Simple Suppers: Findings from a Family Meals Childhood Obesity Prevention Intervention

Home , Meal preparation, Reduction (cooking)

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Catherine Ann Rogers, MS, RDN, LD

Ohio State University Nutrition Graduate Program

The Ohio State University

2017

Dissertation Committee:

Carolyn W. Gunther, PhD (Advisor)

Sarah Anderson, PhD

Carla Miller, PhD, RD

Keeley Pratt, PhD, IMFT

Catherine Ann Rogers

2017

Abstract

Background: Given the ongoing childhood obesity public health crisis and potential protective effect of family meals, there is need for additional family meals research specifically experimental studies with expanded health outcomes that focus on the at-risk populations in highest need of intervention. Future research, specifically intervention work, would also benefit from an expansion of the target age rage to include younger children who are laying the foundation of their eating patterns and are capable of participating in family meal preparations. The purpose of this dissertation research was to address this research gap by developing and assessing the effectiveness of a 10-week multi-component family meals intervention targeting underserved families with children

4-10 years old, aimed at eliciting positive changes in child diet and weight status.

Methods: A 10-week family meals program (Simple Suppers) designed for underserved families with 4-10 year old children from racially diverse backgrounds was implemented as a pre-test-post-test, multi-cohort, quasi-experimental trial with waitlist control. The

10, 90-minute program lessons were delivered weekly over the dinner hour at a faith- based community center. Session components included: a) interactive group discussion of strategies to overcome family meal barriers, plus weekly goal setting for caregivers; b) engagement in age- appropriate food preparation activities for children; and c) group

ii family meal for caregivers and children. Main outcome measures were change in: child diet quality and child standardized body mass index (BMI). Caregiver diet and BMI, along with frequency of family meals, were also assessed.

Results: At baseline, 95 families (children: n=126; caregivers: n=95) enrolled in the study. Among child participants, approximately 62% were female, 60% identified as

Black, mean age was 6.9 years old, and mean BMI z-score was 0.69. Among caregiver participants, 98% were female, 62% identified as Black, 50% were between 31-40 years old, and mean BMI was 33.0. Approximately 40% of participating families were reliant on the Special Supplemental Nutrition Assistance Program for Women, Infants, and children (WIC), the Supplemental Nutrition Assistance Program (SNAP), and/or the

National School Lunch Program (NSLP). Generalized linear models, adjusted baseline values and demographics, demonstrated a program attendance intervention effect on child

BMI z-score, with children attending >7 classes having a significant decrease in BMI z- score post-test, which was maintained during the follow-up period. Similarly, a significant decrease in caregiver BMI was observed among intervention caregivers at post-test, regardless of level of attendance, and maintained during the follow-up period.

Conclusions: Participation in Simple Suppers at a 70% level led to a decreased child BMI z-score, which was maintained at the 10-week follow-up. In addition, participation in the

Simple Suppers program led to a decreased BMI among caregivers, regardless of

iii level of attendance, which was maintained at the 10-week follow-up. Results from this study demonstrate the potential for engagement in an evidence-based family meals program to positively impact child weight status among a racially diverse sample of school-aged children, of significance given the ongoing childhood obesity epidemic.

iv Dedication

May the road rise to meet you,

may the wind be ever at your back.

May the sun shine warm upon your face,

and the rains fall soft upon your fields.

And until we meet again,

may God hold you in the palm of his hand.

This dissertation is dedicated to a man who has taught me the strength of positivity & has shown me that in every situation we face, we are presented with an opportunity to learn.

I love you from the bottom of my heart, Grampie.

v Acknowledgements

A sincere thank you to my adviser, Carolyn Gunther, and dissertation committee members, Sarah Anderson, Carla Miller, and Keeley Pratt, for their continuous guidance, support, and encouragement. I also thank our community collaborator Vineyard

Community Center (VCC) and VCC staff, Daniel Nathan and Maria Broeckel for their support of and dedication to the Simple Suppers program. I would also like to extend unending gratitude to the dozens of undergraduate and graduate students and dietetic interns who invested endless hours to help make this study a success. Lastly, thank you to the Cardinal Health Foundation for making this study possible through their provision of funding.

vi Vita

Education

2009...... M.S. Human Nutrition, Case Western Reserve University

2005...... B.S. Dietetics, The Ohio State University

Research Experience

Aug, 2013 to Present…...... Graduate Research Assistant, The Ohio State University

Apr, 2014-July, 2014……... Research Dietitian, The Ohio State University

Jan, 2012-Mar, 2013…….... Research Associate, Case Western Reserve University

Aug, 2009-Dec, 2010…...... Graduate Associate, Case Western Reserve University

Professional Experience

Aug, 2016-Dec, 2016…..... Nutrition Intern, Nestlé Research & Development Center

Aug, 2009-Aug, 2010…...... Dietetic Intern, University Hospitals of Cleveland

Higher Education Teaching Experience

Fall, 2013-Present……...... Undergraduate Student Mentor, The Ohio State University

Fall, 2013-2016…...... Guest Lecturer (Medical Nutrition Therapy, Type I Diabetes

& Gestational Diabetes), The Ohio State University

Fall, 2013…………………….... Guest Lecturer (Public Health Nutrition, Maternal & Child

Nutrition), The Ohio State University

vii Jan, 2011-Dec, 2011……..... Adjunct Professor, The Art Institute of Pittsburgh

Certifications

Sept, 2010-Present……...... Registered Dietitian (accreditation: Academy of Nutrition &

Dietetics

Publications

Rogers C, Anderson S, Dollahite J, Hill T, Holloman C, Miller C, Pratt K, Gunther C. Methods and Design of a 10-week multi-component family meals intervention: a two group quasi- experimental effectiveness trial. BMC Public Health, 2017. 17:50.

Rose AM, Wagner AK, Kennel JA, Pennywitt J, Battista-Hesse M, Miller CK, Murray RD,

Rogers C, Gunther CW. Determining the feasibility and acceptability of a nutrition education and cooking program for preschoolers and their families delivered over the dinner hour in a low-income day care setting. Infant, Child, and Adolescent Nutrition,

2014. 6;3:144-151.

Major Field of Study

Ohio State University Nutrition Graduate Program

viii Table of Contents

Abstract…………………………………………………………………… ii Dedication………………………………………………………………… v

Acknowledgements………………………………………..……………… vi

Vita………………………………………………………..……………… vii

List of Tables……………………………………………..………………. x

List of Figures………………………………………………...... xii

Chapter 1: Introduction ………………………………………………….. 1

Chapter 2: Literature Review…………………………………………….. 15

Chapter 3: Theoretical Framework…………………………...... 59

Chapter 4: Methods and Design of a 10-Week Multi-Component Family Meals Intervention: a Two Group Quasi- Experimental Effectiveness Trial…………………………………………… 67 Chapter 5: Child Findings from a 10-Week Multi-Component Family Meals Intervention for Underserved Families with Children 4-10 Years Old………………………………………………. 109 Chapter 6: Caregiver and Family Findings from a 10-Week Multi- Component Family Meals Intervention for Underserved Families with Children 4-10 Years Old…………………….... 169 Chapter 7: Discussion…………………………………………………..... 245

Bibliography……………………………………………………………… 261

Appendix A: Simple Suppers Intervention Participant Recruitment Recruitment Materials……………………..………………. 283 Appendix B: Simple Suppers Intervention Informed Consent and Assent Forms………………………..…...... 286 Appendix C: Simple Suppers Intervention: Data Collection Materials…………………………………………………… 294 ix List of Tables

Table 1. Family Meals and Child Diet Quality………………………….. 18

Table 2. Child Involvement in Food Preparation and Child Diet Quality.. 28

Table 3. Home Environment and Child Diet Quality…………………….. 38

Table 4. Table 4. Parent Self-Efficacy for Healthy Dietary Practices……. 52

Table 5. Simple Suppers Intervention Study Design: Two- Group, Staggered Cohort, Quasi-Experimental Design…………………. 74 Table 6. Overview of Formulated Program Objectives at Each Level of Intervention……………………………………………………… 79 Table 7. Matrix of Change Objectives by Level of Intervention for Program Objective 1 of the Simple Suppers Intervention………. 80 Table 8. Matrix of Change Objectives by Level of Intervention for Program Objective 2 of the Simple Suppers Intervention………. 82 Table 9. Theory-Based Methods and Practical Strategies to Achieve the Change Objectives for Selected Program Objectives of the Simple Suppers Intervention…………………………………….. 87 Table 10. Simple Suppers Topics and Goals by Weekly Lesson…………. 101

Table 11. Baseline Child Participant Characteristics……………………… 122

Table 12. Intervention Period: By-Group Child Outcomes at Baseline T0 and Post-Test, T1……..………………………………………… 127 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1…………….………………………………….... 129 Table 14. Intervention Period: Within-Group Child Outcomes at Baseline T0 and Post-Test T1…………………………………... 137

x List of Tables (cont.)

Table 15. Intervention Period: By-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1During the Follow- 139

Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test T1………………… 141

Table 17. Intervention Period: Within-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1…….………………. 150 Table 18. Follow-Up Period: By-Group (Low vs. High Attenders) Child Outcomes at Post-Test T1 and Follow-Up T2………………….. 152 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes…………………………………. 154 Table 20. Follow-Up Period: Within-Group Differences in Child Outcomes………………………………………………………... 162

Table 21. Baseline Caregiver Characteristics……………………………... 182

Table 22. Intervention Period: By-Group Change in Caregiver and Family Outcomes from Baseline T0 to Post-Test T1…………………… 186

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test T1………………………………. 188 Table 24. Intervention Period: Within-Group Caregiver and Family Outcomes from Baseline T0 to Post-Test T1…………………… 202 Table 25. Intervention Period: By-Group (Level of Attendance) Caregiver And Family Outcomes Baseline T0 to Post-Test T1……………. 204 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test T1………………………………. 206 Table 27. Follow-Up Period: By-Group (Low vs. High Attenders) Caregiver and Family Outcomes………………………………... 221

Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes………………... 223

xi List of Figures

Figure 1. Social Cognitive Theory as the Theoretical Framework for the Simple Suppers Program………………………………………… 64

xii Chapter 1. Introduction

1 U.S. Childhood Obesity Statistics

The prevalence of childhood obesity has grown substantially over the past three decades worldwide [1]. In the U.S., the incidence of childhood obesity has more than doubled over the past three decades [2]. Despite a recent plateau in rates of childhood obesity in the U.S., approximately 17% (12.7 million) of America’s youth are obese.

While rates of adult obesity are higher than childhood obesity in the U.S., in relative terms, increases in overweight among children exceeds that of adults [3]. Since 1980, the prevalence of obesity among children 2-5 years old has more than doubled (5.0% to

12.4%), nearly tripled (6.5% to 17.0%) among children 6-11 years old, and more than tripled (5.0% to 17.6%) among adolescents 12-19 years old [4]. This rapid increase in obesity seen across the childhood and adolescent years has led childhood obesity to develop into an epidemic in the U.S., affecting America’s youth across all demographics

[5].

While all children living in the U.S. are at risk for obesity, racial/ethnic disparities place Black and Hispanic children at an increased risk for the disease [6]. Between 1986 and 1998, prevalence of childhood obesity increased 50% in non-Hispanic White children, while increasing more than 120% among Black and Hispanic children. In 2011-

2012, Hispanic children suffered the highest burden of obesity (22.4%), followed by non-

Hispanic Black children (20.2%) [2].

2 Along with minority children being disproportionately affected by the childhood obesity epidemic, children from low-income households, regardless of race/ethnicity, also face a greater disease-burden [5]. In 2012, more than 16% of the American population, and almost 20% of American children, lived in poverty [7]. Americans living in the most impoverished counties in the United States are at highest risk for obesity [8]. In counties where poverty rates are >35%, obesity rates are 145% higher than obesity rates in affluent counties. The increased risk of obesity in poverty-stricken neighborhoods is associated with poor diet quality and a sedentary lifestyle [9, 10]. The built environment in low-income neighborhoods may explain these negative trends in diet and physical activity, as they typically have poor access to fresh and nutritious foods [9], safety concerns, and limited access to recreational parks [10].

The proportion of households classified as moderately low-income has also been on the rise in the U.S. over the past five years [11]. Like children from low-income households, children from moderately low-income households are at an increased risk of obesity due to poor diet quality, a sedentary lifestyle, and the built environment.

However in addition to these factors, moderately low-income households are often ineligible for many resources (e.g., food vouchers, free/reduced childcare, free/reduced healthcare) that their low-income household counterparts have access to due to their income level. For this reason, moderately low-income households are often unable to meet their basic needs, which increases their risk for obesity and other morbidities.

3 Defining Childhood Obesity

Childhood obesity is a disease characterized by excess adiposity in youth 2-19 years of age [12]. Body mass index (BMI), defined as weight in kilograms divided by height in meters squared (kg/m2), is a measure of relative weight based on an individual’s mass and height. Body mass index is a surrogate measure of adiposity, attributing excess weight to fat mass. In adults, BMI is commonly used to assess adiposity, as it is strongly correlated with total adiposity in most adult populations. However, because BMI is a measure of weight and height, factors such as age, sex and race/ethnicity can alter the relationship between BMI and adiposity in adults.

In children, assessing adiposity is challenging. Throughout childhood, as rates of growth and development fluctuate, the relationship between weight, height, and adiposity differs significantly [12–16]. Along with age, sex, and race/ethnicity weakening the relationship of BMI with adiposity, in children, continuous changes in height and sexual maturation also abate this association [13, 16]. However, despite the many challenges of using BMI to assess child adiposity, the International Obesity Task Force has concluded that BMI percentile is still a reasonable measure to assess adiposity in children for clinical and research purposes [14].

For adults, BMI ranges have been established by the World Health Organization to classify adult weight status (Underweight: <18.5; Healthy weight: 18.5-24.9;

Overweight: 25-29.9; Obese: >30) [17]. These ranges aim to guide adults in establishing

4 and maintaining a weight status that promotes optimal health and well-being. For children, growth charts are used to establish BMI ranges for optimal health because they account for variables, such as age, sex, and vertical growth, that are known to alter the relationship between BMI and adiposity in children [12, 14, 15]. The Centers for Disease

Control and Prevention’s (CDC) age- and sex-specific growth charts include weight classification percentiles to classify child weight status [18]. Using the CDC’s BMI-for- age child growth charts, a child is defined as overweight with a BMI-for-age percentile greater than the 85th percentile and less than the 95th percentile and obese with a BMI-for- age percentile greater than or equal to the 95th percentile.

Etiology of Childhood Obesity

In order to address the problem of childhood obesity, it is critical to identify factors contributing to the epidemic. Childhood obesity is a complex disease, influenced by biological, intrapersonal, interpersonal, environmental and societal factors [19]. It is well established that obesity is a result of energy imbalance, in which energy intake exceeds energy expenditure [20]. However, the factors contributing to the recent escalation of energy imbalance among children remains unclear [13].

While numerous biological factors have consistently been associated with energy imbalance and obesity, such factors are likely not considered plausible explanations for the childhood obesity epidemic [10, 12]. For example, several endocrine and neurological syndromes, such as Praeder Willi and Mauriac, are associated with an

5 increased risk of obesity, however less than 5% of obesity cases in the U.S. are accounted for by these endogenous factors [10]. Genetics are another biological factor that have been associated with obesity, with some studies demonstrating 25% to 40% of BMI is heritable [12]. While genetic changes are likely occurring, changes in the human gene pool occur over extended periods of time and therefore cannot explain the rapid increase in childhood obesity that has occurred over the past thirty years [5, 12].

Most children are born with the innate ability to adjust energy intake to match energy expenditure, fostering healthy weight maintenance [12]. As children age, self- regulation of energy intake diminishes as food intake is influenced by both energy needs and external cues [20]. Eating behaviors that fuel development of obesity are established early in life and become difficult to change thereafter [21]. Parents play a major role in shaping food choices and eating behaviors of their children [22]. Parent role modeling and making certain foods available have been shown to directly impact a child’s dietary intake and weight status [22]. Poor parent diet quality, restrictive and permissive parental feeding behavior, pressure to eat nutrient-dense foods, and using food as a reward have been associated with unhealthy dietary habits in children [23].

Along with negative parent feeding behaviors, the obesogenic environment of the

U.S. also fosters adoption of poor dietary and physical activity behaviors [24].

Availability of fast food establishments and restaurants and frequency of consuming meals away-from-home have steadily increased over the past 50 years [25]. In the 1960s,

6 21% of a family’s food budget was spent on dining out [26]. In 2008, American families were spending 42% of their food budget eating away-from-home [27]. Between 1972 and 1997, the number of full-service restaurants increased by 35% and the per capita number of fast-food restaurants doubled [28]. Consuming foods away-from-home is consistently associated with poor diet quality and increased risk of obesity [29]. While there are many sources of away-from-home meals, such as carry-out establishments and restaurants, fast food establishments are the most common source of away-from-home meals [30]. Children may be more vulnerable to consume at fast food establishments, as the fast-food industry specifically targets children in their marketing of unhealthy products [25].

Similar to the increased availability and targeted marketing of the fast-food industry, the sugar-sweetened beverage (SSB) industry has made a great effort to increase

SSB availability and target marketing materials to children [25]. Sugar-sweetened beverages are drinks sweetened with sugar, high-fructose corn syrup, or other caloric sweeteners, such as soft drinks, flavored juice drinks, sports drinks, and sweetened tea.

Sugar-sweetened beverages are the leading source of empty calories in children’s diets in the U.S. [31, 32], with approximately 11% of calories consumed by youth coming from

SSBs [33]. In children, each 8-ounce SSB consumed per day increases the odds of becoming obese by 60% [34]. Increased SSB consumption among children in the U.S. is likely a result of youth-targeted marketing and increased SSB availability [32, 35, 36].

The SSB industry targets children considerably through advertisements and promotions

7 [35, 37]. Unfortunately, this marketing appears to be very effective, as SSB advertising exposure is associated with increased SSB consumption among children [37]. Increased availability and access to SSBs may also be a factor contributing to the rise in child SSB intake. The SSB industry has increased SSB availability and access for children by offering SSBs in vending machines and snack lines in schools [38]. In 2008-2009, SSBs could be purchased in vending machines in about 40% of elementary schools [36] and almost all middle and high schools in the U.S. [38]. Despite the 2010 Healthy Hunger-

Free Kids Act producing some improvements in limiting school SSB availability, as of

2012, only 44% of school districts in the U.S. had banned SSBs from school vending machines [39].

While a change in children’s dietary intake has contributed to the energy imbalance associated with childhood obesity, it is not the sole explanatory factor.

Changes in physical activity and daily energy expenditure over the past thirty years have also contributed to energy imbalance among U.S. children [40]. In schools, children have fewer opportunities for physical activity as outdoor recess and physical education requirements have steadily declined [41]. School transportation is now offered to the majority of students, with less than 15% of students in the U.S. walking or riding a bike to school [42]. Over the past thirty years, children’s leisure time activities have transformed from active play to more sedentary activities, such as television watching and computer games. Daily screen time has increased to more than 7 hours per day in youth, with distinct discrepancies existing between minority and majority youth [43]. In

8 comparison to non-Hispanic White children, whose daily screen time is about 6 hours,

Black and Hispanic children average greater than 9 hours of screen time per day. Cross- sectional data has consistently demonstrated a positive correlation between frequency of sedentary leisure activities and child BMI [24, 44, 45]. Interventions have more recently demonstrated a protective effect of physical activity on child BMI, and decreased sedentary activity time being associated with reduction in child BMI [44].

Improving obesity-related behaviors among America’s youth is challenging, as it requires behavioral, social, environmental, and societal changes [46]. However, despite the challenges, the recent plateau in childhood obesity prevalence is an indication that national, state, and local initiatives are having a positive impact and efforts must continue to decrease rates of childhood obesity and prevent the devastating consequences associated with the disease.

Consequences of Childhood Obesity

Obesity is currently the second leading cause of death in the U.S. and is likely to become the primary cause of death in the near future [47]. If obesity prevalence continues at its current rate, life expectancy will decline in the U.S. for the first time in its existence [48]. Childhood obesity, specifically, has detrimental effects, resulting in an array of adverse short- and long-term health, social, academic, and economic consequences.

9 Children suffering from obesity are at an increased risk for cardiovascular disease, due to an increased risk for hypertension and elevated serum cholesterol [49]. The risk of impaired glucose tolerance, insulin resistance, and type II diabetes is also significantly elevated in obese children [50]. Obese children can also experience physical discomfort, caused by joint and musculoskeletal complications, and compromised breathing, due to sleep apnea and asthma, which can discourage them from engaging in physical activity and foster a sedentary lifestyle [51, 52]. Along with physical complications, childhood obesity can invoke serious social, psychological, and academic problems, including discrimination, poor self-esteem, depression, and poor school performance [50, 53, 54].

Of concern, the detrimental physical, social, psychological and academic consequences associated with childhood obesity have potential to persist throughout a child’s entire life [53]. Obese children are more likely to be obese during adulthood [55–

57] and suffer from health risks associated with adult obesity, such as cardiovascular disease, type II diabetes, and some cancers [58]. Discrimination, poor self-esteem, and depression that originate during childhood as a result of childhood obesity can endure throughout adolescence and adulthood. The short- and long-term physical, social, and psychological consequences associated with childhood obesity can significantly compromise the health, well-being, and quality of life of individuals affected by this serious disease.

Along with health-related consequences of childhood obesity, the economic

10 burden of the disease is tremendous and continues to rise. The economic burden of obesity includes healthcare costs, health and emergency safety equipment costs, decreased worker productivity and absenteeism and higher workers’ compensation claims. In the U.S., obesity-related medical treatment costs between $147 and $210 billion annually, with the majority being used to treat obesity-related diseases [59].

Childhood obesity alone costs $14.1 billion annually [60]. For children treated under

Medicaid, the average annual total health cost for a child is $2,446. The annual cost for an obese child on Medicaid averages out at an annual cost of $6,730. The number of children hospitalized with a diagnosis of obesity has nearly doubled between 1999-2005, with total costs of children hospitalized with an obesity-related complication increasing from $125.9 million in 2001 to $237.6 million in 2005 [60].

Current Childhood Obesity Research

In order to slow the progression of the childhood obesity epidemic, interventions focusing on prevention are needed [61]. With the determinants of childhood obesity being numerous and multi-faceted, multicomponent (e.g. eating, physical activity, parenting), multi-setting (home, school community), and multi-level (e.g. individual, family, group) interventions integrating behavioral and environmental approaches have demonstrated to be most efficacious and are now the recommended approach for childhood obesity prevention interventions [40, 62–64].

As most obesity-related behaviors are established early in life [21], early

11 intervention and parent/family engagement are essential [62, 63]. The American

Academy of Pediatrics (AAP) Expert Committee on Childhood Obesity

Recommendations advocates for family-based approaches to childhood obesity prevention to develop and establish sustainable healthy home environments that support healthy dietary and physical activity behaviors [40, 65–67]. As established by the AAP, the specific core behaviors associated with a decreased risk for childhood obesity, include limiting consumption of SSBs, consuming recommended quantities of fruits and vegetables, and participating in family meals regularly [40].

Family mealtime has become a growing area of research for childhood obesity prevention [68–74]. While lasting only twenty minutes, family meals provide an opportunity for developing and fostering social and dietary habits that influence and shape the health of a child [72]. Cross-sectional research has demonstrated positive associations between family meals and child diet quality, demonstrated by increased fruit and vegetable intake,[75–79] decreased SSB intake [77, 78], and decreased frequency of skipping breakfast [75]. Family meals have also been inversely associated with child

BMI across various demographics [80–82].

The American Academy of Pediatrics recommends participation in family meals as a childhood obesity prevention strategy due to the literature demonstrating a protective effect of participation in healthy mealtime routines on child diet and weight [40].

However, the current evidence linking family meals with improved child dietary intake

12 (increased fruit and vegetable intake, decreased sugar-sweetened beverage (SSB) intake) and weight status (decreased body mass index (BMI; (weight (kg)/height (m)2)) z-score) has significant limitations. The majority of the family meals literature – specifically in the area of childhood obesity prevention – represents observational studies, demonstrating only an associative relationship of family meals with child diet and weight status [75, 77,

83, 84]. What’s more, racial and ethnic differences have been highly understudied; given that the segment of the United States (US) child population with high prevalence of obesity is racial and ethnic minorities [2], it has been suggested that this is an area in which additional research is needed. Similarly, the existing family meals intervention research (i.e., studies designed specifically to examine the cause and effect relationship between family meals and child diet and weight status), while strong with regard to study design, is limited and primarily targets non-Hispanic White children (8 to 12 years old), particularly from well-educated families [73, 74, 85].

In addition, the majority of the current research fails to examine the child health impact of family meals beyond BMI (e.g., central adiposity and blood pressure (BP)), with only a small number of studies including additional outcomes (e.g., disordered eating) [86–89]. Given the ongoing childhood obesity public health crisis [90] and the potential protective effect of family meals, there is need for additional family meals research, specifically experimental studies with expanded health outcomes that focus on the at-risk populations in highest need of intervention. Future research, specifically intervention work, would also benefit from an expansion of the target age range to include

13 younger children (4-7 year olds), who are laying the foundation of their eating patterns

[91], and are capable of participating in family meal preparations [92].

The purpose of this dissertation was to address this gap in the literature by assessing the effectiveness of a 10-week multi-component family meals intervention study, Simple Suppers, aimed at eliciting positive changes in child dietary intake and weight status. The Simple Suppers study was a two group quasi-experimental trial with staggered cohort design that targets underserved families with elementary school age children (4-10 years) and includes an examination of health outcomes beyond weight status.

14 Chapter 2. Literature Review

15 Literature Review Introduction

Cross-sectional and longitudinal data demonstrate a positive association between family meal frequency and child diet quality. This literature review will first discuss evidence supporting the positive association between family meals and child diet quality, and then will identify and discuss potential underlying mechanisms contributing to this association. In doing so, gaps in the family meals literature warranting further investigation will be identified and discussed.

16 Family Meals and Child Diet Quality

In 2007, the American Academy of Pediatrics Expert Committee established core behaviors associated with a decreased risk for childhood obesity. Among the many obesity-preventing behaviors identified (e.g., encouraging fruit and vegetable intake, limiting SSBs), eating meals as a family was emphasized [40]. Foods eaten outside the home are generally characterized as energy dense and nutrient poor [93], while meals prepared and eaten at home are typically of a much higher diet quality [94]. Thus, participation in family meals at home may provide a protective effect against inappropriate weight gain in children through reduced energy intake and improved diet quality [95]. Findings from cross-sectional, longitudinal and intervention research are summarized in Table 1. A brief review of the studies follows.

Table 1. Family Meals and Child Diet Quality Author Sample Sample Size Study Description Outcomes Findings Cross-Sectional Studies Woodruff SJ •6th, 7th & 8th •6th grade Family meal freq & diet quality assessed •Child diet quality High freq family dinners positively et al., 2010 graders (n=372) via the Food Behavior Questionnaire & a •Breakfast & lunch associated with child diet quality •Public •7th grade single 24-hr dietary recall. HEI-C was consumption (p<0.001) & negatively associated schools in (n=429) calculated for each participant •Family meal freq with skipping breakfast & lunch Ontario •8th grade (p<0.001) (n=487) Christian M •Students in •Students (n= Dietary intake & HFE assessed via child & •Child FV intake Child FV intake associated with et al., 2012 London 2383) parent completed CADET questionnaire •Family meal frequency family meal freq (p<.05), parent FV primary •Schools (n=52) sent home to participating families •Parent daily FV intake intake (P<.05) & home FV schools availability (p<.05) Utter J et al., •Students n=8,734 Secondary analysis (random sample of •Family meal (dinner) Family meals frequency >5 d/wk 18 2013 (13-17 yrs participants) of Youth ’07 nationally frequency (60%). Family meal frequency was old) representative survey. Child completed •F&V intake positively associated with F&V •Completed survey assessing family meal frequency •BMI intake (p<0.001) New Zealand and dietary intake. BMI assessed by Youth ’07 research team. survey •White (70%)

Fink S et al., •Parents of •Parents Data source: North Carolina Child Health •Child dietary intake Participating in >5 family meals/ wk 2014 children (n=1992) & Family Monitoring annual phone survey. (FV, SSBs) was associated with lower SSB birth-17 yrs Assessed child dietary intake, family meal •Family meal frequency intake in children (p<0.05), freq & HFE. increased vegetable intake in children & adolescents (p<0.05) & increased fruit intake in adolescents (p<0.05) Table 1. (continued)

18 Table 1. Family Meals and Child Diet Quality (continued) Trofolz AC •Minority n=120 families Data from the Family Meals LIVE! Study. •Home F&V availability •Home F&V availability & et al., 2016 failies with 6- Home food inventory, anthropometric •Home F&V accessibility were significantly 12 yr old measures, survey data collected from accessability associated with F&V at dinner •Parents of families via researcher. Foods served at •Weekly freq of F&V (p<0.05) children meals assessed via direct observation. served at dinner •Having fresh fruit (other than birth-17 yrs Families tape-recorded 8 family meals •Caregiver BMI canned or frozen) was most strongly •Black using an iPad. •Weekly caregiver food associated with serving F&V at (74%); White prep freq dinner (18%) •Weekly child food prep •Higher caregiver vegetable intake •Low income freq was associated with vegetables at (73%) dinner (p>0.05) •Caregiver meal planning

19 associated with fruits at dinner

(p<0.05) Longitudinal Studies Larson N et •11th, 12th •Students •Baseline: FFQ & HFE survey completed •Family meal frequency •Higher family meal freq in al., 2007 graders in (n=2052) during school day •Dietary intake (FV, adolescence predicted higher enrolled •Schools (n=31) •5 yr Follow-up: FFQ & HFE survey milk, whole grains, intakes of fruits/ vegetables & lower schools in completed online or by mailed surveys SSBs) intakes of SSBs during adulthood. MN Larson N et • 11th, 12th •Students •Baseline: FFQ & HFE survey completed •Family meal frequency More frequent family meals during al., 2012 graders in (n=2052) during school day •Dietary intake (FV, adolescence predicted higher freq of enrolled •Schools (n=31) •5 yr Follow-up: FFQ & HFE survey milk, whole grains, shared meals in adulthood. More schools in completed online or by mailed surveys SSBs) frequent shared meals in adulthood MN associated with better diet quality Table 1. (continued)

Table 1. Family Meals and Child Diet Quality (continued) Interventions Fulkerson J •Parent-child n=44 HOME intervention: 5 bi-weekly, 90- •Feasibility •High feasibility (95% parent very et al., 2010 dyads •Intervention minute sessions covering: interactive •Family dinner freq satisfied) •Child 8-10 n=22 •Control nutrition education, taste testing, cooking •Child diet quality •Child diet: improvements in FV years old n=22 skill building, parent discussion groups, •Parental SE intake (p=0.08) hands-on meal prep. Outcomes assessed •Child food prep skills •Child food prep skills increased baseline & post-intervention. •Child BMI z-score significantly (p<0.01) •Home food availability

Fulkerson J •Parent-child n= 160 dyads 10 monthly group session (nutrition •Fidelity •High fidelity (Avg attendance: et al., 2015 dyads •Intervention n= education; hands-on meal planning/prep; •Dose 68% lessons) •Child 8-12 81 goal setting phone calls) & 5 goal-setting •Child BMI z-score •Child BMI z-score: no significant years old •Control n= 79 phone calls. Outcomes assessed baseline, •Family dinner freq treatment group differences at post- •BMI >50% post-intervention (12-months) & follow-up test or follow-up. Prepubescent 20 •White (71%) (21 months post baseline) children in intervention group had •Economic significantly lower BMI z-scores assistance than controls (35%)

Robson SM •Caregiver/ch n=6 (caregiver- 10 wk cooking intervention on reducing •Caregiver & child •Proportion of dinners consumed by et al., 2016 ild (3-10 yrs child dyads) eating dinner away from home, energy dietary intake (energy; caregiver/child dyads away from old) dyads intake and improving diet quality. total & sat fat; sodium; home significantly decreased (56% •Dinner away Delivered at kitchen on university campus. fast food; restaurant) to 25%; p<0.05) from home First 6 lessons (parents only); last 4 lessons •Caregiver SE preparing •Dyad cholesterol at dinner >3x/wk (parents & child). Included hands on food healthy meals by visual significantly decreased (p<0.05) •White prep and behavior management education. analog scale (66%); Black Pretest-posttest design. •Caregiver BMI (33%) •Child BMI z-score HEI-C: Healthy Eating Index-Canada HFE: home food environment CADET: Child and Diet Evaluation Tool

20 Cross-Sectional Studies Cross-sectional studies examining the relationship between family meal frequency and child diet quality consistently demonstrate a positive association, with increased family meal frequency being associated with higher child diet quality [75–77]. The positive association between family meal frequency and child diet quality has been consistently demonstrated across various age groups, including preschoolers [77], grade- school children [75, 76], and adolescents [77].

The majority of cross-sectional studies investigating the association between family meals and child diet quality define family meals in terms of frequency per week

[75–77]. While numerous alternative family meal outcomes could be assessed, such as family members present at meals or location of meals, frequency per week is consistently used to define family meals in cross-sectional studies.

Great variability exists in the definition and assessment of child diet quality among cross-sectional studies investigating its relationship with family meal frequency

[75–77]. Daily fruit and vegetable intake [76, 77], frequency of breakfast consumption

[75], and daily SSB consumption [77] have all been used to define child diet quality in cross-sectional studies looking for associations between child diet and family meal frequency. Due to differences in child diet outcomes, different dietary assessment methods, such as 24-hour dietary recalls [75] and food frequency questionnaires [76, 77], have been used to assess child dietary intake. However despite variations in defining and

21 assessing child diet quality, cross-sectional studies investigating the relationship between child diet quality and family meal frequency have consistently demonstrated a positive association between family meal frequency and child diet quality [75–77].

Longitudinal Studies With cross-sectional studies demonstrating a consistent positive association between family meal frequency and child diet quality across all age groups, researchers began investigating this relationship longitudinally [79, 96]. Studies aiming to assess if family meal frequency during adolescence predicted dietary behaviors in young adults have demonstrated increased family meal frequency during adolescence to be predictive of higher diet quality and social eating behaviors during young adulthood.

Interventions With family meals research still in its infancy, results from interventions aiming to improve child diet quality through family meals are limited. The Healthy Home

Offerings via the Mealtime Environment (HOME) study was a community-based, randomized controlled, childhood obesity prevention intervention aiming to improve child diet through family meals [73]. The HOME intervention targeted 8 to 10 year old children and their parent(s) and included 5 bi-weekly, 90-minute lessons over the course of 10 weeks. Lesson content included: interactive nutrition education, taste testing, hands-on meal preparation skill building, and a parent education. Child-level

(anthropometric and dietary) and family-level (family meal frequency) data was collected at baseline and post-intervention in families receiving the HOME intervention

22 (intervention group) and families not receiving the intervention (control group).

Compared to children in the control group (n=22), children participating in the HOME intervention (n=22) reported greater food preparation skill development and trends suggesting higher fruit and vegetable intake (p=0.08) at the conclusion of the intervention. Improvements in family dinner frequency (p=0.87), parental self-efficacy

(SE) for making healthful changes (p=0.55) and home fruit/vegetable availability

(p=0.12) did not significantly improve among intervention families. As the first intervention aiming to improve child diet quality through family meals, the HOME study demonstrated high feasibility of a family meals nutrition education and cooking intervention. Furthermore, the trends towards positive family mealtime behaviors observed during the HOME intervention provides potential efficacy for improving child diet quality through family meal interventions.

The Fulkerson J et al. research team that led the HOME study followed up that study with the HOME Plus research intervention [73, 74, 85]. The HOME study served as the pilot study for the HOME Plus study, as findings from the HOME study were used to modify the original intervention and develop the HOME Plus study. The HOME Plus study targeted families with one child 8-12 years old with a BMI >50 percentile [74].

The intervention included 10 monthly lessons that included: nutrition education and hands-on food preparation. Participating families also received 5 motivational goal- setting phone calls throughout the intervention.

23 Data was assessed at baseline, post-intervention and follow-up (9-months post- intervention). Program fidelity was measured by assessing average lesson attendance.

Participant measures included: child BMI z-score, home food availability, nutrition quality of foods served at family meals, child dietary intake (assessed via 3, 24-hour dietary recalls); home food environment (including family dinner frequency); and child cooking skills.

One hundred sixty families were recruited for the Home Plus study.

Approximately 70% of the participants identified themselves as being White and 35% reported being on economic assistance. The Home Plus study demonstrated high fidelity, with participants attending on average 68% of the offered lessons. No significant treatment group differences were observed at post-test or follow-up. However prepubescent children in the intervention group did have significantly lower BMI z- scores than controls. Based on these results, the Home Plus intervention demonstrated potential efficacy for family meals interventions significantly improving child BMI z- scores. Furthermore, this intervention demonstrated that family meals interventions may be most effective with younger (<10 years old) children.

Similar to the Fulkerson et al. team, the Robson SM et al. team led an intervention that aimed to improve child diet quality through family meals [97]. Caregiver-child dyads who consumed dinner away from home >3 times per week were recruited for this

24 study. In addition, to participate in this study, caregivers had to be overweight (BMI >25 kg/m2) and the participating child had to be 3-10 years old.

The intervention was a 10-week cooking program that aimed to reduce eating dinner away from home and improve caregiver and child diet quality. The intervention was delivered at a university campus kitchen to six parent-child dyads. Two-thirds of the participating dyads were White and one-third was Black. Only caregivers attended the first six lessons, where they learned hands-on food preparation skills and strategies to manage child mealtime behavior. Both caregivers and child participants attended the remaining four lessons, where parents used the skills learned during the first six lessons to prepare and serve family meals to their child. Child and caregiver outcomes (dietary intake; frequency of dining away from home) were assessed at baseline and post-test.

At post-test, the proportion of dinners consumed away from home by participating caregiver/child dyads significantly decreased (56% to 25%; p<0.05). However, significant improvements in child or caregiver dietary intake (energy; total fat; saturated fat; sodium) were not observed at the end of the 10-week intervention. The Robson et al. intervention does provide some promising potential for family meals intervention to make improvements on the location of family meals (home vs. away from home).

However, the absence of improvements in child and caregiver diet quality, in addition to the small sample size and absence of a control group, warrants further research to assess

25 whether family meals interventions aiming to increase the frequency of family meals eaten at home can improve child diet quality.

26 Potential Underlying Mechanisms for the Association Between Family Meals and Child Diet Quality

Child Involvement in Food Preparation

Eating behaviors that fuel development of obesity are established early in life and become difficult to change thereafter [21]. Establishing healthy dietary behaviors during childhood is therefore critical to prevent excessive weight gain in children. Family meals provide several opportunities for children to develop healthy dietary behaviors, through family meal food preparation, mealtime interactions, and clean-up [72]. Cross-sectional

[78, 98–102] and intervention [103–109] studies have demonstrated evidence of a positive association between child involvement in food preparation and child diet quality.

The major features and findings of studies examining the association between child involvement in food preparation and child diet quality are summarized in Table 2.

27 Table 2. Child Involvement in Food Preparation and Child Diet Quality Author Sample Sample Size Study Description Outcomes Findings Cross-Sectional Studies Freq home food prep was Larson N et • 5th-12th graders •Students (n=4746) Students completed Project EAT survey •Freq of home food prep associated with lower al., 2006 •11-18 yrs old •Schools (n=31) assessing home food prep & Harvard involvement intakes of fat (p<0.01) & •Attending public Youth FFQ during school day •Family meal freq higher intakes of FV school in •Child diet quality (p<0.05) Minneapolis/St Paul th Chu YL et • 5 graders • Students (n=3398) Children completed survey assessing •Freq of home food prep More freq home food prep al., 2012 •10-11 yrs old •Schools (n=151) frequency of home food prep involvement associated with increased •Attending public involvement & rated preference for 12 •FV preference FV preference (p<.05) school in Alberta, FVs during school day

28 Canada th Chu YL et • 5 graders • Students (n=3398) Children completed Project EAT survey •Freq of home food prep Involvement in home food al., 2014 •10-11 yrs old •Schools (n=151) assessing home food prep & Harvard involvement prep positively associated •Attending public Youth FFQ. Diet quality calculated using •Child diet quality (p<0.001) with child diet school in Alberta, DQI-I quality Canada

Slater J, •12-17 yrs old •Adolescents Data from the Canadian Community •66% involved in Freq home food prep Mudryj AN, •Youth/caregivers (n=10,098) Health Survey 2013 Rapid Response on choosing meals & grocery positively associated with 2016 completed Canadian •Canadian provinces Food Skills. Assessed frequency of child shopping family meal frequency Community Health (n=10) participation in family meals food prep •33% involved in food survey and frequency of family meals prep •75% ate family meal together almost every day Table 2. (continued)

28 Table 2. Child Involvement in Food Preparation and Child Diet Quality (continued) Interventions Brown BJ & •Youth (>7 yrs old) •Youth (n=229) Extension produce cooking program •Daily FV intake Significant increase in Hermann •Adults •Adults (n=373) providing education on FV prep skills daily FV intake among JR, 2005 •Live in county in •Counties (n=28) (demonstrations & hands-on), food safety adult (p<0.05) & child Oklahoma & nutrition related to produce. (p<0.05) participants implementing Participants completed pre- & posttest program questionnaires

Cunningham • Fourth graders •Students (n=257; 10-wk school food & nutrition education •FV preferences Among CWK-CT students: -Sabo L et •Aged 8-12 yrs old 75% White) program. Intervention classes received 6 •Cooking SE •Fruit (p=.087) & al., 2013 •Attending °CWK-CT (n=129) cooking/tasting lessons (3, 1-hr FV •Cooking attitudes vegetable (p=.007) participating public °Control (n=118) tastings & 3, 2-hr cooking classes). preference increased schools in Colorado •Schools (n=4) Surveys completed at baseline & post- •Cooking attitude & SE 29 intervention. increased

Gibbs L et •Parent-child dyads • Students (n= 764) Intervention schools: kitchen garden •Willingness to try Participation in food al., 2013 •3rd-6th graders °Intervention (n=463) program (45 min/wk garden class & 90 new foods preparation increased child •Aged 8-12 yrs old °Control (n=280) min/wk cooking) built into curriculum. •Child dietary intake willingness to try new •Attending public •Schools (n=12) Cooking class included hands-on food foods (p=0.001) schools in Victoria, °Intervention (n=6) prep. Parents completed baseline & post- Australia °Control (n=6) intervention surveys.

Cunningham • Fourth graders •Students (n= 961; School food & nutrition education •FV preferences FV preference increased in -Sabo L et •Aged 8-12 yrs old 84% Hisp.) program with 2 intervention arms. CWK- •Cooking SE both intervention arms al., 2014 •Attending °CWK-CT (n=442) CT: cooking (5, 2-hr classes with hands- •Cooking attitudes (p<0.05) with greatest participating public °CWK-T (n=226) on FV prep/cooking) & tasting (5, 1-hr increase in CWK-CT arm schools in New °Control (n=312) classes with FV tasting). CWK-T: Mexico •Schools (n=11) tasting-only. Surveys completed at baseline & post-intervention. Table 2. (continued)

29 Table 2. Child Involvement in Food Preparation and Child Diet Quality (continued) Anderson •Families with >1 •Families (n=29; TST program: 16 weekly 2-hr classes, •Child diet (FV; SSB) •Significant increase in child JD et al., child aged 7-17 yrs 79% Hispanic/ involving: cooking/ eating; nutrition •Physical activity fruit (p=0.006) & vegetable 2014 w/ BMI >85%tile Latino) education activities; & physical activity. •Child BMI (p<.001) intake & physical °Parents (n=33) Surveys completed baseline & post- activity (p=0.001) °Children (n=62) intervention.

van der •Parent-child •Parent-child dyads Intervention: child cooks meal with •Weighed food intake Children in intervention Horst K et dyads (n=47) parent. Control: parent cooks meal. (grams) group ate significantly more al., 2014 • Children aged 6- °Intervention (n=25) Meal= cauliflower, chicken fingers, pasta, •Total energy intake salad (41.7 g; 76.1%), 10 yrs °Control (n=22) salad. Children served same portions of chicken (21.8 g; 27%) & total food. Meal components measured before energy (+85 kcal; 24.4%) & after meal. FV: fruit and vegetables SSB: sugar-sweetened beverages

30 DQI-I: Diet Quality Index-International

SE: self-efficacy TST: Taking Steps Together

30 Cross-Sectional Studies Cross-sectional studies investigating the relationship between child diet quality and child involvement in food preparation define the child diet quality outcome in a variety of ways [78, 110–112]. While some investigators define child diet quality based on food preference [110] and others use actual/reported intake [78, 100], results consistently demonstrate a positive association between child involvement in food preparation and child diet quality, regardless of how the child diet quality outcome is defined [78, 100, 110, 112].

Intervention Studies Intervention studies aiming to improve child diet quality by engaging children in food preparation have increased significantly over the past five years. While similarities exist among interventions aiming to improve child diet quality through child engagement in food preparation, distinct differences are apparent in their design and implementation.

Similar to cross-sectional studies investigating the relationship between child involvement in food preparation and child diet quality, child dietary outcomes and assessment methods vary among interventions studying this relationship [103–108].

Cooking with Kids (CWK) is a school-based food and nutrition education program aiming to increase child fruit and vegetable preference among 8-12 year old children

[105, 107]. Two versions of the CWK program were implemented in eleven public schools in New Mexico, to a predominantly Hispanic sample of students (n=961) [107].

The first version of CWK was delivered to 539 students and included 5, 2-hr classes with

31 hands-on fruit and vegetable preparation/cooking and tasting (CWK-CT). The second version of CWK was delivered to 294 students and included 5, 1-hr classes with fruit and vegetable tastings only (CWK-T). The remaining 397 students were assigned to the control group and did not receive either of the CWK programs. Child fruit and vegetable preference increased significantly among children in both versions of CWK compared to children in the control group (p<0.05), with the greatest improvements observed among children participating in the CWK-CT program that included child food preparation

(p=0.045).

While child involvement in food preparation through the CWK-CT program was demonstrated to be efficacious in increasing child fruit and vegetable preference among

8-12 year old Hispanic children, its effect on non-minority children remained unknown.

To assess the impact of child involvement in food preparation through the CWK-CT program on non-minority children, the CWK-CT program was implemented in four public schools in Colorado to a sample (n=257) of predominantly White (75% ), 8-12 year old students [105]. The CWK-CT program included 6 cooking and tasting lessons

(3, 1-hr fruit and vegetable tastings and 3, 2-hr cooking classes) delivered over 10 weeks.

One hundred thirty-seven students received the CWK-CT program, while 120 students served as comparison controls and did not receive the program. Fruit and vegetable preference was assessed in all students at baseline and post-intervention. Among students participating in the CWK-CT program, there was a much greater increase in fruit

32 (p=0.087) and vegetable preference (p=0.007) at post-intervention compared to students not participating in the CWK-CT program.

Along with assessing the impact of child food preparation involvement on child diet quality by assessing child fruit and vegetable preference, child willingness to try new foods has been used as an alternative child diet quality measure [106]. The efficacy of improving children’s willingness to try new foods through a school-based kitchen-garden program was assessed in a sample (n=764) of 8-12 year old school children. Students

(n=463) attending intervention schools (n=6) received the kitchen-garden program, which included a 45-minute garden class and 90-minute cooking/food preparation class each week as a part of the school curriculum. Students (n=280) attending the control schools

(n=6) did not receive the kitchen-garden program. Among students attending the intervention schools and participating in the kitchen-garden program, willingness to try new foods increased significantly (p=0.001) compared with students who did not receive the kitchen-garden food preparation/cooking program.

Impact of child involvement in food preparation on child diet quality has also been assessed through self- and parent-report of child dietary intake [103, 104].

Oklahoma Cooperative Extension aimed to evaluate the impact of an interactive, hands- on fruit and vegetable preparation and nutrition class series on increasing fruit and vegetable intake among youths and adults [104]. The bi-weekly class series, implemented in 28 counties in Oklahoma included 8 classes delivered by Extension

33 Educators to groups of youth (n=229) and adult (n=373) participants. Each class included hands-on fruit and vegetable preparation, taste testing, and fruit and vegetable nutrition education. Participants completed an FFQ at baseline and post-intervention.

Following the produce class series, daily fruit intake increased significantly in youth

(p<0.0001) from 1.1 servings per day to 2.3 servings per day. Daily vegetable intake also increased significantly (p<0.0001) in youth following the produce class series, from 1.4 servings per day to 2.4 servings per day.

Effects of child involvement in food preparation on child diet quality has also been assessed through a community-based healthy lifestyle intervention targeting low- income ethnic minorities in Minneapolis, Minnesota [103]. Families with at least one child between 7-17 years of age with a BMI-for-age >85%tile were referred by physicians at a child-care clinic to the Taking Steps Together (TST) program. The 16- week TST program included weekly lessons that engaged families in wellness educational activities, physical activity and food preparation and consumption. Impact of the TST program was evaluated for improving obesity-related behaviors, including reducing screen time, increasing physical activity, increasing intake of fruits and vegetables and decreasing intake of SSBs. Participating families (n=29; 79%

Hispanic/Latino) completed a series of questionnaires, including FFQs, at baseline and post-intervention. By the end of the intervention, fruit (p=0.006) and vegetable

(p<0.001) intake increased significantly among youth participants.

34 While self-report of dietary intake is commonly used in clinical and community settings, self-report dietary intake data is subject to reporter bias and may therefore lack accuracy [110, 111, 113–115]. Direct observation and weighed food records are often used in laboratory research settings to eliminate potential reporter bias [110, 113]. The impact of child involvement in food preparation on child dietary intake was assessed via direct observation and weighed food quantities to assess the impact of child involvement in food preparation on child dietary intake [108]. A sample of parent-child (aged 6-10) dyads (n=47) were randomized to an experimental ‘child cooks’ intervention group

(n=25) or ‘parent cooks’ control group (n=22) to examine the effect of child involvement in meal preparation on child dietary intake. Children in the ‘child cooks’ experimental group prepared a meal (cauliflower, pasta, chicken fingers, garden salad) with the help of their parent in a research kitchen. Children in the ‘parent cooks’ control group watched television while their parent prepared the meal in the research kitchen. Weighed portions of the prepared meal were served to each parent-child dyad in the research kitchen. Child plate waste was weighed after the meal to assess consumption of each meal component.

Children who were involved in food preparation in the ‘child cooks’ experimental group consumed significantly more salad (41.7 grams; p=0.008), chicken (21.8 grams; p=0.025) and total energy (+84.6 kcal; p=0.007) compared with children who were not involved in food preparation.

Engaging children in food preparation of nutritious foods may be a mechanism to improve child diet quality. Despite differences in study design and child diet outcomes,

35 the improvements in child diet outcomes resulting from child food preparation interventions provide evidence for a promising approach to improve child diet quality.

36 Home Environment

Prevention and treatment of obesity requires adoption of lifelong healthy eating habits and an active lifestyle [67]. The Expert Committee on Childhood Obesity emphasizes the need to engage parents and alter the home environment to integrate sustainable healthy dietary habits to prevent and treat childhood obesity [116]. The home environment has the potential to alter children’s dietary habits in numerous ways, with parents playing a very influential role in creating the home environment [67]. Parents influence the home food environment by making foods available or unavailable in the home [67, 117–121], establishing mealtime routines and expectations [118, 122–124], creating the physical and emotional environment of mealtimes [80, 125], and role modeling dietary behaviors [126–131]. Family-based childhood obesity interventions have shown high efficacy when they integrate changes in the home and family environment, as parental support, family functioning and the home environment are critical determinants of child behavior change [66, 67, 132]. The major features and findings of studies examining the association between home environment and child diet quality are summarized in Table 3.

