Abstract
Objective
To examine rates and patterns of health-risk behavior (e.g. sexuality, depression/suicidality, substance use, delinquency) among a national probability sample of youth active to the child welfare/child protective services system. Recent federal legislation, P.L. 110–351, encourages child welfare systems, Medicaid, and pediatric experts to collaborate to assure youth entering foster care receive comprehensive health examinations.
Methods
Analysis of baseline caregiver, caseworker and child interviews, and assessment data for a subsample (n=993) of youth, ages 11–15 years, from the National Survey of Child and Adolescent Well-Being, a national probability sample of children and adolescents undergoing investigation for abuse or neglect.
Results
Almost half of the sample (46.3%) endorsed at least one health-risk behavior. On Poisson multivariate regression modeling, factors related to higher rates of health-risk behaviors included older age, female gender, abuse history, deviant peers, limited caregiver monitoring, and poor school engagement.
Conclusion
Given the heightened vulnerability of this population, early screening for health-risk behaviors must be prioritized. Further research should explore specific subpopulations at risk for health-risk behaviors and possible interventions to change these youths’ trajectories.
INTRODUCTION
Multiple professional organizations serving children have long championed comprehensive health and mental health examinations of children entering foster care. In fact, some groups such as the American Academy of Pediatrics have identified foster care as one of their top priority areas. These advocacy efforts culminated in the passage of Public Law 110–351 in October 2008, which called for enhanced collaboration between child welfare, Medicaid, and pediatric experts to improve health care for children in foster care.1
As planning begins for these collaborative initiatives, identifying health priority areas for assessment, prevention and treatment becomes critical. Traditionally, policy statements have focused on the unique needs of young children entering foster care. 2 The early adolescent entering foster care has received much less attention, even though 113,510 of 463,000 youth in foster care were between 11 and 15 years of age in 2008.3
Health-risk behaviors among the general U.S. youth population are a pediatric public health concern and the target of 28 Critical Health Objectives in Healthy People 2010. 4 While operationalized differently across studies, consistent domains of health-risk behaviors include sexuality, depression/suicidality, tobacco and substance use, and delinquency. 5 These behaviors commonly co-occur, 6, 7 are responsible for much of the morbidity and mortality during adolescence and early adulthood, 8, 9 and have increasingly been linked to poor health status in adulthood. 10 Health-risk behaviors have also been documented as occurring at younger ages 11 highlighting the importance of early prevention and intervention.
An explicit aim of risk and resilience research is screening coupled with the development of targeted and empirically guided interventions to alter risk trajectories and reduce engagement in risk-taking behaviors. Youth who come in contact with child welfare and protective service systems (hereafter, termed “child welfare”) because of allegations of abuse or neglect represent a unique subset of youth at risk for health-risk behaviors. Studies in the general youth population suggest family and school contexts are critical forces influencing behavior. 7 Youth in child welfare are, by definition, at risk for poor family and school contexts. Their experiences of maltreatment, instability, and family dysfunction along with high rates of poverty, poor health, developmental delays, mental health problems, and educational deficits heighten their vulnerability to health-risk behavior engagement. 12 Furthermore, in retrospective studies with adults, childhood abuse and neglect have been found to be markers of adult risk behaviors including tobacco use, substance abuse, sexually transmitted diseases, unintended pregnancy, criminal actions, mental health symptomatology, and work absenteeism.10
To date, few studies have examined health-risk behaviors among youth entering child welfare. A limited number of studies focus on youth in foster care and suggest these youth engage in health-compromising behaviors at an earlier age, with an increased frequency and intensity compared to youth in the general population. 13, 14 Not surprisingly, outcomes for youth transitioning out of foster care at age 18 years have been dismal with low levels of educational attainment; high rates of unemployment, homelessness, welfare dependency, early pregnancy, and criminal justice involvement. 15 However, it is not known how many youth are already displaying risk-taking behaviors around the time of investigation for abuse and neglect or if these behaviors reflect a certain subset of youth who remain in foster care and age out at 18 years.
It is also not known if rates of risk-taking behaviors are higher for youth placed in foster care compared to those youth who remain in their home of origin or are placed informally with kin (termed “in-home” hereafter). While organizations such as the American Academy of Pediatrics, 16 the Child Welfare League of America 17 and the American Academy of Child and Adolescent Psychiatry 17 have affirmed the need for conducting comprehensive evaluations for all youth entering foster care, experts have advocated for evaluations for all children undergoing investigation for abuse and neglect, regardless of whether or not they are placed in foster care.12
The present analysis uses data from the National Survey of Child and Adolescent Well-Being (NSCAW), the first national probability study of 5504 youth referred to child welfare, to obtain baseline estimates of nine health-risk behaviors in four broad domains - sexuality, depression/suicidality, tobacco and substance use, and delinquency. Because the cohort was targeted toward youth age 15 or less at investigation, this sample permits examination of health-risk behaviors in early adolescence among youth investigated for abuse and neglect who remain in-home following investigation or are placed in foster care. NSCAW also allows comparison of rates of risk behaviors across a range of risk and protective factors in the literature affecting developmental outcomes in the general youth population. 18 Three studies have examined specific risk behaviors in the NSCAW sample (e.g. delinquency, 19, 20 depression and substance use)21 but have not investigated health-risk behaviors in a more comprehensive manner. In addition, no comparisons were made to reports of health-risk behaviors in similar types of surveys completed with youth in the general population. Among the national studies examining risk behaviors in youth in the late 1990’s (i.e. the Youth Risk Behavior Survey, the Monitoring the Future Survey, and the National Longitudinal Study of Adolescent Health (ADD Health)), 7, 8, 22 ADD Health was identified as the national study with the most similar measures to the NSCAW study; comparisons are drawn in the Discussion where possible.
