Adult well-being of foster care alumni: Comparisons to other child welfare recipients and a non-child welfare sample in a high-risk, urban setting

https://doi.org/10.1016/j.childyouth.2012.11.016Get rights and content

Abstract

Research has shown that children placed in foster care fall below population norms on many indicators of well-being. Yet few studies have been designed to distill the effects of foster care from conditions that precede foster care. Based on the available evidence, it is also uncertain whether the purported effects of foster care are lasting. This study used data gathered prospectively from an economically disadvantaged, urban cohort to examine whether foster care is associated with decreased educational and economic attainments as well as increased criminality in adulthood. Individuals who were placed in foster care after an indicated allegation of maltreatment were compared to three naturally occurring groups: (a) maltreatment victims who did not reside in foster care, (b) individuals without an indicated maltreatment allegation who were raised in a household with a Child Protective Services (CPS) record, and (c) individuals without an individual or household record of CPS involvement. Using multiple estimation procedures, we found that all participants with a CPS record fared worse in adulthood than their peers without a CPS record. Despite their poor outcomes, foster children functioned as well as other CPS recipients who did not reside in foster care. Our findings indicate that caution is warranted when attributing dysfunction observed in foster children to the effects of foster care. Implications for prevention and intervention within the context of child welfare are discussed.

Highlights

► Compared to a non-child welfare sample, foster children had poor adult outcomes. ► Recipients of in-home child welfare services also had poor adult outcomes. ► Likewise, non-maltreated children with a CPS history fared poorly in adulthood. ► The outcomes of foster care alumni did not differ from other child welfare groups. ► Results implicate risks prior to CPS involvement as predictors of long-term well-being.

Introduction

Beginning in the early 1980s the number of children living in out-of-home (i.e., foster) care in the U.S. climbed steadily for nearly two decades, reaching a peak level of 568,000 in 1999 (U.S. House of Representatives, 2000). Placements have since declined, but in 2010 alone more than 250,000 children were placed in foster care and over 400,000 children resided in foster care at year's end (U.S. Department of Health and Human Services, 2012). Although most children who receive child protective services are not removed from their homes, the $25 billion spent on foster care annually accounts for the majority of dedicated federal child welfare spending (DeVooght, Allen, & Geen, 2008). Considering its public costs and implications for the well-being of vulnerable children and families, the need to scrutinize foster care is evident.

With the passage of the 1997 Adoption and Safe Families Act (Public Law 105–89), child well-being became a statutory goal of foster care. Empirical evidence suggests that this objective remains largely unfulfilled, inasmuch as foster children fare poorly compared to their peers in the general population on most metrics of well-being (McDonald et al., 1996, Pinderhughes et al., 2007). It remains unclear, however, what effects foster care has on children, net of other related factors such as child abuse and neglect (i.e., CAN). Therefore, comparisons of foster children to non-maltreated children, even those exposed to high levels of ecological risk, may reveal more about the effects of CAN and its attendant risks than about the effects of foster care per se.

Below, we present hypothesized mechanisms through which foster care may influence the well-being of children for better and worse. We then review empirical research that has examined the effects of foster care across three outcome domains: educational attainment, economic welfare, and crime. Finally, we discuss methodological challenges to assessing the effects of foster care and describe how data from the Chicago Longitudinal Study will be used to address these challenges and contribute to the literature.

Placing children in foster care may promote their well-being in multiple ways, three of which seem most plausible. First, children who face an imminent risk of harm may benefit from an alternative residential setting where they are unlikely to experience abuse or neglect.1 Second, while providing a haven, substitute care providers are expected to provide a secure, nurturing environment that should enrich a child's development over time. Third, children may receive therapeutic services that support their adjustment to foster care and reduce social, emotional, and cognitive deficits that may have emerged prior to placement. Following out-of-home placement, services may also be initiated that catalyze change in the child's family of origin and hasten reunification.

Conversely, there are no less than three ways foster care might be harmful or at least fall short of its mandate to promote child well-being. First, being removed from home may disrupt family connections and attachments to caregivers (Stovall-McClough & Dozier, 2004). This is an acute concern for children in foster care, many of whom have previously experienced harsh, inattentive, or inconsistent caregiving. Second, a child's capacity to adapt to family separation may be undermined by placement disruptions in foster care, which occur frequently (Fong, Schwab, & Armour, 2006). Third, evidence-based interventions are implemented infrequently in child welfare settings (Chaffin and Friedrich, 2004, Horwitz et al., 2010), and the services available may be delivered inconsistently and ineffectively due to conditions that often afflict child welfare systems. For instance, worker retention is an ongoing challenge for many agencies (DePanfilis and Zlotnik, 2008, Mor Barak et al., 2001), as is maintaining a sufficient corps of foster care providers capable of meeting service demands (Baum, Crase, & Crase, 2001).

