at the intervention that was limited in time and essentially national level, including an analysis of State plans.) focused on knowledge dissemination (St. Pierre, The programs in Georgia and Nebraska were firmly 1982). Other factors that complicate interpretation of established and were widely respected by NET staff at the study’s findings are the exposure of some non- the regional and national levels. NET students to other (non-NET) nutrition education The evaluation of the Nebraska NET program focused activities during the treatment period and the signifi- on assessing how well the program was implemented, cant nutrition education that had been conducted in as well as its impact on children’s nutrition-related many classrooms before the NET intervention. knowledge, attitudes, preferences, and eating habits. Other Studies of NET’s Impact Nebraska offered a statewide curriculum that was experience-oriented in the primary grades, but knowl- The literature search identified three small, local stud- edge-oriented in grades 4 through 6. Twenty schools ies that examined the impact of NET interventions on were selected from 98 volunteers and were randomly children’s nutrition-related knowledge, attitudes, assigned to treatment and control groups. A pre- and and/or eating behaviors.168 One of the earliest studies post-test design was used, with the pretest conducted examined the NET program in Tennessee. Tennessee’s immediately after the 10-week treatment concluded.167 first NET Plan included a detailed evaluation plan No followup measures were collected. (Banta and Cunningham, 1982). Assessment instruments were developed for students, parents, teachers, At pretest, treatment and control groups were equivalent school administrators, and foodservice personnel. All on all outcome measures (St. Pierre et al., 1981). At instruments included self-reported measures of nutri- post-test, NET participants in all grades had statistically tion behaviors. A plate waste study was devised to measure student-level changes in food consumption. 167Nebraska’s Experience Nutrition curriculum had 11 segments, designed to be used sequentially for grades K through 6 with an expected cumulative impact on children’s behavior. However, the evaluation was only able to test 168The literature search revealed several published studies that evaluated for an immediate effect of the particular segment used in each classroom (St. the impact of NET on the knowledge and attitudes of teachers and/or foodser- Pierre et al., 1981). vice personnel. These results are summarized elsewhere (Olson, 1994).

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 317 Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program

Baseline data were collected in 36 elementary likely than their non-NET counterparts to report an schools—2 treatment schools and 2 control schools in improvement in the quality of foods chosen for snacks each of 9 so-called development districts. No details eaten away from home. Effects on reported quality of are available in the published literature on the school at-home snacks were inconsistent. No effects were selection process or on the characteristics or compara- detected for any of the other parent-reported measures bility of treatment and control groups (Banta and or for the teacher-reported measures. Cunningham, 1982; Banta et al., 1984). Shannon and Chen (1988) conducted a 3-year study of The first post-test was conducted after less than 1 year the NET program in Pennsylvania. The authors assessed of NET interventions. At this point, fourth and sixth the knowledge, attitudes, and self reported behaviors graders in treatment schools were more likely than those of children as they progressed through grades 3, 4, and in control schools to report positive eating behaviors. In 5. Districts that responded to an invitation to join the addition, first graders in treatment schools were more study were grouped by geographic region and ranked likely than those in control schools to report having according to nutritional need and community socioeco- eaten the school lunch. However, plate waste studies nomic status. The 12 neediest districts were offered found no significant differences in food consumption participation in the study, but two declined. treatment and control students at any grade level. At the last followup (year 3 of the study), significant differ- Schools in the remaining 10 districts were then randomly ences were noted between NET and non-NET students assigned to treatment and control groups. Initial assign- for knowledge gain (at all grade levels) and for atti- ments were adjusted because of administrative contin- tude scores (at four grade levels). Again, however, no gencies. For example, one principal supervised three of significant differences were detected in self-reported the small schools and wanted all of them to be in the eating behaviors or in plate waste (Banta et al., 1984). same group. The resulting baseline sample included 17 treatment schools with 879 students in 39 third-grade Also in the early 1980s, Gillespie (1984) studied three classrooms and 18 control schools with 828 students in NET interventions in New York State. Three schools that 36 classrooms. Students in the treatment group received had received mini-grants from the New York State NET 9-12 weeks of nutrition education each year based on program (and used them for very different activities) the Nutrition in a Changing World curriculum. were matched with control schools based on size (total enrollment), community socioeconomic indicators, staff In the end, the treatment group had significantly greater interest in nutrition education, and type of food service. knowledge gains than the control group, as well as significantly greater improvements in some attitude meas- The study included pre- and post-test assessments of ures. However, no significant impacts were detected for students' nutrition knowledge and attitudes. In addi- eating behaviors. Eating behaviors for both treatment tion, after the intervention was over, parents were and control groups significantly improved over time. asked whether they observed changes in the foods The authors concluded that “it is difficult to demon- their children ate, their children’s interest in eating strate that increased nutrition knowledge dramatically nutritious foods, or their children’s understanding of affects nutrition attitudes and eating behaviors.” nutrition. Similarly, teachers were asked whether they noticed changes in children’s food choices or their attitudes toward nutrition. Summary Since it was established in 1977, the NET program has After controlling for differences in baseline scores, provided fluctuating support for nutrition education in Gillespie found no significant improvement in nutri- school classrooms and cafeterias. Programs have been tion knowledge or attitudes among NET students. Both State-defined and have varied considerably across groups of students showed significant gains in knowl- States. Most programs have aimed at improving chil- edge and attitude measures at post-test, and the differ- dren’s knowledge and attitudes as a means of ultimate- ence in relative size of the gains made by each group ly influencing their behavior because of assumptions was not significantly different. that improved knowledge and/or attitudes will lead to behavior change, because of resource constraints, or With regard to parental reports of children’s eating because of questions about whether behavioral out- behaviors, Gillespie found that parents in NET schools comes constitute an appropriate goal for school-based that had the most intensive intervention were more nutrition education.

