Repository logo

The impact of institutionalisation and deinstitutionalisation on children’s development – A systematic and integrative review of evidence from across the globe

Accepted version

No Thumbnail Available



Change log


Duschinsky, Robert 
Reijman, Sophie 


Background: Millions of children worldwide are brought up in institutional care settings rather than families. These institutions vary greatly both in terms of their organizational principles and structure and in the quality of care they provide. Although universally recognized as providing sub-optimal caregiving environments there is less consensus on how to interpret the evidence relating to the size, range and persistence of their impact on children’s development and well-being. This has led to disagreement as to whether policy should focus on eliminating, transforming or improving institutions. Methods: We reviewed the literature on child institutionalisation and deinstitutionalisation from a global perspective. This included a survey of historical and cultural trends and current levels of institutional care, a systematic review and meta-analysis of developmental sequelae, and a largely qualitative review of factors found to predict individual variations in such outcomes. Findings: The numbers of children in institutional care have varied enormously over the years and from region-to-region - driven by a range of political, cultural and socio-economic factors. Currently, although precise and up-to-date figures are not available, millions of children across the globe are known to be housed in institutions. We found strong negative associations between institutional care and children’s development, especially in relation to developmental levels of physical growth, cognition and attention. Significant but smaller associations were found for socio-emotional development and mental health. Leaving institutions for foster/family care is associated with significant recovery for some (e.g., growth and cognition) outcomes but not for others (e.g., attention). The length of time in institutions was associated with increased risk of adverse sequelae and diminished chance of recovery. However, it was not possible to disentangle the association between developmental outcomes and the duration of institutional care as opposed to its timing, as would be required to establish the precise boundaries of sensitive periods. Interpretation: Every effort should be made to minimise children’s exposure to institutional care. There is an urgent need to reduce the numbers of children entering and to increase the numbers leaving institutions. Where institutional care is considered absolutely necessary, length of stays should be as short as possible, even if care is adequate. To this end, preventive approaches should be promoted, keeping children in birth families when possible. When not possible, family-based care alternatives should be supported, including extended kinship networks, adoption and stable high quality fostering. Policy recommendations to support the implementation of these care reform goals at the global, regional and local levels, will be set out in a separate policy document following on from this paper.



Adolescent, Attention, Child, Child Development, Child, Preschool, Cognition, Cost-Benefit Analysis, Deinstitutionalization, Developmental Disabilities, Female, Humans, Infant, Institutionalization, Male

Journal Title

The Lancet Psychiatry

Conference Name

Journal ISSN


Volume Title


Wellcome Trust (103343/Z/13/A)
Work supported by Wellcome Trust Investigator Award (Grant WT103343MA).