METHODS
Study Design
The NSCAW study examined the characteristics, needs, and outcomes of children and families who underwent investigation for alleged abuse and neglect. For the current analysis, we chose a cross-sectional approach, investigating risk behaviors and risk and protective factors at Wave 1, approximately five months following maltreatment investigation.
Sample
NSCAW used a stratified two-stage cluster sampling strategy to select 100 primary sampling units (PSU) from a national sampling frame, with the probability of PSU selection proportional to the size of the PSU’s service population. Of the 100 PSUs identified, the NSCAW study ultimately collected child-level data in 92 PSUs representing 96 counties in 36 states. In participating counties, children were randomly selected from among the population of children ages birth to 15 years, with an investigation opened during a 15-month period beginning October 1999. Information regarding the sampling design and recruitment process are available on-line.(http://www.acf.hhs.gov/programs/opre/abuse_neglect/nscaw.html). This research reports on a subset of youth (n=993/5501) who were at least eleven years of age at the Wave 1 interview.
Procedures
Field representatives conducted face-to-face interviews with youth, biological parents, caregivers, and caseworkers at four waves (baseline, 12 months, 18 months, and 36 months). Wave 1 interviews with child welfare workers and caregivers were completed an average of 5.1 months (SD= 2.7) and 5.5 months (SD=2.8), respectively, after investigation.
Measures
Health-risk behaviors
These behaviors were operationalized as nine dichotomous risk behavior variables: sexual behaviors (sexual relations; getting someone pregnant/becoming pregnant), emotional distress (depression, suicidality), substance use (tobacco, alcohol, marijuana use), and delinquency (fighting, carrying a weapon).
Risk and protective factors
Considerable theoretical and empirical work has shown that biological, psychological, and social factors at multiple levels (individual, family, school, peer group, and community) contribute to health-risk behaviors in the general youth population. For these analyses, we examined: youth demographics (age, gender, ethnicity), placement-related variables (maltreatment type(s), placement at Wave 1), youth cognitive and behavioral functioning, and association with deviant peers. Caregiver factors included caregiver education and parental risk assessment. Protective factors included caregiver monitoring, youth religiosity, school engagement, and future expectations (see Table 1 for a detailed description of variables).
Table 1.
Health-Risk Behaviors Reported and Independent Variables Examined in a Subsample (N=993) of Youth ages 11 and older in the NSCAW Study
| Variables | Measure | Respondent | Description of Measure or Item |
|---|---|---|---|
| Sexual Behaviors | |||
| Sexual Relations | LongSCAN | Youth | Single item: Have you ever had sexual intercourse? |
| Got Someone Pregnant/or Pregnant | LongSCAN | Youth | Single item: Have you ever been pregnant/gotten someone pregnant? |
| Emotional Distress | |||
| Depression | Children’s Depression Inventory | Youth | A 27-item rating instrument of depressive symptoms with cutoff T-score (65) for determining clinically significant depression. |
| Suicidality | Youth Self-Report | Youth | Single item: I have deliberately tried to hurt or kill myself. |
| Substance Use | |||
| Tobacco Use | Drug Free School Community Act Outcome Study Questions | Youth | Single item: In the last 30 days, on how many days did you smoke a cigarette? |
| Alcohol Use | Drug Free School Community Act Outcome Study Questions | Youth | Single item: In the last 30 days, on how many days did you drink an alcoholic beverage? |
| Marijuana Use | Drug Free School Community Act Outcome Study Questions | Youth | Single item: In the last 30 days, on how many days did you use marijuana or hashish? |
| Delinquency | |||
| Fight | Modified Self Report of Delinquency | Youth | Single item: In the past 6 months, have you hit someone with the idea of hurting them? |
| Carry a weapon | Modified Self Report of Delinquency | Youth | Single item: In the past 6 months, have you carried a hidden weapon? |
| Independent Variables | |||
| Youth Demographics | Not applicable | Caseworker report from case records | Age, gender, race/ethnicity. |
| Youth Maltreatment | Modified Maltreatment Classification Scale | Caseworker report from case records | Indicator (i.e. not mutually exclusive) variables for the following categories: neglect/abandonment, physical abuse, sexual abuse, emotional abuse. |
| Youth Placement | Not applicable | Caseworker report from case records | Derived variable of placement disposition following investigation: remain in-home of origin or removal to out-of-home care. Children legally placed with kin were labeled as “out-of-home care”. |
