Cumulative Risks of Foster Care Placement for U.S. Children: Evidence from Birth Cohort and Synthetic Cohort Approaches

Background: Foster care placement is a far-reaching intervention that can have substantial long-term effects on children and families. Commonly cited lifetime prevalence estimates are based on synthetic cohort life tables, which rely on a stationarity assumption that age-specific rates are stable over time.
Objective: To estimate lifetime prevalence of foster care placement using a birth cohort life table approach, compare it to synthetic cohort estimates, and explore heterogeneity and dynamics in placement.
Participants and Setting: U.S. children born between 2000 and 2002.
Methods: We estimate lifetime prevalence for the 2000-2002 birth cohorts using a birth cohort life table approach. These cohorts are observed through age 18, allowing estimation of placement using uncensored childhood spells.
Results: Using the birth-cohort approach, an estimated 5.0% of children born between 2000 and 2002 experienced foster care placement by age 18. Risks varied substantially by ethnoracial group (Asian/Pacific Islander: 1.6%; Black: 10.0%; Native American: 12.0%; White: 3.9%). Synthetic-cohort estimates exceeded birth-cohort estimates by 16.7% (an absolute difference of 0.8–0.9 percentage points), with the largest differences at older ages and for placements attributed to child behavioral problems, non-kin family foster care, and congregate care. Substantial interstate heterogeneity was also observed.
Conclusions: Birth cohort estimates indicate that 5% of children in the U.S. born between 2000 and 2002 were placed in foster care over their childhoods; synthetic cohort estimates are somewhat higher but are qualitatively similar. Comparisons of the two methods shed light on the changes in the use of different types of foster care over time and across geographies.