CoChat

Resilience following severe early life neglect: A review of longitudinal evidence from the Bucharest Early Intervention Project

Open access

The DOI matches its Crossref record; 71 of the 71 references it cites were found in a registry or an index; 80 of the 117 sources its claims cite were checked for support.

Abstract

Early institutional care is associated with long-lasting adverse outcomes across neurodevelopmental, cognitive, socioemotional, and physical health domains. However, less is known about how resilience emerges following this type of severe neglect. In this review, we define resilience as a descriptive inference reflected in better-than-expected adaptation following adversity among individuals exposed to comparable adversity, and we argue that progress in resilience science requires designs that can test causal mechanisms. We highlight how the randomized foster care intervention in the Bucharest Early Intervention Project (BEIP) provides rare causal evidence on modifiable processes that support resilience following early institutional rearing. Drawing on BEIP findings, we synthesize evidence that improving early caregiving environments leads to more favorable developmental outcomes during childhood and adolescence in multiple domains, while also illustrating that recovery is partial and not uniform across domains. We conclude with key lessons learned about resilience from the BEIP and outline future directions leveraging adult and intergenerational follow-up data to refine how resilience is defined, measured, and mechanistically explained in the context of severe early neglect. By leveraging rare experimental evidence, the BEIP provides a blueprint for identifying modifiable mechanisms through which early intervention can shape resilience across the lifespan following early adversity.

Overview

CoChat Summary

TL;DR

This narrative review uses the Bucharest Early Intervention Project (BEIP) to argue that resilience after severe early institutional neglect is better understood as a dynamic, domain-specific inference shaped by modifiable caregiving conditions rather than a fixed child trait. It concludes that randomized foster care improves many outcomes but recovery is partial, with timing and stability of caregiving especially important.

Key findings

  • The BEIP randomized 136 institutionalized infants/toddlers (6–31 months) to high-quality foster care (FCG) or care-as-usual (CAUG), with 72 never-institutionalized children as a comparison group; follow-ups span 30, 42, and 54 months and 8, 12, 16, and 22 years.
  • Foster care intervention was linked to improvements across neurobiological, cognitive, socioemotional, and physical domains, often moving outcomes closer to never-institutionalized children, but effects were partial and not uniform: the review’s Table 1 rates attachment/socioemotional development and attachment-disorder symptoms as showing “strong” improvements, IQ/language and physical growth as “moderate to strong,” and internalizing symptoms as “weak” for boys but “strong” for girls.
  • Earlier placement into foster care (before 24 months) and greater placement stability were consistently linked to more resilient outcomes, whereas later placement and more disruptions were linked to less favorable outcomes.
  • The review argues that caregiving quality and stability, supported by intensive professional infrastructure, were likely active ingredients, rather than family placement or reduced institutional exposure alone.

Methods

  • Design: narrative review synthesizing longitudinal and randomized evidence from the BEIP, described as the only randomized trial evaluating family-based care as an alternative to institutional care for children exposed to severe early deprivation.
  • Sample/data: BEIP enrolled 136 institutionalized infants and toddlers from six Romanian institutions, randomized to FCG or CAUG, plus 72 never-institutionalized community children; the comparison group was partially refreshed at later waves, so its composition and size varied across assessments.
  • Analysis: qualitative synthesis of BEIP findings across domains, including a table of intervention effects with authors’ qualitative magnitude ratings; no new data were created or analyzed, and no formal meta-analysis or standardized effect-size thresholds were used.

Limitations

  • The paper states that the BEIP design cannot definitively isolate the active ingredients of foster care and cannot fully disentangle reduced institutional exposure, family setting, caregiving quality, and professional support.
  • The paper states that no pre-registration was completed because it is a narrative review, and no new data were created or analyzed.
  • Evident from the text: the effect-magnitude ratings are qualitative and not based on a formal meta-analysis or standardized thresholds; the never-institutionalized comparison group was partially refreshed, so its composition and sample size varied across assessments; and most BEIP work has focused up to adolescence, with adult and intergenerational findings still emerging.

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

Integrity signals

Registry

Crossref · 10.1017/s0954579426101837

Cites

71 of 71 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

51 Valid · 3 Not valid · 26 Not checked · 70 Own results and interpretations

74 more claims · 29 with their cited sources checked

Create a free account to see all 150 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 44
  • Probably not a substantial claim. 38
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Citations

Cites

Metadata

Authors
Holt-Gosselin, B., Adri, F., Humphreys, K. L., & Nelson, C. A.
Affiliations
  • Boston Children's Hospital (US)
  • Vanderbilt University (US)
Corresponding author
Nelson, Charles Alexander
Venue
Development and Psychopathology
Published
2026-08-12
Pages
1-11
ISSN
  • 0954-5794 (ISSN-L)
  • 1469-2198
OpenAlex
W7202265227
Cited by
0
References
64
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Development and Psychopathology (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
  • National Science Foundation (CAREER, 2042285)
  • John D. and Catherine T. MacArthur Foundation
  • Binder Family Foundation (R01MH091363)
  • National Institute of Mental Health (R01MH091363)
Topics
  • Social Sciences › Sociology and Political Science › Child Welfare and Adoption
  • Social Sciences › Psychology › Clinical Psychology › Child Abuse and Trauma
  • Social Sciences › Psychology › Clinical Psychology › Resilience and Mental Health
Keywords
resilience, institutional care, Bucharest Early Intervention Project, neurodevelopmental outcomes, social-emotional development, cognitive development, physical health, recovery from adversity, causal mechanisms, early intervention, intergenerational effects

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-29, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-29 (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

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.