| Youth Behavioral Functioning | Child Behavior Checklist (CBCL), Social Skills Rating System Parent & Teacher Forms (SSRS – P/T), Vineland Adaptive Behavior Scale – Screener (VABS) | Caregiver and Teacher | Derived variable based on the CBCL, SSRS-P/T, and the VABS; the lowest score was used to determine severity; social skills were coded as severely impaired if above the clinical cut point on the CBCL (≥64), or 2 standard deviations below the mean on the SSRS or the VABS. |
| Youth Cognitive Functioning | Kaufman Brief Intelligence Test (K-BIT); Mini-Battery of Achievement (MBA);Preschool Language Scale 3 | Youth | Derived variable based on the lowest K-BIT, PLS 3 and MBA score. K-BIT: standardized assessment tool compromised of two subsets; vocabulary (expressive vocabulary and definitions) & matrices (ability to perceive relationships and complete analogies). MBA: Standardized test of academic achievement; two subtests including reading and mathematics. PLS 3: Standardized assessment tool comprised of expressive communication and auditory comprehension. A score less than 70 on any of these tools was coded as “severe cognitive impairment”. |
| Caregiver Education | Not applicable | Caregiver | Highest school degree achieved by primary caregiver. |
| Deviant Peers | Youth Self Report, CBCL | Youth & Caregiver | Single variable: Parent or youth endorsement of hanging around with kids who get in trouble. |
| Parental Risk Assessment | Derived Measure for NSCAW | Caseworker report from case records | Derived variable of caseworker assessment of family risk factors, including caregiver substance abuse, mental and physical health problems, arrests or detentions, parenting style, family stress, and social support. |
| Caregiver Monitoring | Parental Monitoring Scale | Youth | 6 items (5-point Likert scale) examining the extent to which the primary caretaker monitors youth behavior. |
| Youth Religiosity | LongSCAN | Youth | 2-items (4-point Likert scale) regarding importance of spirituality & religious attendance. |
| School Engagement | Drug Free School Community Act Outcome Study Questions | Youth | 11-items (4-point Likert scale) addressing enjoyment of school, completion of assignments and homework, relationships with teachers and peers. |
| Future Expectations | Adolescent Health Survey | Youth | 8-items (6-point Likert scale) regarding what youth thinks will happen in the future (e.g. live to 35, marry, graduate from high school, have a good job, have a family, have a child before age 18). |
Analyses
Initial analyses aimed to determine the prevalence of nine health-risk behaviors across four domains. Bivariate analyses examined the relationship between each of the risk and protective factors and health-risk behaviors. Age was categorized into three groups (age 11, ages 12–14, and age 15) to facilitate broad comparisons with the 1997 ADD Health study results (see Discussion). Next, we created four combined variables from the nine risk behavior variables: (1) any sexual behaviors, (2) any depression, (3) any delinquency, and (4) any substance use. Based on these, we created clusters of youth based on different types of risk behaviors to investigate patterns of co-occurring behaviors. In preparation for modeling, we looked at the 187 youth who were missing any data. Only one variable, caregiver education, was significant and this was at a level of p = 0.05. Finally, we computed a cumulative risk score based on engagement in health-risk behaviors (range 0–9), which served as the outcome variable in the subsequent multivariate analysis. We used Poisson regression, a statistical method appropriate when rates of events – such as the number of health-risk behaviors – vary across observations and show a skewed distribution. 23 Coefficients are reported as rate ratios and refer to the individual variables’ rate ratios after adjusting for all other covariates. Sample weights (Wave 1) and the two-stage cluster sample design were accounted for using the statistical software SUDAAN (version 10.0). 24 All variables were entered simultaneously into the multivariate model.
RESULTS
Sample Characteristics
The majority of youth were between the ages of 12 and 14 (71.5%) with a mean age of 12.67 (se = 0.07) (see Table 2). Over half of the sample was female (57.4%). In terms of ethnicity of the respondents, half were Caucasian (50.9%), 27.1% were African American, and 16.6% were Hispanic. The most common maltreatment types were neglect (49.8%) and physical abuse (41.5%). The majority (87.8%) were living in-home; only 12.2% were in foster care at Wave 1. About half the youth fell into the clinical range for behavioral functioning, and almost one-fifth scored in the severe range for cognitive functioning. Close to three-quarter of caregivers (71.0%) had at least a high school diploma. Two-thirds of youth (62.3%) reported “hanging out with deviant peers.” Parental risk assessments conducted by caseworkers indicate an average risk rate level of 0.3 with a range from 0.047 to 2.00, which means that families presented with 6–7 risk factors on average. Regarding protective factors, a score of 4.1 on caregiver monitoring indicated that youth in the sample reported being monitored “pretty often.” Youth attended religious services or were engaged in spiritual activities during the past year on average once or twice a month. Youth also reported to be “often” engaged in school activities. Finally, an average score of 4 on the future expectation scale indicated that youth in this sample felt “pretty likely” to have positive life experiences, (e.g. complete high school, have a family, or obtain a good job).