Although children in foster care often fall short of developmental and functional norms, they are also exposed to higher-than-average levels of ecological risk. Thus, comparing foster children to the child population at large offers limited insight into the actual impact of foster care. In the subsequent review we turn to research that has attempted to quantify the effects of foster care, including studies that have compared foster children to matched samples of similar children who were not placed in foster care.

Academic performance is a widely adopted indicator of well-being that forecasts psychological, social, and economic outcomes over the life course. An early synthesis of research by McDonald et al. (1996) revealed that foster care alumni are at risk of poor school performance and low educational attainment, findings that have been replicated by contemporary research (Courtney et al., 2010, Pecora et al., 2006). These results have prompted speculation that children's educational progress may be hindered by foster care, due partly to a lack of emphasis on academic performance and a failure to mitigate the disruptive influence of school transfers (Jackson and McParlin, 2006, Zetlin et al., 2004).

Such conclusions may be premature, however, because most research has compared foster children to the general population or to poorly matched samples. One exception, a seminal study by Runyan and Gould (1985), found that children in out-of-home care had higher rates of school attendance and lower rates of school failure compared to maltreated children who received in-home services. Using multivariate analysis to examine data from the National Survey of Child and Adolescent Well-being (NSCAW), Stahmer et al. (2009) also discovered that most children receiving child welfare services made cognitive gains over time, and rates of cognitive change were unrelated to placement status. In a separate analysis of NSCAW data, Berger, Bruch, Johnson, James, and Rubin (2009) used multiple estimation procedures, including propensity score matching, and found that out-of-home placement did not significantly alter subsequent cognitive performance. Berzin (2010) also used propensity score matching to assess the long-term outcomes of foster children using the National Longitudinal Survey of Youth National Longitudinal Survey of Youth 1997. Results showed that rates of educational attainment among foster care alumni and non-foster youth did not differ once the groups were appropriately matched. Further, a study by Smithgall, Jarpe-Ratner, and Walker (2010) revealed that most foster children who struggle academically follow a developmental trajectory that is established prior to placement in foster care.

Given the strong association between educational and economic attainments, it is expected that foster care will effectuate similar outcomes in both domains. Supporting this hypothesis, McDonald et al. (1996) concluded that foster care alumni were more likely than individuals who were never placed in foster care to receive public assistance. Current research has also shown that, compared to population averages, foster youth have poorer labor market outcomes, including higher rates of unemployment and poverty and lower rates of health insurance (Courtney et al., 2010, Dworsky, 2005, Macomber et al., 2008, Pecora et al., 2006).

Few studies have compared the economic attainments of foster children to children with a child welfare history who were not placed in foster care. One investigation of data from California, Illinois, and South Carolina found that youth who aged out of foster care earned less than other underprivileged youth or than foster youth who were reunified with their parents (Goerge et al., 2002). Another study by Doyle (2007) found no overall differences in employment or earnings between maltreated children who were placed in foster care and maltreated children who were never placed in care.2 Finally, Berzin (2010) discovered that rates of public assistance were higher among foster care alumni than among other NLSY sample members with similar demographic profiles. Like her findings for educational attainment, however, Berzin found no significant discrepancies between groups once propensity score matching was employed.

Foster children often exhibit mental health and behavioral problems, which are associated with later antisocial behavior and crime (e.g., Clausen et al., 1998, Halfon et al., 1992, Keil and Price, 2006). Some evidence suggests out-of-home care may exacerbate these problems (Maluccio and Fein, 1985, Minty, 1999), including a study by Lawrence, Carlson, and Egeland (2006) that showed young children manifested greater psychological and behavioral deficits upon exiting foster care than maltreated children who were never placed in foster care. Other well-designed studies have discovered that out-of-home placement is associated with long-term behavioral consequences, including juvenile delinquency (Doyle, 2007, Ryan and Testa, 2005) and adult crime (Courtney and Dworsky, 2006, English et al., 2002).

Yet a comparable body of research has shown that crime and its earlier behavioral correlates do not vary as a function of placement status. For example, three recent studies uncovered no proximal differences in behavior problems between children receiving in-home services and children placed in out-of-home care (Berger et al., 2009, Mennen et al., 2010, Stahmer et al., 2009). Other studies have also called into question whether out-of-home placement leads to distal consequences such as delinquency and crime (Berzin, 2010, DeGue and Spatz Widom, 2009, Runyan and Gould, 1985). Moreover, some evidence suggests that out-of-home placement actually reduces the risk of juvenile arrest or incarceration for racial and ethnic minority youth (Jonson-Reid & Barth, 2000; Lemmon, 2006, McMurtry and Lie, 1992).

