Summary
Background
Early social deprivation can negatively affect domains of functioning. We examined psychopathology at age 12 years in a cohort of Romanian children who had been abandoned at birth and placed into institutional care, then assigned either to be placed in foster care or to care as usual.
Methods
We used follow-up data from the Bucharest Early Intervention Project (BEIP), a randomised controlled trial of abandoned children in all six institutions for young children in Bucharest, Romania. In the initial trial, 136 children, enrolled between ages 6–31 months, were randomly assigned to either care as usual or placement in foster care. In this study we followed up these children at age 12 years to assess psychiatric symptoms using the Diagnostic Interview Schedule for Children (4th edition; DISC-IV). We also recruited Romanian children who had never been placed in an institution from paediatric clinics and schools in Bucharest as a comparator group who had never been placed in an institution. The primary outcome measure was symptom counts assessed through DISC-IV scores for three domains of psychopathology: internalising symptoms, externalising symptoms, and attention-deficit hyperactivity disorder (ADHD). We compared mean DISC-IV scores between trial participants and comparators who had never been placed in an institution, and those assigned to care as usual or foster care. Analyses were done by modified intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00747396.
Findings
We followed up 110 children from the BEIP trial between Jan 27, 2011, and April 11, 2014, and 49 children as comparators who had never been placed in an institution. The 110 children who had ever been placed in an institution had higher symptom counts for internalising disorders (mean 0·93 [SD 1·68] vs 0·45 [0·84], difference 0·48 [95% CI 0·14–0·82]; p=0·0127), externalising disorders (2·31 [2·86]vs 0·65 [1·33], difference 1·66 [1·06–2·25]; p<0·0001), and ADHD (4·00 [5·01] vs 0·71 [1·85], difference 3·29 [95% CI 2·39–4·18]; p<0·0001) than did children who had never been placed in an institution. Compared with 55 children randomly assigned to receive care as usual, the 55 children in the foster-care group had fewer externalising symptoms (mean 2·89 [SD 3·00] for care as usual vs 1·73 [2·61] for foster care, difference 1·16 [95% CI 0·11 to 2·22]; p=0·0255), but symptom counts for internalising disorders (mean 1·00 [1·59] for care as usual vs 0·85 [1·78] for foster care, difference 0·15 [−0·35 to 0·65]; p=0·5681) and ADHD (mean 3·76 [4·61] for care as usual vs 4·24 [5·41] for foster care, difference −0·47 [–2·15 to 1·20; p=0·5790) did not differ. In further analyses, symptom scores substantially differed by stability of foster-care placement.
Interpretation
Early foster care slightly reduced the risk of psychopathology in children who had been living in institutions, but long-term stability of foster-care placements is an important predictor of psychopathology in early adolescence.
Funding
National Institute of Mental Health and the John D and Catherine T MacArthur Foundation.