Table 2.
| Risky Behaviors | Sexual Relations | Got someone Pregnant/or Pregnant | Depression | Suicidality | Tobacco Use | Alcohol Use | Marijuana Use | Fight | Carry Weapon | Total |
|---|---|---|---|---|---|---|---|---|---|---|
| Categorical Variables | N % | N % | N % | N % | N % | N % | N % | N % | N % | N % |
| Youth Age | *** | * | ** | *** | ** | ** | ||||
| 11 | 19(5.3) | 2(0.7) | 13(4.2) | 14(7.7) | 11(2.0) | 15(10.9) | 7(1.8) | 15(7.7) | 6(0.7) | 231(22.1) |
| 12–14 | 177(28.5 | 31(4.3) | 77(15.6) | 44(7.4) | 143(23.0) | 98(16.6) | 68(8.7) | 102(13.4) | 55(7.0) | 686(71.5) |
| 15 | 43(58.5) | 12(18.7) | 10(16.8) | 6(14.4) | 29(39.4) | 18(27.6) | 17(23.3) | 16(22.2) | 15(21.6) | 76(6.4) |
| Youth Gender | * | |||||||||
| Male | 86(24.9) | 11(3.8) | 26(7.8) | 21(6.7) | 69(18.1) | 53(15.5) | 37(7.7) | 54(15.2) | 38(8.9) | 417(42.6) |
| Female | 153(25.9) | 34(4.9) | 74(17.1) | 43(8.8) | 114(20.5) | 78(16.5) | 55(8.5) | 79(11.0) | 38(5.0) | 576(57.4) |
| Youth Race/Ethnicity | *** | ** | ||||||||
| African American | 85(30.7) | 15(3.1) | 25(2.6) | 16(7.9) | 34(9.4) | 37(13.1) | 25(8.0) | 38(11.6) | 21(5.1) | 295(27.1) |
| Caucasian | 112(26.4) | 21(5.6) | 49(16.6) | 28(8.2) | 110(26.5) | 64(15.0) | 43(7.4) | 63(14.7) | 36(8.0) | 460(50.9) |
| Hispanic | 22(17.8) | 4(3.7) | 17(19.4) | 10(6.3) | 19(14.6) | 20(27.7) | 13(7.6) | 16(7.9) | 10(5.8) | 149(16.6) |
| Other | 20(12.5) | 5(3.0) | 9(14.5) | 10(10.1) | 20(19.1) | 10(7.7) | 11(17.4) | 16(15.8) | 9(4.2) | 89(5.4) |
| Youth Maltreatment | ||||||||||
| Neglect | 113(21.2) | 20(3.6) | 44(9.2) | 33(7.1) | 82(18.1) | 64(11.9) | 45(8.3) | 72(10.7) | 36(5.1) | 505(49.8) |
| Emotional Abuse | 22(10.4)* | 4(0.9)** | 17(16.2) | 11(8.5) | 21(9.8)* | 14(3.7)*** | 10(5.2) | 22(12.5) | 8(2.4)* | 148(11.9) |
| Physical Abuse | 90(32.4) | 15(6.1) | 38(17.2) | 19(7.9) | 79(24.8) | 55(19.5) | 39(9.1) | 58(16.1) | 32(9.6) | 360(41.5) |
| Sexual Abuse | 79(50.8)** | 18(9.5) | 33(14.4) | 19(9.1) | 45(19.1) | 24(14.5) | 19(4.4) | 20(9.8) | 15(11.2) | 213(15.3) |
| Youth Placement | ||||||||||
| In-Home | 150(25.2) | 32(4.8) | 63(12.6) | 41(7.7) | 134(19.6) | 98(16.4) | 61(8.0) | 90(11.4) | 50(6.1) | 720(87.8) |
| Out-of-Home | 89(27.6) | 13(2.1) | 37(16.6) | 23(9.2) | 49(18.7) | 33(13.2) | 31(8.9) | 43(23.5) | 26(10.2) | 273(12.2) |
| Youth Behavioral Functioning | ||||||||||
| Severe | 139(32.5)* | 30(6.8) | 67(16.7) | 45(11.4) | 117(21.7) | 78(16.8) | 57(10.6) | 80(13.8) | 45(6.3) | 480(48.9) |
| Moderate/Normal | 100(18.7) | 15(2.3) | 33(9.7) | 19(4.5) | 66(17.3) | 53(15.3) | 35(5.8) | 53(11.8) | 31(7.0) | 513(51.1) |
| Youth Cognitive Functioning | ||||||||||
| Severe | 52(24.6) | 18(8.6) | 33(20.1) | 12(10.3) | 34(19.6) | 33(20.7) | 21(15.0) | 18(13.0) | 16(8.9) | 197(18.8) |
| Moderate/Normal | 187(26.0) | 27(3.6) | 67(11.6) | 51(6.9) | 149(19.7) | 98(15.3) | 71(6.7) | 114(12.9) | 60(6.2) | 796(81.2) |
| Caregiver Education | * | |||||||||
| < High School (HS) | 50(27.8) | 15(10.7) | 31(19.6) | 13(8.6) | 42(23.4) | 31(20.7) | 20(12.4) | 25(12.0) | 14(7.2) | 247(29.0) |
| HS Diploma/Equivalent | 100(19.1) | 21(2.7) | 33(8.2) | 28(9.8) | 81(17.4) | 54(14.2) | 37(6.4) | 59(15.1) | 40(9.3) | 427(39.5) |
| > High School (HS) | 89(31.1) | 9(1.0) | 36(13.4) | 23(4.8) | 60(18.3) | 46(14.1) | 35(6.5) | 49(10.8) | 22(2.8) | 319(31.5) |
| Deviant Peers | *** | ** | ** | *** | ** | ** | *** | |||
| Yes | 180(33.1) | 32(5.9) | 78(17.3) | 54(11.9) | 148(26.7) | 96(17.0) | 76(11.4) | 101(17.0) | 61(10.0) | 605(62.3) |
| No | 59(13.2) | 13(2.1) | 22(6.2) | 10(1.3) | 35(7.3) | 35(14.4) | 16(3.0) | 32(5.9) | 15(1.2) | 388(37.7) |
| Continuous Variables n (mean) | ||||||||||
| Parental Risk Assessment | 239(0.3) | 45(0.3) | 100(0.3) | 64(0.3) | 183(0.3) | 131(0.3) | 92(0.3)* | 133(0.3) | 76(0.3) | 993(0.3) |
| Caregiver Monitoring (1–5) | 239(3.7)*** | 45(3.4)*** | 100(3.7)** | 64(3.5)** | 183(3.7)** | 131(3.8)* | 92 (3.3)*** | 133 (3.8)* | 76(3.6) | 993(4.1) |
| Youth Religiosity (1–4) | 239(2.8)* | 45(3.1) | 100(2.8)* | 64(2.9) | 183(2.9) | 131(2.9)* | 92 (3.0) | 133 (3.1) | 76 (3.0) | 993 (3.1) |
| School Engagement (1–4) | 239(2.8)** | 45(2.8) | 100(2.6)*** | 64(2.7) | 183(2.7)*** | 131 (2.8)** | 92 (2.7)* | 133 (2.8)* | 76 (2.6)* | 993 (3.0) |
| Future Expectations (1–6) | 239(3.8)* | 45(3.4)** | 100(3.5)** | 64(3.5)* | 183(3.8) | 131(3.8) | 92 (3.5)* | 133 (3.7) | 76 (3.5)* | 993 (4.0) |
| Total | 239(25.5) | 45(4.5) | 100(13.1) | 64(7.9) | 183(19.5) | 131(16.1) | 92(8.2) | 133(12.8) | 76(6.6) | 100 |
| National Population N | 122294 | 21434 | 65707 | 39528 | 92840 | 76724 | 39724 | 61491 | 32218 | 100 |
See table 1 for complete description of each variable; note all percentages are weighted to account for the study’s sampling plan and raw numbers will not always correspond with percentages, given the weighted nature of the data.
