In the Wake of Childhood Maltreat- a Decade, Recording Their Lives in Detail

ENT OF M JU U.S. Department of Justice T S R T A I P C E E D

O J Office of Justice Programs C S F A V M F O I N A C I J S R E BJ G O OJJ DP O F PR Office of Juvenile Justice and Delinquency Prevention JUSTICE

Shay Bilchik, Administrator August 1997

In the Wake From the Administrator of Childhood The victimization of the weak by the strong—in this case, of children by adults—is one of the most shameful Maltreatment constants in human history. Unfortu- nately, contemporary American society is not immune from this Barbara Tatem Kelley, Terence P. Thornberry, Ph.D., repugnant behavior. The National and Carolyn A. Smith, Ph.D. Committee to Prevent Child Abuse estimates that 1 million children This Bulletin is part of the Office of achievement, and mental health prob- suffered maltreatment in the United Juvenile Justice and Delinquency Prevention lems. Also, a history of childhood mal- States in 1995 alone. Because (OJJDP) Youth Development Series, which treatment appeared to be a significant countless cases of suspected child presents findings from the Program of Re- risk factor for the development of mul- abuse and neglect remain unre- search on the Causes and Correlates of tiple problems in adolescence. ported, the actual number of abused and neglected children is unknown. Delinquency. Teams at the University at Throughout history, the lives of some What is known, however, is that Albany, State University of New York; the children have been diminished by their University of Colorado; and the University maltreatment in childhood increases exposure to various forms of maltreat- the risk of problems in adolescence, of Pittsburgh collaborated extensively in ment, such as physical abuse, neglect, designing the studies. At study sites in including juvenile delinquency, drug and sexual exploitation. As a new century use, poor performance in school, Rochester, New York; Denver, Colorado; approaches, Americans need not look be- and Pittsburgh, Pennsylvania, the three teen pregnancy, and emotional and yond their own backyards to see the faces mental health disorders. research teams have interviewed 4,000 of children subjected to maltreatment, participants at regular intervals for nearly most often at the hands of parents and In the Wake of Childhood Maltreat- a decade, recording their lives in detail. other caretakers. ment explores the connections Findings to date indicate that preventing between childhood maltreatment and delinquency requires accurate identification subsequent problem behaviors. The of the risk factors that increase the likeli- National Estimates of findings of the Rochester Youth hood of delinquent behavior and the pro- Child Maltreatment Development Study reported in this tective factors that enhance positive Bulletin are particularly valuable adolescent development. How extensive is the problem of child maltreatment in the United States? This because they come from a general This Bulletin presents findings on the cannot be determined with certainty, population sample, which allowed the relationship between childhood maltreat- because a substantial but unknown researchers to examine how mal- ment and subsequent adolescent problem proportion of child maltreatment cases treated youth differ from the general behaviors in the Rochester Youth Devel- are never brought to the attention of adolescent population. The study opment Study. The researchers found that State and local Child Protective Services. results provide helpful insights into the subjects who had experienced mal- Therefore, the following national estimates the cost survivors of childhood treatment during childhood were signifi- of officially reported and substantiated maltreatment often pay—a cost that cantly more likely to display a variety of cases of child maltreatment undoubtedly is always too high. problem behaviors during adolescence, undercount the actual number of victim- Shay Bilchik including serious and violent delinquency, ized children. The following estimates Administrator teen pregnancy, drug use, low academic were derived from annual surveys of all 50 States and the District of Columbia glect, known and unknown, is about 2,000 relates of Delinquency. To determine how conducted since 1986 by the National (or some 5 children every day) (U.S. De- maltreated youth differ from the general Committee to Prevent Child Abuse partment of Health and Human Services, adolescent population, the researchers (NCPCA) (Lung and Daro, 1996). Administration for Children and Families, asked: First, how many cases of maltreatment 1995). ◆ What is the range and extent of prob- of children are reported in the United The focus of this Bulletin is not on the lems found among adolescents who States? In 1995, NCPCA estimated that: victims whose lives are tragically ended were maltreated as children, compared ◆ There were 3,111,000 children reported as a direct result of child abuse and ne- with those who were not? to Child Protective Services as alleged glect. Rather, it concentrates on the ma- ◆ To what extent is child maltreatment a victims of maltreatment. jority of child maltreatment victims who risk factor for the development of mul- have experienced maltreatment during tiple problems in adolescence? ◆ Reports have steadily risen over the critical stages of child development. past decade, with a 49% increase from The researchers found that maltreat- 1986 to 1995. ment during childhood significantly Adolescent ◆ Forty-six out of every one thousand increased the risk for a variety of adoles- American children (under 18 years old) Consequences of cent problem behaviors, including delin- were victims of reported child mal- Child Maltreatment quency, teen pregnancy, drug use, low treatment in 1995, reflecting fairly academic achievement, and mental health In addition to the immediate and short- problems. steady growth from the reported rate term harm of child maltreatment, there of 33 per 1,000 children in 1986. are long-term ramifications of childhood Also, youth who had been victims of Second, how many reports were sub- victimization. Research has shown that child maltreatment were more likely than stantiated? NCPCA estimated that: the consequences of child maltreatment nonmaltreated youth to exhibit multiple problem behaviors. ◆ Approximately 1 million children were extend into adolescence and beyond found to be victims of maltreatment (Widom, 1994). Survivors of child mal- Subsequent analyses are planned to each year from 1992 through 1995. treatment experience many problems in address additional research questions: the course of adolescent development. A ◆ ◆ What key risk factors appear to in- Fifteen of every one thousand children variety of negative teenage outcomes, were substantiated victims of child crease the likelihood that a maltreated such as delinquency, pregnancy, alcohol child will become a troubled youth? maltreatment in each of these years. and drug abuse, school failure, and emo- ◆ ◆ About one-third of the reports of child tional and mental health problems, have What protective factors serve to en- maltreatment were confirmed. been identified. hance youth resilience despite a history of child maltreatment? ◆ Of the substantiated 1995 reports, 54% Many researchers have utilized conve- involved neglect, 25% physical abuse, nience samples of incarcerated delin- This research clearly has implications 11% sexual abuse, 3% emotional mal- quents, drug rehabilitation clients, for policy and intervention efforts that treatment, and the remainder involved pregnant teenagers, or institutionalized target those challenging cases of adoles- other forms of maltreatment. (The mental patients to conduct retrospective cents who have problems in more than breakdown of reported 1995 cases was analyses of child maltreatment. Such one area of life. The findings should be of essentially identical.) studies have generally concluded that interest to all persons in a position to pre- these troubled populations more fre- vent child maltreatment and to those who Third, how often do children die as a quently report histories of child maltreat- could intervene in the lives of victimized result of abuse or neglect? NCPCA cautions ment than would be expected in the children and youth. Therefore, this Bulle- that the number of fatalities officially re- general population. tin is intended not only for policymakers corded as child maltreatment is considered and practitioners but also for parents, an undercount of the actual incidence. Several methodological limitations af- who are in the best position to protect The NCPCA survey estimates that: fect the confidence that can be placed in and nurture their children’s development. the conclusions drawn from these types ◆ During the past 10 years, more than of studies, however. Few studies establish three children died each day as a an adequate comparison group in their Overview of the result of parental maltreatment. samples or measure an adequate range of Study Design ◆ Abuse is the most common cause of risk factors, protective factors, and prob- The Rochester investigation of mal- death (48%), followed by neglect (37%) lem behaviors. and a combination of abuse and treatment outcomes was designed to con- neglect (15%). sider several different domains of Rochester Youth adolescent problem behavior utilizing a ◆ The majority of victims (85%) are un- Development Study general population sample. Specific prob- der age 5, and nearly half (45%) of the lem areas that are measured include: victims never reach their first birthday. The Rochester Youth Development Study, conducted by researchers at the ◆ Moderate, serious, and violent In a 1995 report, the U.S. Advisory University at Albany, State University of delinquency. Board on Child Abuse and Neglect con- New York, is one of three collaborative ◆ cluded that the most realistic estimate of Teen pregnancy. longitudinal projects under OJJDP’s Pro- annual child deaths from abuse and ne- ◆ Drug use. gram of Research on the Causes and Cor-