37 Table 3. Home Environment and Child Diet Quality Author Sample Sample Size Study Description Outcomes Findings Cross-Sectional Studies Home Food Availability & Parent Role Modeling Kratt P et •Parent-child dyads •Parent-child Parents sent home FV availability •Child FV intake Home availability of FV positively al., 2000 •Child: 4th grader dyads survey & FFQ. Children completed •Parent FV intake associated with parent (p=.001) & (n= 1196; 5, 24-hr dietary recalls to assess •Home FV availability child (p=.01) FV intake 83% White) dietary intake. Cullen KW •4th, 5th & 6th graders • Students Students completed food record for •Child dietary intake Home FJV accessibility & et al., 2003 •Attending (n=225; 75% 6 consecutive days at school. Data (FJV) availability accounted for 10% participating Hispanic) collection day #6, students •Home FJV availability variance in child FJV intake. Home elementary schools in •Schools (n=9) completed home FJV availability •Home FJV accessibility FJV availability & accessibility Houston, Texas & accessibility questionnaire. significantly predicted child FJV intake (p<0.05). Bere E & •Parent-child dyads •Parent-child Students completed FFQ & FV •Child FV intake & Accessibility (p<.01), modeling th th 38 Klepp K, •6 & 7 graders dyads behavior/environment survey at preferences (p=.03) & parent FV intake p<.01)

2004 •Attending (n=1950) school. Survey assessing FV •Parent FV intake & were positively associated with participating primary behavior/ environment & FFQ sent modeling child FV intake. school in Norway home for parent completion. •Home FV accessibility Campbell K •Adolescent-parent •Adolescent- Survey packets were mailed to •Adolescent & parent diet Maternal intake & home availability et al., 2007 dyads parent dyads home. Adolescents completed an •Home food availability of unhealthy foods were inversely •Adolescents (12-13 (n=345) FFQ &HFE survey. Parents associated with adolescent diet yrs old) completed an FFQ, HFE & the quality Child Feeding Questionnaire.

DeBourdeau •European primary •Students Children completed survey •Child FV intake Child FV intake associated with dhuij I et al., school children (n=13,305) assessing FV intake & potential •Parent FV modeling parent modeling (p<.02) & home 2008 •Countries predictors (social & •Home FV availability FV availability (p<.02). (n=9) environmental) of FV intake during school Table 3. (continued)

38 Table 3. Home Environment and Child Diet Quality (continued) Wyse R et •Parents of 3-5 yr old •Parents Phone survey completed by parents •Child FV intake Child FV intake positively al., 2011 children (n=396) assessing child diet (Children’s •Parent FV intake associated with parent FV intake •Child attends non- Dietary Questionnaire) & HFE •FV availability (p=0.005), FV availability (p=.006) govt preschool New (Child Feeding Questionnaire) •FV accessibility & FV accessibility (p=0.012) South Wales, Aust Wang L et •Parent-child dyad •Parent-child Clinic visit: child height & weight •Child dietary intake Availability of unhealthy foods & al., 2013 •Overweight/obese dyads (n=67; assessed. Parents completed HFE (FV, sweets, fats) inappropriate parent modeling child (>85%tile) 90% White) survey & Eating Pattern •Home food associated with increased risk for •Aged 5-11 yrs •Clinics (n=3) Questionnaire to assess child diet. availability (FV, poor child diet quality (p<0.01). •Patient at sweets, fats) participating clinic in •Parent modeling Appalachia US

Couch SC et •Parent-child dyads •Parent-child Parents & children completed 3, •Child dietary intake Child FV intake positively 39 al., 2014 •Child: 6-11 yrs old dyads (n=699) 24-hour dietary recalls via phone. •Parent modeling associated with parent modeling

DASH diet quality score was •Home food (p<.001). Child DASH score calculated. Parents completed a availability positively associated with home HFE survey. Child height & •Child BMI availability of healthy foods weight assessed at child’s home. (p<0.01).

Draxten M •Parent-child dyads n= 160 dyads Baseline HOME Plus data to •Correlation between Significant correlation between et al., 2014 from HOME Plus assess association between parent parent role modeling of parent and child fruit and salad •Child: 8-12 yrs old and child F&V intake. Child F&V intake at dinner & intake at dinner (p<0.05). •BMI >50%tile dietary intake assessed via 24 hr child F&V intake •White (71%) recalls. Parent intake assessed via •Economic assistance parent role modeling survey. (35%)

Table 3. (continued)

39 Table 3. Home Environment and Child Diet Quality (continued) Couch SC et •Parent-child (6-11 yrs N= 699 Participants from Neighborhood •Home food availability Child F&V positively associated with al., 2014 old) dyads parent-child Impact Kids study. Parents •DASH score parent modeling (p<0.001). DASH •White (81%) dyads completed HFE survey. Kids Child dietary intake score was positively associated with •Hispanic (17%) completed 3, 24hr dietary recalls (servings: F&V; SSBs) healthful home food availability •Low income (14%) & BMI was measured. •Parent modeling (p<0.001). Child BMI z-score •Child BMI z-score negatively associated with parent modeling (p<0.05). Television Viewing During Mealtime Coon K et •Parent-child dyads •Parent-child Parents completed HFE survey. •Child dietary intake Mealtime TV viewing: al., 2001 •4th, 5th & 6th graders dyads Children participated in 3, 24-hr •Presence of TV during •Positively associated with (p<.05): red (n=91) dietary recalls (1 in-person & 2 meals meat, pizza, snacks, SSBs telephone) •Inversely associated with (p<.05): FV Feldman SF •5th-12th graders • Students Students completed HFE survey •Family meal freq Mealtime TV viewing: et al., 2007 •Attending (n=4746) & Youth FFQ during gym & •TV viewing during meals •Positively associated with (p<.05):

40 participating middle or •Schools health classes. •Adolescent dietary intake SSBs

high school in (n=31) •Inversely associated with (p<.05): Minneapolis, MN vegetable, calcium-rich foods Colapinto •Parent-child dyads •Parent-child Parents sent HFE survey to •Portion size preference Frequent meal TV viewing associated CK et al., •5th graders dyads complete. Students completed •Child dietary intake with larger portion preference of 2007 •Attending public (n=4966) Youth FFQ, portion size survey & •Child BMI unhealthy foods (p<.05), poor diet primary schools in •Schools had height & weight measured at quality (p<.05) & increased energy Nova Scotia (n=291) school. intake (p<.05).

Liang T et •Parent-child dyads •Parent-child Study-specific survey assessing •TV viewing during meals Mealtime TV viewing: al., 2009 •5th graders dyads home environment was sent home •Child dietary intake (FV, •Positively associated with (p<.05): •Attending public (n=4966) for parents to complete. Students SSBs, Diet Quality Index) SSBs, fat, snack foods primary schools in •Schools completed the Youth/ Adolescent •Inversely associated with (p<.05): FV, Nova Scotia (n=291) FFQ in school. overall diet quality Table 3. (continued)

40 Table 3. Home Environment and Child Diet Quality (continued) Powell F et •Mother/child (2-4 yrs n=75 dyads Typical mealtime observed by a •Mealtime structure & •Children whose mothers ate with al., 2016 old) dyads researcher in the home. environment them and ate the same foods refused •London Mealtimes were coded for •Child eating behaviors fewer foods and were easier to feed •White (97%) mealtime structure and (Child mealtime coding •No mealtime TV viewing was environment and child eating scheme used) associated with fewer fussy eating behaviors. behaviors

Trofholz AC •Families with 6-12 yr n=120 families Data collected during home visits •Meal diet quality Presence of TV was negatively et al., 2017 old child via surveys/interviews (24 hr •Child dietary intake associated with the dietary •Overweight (50% of dietary recalls, anthropometry) (HEI score) healthfulness and emotional children) and direct observation (8 home •Weekly fast food atmosphere of the meal and child diet •Healthy weight videos of family meals) intake quality. TV viewing was positively (50%) •Child BMI associated with fast food for family •Low-income •TV viewing at family meals

41 •Minority meals

Robertson • Families with >1 •Families Families for Health: 12-weekly •Child BMI z-score Significant reduction in child BMI z- W et al., overweight/ obese (n=21) 2.5-hour lessons with parallel •Child exposure to score (p=0.008) & unhealthy food 2008 child (>85%tile) parent-child programming. unhealthy food exposure (p=.001). •7-11 years old Baseline & post-intervention data •Child FV intake collected at child- & parent-level. Haire-Joshu • Families with >1 •Families GRCT. Intervention sites: receive •Child FV intake Intervention families: improvements D et al., child (2-5 years old) (n=1306) •PAT H5-KIDS + PAT. Control sites: •Parent FV intake in child FV intake (p=0.05), parent 2008 •Family attending a sites (n=16) receive PAT only. PAT: child •Home FV availability FV intake (p=0.04) & home FV participating PAT site °Intervention development through parent availability (p=0.01). Parents’ FV in SE Missouri (n=8) teaching program delivering 5 intake change was significant °Control (n=8) group & 5 home visit sessions predictor of child’s FV intake annually. H5-KIDS: 4 home visits change. & 4 tailored nutrition newsletters annually. Table 3. (continued)

41 Table 3. Home Environment and Child Diet Quality (continued) Rosenkranz •Parent-child dyads •Parent-child Girls attended 4 weekly, 2-hr •Family meal freq Family meal frequency significantly R et al., •6-12 years old dyads lessons & learned family meals •Eating while watching increased at post-intervention 2009 •Girls (n=30) topics through interactive skill TV (p=0.008) •Attending •Day care building & activities. Parents •Child dietary intake participating summer programs completed a HFE survey at day care program baseline & post-intervention. Story M et •Parent-child dyads •Parent-child RCT. Intervention group: 12-wk •Child dietary intake Intervention group significantly al., 2012 • African-American dyads program Child (2 lessons/wk): (FV, SSBs) decreased availability of high-fat girls (n=54) obesity-prevention activities & •Food availability (high- foods (p=0.001) & increased •Aged 8-10 °Intervention skill building. Parents (bi- fat, low-fat foods) availability of low-fat foods (p=0.07) •Child attends (n=26) weekly lessons & newsletters): •Child BMI & low-fat food prep practices participating MN °Control (n=28) healthy HFE focus. Outcomes (p=0.01). elementary school • Schools (n=3) assessed at baseline & post- intervention.

42 Tabak RG et • Parents with >1 child •Parents (n=43) Intervention parents received 2 •Child BMI z-score Intervention parents increased home

al., 2012 2-5 years old °Intervention phone calls with dietitian & 4 •Child dietary intake vegetable availability (p=0.02) & (n=22) tailored newsletters (focus: •Home vegetable offering FV for snacks (p=0.04) °Control (n=21) vegetable availability, modeling availability & family meals) over 4 months. •Offering FV Outcomes assessed at baseline & post-intervention. Cullen KW •Parent/child (8-12 yrs n=126 dyads Randomized (control & •Parent BMI (self- •Intervention families had et al., 2016 old) dyads intervention) clinical trial. Eight- report) significantly lower (p<0.05) 100% •Black (100%) session, web-based program •Home food availability fruit juice intake at post-test •Low-income (57%) healthy home food environments •Family food prep •Menu planning skills were & dietary behaviors (designed practices significantly higher (p<0.05) among for parents). Data collected at: •Menu planning intervention parents at follow-up baseline; post-intervention; 4 •Parent SE •Home fruit availability improved in month follow-up. •Child diet ( YRBS) both the intervention & control dyads FV: fruit and vegetables FFQ: food frequency questionnaire HEI: Healthy Eating Index FJV: fruit, juice and vegetables H5-KIDS: High 5 for Kids PAT: Parents as Teachers HFE: home food environment

42 Cross-Sectional Studies Throughout childhood, much of a child’s eating behavior takes place in the home environment, causing the home food environment to have a potentially significant influence on the dietary habits of a child [133, 134]. Cross-sectional studies examining the relationship between home food environment and child diet quality have investigated this relationship by looking at various aspects of the home food environment.

-Home Food Availability- Cross-sectional studies have aimed to determine if a relationship exists between home food availability and child diet quality. In studies investigating this potential relationship, child diet quality is usually defined in terms of daily servings of specific foods (e.g. fruits, vegetables)[118, 121, 133, 135–138] or by calculating a diet quality index score for a child’s total dietary intake using data collected from 24-hour dietary recalls or FFQs [139]. Regardless of the methodology used, increased home availability of nutritious foods/beverages has been demonstrated to be positively associated with a better child diet [118, 121, 133, 135–139].

-Parent Role Modeling of Dietary Behaviors- Parents are frequently referred to as ‘agents of change’ in childhood obesity interventions, as they have a crucial role in establishing and fostering the dietary behaviors of their children [140]. Creating the home food environment is one of the most influential roles parents have in shaping the dietary habits of their children. Cross- sectional data demonstrate an association between parent role modeling of dietary behaviors and child diet quality, with positive parent role modeling associated with

43 increased child diet quality [118, 136–138] and unhealthy parent role modeling associated with poor child diet quality [121, 133, 138].

-Television Viewing During Mealtime- Cross-sectional, longitudinal and epidemiological studies have demonstrated an association between television viewing and childhood obesity [141–144]. Several hypotheses exist to explain this relationship, including displacement of time for physical activity and advertising of unhealthy foods and beverages [145]. Several cross-sectional studies have demonstrated an association between total daily television viewing and greater caloric intake or poorer diet in children [146–148]. Once this association between total daily television viewing and child diet quality was identified, cross-sectional studies began to investigate the relationship between television viewing during mealtime and child diet quality [121, 149–151].

Initial studies examining the relationship between mealtime television viewing and child diet quality indicate a significant, inverse relationship exists [121, 149, 150,

152]. Indicators of poor diet quality, including pizza, salty snacks, sweets [121, 150], and

SSBs [121, 150, 152], have been shown to be positively associated with television viewing during mealtime, while nutrient-rich foods, such as fruits, vegetables, and low- fat dairy, have been shown to be negatively associated with mealtime television watching

[121, 150, 152]. In addition, recent cross-sectional data demonstrate that children who frequently watch television during mealtime consume larger portions of energy-dense,

44 nutrient-poor foods compared with children who rarely watch television during mealtime

[149].

Interventions Interventions aiming to improve child diet quality through the home environment are frequently family-based interventions that target multiple components of the home environment. Families for Health is a 12-week nutrition and physical activity educational program targeting overweight and obese children (7-11 years old) and their parents [153].

Weekly 2 ½ hour lessons include parallel programming for children and parents that use facilitated discussions, role-playing, hands-on learning and goal setting to provide education on creating a healthy home environment. Efficacy of Families for Health on improving the home environment was assessed in 21 families using child anthropometric and home environment data, collected at baseline and post-intervention. Following the

12-week intervention, participating families reported significant improvements in the home environment, including reduced exposure to unhealthy foods (p=0.001) and increased structure to family meals (p=0.001). Despite change in child fruit and vegetable intake not being significant (p=0.119), child BMI z-score was significantly reduced over the course of the intervention (-0.18; p<0.008). While the Families for

Health program has demonstrated positive results at the child- and family-level, the lack of a control group warrants further assessment of the program.

45 Similar to the Families for Health program, High 5 for Kids (H5-KIDS) is a family-based intervention that aims to improve child diet quality through the home environment [154]. The H5-KIDS program was developed in partnership with Parents as

Teachers (PAT), a nationwide child development program that teaches parents of underserved families to serve as role models and educators for their young children [155].

The PAT program delivers positive parent-child communication and child development education through 5 home visits, quarterly newsletters and on-site group activities annually. The H5-KIDS program was designed to be incorporated into an existing PAT program in Missouri [154].

Aiming to improve child daily fruit and vegetable intake through the home environment, the H5-KIDS program targets the intrapersonal level (e.g., parent knowledge), interpersonal interactions exchanged between the parent and child (e.g., role modeling) and the physical environment of the home (e.g., home fruit and vegetable availability). Efficacy of the H5-KIDS intervention was assessed through a randomized control trial, in which families were randomly assigned to receive either PAT+H5-KIDS

(intervention) or just PAT (control). Child anthropometrics and home environment data was collected at baseline and post-intervention in control and intervention families.

When compared to control parents, H5-KIDS parents reported a significant increase in fruit and vegetable intake (p=0.05) and home fruit and vegetable availability (p=0.01).

Among H5-Kids children, fruit and vegetable intake increased significantly in normal weight children (p=0.02), but not in overweight children (p=0.48). Despite the

46 insignificant improvement in child fruit and vegetable intake among overweight H5-

KIDS children, the H5-KIDS intervention demonstrated the importance of parent role modeling, as parent’s change in fruit and vegetable intake was a significant predictor

(p=0.001) of child’s change in fruit and vegetable intake among H5-KIDS families.

The Girls Enrichment Multi-site Study (GEMS) is a National Heart, Lung and

Blood Institute program testing the efficacy of four interventions aiming to prevent excess weight gain in 8-10 year old African American girls [156]. One of the four

GEMS interventions, the Minnesota GEMS intervention, aimed to improve the home food environment and obesity-related behaviors of 8-10 year old African American girls through an after-school parent-child intervention [157]. Participating girls attended the after-school program twice a week for 12 weeks, learning healthy dietary and physical activity behaviors through interactive skill building activities, goal-setting and reinforcement. Parents attended bi-weekly sessions and received weekly information packets providing education on overcoming barriers to adopting healthy home practices to create a healthy home environment. Dietary and physical activity outcomes were assessed at the child- and family-level among families participating in the Minnesota

GEMS intervention (intervention group) and families not receiving the intervention

(control group). Compared to families not receiving the intervention, availability of higher-fat foods significantly decreased (p=0.001) and availability of lower-fat foods

(p=0.07) and low-fat food practices increased (p=0.01) among families receiving the

Minnesota GEMS intervention.

47 While interventions aiming to improve the home food environment are frequently implemented at the family-level, parent-targeted interventions have been demonstrated to be an effective strategy for modifying the home food environment. The 4-month Family

Ties to Health intervention aimed to improve the home food environment in families with young children by working with parents to increase availability and access to healthful foods in the home [158]. Two personalized phone calls with parents and four monthly tailored newsletters directed to parents aimed to increase home vegetable availability, positive parent role modeling, and family meal frequency through problem-solving, personalized feedback, and provision of support and encouragement. Data was collected on child anthropometrics, child dietary intake and the home environment at baseline and post-intervention in families randomly assigned to receive the Family Ties to Health program (n=intervention group) and not receive the intervention (control group).

Compared to control families, home vegetable availability (p=0.02), offering fruits and vegetables for a snack (p=0.04) and parent SE for making healthy changes to the home environment (p=0.02) significantly increased in families participating in the Family Ties to Health program.

Another parent-targeted, web-based intervention aiming to promote healthy home food environments and dietary behaviors among Black families was led by Cullen KW et al [159]. The eight-session, nutrition education web-based program was assessed for efficacy in a randomized clinical trial of 126 Black parent/child (8-12 years old) dyads.

Outcomes, which included parent self-report BMI, home fruit and vegetable availability,

48 family food preparation practices, menu planning, and parent self-efficacy for healthy dietary behaviors, were assessed at baseline, post-intervention, and 4-month follow-up.

Dyads receiving the web-based program had a significantly lower (p<0.05) intake of 100% fruit juice post-intervention. In addition, menu planning skills were significantly higher (p<0.05) among intervention parents, compared with control parents, at the 4-month follow-up. Finally, home fruit availability increased from baseline to post-intervention among all (control and intervention) participating dyads.

Along with targeting parents as agents of change for the home food environment, interventions have aimed to improve the home food environment by targeting children

[160, 161]. A study aiming to improve the home food environments of families with 6-

12 year old girls targeted young girls as the agent of change for improving the home food environment of participating families [161]. A healthy home environment promotion program was integrated into five summer day care programs for 4 weeks. Participating girls learned child-friendly ways to improve the home food environment each week (e.g., asking permission to turn off the television during meals; asking to buy fruits and vegetables at the store) by participating in hands-on skill building activities. Parents of participating children (n=30) completed a home food environment survey at baseline & post-intervention. Despite a significant increase (p=0.008) in frequency of family meals among families of participating girls, only modest improvements in child diet and other home food environment outcomes were observed.

49 Multi-component interventions aiming to improve the home food environment provide promising results on child diet quality. While targeting children as ‘agents of change’ for improving the home food environment have demonstrated some positive outcomes, parent engagement is essential to augment these positive outcomes.

50 Parent Self-Efficacy for Healthy Dietary Practices

Self-efficacy, an individual’s level of confidence in performing a behavior, is often considered a precursor to behavior change [162–165]. According to the Social

Cognitive Theory of behavior change, an individual’s level of SE will determine the amount of effort expended to achieve a desired outcome and the coping mechanisms used when barriers interfere with progress towards achieving a desired outcome [162, 165,

166]. Self-efficacy is considered the ‘central cognitive core’ to parenting competence, as it drives a parent’s beliefs about whether they can influence their child’s health and well- being [167]. Parents maintaining low levels of SE for a given behavior have a tendency to give up much sooner when faced by adversity because they lack confidence in their ability to perform the behavior [162, 163]. With parents serving as ‘agents of change’ for childhood obesity [140], a parent’s SE for healthy dietary practices can be a pivotal factor in determining a child’s ability to achieve and maintain healthy dietary habits throughout childhood [168]. The major features and findings of studies examining the association between parent self-efficacy and child diet quality are summarized in Table 4.

Table 4. Parent Self-Efficacy for Healthy Dietary Practices and Child Diet Quality Author Sample Sample Size Study Description Outcomes Findings Cross-Sectional Cullen •Parent-child •Parent-child Children completed food records on 3 •Child dietary intake Parent SE for healthy eating KW et al., dyads dyads consecutive days at school. Parents •Parent SE for healthy behaviors was positively 2000 •4th-6th graders (n= 109) completed questionnaires during home dietary practices associated with child fruit •Attending •Schools (n=7) visits by trained data collectors. •Menu planning intake (p<.05), home FV participating •Food prep practices availability (p<.05) & middle school accessibility (p<.05). Houston, TX Campbell •First-time • First-time Parents sent an FFQ for child’s dietary •Child dietary intake Mothers of young children had K et al., mothers of mothers of intake & survey assessing: SE for •Parent SE for promoting significantly higher SE for 2010 children 6-24 children 6-24 promoting healthy eating; SE for healthy eating limiting non-core foods months months (n=75) limiting non-core foods; SE for limiting •Parent SE for limiting (p<0.001). Maternal SE 52 •Mothers of 4-5 •Mothers of 4-5 TV viewing non-core foods positively associated with older year old children year old children •Parent SE for limiting TV child FV intake (p<.005) & (n=84) inversely associated with sweet intake (p<0.05). Faught E •Parent/child (10- n=8,388 dyads Parents & their child completed health •Child BMI •Parent self-efficacy for et al., 11 yrs old) dyads surveys. Parent survey assessed self- •Child diet quality (Diet healthy dietary practices was 2015 •Children efficacy for healthy dietary practices & Quality Index- negatively associated with child randomly selected how much they encourage healthy International) weight status & positively from elementary eating. Children & their parents •Child FV intake (from associated with child diet school Alberta, completed a FFQ to assess child dietary Harvard Youth/Adolescent quality Canada intake. Weights & heights were assessed FFQ) on child participants. •Parent self-efficacy •Parent encouragement for child to eat healthy Table 4. (continued)

52 Table 4. Parent Self-Efficacy for Healthy Dietary Practices and Child Diet Quality Interventions West F et •Families with >1 •Families RCT: Intervention group parents received 9, Intervention group children al., 2010 child with a BMI- (n=101) 90-min. group sessions & 3, 20-minute phone •Child BMI z-score demonstrated significant for-age >85th% sessions focusing on building parent SE for •Parent SE for positive reductions in BMI z-score •Child, 4-11 yrs obesity-prevention behaviors. Data collection eating behaviors (p<0.001) & parent SE for old occurred at baseline & post-intervention. positive eating behaviors increased significantly (p=0.002) Davison •Parent-child •Parent-child 3-month program including: health campaign; •Parent SE to provide Parent SE to provide healthy KK et al., dyads dyads personalized nutrition letters; Parents Connect healthy foods foods significantly increased 2013 • Child, 2-5 yrs (n=154) for Healthy Living (6-wk parent-led program •Child BMI •Child (p<0.01). Rates of child obesity old •HS (n=5) to promote parent networking, skill-building dietary intake (p<0.01), total daily energy intake •Enrolled at a & problem-solving for diet & physical activity •Daily TV viewing (p<0.01) & daily TV viewing participating HS behaviors). (p<0.01) significantly decreased. Jurkowski •Parent-child •Parent-child Parent empowerment assessed in parents •Parent empowerment Significant improvement in parent

53 J et al., dyads dyads enrolled in CHL at baseline & post- •Parent SE empowerment related to child

2014 • Child, 2-5 yrs (n=154) intervention using Sprietzer Empowerment •Child BMI weight (p<0.01) & diet (p<0.001). old •HS (n=5) Scale •Child dietary intake Change in parent empowerment •Enrolled at a •Daily TV viewing predicted positive dietary participating HS parenting practice Powell C •Parent/child n= 633 dyads Cluster randomized trial aiming to increase •Child lunch content •Significant increase (p<0.05) in et al., dyads (2-6 yrs) (from 30 sites) parents’ packing of fruit, vegetables and (direct observation) packed vegetables from baseline 2016 •Texas whole grains. Five-wk intervention, followed •Parent self-efficacy for to post-intervention •White (72%) by 1-wk booster 3-months post-intervention. packing healthy child •Significant increase (p<0.05) & •College (82%) Intervention included: newsletters sent home lunches maintenance throughout follow- •Annual income to parents; child classroom activities; trainings up period of packed whole-grains >$100,000 (57%) for site staff. Data collected baseline, post- •Significant increase in parent intervention, pre-booster, post-booster. self-efficacy for packing healthy lunches RCT: randomized control trial SE: self-efficacy CHL: Communities for Healthy Living HS: Head Start FFQ: food frequency questionnaire

53 Cross-Sectional Studies Cross-sectional studies have investigated the relationship between parent SE for healthy dietary practices and child dietary intake, specifically child intake of healthy foods (e.g., fruits and vegetables), and unhealthy foods (e.g., sweets and salty snacks)

[169, 170]. Parent SE for planning, providing and role modeling fruit and vegetable intake has been shown to be positively associated with child fruit and vegetable intake[169, 170] and home fruit and vegetable availability and accessibility [169].

Maternal SE for healthy dietary practices has also been shown to be inversely associated with child dietary intake of unhealthy foods (e.g., sweets and salty snacks) [170].

Interventions While interventions aiming to improve child diet quality by increasing parent SE for healthy dietary practices are limited, the few interventions in this area have demonstrated promising results [171–173]. Communities for Healthy Living (CHL) is a parent-targeted intervention aspiring to improve obesity-related behaviors of young children from low-income families [173]. Implemented through five Head Start centers in upstate New York, the intervention aims to improve the weight status, dietary intake and sedentary behaviors of young children by equipping parents with the knowledge, skills and confidence to model and encourage healthy dietary behaviors at home.

Intervention components include: a health communication campaign to increase awareness and dispel myths about childhood obesity; personalized letters sent home to parents reporting child health indicators; integration of nutrition counseling into Head

Start family engagement activities; and the Parents Connect for Healthy Living Program.

54 The Parents Connect for Healthy Living Program is a 6-week parent-led program promoting advocacy, parent networking, skill building, and problem solving for diet and physical activity behaviors. The efficacy of the CHL intervention on improving child- and parent-level outcomes was assessed at baseline and post-intervention in 154 participating families. Compared with baseline, rates of child obesity (p<0.01), total daily energy intake (p<0.01), and daily TV viewing (p<0.01) significantly decreased among children of participating parents. A significant improvement (p<0.05) in parent

SE to promote healthy eating was observed among participating parents at post- intervention. A significant intervention dose effect was identified for parent SE to promote healthy eating, with a higher intervention dose predicting greater improvements in parent SE (p<0.005).

The positive parent SE outcome of the CHL intervention was further assessed to determine the effect of parent empowerment resulting from the CHL intervention on obesity-prevention parenting practices [172]. Childhood obesity parent empowerment, or the ability of a parent to control and influence their child’s risk factors [174] for obesity, is developed by increasing parent SE for controlling and influencing child risk factors for obesity [172]. Parent empowerment was assessed in the 154 parents enrolled in the CHL intervention at baseline and post-intervention by completion of the Sprietzer

Empowerment Scale [175]. Following the CHL intervention, a significant improvement in parent empowerment related to child weight (p<0.01) and diet (p<0.001) was observed among participating parents. Change in parent empowerment predicted healthy diet-

55 related parenting practices, including frequency of offering fruit (p<0.001), vegetables

(p<0.001) and not serving fast food (p<0.001) at post-intervention.

Similar to the parent-focus of the CHL intervention, the Group Lifestyle Triple P is a lifestyle-specific parenting program aiming to improve child dietary and activity patterns by increasing parents’ skills and confidence in managing weight-related behaviors of their child [171]. Group Lifestyle Triple P is a 12-week intervention with nine 90-minute group sessions and three 20-minute telephone sessions. Parents acquire knowledge and skills in strategies related to nutrition, physical activity and positive parenting through hands-on skills training, self-regulation, self-evaluation, and problem solving. To assess efficacy of Group Lifestyle Triple P, the intervention was implemented as a waitlist RCT, with child- and parent-level outcomes assessed in the intervention and waitlist control group at baseline and post-intervention. Compared to children in the control group, children of families allocated to the Group Lifestyle Triple

P intervention had significantly greater reductions in BMI z-score (p<0.001) and decreased frequency of obesity-related dietary behaviors (p<0.001). Parent SE for healthy eating behaviors significantly increased (p=0.002) in parents receiving the Group

Lifestyle Triple P compared to parents who did not receive the intervention.

The Lunch is in the Bag intervention aimed to increase parents’ packing of fruits, vegetables, and whole grains in their children’s lunches [176]. The study enrolled 633 parent-child (2-6 years old) dyads from 30 Early Child Care Education Centers in Texas.

The intervention was implemented as a clustered randomized trial, with 15 control

56 centers and 15 intervention centers. Parent-child dyads attending the intervention centers received the 5-week Lunch is in the Bag intervention and a one-week booster session three months post-intervention.

The intervention included weekly nutrition education newsletters sent to parents and child classroom activities that aimed to encourage children to ask for, eat and enjoy fruits, vegetables and whole grains. Staff at participating sites also received an intervention toolkit and training to help encourage participating parents and children to consume diets rich in fruits, vegetables and whole grains. Data was collected via direct observation of child packed punches and parent completion of a self-efficacy questionnaire at baseline, post-intervention and follow-up.

Following the intervention, there was a significant increase (p<0.05) in packed vegetables from baseline, however this increase was not maintained at follow-up.

However the significant increase (p<0.05) that was observed in packed whole-grains during the intervention period, was maintained during the follow-up period. Parent self- efficacy for packing healthy lunches containing fruits, vegetables, and whole-grains also significantly increased (p<0.05) from baseline to post-intervention.

57 Discussion

Developing and implementing sustainable, efficacious childhood obesity prevention interventions is a high priority due to the detrimental consequences of childhood obesity [55, 60, 177]. The AAP Expert Committee on Childhood Obesity has established core behaviors associated with a decreased risk for childhood obesity, including limiting consumption of SSBs, consuming recommended quantities of fruits and vegetables and participating in family meals regularly [40]. Family meals may have a ‘protective role’ against childhood obesity by providing a regular opportunity to instill positive eating behaviors in children. There is evidence that various aspects of family meals, including frequency [75–77, 79, 96], child involvement in food preparation [78,

99, 100, 103–108], parent role modeling [121, 133, 136–138, 178], home food availability [121, 133, 135–139, 178], and parent SE for healthy dietary behaviors [169–

173], contribute to positive child eating behaviors and improvements in child weight status.

To date, relatively few efforts have been made to design evidence-based family meals programs that equip parents of children with the knowledge, skills, and other resources needed to establish healthy dietary behaviors in children through family meals to improve child diet quality and weight status. The primary objective of the proposed research study is to implement and test the efficacy of a 10-week, evidence-based nutrition education and food preparation/cooking program (Simple Suppers, SS) aimed at increasing positive eating behaviors in underserved families with school-aged children.

58 Chapter 3. Theoretical Framework

59 Childhood Obesity Interventions & the Social Cognitive Theory

Childhood obesity is a complex disease, as it is influenced by biological, intrapersonal, interpersonal, environmental and societal factors [22]. The complex nature of the disease makes designing and implementing sustainable and efficacious interventions challenging. While there appears to be an infinite number of childhood obesity prevention interventions that exist, mixed and only modest findings have resulted from existing interventions [179].

Childhood obesity interventions designed with a theoretical framework have demonstrated to be most successful in producing positive intervention outcomes [180].

The theoretical frameworks used for childhood obesity interventions have varied based on the target population, goals, and outcomes of the research. Among successful family- based childhood obesity interventions targeting children 4-10 years of age, the Social

Cognitive Theory (SCT) frequently serves as the theoretical framework to promote behavior change among this target audience, as it account for both personal and environmental factors [179].

The SCT of behavior change explains human behavior as an interaction between personal, behavioral, and environmental factors [163, 164, 181, 182]. Reciprocal determinism represents this bidirectional interaction of behavioral determinants

(personal, behavioral, and environmental factors) in the SCT. According to the SCT, an individual’s behavior is guided by their expectations, beliefs, self-perceptions, goals and

60 intentions for a given behavior [181, 182]. In turn, the outcomes of an individual’s actions influence their expectations and beliefs of the behavior. The SCT’s link between environmental and personal factors is demonstrated by the way in which one’s expectations, beliefs, emotions, and cognitive capabilities are established and shaped by social influences, through modeling, instruction, and encouragement [162]. The interaction between environmental factors and behavior is demonstrated when an individual’s behavior alters the external environment, and is in turn, altered by the environment it creates [163]. As a result of the bidirectional relationship between behavior and environment, individuals are ‘products and producers of their environment’

[162].

Constructs of the Social Cognitive Theory Personal Constructs Personal constructs of the SCT include behavioral capability, outcome expectations, SE and self-regulation [181, 182]. An individual’s capability to perform a given behavior is dependent on having the knowledge and skills required to perform the behavior (behavioral capability) [162, 163]. Acquiring knowledge and skills to perform a behavior can be accomplished in various ways, including skills training and mastery experience.

Outcome expectations are the beliefs of an individual about the likelihood and value of the consequences of a behavior [162, 166]. Outcome expectations are created when an individual considers the costs and benefits associated with a behavior [166].

61 Outcome expectations can either increase or decrease the likelihood of a behavior, depending on the consequences an individual associates with the outcomes of the behavior [162]. Adoption of a healthy behavior can be encouraged by increasing positive outcome expectations and minimizing negative outcome expectations associated with a healthy behavior [166]. Positive outcome expectations can be instilled through mastery experience, in which an individual successfully participates in a desired behavior repeatedly, and through vicarious experience, in which an individual observes a peer’s successful attainment of a behavior.

Self-efficacy is an SCT construct that defines the confidence level of an individual in performing or engaging in a given behavior [162, 163, 165, 166]. An individual’s level of SE is behavior-specific, as an individual can demonstrate significantly different levels of SE depending on the behavior of interest [165]. Self- efficacy often serves as a proxy to behavior change, with increasing levels of SE increasing the likelihood of behavior performance. Breaking a behavior down into small, doable steps and celebrating success with accomplishing each step can increase SE for a behavior [166]. Self-efficacy can also be attained through mastery and vicarious experience [162, 183].

Self-regulation is a construct of the SCT that includes self-monitoring of a behavior and its results [162, 181]. Self-regulation of behavior can be achieved through goal-setting, self-observation, self-assessment and self-reinforcement [162]. Participation

62 in a desired behavior can be fostered by guiding an individual in establishing small, incremental goals and regularly monitoring progress and successes in working towards reaching these goals [166]. Coupling social support with self-regulation can further abet behavior performance.

Environmental Constructs Environmental constructs of the SCT include external and physical factors that influence behavior, such as observational learning, modeling and facilitation [163, 165,

182]. Observational learning is a process in which behavior performance results from observing, retaining and replicating behavior performance of others. Modeling, a type of observational learning, can influence behavior performance by actively demonstrating a behavior and its outcomes to observers [163]. Modeling can strengthen or weaken performance of a behavior by creating outcome expectations for behavior performance

[166]. Behaviors can be encouraged through modeling by demonstrating positive outcomes as a result of a given behavior.

Providing tools, resources and social support to make new behaviors easier to adopt describes the construct of facilitation [162]. Equipping individuals with tools, resources and social support reduces barriers to behavior performance and increases the likelihood of behavior adoption.

63 Social Cognitive Theory: Theoretical Framework for Simple Suppers

The SS program was developed with the SCT serving as the overarching theoretical framework (Figure 1). With the goal of improving child weight status by increasing child diet quality, the SS program targeted personal, socio-environmental and behavioral factors that are associated with improved diet quality in children.

Figure 1. Social Cognitive Theory as the Theoretical Framework for the Simple Suppers Program

Adult Learning Theory

While the SCT served as the overarching theoretical framework for the SS intervention, the caregiver education curriculum was developed using Adult Learning

64 Theory (ALT). Adult Learning Theory assumes five characteristics about adult learners:

1) Self-concept; 2) Learner experience; 3) Respect and readiness to learn; 4) Orientation to learn; 5) Motivation to learn [184].

As a person matures, their self-concept transitions from being a dependent learner, to being much more independent and self-directed in the learning process. In addition, adults bring life experiences and knowledge to learning experiences and readiness to learn becomes oriented towards the developmental tasks of their social roles.

As a person matures, they also appreciate problem-centered learning rather than content or subject-centered learning and immediate application of learned knowledge assists with retention of information. Finally, motivation to learn becomes internal and this internal motivation can be enhanced through creating and maintain a respectful learning environment.

These assumptions of adult learning lead to the development of the four principles of the ALT, which guided the development of the caregiver education curriculum of the

SS program. First, ALT encourages engagement of adult learners in the planning and evaluation of instruction, as adult learners thrive when they are involved in the planning and evaluation of their instruction. Extensive formative research with the target population was conducted in developing the caregiver curriculum of the SS program.

Furthermore, caregivers’ feedback and evaluation of the program is requested at the conclusion of each program. Second, ALT guides educators in building instruction

65 around learner experiences. The caregiver curriculum of the SS program utilizes the 4-A approach (Anchor, Add, Apply, Away) to root new knowledge in the life experiences of learners (anchor) and build new knowledge (add) through application (apply) and goal- setting (away) for knowledge retention. The 4-A approach further helps educators demonstrate the relevance and impact of new knowledge by drawing on past experiences of the learners and by providing immediate application of new knowledge, which enables the delivery of knowledge to be done in a problem-centered format.

66 Chapter 4. Methods and Design of a 10-Week Multi-Component Family Meals Intervention: a Two Group Quasi-Experimental Effectiveness Trial

67 Abstract

Background: Given the ongoing childhood obesity public health crisis and potential protective effect of family meals, there is need for additional family meals research, specifically experimental studies with expanded health outcomes that focus on the at-risk populations in highest need of intervention. Future research, specifically intervention work, would also benefit from an expansion of the target age range to include younger children, who are laying the foundation of their eating patterns and capable of participating in family meal preparations. The purpose of this paper is to address this research gap by presenting the objectives and research methods of a 10-week multi-component family meals intervention study aimed at eliciting positive changes in child diet and weight status.

Methods: This will be a group quasi-experimental trial with staggered cohort design.

Data will be collected via direct measure and questionnaires at baseline, intervention completion (or waiting period for controls), and 10-weeks post-intervention. Setting will be faith-based community center. Participants will be 60 underserved families with at least 1, 4-10 year old child will be recruited and enrolled in the intervention (n=30) or waitlist control group (n=30). The intervention (Simple Suppers) is a 10-week family meals program designed for underserved families from racial/ethnic diverse backgrounds.

The 10, 90-minute program lessons will be delivered weekly over the dinner hour.

Session components include: a) interactive group discussion of strategies to overcome family meal barriers, plus weekly goal setting for caregivers; b) engagement in age- appropriate food preparation activities for children; and c) group family meal for

68 caregivers and children. Main outcome measures are change in: child diet quality; child standardized body mass index; and frequency of family meals. Regression models will be used to compare response variables results of intervention to control group, controlling for confounders. Analyses will account for clustering by family and cohort. Significance will be set at p<0.05.

Discussion: This is the first experimentally designed family meals intervention that targets underserved families with elementary school age children and includes an examination of health outcomes beyond weight status. Results will provide researchers and practitioners with insight on evidence-based programming to aid in childhood obesity prevention.

69 Background

However, the current evidence linking family meals with improved child dietary intake

(increased fruit and vegetable intake, decreased sugar-sweetened beverage (SSB) intake) and weight status (decreased body mass index (BMI; (weight (kg)/height (m)2)) z-score) has significant limitations. The majority of the family meals literature – specifically in the area of childhood obesity prevention – represents observational studies, demonstrating only an associative relationship of family meals with child diet and weight status [75, 77,

83, 84].

What’s more, racial and ethnic differences have been highly understudied; given that the segment of the United States (US) child population with high prevalence of obesity is racial and ethnic minorities [185], it has been suggested that this is an area in which additional research is needed. Similarly, the existing family meals intervention research

(i.e., studies designed specifically to examine the cause and effect relationship between family meals and child diet and weight status), while strong with regard to study design, is limited and primarily targets non-Hispanic White children (8 to 12 years old), particularly from well-educated families [73, 74]. In addition, the majority of the current research fails to examine the child health impact of family meals beyond BMI (e.g., central adiposity and blood pressure (BP)), with only a small number of studies including

70 additional outcomes (e.g., disordered eating) [86–89].

Given the ongoing childhood obesity public health crisis [90] and the potential protective effect of family meals, there is need for additional family meals research, specifically experimental studies with expanded health outcomes that focus on the at-risk populations in highest need of intervention. Future research, specifically intervention work, would also benefit from an expansion of the target age range to include younger children (4-7 year olds), who are laying the foundation of their eating patterns [91], and are capable of participating in family meal preparations [92].

The purpose of this paper is to address this gap in the literature by presenting the objectives and research methods of a 10-week multi-component family meals intervention study, Simple Suppers, aimed at eliciting positive changes in child dietary intake and weight status. The Simple Suppers study is a two group quasi-experimental trial with staggered cohort design that targets underserved families with elementary school age children (4-10 years) and includes an examination of health outcomes beyond weight status.

Methods

Objectives and Hypotheses

The objectives of this study with related hypotheses will be as follows:

71 Objective 1. Assess the impact of Simple Suppers on children and caregivers of participating families relative to children and caregivers of families in the control group.

Hypothesis 1.1 Diet quality, BMI z-scores and BMI, waist circumference (WC) z-

scores and WC, and BP z-scores and BP will improve more from baseline to post-

intervention among children and caregivers, respectively, participating in the

intervention than in the controls.

Hypothesis 1.2 Diet quality, BMI z-scores and BMI, WC z-scores and WC, and

BP z-scores and BP improvements will be maintained during the follow-up period

among children and caregivers, respectively, participating in the intervention.

Objective 2. Assess the impact of Simple Suppers on the family meals environment of participating families relative to the controls.

Hypothesis 2.1 Frequency of family meals (breakfast and dinner), TV viewing

during meals, and eating family meals in a dining area will improve more from

baseline to post-intervention among families participating in the intervention than

in the controls.

Hypothesis 2.2 Frequency of family meals (breakfast and dinner), TV viewing

during the meals, and eating family meals in a dining area improvements will be

maintained during the follow-up period among families participating in the

intervention.

Study Design

72 The study will be implemented over 12-months as a two-group (intervention; waitlist control) quasi-experimental trial using a staggered cohort design (Table 5). At each of three time periods, separated by 10 weeks, a cohort of 20 families will be recruited. Each cohort will be divided into an intervention and waitlist control group (10 families in each). Consequently, a total of 60 families (30 in the intervention group and

30 in the waitlist control group) will be enrolled. Upon confirmation of study eligibility, a baseline data collection appointment will be scheduled at the participating family’s home or the community center during the two weeks preceding intervention commencement. Data will be collected on the primary food preparing caregiver and all children 4-10 years old. Written caregiver consent and child assent will be obtained.

Data will be collected on all outcomes via direct measure and questionnaires at baseline

(time point 0, T0), 10-week post-test (time point 1, T1), and 10-week follow-up (time point 2, T2).

Repeatability of the intervention (replication) will be evaluated by assessing measures on the waitlist control group at T1 and T2. Assessments will last up to 90 minutes. A team of trained research staff, blinded from group assignment, will facilitate data collection. Caregiver participants will receive a $25 grocery store gift card at each data collection point for their participation in the research. All study materials and procedures have been approved by the Institutional Review Board at Ohio State

University.

Table 5. Simple Suppers Intervention Study Design: Two-Group, Staggered Cohort Quasi-Experimental Design

Following baseline data collection, families will decide whether to enroll in either the upcoming 10-week session of Simple Suppers (intervention group) or to wait for 10- weeks (waitlist control group) after which time they would begin the Simple Suppers program. Randomization of families is not feasible because of scheduling conflicts with participating families, the desire of families to participate in the program with families they know, and the need to establish trust with the site/participating families; thus, to preserve sample size and establish trust with the site/participating families, the personal preference of participating families will determine group membership.

74 Setting

A faith-based community center will serve as the setting for the Simple

Suppers intervention. The question of “who is my neighbor?” is central to the mission and ministries of the center, which has approximately 10,000 visits per month for programming. The most recent service area census tracts demonstrate the following statistics in the center’s immediately surrounding neighborhoods: median household income is $32,307 to $58,490, compared to $51,890 in the broader county; number of families falling below the poverty line ranges from 10.7% to

24.9%, compared to 13.2% in the broader county; higher percentage of racial and ethnic minorities than the county as a whole, with 41.8% being Black compared to

21.2 in the county; and a high percentage of households that are families (58.7%).

Participants

Participants will be recruited in-person at community center events, center newsletter advertisements, and posters displayed in center. Information on recruitment materials will direct interested families to contact the research team for a screening evaluation to determine study eligibility. To be eligible for inclusion, caregivers should be the primary food preparer in the home; be responsible for at least one child 4-10 years of age; speak English as the primary language in the home; and have lived in the U.S. for at least one year. Families with one or more family members following a restrictive or therapeutic diet will be excluded.

75 Intervention

The Intervention Mapping protocol was utilized in the development of the

Simple Suppers intervention [186, 187]. Formulation of proximal program objectives occurred as the first step in the mapping process. Based on the current evidence linking family meals with improved child diet and weight status [75, 77, 83, 84], the following program objectives were formulated: 1) ‘Increase frequency of family meals prepared in the home (>5 days/week)’ and 2) ‘Improve child diet quality (significantly increase

Matrices containing the behavioral performance objectives relating to each program objective were created for each level of intervention: individual (child) and interpersonal (caregiver) (Table 6). Development of the performance objectives were guided by the evidence-based 2010 Dietary Guidelines for Americans guidelines for families and children [188]. For example, under program objective 1) (family meals), the performance objective at the individual (child) level was ‘Children participate in cooking activities’ and at the interpersonal (caregiver) level, ‘Caregivers identify health benefits of regular family meals prepared in the home’.

After formulation of performance objectives, a list of personal determinants for each performance objective was generated based on the theoretical foundation of the

76 Simple Suppers program – the Social Cognitive Theory, which posits that behavior change is a function of a reciprocal relationship between personal (e.g., behavioral capabilities and cognitive factors, such as self-efficacy and self-evaluation) and environmental (e.g., norms, modeling, and reinforcement) factors [163, 181].

Next, personal determinants were selected for children at the individual level and caregivers at the interpersonal level based on importance (i.e., strength of the association of the determinant with the behavior) and changeability (i.e., likelihood that the intervention may impact the determinant) [186]. The personal determinants included: behavioral capability; self-efficacy; self-evaluation; and norms, modeling, and reinforcement (Table 7). The performance objectives were then crossed with the selected determinants, which resulted in matrices of change objectives (Tables 7 and 8). The change objectives stated precisely what needs to change in the determinants’ behavioral outcomes in order to accomplish the performance objectives. They were developed using action words and followed by a statement of what is expected to result from the intervention [186, 187]. Because two target groups were selected, two difference matrices of change were developed under each program objective. For example, for program objective 1) (family meals), on the individual (child) level, the performance objective for children that stated ‘Children participate in meal preparation activities’ was crossed with the determinant ‘behavioral capability’, which resulted in the change objective that ‘children practice cooking skills during Simple Suppers and at home’. An example on the interpersonal (caregiver) level, also for program objective 1) (family

77 meals), is as follows: the performance objective for caregivers that stated ‘Caregivers identify health benefits of regular family meals prepared in the home’ was crossed with the determinant ‘behavioral capability’, which resulted in the change objective that

‘Caregivers know benefits of regular family meals prepared at home.’