318 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program

Several studies provide compelling evidence that NET in knowledge, attitudes, and behavior across a variety nutrition education activities can improve, at least for of health areas. the short-term, children’s nutrition knowledge and attitudes, but there is limited evidence that NET programs In addition, research has shown that teacher training affect children’s eating habits. increases teachers’ interest in teaching nutrition, as well as the time they devote to it. Adding parent participation This finding holds true for most school-based nutrition to classroom instruction was found to enhance program education programs, including programs not sponsored impact, particularly in the earlier grades, and particu- by NET and not based on the KABINS model (Contento larly if parents and children worked together. Contento et al., 1992). In a comprehensive review of research on and her colleagues (1992) conclude that, in most eval- school-based nutrition education implemented in the uations of nutrition education programs, “the effective- 1980s and early 1990s (most of which was not specifi- ness of nutrition education was not given a fair test.” cally sponsored by NET), Contento and her colleagues found that, on average, these interventions provided The TN Initiative is well-conceived in building on the 10-15 hours of instruction over a period of 3-15 weeks. NET experience and in incorporating a multi-pronged, The programs that were most successful, however, theory-driven focus of behavioral change. Results of tended to include longer (more intensive) interventions. the pilot implementation project, though preliminary The year-long Know Your Body curriculum, for exam- and certainly not generalizable, are promising. ple, has been found to produce not only behavioral changes, but also measurable physiological improve- Future research should examine the impact of the TN ments. The Food ...Your Choice curriculum, which Initiative in a larger number of schools where the pro- includes activities for all grades that can be included in gram is firmly established. Examination of program subjects already being taught, has induced elementary impacts on nutrition-related behaviors should move students to eat significantly more fruits, vegetables, beyond the self-administered questionnaires and cafete- protein foods, and vitamin-A-containing foods. ria observations employed in the evaluation of the TN pilot project to include more sophisticated dietary assess- The relationship between extended intervention periment techniques that will provide information on food ods and behavioral change is supported by the results and nutrient intake both in and out of school. Given the of the School Health Education Evaluation (Connell et multi-modal nature of the TN Initiative and the likeli- al., 1985). This nationally recognized study found that hood that students will receive varying “doses” of the 30 hours of classroom instruction were required to program’s intervention components, a process study achieve “medium” effects for general health practices, that clearly documents how the program is implement- 40 hours were required for changes in attitudes, and 50 ed and, to the extent feasible, the amount of exposure classroom hours were required to achieve stable levels to the program for each child, is also very important.