2 ◆ Low academic achievement. comes. This analysis does not explore the maltreatment as a whole. However, the ◆ Mental health problems. separate but related topic of adolescent researchers did not attempt to separate maltreatment (Garbarino, 1989), which the outcomes of the seven specific sub- Sample may have continued from childhood or types of maltreatment, as the number of begun after age 12. incidents in each subtype was insufficient The sample of 1,000 youth (75% boys The term “child maltreatment” encom- for meaningful statistical analysis. Fur- and 25% girls) was drawn from public thermore, many of the victims were school students attending seventh and passes a fairly broad spectrum of abuse and neglect scenarios. For the purposes subjected to more than one type of eighth grade in Rochester, New York, in maltreatment. spring 1988. African-Americans make up of this analysis, maltreatment included the majority of the sample (68%), while seven subtypes: physical abuse, sexual Recent research provides some indica- 17% are Hispanic and 15% are white. This abuse, physical neglect, lack of supervi- tions that the overall experience of mal- sample was stratified to proportionally sion, emotional maltreatment, educa- treatment, rather than the specific overrepresent those students considered tional maltreatment, and moral-legal subtype, is what has the greatest impact at high risk for delinquency and drug use, maltreatment. The researchers utilized on adolescent outcomes. For instance, in based upon both their gender and their the maltreatment classification and Widom’s (1996) study of a larger sample residence in high-crime neighborhoods. 5-point severity rating scales developed of 1,575 court cases of child maltreat- However, all data analyses presented here by Cicchetti and Barnett (1991) to code ment, the experience of neglect appeared have been statistically weighted to repre- the maltreatment case records. Table 1 to be as damaging as physical abuse. sent the general cohort of all seventh and contains definitions of maltreatment sub- Child victims of neglect were almost as eighth graders in the Rochester public types and examples from the severity rat- likely as physically abused victims to be schools. ing scales of the least and most serious arrested for violent crimes as teenagers types of cases (Barnett et al., 1993). and young adults. Both neglect and abuse Data Collection The researchers recognize that reliance victims were more likely to be arrested for and involved in violent crimes than Researchers commenced data collec- on substantiated records of maltreatment is likely to produce conservative estimates members of the matched nonmaltreated tion in spring 1988 by privately interview- control group. ing youth and their primary caretakers. of the prevalence and frequency of this The interviews were repeated every 6 often undetected or unreported type of Table 2 presents the prevalence of months. (Each 6-month period is referred victimization. Furthermore, although maltreatment for major demographic sub- to as a “wave.”) This report uses inter- most of the sample youth had resided in groups in the Rochester study. There are view data collected in waves two through Monroe County since birth, a small pro- no statistically significant differences in eight. At wave two, the subjects were in portion had not. The researchers also the prevalence of maltreatment by sex or the fall semester of the eighth or ninth expect there may be a few child maltreat- race. Unfortunately, the total number of grade and were 14 years old, on average. ment victims lacking Monroe County maltreatment incidents was not large At wave eight, they were in the fall semes- records who have been inaccurately enough to allow for sex-by-race compari- ter of 11th or 12th grade and were 17 categorized as nonmaltreated in this sons. (Throughout this discussion years old. In addition to data collected in sample. Any bias that results from this “statistical significance” is defined as the face-to-face interviews, data were col- misclassification would be quite small probability that the observed relationship lected from the Rochester public schools, and would work against the study hypoth- would occur strictly by chance less than Police Department, Department of Social esis. That is, it would make it harder, not 5% of the time (probability, or p < .05).) Services, and other agencies that had easier, to find that early maltreatment is There are, however, significant differences contact with the subjects. related to later delinquency. by social class and family structure. Twenty percent of youth reared in disadvantaged families—those in which the Assessment of Prevalence of Child principal wage earner was unemployed, Maltreatment Maltreatment welfare was received, or income was be- low the poverty level—were victims of Data on maltreatment were obtained Of the total sample of 1,000 youth, 14% maltreatment, while only 8% of the from Child Protective Services records of have a child maltreatment record and 86% nondisadvantaged respondents were mal- the Monroe County Department of Social do not. These subjects were maltreated in treated. The largest difference observed Services, the county of residence for all a total of 219 separate incidents (an aver- is in family structure. Only 3% of the boys subjects at the start of the project. For age of 1.5 incidents per maltreated sub- and girls who resided with both biological each subject, the researchers recorded ject). This rate is somewhat higher than parents at the beginning of the Rochester any instance of substantiated abuse or prevalence rates reported in national data study (spring 1988) had a history of mal- maltreatment from birth through 1992. (U.S. Department of Health and Human treatment, but 19% of those in other fam- For the purposes of this analysis, the Services, National Center on Child Abuse ily situations had been maltreated. term “child” refers to someone who has and Neglect, 1997) and is probably related not had his or her 12th birthday. The re- to the urban population of youth in the This demographic information serves searchers established this cutoff so they sample. as a backdrop for examining the central could more accurately assess the impact The 219 separate incidents of maltreat- issue of this analysis—the relationship of maltreatment during childhood on ment constitute a large enough number to between earlier maltreatment and the subsequent adolescent behavioral out- investigate the adolescent outcomes of development of adolescent problem behaviors.

3 Table 1: Defining Child Maltreatment and Rating Its Severity

Examples of Least and Subtype of Maltreatment Brief Definition Most Severe Cases

Physical Abuse A caregiver inflicts a physical injury upon Least—Spanking results in minor a child by other than accidental means. bruises on arm. Most—Injuries require hospitalization, cause permanent disfigurement, or lead to a fatality.

Sexual Abuse Any sexual contact or attempt at sexual Least—A child is exposed to porno- contact that occurs between a caretaker graphic materials. or responsible adult and a child for Most—A caretaker uses force to make a the purposes of the caretaker’s sexual child engage in sexual relations or gratification or financial benefit. prostitution.

Physical Neglect A caretaker fails to exercise a minimum Least—Food is not available for regular degree of care in meeting a child’s meals, clothing is too small, child is not physical needs. kept clean. Most—A child suffers from severe malnutrition or severe dehydration due to gross inattention to his or her medical needs.