Table 6. Overview of Formulated Program Objectives at Each Level of Intervention Program Objective Level of Target group Performance Objectives Intervention 1. Increase frequency of family meals Individual Child PO1. Children participate in cooking activities 1 prepared in the home (>5 days/week) Interpersonal Caregiver PO2. Caregivers identify health benefits of regular family meals prepared in the home PO3. Caregivers plan well-balanced weekly dinner menus that include >1 svg from 3 of the 5 food groups PO4. Caregivers plan when and where family meals will be served in the home PO5. Caregivers use list for grocery shopping PO6. Caregivers use cost-saving strategies for family meals in the home PO7. Caregivers use time-saving strategies for family meals in the

79 home

2. Improve child diet quality Individual Child PO1. Children know health benefits of eating well-balanced meals and (significantly increase HEI score snacks (p<0.05); increase daily svgs of fruits, PO2. Children participate in planning/preparing well-balanced family vegetables to Dietary Guidelines meals >2x/week recommendations;2 significantly decrease Interpersonal Caregiver PO3. Caregivers know benefits of serving well-balanced meals/snacks daily svgs of: SSBs (p<0.05 decrease)3 PO4. Caregivers serve family meal in the home that include >1 svg from 3 of the 5 food groups >1x/ week PO5. Caregivers serve >3 snacks/week that include >1 serving from 2 food groups PO6. Caregivers buy food for planned meals/snacks at grocery store PO: Performance objective HEI: Healthy Eating Index SSB: Sugar sweetened beverage Svg: Serving 1Measured by asking the question, “During the past 7 days, how many times did all or most, of your family eat dinner together?”[7] 2U.S. Departments of Agriculture and Health and Human Services. Dietary Guidelines for Americans, 2010. 7th ed., Washington, DC. December, 2010 [18] 3Measured by 24-hour dietary recall [189, 190]

Table 7. Matrix of Change Objectives by Level of Intervention for Program Objective 1 of the Simple Suppers Intervention Program objective 1: Increase frequency of family meals prepared in the home (>5 days/week)1 Level of Performance Personal Determinants Intervention Objectives Behavioral Capability Self-Efficacy Self-evaluation Norms, Modeling, Reinforcement Individual (child) PO1. Children CO1.1.1 Children practice CO2.1 Children express CO3.1 Children are able to CO4.1.1 Children participate in participate in cooking skills during confidence in determine if they meet their cooking activities at Simple meal preparation Simple Suppers and at participating in cooking weekly goal for participating Suppers family meals 1x/week activities home activities in cooking at home CO4.1.2 Children increase their CO1.1.2 Children are able participation in cooking at home to to participate in age- >1x/week in the home appropriate cooking activities at program and at home Interpersonal PO2. Caregivers CO1.2.1 Caregivers 8

0 (caregiver) identify health identify barriers to family

benefits of home meals at home made regular CO1.2.2 Caregivers know family meals benefits of regular family meals prepared at home PO3. Caregivers CO1.3.1 Caregivers know CO2.3 Caregivers CO3.3 Caregivers are able to CO4.3.1 Caregivers learn to plan, plan well- importance of planning/ express confidence in determine if they meet their prepare and serve well-balanced balanced weekly serving well-balanced planning/serving well- weekly goal for family meals from Simple Suppers dinner menus that dinner menus balanced family meals planning/serving well- Educators include >1 svg CO1.3.2 Caregivers know balanced family meals at CO4.3.2 Caregivers plan, prepare from 3 of the 5 how to plan/serve well- home and serve >1 well-balanced family food groups balanced family meals at meal at home each week home Table 7. (continued)

80 Table 7. Matrix of Change Objectives by Level of Intervention for Program Objective 1 of the Simple Suppers Intervention (continued) PO4. Caregivers CO1.4.1 Caregivers know CO2.4.1 Caregiver CO3.4.1 Caregivers able to CO4.4 Caregivers guided by plan when and importance of mealtime expresses confidence in determine if family mealtime Simple Suppers Educators in where family routines establishing mealtime routines are being established establishing family mealtime meals will be CO1.4.2 Caregivers know routines at home CO3.4.2 Caregivers able to routines during Simple Suppers served at home strategies to minimize CO2.4.2 Caregiver determine if mealtime group family meals mealtime distractions expresses confidence in distractions are minimized CO1.4.3 Caregivers minimizing mealtime plan/establish family distractions mealtime routines PO5. Caregivers CO1.5.1 Caregivers know CO2.5.1 Caregivers CO3.5 Caregivers able to CO4.5 Caregivers develop weekly use list for benefits of using a grocery express confidence determine if they meet their grocery list for planned family grocery shopping list about developing goal to develop and use a list meals CO1.5.2 Caregivers know grocery list for grocery shopping how to develop grocery list CO2.5.2 Caregivers 81 using planned family meals express confidence in

using list for grocery shopping PO6. Caregivers CO1.6 Caregivers know CO2.6 Caregivers use cost-saving how to use cost-saving express confidence in strategies for strategies to plan/prepare preparing and serving family meals at family meals at home family meals at home on home a budget PO7. Caregivers CO1.7 Caregivers know CO2.7 Caregivers use time-saving how to use time-saving express confidence in strategies for strategies to plan/prepare preparing and servings family meals at family meals at home family meals at home home when time is limited PO: Performance objective CO: Change objective HEI: Healthy Eating Index Svg: Serving SSB: Sugar sweetened beverage 1Measured by asking the question, “During the past 7 days, how many times did all or most, of your family eat dinner together?”[73]

81 Table 8. Matrix of Change Objectives by Level of Intervention for Program Objective 2 of the Simple Suppers Intervention Program objective: Improve child diet quality (significantly increase HEI score (p<0.05); increase daily svgs of fruits, vegetables to Dietary Guidelines recommendations; significantly decrease daily svgs of: SSBs (p<0.05 decrease)1 Level of Performance Personal Determinants Intervention Objectives Behavioral Capability Self-Efficacy Self-evaluation Norms, Modeling, Reinforcement Individual (child) PO1. Children know CO1.1 Children know CO2.1 Children express health benefits of eating health benefits of eating a confidence in knowing well-balanced variety of nutritious foods health benefits of eating meals/snacks well-balanced meals/snacks PO2. Children CO1.2.1 Children can CO2.2 Children express CO3.2 Children are able CO4.2.1 Children participate in participate in identify food group confidence in to determine if they cooking a well-balanced family planning/preparing well- sources in meals/snacks participating in meet their weekly goal meal/snack with peers of the balanced family meals/ CO1.2.2 Children are able meal/snack for participating in same age 1x/week during

82 snacks >2x/week to perform age-appropriate planning/preparation family meal/snack Simple Suppers

coking skills preparation CO4.2.2 Children participate in cooking well-balanced family meals/ snacks at home >1x/week Interpersonal PO3. Caregivers know CO1.3.1 Caregivers CO2.3 Caregivers (caregiver) benefits of serving well- identify barriers to offering express confidence in balanced meals/snacks well-balanced knowing benefits of meals/snacks and know serving well-balanced strategies to overcome meals/snacks identified barriers CO1.3.2 Caregivers know short- and long-term consequences of not serving well-balanced meals/snacks Table 8. (continued)

82 Table 8. Matrix of Change Objectives by Level of Intervention for Program Objective 2 of the Simple Suppers Intervention (continued) PO4. Caregivers serve a CO1.4.1 Caregivers know CO2.4 Caregivers CO3.4.1 Caregivers set CO4.4 Caregivers plan >1 family meal that includes importance of including a express confidence in goal to serve a family family meal/week that includes >1 serving from 3 of the variety of foods in meals planning/preparing well- meal that includes >1 >1 serving from 3 of the 5 food 5 food groups >1x/ week CO1.4.2 Caregivers know balanced family meals serving from 3 of the 5 groups >2 strategies to incorporate CO2.4.2 Caregivers food groups >1x/week foods from 3 food groups express confidence in CO3.4.2 Caregivers into family meals eating/serving well- monitor goal progress balanced family meals and determine if meeting established goal PO5. Caregivers serve CO1.5.1 Caregivers know CO2.5.1 Caregivers CO3.5.1 Caregivers set CO4.5 Caregivers plan >3 >3 snacks/ week that importance of express confidence in goal to serve >3 snacks/ week that include >1 include >1 serving from eating/serving well- planning well-balanced snacks/week that serving from 2 food groups 2 food groups balanced snacks snacks include >1 serving from CO1.5.2 Caregivers are CO2.5.2 Caregivers 2 food groups 81 83 able to plan >3 express confidence in CO3.5.2 Caregivers

snacks/week that include eating/serving well- monitor goal progress >1 serving from 2 food balanced snacks and determine if groups meeting established goal PO6. Caregivers buy CO1.6.1 Caregivers plan CO2.6.1 Caregivers CO3.6.1 Caregivers set CO4.6 Using list for grocery food for planned well-balanced family express confidence in goal to develop and use shopping becomes norm for meals/snacks at grocery meals and snacks developing grocery list list for grocery shopping caregivers store CO1.6.2 Caregivers CO2.6.2 Caregivers each week prepare grocery list using express confidence in CO3.6.2 Caregivers planned meals/snacks using list for grocery monitor goal progress shopping and determine if meeting established goal PO: Performance objective CO: Change objective HEI: Healthy Eating Index Svg: Serving SSB: Sugar sweetened beverage 1U.S. Departments of Agriculture & Health and Human Services. Dietary Guidelines for Americans, 2010. 7th ed., Washington, DC. December, 2010 [18]

83 Next, theory-based methods to influence change in the determinants at the individual

(child) and interpersonal (caregiver) level were selected based on the theoretical framework of the intervention (Social Cognitive Theory)[163, 166] and in reference to methods described by Bartholomew et al [186, 187]. For identifying theory-based methods to influence determinants at the interpersonal (caregiver) level, the Adult

Learning Theory, which purports that adult learning is most effective when a collaborative, problem-based approach was also referenced [191, 192]. A list of all change objectives that were linked with a specific determinant was made, and the theoretical methods were then matched with the corresponding determinant (Table 9).

Finally, practical strategies were designed to put the theoretical methods into practice

(Table 9). For example, under the family meals program objective, on the individual

(child) level, the result of crossing the performance objective ‘children participate in meal preparation activities’ with the determinant ‘behavioral capability’ was the change objective ‘children are able to participate in age-appropriate cooking skills’. The selected theory-based method that corresponded to the determinant ‘behavioral capability’ in order to achieve the change objective was facilitation.

This theory-based method was then translated into a practical strategy. In this case, a practical strategy that was chosen for the method facilitation was to ‘Learn age appropriate cooking skills at each Simple Suppers lesson’. An example on the interpersonal (caregiver) level, also under the family meals program objective,

(caregiver) level is as follows: the result of crossing the performance objective

84 ‘Caregivers identify health benefits of regular family meals prepared in the home’ with the determinant behavioral capability was the change objective ‘Caregivers know benefits of regular family meals prepared at home’. The selected theory-based method that corresponded to the determinant behavioral capability in order to achieve the change objective was active learning. This theory-based method was then translated into a practical strategy. In this case, a practical strategy that was chosen for the method active learning was: ‘Educators use the 4A method (participants think about their experience with a topic (Anchor), learn new information (Add), reinforce learning through hands-on activities (Apply), and set goals to utilize new knowledge at home (Away)) to lead weekly caregiver discussions [192, 193].

The next step was to develop the Simple Suppers curriculum in direct reference to the results produced from the aforementioned Intervention Mapping (Table 10). The initial draft was reviewed by field experts using a nutrition education curriculum assessment tool

[194]. Curriculum modifications were then made using reviewer feedback (e.g., incorporating additional hands-on learning activities in the caregiver component to enhance interactive nature of curriculum), after which additional pilot testing occurred and subsequent curricular revisions were made [195].

Finally, the Simple Suppers program design was developed with feedback from program adopters (faith-based community center staff), implementers, and the target population [196] (e.g., utilizing two (versus one) educators for the caregiver component

85 and incorporating site-based staff into the staffing structure). Each 90-minute lesson is delivered weekly over the dinner hour. Session components include: a) interactive group discussion and goal setting with caregivers; b) hands-on activities with children; and c) group family meal with caregivers and children.

Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention Program Objective Level of Determinant Change Theory-based Theory Practical Strategy Intervention Objective Method 1. Increase frequency Individual Behavioral CO1.1.1, • Facilitation • SCT • Learn new age- of family meals (child) capability CO1.1.2 appropriate cooking prepared in the home skills at each Simple (>5 days/week)1 Suppers lesson • Discuss food safety and cleanup with Educators CO1.1.1, • Vicarious • SCT • Children divided into CO1.1.2 learning three age groups (4-5 yr

87 olds; 6-8 yr olds; 9-10 yr

olds) for nutrition education & engagement in food preparation CO1.1.1, • Mastery • SCT • Learned food prep skills CO1.1.2 experience accrued/ practiced over lessons Self-efficacy CO2.1 • Facilitation • SCT • Educators provide guidance & feedback as children learn/ practice food prep skills CO2.1 • Vicarious • SCT • Participate in cooking learning activities with peers of the same age CO2.1 • Mastery • SCT • Practice cooking skills experience learned during Simple Suppers at home Table 9. (continued)

87 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention Self-evaluation CO3.1 • Self- • SCT • Establish weekly goal monitoring during Simple Suppers to practice newly learned cooking skill at home • Weekly goals are reinforced by sharing goal with caregivers during Simple Suppers family meal CO3.1 • Feedback • SCT • Discuss cooking skills used at home during past week with Educators

88 and peers during Simple

Suppers Norms, CO4.1.1, • Facilitation • SCT • Engage in family meal modeling, CO4.1.2 cooking activities with reinforcement peers and Educators during Simple Suppers • Decorate/wear aprons for food prep during Simple Suppers and at home • Share cooking skills learned each week with caregivers at start of Simple Suppers group family meals Table 9. (continued)

88 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention • Lead cleanup at Simple Suppers family meals • Families receive take- home cooking utensil during each Simple Suppers lesson CO4.1.1, • Mastery • SCT • Repeated engagement in CO4.1.2 experience family meal cooking during Simple Suppers • Weekly goal established to engage in family meal food prep at home Interpersonal Behavioral CO1.2.1, • Active • ALT • Educators use 4A 89 (caregiver) capability CO1.2.2, learning method to lead weekly CO1.3.1, caregiver discussions CO1.3.2, • Educators engage CO1.4.1, caregivers in games, CO1.4.2, meal planning & goal- CO1.4.3, setting related to weekly CO1.5.1, lesson topics CO1.5.2, CO1.6, CO1.7 CO1.3.2, • Facilitation • SCT • Educators provide CO1.4.2, • resources (e.g., recipe CO1.4.3, book, coupons, store CO1.5.2, ads) to plan family CO1.6, meals using skills CO1.7 learned at each lesson Table 9. (continued)

89 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention CO1.2.1, • Problem • ALT • Caregivers set weekly CO1.2.2, solving goals & discuss CO1.3.2, successes/challenges CO1.4.2, with meeting goals with CO1.4.3, Educators & other CO1.5.2, caregivers CO1.6, • Educators & caregivers CO1.7 provide suggestions to help peer caregivers overcome challenges preventing them from reaching their goals CO1.2.1, • Vicarious • SCT • Caregivers acquire new 90 CO1.2.2, learning knowledge through peer CO1.3.1, discussions CO1.3.2, • Caregivers participate in CO1.4.1, games, goal-setting & CO1.4.2, menu planning with peer CO1.4.3, caregivers CO1.5.1, CO1.5.2, CO1.6, CO1.7 CO1.3.2, • Mastery • SCT • Caregivers plan >1 CO1.4.3, experience family meal using skills CO1.5.2, • learned each week to CO1.6, practice skills at home CO1.7 Table 9. (continued)

90 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention Self-efficacy CO2.3, • Feedback • SCT • Discuss challenges and CO2.4.1, successes with weekly CO2.4.2, family meals goal. CO2.5.1, • Problem solve with CO2.5.2, peers to overcome CO2.6, challenges CO2.7 CO2.3, • Social • SCT • Post goal successes and CO2.4.1, support challenges throughout CO2.4.2, week on Simple Suppers CO2.5.1, Facebook page. Peers CO2.5.2, and Educators provide CO2.6, praise/support/ CO2.7 encouragement 91 CO2.3, • Modeling • SCT • Caregivers plan family

CO2.4.1, meals for upcoming CO2.4.2 week with peer caregivers during weekly lessons • Caregivers observe Educators facilitating group family meal during weekly lessons CO2.3, • Mastery • SCT • Caregivers participate in CO2.4.1, experience group family meals CO2.4.2, during weekly lessons • Caregivers plan and set weekly goals to have family meals at home Table 9. (continued)

91 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention Self-evaluation CO3.3, • Self- • SCT • Set individualized CO3.4.1, monitoring weekly SMART goals CO3.4.2, aligned with lesson CO3.5 topics CO3.3, • Feedback • SCT • Goals are reinforced by CO3.4.1, caregivers sharing their CO3.4.2, weekly goals CO3.5 • Educators and peers provide feedback/ assure appropriateness • Discuss previous week’s goal successes and challenges at beginning of each lesson. 92 Caregivers problem

solve together to overcome challenges Norms, CO4.3.1, • Facilitation • SCT • Simple Suppers group modeling, CO4.3.2, • family meals follow reinforcement CO4.4, • routine/establish norm CO4.5 for family meals • Provide weekly take- home cooking utensil to facilitate cooking at home Table 9. (continued)

92 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention CO4.4 • Mastery • SCT • Educators guide experience caregivers in establishing mealtime routine during Simple Suppers and at home Improve child diet Individual Behavioral CO1.2.1 • Facilitation • SCT • Before Simple Suppers quality (significantly (child) capability family meals, children increase HEI score name foods from each (p<0.05); increase food group in the daily svgs of fruits, upcoming family meal vegetables to Dietary Guidelines recommendations; significantly decrease 93 daily svgs of: SSBs

(p<0.05 decrease)2 CO1.1, • Vicarious • SCT • Discuss food groups and CO1.2.2 learning benefits of healthy eating with Educators and peers at Simple Suppers • Learn to cook a variety of foods with peers CO1.1, • Mastery • SCT • Children learn food prep CO1.2.2 experience skills & become familiar with a variety of food while helping prepare Simple Suppers family meals Table 9. (continued)

93 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention Self-efficacy CO2.1, • Facilitation • SCT • Learn health benefits of CO2.2 foods through interactive discussions & food prep • Engage in planning/ preparing well-balanced meals/ snacks during Simple Suppers and at home >2x/wk CO2.2 • Vicarious • SCT • Engage in food prep learning with peers of the same age • Eat Simple Suppers

94 group family meals with peers

Self-evaluation CO3.2 • Self- • SCT • Establish weekly goal monitoring during Simple Suppers to try a new food at home • Weekly goal reinforced by sharing goal with caregivers during Simple Suppers family meal Table 9. (continued)

94 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention CO3.2 • Feedback • SCT • Discuss new foods tried at home during past week with Educators and peers during weekly Simple Suppers lesson Norms, CO4.2.1, • Facilitation • SCT • Foods from >3 food modeling, CO4.2.2 groups served at Simple reinforcement Suppers family meals • Eat Simple Suppers family meals with family and peers • Children & caregivers establish weekly goal to

95 engage in preparing

well-balanced meals at home >1x/wk Interpersonal Behavioral CO1.3.1, • Active • ALT • Educators use 4A (caregiver) capability CO1.3.2, learning method to lead caregiver CO1.4.1, discussions CO1.4.2, • Caregivers learn skills to CO1.5.1, serve nutritious CO1.5.2, meals/snacks through CO1.6.1, discussions, problem CO1.6.2 solving, games, meal planning, goal setting Table 9. (continued)

95 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention CO1.4.2, • Facilitation • SCT • Caregivers plan CO1.5.2 • Caregivers provided with take-home recipe book of nutritious recipes • Families receive take- home cooking utensil during each lesson CO1.3.1, • Problem • ALT • Discuss challenges and CO1.3.2, solving successes with serving CO1.4.1, well-balanced CO1.4.2, meals/snacks CO1.6.1 • Problem solve with

96 peers to overcome

challenges CO1.4.2, • Vicarious • SCT • Simple Suppers group CO1.5.2, learning family meals contain >1 CO1.6.1 svg from all 5 food groups • Caregivers observe Educators serving/ engaging children in preparing well-balanced family meals Table 9. (continued)

96 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention CO1.4.2, • Mastery • SCT • Caregivers plan >1 well- CO1.5.2, experience balanced (contains >1 CO1.6.1, svg from 3 food groups) CO1.6.2 family meal per week during each Simple Suppers lesson using skills acquired each lesson • Learned skills repeated in caregiver family meal planning

97 Self-efficacy CO2.3, • Feedback • SCT • Discuss challenges and CO2.4.1, successes with serving CO2.4.2,, well-balanced meals/ CO2.5.1, snacks. CO2.5.2, • Problem solve as a CO2.6.1, group to overcome CO2.6.2, challenges CO2.4.1, • Social • SCT • Plan weekly family CO2.4.2, support meals with peers during CO2.5.1, Simple Suppers lessons CO2.5.2, CO2.6.1 Table 9. (continued)

97 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention • • • Post weekly successes and challenges on Simple Suppers Facebook page. Peers and Educators provide praise/support/ encouragement CO2.3, • Modeling • SCT • Educators serve Simple CO2.4.1, Suppers group family CO2.4.2 meals with >1 svg from all 5 food groups • Simple Suppers group family meals eaten with

98 Educators and peers

CO2.3, • Mastery • SCT • Families eat a well- CO2.4.1, experience balanced family meal CO2.4.2 during Simple Suppers group family meals • Caregivers plan >1 family meal >1 svg from 3 food groups each lesson for upcoming week Self-evaluation CO3.4.1, • Self- • SCT • Set individualized CO3.4.2, monitoring weekly SMART goal to CO3.5.1, serve set number of CO3.5.2, family meals at home CO3.6.1, with >1 svg from >3 CO3.6.2, food groups Table 9. (continued)

98 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention • Caregivers plan menus for the number of family meals they made their goal for the upcoming week during Simple Suppers • Goals are reinforced by sharing weekly goal and planned menus during Simple Suppers each week. Educators and peers provide feedback/ assure appropriateness CO3.4.2, 99 • Feedback • SCT • Discuss previous week’s CO3.5.2, goal successes and

CO3.6.2 challenges at beginning of each Simple Suppers lesson. Caregivers problem solve together to overcome challenges Norms, CO4.4, • Facilitation • SCT • All Simple Suppers modeling, CO4.5, group family meals reinforcement CO4.6 contain >1 svg from all 5 food groups • Receive Simple Suppers cookbook with kid- friendly, well-balanced meals Table 9. (continued)

99 Table 9. Theory-based methods and practical strategies to achieve the change objectives for selected program objectives of the Simple Suppers intervention PO: Performance objective CO: Change objective HEI: Healthy Eating Index Svg: Serving SSB: Sugar sweetened beverage ALT: Adult Learning Theory SCT: Social Cognitive Theory 1Measured by asking the question, “During the past 7 days, how many times did all or most, of your family eat dinner together?”[73] 2U.S. Departments of Agriculture and Health and Human Services. Dietary Guidelines for Americans, 2010. 7th ed., Washington, DC. December, 2010 [18]

100

100 Table 10. Simple Suppers Topics and Goals by Weekly Lessons

Lesson Topic Broad Goal for Upcoming Week 1 Making family mealtime fun! Play 1 family meal-friendly game during mealtime at 2 family meal occasions 2 Planning family meals on a budget Use 1 cost-saving strategy to plan and serve 1 well- balanced family meal at 1 family meal occasion 3 Timesaving strategies for family meals Use 1 timesaving strategy to plan and serve 1 well- balanced family meal at 1 family meal occasion 4 Connecting with your child through family meals Involve child in 1 mealtime activity at 2 family meal occasions 5 Planning well-balanced family meals Serve a family meal with 1 serving of whole grains, vegetables, and protein at 1 family meal occasion 6 Rethink your drink Serve 1 well-balanced family meal with low-fat/no sugar added beverages

101 7 Making healthy cooking tasty & easy Use 1 healthy cooking method to plan and serve 1 well-balanced family meal at 1 family meal occasion

8 Serving & eating healthy portions Serve 1 well-balanced family meal with healthy portion sizes at 1 family meal occasion 9 Eating healthy when eating away-from-home Eat 1 well-balanced, nutritious meal away-from-home at 1 family meal occasion 10 Planning fun & healthy snacks Serve 2 planned, pre-portioned well-balanced snacks to your child

101 Outcome Measures

Children and Caregivers

Diet quality. Dietary intake will be assessed by conducting three, nonconsecutive

(two weekdays, one weekend day) 24-hour (24hr) dietary recalls using the USDA’s 5- step multi-pass dietary recall method [197]. At each data collection time point, the first dietary recall will be conducted during the in-person data collection visit, the remaining two will be conducted via telephone within two weeks of the initial in-person recall. For the child dietary recalls, caregivers will provide assistance, as caregiver-assisted 24hr recalls, collected in this way (i.e., relying on three days and utilizing the multi-pass method), provide the most accurate estimate of dietary intake among children 4 to 11 years of age [189]. Caregiver 24hr dietary recalls will be conducted independently following the child recall(s). Typical daily dietary intake will be determined by averaging dietary intake across the three recalls at each time point to determine daily servings of fruit, vegetables, and SSB. Diet quality will be assessed at each point by calculating a

Healthy Eating Index 2010 score using the three 24hr dietary recalls collected [198].

Anthropometric assessments. Standardized procedures will be used to assess height and weight on all participating children and caregivers via calibrated stadiometers

(Hopkins portable road rod stadiometer) and scales (BFHA-B400SV digital scale), respectively [199, 200]. Body mass index will be calculated using measured heights and weights. Centers for Disease Control and Prevention (CDC) age- and sex-adjusted BMI growth charts will be used to determine BMI z-scores for children to adjust for expected

102 healthy growth and weight gain [199, 201]. Waist circumference will be measured on all participating children and caregivers with a tape measure at the uppermost lateral border of the hip crest (ilium) [199]. To adjust for expected growth among child participants, child WC z-scores will be determined using CDC age- and sex-specific growth charts

[202].

Blood pressure. Blood pressure will be assessed on all participating children and caregivers via automated, calibrated BP monitors (Panasonic EW3109W). Age-, sex-, and height-adjusted National Heart, Lung, and Blood Institute (NHLBI) charts will be used to appropriately classify child BP [203].

Personal determinants. We will also assess immediate intervention targets relating to behavioral capabilities. For child participants, food preparation skills and frequency of involvement will be assessed at each data collection point by caregiver completion of an age-appropriate food preparation skills questionnaire designed to assess both skill ability and frequency of involvement in practicing the skill. Working from an existing validated questionnaire designed to assess child food preparation skills (ability) among 8-10 year olds [73], three versions of the questionnaire (4-5 year old questionnaire; 6-8 year old questionnaire; 9-10 year old questionnaire) were developed to accurately assess child food preparation skills (ability) according to age appropriateness.

Assessment of frequency of involvement in practicing each food preparation skill was added to these modified questionnaires. The resulting questionnaires assessing a child’s ability to

103 participate (8 items; 4-point scale; strongly agree to strongly disagree) and frequency of participation (8 items; 5-point scale; 0 times to 7+ times) in age-appropriate food preparation skills (during the past 30 days) included 16 items.

Among caregiver participants, menu planning skills and frequency will be assessed at each data collection point by caregiver completion of an existing menu planning questionnaire [169] to evaluate immediate intervention targets relating to behavioral capabilities. The 9-item menu planning questionnaire, which has demonstrated adequate internal consistency (a= 0.68) and high test-retest reliability (Pearson test- retest= 0.89), asks respondents to rate statements regarding menu planning, meal decision-making, and grocery shopping using a 4-point scale (‘never,’ ‘sometimes,’

‘often,’ ‘always’).

A key affective variable - caregiver self-efficacy for healthy dietary practices related to family meals - will be assessed using an existing 12-item, 10-point scalar (0= not at all confident; 10= extremely confident) questionnaire [204]. The caregiver self- efficacy questionnaire, which will be completed by caregiver participants at each data collection point, has demonstrated high internal consistency (a= 0.88) among a sample of caregivers of 4-6 year old children. Tests of internal consistency will be run on all of the aforementioned questionnaires. Caregivers will also complete a brief food security questionnaire at each data collection point (6-item Short Form of the USDA Home Food

Security Survey) [205] and a demographics questionnaire to assess key participant

104 characteristics (age, race/ethnicity, education, employment, income) at baseline.

Home Environment

Family meals. Weekly frequency of shared family dinners, shared family breakfasts, television viewing during family meals, and eating family meals in a dining area will be assessed via caregiver reports with 4, 5-point scalar (0=never; 5=7 times) items from previous family meals research [206, 207].

Process Measures

Feasibility (program dose and fidelity) and acceptability will be assessed prospectively throughout the study as process outcomes. Program dose will be assessed by collecting weekly attendance (family and individual level) and tracking presence of caregiver/child dyads at each weekly lesson. Participants who demonstrate irregular attendance and/or discontinue participation will be contacted to learn underlying reasons for absence. To determine program fidelity, a trained observer will complete a program specific fidelity tool at the end of each weekly lesson, which will include a checklist of key program components, activities, and leader characteristics. Acceptability of the program will be measured with a caregiver-completed 5-item satisfaction survey administered at the end of the 10-week program [208]. At the end of programming, interviews will be conducted with a subset of caregivers to learn their perceptions of program strengths and weaknesses.

105 Sample Size and Data Analysis

Sample size was determined by examining the power of the test for comparing increases in frequency of family meals (day per week) of the intervention and waitlist control group. The data used to estimate power come from a previous pilot study, in which the main outcome of interest was the change in frequency of family dinners prepared and eaten at home together (weekly basis) from baseline to post-intervention

[209]. Change in frequency of family dinners was used to power the current study because there is strong evidence that it has a downstream effect on the outcome of interest, child BMI [83, 209–211], and there are no previous studies that show a causal effect of family dinners on BMI. Based on these data, assuming 20% attrition [209], with an expected effect size of 0.7071, there will be 80% power to detect a difference in frequency of family dinners of 3 days per week with 30 families per group for a total sample size of 60 families at α = 0.05. Because the sample size in the previous pilot study was small and uncertainty about estimated effect size was large, we used a conservative estimate of effect size (i.e., the lower bound of a 95% confidence interval) for the power calculation.

Data from each of the three cohorts will be pooled and the intervention tested by comparing change (T1-T0) in diet quality, anthropometric measures, and blood pressure of child and caregiver participants in the intervention compared to participants in the waitlist control (hypotheses 1.1 and 2.1). Multiple regression models will be used to determine the association between the difference in the response variables of interest

106 between the intervention and control group, controlling for potential confounders

(race/ethnicity, income, cohort, intervention dose), from baseline (T0) to 10-week post- test(T1) and 10-week follow-up (T2). For families in which data will be collected on multiple children, the effect of family will also be controlled by including a random effect for family.

Sustainability of intervention effects will be tested by pooling intervention group data from each of the three cohorts, comparing change (T2-T1) in diet quality, anthropometric measures, and blood pressure among intervention group participants at the end of the 10-week follow-up period (hypothesis 1.2 and 2.2). Intervention replication will be assessed by pooling waitlist control group data from each of the three cohorts, comparing post-program change in diet quality, anthropometric measures, and blood pressure among waitlist control participants (T2-T1) to intervention participants

(T1-T0). Significance will be set at p<0.05.

Discussion

We may encounter challenges engaging and developing trust with the target population, an issue that is common to intervention research with underserved families

[212, 213]. However, this study was designed to minimize this potential barrier by implementing the intervention at a local faith-based community center, which has established relationships with the target population. In addition, this study will engage current staff from the faith-based community centers to serve as educators in delivering

107 the intervention. Grounding the caregiver component in Adult Learning Theory will further enhance our abilities to engage with families, as this approach is designed to present new information in a non-threatening, approachable way.

Another limitation is the lack of randomization study design. Randomization was not appropriate for this study because preserving sample size and developing trust with the site/participating families was paramount [213–215]. We will overcome this limitation by assessing potential between- group differences at baseline and, if identified, will be controlled for in the analyses.

108 Chapter 5. Child Findings from a 10-Week Multi-Component Family Meals Intervention for Underserved Families with Children 4-10 Years Old

109 Abstract

Background: Family meals have become a growing area of research for childhood obesity prevention, as they provide an opportunity for developing and fostering social and dietary habits that influence and shape the health of a child. Cross-sectional and longitudinal research has demonstrated positive associations between family meals and child diet quality, and some on child weight status. However, there is need for additional family meals research, specifically experimental studies with expanded health outcomes that focus on the at-risk populations in highest need of intervention. Expanding the target age range to include younger children, who are laying the foundation of their eating patterns and are capable of participating in family meal preparations, would further benefit this line of research. Therefore, the purpose of this study was to assess the effectiveness of a

10-week multi-component family meals intervention study aimed at eliciting positive changes in child diet and weight status.

Methods: A 10-week family meals program (Simple Suppers) was implemented as a quasi-experimental trial with staggered cohort design. Data was collected via direct measure and questionnaires at baseline, intervention completion (or waiting period for controls), and 10-weeks post-intervention. Setting was a faith-based community center.

Child participants included 126, 4-10 year old children with racially diverse backgrounds from underserved families. The 10, 90-minute program lessons were delivered weekly over the dinner hour at a faith-based community center. Session components included: a) interactive group discussion of strategies to overcome family meal barriers, plus weekly

110 goal setting for caregivers; b) engagement in age-appropriate food preparation activities for children; and c) group family meal for caregivers and children. Main outcome measures were change in: child diet quality; and child standardized body mass index

(BMI). Regression models were used to compare response variables results of intervention to control group, controlling for confounders. Analyses accounted for clustering by family. Significance was set at p<0.05.

Results: At baseline, 126 children from 95 families enrolled in the study; mean age was

6.9 years old, approximately 62% were female, mean BMI z-score was 0.69, 60% identified as Black, and approximately 42% were reliant on WIC, SNAP, and/or NSLP federal food assistance programs. Generalized linear mixed models, adjusted for baseline values and demographics, demonstrated an intervention effect on child whole fruit intake, with children attending >7 lessons having a significantly higher whole fruit intake at post-test, which was maintained at the follow-up period. A similar intervention effect was observed regarding child BMI z-score, with intervention children attending >7 classes having a significant decrease in BMI z-score, which was maintained during the follow-up period.

Conclusions: Participation in Simple Suppers at a 70% level led to a significantly higher whole fruit intake and decreased BMI z-score, which were maintained at the 10-week follow-up. Results from this study demonstrate the potential for engagement in an evidence-based family meals program to positively impact child weight status and dietary intake among a racially diverse sample of school-aged children from underserved

111 families. Of significance given the ongoing childhood obesity epidemic.

Background

Prevalence of childhood obesity has grown substantially over the past three decades worldwide [216]. Despite a recent plateau in rates of childhood obesity in the

U.S., approximately 17% (12.7 million) of America’s youth are obese [2]. In addition, racial/ethnic, socioeconomic, and age disparities still exist, as children of racial/ethnic minority, from underserved households and between 6-19 years old are at an increased risk for obesity.

(home, school, community), and multi-level (e.g. individual, family, group) interventions integrating behavioral and environmental approaches have demonstrated to be most efficacious and are now the recommended approach for childhood obesity prevention interventions [40, 62–64].

As most obesity-related behaviors are established early in life [21], early intervention and parent/family engagement are essential [62, 63]. The American

Academy of Pediatrics (AAP) Expert Committee on Childhood Obesity

112 Recommendations advocates for family-based approaches to childhood obesity prevention to develop and establish sustainable healthy home environments that support healthy dietary and physical activity behaviors [40, 65, 67, 140].

Family mealtime has become a growing area of research for childhood obesity prevention [68, 70–73, 85, 217]. While lasting only twenty minutes, family meals provide an opportunity for developing and fostering social and dietary habits that influence and shape the health of a child [72]. Cross-sectional and longitudinal research has demonstrated positive associations between family meals and child diet quality, demonstrated by increased fruit and vegetable intake [75–79], and decreased SSB intake

[77, 78], and some on child weight status [80–82]. However, the current evidence linking family meals with improved child health outcomes has significant limitations. The majority of the family meals literature represents observational studies, targeting non-

Hispanic White children (8 to 12 years old), predominantly from well-educated families

[73, 85]. In addition, the majority of the current research fails to examine the child health impact of family meals beyond BMI (e.g., central adiposity and blood pressure (BP)), with only a small number of studies including additional outcomes (e.g., disordered eating) [86–89].

113 populations in highest need of intervention. Future research, specifically intervention work, would also benefit from an expansion of the target age range to include younger children (4-7 year olds), who are laying the foundation of their eating patterns [91], and are capable of participating in family meal preparations [92].

The purpose of this study was to address this gap in the literature by assessing the effectiveness of a 10-week multi-component family meals intervention study, Simple

Suppers, aimed at eliciting positive changes in child dietary intake and weight status. Key caregiver and family meal environment outcomes were also assessed in this study, however this paper focuses exclusively on child outcomes. The Simple Suppers study was a two group quasi-experimental trial with staggered cohort design that targets underserved families with elementary school age children (4-10 years) and includes an examination of health outcomes beyond weight status [218].

Methods

Objectives and Hypotheses

The objective of this study with related hypotheses were as follows:

Objective 1. Assess the impact of Simple Suppers on children of participating families relative to children of families in the control group.

Hypothesis 1.1 Diet quality, BMI z-score, waist circumference (WC) z-score, and

BP z-score will improve more from baseline to post-intervention among children

participating in the intervention than in the controls.

114 Hypothesis 1.2 Diet quality, BMI z-score, WC z-score, and BP z-score

improvements will be maintained during the follow-up period among children

participating in the intervention.

Study Design

The Simple Suppers study, which aimed to improve child diet quality and weight status by improving the quality of family meals among underserved families with children 4-10 year old, was implemented over 12-months as a two-group (intervention; waitlist control) quasi-experimental trial using a staggered cohort design [218]. At each of three time periods, separated by 10 weeks, a cohort of families was recruited. The original recruitment goal of approximately 20 families per cohort was increased to 30 families per cohort as a result of additional funding. Families self-selected their group assignment (intervention, waitlist control) by being offered the opportunity to participate in the program during the first (intervention group) or second 10-week session (waitlist control group).

Upon confirmation of study eligibility, a baseline data collection appointment was scheduled at the participating family’s home or the faith-based community center during the two weeks preceding intervention commencement. For participating families, data was collected on all children 4-10 years old via direct assessment and caregiver competed questionnaires at baseline (time point 0, T0), 10-week post-test (time point 1, T1), and 10- week follow-up (time point 2, T2). Written caregiver consent and child assent was

115 obtained at baseline. A team of trained research staff, blinded from group assignment, facilitated data collection. Participating families received a $25 grocery store gift card at each data collection point for their participation in the research. All study materials and procedures were approved by the Institutional Review Board at Ohio State University.

Setting

The Simple Suppers intervention was implemented at a faith-based community center, as community centers have been demonstrated to be efficacious sites for family- based childhood obesity prevention programing [73, 103, 104].

Participants

Staff and volunteers recruited families through faith-based community center events, center newsletter advertisements, and posters displayed in the center. To be eligible for inclusion, caregivers had to be the primary food preparer in the home; be responsible for at least one child 4-10 years of age; speak English as the primary language in the home; and have lived in the U.S. for at least one year. Families with one or more family members following a restrictive or therapeutic diet were excluded. For eligible families with multiple children between 4-10 years old, all eligible children were invited to participate.

Intervention

The Intervention Mapping protocol was utilized in the development of the Simple

116 Suppers intervention [186, 187, 218]. Based on the current evidence linking family meals with improved child diet and weight status [75, 77, 83, 84], program objectives for the

Simple Suppers intervention included: 1) ‘Increase frequency of family meals prepared in the home (>5 days/week)’ and 2) ‘Improve child diet quality (significantly increase

Healthy Eating Index (HEI) score (p<0.05); increase servings of fruits and vegetables to meet Dietary Guidelines recommendations; significantly decrease daily servings of sugar sweetened beverages (p<0.05).’ The Social Cognitive Theory, which posits that behavior change is a function of a reciprocal relationship between personal (e.g., behavioral capabilities, such as food preparation skills/frequency) and environmental (e.g., norms, modeling, and reinforcement) factors, served as the theoretical foundation for the Simple

Suppers intervention [163, 181].

The Simple Suppers program included 10, 90-minute lessons delivered weekly over the dinner hour. Each lesson focused on a family meals topic identified in the literature as a barrier to family meals for families with young children (e.g., timesaving strategies for family meals; connecting with your child through family meals). Session components included: a) interactive group discussions and goal setting with caregivers; b) hands-on food preparation activities with children; and c) group family meal with caregivers and children.

Outcome Measures

Diet Quality. Dietary intake was assessed by conducting three, caregiver-assisted,

117 nonconsecutive (two weekdays, one weekend day) 24-hour (24hr) dietary recalls using

USDA’s 5-step multi-pass dietary recall method [189, 197]. At each data collection time point, the first dietary recall was conducted during the in-person data collection visit, and the remaining two were conducted via telephone within two weeks of the initial in-person recall. Typical daily dietary intake was determined by averaging dietary intake across the three recalls at each time point to determine daily servings of fruit, vegetables, and SSB.

Diet quality was assessed at each point by calculating a Healthy Eating Index 2010 score using the three 24hr dietary recalls collected [198].

Anthropometric assessments. Standardized procedures were used to assess height and weight on all participating children [199, 200]. Body mass index was calculated using measured heights and weights. Centers for Disease Control and Prevention (CDC) age- and sex-adjusted BMI growth charts were used to determine BMI z-scores for children to adjust for expected healthy growth and weight gain [199, 201]. Waist circumference was measured on all participating children with a tape measure at the uppermost lateral border of the hip crest (ilium) [199]. To adjust for expected growth among child participants, child WC z-scores were determined using CDC age- and sex- specific growth charts [202].

Blood Pressure. Blood pressure was assessed on all participating children and via automated, calibrated BP monitors (Panasonic EW3109W). Age-, sex-, and height- adjusted National Heart, Lung, and Blood Institute (NHLBI) charts were used to

118 appropriately classify child BP [203].

Personal Determinants. Child food preparation skills and frequency of involvement were assessed at each data collection point by caregiver completion of an age-appropriate food preparation skills questionnaire designed to assess both skill ability and frequency of involvement in practicing the skill. Adapted from a previously validated questionnaire [73], three versions of the questionnaire were developed to align with the three program age groups, with each demonstrating acceptable or good internal consistency (4-5 yrs old: a= 0.79; 6-8 yr olds: a= 0.84; 9-10 yr olds: a= 0.87).

Process Measures

Feasibility (program dose and fidelity) and acceptability were assessed as process outcomes. To determine program feasibility, program dose was assessed by collecting weekly attendance. Program fidelity was determined by having a trained observer complete a program specific fidelity tool at the end of each weekly lesson, which included a checklist of key program components, activities, and leader characteristics.

Acceptability of the program was measured with a caregiver-completed satisfaction survey (yes/no) administered at the end of the 10-week program [218].

Data Analysis

Data from each of the three cohorts was pooled and the intervention tested by comparing change (T1-T0) in diet quality, anthropometric measures, blood pressure, and food preparation skills and frequency of children in the intervention compared to

119 participants in the waitlist control (hypothesis 1.1). Generalized linear mixed models were used to determine the association between the difference in the response variables of interest between the intervention and control group, controlling for potential confounders

(race, income, cohort, intervention dose), from baseline (T0) to 10-week post-test (T1).

Sustainability of intervention effects were tested by pooling intervention group data from each of the three cohorts, comparing change (T2-T1) in diet quality, anthropometric measures, blood pressure, and child food preparation skills and frequency among intervention group participants at the end of the 10-week follow-up period (hypothesis 1.2).

For families in which data were collected on multiple children, the effect of family was controlled by including a random effect for family. For dietary outcomes, while the protocol was to collect 3 dietary recalls at each time point, this was often not feasible due to challenges contacting participants and participants’ requesting not to complete the dietary recalls. In these instances, if two recalls were conducted, the average of the two recalls would be used and if a single 24hr dietary intake was collected, the single recall was used. Multiple imputations were used to deal with missing data with the exception of the dietary outcomes, due to great variability in dietary intake.

Results

Baseline Characteristics

One-hundred-nine families were recruited, resulting in 140 enrolled children at baseline. Family-level program retention was 87.5%, resulting in 126 children

120 completing the program. Descriptive summaries of sample baseline measures are presented in Table 11. Among the 126 child participants, about 40% were overweight or obese, average age was approximately 7 years old, about 60% were Black and approximately 42% were reliant on the WIC, SNAP, and/or NSLP federal food assistance programs. Baseline participant characteristics did not differ by level of attendance, with the exception of race, which was controlled for in the data analyses.

121 Table 11. Baseline Child Participant Characteristics Characteristics Total Intervention Control p-value (n=126) (n=87) (n=39) Age (years)a,b 6.9 (1.9) 6.8 (1.9) 7.2 (2.1) 0.239 Genderc,d (% female) 79 (62%) 54 (62%) 25 (64%) 0.905 Dietary Intakea,b,e,f (n=110)g (n=74)g (n=36)g Total Fruith 1.74 (1.5) 1.68 (1.2) 1.86 (1.8) 0.538 Whole Fruit 1.23 (1.1) 1.15 (0.9) 1.39 (1.5) 0.283 Total Vegetable 1.70 (1.4) 1.69 (1.4) 1.70 (1.6) 0.987 Total Sugar-Sweetened Beverage 0.62 (1.2) 0.55 (1.2) 0.75 (1.1) 0.414 Total Energy 1752.10 (679.1) 1698.09 (553.3) 1863.11 (882.5) 0.233 HEI Total Scorei 52.57 (10.8) 52.24 (10.5) 53.25 (11.4) 0.651 HEI Total Fruit Scorej 2.96 (2.0) 3.07 (1.9) 3.13 (2.1) 0.800 122 HEI Whole Fruit Scorek 3.09 (2.0) 2.99 (1.9) 2.89 (2.2) 0.896

HEI Vegetable Scorel 2.07 (1.4) 2.18 (1.4) 1.84 (1.5) 0.253 Anthropometrics/Blood Pressurea,b (n=126)g (n=87)g (n=39)g BMI z-score 0.69 (1.2) 0.66 (1.1) 0.75 (1.3) 0.693 WC z-score 0.70 (1.1) 0.72 (0.9) 0.63 (1.3) 0.671 Systolic BP z-score 1.46 (1.2) 1.46 (1.3) 1.47 (1.1) 0.954 Diastolic BP z-score 1.23 (1.2) 1.29 (1.3) 1.12 (1.3) 0.473 Personal Determinants (n=126)g (n=87)g (n=39)g Food Preparation Skillsm 22.4 (5.1) 22.03 (5.1) 23.21 (4.8) 0.230 Food Preparation Frequencyn 18.20 (5.6) 18.28 (5.8) 18.03 (5.1) 0.816 Table 11. (continued)

122 Table 11. Baseline Child Participant Characteristics (continued) Racec,d (n=123)g (n=84)g (n=39)g Black 73 (60%) 53 (63%) 20 (51%) 0.007 White 32 (25%) 15 (17%) 17 (41%) Othero 18 (15%) 16 (20%) 2 (8%) Household Income Statusc,d,p (n=119)g (n=82)g (n=37)g Low-Income 50 (42%) 32 (39%) 18 (49%) 0.423 Non-Low-Income 69 (58%) 50 (61%) 19 (51%) Home Food Securityc,d,q (n=124)g (n=85)g (n=39)g High/Marginal Food Security 79 (64%) 48 (56%) 31 (80%) 0.083 Low Food Security 23 (18%) 20 (24%) 3 (7%) Very Low Food Security 22 (18%) 17 (20%) 5 (13%) aValues are mean (SD) bOne-Way ANOVA cValues are n (%) dChi-square 123 eDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] f

Participant completion of 24hr dietary recalls: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)) gParticipant response rate is outcome variable-dependent, and therefore sample sizes are provided for each outcome variable hTotal Fruit= whole fruit + 100% fruit juice iHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] jHealthy Eating Index (HEI) total fruit score, 0-5 with higher scores representing better intake k Healthy Eating Index (HEI) whole fruit score, 0-5 with higher scores representing better intake l Healthy Eating Index (HEI) vegetable score, 0-5 with higher scores representing better intake mFood Preparation Skills, 0-32 scale with higher score representing increased skills nFood Preparation Frequency, 0-40 scale with higher score representing increased frequency oIncludes participants who did not identify with either Black or White pLow-income defined as participation in one or more of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) qUSDA 6-item Short Form Home Food Security Questionnaire (Scoring: High/Marginal food security= 0-1; Low food security= 2-4; Very low food security= 5- 6) [219]

123 Intervention Effects on Child Diet Quality

Daily total fruit intake did not differ by group assignment (Tables 12 and 13. β

(SE)= 0.71 (0.6), p=0.273) at post-test; however, a significant increase was observed by level of attendance, with the high attenders having a significantly higher daily total fruit intake compared to the low attenders (Tables 15 and 16. β (SE)= 2.26 (0.7), p=0.002).

Whole fruit intake significantly increased among children in the high attendance group relative to those in the low attendance (Table 16. β (SE)= 2.25 (0.7), p=0.002) and control (Table 16. β (SE)= 1.42 (0.7), p=0.039) groups at post-test. No significant within-group differences were observed at post-test (Table 17.) At 10-wk follow-up, daily total fruit intake did not significantly differ among intervention children (Tables 18 and 19. β (SE)= 0.17 (0.4), p=0.704), with no differences observed by level of attendance

(Table 18. β (SE)= -1.33 (1.4), p=0.344). Similarly, daily whole fruit intake did not significantly differ among intervention children at the 10-week follow-up; however, differences by level of attendance were observed, with a significantly greater decrease in daily whole fruit intake among children in the high attendance group relative to children in the low attendance group (Tables 16. β (SE)= -3.10 (0.9), p=0.003). Intervention effects were not observed at post-test or 10-week follow-up on daily total vegetable intake, daily SSB intake, daily energy intake, or HEI score.

Intervention Effects on Anthropometric and Blood Pressure Outcomes

Following the 10-week intervention, a significant decrease in child BMI z-score was observed among children in the high attendance group relative to children in the low

124 attendance (Table 16. β (SE)= -0.41 (0.1), p=0.003) and control (Tables 16. β (SE)= -0.26

(0.1), p=0.048) groups. During the follow-up period, BMI z-score was maintained among intervention children, with no significant differences existing by level of attendance (Table 18. β (SE)= -0.06 (0.1), p=0.624).

While a decrease in WC z-score was observed among intervention children (Table

13. β (SE)= -0.16 (0.1), p=0.166) at post-test, this decrease was not significant.

However, at the end of the 10-week intervention, a significant decrease in systolic BP z- score was observed by level of attendance, with high attenders having a significant decrease in systolic BP relative to controls (Tables 15. β (SE)= -0.60 (0.2), p=0.047).

During the follow-up period, systolic BP z-score was maintained among intervention children, with no differences by level of attendance.

Intervention Effects on Child Food Preparation Skills/Frequency

At post-test, child food preparation skills significantly increased (Tables 12. β

(SE)= 4.83 (0.9), p=<0.001) among intervention children relative to controls. Food preparation skills were maintained among intervention children at 10-week follow-up

(Tables 14. β (SE)= -0.15 (0.5), p=0.771), with no difference by level of attendance

(Tables 15. β (SE)= 1.18 (1.1), p= 0.292).

Similarly, child food preparation frequency significantly increased among intervention children (Table 18. β (SE)= 3.53 (1.2), p=0.003) at post-test. At 10-week

125 follow-up, child food preparation frequency was maintained among intervention children

(Table 18. β (SE)= 2.02 (1.1), p=0.059), with no significant differences by level of attendance (Table 19. β (SE)= 2.34 (2.2), p=0.287).

Process Outcomes

Intervention children (low (<7lessons) and high (>7 lessons) attenders) attended on average 71% of Simple Suppers lessons. Approximately 95% of lessons were delivered as intended and child participants were engaged in the program 96% of the time. At program completion (T1), 100% of the participating caregivers reported their child enjoyed participating in the program.

