The Quality of Foster Care

(Originally published January 2012)

One of the most difficult challenges for child welfare scholars is to fairly and accurately summarize existing research regarding effects of foster care on children’s developmental outcomes. I occasionally read confident assertions regarding foster care outcomes based on a single study conducted in one state or community; but because of the large body of research with mixed findings on this subject, there cannot be a single definitive study. Scholars must sort through multiple studies and exercise judgment regarding the weight to give to various studies and their findings and conclusions. In addition, foster care is an abstraction that refers to thousands of foster care systems in different communities around the world. It is plausible, and even likely, that different foster care systems provide different quality of care; foster children may receive excellent care in some communities, indifferent care in others and deplorable care in some communities. Foster care is not a homogenous entity.


Furthermore, there are a number of factors that have been found in one or more studies to influence foster care outcomes: length of stay in out-of-home care, number of unplanned placement changes, age of children at entry into care, behavior problems of children, relationship of caregivers to children in their care, characteristics of foster parents and quality of the home environment, among others.


In recent years, possibly the most important studies of foster care and foster care outcomes have been based on the National Study of Child and Adolescent Well Being (NSCAW), a comprehensive longitudinal study of 5501 children served by child welfare agencies in their parents’ homes or in out- of- home care; 1890 of these children had at least one placement during the study period. NSCAW contains a nationally representative sample of children served by child welfare agencies around the country; as such, findings of NSCAW studies conducted by a number of distinguished scholars must be taken seriously. 


David Rubin’s NSCAW study (2007) of placement stability and foster children’s behavioral problems includes some sobering findings:


  • 21% of children in out- of- home care were “unstable”, meaning that they had not been in a home or residential setting for 9 continuous months during the 18 month study period; 31% of children and youth 10 and older were classified as “unstable” after 18 months in out- of- home care.

  • 19% of children in out- of- home care experienced more than four placements.

  • Placement stability during the first 18 months of care was significantly related to all permanency outcomes, i.e., reunification, adoption, still in care.

  • Children with good Child Behavior Checklist (CBCL) scores at baseline were far more likely to achieve early stability or late stability in out- of- home care than children with poor initial CBCL scores.

  • Children with normal behavior at baseline who achieved early stability in care were more likely to continue to have positive CBCL scores at the 36 month follow- up than reunified children (85% vs.70%).

  • Most children with poor CBCL scores at baseline continued to have poor CBCL scores at the 36 month follow– up regardless of whether they were placed early or late in stable homes.

  • Still, 40% of children with poor CBCL scores at baseline had good CBCL scores at the 36 month follow up. 


In summary, placement stability had a large effect on permanency outcomes for all groups of children, but a limited effect on child development for the most troubled children. To paraphrase a perceptive English scholar, many of the children who needed the most help with developmental challenges benefited the least from out- of- home care even when they experienced stable placements, according to this study. Nevertheless, 40% of children in out- of- home care with behavior problems in the initial assessment had greatly improved behavioral functioning at the 36 month follow up, a considerable rate of developmental repair. 


In another study of NSCAW data, Stahmer, et al (2009) found that young children (12- 47 months of age) in out of- home care made modest improvements on developmental measures, “but these gains were not consistently related to any specific intensity of CW (i.e., child welfare) involvement. According to the authors, after controlling for children’s baseline developmental level, children’s age, maltreatment type, gender and the home environment, there were few differences in children’s developmental progress between children placed in out- of- home care and children remaining in the parents’ home with or without services. However, given that these researchers controlled for quality of the home environment in which children were living, and that children in out- of- home care were found to have more nurturing and stimulating home environments than children remaining in the parents’ home with services, this conclusion seems highly questionable. What if researchers controlled for the quality of medical care in a study of recovery from disease, and then concluded that doctors’ visits had little effect on rates of recovery? 


Possibly the most distressing finding in the study is that “children with lower BL (i.e., baseline) developmental / cognitive scores … placed in out- of- home care

were found to experience relatively less positive change than children with comparable BL scores who remained at home with no service.” Again, the children who had the most developmental challenges benefited the least from out- of- home care, just the opposite of what many practitioners and advocates assume.  


This study’s most important finding is that quality of the home environment, whether the parents’ home or foster care, was positively related to children’s language development and improvements in adaptive behavior. The authors conclude “that foster care does not have a negative effect on child functioning but dies not facilitate development either.” In my view, a more defensible interpretation of the study’s findings is that out- of- home care helps children’s development when they are placed in nurturing homes and provided enriched learning opportunities. Safe but non – nurturing care in a home with limited intellectual stimulation is not likely to do much, if anything, for young children’s development.


The most concerning NSCAW studies regarding the quality of foster care have been provided by Richard Barth, Rebecca Green, Mary Bruce Webb and a few other colleagues of these scholars (2007; 2008). These analyzes found that non- kin and kin foster parents in NSCAW were less well educated, poorer and older than parents in the general population. These scholars state that “Any general conceptualization of foster parents as being predominantly middle class and educated is clearly out of date …” More than one in five foster families in NSCAW had incomes below the poverty line.