Lack of Supervision A caretaker does not take adequate Least—An 8-year-old is left alone for precautions (given a child’s particular short periods of time (i.e., less than 3 emotional and developmental needs) to hours) with no immediate source of ensure his or her safety in and out of the danger in the environment. home. Most—A child is placed in a life- threatening situation without adequate supervision.

Emotional Maltreatment Persistent or extreme thwarting of a Least—A caretaker often belittles or child’s basic emotional needs (such as ridicules a child. the need to feel safe and accepted). Most—A caretaker uses extremely restrictive methods to bind a child or places a child in close confinement such as a closet or trunk for 2 or more hours. Educational Maltreatment A caretaker fails to ensure that a child Least—A caretaker allows a child to miss receives adequate education. school up to 15% of the time (when he or she is not ill and there is no family emergency). Most—A caretaker does not enroll a child in school or provide any educational instruction.

Moral-Legal Maltreatment A caretaker exposes or involves a child in Least—A child is permitted to be present illegal or other activities that may foster for adult activities, such as drunken delinquency or antisocial behavior. parties. Most—A caretaker causes a child to participate in felonies such as armed robbery.

Source: Adapted from Barnett et al. (1993).

4 ate, serious, violent, and general. Table 3 Table 2: Prevalence of Reported Maltreatment Prior to Age 12 lists the actual interview questions. In the by Demographic Characteristics table, the minor, moderate, and serious items are mutually exclusive; the general delinquency index includes all of these Reported No items and is, therefore, not included in Maltreatment Maltreatment the table. The violent offenses are also categorized in either the moderate or the Total Sample 14% 86% serious index.

Demographic Characteristics Findings Regarding the Prevalence of Delinquency Sex Based upon examination of official po- Male 13% 87% lice records and self-reports, maltreated Female 14% 86% children in this sample were significantly more likely to become involved in delin- Race quency at some time. Figure 1 displays White 16% 84% the relationship between prevalence rates for child maltreatment and delinquency. African American 14% 86% While 32% of the nonmaltreated youth Hispanic 9% 91% have official records of delinquency, a history of maltreatment raises this figure to Disadvantaged Family 45%. Maltreated youth also give higher Yes 20% 80% self-reports of having been involved (at No 8% 92% any time) in each delinquency category: minor, moderate, serious, violent, and Family Structure general. Two Biological Parents 3% 97% The researchers further analyzed the Other Family Structures 19% 81% prevalence data, controlling for the following variables generally considered to influence delinquency rates: the subject’s Maltreatment and The official measure of delinquency is sex, race/ethnicity, family disadvantage, based on the number of times each sub- family structure, and mobility. In other Delinquency ject had an official contact with police as words, the researchers wanted to see if It is commonly observed in research a juvenile or an arrest as an adult. Official the effect of maltreatment on delinquency literature that “lives containing delin- contacts include cases in which the juve- outcomes was spurious and would be quent behavior often seem to contain nile was “warned and released” by the “washed out” if the maltreated subjects maltreatment as well” (Garbarino and police and an official record of the event were comparable to the nonmaltreated Plantz, 1985). Most research on the rela- was maintained and cases in which the subjects along these variables. With these tionship of maltreatment to delinquency juvenile was referred to family court. Data variables held constant, maltreated youth either relied exclusively on official were collected from the files of the Roch- still displayed significantly higher preva- records to measure delinquency or used ester Police Department, which maintains lence rates of delinquency in terms retrospective designs requiring subjects a registry of all such contacts from all of three measures: official records, self- to recall their juvenile misdeeds. Results police agencies in Monroe County. The reported moderate delinquency, and have varied regarding whether a mal- official data cover the time from the self-reported violence. treated individual is more likely to engage subject’s first official contact through in specific categories of delinquency such 1992, when the subjects averaged 17 Findings Regarding the as minor, serious, or violent crimes. years of age. Frequency of Delinquency Self-reported delinquency was as- Do maltreated subjects commit delin- Measurement of sessed in the face-to-face interviews. At quent acts more often than nonmaltreated Delinquency each of seven interview waves, the sub- subjects? In all of the measured categories— The Rochester researchers distinguished jects were asked to report their delin- that is, official records and self-reports their study from previous investigations quent activities for the previous 6 months (minor, moderate, serious, violent, and of the relationship of maltreatment to (the period between interviews). Prior to general)—maltreated subjects engaged in delinquency by utilizing a general popula- analyzing these data, the researchers delinquent activities significantly more tion sample with measurement of official screened out any reports of trivial actions frequently. records of delinquency (both “current” that would be ignored by local law en- Again, the researchers controlled for self-reported offenses). (See Smith and forcement officials. In this analysis, the sex, race/ethnicity, family disadvantage, Thornberry (1995) for a detailed presen- researchers examine responses by youth family structure, and mobility. Even with tation of these results.) to questions that were assigned to five these controls in place, maltreated subjects categories of delinquency: minor, moder- displayed significantly higher frequencies

5 Table 3: Interview Items for Self-Reported Delinquency Indexes*

Minor Moderate Serious Violent Since we interviewed you last time, have you . . . Delinquency Delinquency Delinquency Delinquency

1. Carried a hidden weapon? 2. Been loud or rowdy in a public place where someone complained and you got in trouble? 3. Been drunk in a public place? 4. Damaged, destroyed, marked up, or tagged somebody else’s property on purpose? 5. Set fire or tried to set fire to a house, building, or car on purpose? 6. Gone into or tried to go into a building to steal or damage something? 7. Tried to steal or actually stolen money or things worth $5 or less? 8. Tried to steal or actually stolen money or things worth $5–$50? 9. Tried to steal or actually stolen money or things worth $50–$100? 10. Tried to steal or actually stolen money or things worth more than $100? 11. Tried to buy or sell things that were stolen? 12. Taken someone else’s car or motorcycle for a ride without the owner’s permission? 13. Stolen or tried to steal a car or other motor vehicle? 14. Forged a check or used fake money to pay for something? 15. Used or tried to use a credit card, bank card, or automatic teller card without permission? 16. Tried to cheat someone by selling them something that was not what you said it was or that was worthless? 17. Attacked someone with a weapon or with the idea of seriously hurting or killing them? 18. Hit someone with the idea of hurting them? 19. Been involved in gang or posse fights? 20. Thrown objects such as rocks or bottles at people? 21. Used a weapon or force to make someone give you money or things? 22. Made obscene phone calls? 23. Been paid for having sexual relations with someone? 24. Physically hurt or threatened to hurt someone to get them to have sex with you? 25. Sold marijuana/reefer/pot? 26. Sold hard drugs such as crack, heroin, cocaine, or LSD/acid? Total Number of Items 2986

* This table lists the actual interview questions. The minor, moderate, and serious items are mutually exclusive; all 26 items are included in the general delinquency index and, therefore, it is not shown.