126 Table 12. Intervention Period: By-Group Child Outcomes at Baseline T0 and Post-Test T1 Outcomes Baseline T0 Post-Test T1 Mean (SD) Change from Baseline T0 to Post-Test T1 Control Intervention Control Intervention Control Intervention (n=39) (n=87) (n=39) (n=87) (n=39) (n=87) Dietary Intakea,b,c,d Total Fruit intake (svgs/d)e 1.86 (1.8) 1.68 (1.2) 1.48 (1.2) 1.71 (1.8) -0.38 (1.9) 0.04 (2.1) Whole fruit intake (svgs/d) 1.39 (1.5) 1.15 (0.9) 1.12 (1.2) 1.33 (1.7) -0.27 (1.7) 0.18 (1.8) Vegetable intake (svgs/d) 1.70 (1.6) 1.69 (1.4) 1.60 (1.7) 2.05 (1.6) -0.10 (2.8) 0.36 (1.4) Sugar sweetened beverage intake 0.75 (1.1) 0.55 (1.2) 4.23 (16.4) 0.54 (0.9) 3.48 (16.8) -0.01 (1.1) (svgs/d) Energy intake (kcals/d) 1863.11 (882.5) 1698.09 (553.3) 1729.63 (719.9) 1728.68 (609.7) -133.48 (1013.5) 30.59 (744.2) HEI total scoref 53.25 (11.4) 52.24 (10.5) 55.81 (9.4) 53.21 (11.6) 2.56 (14.6) 0.97 (16.1) HEI total fruit scoreg 3.13 (2.1) 3.07 (1.9) 2.73 (2.6) 2.72 (3.1) -0.40 (2.0) -0.35 (1.9) HEI whole fruit scoreh 2.89 (2.2) 2.99 (1.9) 2.40 (1.9) 2.41 (2.0) -0.49 (1.9) -0.58 (2.0) HEI total vegetable scorei 1.84 (1.5) 2.18 (1.4) 1.80 (1.6) 2.38 (1.6) -0.04 (1.6) 0.20 (1.4) 127 Anthropometric/BPa,j

BMI z-score 0.75 (1.3) 0.66 (1.1) 0.69 (1.4) 0.53 (1.2) -0.06 (0.3) -0.13 (0.6) WC z-score 0.63 (1.3) 0.72 (0.9) 0.71 (1.2) 0.79 (0.9) 0.08 (0.9) 0.07 (0.6) Systolic BP z-score 1.47 (1.1) 1.46 (1.3) 1.70 (1.6) 1.27 (1.4) 0.23 (1.9) -0.19 (1.6) Diastolic BP z-score 1.12 (1.3) 1.29 (1.3) 1.39 (1.8) 1.09 (1.3) 0.27 (2.1) -0.20 (1.9) Behavioral Capability/Cognitivea,b Food preparation skillsk 23.21 (4.8) 22.03 (5.1) 22.85 (5.5) 26.89 (4.6) -0.36 (4.2)m 4.86 (5.1)n Food preparation frequencyl 18.03 (5.1) 18.28 (5.8) 19.16 (7.5) 22.90 (6.3) 1.13 (6.4)m 4.62 (5.7)n aValues are Mean (SD) bBy-group differences determined by generalized linear mixed modeling controlling for: cohort; household income; child race, sex, age; group assignment; participant id(family id); baseline value (Y (outcome variable)= outcome at T1) cDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dParticipant completion of 24hr dietary recalls: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)) eTotal Fruit= whole fruit + 100% fruit juice fHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] Table 12. (continued)

127

Table 12. Intervention Period: By-Group Child Outcomes at Baseline T0 and Post-Test T1 (continued) gHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake hHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake jBy-group differences determined by generalized linear mixed modeling controlling for: cohort; household income; child race; baseline value; group assignment; participant id(family id) (Y (outcome variable)= outcome at T1) kFood Preparation Skills, 0-32 scale with higher score representing increased skills lFood Preparation Frequency, 0-40 scale with higher score representing increased frequency m,nDifferent superscripts indicate by-group difference in participant outcome of p<0.01

128

Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 Intervention Period (T0 à T1) Outcomes Covariates β (SE) p-value Confidence Interval Child Dietary Intakea,b,c Fruit intake (svgs/d) Group Assignmentd -Intervention 0.71 (0.6) 0.273 (-0.58, 1.99) -Control 0 - - Cohorte -One 0.53 (0.8) 0.487 (-0.99, 2.04) -Two 0.42 (0.7) 0.598 (-1.17, 2.01) -Three 0 - - Incomef -Low -0.05 (0.6) 0.930 (-1.25, 1.14) -Non-Low 0 - - Raceg 129 -Other -1.18 (0.9) 0.234 (-3.15, 0.79)

-Black -0.98 (0.7) 0.181 (-2.44, 0.47) -White 0 - - Sexh -Male (-0.68 (0.5) 0.219 (-1.77, 0.42) -Female 0 - - Child Age 0.07 (0.1) 0.513 (-0.14, 0.29) Fruit intake, T0 0.16 (0.2) 0.357 (-0.19, 0.51) Whole Fruit intake (svgs/d) Group Assignmentd -Intervention 0.90 (0.6) 0.150 (-0.34, 2.13) -Control 0 - - Cohorte -One 0.70 (0.7) 0.329 (-0.73, 2.14) -Two 0.30 (0.8) 0.692 (-1.21, 1.81) -Three 0 - - Table 13. (continued)

129 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Incomef -Low -0.18 (0.6) 0.753 (-1.32, 0.96) -Non-Low 0 - - Raceg -Other -1.39 (0.9) 0.145 (-3.27, 0.50) -Black -1.16 (0.7) 0.097 (-2.55, 0.22) -White 0 - - Sexh -Male -0.59 (0.5) 0.283 (-1.70, 0.51) -Female 0 - - Child Age 0.01 (0.1) 0.993 (-0.21, 0.21) Vegetable intake, T0 0.21 (0.2)) 0.373 (-0.26, 0.68) Vegetable intake (svgs/d) Group Assignmentd -Intervention 0.61 (0.6) 0.310 (-0.59, 1.81) -Control 0 - - 130 Cohorte

-One -0.82 (0.7) 0.245 (-2.23, 0.58) -Two -0.37 (0.7) 0.614 (-1.82, 1.09) -Three 0 - - Incomef -Low 0.37 (0.5) 0.512 (-0.75, 1.48) -Non-Low 0 - - Raceg -Other -0.84 (0.9) .363 (-2.68, 1.00) -Black -0.54 (0.6) 0.419 (-1.88, 0.80) -White 0 - - Sexh -Male 0.50 (0.5) 0.352 (-0.58, 1.58) -Female 0 - - Child Age 0.03 (0.1) 0.782 (-0.19, 0.26) Vegetable intake, T0 0.28 (0.2) 0.113 (-0.07, 0.64) Table 13. (continued)

130

Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Sugar sweetened beverage intake Group Assignmentd (svgs/d) -Intervention -0.29 (0.3) 0.365 (-0.94, 0.35) -Control 0 - - Cohorte -One 0.06 (0.3) 0.877 (-0.72, 0.84) -Two 0.26 (0.4) 0.526 (-0.57, 1.11) -Three 0 - - Incomef -Low 0.05 (0.3) 0.865 (-0.54, 0.64) -Non-Low 0 - - Raceg -Other 1.05 (0.5) 0.034 (0.08, 2.02) -Black 0.53 (0.4) 0.145 (-0.19, 1.26) -White 0 - - Sexh 131 -Male 0.19 (0.3) 0.513 (-0.39, 0.78)

-Female 0 - - Child Age 0.02 (0.1) 0.740 (-0.09, 0.13) SSB intake, T0 0.15 (0.2) 0.382 (-0.19, 0.48) Energy intake (kcals/d) Group Assignmentd -Intervention -51.56 (217.7) 0.814 (-490.53, 387.41) -Control 0 - - Cohorte -One 302.25 (262.3) 0.256 (-226.78, 831.28) -Two 287.12 (261.5) 0.278 (-240.26, 814.50) -Three 0 - - Incomef -Low 185.9 (218.4) 0.399 (-243.48, 626.43) -Non-Low 0 - - Table 13. (continued)

131

Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Raceg -Other 186.0 (335.9) 0.583 (-491.36, 863.39) -Black 387.38 (246.7) 0.124 (-110.09, 884.85) -White 0 - - Sexh -Male 96.74 (193.5) 0.620 (-293.53, 487.00) -Female 0 - - Child Age 77.7 (42.9) 0.077 (-8.81, 164.27) Energy intake, T0 -0.17 (0.2) 0.302 (-0.49, 0.16) HEI scorei Group Assignmentd -Intervention 0.15 (4.4) 0.973 (-8.98, 8.68) -Control 0 - - Cohorte -One -4.72 (4.9) 0.344 (-14.68, 5.24) 132 -Two -7.93 (5.2) 0.138 (-18.52, 2.65)

-Three 0 - - Incomef -Low 3.53 (4.0) 0.385 (-4.58, 11.65) -Non-Low 0 - - Raceg -Other -5.33 (6.4) 0.411 (-18.29, 7.62) -Black -3.56 (4.6) 0.467 (-13.34, 6.22) -White 0 - - Sexh -Male -4.15 (4.0) 0.307 (-12.23, 3.94) -Female 0 - - Child Age 0.71 (0.7) 0.349 (-0.80, 2.22) HEI Score, T0 -0.08 (0.2) 0.683 (-0.46, 0.31) Table 13. (continued)

132 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Child Anthropometrics/Blood Pressurej BMI z-score Group Assignmentd -Intervention -0.09 (0.1) 0.465 (-0.32, 0.15) -Control 0 - - Cohorte -One -0.01 (0.1) 0.935 (-0.28, 0.15) -Two -0.12 (0.1) 0.371 (-0.39, 0.15) -Three 0 - - Incomef -Low 0.10 (0.1) 0.370 (-0.12, 0.32) -Non-Low 0 - - Raceg -Other -0.20 (0.2) 0.273 (-0.57, 0.16) -Black 0.04 (0.1) 0.748 (-0.22, -.31)

133 -White 0 - - BMI z-score, T0 0.94 (0.1) <0.001 (0.85, 1.0)

Waist Circumference z-scoreh Group Assignmentd -Intervention -0.16 (0.1) 0.166 (-0.39, 0.07) -Control 0 - - Cohorte -One -0.20 (0.1) 0.120 (-0.45, 0.05) -Two 0.04 (0.1) 0.748 (-0.22, 0.30) -Three 0 - - Incomef -Low -0.07 (0.1) 0.514 (-0.29, 0.14) -Non-Low 0 - - Raceg -Other 0.41 (0.2) 0.020 (0.07, 0.75) -Black 0.24 (0.1) 0.058 (-0.01, 0.49) -White 0 - - WC z-score, T0 0.79 (0.1) <0.001 (0.69, 0.89) Table 13. (continued)

133 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Systolic BP z-score Group Assignmentd -Intervention -0.38 (0.3) 0.157 (-0.92, 0.15) -Control 0 - - Cohorte -One -0.01 (0.3) 0.994 (-0.60, 0.59) -Two -0.15 (0.3) 0.635 -0.76, 0.47) -Three 0 - - Incomef -Low 0.87 (0.3) 0.001 (0.37, 1.37) -Non-Low 0 - - Raceg -Other -0.56 (0.4) 0.181 (-1.28, 0.26) -Black 0.46 (0.3) 0.124 (-0.13, 1.0) -White 0 - - SysBP z-score, T0 0.302 (0.1) 0.002 (0.12, 0.49) 134 Diastolic BP z-score Group Assignmentd

-Intervention -0.11 (0.4) 0.782 (-0.89, 0.67) -Control 0 - - Cohorte -One 0.22 (0.4) 0.608 (-0.62, 1.06) -Two 0.41 (0.4) 0.355 (-0.46, 1.27) -Three 0 - - Incomef -Low 1.29 (0.3) 0.001 (0.57, 2.01) -Non-Low 0 - - Raceg -Other -1.20 (0.6) 0.056 (-2.42, 0.03) -Black 0.09 (0.4) 0.831 (-0.74, 0.92) -White 0 - - Table 13. (continued)

134 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) DysBP z-score, T0 0.12 (0.1) 0.340 (-0.13, 0.38) Child Personal Determinantsa Food Preparation Skillsk Group Assignmentd -Intervention 4.83 (0.9) <0.001 (3.10, 6.55) -Control 0 0 - Cohorte -One 1.44 (0.9) 0.134 (-0.45, 3.33) -Two 1.93 (1.0) 0.052 (-0.02, 3.89) -Three 0 - - Incomef -Low -2.21 (0.8) 0.008 (-3.83, -0.59) -Non-Low 0 - - Raceg -Other -2.88 (1.3) 0.026 (-5.42, -0.34) -Black -2.98 (1.0) 0.002 (-4.87, -1.08) 135 -White 0 - - h Sex -Male -0.99 (0.8) 0.209 (-2.55, 0.57) -Female 0 0 0 Child Age -0.02 (0.2) 0.932 (-0.38, 0.35) Food prep skills, T0 0.42 (0.1) <0.001 (0.27, 0.57) Food Preparation Frequencyl Group Assignmentd -Intervention 3.53 (1.2) 0.003 (1.22, 5.83) -Control 0 - - Cohorte -One 3.66 (1.3) 0.006 (1.09, 6.23) -Two -1.38 (1.3) 0.301 (-4.00, 1.25) -Three 0 - - Table 13. (continued)

135 Table 13. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Incomef -Low 1.52 (1.1) 0.168 (-0.65, 3.69) -Non-Low 0 - - Raceg -Other 0.05 (1.8) 0.977 (-3.44, 3.54) -Black -2.82 (1.3) 0.030 (-5.35, -0.28) -White 0 - - Sexh -Male -1.74 (1.1) 0.114 (-3.90, 0.43) -Female 0 - - Child Age -0.38 (0.3) 0.145 (-0.89, 0.13) Food prep freq, T0 0.55 (0.1) <0.001 (0.37, 0.73) aBy-group differences determined by generalized linear mixed model controlling for: cohort; household income; sex; race; age; baseline value; participant id(family id); group assignment (Y (outcome variable)= outcome at T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step 136 multi-pass method [189, 197] c Participant completion of 24hr dietary recalls at T0: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)). Participant completion of 24hr dietary recalls at T1: 0 recalls (n=47 (39%)); 1 recall (n=68 (58%)); 2 recalls (n=4 (4%)); 3 recalls (n=0 (0%)) dGroup assignment: control group (n=39); intervention group (n=87) eCohort: one (n= 35); two (n=29); three (n=62) fIncome: low (n= 50); non-low (n=69). Low-income defined as participation in >1 of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) gRace: other (n= 18); Black (n= 73); White (n=32). Other category includes participants who did not identify with Black or White hSex: male (n=47); female (n= 79) iHealthy Eating Index, 0-100 scale, Higher score represents better diet quality [198] jBetween group differences determined by generalized linear mixed model controlling for: cohort; household income; race; baseline value; participant id(family id); group assignment (Y (outcome variable)= outcome at T1) kFood Preparation Skills, 0-32 scale with higher score representing increased skills lFood Preparation Frequency, 0-40 scale with higher score representing increased frequency

136 Table 14. Intervention Period: Within-Group Child Outcomes at Baseline T0 and Post-Test T1 Outcomes Control (n=39) Intervention (n=87) Baseline T0 Post-Test T1 Baseline T0 Post-Test T1 Dietary Intakea,b,c,d Fruit intake (svgs/d) 1.86 (1.8) 1.48 (1.2) 1.68 (1.2) 1.71 (1.8) Whole fruit intake (svgs/d) 1.39 (1.5) 1.12 (1.2) 1.15 (0.9) 1.33 (1.7) Vegetable intake (svgs/d) 1.70 (1.6) 1.60 (1.7) 1.69 (1.4) 2.05 (1.6) Sugar sweetened beverage intake (svgs/d) 0.75 (1.1) 4.23 (16.4) 0.55 (1.2) 0.54 (0.9) Energy intake (kcals/d) 1863.11 (882.5) 1729.63 (719.9) 1698.09 (553.3) 1728.68 (609.7) HEI total scoree 53.25 (11.4) 55.81 (9.4) 52.24 (10.5) 53.21 (11.6) HEI total fruit scoref 3.13 (2.1) 2.73 (2.6) 3.07 (1.9) 2.72 (3.1) HEI whole fruit scoreg 2.89 (2.2) 2.40 (1.9) 2.99 (1.9) 2.41 (2.0) 137 HEI total vegetable scoreh 1.84 (1.5) 1.80 (1.6) 2.18 (1.4) 2.38 (1.6) a,b Anthropometric/BP BMI z-score 0.75 (1.3) 0.69 (1.4) 0.66 (1.1)k 0.53 (1.2)l WC z-score 0.63 (1.3) 0.71 (1.2) 0.72 (0.9) 0.79 (0.9) Systolic BP z-score 1.47 (1.1) 1.70 (1.6) 1.46 (1.3) 1.27 (1.4) Diastolic BP z-score 1.12 (1.3) 1.39 (1.8) 1.29 (1.3) 1.09 (1.3) Personal Determinants,b Food preparation skillsi 23.21 (4.8) 22.85 (5.5) 22.03 (5.1)m 26.89 (4.6)n Food preparation frequencyj 18.03 (5.1) 19.16 (7.5) 18.28 (5.8)m 22.90 (6.3)n aValues are Mean (SD) bWithin-group differences determined by generalized linear mixed modeling controlling for: participant id(family id) (Y (outcome variable) = (outcome at T1 – outcome at T0)) cDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dParticipant completion of 24hr dietary recalls: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)) eHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] fHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake gHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake Table 14. (continued)

137 Table 14. Intervention Period: Within-Group Child Outcomes at Baseline T0 and Post-Test T1 (continued) hHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake iFood Preparation Skills, 0-32 scale with higher score representing increased skills jFood Preparation Frequency, 0-40 scale with higher score representing increased frequency k,lDifferent superscripts indicate within-group difference in participant outcome of p<0.05 m,nDifferent superscripts indicate within-group difference in participant outcome of p<0.001

138

Table 15. Intervention Period: By-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1 Outcomes Baseline T0 Post-Test T1 Mean (SD) Change from Baseline T0 to Post-Test T1 Control Low Att.a High Att.a Control Low Att.a High Att.a Control Low Att.a High Att.a (n=39) (n=33) (n=54) (n=39) (n=33) (n=54) (n=39) (n=33) (n=54) Dietary Intake,b,c,d,e Total Fruit intake 1.86 (1.8) 1.76 (1.2) 1.62 (1.3) 1.48 (1.2) 1.18 (1.0) 1.79 (1.9) -0.38 (1.9) -0.58 (1.6)p 0.17 (2.1)q (svgs/d)f Whole fruit intake 1.39 (1.5) 1.07 (0.9) 1.20 (0.9) 1.12 (1.2) 1.03 (1.2) 1.38 (1.9) -0.27 (1.7)n -0.04 (1.3)p 0.18 (1.9)o,q (svgs/d) Vegetable intake 1.70 (1.6) 1.67 (1.4) 1.72 (1.4) 1.60 (1.7) 2.24 (1.7) 2.02 (1.7) -0.10 (2.8) 0.57 (1.2) 0.30 (1.6) 139 (svgs/d)

SSB intake (svgs/d) 0.75 (1.1) 0.70 (1.7) 0.45 (0.8) 4.23 (16.4) 0.75 (1.4) 0.51 (0.9) 3.48 (16.8) 0.05 (1.7) 0.06 (1.0) Energy intake (kcals/d) 1863.1 1749.04 1663.36 1729.6 1844.66 1709.36 -133.4 95.62 46.36 (882.5) (593.4) (528.4) (719.9) (663.5) (608.2) (1013.5) (586.1) (773.9) HEI total scoreg 53.25 56.39 (15.4) 50.68 (10.1) 55.81 (9.4) 56.88 (15.4) 52.70 2.56 (14.6) 0.49 (16.1) 2.02 (16.2) (11.4) (11.1) HEI total fruit scoreh 3.13 (2.1) 3.03 (2.1) 2.97 (1.9) 2.73 (2.6) 2.22 (2.1) 2.44 (2.0) -0.40 (2.0) -0.81 (1.9) -0.53 (1.9) HEI whole fruit scorei 2.89 (2.2) 3.04 (1.9) 3.10 (1.9) 2.40 (1.9) 2.50 (2.7) 2.76 (3.1) -0.49 (1.9) -0.54 (2.4) -0.34 (2.2) HEI total vegetable 1.84 (1.5) 2.15 (1.4) 2.21 (1.5) 1.80 (1.6) 2.32 (1.6) 2.38 (1.7) -0.04 (1.6) 0.17 (1.4) 0.17 (1.6) scorej Anthropometric/BPb,k BMI z-score 0.75 (1.3) 0.54 (1.1) 0.76 (1.1) 0.69 (1.4) 0.60 (1.1) 0.48 (1.3) -0.06 (0.3)n 0.06 (0.5)p -0.28 (0.7)o,q WC z-score 0.63 (1.3) 0.87 (0.8) 0.70 (0.9) 0.71 (1.2) 0.75 (0.8) 0.82 (0.9) 0.08 (0.9) -0.12 (0.6) 0.12 (0.7) Systolic BP z-score 1.47 (1.1) 1.18 (1.5) 1.55 (1.3) 1.70 (1.6) 1.40 (1.4) 1.19 (1.3) 0.23 (1.9)n 0.22 (1.6) -0.36 (1.6)o Diastolic BP z-score 1.12 (1.3) 1.15 (1.5) 1.41 (1.4) 1.39 (1.8) 1.06 (1.1) 1.11 (1.4) 0.27 (2.1) -0.09 (1.7) -0.30 (1.9) Pers. Determinantsb,c Food prep skillsl 23.21 (4.8) 21.94 (4.6) 22.09 (5.5) 22.85 (5.5) 28.00 (5.3) 23.2 (4.1) -0.36 (4.2)p 6.06 (4.7)q 1.14 (5.2)q Food prep frequencym 18.03 (5.1) 17.25 (5.4) 18.87 (5.9) 19.16 (7.5) 23.47 (6.1) 22.5 (6.5) 1.13 (6.4)n,p 6.22 (5.4)o 3.69 (5.7)q Table 15. (continued)

139 Table 15. Intervention Period: By-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1 (continued) aLevel of attendance based on total number of lessons attended bValues are Mean (SD) cBy-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, age; group assignment; baseline value; group assignment (Y (outcome variable)= outcome at T1) dDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] eParticipant completion of 24hr dietary recalls at T0: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)). Participant completion of 24hr dietary recalls at T1: 0 recalls (n=47 (39%)); 1 recall (n=68 (58%)); 2 recalls (n=4 (4%)); 3 recalls (n=0 (0%)) fTotal Fruit= whole fruit + 100% fruit juice gHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] hHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake jHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake k

140 Between group differences determined by generalized linear mixed model controlling for: cohort; household income; race; baseline value; participant id(family id); group assignment (Y (outcome variable)= outcome at T1) l Food Preparation Skills, 0-32 scale with higher score representing increased skills mFood Preparation Frequency, 0-40 scale with higher score representing increased frequency n,oDifferent superscripts indicate by-group difference in participant outcome of p<0.05 p,qDifferent superscripts indicate by-group difference in participant outcome of p<0.01

140 Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 Intervention Period (T0 à T1) Outcomes Covariates β (SE) p-value Confidence Interval β (SE) p-value Confidence Interval Child Dietary Intakea,b,c Fruit intake Attendanced (svgs/d) -High 1.27 (0.7) 0.068 (-0.10, 2.63) 0 - - -Low -0.99 (0.8) 0.246 (-2.70, 0.71) -2.26 (0.7) 0.002 (-3.67, -0.85) -Control 0 - - -1.27 (0.7) 0.068 (-2.63, 0.10) Cohorte -One 1.0 (0.8) 0.226 (-0.64, 2.63) -Two 0.71 (0.8) 0.407 (-0.99, 2.41) -Three 0 - - Incomef -Low 0.09 (0.6) 0.888 (-1.15, 1.33) -Non-Low 0 - - Raceg 141 -Other -1.42 (1.0) 0.165 (-3.46, 0.61)

-Black -1.08 (0.8) 0.156 (-2.60, 0.43) -White 0 - - Sexh -Male -0.73 (0.5) 0.133 (-1.69, 0.23) -Female 0 - - Child Age 0.10 (0.1) 0.234 (-0.07, 0.26) Fruit intake, T0 0.12 (0.2) 0.463 (-0.20, 0.44) Whole fruit intake Attendanced (svgs/d) -High 1.42 (0.7) 0.039 (0.07, 2.76) 0 - - -Low -0.83 (0.8) 0.323 (-2.52, 0.85) -2.25 (0.7) 0.002 (-3.62, -0.88) -Control 0 - - -1.42 (0.7) 0.039 (-2.76, -0.07) Cohorte -One 1.11 (0.8) 0.170 (-0.49, 2.70) -Two 0.52 (0.8) 0.534 (-1.15, 2.19) -Three 0 - - Table 16. (continued)

141 Table 16. Intervention Period: By-Group Differences (Level of Attendance) in Child Outcomes at Post-Test, T1 (continued) Incomef -Low -0.06 (0.6) 0.924 (-1.27, 1.15) -Non-Low 0 - - Raceg -Other -1.61 (1.0) 0.109 (-3.61, 0.38) -Black -1.22 (0.7) 0.101 (-2.70, 0.25) -White 0 - - Sexh -Male -0.67 (0.5) 0.158 (-1.61, 0.27) -Female 0 - 0 Child Age 0.04 (0.1) 0.580 (-0.11, 0.20) Whole fruit intake, 0.10 (0.2) 0.630 (-0.32, 0.52) T0 Vegetable intake Attendanced (svgs/d)a -High 0.54 (0.6 0.391 (-0.72, 1.81) 0 - - 142 -Low 0.83 (0.9) 0.336 (-0.89, 2.55) 0.29 (0.8) 0.719 (-1.31, 1.88)

-Control 0 - - -0.54 (0.6) 0.391 (-1.81, 0.72) Cohorte -One -0.88 0.7) 0.229 (-2.34, 0.58) -Two -0.40 (0.7) 0.585 (-1.89, 1.08) -Three 0 - - Incomef -Low 0.35 (0.6) 0.535 (-0.78, 1.49) -Non-Low 0 - - Raceg -Other -0.82 (0.9) 0.381 (-2.69, 1.05) -Black -0.53 (0.7) 0.438 (-1.88, 0.83) -White 0 - - Table 16. (continued)

142

Table 16. Intervention Period: By-Group Differences (Level of Attendance) in Child Outcomes at Post-Test, T1 (continued) Sexh -Male 0.51 (0.7) 0.351 (-0.58, 1.60) -Female 0 - - Child Age 0.03 (0.1) 0.772 (-0.20, 0.27) Vegetable intake, 0.28 (0.2) 0.125 (-0.08, 0.64) T0 Sugar sweetened Attendanced beverage intake -High -0.31 (0.3) 0.361 (-0.99, 0.37) 0 - - (svgs/d) -Low -0.23 (0.4) 0.602 (-1.13, 0.66) 0.08 (0.4) 0.844 (-0.73, 0.89) -Control 0 - - 0.31 (0.3) 0.361 (-0.37, 0.99) Cohorte -One 0.04 (0.4) 0.915 -0.77, 0.85) -Two 0.26 (0.4) 0.547 (-0.60, 1.11)

143 -Three 0 - - Incomef -Low 0.04 (0.3) 0.881 (-0.55, 0.64) -Non-Low 0 - - Raceg -Other 1.06 (0.5) 0.035 (0.08, 2.04) -Black 0.54 (0.4) 0.147 (-0.20, 1.27) -White 0 - - Sexh -Male 0.20 (0.3) 0.511 (-0.40, 0.79) -Female 0 - - Child Age 0.02 (0.1) 0.733 (-0.09, 0.13) SSB intake, T0 0.14 (0.2) 0.396 (-0.19, 0.48) Table 16. (continued)

143

Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 (continued) Energy intake Attendanced (kcals/d) -High -74.07 (230.0) 0.749 (-538.24, 390.09) 0 - - -Low 21.87 (313.1) 0.945 (-610.1, 653.8) 95.94 (290.7) 0.743 (-490.81, 682.69) -Control 0 - - 74.07 (230.0) 0.749 (-390.09, 538.24) Cohorte -One 279.8 (273.1) 0.311 (-271.26, 830.87) -Two 272.54 (267.5) 0.314 (-267.31, 812.39) -Three 0 - - Incomef -Low 180.52 (221.3) 0.419 (-266.04, 627.09) -Non-Low 0 - - Raceg

144 -Other 195.45 (340.5) 0.569 (-491.68, 882.58) -Black 394.51 (250.1) 0.122 (-110.25, 899.26) -White 0 - - Sexh -Male 101.29 (196.1) 0.608 (-294.35, 496.94) -Female 0 - - Child Age 78.28 (43.4) 0.079 (-9.39 165.92) Energy intake, T0 -0.17 (.2) 0.308 (-0.49, 0.16) HEI scorei Attendanced -High -0.46 (4.6) 0.920 (-9.72, 8.79) 0 - - -Low 0.99 (6.1) 0.886 (-11.43, 13.19) 1.34 (5.5) 0.807 (-9.66, 12.35) -Control 0 - - 0.46 (4.6) 0.920 (-8.79, 9.72) Cohorte -One -5.00 (5.1) 0.335 (-15.35, 5.35) -Two -8.08 (5.3) 0.138 (-18.86, 2.70) -Three 0 - - Table 16. (continued)

144

Table 16. Intervention Period: By-Group Differences in Child Outcomes at Post-Test, T1 (continued) Incomef -Low 3,43 (4.1) 0.404 (-4.79, 11.65) -Non-Low 0 - - Raceg -Other -5.20 (6.5) 0.428 (-18.31, 7.91) -Black -3.50 (4.9) 0.479 (-13.39, 6.39) -White 0 - - Sexh -Male -4.07 (4.1) 0.322 (-12.27, 4.13) -Female 0 - - Child Age 0.71 (0.8) 0.359 (-0.83, 2.48) HEI score, T0 -0.08 (0.2) 0.688 (-0.47, 0.31)

145 Child Anthropometrics/Blood Pressurej BMI z-score Attendanced

-High -0.26 (0.1) 0.048 (-0.52, -0.01) 0 - - -Low 0.15 (0.1) 0.310 (-0.14, 0.43) 0.41 (0.1) 0.003 (0.14, 0.68) -Control 0 - - 0.26 (0.1) 0.048 (0.01, 0.52) Cohorte -One -0.02 (0.1) 0.873 (-0.28, 0.24) -Two -0.11 (0.1) 0.428 (-0.38, 0.16) -Three 0 - - Incomef -Low 0.05 (0.1) 0.640 (-0.17, 0.27) -Non-Low 0 - - Raceg -Other -0.13 (0.1) 0.481 (-0.49, 0.23) -Black 0.14 (0.1) 0.323 (-0.14, 0.41) -White 0 - - BMI z-score, T0 0.94 (0.0) <0.001 (0.85, 1.0) Table 16. (continued)

145 Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 (continued) Waist Attendanced Circumference z- -High -0.10 (0.1) 0.447 (-0.35, 0.15) 0 - - score -Low -0.23 (0.1) 0.095 (-0.51, 0.04) -0.14 (0.1) 0.291 (-0.40, 0.12) -Control 0 - - 0.10 (0.1) 0.447 (-0.15, 0.35) Cohorte -One -0.20 (0.1) 0.130 (-0.45, 0.06) -Two 0.05 (0.1) 0.722 (-0.21, 0.31) -Three 0 - - Incomef -Low -0.07 (0.1) 0.553 (-0.28, 0.15) -Non-Low 0 - - Raceg -Other 0.40 (0.2) 0.024 (0.05, 0.75) -Black 0.22 (0.1) 0.90 (-0.04, 0.48) -White 0 - - 146 WC z-score, T0 0.79 (0.1) <0.001 (0.69, 0.90) d Systolic BP z-score Attendance -High -0.60 (0.2) 0.047 (-1.18, -0.01) 0 - - -Low -0.13 (0.3) 0.693 (-0.78, 0.52) 0.47 (0.3) 0.132 (-0.14, 1.08) -Control 0 - - 0.60 (0.2) 0.047 (0.01, 1.18) Cohorte -One -0.03 (0.2) 0.923 (-0.62, 0.56) -Two -0.16 (0.3) 0.600 (-0.78, 0.45) -Three 0 - - Incomef -Low 0.84 (0.2) 0.001 (0.33, 1.34) -Non-Low 0 - - Table 16. (continued)

146 Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 (continued) Raceg -Other -0.51 (0.4) 0.224 (-1.34, 0.32) -Black 0.54 (0.3) 0.083 (-0.07, 1.14) -White 0 - - SysBP z-score, T0 0.31 (0.1) 0.001 (0.12, 0.50) Diastolic BP z- Attendanced score -High -0.12 (0.4) 0.776 (-0.95, 0.71) 0 - - -Low -0.08 (0.5) 0.888 (-1.13, 0.98) 0.04 (0.5) 0.927 (-0.91, 0.99) -Control 0 - - 0.12 (0.4) 0.776 (-0.71, 0.95) Cohorte -One 0.21 (0.4) 0.621 (-0.64, 1.07) -Two 0.40 (0.4) 0.367 (-0.48, 1.28) -Three 0 - - Incomef -Low 1.29 (0.3) 0.001 (0.56, 2.01) 147 -Non-Low 0 - - g Race -Other -1.19 (0.6) 0.062 (-2.44, 0.06) -Black 0.10 (0.4) 0.810 (-0.76, 0.97) -White 0 - - DysBP z-score, T0 0.12 (0.1) 0.345 (-0.14, 0.38) Child Personal Determinantsa Food Preparation Attendanced Skillsk -High 4.04 (0.9) <0.001 (-4.01, -0.62) 0 - - -Low 5.80 (1.1) <0.001 (-3.96, -0.66) 1.76 (1.0) 0.073 (-0.16, 3.67) -Control 0 - - -4.04 (0.9) <0.001 (-5.93, -2.15) Cohorte -One 1.31 (0.9) 0.173 (-5.26, -0.16) -Two 1.89 (1.0) 0.060 (-4.64, -0.76) -Three 0 - - Table 16. (continued)

147 Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 (continued) Incomef -Low -2.31 (0.8) 0.006 (2.15, 5.93) -Non-Low 0 - (3.71, 7.88) Raceg - -Other -2.71 (1.3) 0.038 -Black -2.70 (1.0) 0.007 (-0.58, 3.21) -White 0 - (-0.08, 3.85) Sexh - -Male -1.18 (0.8) 0.136 (-2.75, 0.38) -Female 0 - - Child Age -0.03 (0.2) 0.868 (-0.40, 0.34) Food prep skills, T0 0.42 (0.1) <0.001 (0.27, 0.57) Food Preparation Attendanced Frequencyl -High 3.16 (1.3) 0.016 (0.61, 5.71) 0 - -

148 -Low 4.09 (1.4) 0.005 (1.28, 6.91) 0.94 (1.3) 0.480 (-1.68, 3.55) -Control 0 - - -3.16 (1.3) 0.016 (-5.71, -0.61) Cohorte -One 3.62 (1.3) 0.007 (1.02, 6.22) -Two -1.36 (1.3) 0.313 (-4.01, 1.30) -Three 0 - - Incomef -Low 1.40 (1.1) 0.215 (-0.83, 3.64) -Non-Low 0 - - Raceg -Other 0.17 (1.8) 0.924 (-3.37, 3.71) -Black -2.63 (1.3) 0.050 (-5.25, 0.00) -White 0 - - Table 16. (continued)

148 Table 16. Intervention Period: By-Group (Level of Attendance) Differences in Child Outcomes at Post-Test, T1 (continued) Sexh -Male 01.79 (1.1) 0.110 (-3.99, 0.41) -Female 0 - - Child Age -0.37 (0.3) 0.157 (-0.89, 0.15) Food prep freq, T0 0.55 (0.1) <0.001 (0.37, 0.74) aBy-group differences determined by generalized linear mixed model controlling for: cohort; household income; sex; race; age; baseline value; participant id(family id); level of attendance (Y (outcome variable)= outcome at T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] cParticipant (n=126) completion of 24hr dietary recalls at T0: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)). Participant (n=126) completion of 24hr dietary recalls at T1: 0 recalls (n=47 (39%)); 1 recall (n=68 (58%)); 2 recalls (n=4 (4%)); 3 recalls (n=0 (0%)) dAttendance (level of attendance based on total number of lessons attended): control group (n=39); low (n=33); high (n=54) eCohort: one (n= 35); two (n=29); three (n=62) fIncome: low (n= 50); non-low (n=69). Low-income defined as participation in >1 of the following federal food assistance programs: Special

149 Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) gRace: other (n= 18); Black (n= 73); White (n=32). Other category includes participants who did not identify with Black or White hSex: male (n=47); female (n= 79) iHealthy Eating Index, 0-100 scale, Higher score represents better diet quality [198] jBy-group differences determined by generalized linear mixed model controlling for: cohort; household income; race; baseline value; participant id(family id); level of attendance (Y (outcome variable)= outcome at T1) kFood Preparation Skills, 0-32 scale with higher score representing increased skills lFood Preparation Frequency, 0-40 scale with higher score representing increased frequency

149 Table 17. Intervention Period: Within-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1 Outcomes Control (n=39) Low Attenders (n=33) High Attenders (n=54) Baseline T0 Post-Test T1 Baseline T0 Post-Test T1 Baseline T0 Post-Test T1 Dietary Intakea,b,c,d Fruit intake (svgs/d) 1.86 (1.8) 1.48 (1.2) 1.76 (1.2) 1.18 (1.0) 1.62 (1.3) 1.79 (1.9) Whole fruit intake (svgs/d) 1.39 (1.5) 1.12 (1.2) 1.07 (0.9) 1.03 (1.2) 1.20 (0.9) 1.38 (1.9) Vegetable intake (svgs/d) 1.70 (1.6) 1.60 (1.7) 1.67 (1.4) 2.24 (1.7) 1.72 (1.4) 2.02 (1.7) Sugar sweetened beverage intake 0.75 (1.1) 4.23 (16.4) 0.70 (1.7) 0.75 (1.4) 0.45 (0.8) 0.51 (0.9) (svgs/d) Energy intake (kcals/d) 1863.1 1729.6 1749.04 1844.66 (663.5) 1663.36 (528.4) 1709.36 (608.2) (882.5) (719.9) (593.4) 150 HEI total scoree 53.25 (11.4) 55.81 (9.4) 56.39 (15.4) 56.88 (15.4) 50.68 (10.1) 52.70 (11.1)

HEI total fruit scoref 3.13 (2.1) 2.73 (2.6) 3.03 (2.1) 2.22 (2.1) 2.97 (1.9) 2.44 (2.0) HEI whole fruit scoreg 2.89 (2.2) 2.40 (1.9) 3.04 (1.9) 2.50 (2.7) 3.10 (1.9) 2.76 (3.1) HEI total vegetable scoreh 1.84 (1.5) 1.80 (1.6) 2.15 (1.4) 2.32 (1.6) 2.21 (1.5) 2.38 (1.7) Anthropometric/BPa,b BMI z-score 0.75 (1.3) 0.69 (1.4) 0.54 (1.1) 0.60 (1.1) 0.76 (1.1)m 0.48 (1.3)n WC z-score 0.63 (1.3) 0.71 (1.2) 0.87 (0.8) 0.75 (0.8) 0.70 (0.9) 0.82 (0.9) Systolic BP z-score 1.47 (1.1) 1.70 (1.6) 1.18 (1.5) 1.40 (1.4) 1.55 (1.3) 1.19 (1.3) Diastolic BP z-score 1.12 (1.3) 1.39 (1.8) 1.15 (1.5) 1.06 (1.1) 1.41 (1.4) 1.11 (1.4) Pers. Determinantsa,b Food prep skillsj 23.21 (4.8) 22.85 (5.5) 21.94 (4.6)m 28.00 (5.3)n 22.09 (5.5)m 23.23 (4.2)n Food prep frequencyk 18.03 (5.1) 19.16 (7.5) 17.25 (5.4)m 23.47 (6.1)n 18.87 (5.9)m 22.56 (6.5)n aValues are Mean (SD) bWithin-group differences determined by generalized linear modeling controlling for: participant id (family id) (Y (outcome variable)= (outcome at T1 – outcome at T0)) cDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dParticipant completion of 24hr dietary recalls at T0: 0 recalls (n=9 (7%)); 1 recall (n=81 (68%)); 2 recalls (n=14 (12%)); 3 recalls (n=15 (13%)). Participant completion of 24hr dietary recalls at T1: 0 recalls (n=47 (39%)); 1 recall (n=68 (58%)); 2 recalls (n=4 (4%)); 3 recalls (n=0 (0%)) eHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] Table 17. (continued)

150 Table 17. Intervention Period: Within-Group (Level of Attendance) Child Outcomes at Baseline T0 and Post-Test T1 (continued) fHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake gHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake hHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake iFood Preparation Skills, 0-32 scale with higher score representing increased skills jFood Preparation Frequency, 0-40 scale with higher score representing increased frequency k,lDifferent superscripts indicate by-group difference in participant outcome of p<0.05 m,nDifferent superscripts indicate by-group difference in participant outcome of p<0.01

151

151 Table 18. Follow-Up Period: By-Group (Low vs. High Attenders) Child Outcomes at Post-Test T1 and Follow-Up T2 Child Outcomes Post-Test T1 Follow-Up T2 Mean (SD) Change from Post-Test to Follow-Up T2 Dietary Intake,a,b,c,d Low Attenderse High Attenderse Low Attenderse High Attenderse Low Attenderse High Attenderse (n=33) (n=54) (n=33) (n=54) (n=33) (n=54) Fruit intake (svgs/d)f 1.41 (0.6) 0.66 (1.2) 0.50 (0.7) 1.09 (1.3) -0.91 (1.3) 0.43 (1.4) Whole fruit intake (svgs/d) 1.40 (0.6) 0.56 (0.8) 0.50 (0.7) 0.78 (1.0) -0.90 (1.2)m 0.22 (1.3)n Vegetable intake (svgs/d) 1.58 (0.2) 2.99 (2.2) 2.72 (1.1) 3.04 (1.8) 1.14 (0.8) 0.05 (2.9) Sugar sweetened beverage intake 0.42 (0.8) 0.35 (0.6) 0.43 (0.9) 0.34 (0.6) 0.01 (1.1) -0.01 (0.7) (svgs/d) Energy intake (kcals/d) 1763.96 (543.8) 1707.88 (860.4) 2107.34 (106.8) 1488.86 (559.8) 343.38 (650.6) -219.02 (1030.7) HEI scoreg 72.08 (14.6) 51.32 (15.2) 47.23 (3.7) 55.75 (10.4) -24.85 (10.8) 4.43 (16.6) HEI Total Fruit scoreh 2.26 (2.4) 2.60 (4.2) 2.27 (1.8) 2.64 (4.1) 0.01 (1.1) 0.04 (4.1) HEI Whole Fruit scorei 2.48 (2.2) 1.71 (2.1) 2.51 (1.9) 1.74 (1.9) 0.03 (1.9) 0.03 (1.8) HEI Total Vegetable scorej 2.19 (1.0) 3.21 (1.7) 2.91 (1.0) 3.62 (1.6) 0.72 (9.1) 0.41 (1.4) 152 Athropometrics/BPa,b

BMI z-score 31.53 (9.1) 31.88 (9.6) 31.16 (9.2) 31.79 (9.3) -0.37 (9.0) 0.63 (1.9) Waist circumference z-score 101.20 (20.6) 100.65 (20.6) 99.19 (18.8) 100.76 (21.0) -2.01 (19.1)) 0.11 (11.9) Systolic BP z-score 121.61 (12.3) 128.66 (16.6) 127.01 (16.4) 126.62 (19.8) 5.40 (14.8)) -2.04 (11.6) Diastolic BP z-score 78.73 (8.9) 81.27 (12.2) 80.28 (11.9) 78.84 (12.8) 1.55 (9.2)) -2.43 (13.3) Personal Determinantsa,b Food Preparation Skillsk 103.02 (11.5) 100.51 (14.2) 104.82 (12.4) 102.13 (14.1) 1.80 (12.2) 1.62 (8.2) Food Preparation Frequencyl 26.55 (4.9) 26.77 (3.8) 26.63 (3.8) 27.32 (3.1) 0.08 (4.4) 0.77 (2.4) aValues are Mean(SD) bWithin-group differences determined by generalized linear modeling controlling for: participant id(family id) cDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dIntervention participants’ (n=63) completion of 24hr dietary recalls at T1: 0 recalls (n=22 (35%)); 1 recall (n=38 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=1 (2%)). Intervention participants’ (n=63) completion of 24hr dietary recalls at T2: 0 recalls (n= 22 (35%)); 1 recall (n=39 (65%)); 2 recalls (n=0 (0%)); 3 recalls (n=0 (0%)) eAttendance based on total number of lessons attended Table 18. (continued)

152 Table 18. Follow-Up Period: By-Group (Low vs. High Attenders) Child Outcomes at Post-Test T1 and Follow-Up T2 (continued) fTotal fruit= whole fruit + 100% fruit juice gHealthy Eating Index (HEI), 0-100 with higher score representing better intake hHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake jHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake k Food Preparation Skills, 0-32 scale with higher score representing increased self-efficacy lFood Preparation Frequency, 0-40 scale with higher score representing increased menu planning m,nBy-Group difference of p<0.05

153

153 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes Follow-Up Period (T1 àT2) Covariates β (SE) p-value Confidence Interval Child Dietary Intakea,b,c Fruit intake (svgs/d) Attendanced -High -1.33 (1.4) 0.344 (-4.19, 1.53) -Low 0 - - Cohorte -One -1.26 (1.3) 0.349 (-4.00, 1.48) -Two -0.32 (1.5) 0.829 (-3.39, 2.75) -Three 0 - - Incomef -Low 0.67 (1.1) 0.563 (-1.70, 3.03) -Non-Low 0 - - Raceg

154 -Other -0.51 (2.7) 0.855 (-6.24, 5.22) -Black -1.97 (2.5) 0.449 (-7.28, 3.35) -White 0 - - Sexh -Male 1.54 (1.1) 0.183 (-0.79, 3.86) -Female 0 - - Child Age -0.08 (0.2) 0.710 (-0.51, 0.35) Whole fruit intake (svgs/d) Attendanced -High -3.10 (0.9) 0.003 (-5.01, -1.20) -Low 0 0 0 Cohorte -One -0.83 (1.4) 0.555 (-3.70, 2.05) -Two -0.23 (1.6) 0.886 (-3.51, 3.06) -Three 0 0 0 Table 19. (continued)

154 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes (continued) Incomef -Low 0.17 (1.2) 0.889 (-2.31, 2.65) -Non-Low 0 0 0 Raceg -Other -0.67 (2.8) 0.817 (-6.59, 5.26) -Black -2.51 (2.6) 0.353 (-8.01, 3.00) -White 0 - - Sexh -Male 0.99 (0.8) 0.221 (-0.64, 2.62) -Female 0 - - Child Age 0.09 (0.1) 0.475 (-0.16, 0.33) Vegetable intake (svgs/d) Attendanced -High -1.43 (1.0) 0.162 (-3.48, 0.62) -Low 0 - - Cohorte 155 -One 0.37 (0.9) 0.701 (-1.62, 2.36)

-Two 0.53 (1.1) 0.623 (-1.70, 2.77) -Three 0 - - Incomef -Low -1.49 (0.8) 0.086 (-3.21, 0.23) -Non-Low 0 - - Raceg -Other 3.14 (1.9) 0.131 (-1.02, 7.30) -Black 2.15 (1.8) 0.258 (-1.71, 6.01) -White 0 - - Sexh -Male 0.06 (0.8) 0.943 (-1.61, 1.73) -Female 0 - - Child Age -0.19 (0.1) 0.198 (-0.50, 0.11) Table 19. (continued)

155 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes (continued) Sugar sweetened beverage intake (svgs/d) Attendanced -High 0.01 (0.0) 0.838 (-0.07, 0.09) -Low 0 0 0 Cohorte -One 0.46 (1.2) 0.707 (-2.05, 2.96) -Two -0.72 (1.4) 0.607 (-3.62, 2.17) -Three 0 - - Incomef -Low 0.88 (1.0) 0.408 (-1.30, 3.06) -Non-Low 0 - - Raceg -Other -0.10 (2.5) 0.969 (-5.30, 5.10) -Black 0.69 (2.3) 0.767 (-4.13, 5.51) -White 0 - - Sexh 156 -Male -2.87 (0.0) <0.001 (-2.94, -2.80)

-Female 0 - - Child Age 0.01 (0.0) 0.704 (-0.01, 0.01) Energy intake (kcals/d) Attendanced -High -255.25 (491.8) 0.610 (-1,284.52, 774.03) -Low 0 - 0 Cohorte -One -69.64 (444.4) 0.877 (-999.87, 860.58) -Two 495.14 (495.8) 0.330 (-542.51, 1,532.79) -Three 0 - 0 Incomef -Low -422.50 (384.6) 0.286 (-1,227.46, 382.45) -Non-Low 0 - 0 Table 19. (continued)

156 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes T2 (continued) Raceg -Other 664.35 (939.6) 0.488 (-1,302.28, 2,630.98) -Black -102.40 (870.6) 0.908 (-1,924.54, 1,719.74) -White 0 - 0 Sexh -Male 260.90 (395.7) 0.518 (-567.40, 1,089.19) -Female 0 - - Child Age 1.82 (75.9) 0.981 (-157.19, 160.82) HEI scorei Attendanced -High 14.45 (10.6) 0.190 (-7.82, 36.71) -Low 0 - - Cohorte -One -6.73 (7.8) 0.397 (-23.02, 9.55) -Two -8.97 (8.5) 0.305 (-26.78, 8.84)

157 -Three 0 - - Incomef

-Low 1.52 (7.0) 0.832 (-13.23, 16.26) -Non-Low 0 - - Raceg -Other 31.95 (18.5) 0.100 (-6.68, 70.57) -Black 25.38 (16.8) 0.148 (-9.82, 60.57) -White 0 - - Sexh -Male -3.82 (7.9) 0.634 (-20.34, 12.71) -Female 0 - 0 Child Age -0.79 (1.9) 0.676 (-4.66, 3.09) Table 19. (continued)

157 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes T2 (continued) Child Anthropometrics/Blood Pressurej BMI z-score Attendanced -High -0.06 (0.1) 0.624 (-0.28, 0.17) -Low 0 - - Cohorte -One -0.02 (0.1) 0.855 (-0.26, 0.22) -Two 0.02 (0.1) 0.854 (-0.30, 0.25) -Three 0 - - Incomef -Low -0.01 (0.1) 0.966 (-0.23, 0.22) -Non-Low 0 - - Raceg -Other -0.07 (0.2) 0.677 (-0.42, 0.28) -Black -0.13 (0.1) 0.378 (-0.42, 0.16) -White 0 - - 158 Waist Circumference z-score Attendanced

-High -0.07 (0.2) 0.685 (-0.41, 0.27) -Low 0 - - Cohorte -One 0.36 (0.2) 0.055 (-0.01, 0.73) -Two 0.30 (0.2) 0.162 (-0.12, 0.71) -Three 0 - - Incomef -Low -0.18 (0.2) 0.301 (-0.52, 0.16) -Non-Low 0 - - Raceg -Other 0.58 (0.3) 0.037 (0.03, 1.12) -Black -0.01 (0.2) 0.965 (-0.45, 0.43) -White 0 - - Table 19. (continued)

158 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes (continued) Systolic BP z-score Attendanced -High 0.38 (0.5) 0.463 (-0.65, 1.41) -Low 0 - - Cohorte -One 0.67 (0.6) 0.228 (-0.43, 1.78) -Two 0.92 (0.7) 0.164 (-0.39, 2.23) -Three 0 - - Incomef -Low -1.39 (0.5) 0.010 (-2.44, -0.34) -Non-Low 0 - - Raceg -Other 2.05 (0.8) 0.017 (0.38, 3.72) -Black 1.63 (0.7) 0.024 (0.22, 3.05) -White 0 - - Diastolic BP z-score Attendanced 159 -High -0.73 (0.6) 0.223 (-1.93, 0.46)

-Low 0 - - Cohorte -One -1.26 (0.6) 0.042 (-2.46, -0.05) -Two -1.25 (0.7) 0.069 (-2.61, 0.10) -Three 0 - - Incomef -Low -1.07 (0.5) 0.057 (-2.16, 0.03) -Non-Low 0 - - Raceg -Other 1.51 (0.8) 0.082 (-0.20, 3.21) -Black 0.32 (0.7) 0.645 (-1.07, 1.71) -White 0 - - Table 19. (continued)

159 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes T2 (continued) Child Personal Determinantsa Food Preparation Skillsk Attendanced -High 1.18 (1.1) 0.292 (-0.50, 3.68) -Low 0 - - Cohorte -One -1.66 (1.2) 0.181 (-4.01, 0.60) -Two -2.27 (1.4) 0.115 (-4.96, 0.19) -Three 0 - - Incomef -Low 1.58 (1.2) 0.184 (-0.17, 4.13) -Non-Low 0 - - Raceg -Other 2.04 (1.8) 0.255 (-1.41, 5.26) -Black 0.75 (1.5) 0.621 (-2.09, 3.37)

160 -White 0 - - Sexh

-Male 1.58 (1.1) 0.140 (-0.09, 3.96) -Female 0 - - Child Age -0.52 (0.2) 0.028 (-0.99, 0.03) Food Preparation Frequencyl Attendanced -High 2.37 (2.2) 0.287 (-2.03, 6.77) -Low 0 - - Cohorte -One -2.90 (2.4) 0.239 (-7.77, 1.97) -Two -2.28 (2.8) 0.424 (-7.93. 3.37) -Three 0 - - Incomef -Low 2.81 (2.3) 0.236 (-1.88, 7.49) -Non-Low 0 - - Table 19. (continued)

160 Table 19. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Child Outcomes T2 (continued) Raceg -Other 7.55 (3.5) 0.036 (0.51, 14.60) -Black 6.48 (2.9) 0.033 (0.53, 12.43) -White 0 - - Sexh -Male 0.92 (2.1) 0.660 (-3.24, 5.08) -Female 0 - - Child Age -1.32 (0.4) 0.004 (-2.21, -0.44) aBy-group differences determined by generalized linear mixed modeling controlling for: cohort; household income; sex; race; age; participant id(family id); level of attendance (Y (outcome variable)= change score of T2-T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] cIntervention participants’ (n=87) completion of 24hr dietary recalls at T1: 0 recalls (n=33 (38%)); 1 recall (n=52 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=0 (0%)). Intervention participants’ (n=87) completion of 24hr dietary recalls at T2: 0 recalls (n= 38 (44%)); 1 recall (n=49 (56%)); 2 recalls (n=0

161 (0%)); 3 recalls (n=0 (0%)) dAttendance (based on total number of lessons attended): low attendance group (n=33); high attendance group (n=54)

eCohort: one (n= 24); two (n=15); three (n=48) fIncome: low (n= 32); non-low (n=50). Low-income defined as participation in one or more of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) gRace: other (n= 14); Black (n= 52); White (n=16). Other category includes participants who did not identify Black or White hSex: male (n=34); female (n= 54) iHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] jBy-group differences determined by generalized linear mixed modeling controlling for: cohort; household income; race; participant id(family id); level of attendance (Y (outcome variable)= change score of T2-T1) kFood Preparation Skills, 0-32 scale with higher score representing increased skills lFood Preparation Frequency, 0-40 scale with higher score representing increased frequency

161 Table 20. Follow-Up Period: Within-Group Differences in Child Outcomes Child Outcomes Low Attendersa High Attendersa (n=33) (n=54) Dietary Intakeb,c,d,e Post-Test T1 Follow-Up T2 Post-Test T1 Follow-Up T2

Fruit intake (svgs/d)f 1.41 (0.6) 0.50 (0.7) 0.66 (1.2) 1.09 (1.3) Whole fruit intake (svgs/d) 1.40 (0.6) 0.50 (0.7) 0.56 (0.8) 0.78 (1.0) Vegetable intake (svgs/d) 1.58 (0.2) 2.72 (1.1) 2.99 (2.2) 3.04 (1.8) Sugar sweetened beverage intake 0.42 (0.8) 0.43 (0.9) 0.35 (0.6) 0.34 (0.6) (svgs/d) Energy intake (kcals/d) 1763.96 (543.8) 2107.34 (106.8) 1707.88 (860.4) 1488.86 (559.8) HEI scoreg 72.08 (14.6) 47.23 (3.7) 51.32 (15.2) 55.75 (10.4) HEI Total Fruit scoreh 2.26 (2.4) 2.27 (1.8) 2.60 (4.2) 2.64 (4.1) HEI Whole Fruit scorei 2.48 (2.2) 2.51 (1.9) 1.71 (2.1) 1.74 (1.9) HEI Total Vegetable scorej 2.19 (1.0) 2.91 (1.0) 3.21 (1.7) 3.62 (1.6) 162 Athropometrics/BPb,c

BMI z-score 31.53 (9.1) 31.16 (9.2) 31.88 (9.6) 31.79 (9.3) Waist circumference z-score 101.20 (20.6) 99.19 (18.8) 100.65 (20.6) 100.76 (21.0) Systolic BP z-score 121.61 (12.3) 127.01 (16.4) 128.66 (16.6) 126.62 (19.8) Diastolic BP z-score 78.73 (8.9) 80.28 (11.9) 81.27 (12.2) 78.84 (12.8) Personal Determinantsb,c Food Preparation Skillsk 103.02 (11.5) 104.82 (12.4) 100.51 (14.2) 102.13 (14.1) Food Preparation Frequencyl 26.55 (4.9) 26.63 (3.8) 26.77 (3.8) 27.32 (3.1) aAttendance based on total number of lessons attended bValues are Mean(SD) cWithin-group differences determined by generalized linear modeling controlling for: participant id(family id) dDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] eIntervention participants’ (n=63) completion of 24hr dietary recalls at T1: 0 recalls (n=22 (35%)); 1 recall (n=38 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=1 (2%)). Intervention participants’ (n=63) completion of 24hr dietary recalls at T2: 0 recalls (n= 22 (35%)); 1 recall (n=39 (65%)); 2 recalls (n=0 (0%)); 3 recalls (n=0 (0%)) Table 20. (continued)

162 Table 20. Follow-Up Period: Within-Group Differences in Child Outcomes (continued) fTotal fruit= whole fruit + 100% fruit juice gHealthy Eating Index (HEI), 0-100 with higher score representing better intake hHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake jHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake k Food Preparation Skills, 0-32 scale with higher score representing increased self-efficacy lFood Preparation Frequency, 0-40 scale with higher score representing increased menu planning

163

163 Discussion

The present study assessed the effectiveness of a 10-week multi-component family meals intervention study aimed at eliciting positive changes in child dietary intake and weight status among 4-10-year-old children from underserved families. The Simple

Suppers study fills a gap in the family meals research as it is the first family meals intervention to target underserved families with children 4-10 years old. The elementary school years have been demonstrated to be a period for increased risk for weight gain

[220], making it a critical period to target for childhood obesity prevention interventions.