Research to date has shown unequivocally that foster children fare poorly relative to their peers in the general population. Comparisons of foster children to well-matched samples of non-foster children have produced mixed results, however, due partially to the methodological inconsistencies and limitations of prior research (Berger et al., 2009, Pecora et al., 2006). Again, investigations that have controlled for the confounding influence of child maltreatment are in the minority. Many studies have also suffered from limited statistical power (Dworsky, 2005), and most research has focused on childhood outcomes of foster children or the proximate functioning youth who age out of foster care.

This investigation adds to the literature by addressing the above limitations. Using prospective panel data from the Chicago Longitudinal Study (CLS), we examine the effects of out-of-home placement on educational and economic attainments as well as crime—three salient manifestations of well-being. We use multiple group comparisons to estimate the effects of foster care. Specifically, we compare individuals with an indicated3 record of CAN who resided in foster care to other maltreated children who did not reside in foster care. We also compare the out-of-home and in-home service groups to: (a) children lacking an indicated CAN allegation despite being raised in a household with a Child Protective Services (CPS) record, and (b) children without an individual or household record of CPS involvement. We use statistical techniques, including multivariate analysis and propensity score matching, to maximize group equivalence, reduce unmeasured variance, and test the sensitivity of our estimates.

Section snippets

Sample and data

The CLS investigates a cohort of 1539 participants who attended full-day kindergarten programs in the Chicago Public Schools in 1985–1986. Using a quasi-experimental design, 989 children who attended Chicago Child–parent Center (CPC) preschools were matched to 550 children enrolled in public schools located in comparable neighborhoods without a CPC site. The CPCs offer educational, family-support, health and nutrition services through 3rd grade to children residing in high-poverty catchment

Descriptive statistics

Table 1 presents comparisons of the original CLS sample (N = 1539) to the composition of the samples analyzed in the present study. Participants were included in each effective sample if their CPS records and outcome data could be verified; recovery rates ranged from 79.6% to 90.1%. While the full CLS sample and respective study samples were largely comparable, there were some significant unadjusted differences. Participants in the educational attainment, public assistance, and income samples

Discussion

This study compared the adult well-being of three groups of individuals with distinct CPS histories to the adult well-being of a fourth group of individuals who, prior to reaching the age of majority, had no personal or household record of child welfare involvement. We also tested pairwise contrasts between the three CPS groups to distill the lasting effects of foster care and other forms of CPS involvement from endogenous factors such as childhood maltreatment.

Several design features

Acknowledgments

We would like to thank Arthur Reynolds for permitting our use of data from the Chicago Longitudinal Study. We also thank Dylan Robertson and Lonnie Berger for their contributions to this work.

References (75)

  • V. Keil et al.

    Externalizing behavior disorders in child welfare settings: Definition, prevalence, and implications for assessment and treatment

    Children and Youth Services Review

    (2006)
  • C.N. Lawrence et al.

    Multiple Response System: Evaluation of Policy Change in North Carolina's Child Welfare System

    Children and Youth Services Review

    (2011)
  • J.H. Littell et al.

    What works best for whom? A closer look at intensive family preservation services

    Children and Youth Services Review

    (2002)
  • H.L. MacMillan et al.

    Interventions to prevent child maltreatment and associated impairment

    The Lancet

    (2009)
  • A.N. Maluccio et al.

    Growing up in foster care

    Children and Youth Services Review

    (1985)
  • F.E. Mennen et al.

    Do maltreated children who remain at home function better than those who are placed?

    Children and Youth Services Review

    (2010)
  • J.P. Mersky et al.

    Maltreatment prevention through early childhood intervention: A confirmatory evaluation of the Chicago Child–parent Center preschool program

    Children and Youth Services Review

    (2011)
  • B. Needell et al.

    Black children and foster care placement in California

    Children and Youth Services Review

    (2003)
  • P. Pecora et al.

    Educational and employment outcomes of adults formerly place in foster care: Results from the Northwest Foster Care Alumni Study

    Children and Youth Services Review

    (2006)
  • J. Ryan et al.

    Matching family problems with specific family preservation services: A study of service effectiveness

    Children and Youth Services Review

    (2004)
  • J. Ryan et al.

    Child maltreatment and juvenile delinquency: Investigating the role of placement and placement instability

    Children and Youth Services Review

    (2005)
  • A.C. Stahmer et al.