6 jects with the less-maltreated subjects Figure 1: Relationship Between Prevalence of Child Maltreatment and when comparing the less-maltreated and Delinquency subjects with the more-maltreated subjects. The more-maltreated subjects were Percentage arrested about twice as often as the less- maltreated subjects along each of the four 79% 80 dimensions. 71% 70 70% 70% In terms of frequency of self-reported serious and violent crime, the researchers 60 observed incremental increases from 56% 56% nonmaltreated subjects to less-maltreated 50 subjects to more-maltreated subjects 45% 45% across all four dimensions of maltreat- 42% ment: frequency, severity, duration, and 40 37% variety. Significant increases appear be- 32% 33% 30 tween the nonmaltreated subjects and more-maltreated subjects. However, the 20 distinctions between the less-maltreated and more-maltreated subjects are fairly 10 modest in terms of self-reported serious and violent crime. 0 Overall, the Rochester study data con- Official Minor Moderate Serious Violent General firm the general finding that there is a Delinquency Delinquency Delinquency Delinquency Delinquency Delinquency relationship between childhood maltreatment and later delinquency. Furthermore, Nonmaltreated Maltreated this analysis offers some support for the Youth Youth strength of the relationship between childhood maltreatment and later serious or violent delinquency, which increases of delinquency in the following four mea- duration, and type. The researchers hy- as the seriousness of the maltreatment sures: official records and moderate, seri- pothesize that children exposed to more increases. ous, and violent self-reported delinquency. extreme maltreatment will have higher With the control variables held con- rates of delinquency than children ex- stant, the researchers concluded that posed to less extreme maltreatment. Maltreatment and Teen childhood maltreatment is a significant Therefore, whatever causal mechanisms Pregnancy link early maltreatment to later delin- and nonspurious risk factor for officially Vobejda (1996) reports that, according quency ought to be strongest as the fre- recognized delinquency, violent self- to the National Center for Health Statis- quency, severity, duration, and variety of reported delinquency, and moderate tics, births to teenage mothers increased maltreatment increase. self-reported delinquency. Maltreatment in every State from the mid-1980’s until is inconsistently related to serious delin- To test this hypothesis, the researchers 1990. Beginning in 1991, rates have dropped quency, being a significant predictor for identified the top third of their maltreated in 46 States, with national birth rates for the frequency but not for the prevalence sample along each of the following four teenage mothers declining for 4 straight of serious delinquency. Finally, childhood dimensions: years. The number of births per 1,000 maltreatment is not a significant risk ◆ Number of maltreatment incidents. girls ages 15 to 19 has fallen from 62 in factor for minor delinquency or general ◆ 1991 to 57 in 1995. Nevertheless, birth delinquency, the overall index containing Severity of all incidents. rates for teenage mothers remain much the full range of delinquency. In short, ◆ Duration of all incidents. higher in the United States than in other maltreatment appears to be a risk factor ◆ Number of subtypes of maltreatment industrialized nations. for the more serious, but not the less seri- across all incidents. Experiencing pregnancy as a teenager ous, forms of delinquency. These more-maltreated subjects were has serious and often deleterious conse- then compared with the remaining less- quences on the lives of both the young Variations in the Maltreat- mother and her baby. Teenage mothers ment Experience and maltreated subjects and the nonmaltreated subjects. and fathers are usually ill equipped to Delinquency Outcomes provide effective parenting and often draw To this point, maltreatment has been The strongest support for the hypoth- heavily upon the resources of their ex- measured as a simple dichotomy, compar- esis was found in analyzing official tended families and communities. For ing adolescents who had official records records of the frequency of delinquency teenage parents raised in dysfunctional or of maltreatment as children with those outcomes. For each of the four dimensions abusive families, parenting problems may who did not. This measure ignores the cited above, there were significant in- be even more evident, and family support fact that maltreatment can vary along creases in officially recorded delinquency limited. Therefore, it is important to such dimensions as frequency, severity, when comparing the nonmaltreated sub-

7 identify risk factors associated with teen Maltreated boys did not report higher A Closer Look at pregnancy. rates of impregnating girls than non- Maltreatment Subtypes Research on the effects of maltreat- maltreated boys. Precursors of teen fa- and Pregnancy ment on child development suggests a therhood merit further inquiry, as this experience is likely to have negative con- It is interesting to note several key number of pathways through which mal- findings from the extended analysis of treatment could heighten the risk for teen sequences in the lives of the fathers and their children. Researchers in the Roches- this topic conducted by Smith (1996a) pregnancy (Smith, 1996a). Although stud- utilizing Rochester study data. Approxi- ies following maltreated children have not ter study are examining a broader range of factors that appear to increase the risk mately one-half of the maltreated females generally found evidence linking maltreat- (21 out of 40) had been exposed to more ment with subsequent teen pregnancies, a of adolescent fatherhood (Smith, 1996b; Thornberry et al., forthcoming). than one type of maltreatment. In this small body of retrospective research with general population sample, meaningful pregnant girls suggests that a high pro- The balance of this discussion focuses comparisons of pregnancy outcomes portion of them have histories of abuse. on the findings regarding pregnancy out- could not be conducted across maltreat- Most researchers have focused on in- comes among the female subjects. Rates ment subtypes. There were simply too vestigating the link between childhood of teen pregnancy are significantly higher few “single-type” maltreatment cases. For sexual abuse and teen pregnancy, hypoth- among girls with a history of childhood example, among the five sexually abused esizing that the trauma of this particular maltreatment. The prevalence of preg- female subjects, four became pregnant type of maltreatment may lead to sexual nancy among maltreated girls is 52% com- as teenagers; however, three were also preoccupation, precocious sexual behav- pared with 34% among nonmaltreated exposed to other forms of maltreatment. ior, and sexual offending. However, many girls. The risk of becoming pregnant is Therefore, it cannot be determined from victims of childhood maltreatment are therefore approximately 50% higher these data whether it is sexual abuse or exposed to multiple types of maltreat- among high school girls who experience other forms of abuse that increase the ment. This implies that researchers maltreatment during their childhood. risk of pregnancy in female adolescents. These prevalence rates are shown in fig- should not focus solely on sexual abuse Girls exposed to multiple types of mal- but should examine the broad range of ure 2 (which also highlights key findings regarding self-reported serious delin- treatment are significantly more likely to maltreatment experiences, resulting de- become pregnant than girls who experi- velopmental disruptions, and possible quency, violence, drug use, lower grade point averages (GPA’s), and mental health enced one type of maltreatment. In addi- links to teen pregnancy. tion to maltreatment per se, it may be the problems). The few studies directly exploring the range of developmental domains affected relationship between childhood maltreatment and subsequent teen pregnancy all have drawn samples on the basis of mal- Figure 2: Relationship Between Prevalence of Child Maltreatment treatment or pregnancy. The Rochester and Various Negative Outcomes During Adolescence study provides a unique opportunity to examine the link between childhood mal- Percentage treatment and teen pregnancy among a 70% general population sample. 70 Measurement of Teen 60 56% Pregnancy 52% Teen pregnancy, for the purposes of 50 this discussion, refers to subjects’ self- 42% 43% reports of having been pregnant at 40 any time. If female subjects answered 34% 33% positively in any interview in waves five 33% 32% (when the average age was 15.3 years) 30 26% through nine (when the average age was 23% 17.3 years), the female subjects were clas- 20 15% sified as having been pregnant. Male subjects were asked if they had ever made a 10 female pregnant. Thus, in this presentation, teen pregnancy involves subjects’ 0 reports of pregnancies up to the average Serious Violence Pregnancy Drug Use Lower GPA Mental age of 17.3 years (high school age). Delinquency (Females) Health Problems Findings Regarding the Prevalence of Teen Nonmaltreated Maltreated Pregnancy Youth Youth For the purposes of this analysis, males and females were separated.