Significant at p <.05.
Significant at p <0.01.
Significant at p <0.001; indicated above rows if categorical variable or next to percentile if continuous or dichotomous variable.
Prevalence and Patterns of Risk-Taking Behaviors
Almost half of the sample indicated at least one health-risk behavior (46.3%). At Wave 1, 25.5% of youth had engaged in sexual intercourse (see Table 2) and 4.5% reported having become pregnant or gotten someone pregnant. Thirteen percent of youth met criteria for depression and 7.9% reported “suicidality.” About one fifth (19.5%) reported smoking, 16.1% reported drinking alcohol, and 8.2% reported smoking marijuana with in the last 30 days. About 13.0% of youth reported getting into fights and 6.6% reported carrying a weapon within the last six months.
On bivariate analyses, the nine health-risk behaviors were predicted by different risk and protective factors, with a few variables fairly consistently associated with elevated rates of multiple health-risk behaviors, such as age, having deviant peers, and lower degrees of caregiver monitoring (see Table 2).
No dominant pattern of health-risk behaviors was identified (see Table 3); the most common pattern involved sexual risk behaviors (8.7%) followed by substance use (7.9%).
Table 3.
Patterns of Health-Risk Behaviors (N=993)
| Risk-Taking Behaviors Endorsed | Weighted Percent and Frequency in Sample in Descending Order N (%)a |
|---|---|
| No risk taking behaviors | 527 (53.7) |
| Sexual behavior only | 78 (8.7) |
| Substance use only | 69 (7.9) |
| Depression only | 55 (5.6) |
| Substance use, sexual, and delinquency | 43 (4.3) |
| Delinquency only | 55 (3.2) |
| Substance use, sexual, and depression | 17 (3.1) |
| Substance use and sexual | 49 (2.6) |
| All risk taking behaviors | 15 (2.2) |
| Substance use and depression | 13 (1.9) |
| Substance use and delinquency | 22 (1.7) |
| Sexual behavior and depression | 12 (1.6) |
| All other risk taking behaviors (weighted percents less than 1.5) | 38 (3.7) |
Note all percentages are weighted to account for the study’s sampling plan and raw numbers will not always correspond with percentages, given the weighted nature of the data.
Multivariate Findings
Poisson regression results indicated that youth ages 12–14 years demonstrated a risk rate 0.57 times that of older youth (95% CI 0.39, 0.84) (see Table 4). Interestingly, males demonstrated a risk rate 0.63 times that of females (95% CI 0.47, 0.86). Youth with histories of physical abuse had a risk rate 1.33 times greater than youth without a history of physical abuse (95% CI 1.03, 1.72). Those who reported “hanging out” with deviant peers had a risk rate 1.67 times greater than those who did not (95% CI 1.19, 2.36). Increased caregiver monitoring lowered the rate by 0.76 times (95% CI 0.65, 0.88); similarly, a higher level of school engagement reduced the rate by 0.56 times (95% CI 0.43, 0.72).
Table 4.
Multivariate Poisson Regression of Categorical and Continuous Variables on Number of Health-Risk Behaviors Among Early Adolescents in Child Welfare(N=993)a
| Variables | Rate Estimate | 95% Confidence Interval |
|---|---|---|
| Categorical Variables | ||
| Child Age*** | ||
| 11*** | 0.25 | 0.13, 0.48 |
| 12–14** | 0.57 | 0.39, 0.84 |
| 15b | -- | -- |
| Gender** | ||
| Male | 0.63 | 0.47, 0.86 |
| Femaleb | -- | -- |
| Race/Ethnicity | ||
| Black/Non-Hispanic | 1.04 | 0.74, 1.47 |
| Hispanic | 1.08 | 0.67, 1.73 |
| Other | 1.03 | 0.62, 1.71 |
| Caucasianb | -- | -- |
| Maltreatment History (Yes/No) | ||
| Neglect | 0.97 | 0.69, 1.37 |
| Emotional Abuse | 0.77 | 0.52, 1.12 |
| Physical Abuse * | 1.33 | 1.03, 1.72 |
| Sexual Abuse | 1.37 | 0.95, 1.98 |
| Child Placement | ||
| In-Home | 0.73 | 0.47, 1.13 |
| Out-of-Homeb | -- | -- |
| Youth Behavioral Functioning | ||
| Severe | 0.93 | 0.67, 1.28 |
| Moderate/Normab l | -- | -- |
| Youth Cognitive Functioning | ||
| Severeb | 0.96 | 0.69, 1.33 |
| Moderate/Normal | -- | -- |
| Caregiver Education | ||
| < High School | 1.22 | 0.83, 1.80 |
| HS Diploma/Equivalent | 0.83 | 0.55, 1.24 |
| > High Schoolb | -- | -- |
| Deviant Peers ** (Yes/No) | 1.67 | 1.19, 2.36 |
| Continuous Variables | ||
| Caregiver Risk Assessment | 0.91 | 0.46, 1.83 |
| Caregiver Monitoring*** | 0.76 | 0.65, 0.88 |
| School Engagement *** | 0.56 | 0.43, 0.72 |
| Religiosity | 0.90 | 0.78, 1.04 |
| Future Expectations | 0.82 | 0.67, 1.00 |
Significant at p <.05.
Significant at p <0.01.
Significant at p <0.001; indicated above rows when categorical variable with 3 or more values examined as a group or next to variable name if comparison between categories, dichotomous variable, or continuous variable.
Indicates referent group.