And while an abundance of the existing family meals research has targeted children 8-12 years old, there have been recent indications that family meals interventions may be most effective among younger, pre-pubescent children (<10 years old) [85]. In addition, several reviews have concluded that the inverse association between family meal frequency and child weight status is pronounced among younger children [221, 222].

The high feasibility and acceptability and positive child effectiveness outcomes of this study demonstrate the great potential for family meals programming amongst underserved families with 4-10 year old children. The significant improvements observed in parent-reported child food preparation skills among intervention children at post-test (p<0.001) is consistent with previous family meals research [73]. However, in addition to assessing child food preparation skills, the present study expanded assessment of child food preparation beyond skills to include frequency, demonstrating a significant

164 increase (p<0.05) in child food preparation frequency among intervention children, regardless of level of attendance during the intervention period.

In addition to being the first family meals intervention to assess child food preparation frequency, to our knowledge, the Simple Suppers study is the first family meals intervention to produce a significant decrease in child BMI z-score. At the end of the 10-week program, there was a significant decrease in child BMI z-score among children who attended seven or more lessons relative to children attending less than seven classes (p=0.048) and control children (p=0.003). Interestingly, the between group comparison of change in BMI z-score at follow-up demonstrated a more significant decrease between children in the low and high attendance groups compared with the control group and high attendance group. This finding may be a result of this study’s waitlist control design, as participants/their caregivers in a waitlist control group often begin making changes during the waitlist period in anticipation of receiving an upcoming intervention. Therefore, the waitlist control group may not have been representative of a true control group because participants/their caregivers may have begun making changes that would impact child weight status during the weight list control period.

The present study had several strengths. The Simple Suppers curriculum, which is evidence- and theory-based, was developed and peer-reviewed by a team of experts in the field and pilot tested for feasibility. Outcomes were assessed using reliable questionnaires and established protocols were utilized by trained researchers to assess

165 anthropometric, biometric, and dietary outcomes. In addition, this study was the first family meals intervention to expand its anthropometric/biometric outcomes beyond

BMI/BMI z-score, to include WC and systolic and diastolic BP. Delivering the intervention at a faith-based community center that serves the intervention’s target population (underserved families) likely contributed to the high feasibility and acceptability of the program, as the literature demonstrates faith-based community centers are highly effective in engaging underserved families in programming.

There are noteworthy limitations to this study. First, the study design lacked randomization as a quasi-experimental design was utilized. Randomization was not appropriate for this study because preserving sample size and developing trust with the site/participating families was paramount [213–215]. Lack of randomization was overcome in the present study by assessing between-group differences at baseline. Child race was the only covariate that significantly differed between the intervention and control groups at baseline, which was controlled for in all of the statistical models. A second limitation of this study was assessment of child dietary intake. The dietary intake assessment protocol was to conduct three, caregiver-assisted, nonconsecutive (two weekdays, one weekend day) 24hr dietary recalls using USDA’s 5-step multi-pass dietary recall method at each time point [189, 197]. Average dietary intake at each time point was to be determined by averaging intake across the three days. However due to challenges contacting participants, three 24hr dietary recalls were not always able to be collected at each time point. In these instances, if two recalls were conducted, the average

166 of the two recalls would be used and if a single 24hr dietary intake was collected, the single recall was used.

The are many opportunities for future research in the area of family meals.

Maintenance of behavior change is consistently a challenge, especially with regards to diet and weight status [223]. In the present study, a 10-week follow-up period aimed to assess maintenance of behavior change post-program. Given the positive intervention effects observed on child BMI z-score during both the intervention (decrease in child

BMI z-score) and follow-up (maintenance of child BMI z-score) periods, extending the follow-up period (up to 5 years post-intervention) would provide greater insight on the potential long-term intervention effects of the program.

However, a low-touch follow-up period intervention has the potential to extend and continue improvements in outcomes observed during the intervention period, rather than just maintenance of outcomes.

Finally, to our knowledge, this study is the first family meals intervention to target a population at increased risk for childhood obesity (i.e., 4-10 year old children of racially diverse backgrounds from underserved families). Previous family meals interventions have primarily targeted White, 8-12 year old children, from economically stable households [73, 85]. While targeting ‘at-risk’ populations often imposes numerous

167 research design and implementation challenges [224], future family meals research should continue to target populations at increased risk for childhood obesity (e.g.,

Hispanic children), to provide assistance to those most in-need and fill these apparent gaps in this line of research.

Conclusions

Participation in Simple Suppers at a 70% level led to a significantly higher whole fruit intake and decreased BMI z-score, which were maintained at the 10-week follow-up. Results from this study demonstrate the potential for engagement in an evidence-based family meals program to positively impact child weight status and dietary intake among a racially diverse sample of school-aged children from underserved families. Of significance given the ongoing childhood obesity epidemic.

168 Chapter 6. Caregiver and Family Findings from a 10-Week Multi-Component Family Meals Intervention for Underserved Families with Children 4-10 Years Old

169 Abstract

Background: Caregivers play a critical role in establishing a home food environment that fosters healthy dietary and lifestyle behaviors for their children. Offering regular, quality family meals is one aspect of the home food environment with which caregivers face many barriers. Given the potential protective effect of family meals on childhood obesity, there is need for additional research aimed at providing caregivers with resources needed to overcome family meal barriers. Therefore, this study purpose was to assess the feasibility, acceptability, and effectiveness of a 10-week multi-component family meals intervention study aimed at eliciting positive changes in caregiver diet and weight status and the family meals environment by equipping caregivers with resources needed to establish regular, quality family meals with their family.

Methods: A 10-week family meals program (Simple Suppers) designed for underserved families with 4-10 year old children from racially diverse backgrounds was implemented as a pretest-posttest, multi-cohort, quasi-experimental trial with waitlist control. The 10,

90-minute program lessons were delivered weekly over the dinner hour at a faith-based community center. Session components included: a) interactive group discussion of strategies to overcome family meal barriers, plus weekly goal setting for caregivers; b) engagement in age-appropriate food preparation activities for children; and c) group family meal for caregivers and children. Main caregiver- and family-level outcome measures were change in: caregiver diet quality; caregiver body mass index (BMI); and frequency of family meals.

170 Results: At baseline, 95 caregivers/families enrolled in the study; approximately 98% were female, 62% identified as Black, 50% were between 31-40 years old, and mean

BMI was 33.0 kg/m2. Generalized linear models, adjusted for baseline values and demographics, demonstrated an intervention effect on caregiver BMI, with intervention caregivers having a significant decrease in BMI (p=0.028), which was maintained during the follow-up period. No significant intervention effects were seen in caregiver dietary intake or frequency of family meals at post-test or 10-week follow-up.

Conclusions: The Simple Suppers study assessed the feasibility, acceptability and effectiveness of a theory- and evidence-based family meals program aiming to produce positive caregiver health and family meals environment outcomes among a sample of caregivers from racially diverse and underserved families with children 4-10 years old.

Participation in Simple Suppers decreased caregiver BMI, which was maintained 10- weeks post-program. Results from this study demonstrate the potential for engagement in an evidence-based family meals program to positively impact caregiver weight status among a racially diverse sample of caregivers of 4-10 year old children. And given the strong correlation between caregiver and child weight status, this study provides promise of having a positive impact on child weight status, as well.

171 Background

Childhood overweight and obesity continues to be a serious public health concern, with about one-third of youth in the U.S. being overweight or obese [185]. In order to slow the progression of the childhood obesity epidemic, prevention interventions focusing on establishing healthy lifestyle behaviors during childhood are imperative.

Caregivers, often referred to as ‘agents of change’ [225], play a critical role in developing a home environment that fosters healthy lifestyle behaviors for their children [226].

Caregivers directly influence the home environment and their child’s dietary and lifestyle behaviors through their nutrition knowledge, influence over food selection, modeling of healthful dietary practices, and eating/meal patterns.

Family mealtime is a facet of the home food environment that has become a growing area of research as a result of cross-sectional and longitudinal studies demonstrating a positive association between family meal frequency and child diet quality [75–79] and some showing an inverse association between family meal frequency and child weight status [81, 217, 227]. A similar positive association has been observed between family meal frequency and caregiver diet quality, with increased family meal frequency associated with increased fruit and vegetable intake [81]. Caregivers view family meals in a positive manner, as they offer an opportunity for caregivers to nourish, converse, and bond with their children [206]. However, caregivers often face many challenges to regularly offering quality family meals, where healthful meals are served in a conducive family meal environment. Caregivers are often unable to offer quality

172 family meals regularly due to challenges with meal planning, food preparation, food pickiness, and mealtime conflicts.

Despite the strong evidence demonstrating a positive association between family meal frequency and child diet quality [75–79], few interventions have sought to provide caregivers with the resources needed to overcome these family meal barriers. In addition, the relationship between family meal frequency and caregiver health outcomes is still unclear. While data has indicated a positive association between family meal frequency and caregiver diet quality may exist, this finding has been limited to a single study [228].

The relationship between family meal frequency and caregiver weight status is even more unclear than diet quality. While some studies have indicated a potential inverse association between family meal frequency and caregiver weight status/body mass index

(BMI) [81, 228], others have found no association [229]. These contradicting findings may be due to suboptimal assessment methodology, as BMI was calculated from self- report data in each of these studies. Furthermore, to our knowledge, assessment of caregiver anthropometric outcomes has been limited to BMI only in the family meals literature.

As caregivers serve as agents of change for childhood obesity prevention and treatment [225], understanding the relationship between family meals and caregiver weight status/BMI and other health indices has the potential to positively impact the weight status and overall health of an entire family. Therefore, the purpose of this paper

173 was to assess the effectiveness of a 10-week multi-component family meals intervention study, Simple Suppers, aimed at eliciting positive changes in caregiver health outcomes

(dietary intake and weight status) and the home environment (family meal frequency and quality). The Simple Suppers study is a two group quasi-experimental trial with staggered cohort design that targets underserved families with elementary school age children (4-10 years) and includes an examination of caregiver health outcomes beyond weight status.

Methods

Objectives and Hypotheses

The objectives of this study with related hypotheses were as follows:

Objective 1. Assess the impact of Simple Suppers on caregivers of participating families relative to caregivers of families in the control group

Hypothesis 1.1. Diet quality, BMI, waist circumference (WC), and blood

pressure (BP) will improve more from baseline to post-intervention among

caregivers participating in the intervention than in the controls.

Hypothesis 1.2. Diet quality, BMI, WC, and BP improvements will be

maintained during the follow-up period among caregivers participating in the

intervention.

Objective 2. Assess the impact of Simple Suppers on the family meals environment of participating families relative to the controls.

174 Hypothesis 2.1. Frequency of family meals (breakfast and dinner), TV viewing

during meals, and eating family meals in a dining area will improve more from

baseline to post-intervention among families participating in the intervention than

in the controls.

Hypothesis 2.2. Frequency of family meals (breakfast and dinner), TV viewing

during meals, and eating family meals in a dining area improvements will be

maintained during the follow-up period among families participating in the

intervention.

Study Design

The Simple Suppers study, which aimed to improve child and caregiver diet quality and weight status by improving the quality of family meals among underserved families with children 4-10 year old, was implemented over 12-months as a two-group

(intervention; waitlist control) quasi-experimental trial using a staggered cohort design

[218]. At each of three time periods, separated by 10 weeks, a cohort of families was recruited. The original recruitment goal of approximately 20 families per cohort was increased to 30 families per cohort as a result of additional funding. Families self-selected their group assignment (intervention, waitlist control) by being offered the opportunity to participate in the program during the first (intervention group) or second 10-week session

(waitlist control group). Upon confirmation of study eligibility, a baseline data collection appointment was scheduled at the participating family’s home or faith-based community center during the two weeks preceding intervention commencement. For participating

175 families, data was collected on the primary meal preparing caregiver via direct assessment and caregiver completed questionnaires at baseline (time point 0, T0), 10- week post-test (time point 1, T1), and 10-week follow-up (time point 2, T2). Written caregiver consent was obtained at baseline. A team of trained research staff, blinded from group assignment, facilitated data collection. Participating families received a $25 grocery store gift card at each data collection point for their participation in the research.

All study materials and procedures were approved by the Institutional Review Board at

Ohio State University.

Setting

Participants

176 Intervention

The Intervention Mapping protocol was utilized in the development of the Simple

Suppers intervention [186, 187, 218]. The Social Cognitive Theory, which posits that behavior change is a function of a reciprocal relationship between personal (e.g., SE and behavioral capabilities, such as menu planning) and environmental (e.g., norms, modeling, and reinforcement) factors, served as the theoretical foundation for the Simple

Suppers intervention [163, 181].

The Simple Suppers program included 10, 90-minute lessons delivered weekly over the dinner hour. Each lesson focused on a family meals topic identified in the literature as a barrier to family meals for families with young children (e.g., timesaving strategies for family meals; connecting with your child through family meals). Session components included: a) interactive group discussion and goal setting with caregivers; b) hands-on food preparation activities with children; and c) group family meal with caregivers and children.

Caregiver Outcome Measures

Diet Quality. Dietary intake was assessed by conducting three, nonconsecutive

(two weekdays, one weekend day) 24-hour (24hr) dietary recalls using USDA’s 5-step multi-pass dietary recall method [197, 230, 231]. At each data collection time point, the first dietary recall was conducted during the in-person data collection visit, and the remaining two were conducted via telephone within two weeks of the initial in-person

177 recall. Typical daily dietary intake was determined by averaging dietary intake across the three recalls at each time point to determine daily servings of fruit, vegetables, and SSB.

Diet quality was assessed at each point by calculating a Healthy Eating Index 2010 score using the three 24hr dietary recalls collected [198].

Anthropometric and Blood Pressure Assessments. Standardized procedures were used to assess heights ad weights [199, 200]. Body mass index was calculated using measured heights and weights. Waist circumference was measured on all participating caregivers with a tape measure at the uppermost lateral border of the hip crest (ilium)

[199]. Blood pressure was assessed on all participating caregivers via automated, calibrated BP monitors (Panasonic EW3109W).

Personal Determinants. Caregiver SE for healthy dietary behaviors was assessed via a questionnaire modified from an existing questionnaire [204]. The 12-item questionnaire, which asked caregivers to rate their confidence for specific healthy family meals dietary behaviors using a 10-point scale, demonstrated good internal consistency

(Cronbach’s alpha, a= 0.84). Menu planning behaviors were assessed using an existing questionnaire that previously demonstrated adequate internal consistency (Cronbach’s alpha, a= 0.68) and high test-retest reliability (Pearson test-retest= 0.89) [169].

Caregivers completed the 9-item menu planning questionnaire by rating statements regarding menu planning, meal decision-making and grocery shopping using a 4-point scale (‘never,’ ‘sometimes,’ ‘often,’ ‘always’).

178 Family Meals Outcome Measures

Frequency of family meals (breakfast and dinner), TV viewing during meals, and eating family meals in a dining area were assessed using a modified existing questionnaire [73]. Weekly frequency of shared family breakfasts and dinners were assessed via two independent questions: Over the past 7 days, how many times did all or most of your family eat (breakfast, dinner) together? Response items included 5 options ranging from never to 7 times. Television viewing during meals was assessed with the following question: Over the past 7 days, how many times has your child watched TV while eating family meals? Eating family meals in a dining area was assessed with the following question: Over the past 7 days, how many times was a family meal eaten at the table in the dining area? Response items for these questions included 5 options ranging from never to 7+ times. The family meals environment questionnaire demonstrated acceptable internal consistency (Cronbach’s alpha, a= 0.76).

Process Outcomes

179 Data Analysis

Data from each of the three cohorts was pooled and the intervention tested by comparing change (T1-T0) in caregiver and family meal outcomes of caregivers in the intervention compared to participants in the waitlist control (hypotheses 1.1 and 2.1).

Generalized linear models were used to determine the association between the difference in the response variables of interest between the intervention and control group, controlling for potential confounders (race, income, cohort, intervention dose), from baseline (T0) to 10-week post-test (T1). Sustainability of intervention effects were tested by pooling intervention group data from each of the three cohorts, comparing change (T2-

T1) in caregiver and family meal outcomes among intervention group participants at the end of the 10-week follow-up period (hypotheses 1.2 and 2.2).

For caregiver dietary outcomes, while the protocol was to collect 3 dietary recalls at each time point, this was often not feasible due to challenges contacting participants and participants’ requesting not to complete the dietary recalls. In these instances, if two recalls were conducted, the average of the two recalls would be used and if a single 24hr dietary intake was collected, the single recall was used. For family meal outcomes, child- and caregiver-level potential confounders were used in the models. For families with multiple children enrolled in the study, the potential confounders were used from the oldest child, as the literature demonstrates older children have a strong influence on family meal frequency [87, 221]. Multiple imputations were used to deal with missing data with the exception of the dietary outcomes, due to great variability in dietary intake.

180 Results

Baseline Descriptives

One-hundred-nine caregivers/families were recruited at baseline. Family-level program retention was 87.5%, resulting in 95 caregivers/families completing the program. Descriptive summaries of sample baseline measures are presented in Table 16.

Among the 95 caregiver participants, 98% were female, 50% were between 31-40 years old, approximately 56% were overweight or obese, about 62% identified as Black and approximately 40% were reliant on the WIC, SNAP, and/or NSLP federal food assistance programs to feed themselves and their family. Baseline participant characteristics did not differ by group assignment.

181 Table 21. Baseline Caregiver Characteristics Characteristics Total Intervention Control p-value (n=95) (n=63) (n=32) Gendera,b (% female) 93 (98%) 63 (99%) 31 (97%) 0.605 Age (years)a,b (n=92)c (n=62)c (n=30)c 18-30 10 (11%) 7 (11%) 3 (10%) 0.184 31-40 46 (50%) 27 (44%) 19 (63%) 41+ 36 (39%) 28 (45%) 8 (27%) Dietary Intaked,e,f,g (n=86)c (n=57)c (n=29)c Total Fruit (svgs/day)h 1.20 (1.2) 1.16 (1.2) 1.27 (1.1) 0.668 Whole Fruit (svgs/day) 0.92 (0.9) 0.84 (0.9) 1.06 (1.0) 0.310 Total Vegetable (svgs/day) 2.96 (2.1) 2.97 (2.0) 2.94 (2.2) 0.951 Total Sugar-Sweetened Beverages (svgs/day) 1.52 (9.2) 1.99 (11.3) 0.59 (1.2) 0.511 182 Total Energy (kcals/day) 1937.24 (804.8) 1896.58 (733.5) 2017.15 (938.1) 0.515 Healthy Eating Indexi 55.72 (12.4) 56.75 (11.8) 53.73 (13.5) 0.290 Anthropometrics/Blood Pressuref,g (n=95)c (n=63)c (n=32)c BMI (kg/m2) 33.0 (10.3) 31.95 (9.6) 35.1 (11.5) 0.158 Waist Circumference (cm) 102.2 (7.7) 101.1 (8.1) 104.2 (7.2) 0.492 Systolic BP (mm Hg) 128.3 (11.9) 127.9 (12.3) 129.2 (9.7) 0.718 Diastolic BP (mm Hg) 81.7 (11.1) 81.1 (10.7) 82.9 (12.0) 0.458 Personal Determinantsf,g (n=95)c (n=63)c (n=32)c Self-efficacy for healthy dietary behaviorsj 94.46 (17.2) 93.26 (16.9) 96.94 (17.7) 0.324 Menu Planningk 24.50 (3.4) 24.35 (3.4) 24.81 (3.4) 0.541 Table 21. (continued)

182

Table 21. Baseline Caregiver Characteristics (continued) Racea,b (n=95)c (n=63)c (n=32)c Black 59 (62%) 39 (62%) 20 (62.5%) 0.051 White 27 (29%) 15 (24%) 12 (37.5%) Otherl 9 (9%) 9 (14%) 0 (0%) Household Income Statusa,b,m (n=91)c (n=62)c (n=29)c Low-Income 36 (40%) 24 (39%) 12 (41%) 0.493 Non-Low-Income 55 (60%) 38 (61%) 17 (59%) Home Food Securitya,b,n (n=95)c (n=63)c (n=32)c High/Marginal Food Security 60 (63%) 35 (56%) 25 (78%) 0.083 Low Food Security 19 (20%) 16 (25%) 3 (9%) Very Low Food Security 16 (17%) 12 (19%) 4 (13%) a

183 Values are n (%) bChi-square eParticipant response rate is outcome variable-dependent, and therefore sample sizes are provided for each outcome variable dDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] eParticipant (n=95) completion of 24hr dietary recalls: 0 recalls (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)) fValues are mean (SD) gOne-Way ANOVA hTotal fruit= whole fruit + 100% fruit juice iHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] jSelf-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased skills kMenu Planning, 0-36 scale with higher score representing increased frequency lIncludes participants who did not identify with either Black or White mLow-income defined as participation in one or more of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) nUSDA 6-item Short Form Home Food Security Questionnaire (Scoring: High/Marginal food security= 0-1; Low food security= 2-4; Very low food security= 5-6) [219]

183 Intervention Effects on Caregiver Diet Quality

During the intervention period, no significant differences were observed in caregiver daily intake of total fruit, whole fruit, vegetables or sugar-sweetened beverages by group assignment (Tables 22 and 23.) or within group (Table 24.), as well as by level of attendance (Table 25 and 26.). Similarly, daily total energy intake and overall diet quality, assessed using HEI, did not differ at post-test or 10-week follow-up by group assignment or level of attendance.

Intervention Effects on Caregiver Anthropometric and Blood Pressure Outcomes

A significant decrease in caregiver BMI was observed at post-test among intervention caregivers relative to controls (Table 23. β (SE)= -0.84 (0.4), p=0.028), which was maintained at 10-week follow-up (Table 27 and 28. p=0.445).

At post-test and 10-week follow-up, significant changes in caregiver WC were not observed by group assignment or level of attendance. While changes in WC were not observed, caregiver systolic BP significantly decreased among intervention caregivers relative to controls at post-test (Table 23. β (SE)= -5.67 (2.8), p=0.043). Improvements in systolic BP at post-test among intervention participants were maintained at the 10- week follow-up. No change in diastolic BP was observed during the intervention or follow-up period by group assignment or level of attendance.

Intervention Effects on Caregiver Personal Determinants

184 A significant increase in caregiver SE for healthy dietary behaviors was observed among intervention caregivers relative to controls at post-test (Table 23. β (SE)= 6.99

(2.7), p=0.012), which was maintained at the 10-week follow-up. While an increase in menu planning was observed among intervention caregivers at post-test, this increase was not statistically significant (Table 23. β (SE)= 0.82 (0.9); p=0.365).

Intervention Effects on Family Outcomes

At post-test, no changes in weekly family breakfast or dinner frequency were observed by group assignment or level of attendance. Similarly, frequency of family meals eaten in a dining area did not significantly differ by group assignment or level of attendance at post-test or 10-week follow-up. However, a significant decrease in mealtime TV viewing was observed among intervention families at post-test, relative to control families, at post-test (Table 23. β (SE)= -1.06 (0.5), p=0.028). Improvements in having family meals without TV were maintained among intervention families at the 10- week follow-up (Tables 27 and 28. p=0.626).

Program Feasibility (program dose and fidelity) and Acceptability

The Simple Suppers program retention rate was 87.5% and intervention caregivers/families attended on average 71% of the program lessons. Approximately

95% of lessons were delivered as intended and caregiver participants were engaged in the program 96% of the time. At program completion (T1), approximately 94% of the participating caregivers reported being satisfied or very satisfied with the program.

185 Table 22. Intervention Period: By-Group Change in Caregiver and Family Outcomes from Baseline T0 to Post-Test T1 Outcomes Baseline T0 Post-Test T1 Mean (SD) Change from Baseline T0 to Post-Test T1 Control Intervention Control Intervention Control Intervention (n=32) (n=63) (n=32) (n=63) (n=32) (n=63) Dietary Intakea,b,c,d Fruit intake (svgs/d) 1.27 (1.1) 1.16 (1.2) 1.55 (1.8) 1.45 (2.3) 0.28 (1.9) 0.29 (2.6) Whole fruit intake (svgs/d)e 1.06 (1.0) 0.84 (0.9) 1.35 (1.7) 1.25 (2.1) 0.29 (1.9) 0.41 (2.3) Vegetable intake (svgs/d) 2.94 (2.2) 2.97 (2.0) 3.21 (2.7) 3.12 (2.1) 0.27 (3.0) 0.15 (2.5) Sugar sweetened beverage 0.59 (1.2) 1.99 (11.3) 0.47 (0.9) 0.51 (0.8) -0.12 (1.5) -1.48 (0.89) intake (svgs/d)

186 Energy intake (kcals/d) 2017.15 (938.1) 1896.58 (733.5) 2057.41 (968.3) 1642.41 (688.1) 40.26 (144.9) -254.47 (223.5) HEI total scoref 53.73 (13.5) 56.75 (11.8) 56.29 (11.2) 56.23 (15.0) 2.56 (12.7) -0.52 (17.4) HEI total fruit scoreg 1.91 (1.8) 1.73 (1.7) 2.26 (2.3) 1.84 (2.1) 0.35 (1.4) 0.11 (1.2) HEI whole fruit scoreh 2.61 (2.2) 2.28 (2.1) 3.21 (4.3) 2.64 (3.8) 0.60 (2.9) 0.36 (2.8) HEI total vegetable scorei 2.77 (1.6) 3.11 (1.6) 3.06 (1.7) 3.46 (1.5) 0.29 (1.6) 0.35 (1.4) Anthropometric/BPa,b BMI (kg/m2) 35.18 (11.5) 31.95 (9.6) 35.45 (11.8) 31.75 0.27 (1.6)m -0.20 (1.5)n WC (cm) 104.22 (7.2) 101.12 (8.1) 103.26 (22.9) 100.85 -0.96 (7.8) -0.94 (12.8) Systolic BP (mm Hg) 129.24 (9.7) 127.91 (12.3) 131.14 (16.4) 126.10 1.98 (12.6)m -2.23 (15.0)n Diastolic BP (mm Hg) 82.90 (12.0) 81.14 (10.7) 83.41 (12.9) 80.35 0.45 (11.9) -0.47 (10.1) Personal Determinantsa,b Self-efficacyj 96.94 (17.7) 93.26 (16.9) 94.76 (16.7) 101.43 (13.2) -2.18 (18.9)m 8.17 (14.6)n Menu planningk 24.81 (3.4) 24.35 (3.4) 25.27 (4.9) 26.69 (4.2) 0.45 (4.2) 2.38 (4.1) Family Meals (meals/wk)a,l Family breakfast frequency 2.73 (2.3) 2.72 (2.3) 2.79 (2.0) 2.88 (2.4) 0.06 (3.4) 0.16 (2.9) Family dinner frequency 5.17 (1.5) 4.73 (2.1) 4.02 (2.0) 4.39 (2.1) -1.15 (2.4) -0.39 (2.8) Family meals in a dining area 3.91 (2.6) 4.11 (2.4) 4.09 (2.5) 4.37 (2.2) 0.19 (2.7) 0.26 (2.4) Family meals with TV 2.50 (2.2) 1.92 (1.8) 2.94 (2.5) 2.08 (2.2) 0.44 (2.2)m 0.16 (2.4)n

Table 22. (continued)

186 Table 22. Intervention Period: By-Group Caregiver and Family Outcomes at Baseline T0 and Post-Test T1 (continued) aValues are Mean (SD) bBy-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, age; group assignment; baseline value; group assignment (Y (outcome variable)= outcome at T1) cDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dParticipant (n=95) completion of 24hr dietary recalls: 0 recalls (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)) eTotal fruit= whole fruit + 100% fruit juice fHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] gHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake hHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake jSelf-Efficacy for Healthy Dietary Behaviors, 0-120 scale with higher score representing increased skills kMenu Planning, 0-36 scale with higher score representing increased menu planning l

187 By-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, and age; oldest child race, sex, and age; baseline value; group assignment (Y (outcome variable)= outcome at T1)

m,nDifferent superscripts indicate by-group difference in participant outcome of p<0.05.

187 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 Intervention Period (T0 à T1) Outcomes Covariates β (SE) p-value CI Caregiver Dietary Intakea,b,c Fruit intake (svgs/d) Group Assignmentd -Intervention -0.45 (0.9) 0.623 (-2.27, 1.38) -Control 0 - - Cohorte -One -0.07 (1.0) 0.948 (-2.19, 2.05) -Two -0.52 (1.2) 0.665 (-2.93, 1.89) -Three 0 - - Incomef -Low -1.00 (0.8) 0.208 (-2.59, 0.59) -Non-Low 0 - - Caregiver Raceg

188 -Other 0.06 (1.3) 0.962 (-2.65, 2.77) -Black -0.64 (0.9) 0.503 (-2.55, 1.28) -White 0 - - Caregiver Sexh -Male -1.95 (2.7) 0.475 (-7.45, 3.54) -Female 0 - - Caregiver Agei -18-30 yrs 0.51 (2.6) 0.847 (-4.78, 5.79) -31-40 yrs 1.21 (0.8) 0.128 (-0.37, 2.79) -41+ yrs 0 - - Fruit intake T0 0.31 (0.4) 0.383 (-0.40, 1.02) Whole fruit intake (svgs/d) Group Assignmentd -Intervention -0.39 (0.8) 0.636 (-2.06, 1.28) -Control 0 - - Table 23. (continued)

188 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One 0.14 (0.9) 0.884 (-1.81, 2.09) -Two -0.65 (1.1) 0.557 (-2.89, 1.58) -Three 0 - - Incomef -Low -0.85 (0.7) 0.239 (-2.29, 0.59) -Non-Low 0 - - Caregiver Raceg -Other -1.14 (1.2) 0.910 (-2.62, 2.35) -Black -0.71 (0.8) 0.409 (-2.43, 1.01) -White 0 - - Caregiver Sexh -Male -1.19 (2.4) 0.617 (-5.98, 3.60)

189 -Female 0 - - Caregiver Agei -18-30 yrs 0.88 (2.4) 0.716 (-3.98, 5.73) -31-40 yrs 1.39 (0.7) 0.061 (-0.07, 2.85) -41+ yrs 0 - - Whole fruit intake T0 0.33 (0.4) 0.379 (-0.43, 1.09) Vegetable intake (svgs/d) Group Assignmentd -Intervention -0.76 (0.9) 0.388 (-2.51, 1.00) -Control 0 - - Cohorte -One -0.54 (0.9) 0.595 (-2.56, 1.49) -Two 0.32 (1.1) 0.780 (-2.00, 2.66) -Three 0 - - Incomef -Low -1.14 (0.7) 0.128 (-2.63, 0.35) -Non-Low 0 - - Table 23. (continued)

189

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other -0.83 (1.3) 0.530 (-3.48, 1.83) -Black -0.99 (0.9) 0.261 (-2.75, -0.77) -White 0 - - Caregiver Sexh -Male 4.96 (2.5) 0.052 (-0.04, 9.96) -Female 0 - - Caregiver Agei -18-30 yrs -2.57 (2.5) 0.216 (-7.69, 2.56) -31-40 yrs 0.38 (0.8) 0.624 (-1.16, 1.91) -41+ yrs 0 - - Veg intake T0 0.25 (0.2) 0.216 (-0.15, 0.65) Sugar sweetened beverage intake Group Assignmentd

190 (svgs/d) -Intervention -0.39 (0.3) 0.179 (-0.96, 0.19) -Control 0 - - Cohorte -One -0.56 (0.3) 0.094 (-1.23, 0.10) -Two -0.05 (0.4) 0.895 (-0.82, 0.72) -Three 0 - - Incomef -Low 0.44 (0.2) 0.080 (-0.06, 0.94) -Non-Low 0 - - Caregiver Raceg -Other 0.29 (0.4) 0.495 (-0.56, 0.94) -Black -0.21 (0.3) 0.467 (-0.81, 0.38) -White 0 - - Caregiver Sexh -Male 4.7 (1.2) 0.001 (2.24, 7.30) -Female 0 - - Table 23. (continued)

190 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Agei -18-30 yrs 0.60 (0.8) 0.466 (-1.06, 2.26) -31-40 yrs 0.31 (0.2) 0.228 (-0.20, 0.81) -41+ yrs 0 - - SSB intake T0 0.30 (0.2) 0.085 (-0.04, 0.64) Energy intake (kcals/d) Group Assignmentd -Intervention -336.99 (281.7) 0.240 (-909.40, 235.40) -Control 0 - - Cohorte -One 368.93 (331.9) 0.274 (-305.65, 1,043.50) -Two -108.52 (362.9) 0.767 (-846.19, 629.14) -Three 0 - - Incomef

191 -Low 44.26 (240.9) 0.855 (-445.23, 533.76) -Non-Low 0 - - Caregiver Raceg -Other 51.56 (407.7) 0.900 (-776.91, 880.04) -Black 485.65 (282.1) 0.094 (-87.55, 1,058.85) -White 0 - - Caregiver Sexh -Male 227.59 (803.2) 0.779 (-1,404.75, 1,859.94) -Female 0 - - Caregiver Agei -18-30 yrs 390.93 (802.7) 0.629 (-1,240.31, 2,022.18) -31-40 yrs 116.73 (237.4) 0.626 (-365.74, 599.21) -41+ yrs - - - Energy intake T0 0.15 (0.2) 0.506 (-0.30, 0.59) HEI scorej Group Assignmentd -Intervention -6.54 (5.4) 0.238 (-17.61, 4.53) -Control 0 - - Table 23. (continued)

191 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -1.53 (6.4) 0.812 (-14.49, 11.43) -Two -0.28 (7.3) 0.970 (-15.08, 14.52) -Three 0 - - Incomef -Low -4.41 (4.8) 0.368 (-14.23, 5.41) -Non-Low 0 - - Caregiver Raceg -Other 2.29 (8.1) 0.780 (-14.22, 18.81) -Black -4.34 (5.7) 0.449 (-15.86, 7.17) -White 0 - - Caregiver Sexh -Male -6.33 (16.0) 0.695 (-38.89, 26.23)

192 -Female 0 - - Caregiver Agei -18-30 yrs -31.53 (16.5) 0.065 (-65.08, 2.03) -31-40 yrs 3.48 (4.7) 0.467 (-6.13, 13.09) -41+ yrs 0 - - HEI Score T0 0.15 (0.3) 0.552 (-0.37, 0.68) Caregiver Anthropometrics/Blood Pressurea BMI Group Assignmentd -Intervention -0.84 (0.4) 0.028 (-1.58, -0.10) -Control 0 - - Cohorte -One -0.28 (0.4) 0.474 (-1.06, 0.49) -Two 0.07 (0.4) 0.864 (-0.77, 0.92) -Three 0 - - Incomef -Low -0.93 (0.3) 0.008 (-1.62, -0.25) -Non-Low 0 - - Table 23. (continued)

192

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other 0.41 (0.6) 0.525 (-0.87, 1.68) -Black 0.87 (0.4) 0.021 (0.14, 1.60) -White 0 - - Caregiver Sexh -Male -0.60 (1.5) 0.697 (-3.63, 2.44) -Female 0 - - Caregiver Agei -18-30 yrs 0.23 (0.6) 0.684 (-0.91, 1.38) -31-40 yrs -0.33 (0.4) 0.362 (-1.04, 0.39) -41+ yrs 0 - - BMI T0 0.99 <0.001 (0.96, 1.03) Waist Circumference Group Assignmentd 193 -Intervention -1.04 (2.8) 0.707 (-6.55, 4.46) -Control 0 - -

Cohorte -One -5.43 (2.9) 0.065 (-11.21, 0.35) -Two 0.08 (3.1) 0.981 (-6.18, 6.33) -Three 0 - - Incomef -Low -8.59 (2.6) 0.001 (-13.76, -3.43) -Non-Low 0 - - Caregiver Raceg -Other 4.58 (4.7) 0.330 (-4.73, 13.88) -Black 3.38 (2.7) 0.220 (-2.07, 8.82) -White 0 - - Caregiver Sexh -Male -2.33 (11.2) 0.836 (-24.71, 20.05) -Female 0 - - Table 23. (continued)

193 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Agei -18-30 yrs 2.17 (4.3) 0.612 (-6.33, 10.68) -31-40 yrs -2.21 (2.7) 0.420 (-7.62, 3.21) -41+ yrs 0 - - WC T0 0.92 (0.1) <0.001 (0.79, 1.05) Systolic BP Group Assignmentd -Intervention -5.67 (2.8) 0.043 (-11.14, -0.19) -Control 0 - - Cohorte -One -4.22 (2.9) 0.147 (-9.95, 1.52) -Two -4.52 (3.2) 0.161 (-10.88, 1.84) -Three 0 - - Incomef

194 -Low 2.93 (2.6) 0.267 (-2.29, 8.15) -Non-Low 0 - - Caregiver Raceg -Other 5.29 (4.7) 0.261 (-4.01, 14.60) -Black 4.29 (2.7) 0.113 (-1.04, 9.61) -White 0 - - Caregiver Sexh -Male -2.69 (11.4) 0.815 (-25.48, 20.10) -Female 0 0 - Caregiver Agei -18-30 yrs -16.71 (4.3) <0.001 (-25.28, -8.14) -31-40 yrs -8.61 (2.6) 0.002 (-13.88, -3.33) -41+ yrs 0 0 0 SysBP T0 0.46 (0.1) <0.001 (0.30, 0.61) Diastolic BP Group Assignmentd -Intervention 0.39 (2.2) 0.863 (-4.03, 4.80) -Control 0 - - Table 23. (continued)

194 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -2.18 (2.3) 0.349 (-6.79, 2.43) Two -4.29 (2.6) 0.098 (-9.38, 0.80) -Three 0 - - Incomef -Low 1.21 (2.1) 0.557 (-2.89, 5.32) -Non-Low 0 - - Caregiver Raceg -Other -1.64 (3.8) 0.665 (-9.14, 5.86) -Black 0.95 (2.2) 0.663 (-3.35, 5.24) -White 0 - - Caregiver Sexh -Male -6.92 (9.2) 0.456 (-25.33, 11.49)

195 -Female 0 - - Caregiver Agei -18-30 yrs -2.34 (3.5) 0.502 (-9.26, 4.58) -31-40 yrs 0.08 (2.1) 0.970 (-4.16, 4.32) -41+ yrs 0 - - DysBP T0 0.44 (0.1) <0.001 (0.28, 0.61) Caregiver Personal Determinants Self-efficacy for Healthy Dietary Group Assignmentd Behaviorsk -Intervention 6.99 (2.7) 0.012 (1.58, 12.42) -Control 0 0 - Cohorte -One 1.41 (2.9) 0.622 (-4.27, 7.09) -Two -0.40 (3.2) 0.901 (-6.71, 5.91) -Three 0 - - Incomef -Low -4.47 (2.4) 0.071 (-9.33, 0.39) -Non-Low 0 - - Table 23. (continued)

195

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other 8.57 (4.6) 0.067 (-0.63, 17.76) -Black 5.40 (2.6) 0.044 (0.15, 10.65) -White 0 - - Caregiver Sexh -Male 8.31 (11.3) 0.465 (-14.22, 30.83) -Female 0 0 0 Caregiver Agei -18-30 yrs -4.24 (4.3) 0.323 (-12.71, 4.24) -31-40 yrs 1.61 (2.6) 0.537 (-3.56, 6.78) -41+ yrs 0 0 - Self-efficacy T0 0.45 (0.1) <0.001 (0.31, 0.58) Menu Planningl Group Assignmentd

1 -Intervention 0.82 (0.9) 0.365 (-0.97, 2.60) 96 -Control 0 - -

Cohorte -One -0.58 (0.9) 0.539 (-2.44, 1.28) -Two -1.67 (1.0) 0.110 (-3.73, 0.39) -Three 0 - - Incomef -Low 1.74 (0.8) 0.034 (0.13, 3.34) -Non-Low 0 - - Caregiver Raceg -Other 2.67 (1.5) 0.084 (-0.36, 5.70) -Black -0.71 (0.9) 0.428 (-2.48, 1.06) -White 0 - - CaregiverSexh -Male 3.99 (3.7) 0.287 (-3.43, 11.42) -Female 0 - - Table 23. (continued)

196 Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Agei -18-30 yrs -2.39 (1.4) 0.094 (-5.21, 0.42) -31-40 yrs 0.31 (0.9) 0.718 (-1.39, 2.02) -41+ yrs - - - Menu planning T0 0.60 (0.1) <0.001 (0.38, 0.83) Family Mealsm Family breakfast frequency (meals/wk) Group Assignmentd -Intervention 0.12 (0.6) 0.847 (-1.10, 1.33) -Control 0 - - Cohorte -One 0.26 (0.7) 0.700 (-1.08, 1.60) -Two -1.15 (0.7) 0.103 (-2.54, 0.24) 197 -Three 0 - - f Income -Low 0.72 (0.6) 0.213 (-0.42, 1.85) -Non-Low 0 - - Caregiver Raceg -Other 1.16 (1.2) 0.216 (-0.93, 4.05) -Black 0.48 (1.2) 0.686 (-1.89, 2.85) -White 0 - - Caregiver Sexh -Male 1.32 (2.7) 0.629 (-4.13, 6.78) -Female 0 - - Caregiver Agei -18-30 yrs 0.35 (1.0) 0.712 (-1.54, 2.25) -31-40 yrs 0.23 (0.6) 0.713 (-1.00, 1.45) -41+ yrs 0 - - Oldest Child Racen -Other -2.69 (1.0) 0.013 (-4.81, -0.58) -Black -1.71 (1.3) 0.187 (-4.26, 0.85) -White 0 - - Table 23. (continued)

197

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Oldest Child Sexo -Male 0.11 (0.6) 0.858 (-1.11, 1.33) -Female 0 - - Oldest Child Agep -0.06 (0.1) 0.670 (-0.35, 0.23) Family Breakfast Freq, T0 0.18 (0.1) 0.145 (-0.06, 0.42) Family dinner frequency (meals/wk) Group Assignmentd -Intervention 0.48 (0.5) 0.354 (-0.55, 1.51) -Control 0 - - Cohorte -One 0.50 (0.6) 0.371 (-0.61, 1.62) -Two 0.92 (0.6) 0.131 (-0.28, 2.11) -Three 0 - - Incomef -Low -0.16 (0.5) 0.737 (-1.11, 0.79)

198 -Non-Low 0 - - Caregiver Raceg -Other 2.37 (1.04) 0.026 (0.29, 4.45) -Black 1.20 (0.9) 0.233 (-0.78, 3.17) -White 0 - - Caregiver Sexh -Male -1.33 (2.3) 0.562 (-5.88, 3.22) -Female 0 - - Caregiver Agei -18-30 yrs -1.78 (0.8) 0.033 (-3.41, -0.15) -31-40 yrs -0.54 (0.5) 0.292 (-1.54, 0.47) -41+ yrs 0 - - Oldest Child Racen -Other -0.87 (0.9) 0.328 (-2.63, 0.89) -Black -0.71 (1.1) 0.506 (-2.84, 1.41) -White 0 - - Table 23. (continued)

198

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Oldest Child Sexo -Male -0.60 (0.5) 0.246 (-1.62, 0.42) -Female 0 - - Oldest Child Agep -0.08 (0.12) 0.515 (-0.32, 0.16) Family Dinner Freq, T0 0.10 (0.1) 0.425 (-0.14, 0.34) Family meals in a dining area Group Assignmentd (meals/wk) -Intervention -0.10 (0.6) 0.857 (-1.23, 1.03) -Control 0 - - Cohorte -One -0.43 (0.6) 0.494 (-1.67, 0.81) -Two -0.06 (0.6) 0.925 (-1.36, 1.23) -Three 0 - - Incomef -Low 0.74 (0.5) 0.168 (-0.32, 1.79)

199 -Non-Low 0 - - Caregiver Raceg -Other -0.45 (1.2) 0.703 (-2.76, 1.87) -Black -1.54 (1.1) 0.170 (-3.76, 0.68) -White 0 - - Caregiver Sexh -Male -0.20 (2.5) 0.938 (-5.28, 4.89) -Female 0 - - Caregiver Agei -18-30 yrs 0.31 (0.9) 0.730 (-1.46, 2.08) -31-40 yrs -0.38 (0.6) 0.496 (-1.49, 0.73) -41+ yrs 0 - - Oldest Child Racen -Other 0.26 (1.0) 0.791 (-1.70, 2.23) -Black 1.41 (1.2) 0.242 (-0.97, 3.78) -White 0 - - Table 23. (continued)

199

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Oldest Child Sexo -Male 0.30 (0.6) 0.595 (-0.83, 1.43) -Female 0 - - Oldest Child Agep -0.12 (0.1) 0.387 (-0.39, 0.15) Dining Area Freq, T0 0.38 (0.1) 0.001 (0.16, 0.59) Family meal TV viewing (meals/wk) Group Assignmentd -Intervention -1.06 (0.5) 0.028 (-2.01, -0.12) -Control 0 - - Cohorte -One -0.16 (0.5) 0.760 (-1.17, 0.86) -Two 0.01 (0.5) 0.998 (-1.05, 1.06) -Three 0 - - Incomef -Low 0.73 (0.4) 0.095 (-0.13, 1.59)

200 -Non-Low 0 - - Caregiver Raceg -Other -3.40 (0.9) 0.001 (-5.29. -1.52) -Black -5.01 (0.9) <0.001 (-6.80, -3.21) -White 0 - - Caregiver Sexh -Male 1.17 (2.1) 0.572 (-2.95, 5.29) -Female 0 - - Caregiver Agei -18-30 yrs -0.38 (0.7) 0.597 (-1.82, 1.05) -31-40 yrs -0.43 (0.5) 0.348 (-1.35, 0.48) -41+ yrs 0 - - Oldest Child Racen -Other 3.87 (0.8) <0.001 (2.27, 5.47) -Black 5.14 (1.0) <0.001 (3.21, 7.07) -White 0 - - Table 23. (continued)

200

Table 23. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Oldest Child Sexn -Male -0.04 (0.5) 0.924 (-0.96, 0.87) -Female 0 - - Oldest Child Agep -0.15 (0.1) 0.183 (-0.37, 0.07) No Mealtime TV Freq, T0 0.55 (0.1) <0.001 (0.34, 0.76) aBy-group differences determined by generalized linear models controlling for: cohort; household income; caregiver sex, race, and age; baseline value; group assignment (Y (outcome variable)= outcome at T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] cParticipant (n=95) completion of 24hr dietary recalls at T0: (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)). Participant completion of 24hr dietary recalls at T1: 0 recalls (n=33 (35%)); 1 recall (n=56 (59%)); 2 recalls (n=6 (6%)); 3 recalls (n=0 (0%)) dGroup assignment: control group (n=32); intervention group (n=63) eCohort: one (n= 35); two (n=25); three (n=35) fIncome: low (n= 36); non-low (n=55). Low-income defined as participation in >1 of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National 201 School Lunch Program (NSLP) g Caregiver Race: other (n= 9); Black (n= 59); White (n=27). Other category includes participants who did not identify with Black or White hCaregiver Sex: male (n=2); female (n= 93) iCaregiver Age: 18-30 yrs (n=10); 31-40 yrs (n=46); 41+ yrs (n=36) jHealthy Eating Index, 0-100 scale, Higher score represents better diet quality [198] kSelf-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased self-efficacy lMenu planning behaviors, 0-36 scale with higher score representing increased menu planning mBy-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, and age; oldest child race, sex, and age; baseline value; group assignment (Y (outcome variable)= outcome at T1) nOldest Child Race: other (n=18); Black (n=54); White (n=23). Other category includes participants who did not identify with either Black or White oOldest Child Sex: male (n=35); female (n=60) pChild (oldest) age: mean (SD)= 7.4 (1.9) yrs

201 Table 24. Intervention Period: Within-Group Caregiver and Family Outcomes from Baseline T0 to Post-Test T1 Outcomes Control (n=32) Intervention (n=63) Baseline T0 Post-Test T1 Baseline T0 Post-Test T1 Dietary Intakea,b,c,d Fruit intake (svgs/d)e 1.27 (1.1) 1.55 (1.8) 1.16 (1.2) 1.45 (2.3) Whole fruit intake (svgs/d) 1.06 (1.0) 1.35 (1.7) 0.84 (0.9) 1.25 (2.1) Vegetable intake (svgs/d) 2.94 (2.2) 3.21 (2.7) 2.97 (2.0) 3.12 (2.1) Sugar sweetened beverage intake 0.59 (1.2) 0.47 (0.9) 1.99 (11.3) 0.51 (0.8) (svgs/d) Energy intake (kcals/d) 2017.15 (938.1) 2057.41 (968.3) 1896.58 (733.5)l 1642.41 (688.1)m HEI total scoref 53.73 (13.5) 56.29 (11.2) 56.75 (11.8) 56.23 (15.0) 202 HEI total fruit scoreg 1.91 (1.8) 2.26 (2.3) 1.73 (1.7) 1.84 (2.1)

HEI whole fruit scoreh 2.61 (2.2) 3.21 (4.3) 2.28 (2.1) 2.64 (3.8) HEI total vegetable scorei 2.77 (1.6) 3.06 (1.7) 3.11 (1.6) 3.46 (1.5) Anthropometric/BPa,b BMI (kg/m2) 35.18 (11.5) 35.45 (11.8) 31.95 (9.6) 31.75 WC (cm) 104.22 (7.2) 103.26 (22.9) 101.12 (8.1) 100.85 Systolic BP (mm Hg) 129.24 (9.7) 131.14 (16.4) 127.91 (12.3) 126.10 Diastolic BP (mm Hg) 82.90 (12.0) 83.41 (12.9) 81.14 (10.7) 80.35 Behavioral Capability/Cognitivea,b Self-efficacyj 96.94 (17.7) 94.76 (16.7) 93.26 (16.9)l 101.43 (13.2)m Menu planningk 24.81 (3.4) 25.27 (4.9) 24.35 (3.4)l 26.69 (4.2)m Family Meals (meals/wk)a,l Family breakfast frequency 2.73 (2.3) 2.79 (2.0) 2.72 (2.3) 2.88 (2.4) Family dinner frequency 5.17 (1.5) 4.02 (2.0) 4.73 (2.1)l 4.39 (2.1)m Family meals in a dining area 3.91 (2.6) 4.09 (2.5) 4.11 (2.4) 4.37 (2.2) Family meals with TV 2.50 (2.2) 2.94 (2.5) 1.92 (1.8) 2.08 (2.2) Table 24. (continued)

202 Table 24. Intervention Period: Within-Group Caregiver and Family Outcomes from Baseline T0 to Post-Test T1 (continued) aValues are Mean (SD) bWithin-group differences determined by paired t-test cDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dParticipant (n=95) completion of 24hr dietary recalls: 0 recalls (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)) eTotal fruit= whole fruit + 100% fruit juice fHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] gHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake hHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake jSelf-Efficacy for Healthy Dietary Behaviors, 0-120 scale with higher score representing increased skills kMenu Planning, 0-36 scale with higher score representing increased menu planning l,mDifferent superscripts indicate within-group difference in participant outcome of p<0.05.