8 by maltreatment that is associated with the maltreated victims may seek escape measure self-reported drug use among a pregnancy risk. Fortunately, not all mal- from the harsh realities of their lives, general population sample of adolescents. treated girls report pregnancies. Maltreated even after the abuse has ended. Those The researchers can now examine whether teenage girls who become pregnant are feeling socially isolated might be willing childhood maltreatment predicts adoles- more likely to display a constellation of to use drugs in order to be part of a group. cent drug use. As the subjects age, the risk factors in early adolescence, includ- Other victims may seek to alleviate suffer- researchers will be able to assess whether ing early substance use, early sexual inti- ing from low self-esteem or mental disor- the drug-related consequences of child macy, and poor academic performance. ders by self-medicating. maltreatment extend into or emerge dur- Although reasonable rationales have ing adulthood. Maltreatment and been put forth as to why maltreatment victims would use drugs, the research on Measurement of Drug Use Other Problem this issue is inconclusive. When Ireland The Rochester study conducted face- Behaviors and Widom (1994) reviewed the literature, to-face interviews to assess drug use. At The most complete analysis of the they found both ambiguous and conflict- each of seven interview waves, the sub- consequences of maltreatment using the ing findings. Research designs typically jects were asked to report their drug use Rochester data has been conducted with are cross-sectional, rather than longitudi- during the previous 6 months. Table 4 delinquency and teen pregnancy as the nal, so it is difficult to draw any causal provides the 10 interview items for drug outcome variables. As discussed above, inferences. There are few adequate con- use. The drug-use index does not include the analyses examine different forms of trol or comparison groups. Many studies alcohol or cigarette use. delinquency, evaluate the role of con- have focused on adult female samples, founding variables, and address different and there is evidence that gender may Findings Regarding the dimensions of maltreatment. affect the drug-use outcomes of child Prevalence of Drug Use maltreatment. Recently, the researchers on the Roch- Youth with a history of childhood mal- ester project have begun to explore the The impact of child maltreatment on treatment are more likely to use drugs relationship of childhood maltreatment to drug use may not be fully realized until than nonmaltreated subjects. Based upon other adolescent outcome variables, but the adult years. In Ireland and Widom’s self-reports, 43% of the Rochester study not yet in as much detail as for the delin- (1994) study, after controlling for relevant adolescents who were maltreated as chil- quency and pregnancy analyses. For demographic variables, child maltreat- dren have used drugs, compared with these other outcomes, more complete ment was a significant predictor of adult, 32% of those who were not maltreated. multivariate models will be tested in the but not juvenile, arrests for alcohol and/ Thus, the risk of using drugs is about one- future. Following are highlights of the or other drug-related offenses. third higher among youth who have a analyses conducted to date on the rela- The Rochester longitudinal design maltreatment history. These drug-use tionship of maltreatment to drug use, low provides an excellent opportunity to rates are displayed in figure 2. academic achievement, and mental health problems. Table 4: Interview Items for Self-Reported Drug Use Maltreatment and Drug Use Since we interviewed you last time, have you . . . The catalysts for the onset of drug abuse in adolescence are many. Teenage 1. Used marijuana/reefer/pot? drug users frequently cite the following current problems in their lives: peer pres- 2. Inhaled things (other than cigarettes) like glue to get high? sure to use drugs, difficulties in school, rebellion against adult authority, lack of 3. Tried LSD/acid/cubes? impulse control, desire for immediate gratification, and low self-esteem. When 4. Tried cocaine/coke/snow (other than crack)? researchers (Dembo et al., 1992) interviewed teenage drug users about earlier childhood experiences, the youth fre- 5. Tried crack? quently reported histories of physical child abuse and sexual victimization. 6. Tried heroin/smack? Dembo concluded that developmental damage caused by child maltreatment 7. Tried angel dust/PCP? increases the likelihood of adolescent drug use and other problem behaviors. 8. Tried tranquilizers such as ludes (Quaaludes) or Valium? A number of explanations have been offered as to why a maltreated child 9. Tried downers such as yellow jackets or red or blue devils? would use drugs in subsequent years (Ireland and Widom, 1994). For instance, 10. Tried uppers/speed such as bennies or black beauties?

9 Maltreatment and Low Measurement of Academic support by parents or other primary care- Achievement takers. Obviously, a child traumatized by Academic Achievement maltreatment, particularly chronic and To assess how well subjects performed Does the childhood experience of mal- severe maltreatment within the family at school, researchers computed their treatment increase the likelihood of aca- setting, suffers (at minimum) a temporary cumulative GPA’s from 1986 to 1989 Roch- demic difficulties and failure during setback in his or her positive psychologi- ester City School District records. In 1989, middle school? Some research has been cal development. More is known about the mean age was 14.4 years, and subjects conducted on adolescent learning prob- short-term consequences than long-term were in the spring semester of either the lems, school failure, and truancy among consequences of maltreatment, particu- eighth or ninth grade. Therefore, the cu- abused populations (Malinosky-Rummell larly in instances where Child Protective mulative GPA is basically a measure of a and Hansen, 1993). Several studies have Services are notified and mental health student’s grades during middle or junior noted that maltreated children and ado- assessments are conducted. high school. In this analysis, a GPA of less lescents exhibit intellectual and academic than 2.0 (less than “C”) is categorized as Although each child responds differently delays and lower scores on tests of intelli- low. to life circumstances and maltreatment gence. Considerably more is known about experiences, child victims often have a the immediate childhood consequences wide range of maladaptive emotional and of maltreatment on academic success Findings Regarding Academic Achievement interpersonal symptoms in common. The than about the long-term adolescent and negative attributes associated with this adult outcomes. The Rochester study shows that school victimized population include anxiety, achievement by maltreated children is Because the experience of maltreat- inattentiveness, impulsiveness, anger, significantly lower than that of youth who ment can delay or disrupt normal child aggression, passivity, withdrawal, depres- do not have a history of childhood mal- development, research literature often sion, self-destructiveness, obsessive- treatment. Poor grades are evident among emphasizes that maltreated children compulsive behavior, and unpopularity 33% of the maltreated group, compared exhibit problems in concentration and (Erickson et al., 1989). Such child victims with 23% in the group who were not mal- achievement in school. Researchers are further described as lacking self- treated. The prevalence rates for low (Erickson et al., 1989) examined the kin- confidence, empathy, and perhaps most GPA’s are presented in figure 2. dergarten experience of maltreated and telling, joy. nonmaltreated children. Maltreated chil- According to the study, then, students What is known about the long-term performing poorly in middle school are dren displayed a host of difficulties in consequences of child maltreatment for adapting to the social environment and considered at increased risk for contin- victims in adolescence? Maltreatment has task demands of school. For instance, ued academic failure in high school, low been linked to a number of mental health educational aspirations, premature children who were maltreated were sig- problems among adolescents, including nificantly more likely than their nonmal- school dropout, and reduced educational increased self-destructive and suicidal and economic opportunities. The Roches- treated peers to need special assistance, behavior, fewer interpersonal competen- create classroom disturbances, perform ter researchers plan additional data col- cies, and more mood disorders, such as lection and analysis to compare the high poorly, and comprehend less and to be anxiety and depression (Malinosky- inattentive, overactive, unpopular, and school experiences of maltreated youth Rummell and Hansen, 1993; Downs, 1993). with those of nonmaltreated youth. aggressive. For at least some victims, the impact of Do such negative outcomes continue childhood maltreatment extends beyond to impede academic success in adoles- Maltreatment and adolescence, as documented by both recent years? Due to the sequential nature Mental Health Problems searchers and clinicians working with of the academic learning process, it is adults with alcoholism and mental illness. likely that maltreated children who never In considering the possible links be- mastered basic skills would have diffi- tween maltreatment and subsequent men- Measurement of Mental culty applying these skills to the more tal health problems, it is useful to employ Health Problems complex tasks found in middle school a developmental approach to psychopa- In the Rochester study, parents were curriculums. Furthermore, adolescents thology (Cicchetti, 1989). Psychopathology asked to assess whether their children with a history of maltreatment might have occurs when there is a lack of integration exhibited certain emotional and behav- less personal capacity to meet social chal- of the various social, emotional, and cog- ioral responses symptomatic of mental lenges in the middle school setting. nitive competencies that underlie healthy or normal adaptation at a particular level illness. The Rochester study investiga- There are indications (Smith, 1996a) of child and adolescent development. In- tors collaborated with their colleagues that girls with histories of child maltreat- terruption of healthy childhood develop- at the Pittsburgh Causes and Correlates ment are less likely to become pregnant if ment by maltreatment may ultimately of Delinquency study site to develop a they aspire to and achieve academic suc- lead to much larger disturbances at sub- shortened version of the Achenbach cess. Other researchers (Zingraff et al., sequent developmental stages. The timing, Child Behavior Checklist without compro- 1994) have noted that adequate school chronicity, and severity of the maltreat- mising the reliability and validity of the performance appears to substantially re- ment are critical factors in assessing their longer, 118-item form. The abbreviated duce the risk of delinquency among mal- potential impact on an individual child’s checklist was advantageous in terms of treated children. Performing well at school psychosocial development. reducing longitudinal survey costs, inter- may indeed be key to adolescent resil- view time, and subject attrition (Lizotte ience following childhood maltreatment. Healthy development requires nurtur- et al., 1992). The shortened behavioral ing, encouragement, supervision, and

