A Survival Analysis of Length of Services and Implications for Retention of Foster Parents

Background and Purpose: In the child welfare system, foster homes play a vital role by providing a resource helping to support a child’s safety, well-being and permanency. As of FY 2015, there were 427, 910 children in foster care, 45% of these children were living in non-relative foster family homes (U.S. Department of Health and Human Services, 2016). Based on the growing demand for out of home placements, a major concern for child welfare agencies is how to recruit and maintain an adequate supply of foster homes. This has forced child welfare research to assess the length of time of fostering and its implications for recruitment, retention, and supports provided to foster parents. The main objective of this study is to examine factors associated with foster parent’s length of service.

Methods: The survey design implements two foster parent surveys, one for continuing and one for exiting foster parents. The state agency provided a random sample of continuing and exiting foster parents, and the university conducted surveys from November 2008 to April 2010 with 478 current (n=255) or exited foster parents (n=223). Three analyses model were used; 1) Bivariate analyses were conducted to examine characteristics of foster parents and significant differences between exited foster parents and continuing foster parents. 2) Kaplan-Meier survival curves investigated the relationship between the length of time foster parents served and each individual characteristic (e.g., age and gender), and tested the differences using log rank tests. 3) Then, multivariate survival analyses were conducted using Cox proportional hazards regression models to examine the impact of parents’ characteristics on duration of being foster parents.

Results: For those who were not married, the mean length of being a foster parent was 201 months, compared to 148 months for those who were married (χ²=7.20, p < 0.01). The mean survival time of being a foster parent for non-white and white respondents was 143 and 200 months, respectively. The difference was also statistically significant (χ²=3.72, p < 0.05). Those who were unemployed tended to foster for longer, with an average length of 184 months as compared to 142 months for those who were employed (χ²=4.05, p < 0.05). People with a college degree and above had a shorter length of being a foster parent (109 months) compared to those who had less than a college degree (189 months; p < 0.001). Results from multivariate survival analyses using Cox proportional hazards regression models showed that agency support had statistically significant impacts on continuing foster services without controlling for other factors.

Conclusions and Implications: Results of this study indicate that agency support had statistically significant impacts on foster parents retention. In order for child welfare agencies to continue to meet the growing needs of the children they serve by providing quality foster care homes, an emphasis must be made in enhancing agency recruitment and screening methods, improving and developing foster parent training, addressing the issues that impact retention, and providing support to foster parents.