CoChat

The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index; 1 of the 11 sources its claims cite were checked for support.

Foster care leads to sustained cognitive gains following severe early deprivation

Open access
Humphreys, Kathryn L., King, Lucy S., Guyon-Harris, Katherine L., Sheridan, Margaret A., McLaughlin, Katie A., Radulescu, Anca, Nelson, Charles A., Fox, Nathan A., Zeanah, Charles H. · 2022 · Proceedings of the National Academy of Sciences

Abstract

This study examined longitudinal data from the Bucharest Early Intervention Project, a randomized controlled trial of foster care as an alternative to institutional care following exposure to severe psychosocial deprivation. We report data from 135 participants assessed in early adulthood (age 18 y). We find that 16 y after randomization occurred, those who had been randomized to high-quality foster care had significantly higher IQ scores (9 points, 0.6 SD) than those randomized to care as usual. Mediation analyses provide evidence that the causal effect of the intervention on cognitive ability in early adulthood could be explained, in part, by higher-quality caregiving and attachment security. These findings indicate that early investment in family care as an alternative to institutional care leads to sustained gains in cognitive ability. Fostering caregiving relationships is a likely mechanism of the intervention. In addition, exploratory analyses indicate that stable placements throughout childhood are associated with the greatest long-term gains in cognitive ability. Whether early interventions for infants and young children lead to lasting change has significant implications for decisions to invest in programs aimed at improving children’s developmental outcomes.

Overview

CoChat Summary

TL;DR

Using the Bucharest Early Intervention Project randomized trial, individuals randomized to foster care after early institutionalization had early-adulthood IQ scores 9.00 points above those assigned to care as usual. The effect persisted nearly 14 years after the intervention ended and was mediated by early caregiving quality and attachment security, though ever-institutionalized adults still scored far below never-institutionalized peers.

Key findings

  • In intent-to-treat analyses at age 18 y, the foster care group (FCG) had a mean FSIQ of 73.49 versus 64.65 in the care-as-usual group (CAUG): B = 9.00, β = 0.45, SE = 4.04, P = 0.026, R² = 0.05; n = 95 (49 randomized to foster care). Significant domain effects were found for verbal comprehension (B = 11.09, P = 0.003) and processing speed (B = 7.75, P = 0.024), while perceptual reasoning and working memory were not statistically significant (working memory P = 0.053).
  • Caregiving quality was higher in the FCG than the CAUG (R² = 0.10) and was associated with higher age-18 FSIQ (R² = 0.13). In mediation models, the indirect effect through caregiving quality was 4.44, and the intervention effect was no longer significant when caregiving quality was included; attachment security at age 42 mo also mediated the effect (indirect effect = 6.42). Caregiving quality explained 49% of the intervention effect on FSIQ, and attachment security explained 71%.
  • Exploratory analyses found a small negative association between age of placement into foster care and age-18 FSIQ (B = −0.81, β = −0.25, SE = 0.42, P = 0.050, R² = 0.06), with earlier placement tending to predict higher FSIQ.
  • Compared with never-institutionalized children (NIG), ever-institutionalized individuals (EIG) had much lower age-18 FSIQ: mean 69.21 versus 94.75, B = 26.21, β = 1.20, SE = 3.01, P < 0.001, R² = 0.30; EIG scores were lower across all four cognitive domains.

Methods

  • The study used data from the Bucharest Early Intervention Project, an RCT (NCT00747396) in which young children living in Romanian institutions were assessed at a mean age of 22 mo at baseline and randomized to high-quality foster care (FCG) or care as usual (CAUG). The current follow-up included 135 participants: 95 from the original trial (46 CAUG, 49 FCG) and 40 never-institutionalized community comparison children (NIG), assessed at a mean age of 18.74 y (SD = 0.65); 47% female/53% male.
  • Cognitive ability at age 18 y was assessed with the WISC-IV, yielding a full-scale IQ composite and four domain scores. Because of floor effects and nonnormality, analyses used Tobit models estimated with robust weighted least squares; five participants too cognitively impaired to complete the WISC-IV had scores imputed to 40 for tables and figures.
  • Caregiving quality was assessed at ages 30 and 42 mo using the Observational Record of the Caregiving Environment, coded from videotaped child–preferred-caregiver interactions and averaged into a composite of sensitivity, stimulation of development, positive regard, flat affect (reverse coded), and detachment (reverse coded). Attachment security was assessed at age 42 mo via a strange situation procedure.

Limitations

  • The paper states that the sample is small and cannot be increased for the ever-institutionalized group; small trials may limit the accuracy of statistical estimates. It also states that prenatal care and pre-/postnatal nutrition could not be examined, and that the NIG is a comparison but not a “control” group for those placed in institutions.
  • The paper states that it used the WISC-IV rather than an adult intelligence instrument, precluding adult age-based norms, and that the liberal substitution approach deviates from Wechsler recommendations; these choices were made to address very low scores and floor effects. It also notes that other environmental aspects beyond the caregiving relationship were not included.
  • The paper states that exploratory analyses of placement stability and age of placement are not intent-to-treat and cannot support causal conclusions; it also notes that the association between placement stability and IQ is challenging to interpret because children and adolescents with lower IQ may be at greater risk for placement instability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1073/pnas.2119318119