10 checklist adopted for use at Rochester baggage than just their presenting offenses. risk factor for multiple problem outcomes. includes seven scales that fall under one Researchers conducted a longitudinal Of the maltreated youth, 32% have three of two broad dimensions: examination of the troubled lifestyles or more negative outcomes, compared (1) Externalizing syndrome—aggressive, of juveniles originally held in a Florida with only 18% in the nonmaltreated delinquent, hostile/withdrawn, and hyper- detention facility (Dembo et al., 1991). group. The risk of multiple negative out- active. Among this high-risk sample, multiple comes increases by nearly 80% among the problems were the norm, rather than the maltreated subjects. (2) Internalizing syndrome—immature, exception. The majority of the detained somatic (physical) complaints, and un- youth came from physically abusive communicative. homes, engaged in delinquency and drug Summary and As shown in table 5, quite a few of the abuse, and experienced academic difficul- Discussion behaviors included in this checklist are ties or school failure. In the past few decades, increasing at- not particularly uncommon among ado- tention has been paid to the phenomenon lescents. What is of clinical concern is not Assessment of Multiple of child maltreatment and its long-term an occasional positive response but Problem Behaviors impact on youth development. A growing rather a significant loading on a particular This analysis looks at whether child- body of research suggests that being mal- scale or syndrome. The checklist is typi- hood maltreatment is a risk factor for hav- treated as a child increases the chances cally used as a diagnostic tool to identify ing multiple problems during adolescence of a variety of developmental problems subjects who fall at the behavioral ex- and utilizes measurement procedures pre- during childhood, adolescence, and adult- tremes on these scales. For the purposes viously described for the following five hood. This presentation of the Rochester of this analysis, adolescents who score in categories of negative outcomes: study data has examined the consequences the top 10% on either the externalizing or of childhood maltreatment in terms of ◆ internalizing syndrome scale are counted Serious delinquency. subsequent adolescent problem behaviors. as having problems suggesting psychopa- ◆ Teen pregnancy. Overall, these results indicate that hav- thology and/or behavioral disturbances. ◆ Drug use. ing a history of childhood maltreatment ◆ Low academic achievement. serious enough to warrant official inter- Findings Regarding Mental vention by Child Protective Services in- Health Problems ◆ Mental health problems. creases the likelihood of problems Mental health problems that affect during adolescent development. Specifi- teenagers include externalized conduct Findings Regarding Multiple cally, subjects with a history of maltreat- problems such as aggressive, hostile, and Problem Behaviors ment were more likely to engage in hyperactive behavior and internalized Figure 3 illustrates the proportion of serious and violent delinquency, use problems such as social isolation, anxiety, maltreated youth and nonmaltreated drugs, perform poorly in school, display and physical distress. One might expect youth with no negative outcomes, one to symptoms of mental illness, and (for such symptoms to be linked with prior two negative outcomes, or three or more girls) become pregnant. Childhood mal- maltreatment, as indeed figure 2 suggests. negative outcomes. treatment is associated with an increased Of subjects whose parents report the risk of at least 25% for each of these in- First, the figure shows that a lower highest number of either externalizing vestigated outcomes. Any one of these or internalizing problems, 15% of proportion of maltreated youth than problems jeopardizes a teenager’s poten- nonmaltreated youth have no problem nonmaltreated teenagers are in this tial for making a successful transition to group, compared with 26% of the mal- outcomes. Still, it is somewhat encourag- adulthood. ing to see that more than one-fourth treated teens. (28%) of youth with histories of child- Maltreatment diminishes the likelihood hood maltreatment are not presenting that children will come through adoles- Maltreatment and any negative adolescent outcomes, as cence with no serious problems. More- Multiple Problem measured in this analysis. In contrast, over, a history of childhood maltreatment 40% of the nonmaltreated youth fall into nearly doubles the risk that teenagers will Behaviors this category; they are more likely to experience multiple problems during Thus far, this Bulletin has attempted to progress through adolescence unscathed adolescence. sort out or isolate specific problem areas by serious problems. In view of the link between maltreat- in the lives of adolescents. Traditionally, Next, maltreated youth (40%) and ment and various problem outcomes, it is this is the way research on adolescent nonmaltreated youth (42%) are almost critical for practitioners and researchers problems has been conducted. It is now equally likely to have only one or two to consider youth holistically, across a time to consider the process of youth de- problems. This suggests that many young- range of areas in which maltreatment may velopment from a holistic perspective. sters, not just those who have official be manifested developmentally. This is Some teenagers have multiple and over- records of childhood maltreatment, have particularly critical for youth who are lapping problems that increase the likeli- a problem in some area of their lives. already experiencing multiple problems. hood that they will not be able to make Implementation of a holistic intervention successful transitions to adult roles and Finally, the figure depicts the group of approach will necessitate a high level of most concern—youth experiencing three responsibilities. coordination among service systems, or more of the five problem areas exam- targeting maltreated children and families Youth who become involved in the ined here. Maltreatment is a significant juvenile justice system often carry more