DISCUSSION
This research contributes to the existing literature by establishing baseline rates of health-risk behaviors and associated risk and protective factors in the first national probability sample of early adolescent youth involved with child welfare. Findings should be interpreted within the context of the study’s reliance on youth self-report. Specifically, validity and reliability of data obtained from youth involved in child welfare about their engagement in health-risk behaviors cannot be assured. Fear of loss of privacy, concerns about repercussions, and issues of social desirability may undermine accurate reporting on sensitive personal behaviors in youth in the general population. 25, 26 Such concerns may be amplified for youth in child welfare who may have experienced first-hand that the revelation of personal events can have significant consequences, namely intervention by a public institution in the form of continued monitoring and supervision, and in some cases removal from their biological family. These methodological concerns warrant further research within this population.
In addition, public health is concerned with identifying populations with heightened risk for poor outcomes. Yet, comparison of our findings with rates reported for large samples of youths in the general U.S. population is complicated by several factors, including differences in age, measures used, and populations from which samples are drawn (usually youth identified through schools). To provide a benchmark for the data presented here, we researched available national studies examining risk behaviors in youth in the late 1990’s including the Youth Risk Behavior Survey, the Monitoring the Future Survey, and the National Longitudinal Study of Adolescent Health (ADD Health). 7, 8, 22 Measures used in the NSCAW study most closely paralleled those in the ADD Health Study. Below we provide data from the ADD Health study as a way of contextualizing our data rather than as a direct comparison.
Overall, our findings indicate that a little under half of the sample engaged in one or more health-risk behaviors. This finding converges with prior work, which found that not all maltreated youth engage in such behaviors. 27 However, rates of health-risk behaviors for this sample appear to be elevated in most areas when contextualized against the ADD Health study. Specifically, in ADD Health, 17% of youth in grades 7 and 8 (often ages 12 through 14) reported having sexual intercourse; pregnancy history was only calculated for sexually active females at least 15 years of age. In our sample, 28.5% of 12–14 years olds had had intercourse, and approximately 4.3% in this group and 18.7% of 15 year olds reported becoming pregnant or getting someone pregnant. With regard to depression, ADD Health reported rates of emotional distress at 17.7% for the previous week and/or year and more than one suicide ideation or attempt in 3.7% of middle school youth. We found clinically significant rates of depression for 15.6% of youth with suicidal ideation or harm among 7.4% of youth aged 12–14. The rate of violence perpetration in ADD Health middle schoolers was 9.2%. By comparison, 13.4% of NSCAW youths, ages 12–14, reported getting into fights, and 7.0% of youths indicated having carried a weapon. It is more difficult to compare rates of tobacco and substance use, given differences in measures across the two studies. In ADD Health, 3.2% of middle schoolers had smoked more than 6 cigarettes per day, 7.3% had drunk alcohol or beer more than two days in the preceding month, and 6.9% had smoked marijuana at least 1 time in the past month. In our sample, 23.0% in the 12–14 age group smoked cigarettes at least once during the past month; 16.6% reported drinking at least one day during the past month, and 8.7% smoked marijuana at least one time during the last thirty days.
Implications
Similar to findings in Grogan-Kaylor and colleagues’ 19 analyses of risk and protective factors specific to delinquency in NSCAW, our more comprehensive examination reinforces the role of preventive services when youth first enter child welfare. Our findings also highlight the need for targeted prevention and interventions efforts for youth interacting with the child welfare system and for those youth identified with active health-risk behaviors.
In addition, our results suggest that all youth in child welfare should be targeted for screening regarding health risk behaviors, no matter what their placement status. Our findings caution that assumptions about a youth’s engagement in health-risk behaviors cannot be made based on their placement status alone. The lack of statistically significant differences between the youth remaining in-home and those in foster care is somewhat surprising. However, recent work has shown that foster care placement has less impact on measures of child well-being than previously suggested, when analytic methods appropriately account for selection bias. 28 At a minimum, all youth ages 11 and up undergoing an investigation for abuse and neglect should be assessed for health-risk behaviors using a standardized assessment approach such as the Guidelines for Adolescent Preventive Services (GAPS).
Factors related to higher rates of health-risk behaviors paralleled those in the general child population, including older age, caregiver education less than high school, deviant peers, limited caregiver monitoring, and poor school engagement. 29–33 Much of the literature on risk behaviors in the general population has focused on males; however, girls are at high risk for depression post-puberty and are beginning to surpass boys with respect to substance use. 34, 35 Wall and Barth’s analyses of delinquent behavior among subjects 11 to 15 years in the NSCAW sample also suggested that female gender was predictive of aggression and delinquency on multivariate modeling. 20 The unique vulnerabilities of girls in child welfare deserve further exploration. The central role of abuse in predicting health-risk behaviors has been previously documented in retrospective and prospective clinical populations. 36 but also address the vulnerability of this population.
Our findings underscore both the need to reinforce protective factors and diminish risk factors. The social development model, which integrates knowledge about the effect of empirical predictors, or risk factors, and protective factors in the development of health-risk behavior 18 suggests the multi-level contributions to health risk behaviors (e.g. individual, family, school, peer group and community). These youth, in particular, need concentrated attention at all of these levels to encourage health-promoting behaviors. It has also been suggested that prevention efforts may need to target multiple behaviors concurrently, given the strong correlation among different risk behaviors and their predictors. 37 Even in this young sample where engagement in health-risk behaviors is emerging, more than one-fifth of youth reported engaging in multiple types of health-risk behaviors.
