The Enduring Intervention: Predictors of Placement into Congregate Care Among Older Children and Adolescents in Non-Relative Foster Care

Background:

Congregate care has persisted despite historical and ongoing efforts to eliminate it in favor of foster family care, though there is little research on predictors of placement into group settings. Since there may be fewer foster family placements available or willing to address the heightened risks facing older youth with intensive service needs, group care may play an important role in promoting desired outcomes in this population. Examining what factors contribute to the risk of group placement may enhance understanding of how this intervention could better meet the needs of those who use it.

Methods:

Using administrative child welfare data, the study population consists of all 10-17 year-old children in California who were in foster care during 2014, and whose last/most recent placement was in a foster family home or group home (N=14,483). Youth in kin/guardian placements were excluded from the analysis.  

The outcome variable is most recent placement type (congregate vs. foster family). Predictor variables (associated with the current stay in care) include: age at entry into the focal placement, gender, race/ethnicity, removal reason, length of stay, prior entries to care, number of total placements, prior placement types, and history of runaway episodes. Unadjusted and adjusted generalized linear models were specified to assess the likelihood of placement into congregate vs. foster family care.

Results:

Just over 34% of 10-17 year-olds in non-relative foster placements in California during 2014 were ultimately placed in congregate care. Bivariate analyses showed significant differences (p<.001) between youth placed in congregate vs. foster family care across demographic and placement characteristics. Compared to children in foster family care, those in congregate care were more likely to be older, male, Black, White, or Native American, and to have entered care due to caretaker absence/incapacity. They were also more likely to have runaway history, a previous group placement, and a higher number of total placements.

Adjusting for all covariates, the factors most strongly associated with increased risk for placement in congregate care were older age, male gender, Black or White race, runaway history, and placement instability. Placement instability was the strongest predictor of congregate care placement; youth with 9 or more placements were more than twice as likely to be in congregate care as youth with 1-2 placements (RR: 2.24; 95% CI: 2.03, 2.48).

Conclusions:

Congregate care persists despite opposition, with more than a third of youth in this study sample ultimately placed in a congregate setting. There are substantial differences between youth placed in congregate care compared to those placed in foster family care, with placement instability emerging as a key predictor of group placement. These findings suggest that, despite general negative regard for congregate care, these settings may play a crucial role in caring for a subset of foster youth who are older and have had undesirable placement experiences. Given the continued reliance on congregate placements in the dependency system, efforts to improve their responsiveness to the needs of the population who utilize them may be preferable to eradicating them from the system of care.