The benefits and pitfalls of different forms of substitute care have rarely been evaluated in comparison with each other. In this study we compared outcomes in youth and young adulthood of long-term foster care and adoption for children who came into the Child Welfare system at a young age. We linked ten national registers with data covering ten national birth cohorts to compare cognitive, educational and self-support outcomes for 900 adoptees with 3100 who grew up in foster care. Outcomes for 900 000 majority population peers were assessed for descriptive purposes. Comparisons adoptees/foster children were done in linear regression models and in Cox regression models with fixed person time. We adjusted the analyses for birth parental related selection/confounding factors (mental health problems, substance abuse and maternal education), and age at placement in substitute care.

Crude outcomes for both groups were substantially weaker than for majority population peers. The foster children fell clearly short of adoptees on all outcomes; school performance at 15, cognitive competence at 18, educational achievement and self-support capability in young adult years, also after adjustments for birth parent related confounders and age at placement in substitute care.