203

203 Table 25. Intervention Period: By-Group (Level of Attendance) Caregiver and Family Outcomes Baseline to Post-Test T1 Outcomes Baseline T0 Post-Test T1 Mean (SD) Change from Baseline T0 to Post-Test T1 Control Low Att.a High Att.a Control Low Att.a High Att.a Control Low Att.a High Att.a (n=32) (n=23) (n=40) (n=32) (n=23) (n=32) (n=32) (n=23) (n=40) Dietary Intake,b,c,d,e Fruit intake (svgs/d)f 1.27 (1.1) 0.74 (1.0) 1.46 (1.3) 1.55 (1.8) 1.59 (2.4) 1.42 (2.3) 0.28 0.85 (2.5) -0.04 (2.5) (2.6) Whole fruit intake (svgs/d) 1.06 (1.0) 0.53 (0.8) 1.06 (0.9) 1.35 (1.7) 1.35 (2.4) 1.22 (2.1) 0.29 0.85 (2.1) 0.16 (2.1) (2.3) Vegetable intake (svgs/d) 2.94 (2.2) 3.09 (2.3) 2.89 (1.8) 3.21 (2.7) 2.58 (1.8) 3.28 (2.2) 0.27 -0.51 (2.3) 0.39 (2.2) (2.5) 204 Sugar sweetened beverage 0.59 (1.2) 0.57 (0.9) 1.02 (1.8) 0.47 (0.9) 0.38 (1.0) 0.54 (1.2) -0.12 -0.19 (0.8) -0.48 (0.8)

intake (svgs/d) (0.89) Energy intake (kcals/d) 2017.15 1828.61 1946.01 2057.41 1532.63 1673.24 40.26 -155.37 -272.77 (938.1) (844.28) (650.6) (968.3) (322.8) (760.4) (223.5) (790.4) (778.6) HEI total scoreg 53.73 55.16 57.93 (10.5) 56.29 (11.2) 59.89 55.22 2.54 4.73 -2.71 (13.5) (13.3) (13.5) (15.2) (17.4) (15.5) (17.0) HEI total fruit scoreh 2.61 (2.2) 1.21 (1.6) 2.11 (1.69) 3.21 (4.3) 1.97 (2.3) 1.80 (4.2) 0.60 0.76 (2.1) -0.31 (2.4) (2.8) HEI whole fruit scorei 1.91 (1.8) 1.49 (2.0) 2.85 (2.0) 2.26 (2.3) 2.16 (2.4) 2.78 (2.1) 0.35 0.67 (4.1) -0.07 (3.2) (1.2) HEI total vegetable scorej 2.77 (1.6) 3.33 (1.7) 2.95 (1.4) 3.06 (1.7) 3.28 (1.2) 3.52 (1.7) 0.29 -0.05 (1.6) 0.57 (1.1) (1.4) Anthropometric/BPb,c BMI (kg/m2) 35.18 31.89 (9.1) 32.12 (9.8) 35.45 (11.8) 31.52 31.87 0.27 -0.37 (1.1) -0.25 (11.5) (9.1) (9.6) (1.5)m (1.6)n WC (cm) 104.22 102.38 101.61 103.26 (22.9) 101.20 100.65 -0.96 -1.18 (9.4) -0.96 (7.2) (18.9) (19.9) (20.6) (20.6) (12.8) (13.8) Table 25. (continued)

204 Table 25. Intervention Period: By-Group (Level of Attendance) Caregiver and Family Outcomes Baseline to Post-Test T1 (continued) Systolic BP (mm Hg) 129.24 124.72 130.55 131.14 (16.4) 121.61 128.66 1.90 -3.11 -1.89 (9.7) (13.9) (17.4) (12.3) (16.6) (15.0)m (14.3)n (16.1) Diastolic BP (mm Hg) 82.90 79.61 (8.2) 81.78 (12.4) 83.41 (12.9) 78.73 81.27 0.51 -0.88- -0.51 (12.0) (8.9) (12.2) (10.1) (8.3) (11.4) Personal Determinantsb,c Self-efficacyk 96.94 94.79 92.39 (16.3) 94.76 (16.7) 103.02 100.51 -2.18 8.23 8.21 (17.7) (18.2) (11.5) (14.2) (14.6)m (16.6)n (13.3) l

205 Menu planning 24.81 (3.4) 24.32 (3.4) 24.37 (3.5) 25.27 (4.9) 26.55 26.77 0.46 (4.1) 2.23 2.40 (3.8) (4.9) (3.8) (4.5) Family Meals (meals/wk)b,c Family breakfast frequency 2.73 (2.3) 2.25 (2.4) 2.98 (2.8) 2.79 (2.0) 2.98 (2.6) 2.82 (2.2) 0.06 (2.9) 0.73 -0.16 (2.7) (2.1) Family dinner frequency 5.17 (1.5) 4.87 (2.2) 4.65 (2.0) 4.02 (2.0) 3.67 (2.2) 4.82 (1.9) -1.15 (2.8) -1.20 0.17 (2.6) (1.9) Family meals in a dining area 3.91 (2.6) 3.71 (2.3) 4.34 (2.4) 4.09 (2.5) 4.75 (2.3) 4.15 (2.1) 0.18 (2.4) 1.04 -0.19 (2.2) (2.2) Family meals with TV 2.50 (2.2) 1.71 (1.8) 2.04 (1.8) 2.94 (2.5) 2.68 (2.8) 1.72 (1.7) 0.44 0.97 -0.32 (2.4)m (1.9) (1.9)n aLevel of attendance based on total number of lessons attended bValues are Mean (SD) cBy-group differences determined by paired t-test dDaily intake averaged across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] eParticipant (n=95) completion of 24hr dietary recalls: 0 recalls (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)) fTotal fruit= whole fruit + 100% fruit juice gHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] hHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake jHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake kSelf-Efficacy for Healthy Dietary Behaviors, 0-120 scale with higher score representing increased skills lMenu Planning, 0-36 scale with higher score representing increased menu planning m,nDifferent superscripts indicate within-group difference in participant outcome of p<0.05.

205 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 Intervention Period (T0 à T1) Outcomes Covariates β (SE) p-value Confidence Interval β (SE) p-value Confidence Interval Caregiver Dietary Intakea,b,c Fruit intake Attendanced (svgs/d) -High -0.53 (0.9) 0.596 (-2.52, 1.47) 0 - - -Low -0.28 (1.2) 0.815 (-2.69, 2.14) 0.25 (1.1) 0.830 (-3.67, -0.85) -Control 0 - - 0.53 (0.9) 0.596 (-2.63, 0.10) Cohorte -One -0.17 (1.1) 0.886 (-2.50, 2.17) -Two -0.58 (1.2) 0.644 (-3.08, 1.93) -Three 0 - - Incomef -Low -1.00 (0.8) 0.212 (-2.62, 0.60) -Non-Low 0 - - g 206 Caregiver Race -Other 0.08 (1.4) 0.954 (-2.68, 2.83) -Black -0.64 (0.9) 0.505 (-2.59, 1.30) -White 0 - - Caregiver Sexh -Male -1.83 (2.8) 0.517 (-7.52, 3.86) -Female 0 - - Caregiver Agei -18-30 yrs 0.58 (2.7) 0.830 (-4.83, 5.98) -31-40 yrs 1.23 (0.8) 0.130 (-0.38, 2.83) -41+ yrs - - - Fruit intake T0 0.30 (0.4) 0.400 (-0.42, 1.03) Whole fruit intake Attendanced (svgs/d) -High -0.42 (0.9) 0.642 (-2.26, 1.41) 0 - - -Low -0.33 (1.1) 0.764 (-2.53, 1.88) 0.10 (1.0) 0.642 (-1.41, 2.26) -Control 0 - - 0.42 (0.9) 0.927 (-2.02, 2.21) Table 26. (continued)

206 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One 0.10 (1.1) 0.923 (-2.05, 2.52) -Two -0.68 (1.1) 0.558 (-2.99, 1.65) -Three 0 - - Incomef -Low -0.85 (0.7) 0.245 (-2.32, 0.61) -Non-Low 0 - - Caregiver Raceg -Other -0.13 (1.2) 0.916 (-2.66, 2.40) -Black -0.71 (0.9) 0.415 (-2.46, 1.04) -White 0 - - Caregiver Sexh -Male -1.15 (2.4) 0.640 (-6.09, 3.80) -Female 0 - - 207 Caregiver Agei

-18-30 yrs 0.90 (2.4) 0.714 (-4.06, 5.87) -31-40 yrs 1.40 (0.7) 0.065 (-0.09, 2.89) -41+ yrs 0 - - Whole Fruit intake 0.33 (0.4) 0.387 (-0.44, 1.10) T0 Vegetable intake Attendanced (svgs/d) -High -0.71 (0.9) 0.461 (-2.63, 1.22) 0 - - -Low -0.86 (1.1) 0.456 (-3.19, 1.46) -0.16 (1.0) 0.885 (-2.36, 2.05) -Control 0 - - 0.71 (0.9) 0.461 (-1.22, 2.63) Cohorte -One -0.47 (1.1) 0.669 (-2.71, 1.76) -Two 0.36 (1.2) 0.763 (-2.07, 2.80) -Three 0 - - Incomef -Low -1.14 (0.7) 0.135 (-2.65, 0.37) -Non-Low 0 - - Table 26. (continued)

207

Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other -0.84 (1.3) 0.533 (-3.54, 1.86) -Black -0.99 (0.9) 0.269 (-2.77, 0.80) -White 0 - - Caregiver Sexh -Male 4.89 (2.5) 0.063 (-0.28, 10.07) -Female 0 - - Caregiver Agei -18-30 yrs -2.61 (2.6) 0.318 (-7.86, 2.63) -31-40 yrs 0.36 (0.8) 0.642 (-1.21, 1.94) -41+ yrs 0 - - Veg intake T0 0.25 (0.2) 0.220 (-0.16, 0.66) d 208 Sugar sweetened Attendance beverage intake -High -0.30 (0.3) 0.346 (-0.92, 0.33) 0 - - (svgs/d) -Low -0.57 (0.4) 0.133 (-1.32, 0.18) -0.27 (0.4) 0.448 (-0.99, 0.45) -Control 0 - - 0.30 (0.3) 0.346 (-0.33, 0.92) Cohorte -One -0.46 (0.4) 0.210 (-1.19, 0.27) -Two 0.02 (0.4) 0.962 (-0.78, 0.81) -Three 0 - - Incomef -Low 0.44 (0.2) 0.081 (-0.06, 0.94) -Non-Low 0 - - Caregiver Raceg -Other 0.27 (0.4) 0.530 (-0.59, 1.13) -Black -0.22 (0.3) 0.464 (-0.81, 0.38) -White 0 - - Caregiver Sexh -Male 4.58 (1.3) 0.001 (1.98, 7.18) -Female 0 - - Table 26. (continued)

208

Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Agei -18-30 yrs 0.53 (0.8) 0.530 (-1.16, 2.21) -31-40 yrs 0.28 (0.3) 0.271 (-0.23, 0.79) -41+ yrs 0 - - SSB intake T0 0.31 (0.2) 0.076 (-0.03, 0.66) Energy intake Attendanced (kcals/d) -High -215.61 (296.5) 0.472 (-818.87, 387.65) 0 - - -Low -646.94 (376.5) 0.095 (-1,412.99, 119.12) -431.33 (352.0) 0.228 (-1,145.46, 282.79) -Control 0 - - 215.61 (296.5) 0.472 (818.87, -387.65) Cohorte -One 509.44 (348.8) 0.154 (-200.10, 1,218.98) -Two -6.40 (369.8) 0.986 (-758.66, 745.86) -Three 0 - - Incomef

209 -Low 70.24 (240.0) 0.772 (-418.06, 558.54) -Non-Low 0 - - Caregiver Raceg -Other 31.99 (404.9) 0.938 (-791.90, 855.87) -Black 469.82 (280.3) 0.103 (-100.36, 1,040.0) -White 0 - - Caregiver Sexh -Male -17.99 (821.9) 0.983 (-1,690.24, 1,930.0) -Female 0 - - Caregiver Agei -18-30 yrs 302.45 (799.9) 0.708 (-1,325.1, 1,930.0) -31-40 yrs 98.21 (236.1) 0.680 (-382.20, 578.62) -41+ yrs 0 - - Energy intake T0 0.22 (0.2) 0.337 (-0.24, 0.68) Table 26. (continued)

209 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) HEI score Attendanced -High -9.48 (5.9) 0.118 (-21.50, 2.54) 0 - - -Low -0.72 (7.2) 0.921 (-15.33, 13.89) 8.76 (7.1) 0.227 (-5.72, 23.25) -Control 0 - - 9.48 (5.9) 0.118 (-2.54, 21.50) Cohorte -One -4.72 (6.8) 0.495 (-18.63, 9.20) -Two -1.95 (7.3) 0.793 (-16.91, 13.01) -Three 0 - - Incomef -Low -5.11 (4.8) 0.298 (-14.93, 4.72) -Non-Low 0 - - Caregiver Raceg -Other 2.52 (8.1) 0.757 (-13.90, 18.94) -Black -5.04 (5.6) 0.379 (-16.54, 6.46) 210 -White 0 - - h

Caregiver Sex -Male -0.93 (16.5) 0.956 (-34.39, 32.63) -Female 0 - - Caregiver Agei -18-30 yrs -27.05 (16.8) 0.117 (-61.20, 7.10) -31-40 yrs 4.36 (4.7) 0.365 (-5.30, 14.02) -41+ yrs 0 - - HEI Score T0 0.05 (0.3) 0.863 (-0.50, 0.60) Caregiver Anthropometrics/Blood Pressure BMI Attendanced -High -0.82 (0.4) 0.048 (-1.63, -0.01) 0 - - -Low -0.88 (0.5) 0.068 (1.83, 0.07) -0.06 (0.4) 0.891 (-0.94, 0.82) -Control 0 - - 0..82 (0.4) 0.048 (0.01, 1.63) Table 26. (continued)

210 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -0.28 (0.4) 0.477 (-1.07, 0.51) -Two 0.07 (0.4) 0.876 (-0.79, 0.92) -Three 0 - - Incomef -Low -0.93 (0.3) 0.009 (-1.62, -0.24) -Non-Low 0 - - Caregiver Raceg -Other 0.41 (0.6) 0.529 (-0.88, 1.69) -Black 0.85 (0.4) 0.031 (0.08, 1.63) -White 0 - - Caregiver Sexh -Male -0.61 (1.5) 0.692 (-3.67, 2.45) -Female 0 - - i 211 Caregiver Age -18-30 yrs 0.26 (0.6) 0.670 (-0.93, 1.44) -31-40 yrs -0.33 (0.4) 0.365 (-1.05, 0.39) -41+ yrs 0 - - BMI, T0 0.99 (0.1) <0.001 (0.96, 1.03) Waist Attendanced Circumference -High -1.15 (3.0) 0.703 (-7.15, 4.85) 0 - - -Low -0.85 (3.5) 0.810 (07.83, 6.14) 0.30 (3.3) 0.927 (-6.24, 6.85) -Control 0 - - 1.15 (3.0) 0.703 (-4.85, 7.15) Cohorte -One -5.43 (2.9) 0.067 (-11.25, 0.38) -Two 0.11 (3.2) 0.974 (-6.22, 6.43) -Three 0 - - Incomef -Low -8.60 (2.6) 0.002 (-13.80. -3.40) -Non-Low 0 - - Table 26. (continued)

211 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other 4.58 (4.7) 0.333 (-4.79, 13.95) -Black 3.46 (2.9) 0.235 (-2.29, 9.21) -White 0 - - Caregiver Sexh -Male -2.26 (11.3) 0.843 (-24.85, 20.34) -Female 0 - - Caregiver Agei -18-30 yrs 2.06 (4.5) 0.645 (-6.82, 10.94) -31-40 yrs -2.21 (2.7) 0.422 (-7.67, 3.25) -41+ yrs 0 - - WC, T0 0.92 (0.1) <0.001 (0.79, 1.05) Systolic BP Attendanced -High -5.01 (3.1) 0.111 (-11.20, 1.18) 0 - -

212 -Low -7.61 (3.6) 0.039 (-14.82, -0.40) -2.60 (3.4) 0.451 (-9.45, 4.24) -Control 0 - - 5.01 (3.1) 0.111 (-1.18, 11.20) Cohorte -One -4.46 (3.0) 0.144 (-10.47, 1.56) -Two -4.74 (3.3) 0.150 (-11.23, 1.75) -Three 0 - - Incomef -Low 3.35 (2.7) 0.218 (-2.03, 8.74) -Non-Low 0 - - Caregiver Raceg -Other 5.69 (4.8) 0.244 (-3.96, 15.33) -Black 4.09 (2.9) 0.167 (-1.74, 9.92) -White 0 - - Caregiver Sexh -Male -3.13 (11.7) 0.790 (-26.38, 20.13) -Female 0 - - Table 26. (continued)

212 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Agei -18-30 yrs -15.94 (4.5) 0.001 (-24.97, -6.91) -31-40 yrs -8.89 (2.8) 0.002 (-14.37, -3.40) -41+ yrs - - - SysBP, T0 0.44 (0.1) <0.001 (-.27, 0.61) Diastolic BP Attendanced -High 1.53 (2.5) 0.540 (-3.42, 6.48) 0 - - -Low -0.70 (2.9) 0.810 (-6.52, 5.12) -2.24 (2.7) 0.417 (-7.69, 3.22) -Control 0 - - -1.53 (2.5) 0.540 (-6.48, 3.42) Cohorte -One -1.65 (2.4) 0.497 (-6.48, 3.18) -Two -4.38 (2.6) 0.097 (-9.57, 0.82) -Three 0 - - Incomef

213 -Low 1.40 (2.1) 0.509 (-2.81, 5.62) -Non-Low 0 - - Caregiver Raceg -Other -2.32 (3.9) 0.553 (-10.07, 5.43) -Black -0.05 (2.4) 0.985 (-4.74, 4.65) -White 0 - - Caregiver Sexh -Male -7.62 (9.4) 0.420 (-26.35, 11.11) -Female 0 - - Caregiver Agei -18-30 yrs -1.47 (3.6) 0.688 (-8.74, 5.80) -31-40 yrs 0.32 (2.2) 0.884 (-4.07, 4.71) -41+ yrs 0 - - DysBP, T0 0.42 (0.1) <0.001 (0.25, 0.59) Table 26. (continued)

213 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Personal Determinants Self-Efficacy Attendanced -High 5.15 (2.9) 0.082 (-0.67, 10.97) 0 - - -Low 10.21 (3.3) 0.003 (3.56, 16.85) 5.05 (3.1) 0.108 (-1.14, 11.24) -Control 0 - - -5.15 (2.0) 0.082 (-10.97, 0.67) Cohorte -One 1.24 (2.8) 0.663 (-4.39, 6.86) -Two 0.03 (2.8) 0.993 (-6.24, 6.29) -Three 0 - - Incomef -Low -4.81 (2.4) 0.051 (-0.65, 0.03) -Non-Low 0 - - Caregiver Raceg -Other 8.76 (4.6) 0.059 (-0.34, 17.86)

214 -Black 6.66 (2.7) 0.017 (1.24, 12.08) -White 0 - - Caregiver Sexh -Male 9.60 (11.2) 0.395 (-12.75, 31.95) -Female 0 - - Caregiver Agei -18-30 yrs -6.02 (4.3) 0.171 (-14.68, 2.65) -31-40 yrs 1.51 (2.6) 0.558 (-3.61, 6.64) -41+ yrs 0 - - Self-efficacy T0 0.45 (0.1) <0.001 (0.31, 0.58) Menu Planning Attendanced -High 0.94 (1.0) 0.341 (-1.00, 2.88) 0 - - -Low 0.60 (1.1) 0.595 (-1.64, 2.85) -0.34 (1.1) 0.751 (-2.43, 1.76) -Control 0 - - -0.94 (1.0) 0.341 (-2.88, 1.00) Table 26. (continued)

214 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -0.56 (0.9) 0.551 (-2.44, 1.31) -Two -1.70 (1.0) 0.107 (-3.78, 0.38) -Three 0 - - Incomef -Low 1.76 (0.8) 0.034 (0.14, 3.38) -Non-Low 0 - - Caregiver Raceg -Other 2.65 (1.5) 0.087 (-0.40, 5.71) -Black -0.80 (0.9) 0.397 (-2.68, 1.07) -White 0 - - Caregiver Sexh -Male 3.91 (3.8) 0.302 (-3.58, 11.40) -Female 0 - - i 215 Caregiver Age -18-30 yrs -2.27 (1.5) 0.128 (-5.21, 0.67) -31-40 yrs 0.32 (0.9) 0.714 (-1.40, 2.03) -41+ yrs 0 - - Menu planning, T0 0.59 (0.1) <0.001 (0.37, 0.82) Family Meals Family Breakfast Attendanced Frequency -High 0.19 (0.7) 0.781 (-1.15, 1.53) 0 - - (meals/wk) -Low -0.01 (0.8) 0.998 (-1.55, 1.54) -0.19 (0.8) 0.801 (-1.68, 1.30) -Control 0 - - -0.19 (0.7) 0.781 (-1.53, 1.15) Cohorte -One 0.28 (0.7) 0.680 (-1.08, 1.64) -Two -1.17 (0.7) 0.102 -2.57, 0.24) -Three 0 - - Incomef -Low 0.73 (0.6) 0.209 (-0.42, 1.88) -Non-Low 0 - - Table 26. (continued)

215 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Caregiver Raceg -Other 1.56 (1.3) 0.218 (-0.94, 4.07) -Black 0.39 (1.2) 0.750 (-2.08, 2.88) -White 0 - - Caregiver Sexh -Male 1.30 (2.8) 0.639 (-4.19, 6.79) -Female 0 - - Caregiver Agei -18-30 yrs 0.42 (1.0) 0.672 (-1.56, 2.41) -31-40 yrs 0.23 (0.6) 0.715 (-1.01, 1.46) -41+ yrs 0 - - Oldest Child Racen -2.72 (1.1) 0.013 (-4.86, -0.58) -Other -1.67 (1.3) 0.201 (-4.26, 0.91)

216 -Black 0 - - -White Oldest Child Sexo 0.14 (0.6) 0.824 (-1.11, 1.39) -Male 0 - - -Female -0.07 (0.2) 0.652 (-0.37, 0.23) Oldest Child Agep 0.17 (0.1) 0.170 (-0.08, 0.42) Family Breakfast Freq, T0 Family Dinner Attendanced Frequency -High 0.72 (0.6) 0.205 (-0.40, 1.84) 0 - - (meals/wk) -Low 0.07 (0.6) 0.913 (-1.21, 1.35) -0.65 (0.6) 0.290 (-1.86, 0.57) -Control 0 - - -0.72 (0.6) 0.205 (-1.84, 0.40) Cohorte -One 0.58 (0.6) 0.303 (-0.54, 1.71) -Two 0.88 (0.6) 0.149 (-0.32, 2.08) -Three 0 - - Table 26. (continued)

216 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Incomef -Low -0.10 (0.5) 0.840 (-1.05, 0.86) -Non-Low 0 - - Caregiver Raceg -Other 2.38 (1.0) 0.025 (0.31, 4.46) -Black 0.91 (1.0) 0.382 (-1.15, 2.95) -White 0 - - Caregiver Sexh -Male -1.42 (2.3) 0.535 (-5.97, 3.13) -Female 0 - - Caregiver Agei -18-30 yrs -1.54 (0.8) 0.073 (-3.23, 0.15) -31-40 yrs -0.51 (0.5) 0.321 -1.51, 0.50) -41+ yrs 0 - -

217 Oldest Child Racen -0.96 (0.9) 0.281 (-2.73, 0.81) -Other -0.59 (1.1) 0.583 (-2.72, 1.55) -Black 0 - - -White Oldest Child Sexo -0.47 (0.5) 0.378 (-1.52, 0.58) -Male 0 - - -Female -0.09 (0.1) 0.455 (-0.34, 0.15) Oldest Child Agep 0.09 (0.1) 0.462 (-0.15, 0.33) Family Dinner Freq, T0 Family meals in a Attendanced dining area -High -0.36 (0.6) 0.562 (-1.60, 0.88) 0 - - (meal/wk) -Low 0.35 (0.7) 0.629 (-1.08, 1.78) 0.71 (0.7) 0.311 (-0.68, 2.09) -Control 0 - - 0.36 (0.6) 0.562 (-0.88, 1.60) Table 26. (continued)

217 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -0.51 (0.6) 0.418 (-1.76, 0.74) -Two 0.02 (0.7) 0.980 (-1.29, 1.32) -Three 0 - - Incomef -Low 0.68 (0.5) 0.207 (-0.38, 1.74) -Non-Low 0 - - Caregiver Raceg -Other -0.45 (1.2) 0.701 (-2.76, 1.87) -Black -1.20 (1.1) 0.307 (-3.52, 1.12) -White 0 - - Caregiver Sexh -Male -0.15 (2.5) 0.953 (-5.24, 4.93) -Female 0 - - i 218 Caregiver Age -18-30 yrs 0.03 (0.9) 0.978 (-1.83, 1.88) -31-40 yrs -0.41 (0.6) 0.462 (-1.52, 0.70) -41+ yrs 0 - - Oldest Child Racen -Other 0.36 (1.0) 0.715 (-1.61, 2.34) -Black 1.28 (1.2) 0.288 (-1.11, 3.67) -White 0 - - Oldest Child Sexo -Male 0.19 (0.6) 0.744 (-0.96, 1.34) -Female 0 - - Oldest Child Agep -0.11 (0.1) 0.432 (-0.38, 0.16) Meal at Table Freq, T0 0.40 (0.1) <0.001 (0.18, 0.62) Family Meal TV Attendanced Viewing -High -1.34 (0.5) 0.011 (-2.36, -0.32) 0 - - (meals/wk) -Low -0.58 (0.6) 0.330 (-1.75, 0.60) 0.77 (0.5) 0.166 (-0.33, 1.86) -Control 0 - - 1.34 (0.5) 0.011 (0.32, 2.36) Table 26. (continued)

218 Table 26. Intervention Period: By-Group Differences in Caregiver and Family Outcomes at Post-Test, T1 (continued) Cohorte -One -0.25 (0.5) 0.627 (-1.26, 0.77) -Two 0.07 (0.5) 0.893 (-0.98, 1.12) -Three 0 - - Incomef -Low 0.65 (0.4) 0.137 (-0.21, 1.51) -Non-Low 0 - - Caregiver Raceg -Other -3.42 (0.9) 0.001 (-5.30, -1.55) -Black -4.67 (0.9) <0.001 (-6.52, -2.83) -White 0 - - Caregiver Sexh -Male 1.28 (2.1) 0.535 (-2.82, 5.38) -Female 0 - - i 219 Caregiver Age -18-30 yrs -0.66 (0.7) 0.375 (-2.14, 0.82) -31-40 yrs -0.46 (0.5) 0.314 (-1.38, 0.45) -41+ yrs 0 - - Oldest Child Racen -Other 3.98 (0.8) <0.001 (2.38, 5.57) -Black 4.99 (1.0) <0.001 (3.07, 6.93) -White 0 - - Oldest Child Sexo -Male -0.19 (0.5) 0.688 (-1.12, 0.74) -Female 0 - - Oldest Child Agep -0.14 (0.1) 0.224 (-0.35, 0.08) Meal Without TV, T0 0.55 (0.1) <0.001 (0.35, 0.76) aBy-group differences determined by generalized linear models controlling for: cohort; household income; caregiver sex, race, and age; baseline value; level of attendance (Y (outcome variable)= outcome at T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] Table 26. (continued)

219

Table 26. Differences in Caregiver and Family Outcomes by Level of Attendance at Post-Test, T1 (continued) cParticipant (n=95) completion of 24hr dietary recalls at T0: (n=6 (7%)); 1 recall (n=59 (65%)); 2 recalls (n=12 (13%)); 3 recalls (n=14 (15%)). Participant completion of 24hr dietary recalls at T1: 0 recalls (n=33 (35%)); 1 recall (n=56 (59%)); 2 recalls (n=6 (6%)); 3 recalls (n=0 (0%))dLevel of Attendance (based on total number of lessons attended): control group (n=32); low attendance group (n=23); high attendance group (n=40) eCohort: one (n= 35); two (n=25); three (n=35) fIncome: low (n= 36); non-low (n=55). Low-income defined as participation in >1 of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) gCaregiver Race: other (n= 9); Black (n= 59); White (n=27). Other category includes participants who did not identify with Black or White hCaregiver Sex: male (n=2); female (n= 93) iCaregiver Age: 18-30 yrs (n=10); 31-40 yrs (n=46); 41+ yrs (n=36) jHealthy Eating Index, 0-100 scale, Higher score represents better diet quality [198] kSelf-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased self-efficacy lMenu planning behaviors, 0-36 scale with higher score representing increased menu planning m 220 By-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, and age; oldest child race, sex, and age; baseline value; group assignment (Y (outcome variable)= outcome at T1) nOldest Child Race: other (n=18); Black (n=54); White (n=23). Other category includes participants who did not identify with either Black or White oOldest Child Sex: male (n=35); female (n=60) pChild (oldest) age: mean (SD)= 7.4 (1.9) yrs

220 Table 27. Follow-Up Period: By-Group (Low vs. High Attenders) Caregiver and Family Outcomes Post-Test T1 Follow-Up T2 Mean (SD) Change from Post-Test to Follow-Up T2 Caregiver Dietary Intakeb,c,d,e Low Attendersa High Attendersa Low Attendersa High Attendersa Low Attendersa High Attendersa (n=23) (n=40) (n=23) (n=40) (n=23) (n=40) Fruit intake (svgs/d)f 1.59 (2.4) 1.42 (2.3) 0.67 (0.7) 1.06 (1.3) -0.92 (1.3) -0.36 (1.4) Whole fruit intake (svgs/d) 1.35 (2.4) 1.22 (2.1) 0.62 (0.7) 0.78 (1.0) -0.73 (1.2) -0.44 (1.3) Vegetable intake (svgs/d) 2.58 (1.8) 3.28 (2.2) 2.23 (1.1) 3.27 (1.8) -0.35 (0.8) -0.01 (2.9) Sugar sweetened beverage intake 0.38 (1.0) 0.54 (1.2) 0.43 (0.9) 0.34 (0.6) 0.05 (1.1) -0.20 (0.7) (svgs/d) Energy intake (kcals/d) 1532.63 (322.8) 1673.24 (760.4) 1759.32 (607.8) 1518.86 (559.8) 226.69 (650.6) -154.38 (1030.7) HEI scoreg 59.89 (13.5) 55.22 (15.2) 54.41 (12.7) 56.32 (10.4) -5.48 (10.8) 1.10 (16.6) HEI Total Fruit scoreh 1.97 (2.3) 1.80 (4.2) 1.49 (1.8) 1.76 (4.1) -0.48 (1.9) -0.04 (4.2) HEI Whole Fruit scorei 2.16 (2.4) 2.78 (2.1) 2.98 (1.9) 2.02 (1.9) 0.82 (2.1) -0.76 (1.9) 221 HEI Total Vegetable scorej 3.28 (1.2) 3.52 (1.7) 2.83 (1.0) 3.70 (1.6) -0.45 (1.1) 0.18 (1.8) b,c Caregiver Athropometrics/BP BMI (kg/m2) 31.52 (9.1) 31.87 (9.6) 31.16 (9.2) 31.79 (9.3) -0.36 (2.0) -0.08 (1.9) Waist circumference (cm) 101.20 (20.6) 100.65 (20.6) 99.19 (18.8) 100.76 (21.0) -2.01 (11.4) 0.11 (11.9) Systolic BP (mm Hg) 121.61 (12.3) 128.66 (16.6) 127.01 (16.4) 126.62 (19.8) 5.40 (14.2) -2.04 (11.6) Diastolic BP (mm Hg) 78.73 (8.9) 81.27 (12.2) 80.28 (11.9) 78.84 (12.8) 1.55 (11.2) -2.43 (13.3) Personal Determinantsb,c Self-efficacyk 103.02 (11.5) 100.51 (14.2) 104.82 (12.4) 102.13 (14.1) 1.80 (11.8) 1.62 (8.2) Menu Planningl 26.55 (4.9) 26.77 (3.8) 26.63 (3.8) 27.32 (3.1) 0.08 (3.9) 0.55 (3.4) Family Meals (meals/wk)b,c Family breakfast frequency 2.97 (2.7) 2.82 (2.2) 4.08 (2.5) 3.88 (2.4) 1.11 (3.8) 1.06 (2.7) Family dinner frequency 3.67 (2.2) 4.82 (1.9) 4.04 (1.9) 4.88 (1.8) 0.37(2.6) 0.06 (2.3) Family meals in a dining area 4.75 (2.3) 4.16 (2.1) 4.85 (2.4) 4.61 (1.8) 0.10 (3.4) 0.45 (2.2) Family meals TV viewing 2.69 (2.8) 1.73 (1.7) 2.35 (2.1) 2.00 (1.9) -0.34 (2.9) 0.27 (2.4)

Table 27. (continued)

221 Table 27. Follow-Up Period: By-Group (Low vs. High Attenders) Caregiver and Family Outcomes (continued) aLevel of attendance based on total number of lessons attended bValues are Mean(SD) cBy-group differences determined by paired samples t-test dDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] eIntervention participants’ (n=63) completion of 24hr dietary recalls at T1: 0 recalls (n=22 (35%)); 1 recall (n=38 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=1 (2%)). Intervention participants’ (n=63) completion of 24hr dietary recalls at T2: 0 recalls (n= 22 (35%)); 1 recall (n=39 (65%)); 2 recalls (n=0 (0%)); 3 recalls (n=0 (0%)) fTotal fruit= whole fruit + 100% fruit juice gHealthy Eating Index (HEI), 0-100 with higher score representing better intake hHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake jHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake kSelf-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased self-efficacy l 222 Menu planning behaviors, 0-36 scale with higher score representing increased menu planning

222 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes Follow-Up Period (T1 àT2) Outcomes Covariates β (SE) p-value Confidence Interval Caregiver Dietary Intakea,b,c Fruit intake (svgs/d) Attendanced -High 1.24 (1.4) 0.409 (-1.96, 4.44) -Low 0 - - Cohorte -One -0.15 (0.9) 0.867 (-2.08, 1.78) -Two -1.70 (1.4) 0.261 (-4.87, 1.48) -Three 0 - - Incomef -Low -0.12 (0.8) 0.879 (-1.84, 1.60) -Non-Low 0 - - Caregiver Raceg

223 -Other 0.86 (1.9) 0.662 (-3.39, 5.11) -Black -0.59 (1.0) 0.577 (-2.86, 1.69) -White 0 - - Caregiver Sexh -Male - -Female 0 Caregiver Agei -18-30 yrs - - - -31-40 yrs 0.17 (0.9) 0.864 (-1.92, 2.25) -41+ yrs 0 - - Whole fruit intake (svgs/d) Attendanced -High 0.93 (1.3) 0.474 (-1.86, 3.73) -Low 0 - - Table 20. (continued)

223 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Cohorte -One 0.12 (0.8) 0.882 (-1.57, 1.80) -Two -1.41 (1.2) 0.284 (-4.18, 1.36) -Three 0 - - Incomef -Low 0.05 (0.7) 0.943 (-1.45, 1.55) -Non-Low 0 - - Caregiver Raceg -Other 0.95 (1.7) 0.581 (-2.76, 4.67) -Black -0.75 (0.9) 0.419 (-2.74, 1.24) -White 0 - - Caregiver Sexh -Male - -Female - i 224 Caregiver Age -18-30 yrs - - -31-40 yrs -0.23 (0.8) 0.780 (-2.05, 1.58) -41+ yrs 0 - - Vegetable intake (svgs/d) Attendanced -High -2.63 (2.99) 0.401 (-9.31, 4.05) -Low 0 - - Cohorte -One -0.92 (1.8) 0.622 (-4.94, 3.11) -Two 1.04 (2.9) 0.734 (-5.58, 7.65) -Three 0 - - Incomef -Low 1.37 (2.9) 0.416 (-2.22, 4.96) -Non-Low 0 - - Table 28. (continued)

224 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Raceg -Other -2.82 (3.9) 0.494 (-11.69, 6.05) -Black -0.19 (2.1) 0.929 (-4.94, 4.55) -White 0 - - Caregiver Sexh -Male - -Female - Caregiver Agei -18-30 yrs - - - -31-40 yrs -2.19 (1.9) 0.287 (-6.53, 2.15) -41+ yrs 0 - - Sugar sweetened beverage intake (svgs/d) Attendanced -High -0.26 (0.7) 0.717 (-1.83, 1.31) -Low 0 - - e 225 Cohort -One 0.10 (0.4) 0.820 (-0.85, 1.04) -Two -1.10 (0.7) 0.145 (-2.66, 0.45) -Three 0 - - Incomef -Low -0.50 (0.4) 0.214 (-1.35, 0.34) -Non-Low 0 - - Caregiver Raceg -Other -0.25 (0.9) 0.799 (-2.33, 1.84) -Black -0.13 (0.5) 0.804 (-1.24, 0.99) -White 0 - - Table 28. (continued)

225 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Sexh -Male - -Female - Caregiver Agei -18-30 yrs - - -31-40 yrs 0.52 (0.5) 0.282 (-0.50, 1.54) -41+ yrs 0 - Energy intake (kcals/d) Attendanced -High -1,439.40 (1,058.1) 0.204 (-3,796.95, 918.15) -Low 0 - - Cohorte -One -611.18 (637.6) 0.360 (-2,031.86, 809.49) -Two 387.34 (1,048.7) 0.720 (1,949.31, 2,723.98) -Three 0 - - f 226 Income -Low -659.11 (568.9) 0.274 (-1,926.90, 608.68) -Non-Low 0 - - Caregiver Raceg -Other -616.82 (1,405.7) 0.670 (-3,748.81, 2,515.18) -Black -710.62 (752.3) 0.367 (-2,386.84, 965.59) -White 0 - - Caregiver Sexh -Male -Female Caregiver Agei -18-30 yrs - - - -31-40 yrs -418.04 (687.9) 0.557 (-1,950.88, 1,114.80) -41+ yrs 0 - - Table 28. (continued)

226 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) HEI scorej Attendanced -High 31.01 (19.0) 0.137 (-11.99, 74.03) -Low 0 - - Cohorte -One 8.36 (11.9) 0.502 (-18.68, 35.41) -Two 8.58 (18.8) 0.659 (-33.98, 51.15) -Three 0 - - Incomef -Low 5.13 (10.5) 0.635 (-18.52, 28.79) -Non-Low 0 - - Caregiver Raceg -Other 2.70 (25.5) 0.918 (-54.96, 60.36) -Black 1.88 (13.6) 0.893 (-28.81, 32.56) -White 0 - - h 227 Caregiver Sex -Male -Female Caregiver Agei -18-30 yrs - -31-40 yrs -5.26 (12.7) 0.688 (-34.02, 23.50) -41+ yrs 0 - - Caregiver Anthropometrics/Blood Pressurea BMI Attendanced -High -0.03 (0.6) 0.965 (-1.18, 1.12) -Low 0 - - Table 28. (continued)

227 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Cohorte -One 0.23 (0.6) 0.701 (-0.95, 1.40) -Two 1.20 (0.7) 0.103 (-0.25, 2.65) -Three 0 - - Incomef -Low 1.17 (0.5) 0.028 (0.13, 2.21) -Non-Low 0 - - Caregiver Raceg -Other 0.52 (0.8) 0.541 (-1.18, 2.22) -Black 0.91 (0.7) 0.164 (-0.39, 2.21) -White 0 - - Caregiver Sexh -Male -0.14 (1.9) 0.944 (-4.00, 3.73) -Female 0 - - i 228 Caregiver Age -18-30 yrs 0.06 (0.9) 0.946 (-1.74, 1.886) -31-40 yrs 1.39 (0.6) 0.018 (0.25, 2.54) -41+ yrs 0 - - Waist Circumference Attendanced -High 3.46 (3.6) 0.343 (-3.81, 10.73) -Low 0 - - Cohorte -One 9.47 (3.7) 0.014 (2.01, 16.93) -Two 0.35 (4.6) 0.939 (-8.84, 9.55) -Three 0 - - Incomef -Low 8.28 (3.3) 0.015 (1.70, 14.86) -Non-Low 0 - - Table 28. (continued)

228 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Raceg -Other -10.78 (5.3) 0.050 (-21.55, -0.02) -Black -3.65 (4.1) 0.376 (-11.87, 4.57) -White 0 - - Caregiver Sexh -Male -1.40 (12.2) 0.909 (-25.88, 23.09) -Female 0 - - Caregiver Agei -18-30 yrs -0.59 (5.6) 0.918 (-11.96, 10.79) -31-40 yrs -0.87 (3.6) 0.811 (-8.11, 6.37) -41+ yrs 0 - - Systolic BP Attendanced -High -0.97 (3.6) 0.791 (-8.26, 6.33) -Low 0 - - e 229 Cohort -One -1.00 (3.7) 0.789 (-8.49, 6.49) -Two 1.39 (4.6) 0.764 (-7.84, 10.62) -Three 0 - - Incomef -Low 7.03 (3.3) 0.038 (0.42, 13.64) -Non-Low 0 - - Caregiver Raceg -Other -4.33 (5.4) 0.425 (-15.14, 6.49) -Black -10.60 (4.1) 0.013 (-18.85, -2.35) -White 0 - - Caregiver Sexh -Male 8.37 (12.2) 0.497 (-16.22, 32.95) -Female 0 - - Table 28. (continued)

229 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Agei -18-30 yrs 5.05 (5.7) 0.379 (-6.38, 16.47) -31-40 yrs 0.44 (3.6) 0.903 (-6.82, 7.71) -41+ yrs 0 - - Diastolic BPa Attendanced -High -4.70 (3.3) 0.166 (-11.41, 2.01) -Low 0 - - Cohorte -One -0.47 (3.4) 0.892 (-7.35, 6.42) -Two 8.38 (4.2) 0.053 (-0.11, 16.87) -Three 0 - - Incomef -Low 4.48 (3.0) 0.145 (-1.60, 10.55) -Non-Low 0 - - g 230 Caregiver Race -Other 10.3 (4.9) 0.042 (0.40, 20.29) -Black 0.78 (3.8) 0.837 (-6.80, 8.36) -White 0 - - Caregiver Sexh -Male 8.91 (11.2) 0.431 (-13.69, 31.53) -Female 0 - - Caregiver Agei -18-30 yrs -9.03 (5.2) 0.090 (-19.53, 1.48) -31-40 yrs -9.07 (3.3) 0.009 (-15.75, -2.38) -41+ yrs 0 - - Caregiver Personal Determinantsa Self-efficacyk Attendanced -High -0.22 (2.7) 0.936 (-5.65, 5.21) -Low 0 - - Table 28. (continued)

230 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Cohorte -One 3.85 (2.8) 0.177 (-1.80, 9.50) -Two 1.93 (3.6) 0.592 (-5.25, 9.12) -Three 0 - - Incomef -Low -3.72 (2.5) 0.149 (-8.81, 1.37) -Non-Low 0 - - Caregiver Raceg -Other -7.26 (4.1) 0.082 (-15.47, 0.95) -Black -0.27 (3.0) 0.928 (-6.37, 5.82) -White 0 - - Caregiver Sexh -Male -18.63 (9.5) 0.056 (-37.78, 0.53) -Female 0 - - Caregiver Agei 231 -18-30 yrs -1.60 (4.4) 0.718 (-10.46, 7.26)

-31-40 yrs -7.43 (2.7) 0.009 (-12.93, -1.94) -41+ yrs 0 - - Menu Planningl Attendanced -High 0.06 (0.7) 0.940 (-1.46, 1.57) -Low 0 - - Cohorte -One -0.89 (0.8) 0.264 (-2.46, 0.69) -Two 0.14 (1.0) 0.891 (-1.86, 2.14) -Three 0 - - Incomef -Low -0.30 (0.7) 0.674 (-1.72, 1.12) -Non-Low 0 - - Table 28. (continued)

231 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Raceg -Other -2.85 (1.1) 0.016 (-5.13, -0.56) -Black 0.08 (0.8) 0.927 (-1.62, 1.77) -White 0 - - Caregiver Sexh -Male -1.99 (2.7) 0.456 (-7.32, 3.34) -Female 0 - - Caregiver Agei -18-30 yrs -0.70 (1.2) 0.570 (-3.17, 1.76) -31-40 yrs -1.07 (0.8) 0.168 (-2.60, 0.47) -41+ yrs 0 - - Family Mealsm Family Breakfast Frequency (meals/wk) Attendanced -High -1.20 (1.1) 0.270 (-3.37, -0.97)

232 -Low 0 - - Cohorte -One -1.34 (1.0) 0.201 (-3.43, 0.74) -Two 0.78 (1.2) 0.536 (-1.75, 3.32) -Three 0 - - Incomef -Low -0.77 (0.9) 0.417 (-2.68, 1.13) -Non-Low 0 - - Caregiver Raceg -Other -0.70 (1.8) 0.700 (-4.32, 2.93) -Black -0.14 (2.1) 0.947 (-4.33, 4.05) -White 0 - - Caregiver Sexh -Male 1.81 (3.7) 0.628 (-5.65, 9.26) -Female 0 - - Table 28. (continued)

232 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Agei -18-30 yrs -1.95 (1.5) 0.209 (-5.04, 1.14) -31-40 yrs -1.79 (1.0) 0.084 (-3.83, 0.25) -41+ yrs 0 - - Child Racen -Other 4.33 (1.8) 0.023 (-0.53, 8.03) -Black 2.49 (2.2) 0.267 (-1.97, 6.95) -White 0 - - Child Sexo -Male -1.30 (1.0) 0.195 (-3.29, 0.69) -Female 0 - - Child Agep 0.07 (0.2) 0.762 (-0.40, 0.55) Family Dinner Frequency (meals/wk) Attendanced -High -0.39 (0.8) 0.646 (-2.10, 1.32)

233 -Low 0 - - Cohorte -One -0.50 (0.8) 0.538 (-2.14, 1.13) -Two 0.52 (1.0) 0.614 (-1.53, 2.57) -Three 0 - - Incomef -Low -0.96 (0.7) 0.201 (-2.45, 0.53) -Non-Low 0 - - Caregiver Raceg -Other 0.25 (1.4) 0.863 (-2.60, 3.09) -Black 3.72 (1.6) 0.028 (0.43, 7.01) -White 0 - - Caregiver Sexh -Male -0.40 (2.9) 0.892 (-6.27, 5.47) -Female 0 - - Table 28. (continued)

233 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Agei -18-30 yrs 0.46 (1.2) 0.702 (-1.97, 2.90) -31-40 yrs 0.36 (0.8) 0.657 (-1.26, 1.97) -41+ yrs 0 - - Child Racen -Other -1.78 (1.4) 0.223 (-4.68, 1.12) -Black -4.16 (1.7) 0.021 (-7.66, -0.66) -White 0 - - Child Sexo -Male 0.49 (0.8) 0.540 (-1.12, 2.11) -Female 0 - - Child Agep 0.15 (0.2) 0.413 (-0.22, 0.53) Family meals in a dining area (meal/wk) Attendanced -High -0.04 (0.9) 0.965 (-1.91, 1.83)

234 -Low 0 - - Cohorte -One 0.55 (0.9) 0.538 (-1.24, 2.35) -Two -0.05 (1.1) 0.965 (-2.23, 2.13) -Three 0 - - Incomef -Low -1.52 (0.8) 0.068 (-3.16, 0.12) -Non-Low 0 - - Caregiver Raceg -Other 0.23 (1.5) 0.883 (-2.89, 3.35) -Black 0.68 (1.8) 0.706 (-2.93, 4.29) -White 0 - - Caregiver Sexh -Male 0.25 (3.2) 0.937 (-6.16, 6.66) -Female 0 - - Table 28. (continued)

234 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Agei -18-30 yrs -0.82 (1.3) 0.538 (-3.48, 1.84) -31-40 yrs 0.93 (0.9) 0.292 (-0.83, 2.68) -41+ yrs 0 - - Child Racen -Other -0.94 (1.6) 0.557 (-4.12, 2.25) -Black -2.57 (1.9) 0.183 (-6.41, 1.26) -White 0 - - Child Sexo -Male -0.21 (0.8) 0.806 (-1.92, 1.50) -Female 0 - - Child Agep 0.08 (0.2) 0.708 (-0.33, 0.48) Family Meals TV Viewing (meals/wk) Attendanced -High -0.27 (0.9) 0.777 (-2.19, 1.65) -Low 0 - - 235 Cohorte

-One -0.46 (1.1) 0.614 (-2.28, 1.36) -Two -0.41 (1.1) 0.717 (-2.69, 1.86) -Three 0 - - Incomef -Low -1.36 (0.9) 0.119 (-3.09, 0.36) -Non-Low 0 - - Caregiver Raceg -Other 2.81 (1.9) 0.149 (-1.05, 6.66) -Black 3.49 (1.9) 0.073 (-0.35, 7.33) -White 0 - - Caregiver Sexh -Male -1.24 (3.4) 0.717 (-8.08, 5.61) -Female 0 - - Table 28. (continued)

235 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) Caregiver Agei -18-30 yrs -1.25 (1.4) 0.367 (-4.03, 1.52) -31-40 yrs -0.22 (0.9) 0.813 (-2.05, 1.62) -41+ yrs - - - Child Racen -Other -1.62 (1.6) 0.307 (-4.78, 1.54) -Black -2.18 (1.9) 0.278 (-6.18, 1.83) -White 0 - - Child Sexo -Male 0.04 (0.9) 0.968 (-1.85, 1.93) -Female 0 - - Child Agep 0.26 (0.2) 0.222 (-0.16, 0.69) aBy-group differences determined by generalized linear modeling controlling for: cohort; household income; caregiver race, sex, and age; level of attendance (Y (outcome variable)= change score of T2-T1) bDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step 236 multi-pass method [189, 197] c Intervention participants’ (n=63) completion of 24hr dietary recalls at T1: 0 recalls (n=22 (35%)); 1 recall (n=38 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=1 (2%)). Intervention participants’ (n=63) completion of 24hr dietary recalls at T2: 0 recalls (n= 22 (35%)); 1 recall (n=39 (65%)); 2 recalls (n=0 (0%)); 3 recalls (n=0 (0%)) dAttendance (based on total number of lessons attended): low attendance group (n=23); high attendance group (n=40) eCohort: one (n= 24); two (n=12); three (n=27) fIncome: low (n= 24); non-low (n=38). Low-income defined as participation in one or more of the following federal food assistance programs: Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC); Supplemental Nutrition Assistance Program (SNAP); National School Lunch Program (NSLP) gCaregiver Race: other (n=6); Black (n= 39); White (n=16). Other category includes participants who did not identify with either Black or White hCaregiver Sex: male (n=1); female (n= 62) iCaregiver Age: 18-30 yrs (n=6); 31-40 yrs (n=27); 41+ yrs (n=28) jHealthy Eating Index (HEI), 0-100 scale with higher score representing better diet quality [198] kSelf-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased self-efficacy lMenu planning behaviors, 0-36 scale with higher score representing increased menu planning Table 28. (continued)

236 Table 28. Follow-Up Period: By-Group (Low vs. High Attenders) Differences in Caregiver and Family Outcomes (continued) mBy-group differences determined by generalized linear modelling controlling for: cohort; household income; caregiver race, sex, and age; oldest child race, sex, and age; level of attendance (Y (outcome variable)= outcome at T1) nChild (oldest) sex: male (n=24); female (n= 39) oChild (oldest) race: other (n=14); non-Hispanic Black (n=37); non-Hispanic White (n=12). Other category includes participants who did not identify with either Black or White pChild (oldest) age: mean (SD)= 7.3 (1.8) yrs

Table 29. Follow-Up Period: Within-Group (Intervention Group) Differences in Caregiver and Family Outcome Post-Test T1 Follow-Up T2 Mean (SD) Change from Post- 237 (n=63) (n=63) Test to Follow-Up T2 a,b,c,d Caregiver Dietary Intake Fruit intake (svgs/d)e 0.74 (1.2) 1.02 (1.2) 0.28 (1.4) Whole fruit intake (svgs/d) 0.76 (0.9) 0.64 (0.9) -0.12 (1.3) Vegetable intake (svgs/d) 2.85 (2.1) 3.01 (1.7) 0.16 (2.8) Sugar sweetened beverage intake (svgs/d) 0.31 (0.7) 0.38 (0.7) 0.07 (0.78) Energy intake (kcals/d) 1713.78 (821.4) 1553.96 (563.2) -159.82 (999.66) HEI scoref 53.63 (16.1) 54.80 (10.2) 1.17 (18.46) HEI Total Fruit scoreg 1.28 (1.7) 1.69 (2.0) 0.41 (1.9) HEI Whole Fruit scoreh 2.23 (3.3) 1.96 (2.0) -0.27 (2.2) HEI Total Vegetable scorei 3.09 (3.3) 3.54 (1.6) 0.45 (2.1) Table 29. (continued)

237 Table 29. Follow-Up Period: Within-Group (Intervention Group) Differences in Caregiver and Family Outcome Caregiver Athropometrics/BPa,b BMI (kg/m2) 31.26 (9.5) 31.18 (9.4) -0.18 (1.9) Waist circumference (cm) 100.52 (20.2) 99.07 (20.6) -0.66 (11.7) Systolic BP (mm Hg) 126.59 (15.6) 127.61(18.7) 0.66 (13.1) Diastolic BP (mm Hg) 80.21 (11.4) 78.92 (12.2) -0.98 (12.7) a,b ÷≥…

238 Personal Determinants Self-efficacyj 102.65 (12.7) 104.09 (12.4) 1.68 (9.6) Menu Planningk 26.88 (4.3) 27.27 (3.4) 0.38 (2.6) Family Mealsa,b Family breakfast frequency (meals/wk) 3.20 (2.4) 4.22 (2.3) 1.02 (3.0)l Family dinner frequency (meals/wk) 4.41 (2.1) 4.68 (1.9) 0.27 (1.9) Family meals in a dining area (meals/wk) 4.69 (2.1) 4.82 (1.9) 0.13 (2.6) Family meals TV viewing (meals/wk) 1.97 (1.8) 1.90 (2.1) -0.07 (2.4) aValues are Mean(SD) bWithin-group differences determined by paired samples t-test cDaily dietary intake determined by averaging daily intake across 3, caregiver-assisted, nonconsecutive 24hr dietary recalls collected using USDA 5-step multi-pass method [189, 197] dIntervention participants’ (n=63) completion of 24hr dietary recalls at T1: 0 recalls (n=22 (35%)); 1 recall (n=38 (60%)); 2 recalls (n=2 (3%)); 3 recalls (n=1 (2%)). Intervention participants’ (n=63) completion of 24hr dietary recalls at T2: 0 recalls (n= 22 (35%)); 1 recall (n=39 (65%)); 2 recalls (n=0 (0%)); 3 recalls (n=0 (0%)) eTotal fruit= whole fruit + 100% fruit juice fHealthy Eating Index (HEI), 0-100 with higher score representing better intake gHealthy Eating Index (HEI) total fruit score, 0-5 with higher score representing better intake hHealthy Eating Index (HEI) whole fruit score, 0-5 with higher score representing better intake iHealthy Eating Index (HEI) total vegetable score, 0-5 with higher score representing better intake j Self-efficacy for healthy dietary behaviors, 0-120 scale with higher score representing increased self-efficacy kMenu planning behaviors, 0-36 scale with higher score representing increased menu planning lWithin-Group difference of p<0.05

238 Discussion

The present study aimed to assess the effectiveness of a 10-week, quasi- experimental family meals intervention with waitlist control on caregiver- and family- level outcomes. The Simple Suppers intervention is unique in that it fills a gap in the family meals research as it is the first family meals intervention to target underserved families with children 4-10 years old. The elementary school years have been demonstrated to be a period for increased risk for weight gain [220], making it a critical period to target for childhood obesity prevention interventions.