11 Table 5: Assessment of Mental Health Problems: Rochester Youth Development Study’s Shortened Child Behavior Checklist

Externalizing Scales Internalizing Scales

Parental Assessment Aggressive Delinquent Hostile/ Hyperactive Immature Somatic Uncommunicative of the Following Child Withdrawn Complaints Behaviors: Acts too young Can’t concentrate Is hyperactive Cries a lot Is cruel to animals Is cruel to others Demands attention Destroys others’ things Is disobedient at school Has poor peer relations Feels persecuted Feels worthless Fights Has bad friends Is impulsive Lies, cheats Is disliked Is anxious Has pains Has headaches Has nausea School work is poor Prefers young kids Screams Is secretive Is self-conscious Is shy/timid Steals outside home Is stubborn Is moody Has temper tantrums Threatens people Is sad Whines Is withdrawn Worries Total Number of Items 12 6 8 6 5 4 8

Source: Adapted from Lizotte et al. (1992).

12 ◆ The relationship of the victim to the Figure 3: Relationship Between Prevalence of Child Maltreatment victimizer. and Number of Negative Outcomes During Adolescence ◆ The question of whether maltreatment was reported and, if so, the agency’s Percentage response. ◆ The range of developmental areas af- 50 fected by experiences of maltreatment. 42% With a better understanding of how 40% 40% 40 maltreatment disrupts or derails healthy development in children and adolescents, 32% those who work with young people will be far better positioned to help guide them 30 28% onto productive pathways to the future. Practitioners providing such guidance represent fields such as Child Protective 20 18% Services, education, pediatrics, juvenile justice, substance abuse treatment, and mental health. These practitioners are 10 urged to foster two interdisciplinary approaches: ◆ A comprehensive assessment of 0 Zero 1 to 2 3 or More troubled adolescents to determine if a Negative Outcomes Negative Outcomes Negative Outcomes history of maltreatment continues to negatively impact their lives. Nonmaltreated Maltreated ◆ Individualized treatment of victimized Youth Youth youth to strengthen protective factors that help minimize their involvement in detrimental behaviors. and working with teenagers experiencing maltreated children and their families, As maltreated children grow older, several related problems. within and across jurisdictions, ranging their status as victims may be overlooked Obviously, not every possible adverse from no service provision to placement of as their behavior becomes more disrup- outcome was detected or assessed in this children outside the home (Lung and tive and problematic. There is evidence presentation of the Rochester data. For Daro, 1996). Information on service re- that some institutionalized delinquent instance, alcohol use, school truancy, sponses and other intervening factors is and substance-abusing youth may have school dropout incidents, serious con- vital to the enhancement of efforts to ac- experienced considerable trauma in con- flicts with parents, running away from curately assess and effectively deal with nection with early maltreatment (Dembo home, and teenage parents’ maltreatment maltreatment victims. et al., 1992). When a child victim becomes of their own children were not included in When researchers look for indicators a juvenile offender, legitimate concerns this analysis of adolescent problem be- that increase the risk of negative out- about protecting public safety and holding haviors. Practitioners and researchers are comes for maltreated youth, they should youth accountable for their behavior can urged to consider assessment of a full also identify protective factors or buffers easily overshadow issues of continued spectrum of adolescent consequences that boost resiliency. As noted in the pre- trauma from childhood maltreatment. As when dealing with victims of maltreatment. sentation of findings, 28% of maltreated the Nation seeks to provide a balanced approach to juvenile justice, policymakers, The Rochester research strengthens subjects had no detected problem. This practitioners, and the public must recog- and clarifies the empirical link between is the most encouraging finding: long-term nize that a punitive response such as se- maltreatment and later problem behav- developmental damage does not appear cure detention or incarceration may place iors among a general population sample to be inevitable, as assessed across five the youth at risk of further victimization. of teens. However, these data do not im- problem areas. This finding is important. Punitive responses also may exacerbate ply that maltreatment leads directly or If the processes that lead to resilience can previous emotional and developmental inevitably to later problems. An important be identified, youth-serving professionals problems resulting from maltreatment. theoretical and practical issue for future will be in a better position to develop ef- In this connection, it is particularly im- research is the identification of the path- fective intervention programs for other portant to provide comprehensive assess- ways connecting early maltreatment to maltreated youth. In doing so, it will be ment services to identify problem areas various adolescent problem behaviors. important to consider factors such as the following: and develop responsive services for Such pathways may be affected by in- ◆ juveniles. tervening factors, including the emergence The developmental stage(s) at which Practitioners attempting to design and of protective factors and the provision victimization occurred. implement effective individualized treat- of effective services. There is evidence of ◆ Variations in the frequency, severity, ment plans must consider the context of great variation in official responses to duration, and type of maltreatment.