Economic Research Service/USDA Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 319 Chapter 16: Team Nutrition Initiative and Nutrition Education and Training Program

References Olson, C.M. 1994. A Review of the Research on the Effects of Training in Nutrition Education on Banta, T., and J.L. Cunningham. 1982. A Chance to Do Intermediaries, Paraprofessionals, and Professionals. It Right: Assessing the Impact on Participants of a State- USDA, Food and Nutrition Service. Wide Nutrition Education Program. Paper presented at the Annual Meeting of the American Educational Shannon, B., and A.N. Chen. 1988. “A Three-year Research Association, New York, NY, March 19-23. School-based Nutrition Education Study,” Journal of Nutritional Education 20(3):114-24. Banta, T.W., J.L. Cunningham, W.W. Jozwiak, et al. 1984. “Comprehensive Evaluation of a State Nutrition St. Pierre, R.G. 1982. “Specifying Outcomes in Education and Training Program,” Education Nutrition Education Evaluation.” Journal of Nutrition 105(4):376-81. Education 14:49-51.

Burghardt, J., A. Gordon, N. Chapman, et al. 1993. St. Pierre, R.G., T.D. Cook, and R.B. Straw. 1981. “An The School Nutrition Dietary Assessment Study: Evaluation of the Nutrition Education and Training School Food Service, Meals Offered, and Dietary Program: Findings from Nebraska,” Evaluation and Intakes. USDA, Food and Nutrition Service. Program Planning 4:335-44.

Connell, D.B., R.R. Turner, and E.F. Mason. 1985. St. Pierre, R.G., and J.A. Glotzer. 1981. An Evaluation “Summary of findings of the School Health Education of the Georgia Nutrition Education and Training Evaluation: health promotion effectiveness, implementa- Program. Cambridge, MA: Abt Associates Inc. tion, and costs,” Journal of School Health 55(8):316-21. St. Pierre, R.G., and V. Rezmovic. 1982. “An Contento, I.R., A.D. Manning, and B. Shannon. 1992. Overview of the National Nutrition Education and “Research Perspective on School based Nutrition Training Program Evaluation,” Journal of Nutrition Education.” Journal of Nutritional Education Education 14(2):61-66. 24(5):247-60. U.S. Department of Agriculture, Food and Nutrition French, L. 2002. Personal communication. Service. 2002a. “Team Nutrition School Enrollment Form.” Available: http://www.fns.usda.gov/tn/Join/ Gillespie, A.H. 1984. “Evaluation of Nutrition Education enrollmntform.pdf. Accessed May 2002. and Training Mini-grant Programs,” Journal of Nutritional Education 16(1):8-12. U.S. Department of Agriculture, Food and Nutrition Service. 2002a. “Team Nutrition Policy Statement.” Kalina, B.B., C.A. Phillips, and H.V. Minns. 1989. Available: http://www.fns.usda.gov/tn/grants/ “The NET Program: a 10-Year Perspective,” Journal TN_PolicyStatement.pdf). Accessed April 2002. of Nutritional Education 21(1):38-42. U.S. Department of Agriculture, Food and Nutrition Lytle, L.A. 1994. Nutrition Education for School-Aged Service. 2002b. “Join the Team: Communication Children: A Review of Research. USDA, Food and Channels.” Available: http://www.fns.usda.gov/tn/Join/ Nutrition Service. Ontheteam/communication.html. Accessed April 2002.

Maretzki, A.N. 1979. “A Perspective on Nutrition U.S. Department of Agriculture, Food and Nutrition Education and Training,” Journal of Nutritional Service. 2002c. NET History. Available: http://www. Education 11(4):176-80. fns.usda.gov/cnd/NET.NETproghist.htm. Accessed April 2002. National Research Council. 1989. Diet and Health: Implications for Reducing Chronic Disease Risk. U.S. Department of Agriculture, Food and Nutrition Washington, DC: National Academy Press. Service. 1998. The Story of Team Nutrition: Pilot Study Outcome Report. Nelsen, B.J. 1992. “The Role of the Federal Government in Promoting Health Through the U.S. Department of Agriculture and U.S. Department Schools: Report from the Department of Agriculture,” of Health and Human Services. 1990. Dietary Journal of School Health 62(4):138-40. Guidelines for Americans, 3rd edition. Home and Garden Bulletin No. 232.

320 Effects of Food Assistance and Nutrition Programs on Nutrition and Health / FANRR-19-3 Economic Research Service/USDA