These authors assert that “About twice as many caregivers of children older than 3 in the general population provide a highly stimulating environment than do kinship or foster care providers.” These authors are concerned with the high percentage of very young foster children living in large households. They comment, “When a young child with a variety of developmental vulnerabilities lives with three or more other children, each of whom may also have special problems, one is entitled to raise questions about the level of individual attention the child will receive.” This comment raises the issue of overcrowding in foster homes resulting from acute and chronic shortages of licensed homes in many communities. In the not so distant past in Washington State, some child welfare offices with severe shortages of foster homes were forced to move a number of children from home to home on a daily basis until a long term placement in a relative’s home or non – kin foster home could be arranged.


In the analysis of NSCAW data published in 2007, Barth, et al found that “About one- fifth of children (in out- of- home care) were rated as experiencing both low responsiveness and high punitiveness at both baseline and eighteen months.” Kin and non- kin caregivers did not differ in rates of what might be described as harsh non – nurturing care. However, in the analysis published in Child Welfare

in 2008, Barth, et al state that “Just under 10% of kin (9.6) and just over 6% (6.2) of non – kin are identified as high in punitiveness and low in responsiveness.” This is a large difference in two analyzes of the same HOME – SF measures that should be clarified by the authors given that the combination of low responsiveness and high punitiveness arguably represents an unacceptable quality of care. If one- fifth of foster parents are providing harsh non- nurturing care to children placed due to abuse and neglect, U.S. foster care systems should be regarded as in crisis; if 6% of foster homes are providing an inadequate quality of non - nurturing care, this is a management problem that requires a timely and thoughtful response, but it’s not a crisis.


Barth, et al, convincingly asserts that “A vision for excellence in foster care is needed.” They advocate for a careful examination of foster care reimbursement rates that leave a fifth of foster parents living in poverty, but acknowledge that states are reluctant to increase the already existing difference between foster care reimbursement rates and the amount low income birth parents receive from welfare to care for children. Nevertheless, to reimburse poor and lower middle class foster parents and unlicensed relative caregivers less than the cost of care, and fail to compensate caregivers for the costs of music lessons, soccer leagues, field trips, etc. is a de facto acceptance of a quality of care middle class parents would not accept for their own children.


Every foster parent needs good training on trauma and how to help traumatized children with issues of trust and emotional regulation. In addition, state and local child welfare agencies need to develop a sizeable cadre (10-15% of licensed homes) of professional foster parents trained and supported to care for children and youth with serious mental health problems. As matters currently stand, foster care has the most developmental benefits for developmentally normal children who lack behavior problems. Some emotionally troubled children are fortunate to be placed with nurturing, skilled and committed foster parents and make impressive developmental gains. However, there is a large group of children and youth with severe behavior problems and other mental health issues who move from home to home to residential / institutional care and are further damaged by their experiences in out- of- home care. Child welfare systems have a moral obligation to improve the quality of foster care and residential care for these children. However, this is more easily said than done.


Barth, Green, Webb, et al conclude that “Foster caregivers are, on the whole, providing marginally adequate environments for foster children. Yet, this is clearly not enough given the untoward educational and developmental outcomes that await foster children if they do not get adequate assistance.” They recommend linking every foster home to a resource center much like the Mockingbird Hub Home model used to a modest extent in Washington State. “Ensuring that all foster parents receive consistent, powerful, supportive in- home training would be another powerful component of that (positive) vision,” they state. Unlike some other research, their NSCAW analyzes have found that kin and non – kin caregivers are more alike than they are different; and they clearly do not believe that an increased rate of kinship care will greatly improve the quality of foster care unless child welfare agencies invest in a vision of foster care excellence. The underling goal of that vision should be that foster care becomes a therapeutic experience that achieves developmental repair for children who need it through skilled, stable and nurturing care.




Barth, Richard, Guo, Shenyang, Green, Rebecca, McCrae, Julie, “Kinship Care and Nonkinship Foster Care,” Chapter 11 in Child Protection: Using Research to Improve Policy and Practice, edited by Ron Haskins, Fred Wulczyn and Mary Bruce Webb, Brookings Institution Press, 2007.


Barth, Richard, Green, Rebecca, Webb, Mary Bruce, Wall, Ariana, Gibbons, Claire and Craig, Carlton, Child Welfare, Volume 87, #3, 2008.


Rubin, David, O’Reilly, Amanda, Hafner, Lauren, Luan, Xianqun and Localio, A. Russell, “Placement Stability and Early Behavioral Outcomes among Children in Out- of- Home Care,” Chapter 10 in Child Protection: Using Outcomes to Improve Policy and Practice, Brookings Institution Press, 2007.


Stahmer, Aubyn C., Hurlburt, Michael, McCue Horwitz, Sarah, Landsverk, John, Zhang, Jinjin, Leslie, Laurel K., “Associations between intensity of child welfare involvement and child development among young children in child welfare,” Child Abuse & Neglect

Volume 33, Issue 9, September 2009.


© 2020 by Dee Wilson Consulting. Proudly created with