Cites

33 of 33 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1. FSIQ scores at age 18 y as a function of ITT group. Distributions of FSIQ scores at age 18 y within participants randomized to care as usual (CAUG) and participants randomized to the foster care intervention (FCG). Distributions are rotated kernel density plots trimmed to the range of the observed values.
Fig. 1. FSIQ scores at age 18 y as a function of ITT group. Distributions of FSIQ scores at age 18 y within participants randomized to care as usual (CAUG) and participants randomized to the foster care intervention (FCG). Distributions are rotated kernel density plots trimmed to the range of the observed values.CC BY-NC-ND · PubMed Central
Fig. 2. Quality of caregiving in early life as a potential mechanism explaining the effect of the foster care intervention on FSIQ scores at age 18 y. ITT group was associated with caregiving quality in early life (A), and caregiving quality was in turn associated with FSIQ scores at age 18 y (B). Distributions are rotated kernel density plots trimmed to the range of the observed values. There was a significant indirect effect for caregiving quality in early life in partially explaining the causal effect of the intervention on IQ scores at age 18 y (C).
Fig. 2. Quality of caregiving in early life as a potential mechanism explaining the effect of the foster care intervention on FSIQ scores at age 18 y. ITT group was associated with caregiving quality in early life (A), and caregiving quality was in turn associated with FSIQ scores at age 18 y (B). Distributions are rotated kernel density plots trimmed to the range of the observed values. There was a significant indirect effect for caregiving quality in early life in partially explaining the causal effect of the intervention on IQ scores at age 18 y (C).CC BY-NC-ND · PubMed Central
Fig. 3. FSIQ scores at age 18 y as a function of age of placement into foster care among those in the FCG.
Fig. 3. FSIQ scores at age 18 y as a function of age of placement into foster care among those in the FCG.CC BY-NC-ND · PubMed Central

Verification

Claims the paper makes

1 Valid · 0 Not valid · 9 Not checked · 38 Own results and interpretations

27 more claims

Create a free account to see all 48 claims and to verify the 10 references still open.

See all claims — free account
  • It may state two separate claims as one. 8
  • Probably not a substantial claim. 13

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 33 references

Research works printed but not linked: 2

  1. M. Brookwood, The Orphans of Davenport: Eugenics, the Great Depression, and the War over Children’s Intelligence (Liveright, 2021).
  2. L. K. Muthén, B. O. Muthén, Mplus User’s Guide (Muthén & Muthén, ed. 8, 2017).

Cited by

See all 37

Metadata

Authors
Humphreys, K. L., King, L. S., Guyon-Harris, K. L., Sheridan, M. A., McLaughlin, K. A., Radulescu, A., Nelson, C. A., Fox, N. A., & Zeanah, C. H.
Affiliations
  • Tulane University (US)
  • Department of Psychiatry and Behavioral Sciences, Tulane University School of Medicine, New Orleans, LA 70112
  • University of North Carolina at Chapel Hill (US)
  • Harvard University (US)
  • Institute of Child Development, Bucharest, 010919 Romania
  • Boston Children's Hospital (US)
  • University of Maryland, College Park (US)
Corresponding author
Humphreys, Kathryn L.
Venue
Proceedings of the National Academy of Sciences
Published
2022-09-12
Volume
119
Issue
38
Pages
e2119318119-e2119318119
ISSN
  • 0027-8424 (ISSN-L)
  • 1091-6490
OpenAlex
W4295309339
Cited by
36
References
32
Citations by year
2026: 7 · 2025: 10 · 2024: 9 · 2023: 6 · 2022: 4
FWCI
8.39
Type
journal article
Language
en
Open access
Locations
  • Proceedings of the National Academy of Sciences (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Carolina Digital Repository (University of North Carolina at Chapel Hill) (submittedVersion, cc-by-nc-nd)
  • UNC Libraries
Funders
  • HHS | NIH | National Institute of Mental Health (MH091363)
  • HHS | Health Resources and Services Administration (T32HP22240)
  • HHS | NIH | Eunice Kennedy Shriver National Institute of Child Health and Human Development (F32HD105385)
  • John D. and Catherine T. MacArthur Foundation (n/a)
  • Jacobs Foundation (2017-1261-05)
Topics
  • Social Sciences › Sociology and Political Science › Child Welfare and Adoption
  • Social Sciences › Education › Early Childhood Education and Development
  • Social Sciences › Psychology › Clinical Psychology › Child Abuse and Trauma
Keywords
foster care, institutional care, early deprivation, psychosocial deprivation, cognitive development, IQ, early intervention, caregiving quality, attachment security, Bucharest Early Intervention Project, early adulthood, child development
MeSH terms
Child, Institutionalized, Child, Preschool, Cognition, Foster Home Care, Humans, Infant, Intelligence Tests, Psychosocial Deprivation, Early Intervention, Educational

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Foster care leads to sustained cognitive gains following severe early deprivation. https://app.cochat.ai/papers/10.1073/pnas.2119318119
The paper
Humphreys, K. L., King, L. S., Guyon-Harris, K. L., Sheridan, M. A., McLaughlin, K. A., Radulescu, A., Nelson, C. A., Fox, N. A., & Zeanah, C. H. (2022). Foster care leads to sustained cognitive gains following severe early deprivation. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.2119318119

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.