However, in addition to assessing the effects of the intervention on the child- level, this study’s mixed methods, multi-component evaluation plan assessed the impact of the intervention on the caregiver- and family-level, as well. While some caregiver- level outcomes have been assessed in a limited number of family meals interventions [73,

232], these studies focused on assessing dietary intake and personal determinants of behavior. Therefore, to our knowledge, this is the first family meals intervention to expand the evaluation plan for caregiver outcomes to not only include dietary intake and personal determinants of behaviors, but also direct assessment of caregiver anthropometric (BMI, WC) and blood pressure (systolic, diastolic) outcomes.

Caregiver- and family-level outcomes were assessed at three time points to determine the effectiveness of the intervention during both the intervention period

(baseline T0 to post-test T1) and the hollow-up period (post-test T1 to follow-up T2).

239 We hypothesized caregiver- and family-level outcomes would improve more among caregivers and families, respectively, participating in the intervention than in the controls during the intervention period and that improvements in outcomes made during the intervention period would be maintained among caregivers and families who participated in the intervention.

The intervention effect on our main caregiver outcome, BMI, demonstrated a significant reduction in caregiver BMI at post-test among caregivers participating in the intervention relative to those in the control group. Surprisingly, not only were improvements in caregiver BMI maintained at 10-week follow-up period, but caregiver

BMI continued to decrease during the follow-up period, although not significant. This finding was interesting in that we saw a significant decrease in caregiver BMI among intervention caregivers relative to control caregivers, with no significant difference in dietary intake at post-test or 10-week follow-up. One potential explanation for this could be the challenges with assessing dietary intake (e.g., inaccurate participant reporting, difficulty remembering dietary intake from previous day), which may have been pronounced in the caregiver dietary intake assessment (versus the child assessment) due to participant burden as caregiver dietary intake was collected after assessing child dietary intake. Another potential mediator of the significant decrease in caregiver BMI at post-test and maintenance at 10-week follow-up could be physical activity; however, data on physical activity was not assessed because physical activity was not addressed in the

SS intervention. Finally, while we did not observe a significant by-group difference in

240 dietary intake at post-test, we did observe a significant within-group decrease in total energy intake from baseline to post-test (-254.17 kcals/day) among caregivers in the intervention group, which could potentially explain the significant decrease in BMI among intervention caregivers at post-test.

The decrease in BMI observed among intervention caregivers at post-test may explain the significant reduction in caregiver systolic BP observed among intervention caregivers relative to controls at post-test. Like BMI, the improvement in systolic BP observed at post-test was maintained among intervention caregivers during the follow-up period, with no differences by level of attendance during the program.

In addition to the intervention effects were observed on caregiver BMI and systolic BP, a significant increase in caregiver SE was observed among intervention caregivers relative to control caregivers following the Simple Suppers program.

Although not significant, caregiver SE continued to increase during the 10-week follow- up period among intervention caregivers, with no difference by level of attendance.

Caregivers face numerous barriers to offering frequent family meals, with work schedules and child after-school activities being the most commonly reported challenges

[233]. These barriers may explain why family meals interventions have consistently faced challenges with increasing the frequency of shared family meals among participating families [73, 74]. For this reason, the Simple Suppers study aimed to improve both the

241 frequency and the quality of shared family mealtime. Quality family mealtime entails multiple factors, including the nutritional quality of the foods being served and eaten, the location of the meal, and the presence/absence of TV. In addition, while previous family meals intervention typically just assessed frequency of family dinners, this intervention expanded its assessment of family meals to include both lunch and dinner.

Consistent with previous family meals interventions [73, 74], intervention effects were not observed on frequency of family dinners during the intervention and follow-up periods. However, while frequency of family breakfast did not change during the intervention period, a significant increase in family breakfast frequency was observed among intervention families during the follow-up period. Following the Simple Suppers program, improvements in family meal quality were also observed among intervention families, as mealtime TV viewing significantly declined relative to control families.

Several limitations should be noted. Lack of randomization in the study design was a limitation in this study. However randomization was not appropriate for this study because preserving sample size and developing trust with the site/participating families was paramount [213, 215]. We overcame this limitation by assessing between-group differences at baseline, of which none existed. A second limitation of this study was assessment of caregiver dietary intake. The dietary intake assessment protocol was to conduct three, nonconsecutive (two weekdays, one weekend day) 24hr dietary recalls using USDA’s 5-step multi-pass dietary recall method at each time point [230, 231]. The

242 first 24hr dietary recall was to be collected in-person during the data collection visit and the remaining two were to be collected via telephone [234]. Average dietary intake at each time point was to be determined by averaging intake across the three days. However due to challenges contacting participants, three 24hr dietary recalls were not always able to be collected at each time point. In these instances, if two recalls were conducted, the average of the two recalls would be used and if a single 24hr dietary intake was collected, the single recall was used.

Lastly, while program retention/completion was high (87.5%), connecting with families to complete the post-test T1 and follow-up T2 data collections was a challenge.

Recruitment and retention of underserved participants in community and research programming is more challenging than recruiting and engaging higher income participants due to issues with transportation, time demands and scheduling conflicts, mistrust with individuals outside of their community, and language and literacy barriers

[235]. This study sought to minimize these barriers by letting participants select the date, time and location (community center or participant home) of data collection visits.

Researchers leading data collection visits were trained to identify and overcome language and literacy barriers with participants to minimize the impact of this potential barrier.

The Simple Suppers program was also strategically designed to build trust with the target population. Implementing the intervention at a faith-based community center that was located in a neighborhood and served as a resource for the target population provided a sense of familiarity for participating families. Program trust was further established by

243 integrating staff of the community center into the staffing structure of the program. In addition, participating families received a grocery store gift card following each data collection visit to thank them for their time. While these efforts were successful in recruiting and retaining the target population in programming, challenges connecting with families (e.g., disconnected telephone numbers, limited access to email/internet, alternating work schedules) to complete the post-test T1 and follow-up T2 data collection visits existed. Future researchers in this area may consider completing post-test data collection immediately following programming (i.e., at the end of the last lesson).

244 Chapter 7. Discussion

245 Childhood obesity continues to be a major public health concern here in the U.S. and globally. While rates of childhood obesity have plateaued in the U.S. over the past decade, the rate at which this plateau has occurred is still significantly higher than thirty years ago and rates of severe obesity continue to rise [2, 90]. Furthermore, there continues to be distinct racial/ethnic and socioeconomic disparities in the prevalence of childhood obesity, with higher rates among Hispanic and Black children and children from underserved households [2, 11, 236]. As the childhood obesity epidemic continues to have detrimental effects on the health and well-being of millions of children across the

U.S., the need for prevention interventions is at an all-time high.

Family meals has become a growing area of research for childhood obesity prevention, as an abundance of cross-sectional studies have demonstrated a positive association between family meal frequency and improved child health outcomes, including child diet quality (increased fruit and vegetable intake; decreased SSB intake)

[86, 221, 237, 238] and weight status (decreased rates of childhood obesity) [217, 221].

The evidence linking family meals with improved child health outcomes is so strong, that the American Academy of Pediatrics Evidence-Based Recommendations on Childhood

Obesity Prevention recommend participation in family meals as a childhood obesity prevention strategy [40]. Furthermore, the recently published 2015 Dietary Guidelines

Advisory Report’s Needs for Future Research included the need to conduct additional family meals research to fill gaps that exist in this growing line of research [239].

246 Family meal frequency in the U.S. has been consistently, relatively high at an average of about 5 family meals per week [240]. However, despite the frequency of family meals being relatively high, the literature demonstrates that caregivers face multiple barriers to offering quality family meals at home on a regular basis [233, 241].

In particular, these barriers have been reported to be greatest among minority and underserved families. And by quality family meals, we mean offering nutritious foods, creating a conducive family meals environment where meals are eaten in a dining area without the TV on, and where family mealtime behavioral expectations are established and followed.

The limited number of family meals interventions that have been implemented have aimed to improve child diet quality and weight status by increasing frequency of family meals [73, 74, 232]. Interestingly, the vast majority of these interventions have been unsuccessful in increasing family meal frequency and have seen minimal improvements in child diet quality and weight status [73, 74]. In considering these interventions, it is important to recognize several limitations of these initial interventions.

First, as previously discussed, the majority of the current family meals research has aimed to improve child diet quality and weight status by increasing frequency of family meals [73, 74, 232]. Overall in the U.S., family meal frequency is relatively high at an average of about 5 family meals per week [240]. While further increasing family meal frequency may result in improved child health outcomes, this may not be feasible,

247 as caregivers face multiple barriers to offering family meals at home on a regular basis, with work schedules and child after-school activities being the greatest barriers [233, 241,

242]. Therefore, in an effort to improve child health outcomes, family meals interventions may need to focus on improving the quality of family meals since increasing the frequency may not be feasible. The Simple Suppers study is unique in that it was specifically designed to help caregivers overcome barriers to family meals in an effort to increase both the frequency and the quality of family meals in order to improve child and caregiver health outcomes [218].

Second, the majority of family meals interventions to date have targeted predominantly White, 8-12 year old children from economically stable households [73,

74], despite the literature demonstrating children of racial/ethnic minority and from underserved households are at highest risk of obesity [11, 236, 243]. The Simple Suppers study aimed to fill this gap in the literature by targeting 4-10 year old children of racial minority from underserved households [218]. A recent family meals intervention targeting 8-12 year old children demonstrated great potential for earlier intervention, as a significant intervention effect was observed on child BMI z-score only among prepubescent children [74]. In addition, the elementary school years have been demonstrated in the literature to be a period of increased risk for weight gain, making childhood obesity prevention interventions critical during these years [220].

248 Lastly, the majority of the current research fails to examine the child health impact of family meals beyond BMI, with only a small number of studies including additional outcomes (e.g., disordered eating) [86–89]. Along with the need to expand the assessment of child health outcomes, the caregiver health impact of family meals is severely understudied. As caregivers serve as ‘agents of change’ and are responsible for establishing the home food environment, understanding the caregiver health impact of family meals has potential to improve the health and well-being of an entire family. This study sought to better understand the child health impact of family meals by expanding assessment beyond BMI, to include WC and BP (systolic and diastolic). In addition, to better understand the impact of family meals on caregiver health, this study included a thorough assessment of caregiver health outcomes, which included BMI, WC, and BP

(systolic and diastolic).

The Simple Suppers study exploratory findings suggest great potential for family meals programming to be an effective strategy to prevent excess weight gain among children, as a significant decrease in BMI z-score was observed among intervention children attending 7 or more classes relative to intervention children attending less than 7 classes (p=0.003) and control children (p=0.048). Interestingly, while level of attendance had a significant impact on child BMI z-score, this trend was not observed in caregivers.

Relative to caregivers in the control group, a significant decrease (p=0.028) in caregiver

BMI was observed among all intervention caregivers, regardless of level of attendance, following the 10-week Simple Suppers program. This finding suggests that program

249 level of attendance may have a greater impact on child weight status than caregivers, with higher levels of program attendance being necessary for improved outcomes.

Expanding child and caregiver health outcomes beyond BMI demonstrated additional positive intervention effects on participant health. Similar to child BMI z- score, a significant decrease in systolic BP (p=0.047) was observed among intervention children attending 7 or more classes relative to control children at post-test. Caregiver systolic BP also decreased significantly (p=0.043) at post-test among all intervention caregivers, regardless of level of attendance, relative to control caregivers. Once again, this finding further suggests that program level of attendance may have greater impact on children than caregivers, with regards to health outcomes.

However, while program attendance had a significant effect on child health outcomes (BMI z-score, systolic BP), this effect was not observed on personal determinants of child behavior (i.e., child food preparation skills/frequency). At post- test, a significant increase in child food preparation skills (p<0.001) and frequency of participation (p=0.003) was observed among all intervention children relative to control children, regardless of level of attendance.

Consistent with the literature [73, 74], improvements in family meal (breakfast and dinner) frequency were not observed. However, in expanding the focus and evaluation plan of the Simple Suppers intervention beyond just family meal frequency,

250 improvements in family meal quality were observed following the 10-week intervention.

Frequency of watching TV during family mealtime significantly declined (p=0.028) among intervention families, relative to control families, with the greatest decline being among families attending 7 or more lessons.

Engagement in the SS family meals program was associated with improvements in weight status among child and caregiver participants, which may have been mediated by improvements in child dietary intake (increased total and whole fruit intake) and behavioral capability (increased food preparation skills and frequency), caregiver self- efficacy for healthy dietary behaviors, and family meal quality (decreased mealtime TV viewing). At the child, caregiver, and family level, these results coincide with theoretical predictors and the literature. At the child level, the SCT and literature demonstrate a positive association between fruit and vegetable intake (behavior) and food preparation skills/frequency (behavioral capability construct), which was observed in this intervention. The literature further demonstrates an inverse association between child fruit and vegetable intake and BMI z-score, which was observed at completion of the SS family meals program as well. At the family level, child fruit and vegetable intake

(behavior) is consistently inversely associated with family mealtime TV viewing (socioenvironmental construct), which again was observed in this intervention. Furthermore, in the present study, a significant decrease in caregiver BMI and increase in self-efficacy

(personal construct) for healthy dietary behaviors was observed at post-test. This

251 observation aligns with the SCT and literature, which consistently demonstrate a positive association between goal achievement (i.e., decreased BMI) and self-efficacy.

The Simple Suppers intervention had several strengths. The program curriculum and intervention were of high quality, as they were evidence- and theory-based and demonstrated high fidelity. The extensive intervention evaluation plan not only expanded child health outcomes beyond BMI, but included a thorough assessment of caregiver health outcomes, as well. All anthropometric and BP outcomes were assessed by trained researchers, rather than self-report data. Furthermore, the evaluation plan assessed sustainability of intervention effectiveness by assessing outcomes at both post-test and

10-week follow-up.

This study also expanded the family meals literature by targeting 4-10 year old child of racially minority from underserved households. Implementing the intervention at a faith-based community center in a neighborhood of the target population fostered high rates of participant recruitment and program retention. Similarly, implementing the intervention in a ‘real-world’ community center demonstrated the potential sustainability of the program through successful partnerships with community organizations.

252 Several limitations of the intervention should be noted. Lack of randomization in the study design was a limitation in this study. However randomization was not appropriate for this study because preserving sample size and developing trust with the site/participating families was paramount [213, 215]. We overcame this limitation by assessing between-group differences at baseline, and controlling for differences in the analyses. A second limitation of this study was assessment of child and caregiver dietary intake. The dietary intake assessment protocol was to conduct three, nonconsecutive (two weekdays, one weekend day) 24hr dietary recalls using USDA’s 5-step multi-pass dietary recall method at each time point [230, 231]. The first 24hr dietary recall was to be collected in-person during the data collection visit and the remaining two were to be collected via telephone [234]. Average dietary intake at each time point was to be determined by averaging intake across the three days. However due to challenges contacting participating families, three 24hr dietary recalls were not always able to be collected at each time point. In these instances, if two recalls were conducted, the average of the two recalls would be used and if a single 24hr dietary intake was collected, the single recall was used. Future studies should consider using alternative dietary assessment methods that can be completed in a single session, such as a food frequency questionnaire.

In the present study, physical activity was not assessed at any of the three data collection time points because the intervention did not educate on or address physical activity in any of its three program components (child education; caregiver education;

253 group family meal). However, because the intervention promoted overall health and wellness, participants may have altered their physical activity behaviors as a result of participating in the intervention [119]. In addition, because engagement in physical activity can mediate changes in weight status [119], future family meals research should assess physical activity to help better understand changes in lifestyle behaviors and changes in participant weight status.

Lastly, while program retention/completion was high (87.5%), connecting with families to complete the post-test T1 and follow-up T2 data collections was a challenge.

[235]. This study sought to minimize these barriers by letting participants select the date, time and location (community center or participant home) of data collection visits.

Researchers leading data collection visits were trained to identify and overcome language and literacy barriers with participants to minimize the impact of this potential barrier.

254 established by integrating staff of the community center into the staffing structure of the program. In addition, participating families received a grocery store gift card following each data collection visit to thank them for their time. While these efforts were successful in recruiting and retaining the target population in programming, challenges connecting with families (e.g., disconnected telephone numbers, limited access to email/internet, alternating work schedules) to complete the post-test T1 and follow-up T2 data collection visits existed. Future researchers in this area may consider completing post-test data collection immediately following programming (i.e., at the end of the last lesson).

The positive impact of the SS intervention at the child, caregiver, and family level demonstrates great potential for family meals programming. However, changes to the program and curriculum has potential to produce even greater outcomes. The fidelity assessment of the caregiver education demonstrated that the goal setting component of the caregiver education curriculum had high fidelity and excellent engagement. As SCT and goal setting theory demonstrate the importance and effectiveness of self-regulation in behavior change, expanding the goal setting portion of the caregiver education curriculum has great potential to produce even greater intervention impacts.

This effectiveness trial demonstrated level of attendance to have a significant on child anthropometric outcomes, but not caregiver anthropometric outcomes. More specifically, the findings demonstrated significant improvements in child BMI and

255 systolic BP z-scores only among children in the high attendance intervention group, while improvements in BMI and systolic BP were observed among all caregiver participants, regardless of level of attendance. These findings demonstrated that children may require more intensive and overall program engagement to produce positive anthropometric outcomes. Given this, the current SS curriculum may have potential to produce an even greater impact on child anthropometric and BP outcomes by increasing the intensity and overall of engagement of child participants in the programming.

Finally, the current SS program had an expansive evaluation plan as it assessed outcomes at the child, caregiver, and family level. With weight status being the main intervention outcome at the child and caregiver level, potential mediators of participant weight status were assessed at the child and caregiver level (i.e., dietary, anthropometric, cognitive/behavioral capability), as well as the family level (family meal frequency and quality). At post-test, improvements in weight status were observed at the child and caregiver level. Post-test improvements in several of our potential mediators (child: dietary intake (increased total and whole fruit), caregiver: cognitive (increased self- efficacy, family: family meal quality (decreased family meal TV viewing)) may explain the improvements in weight status that were observed. Interestingly, by-group improvements in dietary intake (increased total and whole fruit intake) were observed only among high attendance children at post-test, yet improvements in caregiver weight status were observed. Although by-group differences in dietary intake were not observed among intervention caregivers at post-test, weight status improvements among

256 intervention caregivers at post-test may have resulted from a significant within-group decrease in daily total energy intake. While a correlation exists between dietary intake and weight status, the challenges associated with accurately assessing dietary intake, along with the numerous other factors impacting weight status (e.g., physical activity, food environment), further assessment of potential mediators, such as physical activity, caregiver role modeling, and social support, may provide further insight into the positive impact of the intervention on child and caregiver weight status.

Future family meals research is needed in several areas to continue to fill the gaps in this growing line of research. The positive intervention effects observed on child and caregiver weight status, systolic BP and personal determinants (i.e., children: food preparation skills and frequency of participation; caregiver: SE for healthy dietary behaviors) warrants further assessment of the Simple Suppers program. As a community-based program, a critical next step is to assess program sustainability. In this study, the Simple Suppers program was a success because of the mutual partnership established between the faith-based community center where the program was implemented and the research team delivering and overseeing the program. However, to assess program sustainability, the Simple Suppers program should be implemented and assessed for effectiveness without the resources (e.g., funding, staffing, participant incentives) of the research team. One opportunity to assess program sustainability would be to implement the Simple Suppers program through the Supplemental Nutrition

Assistance Program-Education (SNAP-ED). Given that the caregiver component of the

257 Simple Suppers curriculum aligns with the format and theoretical framework of the

SNAP-ED curriculum, there is great potential for assessing the sustainability of the

Simple Suppers program by delivering it through the SNAP-ED program.

Another area of future research for the Simple Suppers study is to extend the follow-up period assessment to determine more long-term intervention effects on child-, caregiver-, and family-level outcomes. In particular, extending the follow-up period on child and caregiver BMI z-score and BMI, respectively, will provide insight on the potential long-term effects of the intervention on child and caregiver weight status. In this study, intervention children attending 7 or more lessons had a significant decrease in

BMI z-score post-program relative to intervention children attending less than 7 lessons and control children, while all intervention caregivers had a significant decrease in BMI relative to control caregivers post-program. These improvements in weight status among intervention children attending 7 or more lessons and intervention caregivers were maintained during the 10-week follow-up period. It is well established in the literature that the inverse association between family meal frequency and child weight status/BMI z-score is strongest among younger children [221, 222] and family meal frequency is inversely associated with child age, with decreasing rates of family meal frequency occurring as children age [233]. Assessing child-, caregiver-, and family-level outcomes up to 5 years post-program will provide insight on whether the positive intervention effects can be sustained and refute the decline in family meal frequency and weakened

258 inverse association between family meal frequency and child weight status that is consistently seen in the literature.

Along with extending the follow-up period, implementing a ‘low-touch’ follow- up period intervention has the potential to produce even greater intervention effects. In the current study, maintenance of behavior was observed among all child-, caregiver-, and family-level outcomes. However, a low-touch follow-up period intervention has the potential to extend and continue improvements in outcomes observed during the intervention period, rather than maintenance of outcomes.

This study was designed as a community-based childhood obesity prevention intervention. For this reason, families with children 4-10 years old were eligible to participate in the study, regardless of child weight status/BMI z-score, resulting in the mean child BMI z-score of all participating children (intervention and control) at baseline to be 0.69 (1.2). It is exceptional to have seen a positive intervention effect on child BMI z-score given that children were, on average, a healthy weight at baseline. If a child weight status/BMI z-score cut point was used in the eligibility criteria of this study (e.g.,

BMI z-score >0, which is equivalent to the 50%tile), a more pronounced intervention effect would be expected among the heavier children [74]. Given the success of the program, it is worthy to consider slightly modifying the program and assessing its effectiveness as a childhood obesity treatment program. With a treatment focus, the

259 program could continue to be implemented in a community setting or be expanded to clinical out-patient settings.

Conclusions

Engagement in the Simple Suppers family meals program was associated with improvements in weight status among child and caregiver participants, which may have been mediated by improvements in child dietary intake (increased total and whole fruit) and behavioral capability (increased food preparation skills and frequency), improvements in caregiver self-efficacy for healthy dietary behaviors, and improvements in the quality of family meals (decreased mealtime TV viewing). Results from this study demonstrate the potential for engagement in an evidence-based family meals program to positively impact child and caregiver weight status among a racially diverse sample of 4-

10 year old children and their caregivers from underserved families. The positive intervention impact observed on child weight status is of utmost importance given the ongoing childhood obesity epidemic.

260 Bibliography

261 1. Description of the Global Burden of Non-Communicable Diseases, Their Risk Factors and Determinants. World Health Organization. Geneva, Switzerland; 2011.

2. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA. 2014; 311:806–814.

3. Popkin BM, Conde W, Hou N, Monteiro C. Is there a lag globally in overweight trends for children compared with adults? Obesity. 2006; 14:1846–b53.

4. Ogden C, Flegal KM, Carroll MD. Prevalence and trends in overweight among U.S. children and adolescents, 1999-2000. JAMA. 2002; 288:1728–1732.

5. Johnson S. The Nation's Childhood Obesity Epidemic: Health Disparities in the Making. Volume 2012, New York; 2012.

6. Strauss R, HA P. Epidemic increase in childhood overweight, 1986-1998. JAMA. 2001; 286:2845–2848.

7. Poverty in the United States, 2012. US Census Bureau. Washington D.C.; 2013. [https://www.census.gov/hhes/www/poverty/about/overview/] Accessed Nov 2016.

8. DeNavas-Walt C, Proctor BD, Smith JC. Income, Poverty and Health Insurance in the United States, 2010. Washington, DC; 2010.

9. Levine JA. Poverty and obesity in the U.S. Diabetes. 2011; 60:2667–2668.

10. Hill J, Peters J. Environmental contributions to the obesity epidemic. Science. 1998; 280:1371–1374.

11. Roberts B, Povich D, Mather M. Low-Income Working Families: The Growing Economic Gap. Working Poor Families Project. Washington, D.C.; 2013. [http://www.workingpoorfamilies.org/wp-content/uploads/2013/01/Winter-2012_2013- WPFP-Data-Brief.pdf] Accessed Oct 2016.

12. Anderson PM, Butcher KF, Kristin F. Childhood Obesity: Trends and Potential Causes. Futur Child. 2006; 16:19–45.

13. Dietz WH, Story MT, Leviton LC. Introduction to issues and implications of screening, surveillance, and reporting of children’s BMI. Pediatrics. 2009; 124 Suppl:S1- 2.

14. Dietz WH, Bellizzi MC. Introduction: the use of body mass index to assess obesity in children. Am J Clin Nutr. 1999, 70:S123–125.

262 15. Freedman DS, Sherry B. The validity of BMI as an indicator of body fatness and risk among children. Pediatrics. 2009; 124 Suppl:S23-34.

16. Himes JH. Challenges of accurately measuring and using BMI and other indicators of obesity in children. Pediatrics. 2009; 124 Suppl:S3-22.

17. Obesity: Preventing and Managing the Global Epidemic. World Health Organization. Geneva, Switzerland; 2000.

18. Kuczmarski R, Ogden C, Guo S, Grummer-Strawn L, Flegal K, Mei Z. 2000 CDC growth chart for the United States: methods and development. Vital Heal Stat. Ser. 2002; 246:1–190.

19. Casues and risks for obesity in children. MedlinePlus. Washington, D.C.; 2017. [http://www.nlm.nih.gov/medlineplus/ency/patientinstructions/000383.htm] Accessed Jan 2017.

20. Fisher JO, Rolls BJ, Birch LL. Children’s bite size and intake of an entrée are greater with large portions than with age-appropriate or self-selected portions. Am J Clin Nutr. 2003; 77:1164–1170.

21. Devaney B, Zieglar P, Pac S, Karwe V, Barr S. Nutrient intakes of infants and toddlers. J Am Diet Assoc. 2004; 35:S14-21.

22. Savage J, Fisher J, Birch LL. Parental influence on eating behavior: conception to adolescence. J Law Med Ethics. 2007; 35:22–34.

23. Scaglioni S, Arrizza C, Vecchi F, Tedeschi S. Determinants of children’s eating behavior. Am J Clin Nutr. 2011; 94(6 Suppl):S2006–11.

24. Leech RM, McNaughton SA, Timperio A. The clustering of diet, physical activity and sedentary behavior in children and adolescents: a review. Int J Behav Nutr Phys Act. 2014; 11:4.

25. Harris J, Schwartz M, Munsell C, Dembek C, Liu S, LoDolce M: Fast Food FACTS 2013: Measuring Progress in Nutrition and Marketing to Children and Teens. Yale Rudd Center for Food Policy & Obesity. New Haven, CT; 2013. [http://www.rwjf.org/content/dam/farm/reports/reports/2013/rwjf408549] Accessed Mar 2016.

26. Jacobs E, Shipp S. How Family Spending Has Changed in the U.S. Mon Labor Rev. 1990; 113:20-7.

263 27. U.S. Consumer Expenditures - 2010. Bureau of Labor Statistics, Washington, DC; 2011. [https://www.bls.gov/news.release/archives/cesan_09272011.pdf] Accessed Dec 2014.

28. Chou S, Grossman M, Saffer H. An economic analysis of adult obesity: Results from the Behavioral Risk Factor Surveillance System. J Health Econ. 2004; 23:565–587.

29. Ma Y. Association between Eating Patterns and Obesity in a Free-living US Adult Population. Am J Epidemiol. 2003; 158:85–92.

30. Lin B, Guthrie J, Frazão E. Nutrient Contribution of Food Away From Home. United States Department of Agriculture, Washington, D.C.; 1995. [https://www.ers.usda.gov/webdocs/publications/aib750/5841_aib750l_1_.pdf?v=41055] Accessed Feb 2015.

31. Reedy J, Krebs-Smith S. Dietary sources of energy, solid fats, and added sugars among children and adolescents in the United States. J Am Diet Assoc. 2010; 110:1477– 1484.

32. Han E, Powell L. Consumption patterns of sugar- sweetened beverages in the United States. J Acad Nutr Diet. 2013; 113:43–53.

33. Jacobson M. Liquid Candy: How Soft Drinks Are Harming Americans’ Health. Center for Science in the Public Interest. Washington, DC; 2005. [http://www.kickthecan.info/sites/default/files/documents/Jacobson%202005_liquid%20c andy.pdf] Accessed Nov 2014.

34. Ludwig D, Peterson K, Gortmaker S. Relationship between consumption of sugar- sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet. 2001; 357:505–508.

35. Harris J, Schwartz M, LoDolce M, Munsell C, Fleming-Milici F, Elsey J, Liu S, Hyary M, Gross R, Hazen C, Dembek C. Sugary Drink FACTS 2014: Some Progress but Much Room for Improvement in Marketing to Youth. Rudd Center for Food Policy & Obesity. New Haven, CT; 2014. [http://www.sugarydrinkfacts.org/resources/sugarydrinkfacts_report.pdf] Accessed Nov 2014

36. Fernandes M. The effect of soft drink availability in elementary schools on consumption. J Am Diet Assoc. 2008; 108:1445–1452.

37. Andreyeva T, Kelly I, Harris J. Exposure to food advertising on television: associations with children’s fast food and soft drink consumption and obesity. Econ Hum Biol. 2011; 9:221–233.

264 38. Wiecha J, Finkelstein D, Troped P, Fragala M, Peterson K. School vending machines use and fast–food restaurant use are associated with sugar-sweetened beverage intake in youth. J Am Diet Assoc. 2006; 106:1624–1630.

39. Results from the School Health Policies and Practices Study. Centers for Disease Control and Prevention. Atlanta, Georgia; 2012. [https://www.cdc.gov/healthyyouth/data/shpps/index.htm] Accessed Oct 2015.

40. Barlow S. Expert committee recommendation regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: summary report. Pediatrics. 2007; 120:S164–S192.

41. Johnson L, Delva J, O’Malley P. Sports participation and physical education in American secondary schools. Am J Prev Med. 2007; 33:S195–207.

42. Barriers to Children Walking to or from School - United States, 2004. Centers for Disease Control and Prevention. Washington, DC; 2005. [https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5438a2.htm] Accessed Oct 2016.

43. Rideout V, Foehr U, Roberts D. Generation M2: Media in the Lives of 8-to-18 Year Olds. Kaiser Family Foundation. Menlo Park, CA; 2010.

44. Tremblay MS, LeBlanc AG, Kho ME, Saunders TJ, Larouche R, Colley RC, Goldfield G, Connor Gorber S. Systematic review of sedentary behaviour and health indicators in school-aged children and youth. Int J Behav Nutr Phys Act. 201;, 8:98.

45. Iannotti RJ, Wang J. Trends in physical activity, sedentary behavior, diet, and BMI among US adolescents, 2001-2009. Pediatrics. 2013; 132:606–14.

46. Koplan J, Liverman C, Kraak V (Eds): Preventing Childhood Obesity: Health in the Balance. Washington, DC: National Academies Press; 2005.

47. Mokdad A, Marks J, Stroup D, Gerberding J. Actual causes of death in the United States, 2000. JAMA. 2004; 291:1238–1245.

48. Olshansky S, Passaro D, Hershow R, Layden J, Carnes B, Brody J, Hayflick L, Butler RN, Allison DB, Ludwig DS. A Potential Decline in Life Expectancy in the United States in the 21st Century. N Engl J Med. 2005; 352:1138–1145.

49. Freedman D, Mei Z, Srinivasan S, Berenson G, Dietz W. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study. J Pediatr. 2007; 150:12–17.

265 50. Whitlock E, Williams S, Gold R, Smith P, Shipman S. Screening and interventions for childhood overweight: a summary of evidence for the US Preventive Services Task Force. Pediatrics. 2005; 116:e125—144.

51. Han J, Lawlor D, Kimm S. Childhood obesity. Lancet. 2010; 375:1737—1748.

52. Taylor E, THeim K, Mirch M, Ghorbani S, M T-K, Adler-Wailes D, Brady S. Orthopedic complications of overweight in children and adolescents. Pediatrics. 2006; 117:2167–2174.

53. Dietz W. Health consequences of obesity in youth: Childhood predictors of adult disease. Pediatrics. 1998; 101: 518–525.

54. Swartz M, Puhl R. Childhood obesity: a societal problem to solve. Obes Rev. 2003; 4:57–71.

55. Biro F, Wien M. Childhood obesity and adult morbidities. Am J Clin Nutr. 2010; 91:S1499–1505.

56. Whitaker R, Wright J, Pepe M, Seidel K, Dietz W. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 1997; 37:869-873.

57. Serdula M, Ivery D, Coates R, Freedman D, Williamson D, Byers T. Do obese children become obese adults? A review of the literature. Prev Med. 1993; 22:167–177.

58. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung, and Blood INstitute. Bethesda, MD; 1998. [https://www.ncbi.nlm.nih.gov/books/NBK2003/]. Accessed Nov 2016.

59. Finkelstein E, Trogdon J, Cohen J, Dietz W. Annual Medical Spending Attributable to Obesity: Payer- and Service-specific Estimates. Health Aff. 2009; 28:822–831.

60. Trasande L, Chatterjee S. The impact of obesity on health service utilization and costs in childhood. Obesity. 2009; 17:1749–1754.

61. Population-Based Approach to Childhood Obesity Prevention. World Health Organization. Geneva, Switzerland; 2012. [http://www.who.int/dietphysicalactivity/childhood/WHO_new_childhoodobesity_PREV ENTION_27nov_HR_PRINT_OK.pdf]. Accessed Oct 2016.

62. Hoelscher DM, Kirk S, Ritchie L, Cunningham-Sabo L. Position of the Academy of Nutrition and Dietetics: interventions for the prevention and treatment of pediatric overweight and obesity. J Acad Nutr Diet. 2013; 113:1375–1394.

266 63. Pratt C, Stevens J, Daniels S. Childhood Obesity Prevention and Treatment Recommendations for Future Research. Am J Prev Med. 2009; 35:249–252.

64. Davis MM, Gance-Cleveland B, Hassink S, Johnson R, Paradis G, Resnicow K. Recommendations for prevention of childhood obesity. Pediatrics. 2007; 120 Suppl: S229-253.

65. Epstein L, Valoski A, Wing RR, Mccurley J. Ten-year outcomes of behavioral family-based treatment for childhood obesity. Heal Psychol.1994; 13:373–383.

66. Golan M, Weizman A, Apter A, Fainaru M. Parents as the exclusive agents of change in the treatment of childhood obesity. Am J Clin Nutr. 1998; 67:1130–1135.

67. Golan M, Crow S. Targeting parents exclusively in the treatment of childhood obesity: long-term results. Obes Res. 2004; 12:357–361.

68. Cook E, Dunifon R. Do Family Meals Really Make a Difference? Cornell University Extension. Ithaca, New York; 2011.

69. Spagnola M, Fiese BH. Family Routines and Rituals. Infants Young Child. 2007; 20:284–299.

70. Fiese BH, Foley KP, Spagnola M. Routine and Ritual Elements in Family Mealtimes: Contexts for Child Well- Being and Family Identity. In: New Directions for Child and Adolescent Development. Volume 2006. WIley Periodicals, Inc.; 2006:67–89.

71. Fiese BH, Hammons A, Grigsby-Toussaint D. Family mealtimes: a contextual approach to understanding childhood obesity. Econ Hum Biol. 2012; 10:365–374.

72. Fiese BH, Schwartz M. Reclaiming The Family Table: Mealtimes and Child Health and Wellbeing. Institute of Education Sciences. Ann Arbor, MI. 2008. [https://eric.ed.gov/?id=ED521697]. Accessed Dec 2016.

73. Fulkerson J a, Rydell S, Kubik MY, Lytle L, Boutelle K, Story M, Neumark-Sztainer D, Dudovitz B, Garwick A. Healthy Home Offerings via the Mealtime Environment (HOME): feasibility, acceptability, and outcomes of a pilot study. Obesity. 2010; 18 Suppl:S69-74.

74. Fulkerson J a, Neumark-Sztainer D, Story M, Gurvich O, Kubik MY, Garwick A, Dudovitz B.The Healthy Home Offerings via the Mealtime Environment (HOME) Plus study: design and methods. Contemp Clin Trials. 2014; 38:59–68.

267 75. Woodruff SJ, Hanning RM, McGoldrick K, Brown KS. Healthy eating index-C is positively associated with family dinner frequency among students in grades 6-8 from Southern Ontario, Canada. Eur J Clin Nutr. 2010; 64:454–460.

76. Christian MS, Evans CEL, Hancock N, Nykjaer C, Cade JE. Family meals can help children reach their 5 a day: a cross-sectional survey of children’s dietary intake from London primary schools. J Epidemiol Community Health. 2013; 67:332–338.

77. Fink SK, Racine EF, Mueffelmann RE, Dean MN, Herman-Smith R. Family meals and diet quality among children and adolescents in North Carolina. J Nutr Educ Behav. 2014; 46:418–422.

78. Larson NI, Story M, Eisenberg ME, Neumark-Sztainer D. Food preparation and purchasing roles among adolescents: associations with sociodemographic characteristics and diet quality. J Am Diet Assoc. 2006; 106:211–218.

79. Larson N, Fulkerson J, Story M, Neumark-Sztainer D. Shared meals among young adults are associated with better diet quality and predicted by family meal patterns during adolescence. Public Health Nutr. 2013; 16:883–893.

80. Wansink B, van Kleef E. Dinner rituals that correlate with child and adult BMI. Pediatr Obes. 2014; 22:e91-95.

81. Chan J, Sobal J. Family meals and body weight. Analysis of multiple family members in family units. Appetite. 2011; 57:517–524.

82. Berge JM, Wall M, Hsueh T, Fulkerson JA, Larson N, Neumark-sztainer D. The Protective Role of Family Meals for Youth Obesity: 10-Year Longitudinal Associations. J Pediatr. 2015; 166:296–301.

83. Gable S, Chang Y, Krull JL. Television watching and frequency of family meals are predictive of overweight onset and persistence in a national sample of school-aged children. J Am Diet Assoc. 2007; 107:53–61.

84. Rollins BY, Belue RZ, Francis LA. The beneficial effect of family meals on obesity differs by race, sex, and household education: the national survey of children’s health, 2003-2004. J Am Diet Assoc. 2010; 110:1335–1339.

85. Fulkerson JA, Friend S, Flattum C, Horning M, Draxten M, Neumark-Sztainer D, Gurvich O, Story M, Garwick A, Kubik MY. Promoting healthful family meals to prevent obesity: HOME Plus, a randomized controlled trial. Int J Behav Nutr Phys Act. 2015; 12:154.

268 86. Hammons AJ, Fiese BH. Is frequency of shared family meals related to the nutritional health of children and adolescents? Pediatrics. 2011; 127:e1565–1574.

87. Fulkerson JA, Neumark-Sztainer D, Hannan PJ, Story M. Family meal frequency and weight status among adolescents: cross-sectional and 5-year longitudinal associations. Obesity. 2008; 16:2529–2534.

88. Neumark-Sztainer D. Eisenberg ME. Fulkerson JA. Story M. Larson N. Family meals and disordered eating. Arch Pediatr Adolesc Med. 2008; 162:17–22.

89. Neumark-Sztainer D, Wall M, Story M, Fulkerson JA. Are family meal patterns associated with disordered eating behaviors among adolescents? J Adolesc Heal. 2004; 35:350–359.

90. Skinner AC, Perrin EM, Skelton JA. Prevalence of obesity and severe obesity in US children, 1999-2014. Obesity. 2016; 24:1116–1123.

91. Birch L, Arbor A, Savage JS, Ventura A. From Infancy to Adolescence. Can J Diet Pr Res. 2009; 68:1–11.

92. Hunter J, Cason K. Kids in the Kitchen. Clemson University. South Carolina; 2008.

93. Guthrie J, Lin B, Frazao E. Role of food prepared away from home in the American diet, 1977-78 versus 1994-96: changes and consequences. J Nutr Educ Behav. 2002; 34:140–150.

94. Mancino L, Todd J, Guthrie J, Lin B. How Food Away From Home Affects Children’s Diet Quality. St. Louis, MI; 2010. [https://ideas.repec.org/p/ags/uersrr/134700.html]. Accessed Oct 2016.

95. Gillman M, Rifas-Shiman S, Frazier A, Rockett H, Cambargo C, Field A, Berkey C, Colditz G. Family dinner and diet quality among older children and adolescents. Arch Fam Med. 2000; 9:235–240.

96. Larson NI, Neumark-Sztainer D, Hannan PJ, Story M. Family meals during adolescence are associated with higher diet quality and healthful meal patterns during young adulthood. J Am Diet Assoc. 2007; 107:1502–1510.

97. Robson SM, Crosby LE, Stark LJ. Eating dinner away from home: Perspectives of middle-to high-income parents. Appetite. 2016; 96:147–153.

98. Appelhans BM, Waring ME, Schneider KL, Pagoto SL. Food preparation supplies predict children’s family meal and home-prepared dinner consumption in low-income households. Appetite. 2014;76:1–8.

269 99. Chu YL, Farmer A, Fung C, Kuhle S, Storey KE, Veugelers PJ. Involvement in home meal preparation is associated with food preference and self-efficacy among Canadian children. Public Health Nutr. 2013; 16:108–112.

100. Chu YL, Storey KE, Veugelers PJ. Involvement in meal preparation at home is associated with better diet quality among Canadian children. J Nutr Educ Behav. 2014; 46:304–308.

101. Kramer RF, Coutinho AJ, Vaeth E, Christiansen K, Suratkar S, Gittelsohn J. Healthier Home Food Preparation Methods and Youth and Caregiver Psychosocial Factors Are Associated with Lower BMI in African American Youth. J Nutr. 2012; 142: 948-954.

102. Woodruff SJ, Kirby AR. The associations among family meal frequency, food preparation frequency, self-efficacy for cooking, and food preparation techniques in children and adolescents. J Nutr Educ Behav. 2013; 45:296–303.

103. Anderson JD, Newby R, Kehm R, Barland P, Hearst MO. Taking Steps Together: A Family- and Community-Based Obesity Intervention for Urban, Multiethnic Children. Health Educ Behav. 2015; 42:194-201.

104. Brown BJ, Hermann. Cooking Classes Increase Fruit and Vegetable Intake and Food Safety Behaviors in Youth and Adults. J Nutr Educ Behav. 2005; 37:104–105.

105. Cunningham-Sabo L, Lohse B. Cooking with Kids positively affects fourth graders’ vegetable preferences and attitudes and self-efficacy for food and cooking. Child Obes. 2013; 9:549–556.

106. Gibbs L, Staiger PK, Johnson B, Block K, Macfarlane S, Gold L, Kulas J, Townsend M, Long C, Ukoumunne O. Expanding children’s food experiences: the impact of a school-based kitchen garden program. J Nutr Educ Behav. 2013; 45:137–146.

107. Cunningham-Sabo L, Lohse B. Impact of a school-based cooking curriculum for fourth-grade students on attitudes and behaviors is influenced by gender and prior cooking experience. J Nutr Educ Behav. 2014; 46:110–120.

108. van der Horst K, Ferrage A, Rytz A. Involving children in meal preparation. Effects on food intake. Appetite. 2014; 79:18–24.

109. Walters LM, Stacey JE. Focus on food: development of the Cooking with Kids experiential nutrition education curriculum. J Nutr Educ Behav. 2009; 41:371–373.

270 110. Bingham S, Gill C, Welch A, Day K, Cassidy A. Comparison of dietary assessment methods in nutritional epidemiology: weighed records v. 24 h recalls. food-frequency questionnaires and estimated diet-records. Br J Nutr. 1994; 72:619–643.

111. Horner NK, Patterson RE, Neuhouser ML, Lampe JW, Beresford SA, Prentice RL. Participant characteristics associated with errors in self-reported energy intake from the Women ’s Health Initiative food-frequency. Am J Clin Nutr. 2002; 76:766–773.

112. Slater J, Mudryj AN. Nuturing Future Generations: Household Food Practices of Canadian Children and Family Meal Participation. Can J Diet Pract Res. 2016; 77:113– 118.

113. Coulston A, Boushey C, Ferruzzi M: Nutrition in the Prevention and Treatment of Disease. Third. Oxford, UK: Elsevier Inc.; 2013.

114. Harnack L, Himes JH, Anliker J, Clay T, Gittelsohn J, Jobe JB, Ring K, Snyder P, Thompson J, Weber JL. Intervention-related bias in reporting of food intake by fifth- grade children participating in an obesity prevention study. Am J Epidemiol. 2004; 160:1117–1121.

115. Miller TM, Abdel-Maksoud MF, Crane LA, Marcus AC, Byers TE. Effects of social approval bias on self-reported fruit and vegetable consumption: a randomized controlled trial. Nutr J. 2008; 7:18.

116. Barlow SE, Dietz WH. Obesity Evaluation and Treatment: Expert Committee Recommendations. Pediatrics. 1998; 102:1–11.

117. Hendrie G, Sohonpal G, Lange K, Golley R. Change in the family food environment is associated with positive dietary change in children. Int J Behav Nutr Phys Act. 2013; 10:4.

118. Wyse R, Campbell E, Nathan N, Wolfenden L. Associations between characteristics of the home food environment and fruit and vegetable intake in preschool children: a cross-sectional study. BMC Public Health. 2011; 11:938.

119. MacFarlane A, Cleland V, Crawford D, Campbell K, Timperio A. Longitudinal examination of the family food environment and weight status among children. Int J Pediatr Obes. 2009; 4:343–352.

120. Ostbye T, Malhotra R, Stroo M, Lovelady C, Brouwer R, Zucker N, Fuemmeler B. The effect of the home environment on physical activity and dietary intake in preschool children. Int J Obes. 2013; 37:1314–1321.

271 121. Wang L, Dalton WT, Schetzina KE, Fulton-Robinson H, Holt N, Ho A, Tudiver F, Wu T. Home food environment, dietary intake, and weight among overweight and obese children in Southern Appalachia. South Med J. 2013; 106:550–557.

122. Haines J, McDonald J, O’Brien A, Sherry B, Bottino CJ, Schmidt ME, Taveras EM. Healthy Habits, Happy Homes: randomized trial to improve household routines for obesity prevention among preschool-aged children. JAMA Pediatr. 2013;167:1072–1079.

