Maternal Stress Before and During Pregnancy and Adverse Birth Outcomes: A Cross‐Sectional Study of Cumulative Perinatal Stress in The Gambia

Background The Developmental Origins of Health and Disease (DOHaD) framework links maternal stress to adverse offspring outcomes. Whether stress before pregnancy carries risk independent of stress during pregnancy is uncertain, and evidence from sub‐Saharan Africa, where chronic stressors are common, is sparse. This study examined the association between maternal stress in each period and adverse birth outcomes in a West African population, and assessed whether the timing of stress could be distinguished. Methods A cross‐sectional study was conducted among 504 mother–child dyads in The Gambia, with maternal stress before and during pregnancy ascertained retrospectively using a 16‐item checklist for each period. The primary outcome was an adverse birth outcome, defined as low birth weight (<2500 g) or preterm birth (<37 weeks), both fixed at delivery and therefore ascertainable for every child irrespective of age at interview. Multivariable logistic regression estimated adjusted odds ratios (aORs) per one‐standard‐deviation increase in stress, adjusting for maternal age, education, household income, co‐wife presence, parity, antenatal care (ANC) attendance, residence, and the interval between birth and interview. Dose–response across stress quartiles and the contribution of individual stressor domains were examined. Results The prevalence of an adverse birth outcome was 23.6% (119/504; low birth weight 18.5%, preterm birth 8.5%). Mean stress scores were 2.8 ± 2.3 before pregnancy and 2.9 ± 2.4 during pregnancy, and the two were strongly correlated ( r  = 0.85, p < 0.001). Each one‐standard‐deviation increase in stress was associated with raised odds of an adverse birth outcome, both before pregnancy (aOR: 1.50, 95% confidence interval [CI]: 1.20–1.88) and during pregnancy (aOR: 1.53, 95% CI: 1.22–1.92). Because stress in the two periods was highly collinear, the two exposures could not be separated when modeled together, and the data did not support a differential effect of one window over the other. A dose–response gradient was evident across quartiles of during‐pregnancy stress, from 12.1% in the lowest quartile to 35.5% in the highest ( p for trend < 0.001). Women reporting high stress in both periods had higher odds of an adverse outcome than those with low stress throughout (aOR 2.79, 95% CI 1.65–4.69). Stressor domains most strongly associated with adverse outcomes were residential instability, pregnancy‐related problems, and violence exposure; reported family support was associated with lower risk. Conclusions Maternal stress in the period surrounding conception was associated with adverse birth outcomes in this Gambian sample, with a graded relationship between accumulated stress and risk. The strong correlation between the two periods meant that a window‐specific effect could not be isolated, so the findings are best read as evidence of cumulative perinatal stress burden rather than of a single critical window. The stressor domains identified and the dose–response pattern offer plausible targets for psychosocial support within maternal health programs in resource‐limited settings, though the cross‐sectional design with retrospective exposure assessment precludes causal inference.