Peer-reviewed article

Social determinants of early childhood development in The Gambia: a multi-level analysis of household, maternal, and healthcare factors.

Bajinka O, Barrow A · 15 Jul 2026

Abstract

Background Early childhood development is a critical determinant of lifelong health and well-being, yet evidence on its social and maternal determinants from West Africa remains limited. We examined the independent contributions of household, maternal, and healthcare factors to maternally reported suspected developmental concerns in children under two years in The Gambia. Methods We conducted a cross-sectional study of 504 mother-child dyads recruited through multi-stage probability sampling from urban and rural settlements in The Gambia (August-October 2021). The primary outcome was maternally reported suspected developmental concerns, assessed by a single binary question combining developmental and growth-related concerns. Cumulative maternal stress was measured using a 32-item binary checklist adapted from the Revised Prenatal Distress Questionnaire (Cronbach's α = 0.79). Hierarchical multivariable logistic regression examined factors at household (settlement type, education, income), maternal (age, stress, polygamy, intimate partner violence, anemia, hypertension), and healthcare (adequate antenatal care, ≥ 8 visits) levels with child age and sex retained as covariates. We formally tested for effect modification by settlement type using a product term. Results Suspected developmental concerns were reported for 69 children (13.7%). High cumulative maternal stress (≥ 6 stressors) was independently associated with the outcome (adjusted odds ratio [aOR] = 2.80, 95% confidence interval [CI]: 1.38-5.70, p = 0.004). Intimate partner violence during pregnancy showed a positive association of borderline significance (aOR = 2.06, 95% CI: 0.99-4.28, p = 0.054). Adequate antenatal care was not associated with the outcome (aOR = 0.81, 95% CI: 0.46-1.41, p = 0.452). The interaction between high stress and settlement type was not statistically significant (interaction p = 0.59); stratified estimates were of similar magnitude in urban (OR = 3.48) and rural (OR = 4.62) settings. Maternal-level factors substantially improved model fit (Δ pseudo-R² = 0.060) compared with household (R² = 0.027) or healthcare factors (ΔR² = 0.002). Conclusion Cumulative maternal psychosocial stress was independently associated with maternally reported suspected developmental concerns in young Gambian children, with a similar association observed in urban and rural settings. Intimate partner violence during pregnancy showed a positive association of borderline statistical significance. The number of antenatal care visits alone was not associated with the outcome, suggesting that visit quantity may be insufficient without quality improvements that address maternal mental health. These findings, while cross-sectional and based on a single-item maternal-report outcome, support further investment in integrated maternal mental health services within antenatal care. Longitudinal studies using validated developmental assessments are needed to establish temporal relationships and causality.