123. Rockett HR. Family dinner: more than just a meal. J Am Diet Assoc. 2007; 107:1498–1501.

124. Woodruff SJ, Hanning RM. Effect of Meal Environment On Diet Quality Rating. Can J Diet Pract Res. 2009; 70:118–124.

125. Lytle L, Hearst MO, Fulkerson J, Murray D. Examining the Relationships Between Family Meal Practices, Family Stressors, and the Weight of Youth in the Family. Ann Behav Med. 2013; 41:353–362.

126. Ohly H, Pealing J, Hayter AKM, Pettinger C, Pikhart H, Watt RG, Rees G. Parental food involvement predicts parent and child intakes of fruits and vegetables. Appetite. 2013; 69:8–14.

127. Tibbs T, Haire-Joshu D, Schechtman K, Brownson R. The relationship between parental modeling, eating patterns, and dietary intake among African-American parents. J Am Diet Assoc. 2001; 101:535–541.

128. Boutelle KN, Birkeland RW, Hannan PJ, Story M, Neumark-Sztainer D. Associations between maternal concern for healthful eating and maternal eating behaviors, home food availability, and adolescent eating behaviors. J Nutr Educ Behav. 2007; 39:248–256.

129. Zuercher JL, Wagstaff DA, Kranz S. Associations of food group and nutrient intake, diet quality, and meal sizes between adults and children in the same household: a cross- sectional analysis of U.S. households. Nutr J. 2011; 10:131.

130. Draxten M, Fulkerson JA, Friend S, Flattum CF, Schow R. Parental role modeling of fruits and vegetables at meals and snacks is associated with children’s adequate consumption. Appetite. 2014; 78:1–7.

131. Santiago-Torres M, Adams AK, Carrel AL, LaRowe TL, Schoeller DA. Home food availability, parental dietary intake, and familial eating habits influence the diet quality of urban Hispanic children. Child Obes. 2014; 10:408–415.

272 132. Epstein L, Valoski A, Wing RR, Mccurley J. Ten-Year Obese Children Follow-up of Behavioral , Family-Based Treatment for. J Am Med Assoc. 1990; 264:2519–2523.

133. Campbell KJ, Crawford DA, Salmon J, Carver A, Garnett SP, Baur LA. Associations between the home food environment and obesity-promoting eating behaviors in adolescence. Obesity. 2007; 15:719–730.

134. Rosenkranz R, Dzewaltowski D. Model of the home food environment pertaining to childhood obesity. Nutr Rev. 2008; 66:123–140.

135. Cullen KW, Baranowski T, Owens E, Marsh T, Rittenberry L, de Moor C. Availability, Accessibility, and Preferences for Fruit, 100% Fruit Juice, and Vegetables Influence Children’s Dietary Behavior. Heal Educ Behav. 2003; 30:615–626.

136. Kratt P, Reynolds K, Shewchuk R. The Role of Availability as a Moderator of Family Fruit and Vegetable Consumption. Health Educ Behav. 2000; 27:471–482.

137. De Bourdeaudhuij I, te Velde S, Brug J, Due P, Wind M, Sandvik C, Maes L, Wolf A, Perez Rodrigo C, Yngve A, Thorsdottir I, Rasmussen M, Elmadfa I, Franchini B, Klepp KI. Personal, social and environmental predictors of daily fruit and vegetable intake in 11-year-old children in nine European countries. Eur J Clin Nutr. 2008; 62:834– 841.

138. Bere E, Klepp KI. Correlates of fruit and vegetable intake among Norwegian schoolchildren: parental and self-reports. Public Health Nutr. 2004; 7:991–998.

139. Couch SC, Glanz K, Zhou C, Sallis JF, Saelens BE. Home food environment in relation to children’s diet quality and weight status. J Acad Nutr Diet. 2014; 114:1569– 1579.

140. Golan M. Parents as agents of change in childhood obesity – from research to practice. Int J Pediatr Obes. 2006; 1:66–76.

141. Danner FW. A National Longitudinal Study of the Association Between Hours of TV Viewing and the Trajectory of BMI Growth Among US Children. J Peiatr Psychol. 2008; 33:1100–1107.

142. Robinson T. Television viewing and childhood obesity. Pediatr Clin North Am. 2001; 48:1017–1025.

143. Robinson TN, Nited HEU, Has ST. Reducing Children’s Television Viewing to Prevent Obesity. JAMA. 1999; 282:1561-1567.

273 144. Boone JE, Gordon-larsen P, Adair LS, Popkin BM. Screen time and physical activity during adolescence : longitudinal effects on obesity in young adulthood. Int J Behav Nutr Phys Act. 2007; 10:1–10.

145. Strasburger VC. Children, adolescents, obesity, and the media. Pediatrics. 2011; 128:201–208.

146. Sonneville KR, Gortmaker SL. Total energy intake, adolescent discretionary behaviors and the energy gap. Int J Obes. 2008; Suppl 6:S19-27.

147. Barr-Anderson DJ, Larson NI, Nelson MC, Neumark-Sztainer D, Story M. Does television viewing predict dietary intake five years later in high school students and young adults? Int J Behav Nutr Phys Act. 2009; 6:7.

148. Wiecha JL, Peterson KE, Ludwig DS, Kim J, Sobol A, Gortmaker SL. When Children Eat What They Watch. Arch Pediatr Adolesc Med. 2006; 160:436–442.

149. Colapinto CK, Fitzgerald A, Taper LJ, Veugelers PJ. Children’s preference for large portions: prevalence, determinants, and consequences. J Am Diet Assoc. 2007; 107:1183– 1190.

150. Coon KA, Goldberg J, Rogers BL, Tucker KL. Relationships Between Use of Television During Meals and Children’s Food Consumption Patterns. Pediatrics. 2014; 107:1–11.

151. Feldman S, Eisenberg ME, Neumark-Sztainer D, Story M. Associations between watching TV during family meals and dietary intake among adolescents. J Nutr Educ Behav. 2007; 39:257–263.

152. Feldman S, Eisenberg ME, Neumark-Sztainer D, Story M: Associations between watching TV during family meals and dietary intake among adolescents. J Nutr Educ Behav. 2007; 39:257–263.

153. Robertson W, Friede T, Blissett J, Rudolf MCJ, Wallis M, Stewart-Brown S. Pilot of “Families for Health”: community-based family intervention for obesity. Arch Dis Child. 2008; 93:921–926.

154. Haire-joshu D, Elliott MB, Caito NM, Nanney MS, Hale N, Boehmer TK, Kreuter M, Brownson RC. High 5 for Kids: The Impact of a Home Visiting Program on Fruit and Vegetable Intake of Parents and Their Preschool Children. Prev Med. 2013; 47:77–82.

155. Parents as Teachers. Parents as Teachers Assoication, 2017. St. Louis, MO. [www.parentsasteachers.org]. Accessed Oct 2016.

274 156. Kumanyika S, Obarzanek E, Robinson T, Beech B. Phase 1 of the Girls health Enrichment Multi-site Studies (GEMS): conclusion. Ethn Dis. 2003; 13(Supplement 1):S88–91.

157. Story M, Sherwood NE, Himes JH, Davis M, Jr DRJ, Cartwright Y, Smyth M, Rochon J. An After-School Obesity Prevention Program For African-American Girls: The Minnesota GEMS Pilot Study. Ethn Dis. 2003; 13(Supplement 1):S54–64.

158. Tabak RG, Tate DF, Stevens J, Maria A, Ward DS. Family ties to health program: a randomized intervention to improve vegetable intake in children. J Nutr Educ Behav. 2014; 44:166–171.

159. Cullen KW, Thompson D, Chen T. Outcome Evaluation of Family Eats : An Eight- Session Web-Based Program Promoting Healthy Home Food Environments and Dietary Behaviors for African American Families. Heal Educ Behav. 2016; 44:1–9.

160. Penney T, Almiron-Roig E, Shearer C, Mclsaac J, Kirk S. Modifying the food environment for childhood obesity prevention: challenges and opportunities. Proc Nutr Soc. 2014; 73:226–236.

161. Rosenkranz RR, Dzewaltowski DA. Promoting better family meals for girls attending summer programs. J Nutr Educ Behav. 2009; 41:65–67.

162. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ. Prentice-Hall, Inc; 1986.

163. Bandura A. Social cognitive theory. In: E. Barnouw (Ed.), International Encyclopedia of Communications (Vol 4, p92-96). New York: Oxford University Press.

164. Bandura A. Self-agency mechanism in human agency. Amer Psychol. 1982; 37:122– 147.

165. Bandura A. Self-Efficacy: Toward a Unifying Theory of Behavioral Change. Psychol Rev. 1977; 84:191–215.

166. Bandura A. Health promotion by social cognitive means. Health Educ Behav. 2004; 31:143–64.

167. Coleman P, Karraker K. Parenting self-efficacy among mothers of school-age children: Conceptualization, measurement, and correlates. Fam Relat. 2000; 49:13–24.

168. Marvicsin D, Danford CA. Parenting efficacy related to childhood obesity: comparison of parent and child perceptions. J Pediatr Nurs. 2013; 28:422–429.

275 169. Cullen KW, Baranowski T, Rittenberry L, Cosart C, Owens E, Hebert D, Moor C DE. Socioenvironmental influences on children’s fruit , juice and vegetable consumption as reported by parents: reliability and validity of measures. Pub Heal Nutr. 2000; 3:345- 356.

170. Campbell K, Hesketh K, Silverii A, Abbott G. Maternal self-efficacy regarding children’s eating and sedentary behaviours in the early years: associations with children’s food intake and sedentary behaviours. Int J Pediatr Obes. 2010; 5:501–508.

171. West F, Sanders MR, Cleghorn GJ, Davies PSW. Randomised clinical trial of a family-based lifestyle intervention for childhood obesity involving parents as the exclusive agents of change. Behav Res Ther. 2010; 48:1170–1179.

172. Jurkowski JM, Lawson HA, Green Mills LL, Wilner PG, Davison KK. The empowerment of low-income parents engaged in a childhood obesity intervention. Fam Community Health. 2014; 37:104–118.

173. Davison KK, Jurkowski JM, Li K, Kranz S, Lawson HA. A childhood obesity intervention developed by families for families: results from a pilot study. Int J Behav Nutr Phys Act. 2013; 10:1–11.

174. Akey T, Marquis J, Ross M. Validation of scores on the psychological empowerment scale: a measure of empowerment for parents of children with a disability. Educ Psychol Meas. 2000; 60:419–438.

175. Sprietzer G. Psychological empowerment in the workplace: dimensions, measurement, and validation. Acad Manag J. 1995; 38:1442–1465.

176. Roberts-Gray C, Briley ME, Ranjit N, Byrd-williams CE, Sweitzer SJ, Sharma SV, Palafox MR, Hoelscher DM. Efficacy of the Lunch is in the Bag intervention to increase parents’ packing of healthy bag lunches for young children : a cluster-randomized trial in early care and education centers. Int J Behav Nutr Phys Act. 2016; 13:1–19.

177. Dietz W. Childhood weight affects adult morbidity and morality. J Nutr. 1998; 128:S411–414.

178. Wyse R, Campbell E, Nathan N, Wolfenden L. Associations between characteristics of the home food environment and fruit and vegetable intake in preschool children: a cross-sectional study. BMC Public Health. 2011; 11:938.

179. Thomas H. Obesity prevention programs for children and youth: why are their results so modest? Health Educ Res. 2006; 21:783–795.

276 180. Waters E, Hall BJ, Brown T, Campbell KJ, Gao Y, Armstrong R, Prosser L, Summerbell CD. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2013; 12.

181. Bandura A. Social Cognitive Theory of Self-Regulation. Organ Behav Hum Decis Process. 1991; 50:248-287.

182. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall, Inc; 1986.

183. Margolis H, McCabe P. Improving self-efficacy and motivation: what to do, what to say. Interv Sch Clin. 2006; 41:218–227.

184. Ogden CL, Carroll MD, Fryar CD, Flegal KM. Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief. 2015; 219:1-8.

185. Bartholomew LK, Parcel GS, Kok G. Intervention mapping: a process for developing theory- and evidence-based health education programs. Heal EducBehav. 1998, 25:545–563.

186. Bartholomew L, Parcel G, Kok G, Gottlieb N. Planning Health Promotion Programs: An Intervention Mapping Approach. Second Ed. San Francisco, CA: Jossey- Bass; 2011.

187. U.S. Department of Agriculture and U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2010. 7th edition. Washington, DC. [https://health.gov/dietaryguidelines/dga2010/dietaryguidelines2010.pdf]. Accessed Oct 2014.

188. Burrows TL, Martin RJ, Collins CE. A systematic review of the validity of dietary assessment methods in children when compared with the method of doubly labeled water. J Am Diet Assoc. 2010; 110:1501–1510.

189. Hajduk C, Driscoll P, Johnson R, Goran M. Validation of the multiple-pass 24-hour dietary recall in young children; in-person vs. telephone-administered interviews. J Am Diet Assoc. 1996; 96:A77.

190. Vella J. Learning to Listen, Learning to Teach: The Power of Dialogue in Educating Adults. San Francisco, CA: Jossey-Bass; 1994.

191. Norris J. From Telling to Teaching: A Dialogue Approach to Adult Learning. North Myrtle Beach, SC: Learning by Dialogue; 1993.

277 192. Islam SN, Paddock JD, Dollahite JS. Navigating for Success: Quality Staff Training Translates Into Improved Program Participant Outcomes. J Nutr Educ Behav. 2014; 47:188–189.

193. Verbon K, Brandt G, Bunsen T, Hobbs H, Johnson F. Promising Practices in Nutrition Education: Choosing the Right Nutrition Education Curriculum for Your Program. Washington State Department of Health. Tumwater, WA; 2009. [http://www.doh.wa.gov/portals/1/Documents/Pubs/345-250- PromisingPracticesNutritionEducation.pdf]. Accessed Oct 2014.

194. Rogers C, Gunther C. Determining the feasibility and acceptability of a family meals pilot study for families with elementary-aged children. In: Russell Klein Research Symposium. Columbus, OH; 2014.

195. Rogers C, Gunther C. A mixed methods study designed to enhance feasibility and efficacy of a field-tested nutrition education/cooking program for families with preschool-aged children. In: Russell Klein Research Symposium. Columbus, OH; 2015.

196. Bliss R. Researchers produce innovation in dietary recall. Agric Res.2004; 52:10–12.

197. Guenther P, Casavale K, Kirkpatrick S, Reedy J. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet. 2014; 113:1–20.

198. Lohman T, Roche A, Martorell R. Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics Books; 1988.

199. National Health and Examination Survey (NHANES): Anthropometric Procedures Manual. Centers for Disease Control and Prevention. Atlanta, GA; 2007. [https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf]. Accessed Nov 2014.

200. CDC National Center for Health Statistics: Growth Charts Z-score Data Files. Centers for Disease Control and Prevention. Atlanta, GA; 2009. [http://www.cdc.gov/growthcharts/zscore.htm] Accessed Jan 2017.

201. Fryar CD, Gu Q, Ogden CL. Anthropometric Reference Data for Children and Adults: United States, 2007-2010. U.S. Depatment of Health and Human Services, Vital and Health Statistics. 2012; Series 11: Number 252. [https://www.cdc.gov/nchs/data/series/sr_11/sr11_252.pdf]. Accessed Jan 2017.

202. Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents. U.S. Department of Health and Human Services, National Heart, Lung, and Blood Institute. Bethesda, MD; 2004.

278 203. Bohman B, Nyberg G, Sundblom E, Schäfer Elinder L. Validity and Reliability of a Parental Self-Efficacy Instrument in the Healthy School Start Prevention Trial of Childhood Obesity. Heal Educ Behav. 2013; 41:392–396.

204. Bickel G, Nord M, Price C, Hamilton W, Cook J. Guide to Measuring Household Food Security, Revised 2000. United States Department of Agriculture. Alexandria, VA; 2000. [https://fns-prod.azureedge.net/sites/default/files/FSGuide.pdf]. Accessed Jan 2017.

205. Fulkerson JA, Story MT, Neumark-Sztainer D, Rydell S. Family meals: perceptions of benefits and challenges among parents of 8- to 10-year-old children. J Am Diet Assoc. 2008; 108:706–709.

206. Neumark-Sztainer D, Hannan PJ, Story M, Croll J, Perry C. Family meal patterns: Associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc. 2003; 103:317–322.

207. Hesketh K, Campbell K. Interventions to prevent obesity in 0-5 year olds: an updated systematic review of the literature. Obesity. 2010; 18:S27–35.

208. Rose AM, Wagner AK, Kennel JA, Miller CK, Holloman C, Pennywitt J, Battista M, Murray RD, Rogers C a., Gunther CW. Determining the Feasibility and Acceptability of a Nutrition Education and Cooking Program for Preschoolers and Their Families Delivered Over the Dinner Hour in a Low-Income Day Care Setting. Infant, Child, Adolesc Nutr. 2014; 6:144–151.

209. Fulkerson JA, Kubik MY, Story M, Lytle L, Arcan C. Are There Nutritional and Other Benefits Associated with Family Meals Among At-Risk Youth? J Adolesc Heal 2009, 45:389–395.

210. Taveras EM, Rifas-Shiman SL, Berkey CS, Rockett HR, Field AE, Frazier AL, Colditz GA, Gillman MW. Family dinner and adolescent overweight. Obes Res. 2005; 13:900–906.

211. Corbie-Smith G, Thomas S, St George D. Distrust, race, and research. Arch Intern Med. 2002; 162:2458–2463.

212. Goode T, Harrison S. Cultural Competence in Primary Health Care: Partnerships for a Research Agenda. National Center for Cultural Competence. Washington, D.C.; 2000.

213. Yancey AK, Ortega AN, Kumanyika SK. Effective Recruitment and Retention of Minority Research Participants. Annu Rev Public Health. 2006; 27:1–28.

279 214. Brannon EE, Kuhl ES, Boles RE, Aylward BS, Ratcliff MB, Valenzuela JM, Johnson SL, Powers SW. Strategies For Recruitment and Retentionof Families fom Low- Income, Ethnic Minority Backgrounds in a Longitudnal Study of Carefiver Feeding and Child Weight. Child Heal Care. 2013; 42:198–213.

215. Alwan A, Armstrong T, Bettcher D, Branca F. Global Status Report on Noncommunicable Diseases. World Health Organization. Geneva, Switzerland; 2011.

216. Berge JM, Wall M, Hsueh T-F, Fulkerson JA, Larson N, Neumark-Sztainer D. The Protective Role of Family Meals for Youth Obesity: 10-Year Longitudinal Associations. J Pediatr. 2015; 201:296-301.

217. Rogers C, Anderson SE, Dollahite JS, Hill TF, Holloman C, Miller CK, Pratt KJ, Gunther C. Methods and design of a 10-week multi- component family meals intervention: a two group quasi-experimental effectiveness trial. BMC Public Health. 2017; 17:50.

218. Blumberg S, Bialostosky K, Hamilton W, Briefel R. The Effectiveness of a Short Form of the Household Food Security Scale. Am J Public Health. 1999; 89:1231–1234.

219. Datar A, Shier V, Sturm R. Changes in body mass during elementary and middle school in a national cohort of kindergarteners. Pediatrics. 2011; 128:e1411-1417.

220. Fulkerson JA, Larson N, Horning M, Neumark-Sztainer D. A Review of Associations Between Family or Shared Meal Frequency and Dietary and Weight Status Outcomes Across the Lifespan. J Nutr Educ Behav. 2014; 46:2–19.

221. Waters E, de Silva-Sanigorski A, Burford B, Brown T, Campbell K, Gao Y, Armstrong R, Prosser L, Summerbell CD. Interventions for Preventing Obesity in Children. Cochrane Database Syst Rev 2011, 12:CD001871.

222. Ory MG, Smith ML, Mier N, Wernicke MM. The Science of Sustaining Health Behavior Change: The Health Maintenance Consortium. Am J Health Behav. 2010; 34:647–659.

223. Pyatak EA, Otr L, Blanche EI, Otr L, Garber SL, Diaz J, Otr L, Blanchard J, Otr L, Florindez L, Clark FA, Otr L. Conducting Intervention Rsearch Among Underserved Populations: Lessons Learned and Recommendations for Researchers. Arch Phys Med Rehabil. 2013; 94:1190–1198.

224. Faith MS, Horn L Van, Appel L, Burke LE, Carson JAS, Franch HA, Jakicic J, Kral T, Odoms-Young A, Wansink B. Evaluating Parents and Adult Caregivers as “Agents of Change ” for Treating Obese Children: Evidence for Parent Behavior Change Strategies and Research Gaps. American Heart Association. Circulation 2012; 125:1186–1207.

280 225. Koplan JP, Liverman CT, Kraak VA. Preventing Childhood Obesity. Washington, D. C.: National Academies Press; 2004.

226. Wansink B, Kleef E Van. Dinner Rituals That Correlate with Child and Adult BMI. Obesity. 2014; 22:91–95.

227. Sobal J, Hanson K. Family meals and body weight in US adults. Public Health Nutr. 2011; 14:1555–1562.

228. Berge JM, Maclehose RF, Loth KA. Family Meals: Associations with Weight and Eating Behaviors Among Mothers and Fathers. Appetite. 2012; 58:1128–1135.

229. Conway JM, Ingwersen LA, Moshfegh AJ. Accuracy of dietary recall using the USDA five-step multiple-pass method in men: an observational validation study. J Am Diet Assoc. 2004; 104:595–603.

230. Conway JM, Ingwersen LA, Vinyard BT, Moshfegh AJ. Effectiveness of the US Department of Agriculture 5-step multiple-pass method in assessing food intake in obese and nonobese women. Am J Clin Nutr. 2003; 77:1171-1178.

231. Cunningham-Sabo L, Lohse B, Smith S, Browning R, Strutz E, Nigg C, Balgopal M, Kelly K, Ruder E. Fuel for Fun: a cluster-randomized controlled study of cooking skills, eating behaviors, and physical activity of 4th graders and their families. BMC Public Health. 2016; 16:444.

232. Martin-Biggers J, Hongu N, Worobey J, Byrd-Bredbenner C. Family meal behaviors and cognitions among parents of preschoolers. FASEB J. 2013, 27:367.

233. Tran K, Johnston R, Soultanakis R, Matthews D. In-person vs telephone- administered multiple-pass 24-hour recalls in women: validation with doubly labeled water. J Am Diet Assoc. 2000; 100:777–783.

234. Schnirer L, Stack-Cutler H. Recruitment and Engagement of Low-Income Populations: Service Provider and Researcher Perspectives. Community-University Partnerships for the Study of Children, Youth and Families. Alberta, Canada; 2011. [http://www.cup.ualberta.ca/wp-content/uploads/2011/07/Recruitment-and-Engagement- of-Low-Income-Populations1.pdf]. Accessed Jan 2017.

235. Singh G, Siapush M, KoganMD. Rising social inequalities in US childhood obesity, 2003-2007. Ann Epidemiol. 2010; 20:40–52.

236. Woodruff S, Hanning R. A review of family meal influence on adolescents’ dietary intake. Can J Diet Pr Res. 2008; 69:14–22.

281 237. Neumark-Sztainer D, Larson N, Fulkerson J, Eisenberg M, Story M. Family meals and adolescents: What have we learned from Project EAT (Eating Among Teens)? Public Health Nutr. 2010; 13:1113–1121.

238. Dietary Guidelines for Americans 2015-2020. 8th Edition.United States Department of Agriculture. Washington, D.C.; 2015. [https://health.gov/dietaryguidelines/2015/resources/2015-2020_Dietary_Guidelines.pdf]. Accessed Dec 2016.

239. Data Resource Center for Child and Adolescent Health. Child and Adolescent Health Measurement Initiative. [http://www.childhealthdata.org]. Accessed Jan 2017.

240. Fulkerson J, Neumark-Sztainer D, Story M. Adolescent and Parent Views of Family Meals. J Am Diet Assoc. 2006; 106:526–532.

241. Fulkerson J, Kubik MY, Rydell S, Boutelle KN. Focus groups with working parents of school-aged children: what’s needed to improve family meals? J Nutr Educ Behav. 2011; 43:189–193.

242. Ogden CL, Lamb MM, Carroll MD, Flegal KM. Obesity and Socioeconomic Status in Children and Adolescents: United States, 2005-2008. U.S. Department of Health and Human Services. Washington, D.C. 2010; 2010. [https://www.cdc.gov/nchs/data/databriefs/db51.pdf]. Accessed Jan 2017.

282

Appendix A: Simple Suppers Intervention Participant Recruitment Materials

283 Appendix A: Family Recruitment Flyer

Simple Suppers Building Healthy Families Through Healthy Meals

What is Simple Suppers? What’s in it for me? • A free 10-week nutrition education • Retail gift cards up to $75 program teaching families to cook and • Weekly prize raffles enjoy family meals together • Dinner is included

• Simple Suppers meets 1 evening/ How do I Join? week for 10 weeks, 90 minute session • Be a family with at least 1 child between 4-10 years of age & willing to Join to learn about affordable, participate in a research study simple, healthy meals for free • Willing to attend a 10-week cooking & • Healthy Meal Planning nutrition education program • Quick and Easy Meal Prep CONTACT • Healthy, Kid-Friendly Recipes • Maria Broeckel at Vineyard (614.259.5296) • Preparing & Eating a Family Meal • Katy at OSU (email [email protected])

Coming to Vineyard Columbus in March, 2015!

284 Appendix A: Recruitment Ad in Site Newsletter

New Nutrition Education/Cooking Program for Families with Young Children Coming to Vineyard March, 2015!

Do you want to learn to make family meals easier, cheaper and funner for the entire family? Do you want to encourage your child to eat healthier at meals? Do you want to learn easy, healthy, kid-friendly recipes?

Simple Suppers is a new nutrition education and cooking program for families with young children coming to Vineyard in March, 2015. The Simple Suppers program is a family meals research program from Ohio State that includes 10 lessons delivered weekly. Each Simple Suppers lesson includes: strategies for planning, preparing and serving fast, healthy, kid-friendly family meals; parent meal planning; nutritious, kid- friendly recipes; child food prep activities; and enjoyment of a meal with your family.

We would love to have you and your family participate in Simple Suppers at Vineyard! To participate, you must have at least 1 child between 4-10 years of age in your family. For more information or to sign-up to participate in Simple Suppers, please contact Katy at: [email protected] or 614-292-5125.

285

Appendix B: Simple Suppers Intervention Informed Consent & Assent Forms

286 Appendix B: Caregiver Informed Consent Form (pg. 1)

CONSENT IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2015 Version: 1

1 2 The Ohio State University Consent to Participate in Research 3 4 Simple Suppers: a novel approach to childhood obesity Study Title: prevention Researcher: Dr. Carolyn Gunther

Sponsor: Cardinal(Health(Foundation;(OSU(Seed(grant 5 6 7 This is a consent form for research participation. It contains important information about 8 this study and what to expect if you decide to participate. 9 Your participation is voluntary. 10 Please consider the information carefully. Feel free to ask questions before making your 11 decision whether or not to participate. If you decide to participate, you will be asked to sign 12 this form and will receive a copy of the form. 13 Purpose: 14 The purpose of the Simple Suppers (SS) program is to equip you with healthy cooking skills 15 and nutrition knowledge to use in preparing family meals and have a better overall diet. 16 17 Procedures/Tasks: 18 The program will occur weekly for 10 weeks. Each SS session will include separate parent 19 and child nutrition education lessons, a cooking lesson, and a group meal. You will be asked 20 to complete multiple questionnaires and participate in 3 diet-recall interviews at 2 time points 21 throughout the 3 month study. We will collect height/weight data on your participating 22 child(ren) at 2 time points during the study. You will have the opportunity to have your height 23 and weight measured at 2 time points during the study as well. All identifying information 24 will be removed from the surveys and the results will be published in aggregate only. 25 26 Duration: 27 You may leave the study at any time. If you decide to stop participating in the study, there 28 will be no penalty to you, and you will not lose any benefits to which you are otherwise 29 entitled. Your decision will not affect your future relationship with The Ohio State 30 University. The program will occur weekly for 10 weeks and each session will last 90 31 minutes. 32 33 Risks and Benefits: 34 The risks of this study are minimal. The benefits are that you may gain healthy cooking skills 35 and nutrition knowledge. 36 37

Page 1 of 3 Form date: 02/11/13

287 Appendix B: Caregiver Informed Consent Form (pg. 2)

CONSENT IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2015 Version: 1

38 39 40 Confidentiality: 41 Efforts will be made to keep your study-related information confidential. However, there may 42 be circumstances where this information must be released. For example, personal information 43 regarding your participation in this study may be disclosed if required by state law. Also, 44 your records may be reviewed by the following groups (as applicable to the research): 45 • Office for Human Research Protections or other federal, state, or international 46 regulatory agencies; 47 • The Ohio State University Institutional Review Board or Office of Responsible 48 Research Practices; 49 • The sponsor, if any, or agency (including the Food and Drug Administration for FDA- 50 regulated research) supporting the study. 51 52 Incentives: 53 You will be asked to complete a packet of surveys and participate in 3 diet-recall interviews at 54 2 time points: 1 week before the start of the program and 1 week after the end of the program. 55 After completing the surveys and interviews 1 week before the program, you will receive a 56 $25 retail gift card. After completing the surveys and interviews 1 week after the program, 57 you will receive a $50 retail gift card. 58 59 Participant Rights: 60 61 You may refuse to participate in this study without penalty or loss of benefits to which you 62 are otherwise entitled. If you are a student or employee at Ohio State, your decision will not 63 affect your grades or employment status. 64 65 If you choose to participate in the study, you may discontinue participation at any time 66 without penalty or loss of benefits. By signing this form, you do not give up any personal 67 legal rights you may have as a participant in this study. 68 69 An Institutional Review Board responsible for human subjects research at The Ohio State 70 University reviewed this research project and found it to be acceptable, according to 71 applicable state and federal regulations and University policies designed to protect the rights 72 and welfare of participants in research. 73 74 Contacts and Questions: 75 For questions, concerns, or complaints about the study, or you feel you have been harmed as a 76 result of study participation, you may contact Dr. Carolyn Gunther (614) 292-5125 77 [email protected]. 78 79 For questions about your rights as a participant in this study or to discuss other study-related 80 concerns or complaints with someone who is not part of the research team, you may contact 81 Ms. Sandra Meadows in the Office of Responsible Research Practices at 1-800-678-6251. 82 Page 2 of 3 Form date: 02/11/13

288 Appendix B: Caregiver Informed Consent Form (pg. 3)

CONSENT IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2015 Version: 1

83 Signing the consent form 84 85 I have read (or someone has read to me) this form and I am aware that I am being asked to 86 participate in a research study. I have had the opportunity to ask questions and have had them 87 answered to my satisfaction. I voluntarily agree to participate in this study. 88 89 I am not giving up any legal rights by signing this form. I will be given a copy of this form. 90

Printed name of subject Signature of subject

AM/PM Date and time

Printed name of person authorized to consent for subject Signature of person authorized to consent for subject (when applicable) (when applicable)

AM/PM Relationship to the subject Date and time 91 92 93 94 Investigator/Research Staff 95 96 I have explained the research to the participant or his/her representative before requesting the 97 signature(s) above. There are no blanks in this document. A copy of this form has been given 98 to the participant or his/her representative. 99

Printed name of person obtaining consent Signature of person obtaining consent

AM/PM Date and time 100

Page 3 of 3 Form date: 02/11/13

289 Appendix B: Caregiver Permission Form (pg. 1)

PARENTAL PERMISSION IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2014 Version: 1

1 2 The Ohio State University Parental Permission 3 For Child’s Participation in Research 4 5 Simple'Suppers'Scale-Up'(S3):'An'expanded'pilot'study'to'determine' Study Title: the'effectiveness'of'participation'in'an'innovative'nutrition' education'and'cooking'program'for'families'with'young'children Researcher: Dr. Carolyn Gunther

Sponsor: Cardinal(Health(Foundation;(OSU(Seed(grant 6 7 This is a parental permission form for research participation. It contains important 8 information about this study and what to expect if you permit your child to participate. 9 Your child’s participation is voluntary. 10 Please consider the information carefully. Feel free to discuss the study with your friends and 11 family and to ask questions before making your decision whether or not to permit your child 12 to participate. If you permit your child to participate, you will be asked to sign this form and 13 will receive a copy of the form. 14 Purpose: 15 The purpose of the Simple Suppers (SS) program is to equip parents and their children with 16 healthy cooking skills and nutrition knowledge to use in preparing family meals and have a 17 better overall diet. 18 19 Procedures/Tasks: 20 The program will occur weekly for 10 weeks. Each SS session will include separate parent 21 and child nutrition education lessons, a cooking lesson, and a group meal. During the child 22 nutrition education lessons, your child will learn several age-appropriate food preparation 23 skills. We will collect height/weight data on your participating child at 2 time points during 24 the study. Your child will be asked to complete a food preference survey and participate in a 25 dietary recall interview, with your help, at 2 time points during the study. All identifying 26 information will be removed from data and the results will be published in aggregate only. 27 28 Duration: 29 Your child may leave the study at any time. If you or your child decides to stop participation 30 in the study, there will be no penalty and neither you nor your child will lose any benefits to 31 which you are otherwise entitled. Your decision will not affect your future relationship with 32 The Ohio State University. Simple Suppers will occur weekly for 10 weeks and each session 33 will last 90 minutes. 34 35 Risks and Benefits:

Page 1 of 4 Form date: 06/03/13

290 Appendix B: Caregiver Permission Form (pg. 2)

PARENTAL PERMISSION IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2014 Version: 1

36 The risks of this study are minimal. The benefits are that your child may gain healthy cooking 37 skills and nutrition knowledge. 38 Confidentiality: 39 Efforts will be made to keep your child’s study-related information confidential. However, 40 there may be circumstances where this information must be released. For example, personal 41 information regarding your child’s participation in this study may be disclosed if required by 42 state law. Also, your child’s records may be reviewed by the following groups (as applicable 43 to the research): 44 • Office for Human Research Protections or other federal, state, or international 45 regulatory agencies; 46 • The Ohio State University Institutional Review Board or Office of Responsible 47 Research Practices; 48 • The sponsor, if any, or agency (including the Food and Drug Administration for FDA- 49 regulated research) supporting the study. 50 51 Incentives: 52 Children will not be directly incentivized. Families will be asked to complete a packet of 53 surveys and participate in dietary recall interviews at the beginning and end of the program. 54 For completing the surveys and participating in the interviews at the beginning of the 55 program, your family will receive a $25 retail gift card. For completing the surveys and 56 participating in the interviews at the end of the program, your family will receive a $50 retail 57 gift card. 58 59 Participant Rights: 60 61 You or your child may refuse to participate in this study without penalty or loss of benefits to 62 which you are otherwise entitled. If you or your child is a student or employee at Ohio State, 63 your decision will not affect your grades or employment status. 64 65 If you and your child choose to participate in the study, you may discontinue participation at 66 any time without penalty or loss of benefits. By signing this form, you do not give up any 67 personal legal rights your child may have as a participant in this study. 68 69 An Institutional Review Board responsible for human subjects research at The Ohio State 70 University reviewed this research project and found it to be acceptable, according to 71 applicable state and federal regulations and University policies designed to protect the rights 72 and welfare of participants in research. 73 74 Contacts and Questions: 75 For questions, concerns, or complaints about the study, or you feel your child has been 76 harmed as a result of study participation, you may contact Dr. Carolyn Gunther (614) 292- 77 5125 [email protected]. 78 79 For questions about your child’s rights as a participant in this study or to discuss other study- 80 related concerns or complaints with someone who is not part of the research team, you may Page 2 of 4 Form date: 06/03/13

291 Appendix B: Caregiver Permission Form (pg. 3)

PARENTAL PERMISSION IRB Protocol Number: 2014B0494 Behavioral/Social Science IRB Approval date: 1/27/2014 Version: 1

84 Signing the parental permission form 85 86 I have read (or someone has read to me) this form and I am aware that I am being asked to 87 provide permission for my child to participate in a research study. I have had the opportunity 88 to ask questions and have had them answered to my satisfaction. I voluntarily agree to permit 89 my child to participate in this study. 90 91 I am not giving up any legal rights by signing this form. I will be given a copy of this form. 92

Printed name of subject

Printed name of person authorized to provide permission for Signature of person authorized to provide permission for subject subject

AM/PM Relationship to the subject Date and time 93 94 95 Investigator/Research Staff 96 97 I have explained the research to the participant or his/her representative before requesting the 98 signature(s) above. There are no blanks in this document. A copy of this form has been given 99 to the participant or his/her representative. 100

Printed name of person obtaining consent Signature of person obtaining consent

AM/PM Date and time 101

Page 4 of 4 Form date: 06/03/13

292 Appendix B: Child Assent Script

You are being asked to be in a research study called Simple Suppers. It is okay to say “No” if you don’t want to be in the study. If you say “Yes” you can change your mind and quit being in the study at any time without getting in trouble. If you decide you want to be in the study, an adult (usually a parent) will also need to give permission for you to be in the study. In Simple Suppers you will learn how to help your family make dinner and learn about healthy eating. Over 10 weeks you’ll come with your family to Simple Suppers for 90 minutes one night a week. Each session will include sharing a healthy dinner with your family that you helped to make. We will measure your height and weight, and, with the help of a parent, you will tell us about what you eat and drink. Everything you tell us will be kept private. You can stop being in the study at any time.

Would you like to be in Simple Suppers?

293

Appendix C: Simple Suppers Intervention Data Collection Materials

294 Appendix C: Data Collection Checklist

295 Appendix C: Demographic Questionnaire (pg. 1)

Parent Name:______Parent ID:______About You and Your Family Child Name:______Child ID:______Date:______Time Point:______

Respondent: (please check) Questions 1-7 are about YOUR CHILD. [] Mother Fill in the appropriate circle. [] Father [] Other______1. How old is your preschooler? ! 2 ! 3 ! 4 ! 5 years

2. Is this child a boy or girl? ! Boy ! Girl

3. Please check (!) only one box about your child.

ETHNICITY OF THIS CHILD Please check (!) only one box. Hispanic or Latino. A person of Cuban, Mexican, Puerto Rican, South or Central American, or " other Spanish culture or origin, regardless of race.

" Not Hispanic or Latino.

4. How would you best describe this child with respect to race?

RACE OF THIS CHILD You may check (!) more than one box. " Black or African American. A person having origins in any of the Black racial groups of Africa. White. A person having origins in any of the original peoples of Europe, the Middle East, or " North Africa. Alaska native or American Indian. A person having origins in any of the original peoples of " North, Central and South America, and who maintains tribal affiliation or community attachment. Asian. A person having origins in any of the original peoples of the Far East, Southeast Asia, or " the Indian subcontinent including for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand and Vietnam. Native Hawaiian or other Pacific Islander. A person having origins in any of the original " peoples of Hawaii, Guam, Samoa, or other Pacific Islands.

" Other. A group not mentioned above. If Other is checked, please describe:

5. What is your relationship to this child? ! Parent (includes step parent/foster) ! Grandparent ! Aunt or uncle ! Sibling ! Other, please specify ______6. On average, how many days of the week does this child live in your home? ! 1-3 days ! 4 or more days

296 Appendix C: Demographic Questionnaire (pg. 2)

Parent Name:______Parent ID:______About You and Your Family Child Name:______Child ID:______Date:______Time Point:______

Respondent: (please check) Questions 1-7 are about YOUR CHILD. [] Mother Fill in the appropriate circle. [] Father [] Other______1. How old is your preschooler? ! 2 ! 3 ! 4 ! 5 years

2. Is this child a boy or girl? ! Boy ! Girl

3. Please check (!) only one box about your child.

" Not Hispanic or Latino.

4. How would you best describe this child with respect to race?

" Other. A group not mentioned above. If Other is checked, please describe:

297 Appendix C: Demographic Questionnaire (pg. 3)

298 Appendix C: Demographic Questionnaire (pg. 4)

299 Appendix C: Demographic Questionnaire (pg. 5)

300 Appendix C: 24-hour Dietary Recalls for Children (pg. 1)

301 Appendix C: 24-hour Dietary Recalls for Children (pg. 2)

302 Appendix C: 24-hour Dietary Recalls for Caregivers (pg. 1)

303 Appendix C.: 24-hour Dietary Recalls for Caregivers (pg. 2)

304 Appendix C: Home Food Environment Questionnaire (pg. 1)

Home%Food%Environment%Questionnaire% % % Part%A.% For%each%of%the%items%below,%please%place%an%“X”%in%the%box%that%best%represents% your%thoughts%or%actions.% % Over%the%past%7%days,%home%many%times…..% 1.%%%…did%all%or%most,%of%your%family%eat%dinner%together?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ ! ! ! ! ! ! 2.%%%…did%all%or%most,%of%your%family%eat%breakfast%together?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 3.%%%…was%at%least%one%parent%present%when%you%child(ren)%ate%dinner?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 4.%%…was%dinner%prepared%at%home%and%eaten%together%as%a%family?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 5.%%…was%your%child(ren)%involved%in%meal%preparation?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 6.%%…as%a%separate%meal%made%for%your%child(ren)%because%he/she%did%not%like% the%foods%prepared%for%the%rest%of%the%family?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 7.%%…was%dinner%for%the%family%purchased%from%a%fastQfood%restaurant,%and% eaten%either%at%the%restaurant%or%at%home?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % % % %

305 Appendix C: Home Food Environment Questionnaire (pg. 2)

8.%…has%your%child(ren)%watched%TV%while%eating%meals?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ % 9.%…has%your%child(ren)%requested%to%watch%TV%while%eating%meals?% ! !!!!!!!!!!Never! !!!!!!!!!!!!!1'2! !!!!!!!!!!!!!3'4! !!!!!!!!!!!!!5'6! ! !7! ! ☐ ☐ ☐ ☐ ☐ ! % Part%B.%% For%the%following%questions,%please%place%an%“X”%in%the%box%that%best%represents% your%family%mealsQrelated%actions.%% % During%the%past%30%days,%how%often%did%you…..% % 10.%Eat%healthy%meals%or%snacks%while%your%child%was%around%(“healthy”% defined%as%fruits,%vegetables,%lowQfat%foods,%lean%meats,%wholeQgrains,%etc.)% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 11.%Or%your%spouse/partner%(if%applicable)%eat%meals%in%the%living%room%or%TV% room?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 12.%Take%a%second%helping%during%meals?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 13.%Eat%unhealthy%snacks%around%your%child(ren)?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 14.%Drink%sugared%drinks%or%nonQdiet%soda%around%your%children?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 15.%Eat%meals%or%snacks%while%standing?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % %

306 Appendix C: Home Food Environment Questionnaire (pg. 3)

16.%Eat%meals%or%snacks%straight%from%the%pot/pan/bowl?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 17.%Eat%meals%or%snacks%while%watching%television,%reading,%or%working?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 18.%Eat%meals%or%snacks%when%you%were%bored?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 19.%Eat%meals%or%snacks%when%you%were%angry%or%in%a%bad%or%sad%mood?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 20.%Eat%meals%or%snacks%in%a%disorderly%way%between%meals?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 21.%Eat%meals%or%snacks%late%in%the%evening%or%at%night?% Never! Sometimes! Often! Always! Do!not!know! ☐! ☐! ☐! ☐! ☐! % % Part%C.%% For%the%following%questions,%please%place%an%“X”%in%the%box%that%best%represents% family%mealtime%in%your%household.%% % % 21.%My%child%has%to%come%and%sit%at%the%table%during%meals.% Strongly!Disagree! Disagree! Agree! Strongly!Agree! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 22.%My%child%knows%what%the%rules%for%mealtime%behavior%are.! Strongly!Disagree! Disagree! Agree! Strongly!Agree! Do!not!know! ☐! ☐! ☐! ☐! ☐! % 23.%We%have%clear%rules%about%behavior%at%mealtime.%%% Strongly!Disagree! Disagree! Agree! Strongly!Agree! Do!not!know! ☐! ☐! ☐! ☐! ☐! %

307 Appendix C: Caregiver Self-Efficacy for Healthy Dietary Behaviors Questionnaire (pg. 1)

308 Appendix C: Caregiver Self-Efficacy for Healthy Dietary Behaviors Questionnaire (pg. 2)

309 Appendix C: Child (4-5 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 1)

Child Food Preparation Skills (Ages 4-5 years) ID: ______Time Point: ______Date: ______Respondent: (please check one) [ ] Mother [ ] Father [ ] Other ______

Please answer the following questions based on your 4-5 year old child by putting a ✓ in the appropriate box.

1a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to%set%the%table$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child set the table… 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 2a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to$wipe%table%after%mealtime%% Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child wiped the table after mealtime…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 3a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to%grease%or%spray%baking%pans$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child greased or sprayed a baking pan…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 4a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to%peel%fruit%(e.g.,%orange,%banana)$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child peeled fruit%(e.g.,%orange,%banana)…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 5a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to%measure%dry%ingredients$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child measured dry ingredients…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ $ $ $

310 Appendix C: Child (4-5 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 2)

6a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to$measure%liquid%ingredients$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child measured liquid ingredients…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 7a.$When$we$prepare$food$at$home,$my$445$year$old$child$is$able%to$cut%soft%foods%with%a%blunt%knife$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child cut%soft%foods%with%a%blunt%knife…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

8a.$When$we$prepare$food$at$home,$my$445$year$old$child$knows%when%to%wash%their%hands,%utensils% and%cooking%surfaces%during%food%prep/cleanup% Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 4-5 year old child independently washed their hands, a utensil or a cooking surface for food safety during food prep/cleanup…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

9. Over the past 30 days, my 4-5 year old child asked to help with food prep activities…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

311 Appendix C: Child (6-8 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 1)

Child Food Preparation Skills (Ages 6-8 years) ID: ______Time Point: ______Date: ______Respondent: (please check one) [ ] Mother [ ] Father [ ] Other ______

Please answer the following questions based on your 4-5 year old child by putting a ✓ in the appropriate box.

1a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$set%the%table%independently$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child set the table independently… 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 2a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$pour%beverages%independently%% Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-10 year old child poured beverages independently…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 3a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$measure%dry%ingredients$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child measured dry ingredients…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 4a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$measure%liquid%ingredients$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child measured liquid ingredients…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 5a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$beat%eggs%with%a%whisk%or%egg% beater$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child beat eggs with a whisk or egg beater…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ $ $

312 Appendix C: Child (6-8 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 2)

$ $ $ 6a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$grate%cheese$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child grated cheese…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 7a.$When$we$prepare$food$at$home,$my$648$year$old$child$is$able%to$cut%soft%foods%with%a%blunt%knife$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child cut%soft%foods%with%a%blunt%knife$…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

8a.$When$we$prepare$food$at$home,$my$648$year$old$child$knows%when%to%wash%their%hands,%utensils% and%cooking%surfaces$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 6-8 year old child independently washed their hands, a utensil or a cooking surface for food safety during food prep/cleanup…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

9. Over the past 30 days, my 6-8 year old child asked to help with food prep activities…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

313 Appendix C: Child (9-10 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 1)

Child Food Preparation Skills (Ages 9-10 years) ID: ______Time Point: ______Date: ______Respondent: (please check one) [ ] Mother [ ] Father [ ] Other ______

Please answer the following questions based on your 4-5 year old child by putting a ✓ in the appropriate box.

1a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to%follow%a%recipe$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child followed%a%recipe$… 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 2a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to$crack%eggs%% Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child cracked%eggs…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 3a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to%core%and%slice%an%apple$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child%cored%and%sliced%an%apple…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 4a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to%grate%cheese$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child grated cheese…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 5a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to$peel%potatoes%or%apples$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child peeled potatoes or apples…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ $ $ $

314 Appendix C: Child (9-10 yrs old) Food Preparation Skills & Frequency Questionnaire (pg. 2)

$ $ 6a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to%cut%chicken%with%a%knife%or% scissors$$$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child cut%chicken%with%a%knife%or%scissors…$ 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐! $ 7a.$When$we$prepare$food$at$home,$my$9410$year$old$child$is$able%to%use%a%thermometer%to% determine%if%meat%is%thoroughly%cooked%%% Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child used%a%thermometer%to%determine%if%meat%was% thoroughly%cooked…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

8a.$When$we$prepare$food$at$home,$my$9410$year$old$child$knows%to%wash%their%hands,%utensils%and% cooking%surfaces%after%handling%raw%meat%or%eggs$ Strongly)Disagree) Disagree) Agree) Strongly)Agree) Do)not)know) ☐! ☐! ☐! ☐! ☐! b. Over the past 30 days, my 9-10 year old child washed their hands, utensils and cooking surfaces after handling raw meat or eggs…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

9. Over the past 30 days, my 9-10 year old child asked to help with food prep activities…% 0)times) 152)times) 354)times) 556)times) 7)or)more)times) ☐! ☐! ☐! ☐! ☐!

315 Appendix C: Menu Planning Questionnaire

Menu Planning Name:______ID:______Date:______Time Point:______Respondent: (please check) [] Mother [] Father [] Other______

Please check the box by each statement that most closely describes whether these things happen in your family.

Never Sometimes Often Always 1. I check food labels for ingredients before purchasing a product for the first time. 2. I read the nutrition facts panel before purchasing a product for the first

time. 3. I plan menus before doing my food shopping. 4. I make out a grocery list before doing the shopping. 5. I compare prices between items when I shop. 6. I check the food ads in the newspaper before going food shopping. 7. My children go grocery shopping with me.

8. My children ask me to buy certain vegetables at the grocery store.

9. My children ask me to buy certain fruits at the grocery store.

316 Appendix C: USDA 6-Item Short Form Food Security Questionnaire

Food Security Questionnaire Name:______ID:______Date:______

Time Point:______Respondent: (please check) [] Mother [] Father [] Other______

How often were the following true for your family in the past 30 days:

1. The food we bought just didn’t last, Sometimes Never true Often true Don’t know and we didn’t have money to get true ! ! ! more. ! 2. We couldn’t afford to eat balanced Sometimes Never true Often true Don’t know meals. true ! ! ! !

In the past 30 days:

3. Did you or other adults in your household ever cut the size of your Yes No Don’t know meals or skip meals because there ! ! ! wasn’t enough money for food?

If YES, in the past 30 days, how many days did this happen? ______days

In the past 30 days:

4. Did you ever eat less than you felt Yes No Don’t know you should because there wasn’t ! ! ! enough money for food?

5. Were you ever hungry but didn’t eat Yes No Don’t know because there wasn’t enough money ! ! ! for food?

317 Appendix C: Child Anthropometric Data Collection Form (pg. 1)

318 Appendix C: Child Anthropometric Data Collection Form (pg. 2)

319 Appendix C: Caregiver Anthropometric Data Collection Form (pg. 1)

320 Appendix C: Caregiver Anthropometric Data Collection Form (pg. 2)

321 Appendix C: Simple Suppers Program Satisfaction Survey

322 Appendix C: Fidelity Tool (pg. 1)

Leader (ADULT) ______Leader (CHILD) ______Date ______Session Number ______

Simple Suppers Fidelity Checklist

Core Component Yes No Comment 1. Parent education: review goal from previous week (lessons 3-10) 2. Parent education: Educator anchored topic with an open discussion 3. Parent education: Educator added new information on the topic (provide handouts when applicable) 4. Parent education: Educator applied new information with interactive activity 5. Parent education: Parents planned a family meal for the upcoming week (lessons 3-10) 6. Child education: each child age group completed a specific food prep skill 7. Group family meal: Educator leads nutrition discussion by identifying foods groups in upcoming family meal 8. Group family meal: Educator leads food safety discussion related to upcoming family meal 9. Group family meal: Families complete family meal food preparation 10. Group family meal: Educator guides parents in establishing family meal behavior expectations 11. Group family meal: Families receive take-home bags

12. Child participants are engaged & involved in the program

13. Adult participants are engaged & involved in the program

14. Child Educators create a positive, interactive environment

15. Adult Educators creates a positive, interactive environment 16. Child Educators exhibit a caring attitude

17. Adult Educators allow time for questions

18. Adult Educators answer questions adequately

18. Adult Educators answer question adequately

323 Appendix C: Fidelity Tool (pg. 2)

Unusual events during the session:

Overall comments:

324