13 the victim’s family and community envi- able and accessible daycare, to support risk youths’ alcohol and other drug use ronment. As noted by Briscoe (1995): parents’ efforts to provide the emotional, and delinquency: A longitudinal model. . . . if we understand that a child who social, and physical support their chil- Violence and Victims 7(3):245–266. is not nurtured is a child who never dren need. Dembo, R., L. Williams, J. Schmeidler, learns to trust, never develops empa- Admittedly, some mothers and fathers and D. Howitt. 1991. Troubled Lifestyles: thy, never accepts responsibility for will not be receptive to even the most in- High-Risk Youth in Florida. Commissioned his behavior, and hurts others with tensive support and training, perhaps by the Office of the Assistant Secretary impunity, then we will figure out a way because of their own lack of maturity, for Educational Research and Improve- to collaboratively make children and mental illness, alcoholism, drug addic- ment. Washington, DC: U.S. Department families a priority and support and em- tion, criminal involvement, or history of of Education. power them to take responsibility to abuse. In certain instances, the best inter- Downs, W.R. 1993. Developmental con- rear healthy and productive children. ests of a child would necessitate at least siderations for the effects of childhood Traditionally, less emphasis has been temporary removal from the biological sexual abuse. Journal of Interpersonal placed on preventing child maltreatment parents, before the child’s normal devel- Violence 8(3):331–345. than reporting and intervening in these opment is severely compromised by maltreatment. Erickson, M.R., B. Egeland, and R. Pianta. cases. Rather than merely responding to 1989. The effects of maltreatment on reports of child maltreatment, Child Pro- While far more is known about the the development of young children. In tective Services must also dedicate their symptoms of maltreatment than ways to D. Cicchetti and V. Carlson (eds.), Child resources to screening and developing deal with it, research is providing more Maltreatment: Theory and Research on the resources (preferably those that begin information about both. Ultimately, the Causes and Consequences of Child Abuse during the prenatal period) for at-risk degree to which the Nation safeguards and Neglect. Cambridge, MA: Harvard parents. children’s futures will be measured in University Press. Educating all future parents, not just terms of how effectively society responds to abuse and neglect when it occurs and Garbarino, J. 1989. Troubled youth, those at high risk for maltreating their troubled families: The dynamics of ado- children, should be a national priority. how committed the American people are to prevention in the first place. lescent maltreatment. In D. Cicchetti and The United States currently places more V. Carlson (eds.), Child Maltreatment: emphasis on educating and licensing driv- Theory and Research on the Causes and ers than on preparing the next generation References Consequences of Child Abuse and Neglect. of parents. Few new parents are fully cog- Barnett, D., J.T. Manly, and D. Cicchetti. Cambridge, MA: Harvard University Press. nizant of just how much time and energy 1993. Defining child maltreatment: The are involved in nurturing newborns. Many Garbarino, J., and M. Plantz. 1985. interface between policy and research. parents lack familiarity with the normal Child maltreatment and juvenile delin- In D. Cicchetti and S.L. Toth (eds.), Child difficulties and stresses inherent in the quency: What are the links? Literature Abuse, Child Development, and Social care and supervision of developing in- review in E. Gray (ed.), Child Abuse: Pre- Policy. Norwood, NJ: Ablex. fants, children, and adolescents. Parents lude to Delinquency? Washington, DC: should be advised that stress factors Briscoe, J. 1995. Cycle of violence and U.S. Department of Justice, Office of Jus- such as financial concerns, fatigue, alco- the cost of child abuse and neglect. Texas tice Programs, Office of Juvenile Justice hol consumption, and drug use can re- Youth Commission Journal (June 1995):8–12. and Delinquency Prevention. duce their tolerance for normal infant and Cicchetti, D. 1989. How research on Ireland, T., and C.S. Widom. 1994. Child- child behavior and increase their propen- child maltreatment has informed the hood victimization and risk for alcohol sity to abuse their children. To prevent study of child development: Perspectives and drug arrests. The International Jour- child maltreatment, parents need to un- from developmental psychopathology. In nal of the Addictions 2(2):235–274. derstand not only the immediate but also D. Cicchetti and V. Carlson (eds.), Child Lizotte, A.J., D.J. Chard-Wierschem, the long-term consequences of their Maltreatment: Theory and Research on the R. Loeber, and S.B. Stern. 1992. Research actions. Causes and Consequences of Child Abuse note: A shortened child behavior check- Furthermore, parents must recognize and Neglect. Cambridge, MA: Harvard Uni- list for delinquency studies. Journal of that their dependents need extensive versity Press. Quantitative Criminology 8(2):233–245. adult supervision and care for the dura- Cicchetti, D., and D. Barnett. 1991. To- Lung, C., and D. Daro. 1996. Current tion of their childhood. Social and work- ward the development of a scientific no- Trends in Child Abuse Reporting and Fatali- related commitments do not relieve a sology of child maltreatment. In W. Grover ties: The Results of the 1995 Annual Fifty parent of the responsibility to provide and D. Cicchetti (eds.), Child Maltreatment: State Survey. Working Paper Number 808, adequate nurturing and supervision. The Theory and Research on the Causes and Chicago, IL: National Committee to Pre- task of raising a child is particularly diffi- Consequences of Child Abuse and Neglect. vent Child Abuse. cult for parents who are isolated and lack Cambridge, MA: Harvard University Press. an extended family or social network from Malinosky-Rummell, R., and D.J. Hansen. which to draw support. Merely educating Dembo, R., L. Williams, W. Wothke, 1993. Long-term consequences of child- parents about their responsibilities will J. Schmeidler, and C.H. Brown. 1992. The hood physical abuse. Psychological Bulletin not suffice—communities must also de- role of family factors, physical abuse, and 114(1):68–79. velop adequate resources, such as afford- sexual victimization experiences in high-

14 Smith, C. 1996a. The link between childhood maltreatment and teenage Acknowledgments pregnancy. Social Work Research 20(3):131–141. This Bulletin was prepared by Barbara Tatem Kelley, M.A., an independent consultant with extensive experience in juvenile justice research and program Smith, C. 1996b. Teen fatherhood. development; Terence P. Thornberry, Ph.D., Professor and former Dean at the In R. Loeber, D.H. Huizinga, and T.P. School of Criminal Justice, University at Albany, State University of New York; and Thornberry (eds.), Program of research Carolyn A. Smith, Ph.D., Assistant Professor at the School of Social Welfare, on the causes and correlates of delin- University at Albany, State University of New York. Dr. Thornberry is Director of the quency: Annual report 1995–1996, submit- Rochester Youth Development Study; Dr. Smith is coprincipal investigator of the ted to the Office of Juvenile Justice and study. Delinquency Prevention. The authors would like to express appreciation to their colleagues on the Roches- Smith, C., and T.P. Thornberry. 1995. ter Youth Development Study. Particular thanks are due to the Monroe County The relationship between childhood mal- Department of Social Services, especially the Commissioner, Diane Larter, and treatment and adolescent involvement in the Research Director, Daniel Ross, for providing access to the records on delinquency. Criminology 33(4):451–481. maltreatment and for helping us properly interpret the data. We would also like to Thornberry, T.P., C. Smith, and G.J. thank Dante Cicchetti of the University of Rochester for his advice about measur- Howard. Forthcoming. Risk factors for ing maltreatment. Most important, of course, we thank the participants in the teenage fatherhood. Journal of Marriage Rochester Youth Development Study for sharing so much of their time with us over and the Family. the years. U.S. Department of Health and Human Services, Administration for Children and Families. 1995. A Nation’s Shame: Fatal Widom, C.S. 1994. Childhood victimiza- Points of view or opinions expressed in this Child Abuse and Neglect in the United tion and risk for adolescent problem be- document are those of the author(s) and do States. Washington, DC: U.S. Advisory havior. In M.E. Lamb and R. Ketterlinus not necessarily represent the official position Board on Child Abuse and Neglect. (eds.), Adolescent Problem Behaviors. or policies of OJJDP or the U.S. Department of U.S. Department of Health and Human New York, NY: Lawrence Erlbaum Justice. Services, National Center on Child Abuse Associates. and Neglect. 1997. Child Maltreatment 1995: Widom, C.S. 1996. The Cycle of Violence The Office of Juvenile Justice and Delin- Reports from the States to the National Revisited. Research Preview. Washington, quency Prevention is a component of the Of- Child Abuse and Neglect Data System. DC: National Institute of Justice. fice of Justice Programs, which also includes Washington, DC: Government Printing Zingraff, M.T., J. Leiter, M.C. Johnsen, the Bureau of Justice Assistance, the Bureau Office. of Justice Statistics, the National Institute of and K.A. Myers. 1994. Mediating effect Vobejda, B. 1996. Research group of good school performance on the Justice, and the Office for Victims of Crime. confirms decrease in births to teens. The maltreatment-delinquency relationship. Washington Post, October 29, 1996, A–3. Journal of Research in Crime and Delin- quency 31(1):62–91.

15 U.S. Department of Justice BULK RATE Office of Justice Programs POSTAGE & FEES PAID Office of Juvenile Justice and Delinquency Prevention DOJ/OJJDP Permit No. G–91 Washington, DC 20531

Official Business Penalty for Private Use $300

Bulletin NCJ 165257