Last, given overall dismal outcomes of youth who emancipate from foster care, much more needs to be learned about specific factors that put subgroups of youth at risk. Our sample is younger than samples used in previous foster youth health-risk studies. 14, 15, 38 A recent study using NSCAW data, investigating sexual risk behaviors at Wave 4 (36-month) when youth were 3 years older, found much higher rates of pregnancy and engagement in sexual intercourse compared to rates reported at Wave 1. 39 While this study identified independent variables associated with risky behaviors, we were unable to examine causality because of its cross-sectional design. Future studies should investigate developmental trajectories of engagement in health-risk behaviors and to determine whether other developmental or time-dependent factors beyond age elevate risk for subgroups of youth, such as youth who remain in foster care for extended periods of time or experience frequent placement changes.
Summary
While a recent report from the AAP Committee on Child Abuse and Neglect and Section on Adoption and Foster Care concurs that youth with histories of maltreatment are at high risk for health-risk behaviors, current policy statements by professional and advocacy organizations regarding comprehensive evaluations for youth in foster care do not specifically highlight the assessment of risk behaviors as part of this evaluation. 40 This study affirms the critical role of screening, prevention, and treatment of health-risk behaviors in these youth, as well as youth who undergo investigation but remain in their home of origin or are placed informally with kin.
Acknowledgments
Acknowledgements/Disclaimers/Disclosures: This study was supported by the Charles H. Hood Foundation (Dr. Leslie, Ms. Kauten); the Administration on Children, Youth, and Families, U.S. Department of Health and Human Services (ACYF/DHHS; Grant No. 90PH0006, Dr. Leslie, Dr. James, Ms. Zhang, Ms. Monn) and NIMH Grant No. R01MH072961 (Dr. Aarons). Drs. Leslie, James, and Aarons, and Ms. Monn and Ms. Kauten all contributed to the conceptualization, analyses, and editing of this document; Ms. Zhang conducted the analyses and edited the manuscript. The National Survey of Child and Adolescent Well-Being (NSCAW) was developed under contract to RTI from the ACYF/DHHS. The information and opinions expressed herein reflect solely the position of the author(s). Nothing herein should be construed to indicate the support or endorsement of its content by ACYF/DHHS.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.United States 110th Congress: Fostering Connections to success and Increasing Adoptions Act of 2008 H.R.6893/P.L. 110–351), 2008.
- 2.American Academy of Pediatrics, Committee on Early Childhood and Adoption and Dependent Care. Developmental issues for young children in foster care. Pediatrics. 2000;106(5):1145–50. [PubMed] [Google Scholar]
- 3.US Department of Health and Human Services, Administration for Children, Youth and Families, Children’s Bureau. [Accessed December 21, 2009];The AFCARS Report 2008: Preliminary estimates. 2009 January; Available at http://www.acf.hhs.gov/programs/cb/stats_research/afcars/tar/report16.htm.
- 4.US Department of Health and Human Services. [Accessed December 21, 2009];Healthy people 2010. doi: 10.3109/15360288.2015.1037530. Available at: http://www.healthypeople.gov/Publications. [DOI] [PubMed]
- 5.Grunbaum JA, Kann L, Kinchen SA, et al. Youth risk behavior surveillance--United States, 2001. J School Health. 2002;72(8):313–28. doi: 10.1111/j.1746-1561.2002.tb07917.x. [DOI] [PubMed] [Google Scholar]
- 6.Bryan A, Stallings MC. A case control study of adolescent risky sexual behavior and its relationship to personality dimensions, conduct disorder, and substance use. J Youth Adolesc. 2002;31(5):387–96. [Google Scholar]
- 7.Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm. Findings from the National Longitudinal Study on Adolescent Health. JAMA. 1997;278(10):823–32. doi: 10.1001/jama.278.10.823. [DOI] [PubMed] [Google Scholar]
- 8.Grunbaum JA, Kann L, Kinchen S, et al. Youth risk behavior surveillance--United States, 2003. MMWR Surveill Summ. 2004;53(2):1–96. [PubMed] [Google Scholar]
- 9.Sells CW, Blum RW. Morbidity and mortality among US adolescents: An overview of data and trends. Am J Public Health. 1996;86(4):513–9. doi: 10.2105/ajph.86.4.513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4):245–58. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
- 11.Barone C, Weissberg RP, Kasprow WJ, Voyce CK, Arthur MW, Shriver TP. Involvement in multiple problem behaviors of young urban adolescents. J Prim Prev. 1995;15(3):261–283. doi: 10.1007/BF02197475. [DOI] [PubMed] [Google Scholar]
- 12.Landsverk JA, Garland AF, Leslie LK. Mental health services for children reported to Child Protective Services. In: Myers JEB, Berliner L, Briere J, Hendrix CT, Jenny C, Reid TA, editors. APSAC Handbook on Child Maltreatment. Thousand Oaks (California): Sage Publications; 2002. pp. 487–507. [Google Scholar]
- 13.Carpenter SC, Clyman RB, Davidson AJ, Steiner JF. The association of foster care or kinship care with adolescent sexual behavior and first pregnancy. Pediatrics. 2001;108(3):E46. doi: 10.1542/peds.108.3.e46. [DOI] [PubMed] [Google Scholar]
- 14.Taussig HN. Risk behaviors in maltreated youth placed in foster care: A longitudinal study of protective and vulnerability factors. Child Abuse Negl. 2002;26(11):1179–99. doi: 10.1016/s0145-2134(02)00391-5. [DOI] [PubMed] [Google Scholar]
- 15.Courtney ME, Dworsky A. Early outcomes for young adults transitioning from out-of-home care in the USA. Child Fam Soc Work. 2006;11(3):209–19. [Google Scholar]
- 16.American Academy of Pediatrics. [Accessed December 21, 2009];AAP strategic plan guides activities on behalf of children, members. 2007 Available at: http://aapnews.aappublications.org/cgi/content/full/28/12/36.