    Associations between intensity of child welfare involvement and child development among young children in child welfare

    Child Abuse & Neglect

    (2009)
  • W. Arulampalam et al.

    Unemployment persistence

    Oxford Economic Papers

    (2000)
  • O. Ashenfelter et al.

    Income, schooling and ability: Evidence from a new sample of identical twins

    Quarterly Journal of Economics

    (1998)
  • R.P. Barth et al.

    Parent training in child welfare services: Planning for a more evidence based approach to serving biological parents

    Research on Social Work Practice

    (2005)
  • R.P. Barth et al.

    Placement into foster care and the interplay of urbanicity, child behavior problems, and poverty

    The American Journal of Orthopsychiatry

    (2006)
  • A.C. Baum et al.

    Influences on the decision to become or not become a foster parent

    Families in Society

    (2001)
  • L. Berger et al.

    Estimating the “impact” of out-of-home placement on child well-being: Approaching the problem of selection bias

    Child Development

    (2009)
  • S.C. Berzin

    Understanding foster youth outcomes: Is propensity scoring better than traditional methods?

    Research on Social Work Practice

    (2010)
  • Child Trends

    Foster care

  • J.M. Clausen et al.

    Mental health problems of children in foster care

    Journal of Child and Family Studies

    (1998)
  • M.E. Courtney et al.

    Early outcomes for young adults transitioning from out-of-home care in the U.S.A.

    Child and Family Social Work

    (2006)
  • M.E. Courtney et al.

    Midwest evaluation of the adult functioning of former foster youth: Outcomes at ages 23 and 24

    (2010, March)
  • S. DeGue et al.

    Does out-of-home placement mediate the relationship between child maltreatment and adult criminality?

    Child Maltreatment

    (2009)
  • K. DeVooght et al.

    Federal, state, and local spending to address child abuse and neglect in SFY 2006

    (2008, December)
  • J.J. Doyle

    Child protection and child outcomes: Measuring the effects of foster care

    American Economics Review

    (2007)
  • M. du Toit et al.

    Interactive LISREL: User's guide

    (2001)
  • Cited by (27)

    • Social networks of youth transitioning from foster care to adulthood

      2019, Children and Youth Services Review
      Citation Excerpt :

      For example, young people with foster care experience may not have the social support necessary during the transition to adulthood and often report the transition as abrupt and occurring earlier than they were ready (Goodkind, Schelbe, & Shook, 2011). Young people who leave foster care without social support often experience an increased risk for homelessness, low educational attainment, unemployment, and young parenthood (Berzin, Rhodes, & Curtis, 2011; Braciszewski & Stout, 2012; Courtney, Dworsky, Lee, & Raap, 2010; Jones, 2011; Mersky & Janczewski, 2013). However, there is tremendous variation in the experiences of such youth.

    • The factors that contribute educational outcomes of adolescents placed in care due to severe behavioral problems

      2019, Children and Youth Services Review
      Citation Excerpt :

      Several international studies confirm that children placed in OOHC, regardless of the type of out-of-home placement, have a risk of remaining less educated (e.g., Berlin et al., 2011; Cameron et al., 2012; Cameron et al., 2018; Vinnerljung et al., 2005). The results from North America (e.g., Courtney and Dworsky, 2006; Jones, 2010; Mersky and Janczewski, 2013; Warburton et al., 2014) are in line with the results from Great Britain (e.g., Biehal et al., 1994), France (Dumaret et al., 2011) and the Nordic countries (e.g., Clausen and Kristofersen, 2008; Österberg et al., 2016; Vinnerljung and Hjern, 2011; Vinnerljung and Sallnäs, 2008). For example, in a study from the United States, adults with a foster care background graduated from high school as often as adults without a history in placement.

    • The association between out-of-home placement and offending behavior among maltreated youth: A systematic review

      2018, Children and Youth Services Review
      Citation Excerpt :

      Data from the Midwest study, a longitudinal multisite study of youth aging out of out-of-home care, revealed that youth who have experienced out-of-home care are at greater risk of criminal involvement at age 21 (Courtney et al., 2007), ages 23–24 (Courtney et al., 2010), and age 26 (Courtney et al., 2011). Additionally, a study by Mersky and Janczewski (2013) showed that individuals with an indicated child abuse or neglect report who were placed in out-of-home care are at an increased risk for criminal offending as compared to the individuals with no Child Protection Services (CPS) record. Another study by Orsi, Lee, Winokur, and Pearson (2018) found that youth with a prior child welfare placement had higher odds of non-permanency (i.e., emancipation, involvement with the state's Division of Youth Corrections, or runaway) as compared to those without a prior placement.

    View all citing articles on Scopus
    View full text