See related article on page 664
One of the most interesting and enduring questions in developmental psychopathology is the extent to which early experiences affect later development. The question also relates to the limits of children’s resilience, and their capacity to recover when good experiences and optimal parenting succeed very poor rearing conditions. Opportunities to study such outcomes are relatively rare and rely on data derived from so-called ‘natural experiments’.1 The paper by Sonuga-Barke and colleagues reports on the one of the most significant studies of this kind, the English Romanian Adoption (ERA) study.1 Rutter and colleagues recognized the theoretical and clinical significance of investigating the long-term outcomes of young infants adopted from Romanian institutions following the fall of the Ceauşcescu regime in 1989 into UK families. These adoptions were often for altruistic reasons and could be seen as humanitarian responses to the media coverage of the appalling conditions that existed in the institutions and the global deprivation these infants were exposed to. The current paper reports on findings up to 11 years.
Previous reports from the ERA and other studies have described both heterogeneity and specificity in outcomes for post-institutionalized adoptees.1,2 Development of children aged less than 6 months at adoption has been found to be comparable with that of domestic adoptees, but residual cognitive impairment, disinhibited attachment, inattention/overactivity, and quasi-autistic symptoms are evident in a substantial minority of children adopted after 6 months. As the current authors argue, the persistence of these effects, at least to 11 years, despite radical changes in the quality of care, suggests they are mediated by fundamental neurobiological changes.
The study has a number of significant strengths, but is not without weaknesses. These are acknowledged and, where possible, addressed by the authors. An unavoidable but significant shortcoming is the absence of any direct measures taken in the institutional settings: the extent to which weight at entry is a valid measure of nutritional intake, for example, is not known. Malnutrition was widespread and the subnutrition threshold adopted in study is low. However, the authors provide convincing arguments and evidence in support of the cut-off selected.
The novel contribution of this paper involves separating out, through group profiling rather than statistical control, the effects of two potential causal factors associated with brain growth and outcomes: subnutrition (indexed by 2 levels of weight at entry, with a cut-off of 1.5 SD below UK age-related norms) and psychosocial deprivation (indexed by 2 levels of duration of deprivation, with a cut-off of 6mo in institutions). The effects of the two factors on subsequent weight and head circumference at entry age, 6, and 11, and on the 4 domains of cognitive impairment and mental health at 11 are summarized in a highly informative table. Additional analyses consider the role of brain growth in mediating the effects of subnutrition and psychosocial deprivation on outcomes at 11.
Findings add to our understanding of the developmental trajectories associated with risk factors and highlight the significant and independent role of psychosocial deprivation in the neurodevelopmental outcomes, but many questions remain unanswered. The role of head circumference in mediating outcomes was found to be relatively limited and complex. The results suggested that brain size may partially account for the effects of duration of deprivation on lowered IQ, but in the sub-nourished group only, and to a lesser extent for the effects of duration of deprivation on inattention/overactivity and disinhibited attachment in the non-subnourished group. Similar dissociation of outcomes has been found in studies of early non-subnourished institutionalized children where profiles of inattention/overactivity and disinhibited social relationships were common, but there was little or no evidence of low cognitive abilities or quasi-autistic symptoms.3–5
The current study investigated the potential role of reduced brain size in previously reported associations between early subnutrition and psychosocial deprivation and later outcomes, but produced no simple answers about the underlying mechanisms. However, the profiles of subgroups are informative and the current set of analyses raises the possibility that different mechanisms may underpin the effects of different risk factors.