- 17.American Academy of Child and Adolescent Psychiatry & Child Welfare League of America. [Accessed December 21, 2009];AACAP/CWLA Foster Care Mental Health Values Subcommittee-Policy Statement. 2007 Available at: http://www.aacap.org/publications/policy/collab01.htm.
- 18.Hawkins JD, Weiss JG. The social development model: An integrated approach to delinquency prevention. J Prim Prev. 1985;6:73–95. doi: 10.1007/BF01325432. [DOI] [PubMed] [Google Scholar]
- 19.Grogan-Kaylor A, Ruffolo MC, Ortega RM, Clarke J. Behaviors of youth involved in the child welfare system. Child Abuse Negl. 2008;32(1):35–49. doi: 10.1016/j.chiabu.2007.09.004. [DOI] [PubMed] [Google Scholar]
- 20.Wall AE, Barth RP. Aggressive and delinquent behavior of maltreated adolescents: Risk factors and gender differences. Stress, Trauma, and Crisis. 2005;8(1):1–24. [Google Scholar]
- 21.Orton HD, Riggs PD, Libby AM. Prevalence and characteristics of depression and substance use in a US child welfare sample. Child Youth Serv Rev. 2009;31(6):649–53. [Google Scholar]
- 22.Johnston LD, O’Malley PM, Bachman JG. The Monitoring the Future National Survey Results on Adolescent Drug Use: Overview of Key Findings. Bethesda, MD: National Institute on Drug Abuse; 2003. (NIH Publication No. 03–5374) [Google Scholar]
- 23.Cameron AC, Trivedi PK. Regression analysis of count data. Cambridge University Press; 1998. [Google Scholar]
- 24.SUDAAN Example Manual, Release 10.0. Research Triangle Park, NC: Research Triangle Institute; 2007. [Google Scholar]
- 25.Turner CF, Ku L, Rogers SM, Lindberg LD, Pleck JH. Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science. 1998;280:867–873. doi: 10.1126/science.280.5365.867. [DOI] [PubMed] [Google Scholar]
- 26.Tourangeau R, Smith TW. Asking sensitive questions: The impact of data collection mode, question format, and question context. Public Opin Quart. 1996;60(2):275–304. [Google Scholar]
- 27.Widom CS, Ames MA. Criminal consequences of childhood sexual victimization. Child Abuse Negl. 1994;18(4):303–18. doi: 10.1016/0145-2134(94)90033-7. [DOI] [PubMed] [Google Scholar]
- 28.Berger LM, Bruch SK, Johnson EI, James S, Rubin D. Estimating the “impact” of out-of-home placement on child well-being: Approaching the problem of selection bias. Child Dev. doi: 10.1111/j.1467-8624.2009.01372.x. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Rai AA, Stanton B, Wu Y, et al. Relative influences of perceived parental monitoring and perceived peer involvement on adolescent risk behaviors: An analysis of six cross-sectional data sets. J Adolesc Health. 2003;33(2):108–18. doi: 10.1016/s1054-139x(03)00179-4. [DOI] [PubMed] [Google Scholar]
- 30.Hawkins JD, Catalano RF, Miller JY. Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychol Bull. 1992;112(1):64–105. doi: 10.1037/0033-2909.112.1.64. [DOI] [PubMed] [Google Scholar]
- 31.Repetti RL, Taylor SE, Seeman TE. Risky families: Family social environments and the mental and physical health of offspring. Psychol Bull. 2002;128(2):330–66. [PubMed] [Google Scholar]
- 32.Oman RF, McLeroy KR, Vesely S, Aspy CB, Smith DW, Penn DA. An adolescent age group approach to examining youth risk behaviors. Am J Health Promot. 2002;16(3):167–76. doi: 10.4278/0890-1171-16.3.167. [DOI] [PubMed] [Google Scholar]
- 33.DiClemente RJ, Wingood GM, Crosby R, et al. Parental monitoring: Association with adolescents’ risk behaviors. Pediatrics. 2001;107(6):1363–8. doi: 10.1542/peds.107.6.1363. [DOI] [PubMed] [Google Scholar]
- 34.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. National results on adolescent drug use: Overview of key findings 2005. Bethesda, MD: 2006. Monitoring the future. NIH Publication No. 06–5882. [Google Scholar]
- 35.National Center on Addiction and Substance Abuse. The formative years: Pathways to substance abuse among girls and young women ages 8–22. New York: National Center on Addiction and Substance Use; 2003. [Google Scholar]
- 36.Auslander WF, McMillen JC, Elze D, Thompson R, Jonson-Reid M, Stiffman A. Mental health problems and sexual abuse among adolescents in foster care: Relationship to HIV risk behaviors and intentions. AIDS and Behavior. 2002;6(4):351–9. [Google Scholar]
- 37.Flay BR, Graumlich S, Segawa E, Burns JL, Holliday MY. Effects of 2 prevention programs on high-risk behaviors among African American youth: A randomized trial. Arch Pediatr Adolesc Med. 2004;158(4):377–84. doi: 10.1001/archpedi.158.4.377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Risley-Curtiss C. Sexual activity and contraceptive use among children entering out-of-home care. Child Welfare. 1997;76(4):475–99. [PubMed] [Google Scholar]
- 39.James S, Montgomery S, Leslie LK, Zhang JJ. Sexual risk behaviors among youth in the child welfare system. Child Youth Serv Rev. doi: 10.1016/j.childyouth.2009.04.014. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.American Academy of Pediatrics; Stirling J, Jr, et al. Committee on Child Abuse and Neglect and Section on Adoption and Foster Care. Understanding the behavioral and emotional consequences of child abuse. Pediatrics. 2008;122(3):667–73. doi: 10.1542/peds.2008-1885. [DOI] [PubMed] [Google Scholar]