Low Omega-3 intake is associated with high rates of depression and preterm birth on the country level

Prenatal depression increases the risk of preterm birth (PTB), yet antidepressant use in pregnancy does not reduce PTB risk and may increase adverse outcomes. Both prenatal depression and PTB are common globally and have long-term consequences. Long-chain omega-3 polyunsaturated fatty acids (LC omega-3 PUFA; EPA and DHA) have shown efficacy against major depressive disorder (MDD) in some contexts, and prenatal omega-3 supplementation can reduce PTB risk, although trial findings are heterogeneous, partly due to not accounting for baseline intakes, sufficiency thresholds, and endogenous conversion of plant-based omega-3 (ALA) to LC omega-3. Given biologic plausibility and observational evidence, the authors hypothesized that countries with low omega-3 intakes would have higher rates of both MDD and PTB. They aimed to test this hypothesis using 2010 country-level data and to explore sufficiency thresholds and the potential scale of LC omega-3 insufficiency worldwide.

Meta-analytic and observational studies generally show inverse associations between habitual omega-3 intake (EPA+DHA and total omega-3 including ALA) and depression, with indications of sufficiency thresholds. RCTs of LC omega-3 for MDD show mixed but overall positive effects in specific contexts, though many did not account for baseline intake, thresholds, or ALA conversion, limiting interpretability. For PTB, prenatal omega-3 supplementation can reduce risk, but results vary across settings, with some trials showing no benefit or increased post-term delivery; again, failure to account for thresholds and baseline intake likely contributes to heterogeneity. Biomarker studies consistently link lower internal LC omega-3 levels with depression and prenatal depression, and low maternal LC omega-3 levels in pregnancy predict shorter gestation and PTB. Mechanistic evidence supports plausibility: low omega-3 affects neurotransmission, neuroinflammation, and neurogenesis (for MDD) and increases uterotonic prostaglandins, trophoblast cell death, and perinatal inflammation (for PTB).

Design: Ecological, cross-sectional analysis of 184 countries in 2010 with complete data on MDD prevalence, omega-3 intake, PTB rate, and country income. Data sources: MDD prevalence from the GBD 2010 (Ferrari et al.), PTB rates and income categories (World Bank Atlas Method, 4-level ordinal rank) from Blencowe et al., and omega-3 intakes (plant-based ALA and seafood-based EPA+DHA) from NutriCoDE (Micha et al.). Exposure metric: Estimated total LC omega-3 PUFA intake (mg/day) per country defined as seafood-based EPA+DHA plus 15% of plant-based ALA to account for endogenous conversion (sex-averaged conversion rate reflecting ~21% in women and ~8% in men). Sensitivity analyses varied the ALA conversion rate (0–25%) and examined plant-based and seafood-based intakes separately. Outliers: Afghanistan (MDD prevalence 22.5/100) and the Maldives (LC omega-3 intake 3918 mg/day) were excluded in some analyses. Statistical analysis: Penalized spline models using GAM (mgcv in R 3.5.0) with generalized cross-validation (GCV) assessed nonlinearity and identified putative thresholds in the omega-3–MDD and omega-3–PTB relationships. Based on spline-identified thresholds (~1000 mg/day for MDD; ~550 mg/day for PTB), separate linear regressions were fit below and above thresholds. MDD models adjusted for country income and PTB rate; PTB models adjusted for country income and MDD prevalence. Additional descriptive and correlation analyses (Spearman) were performed. Software: R 3.5.0 and SAS 9.4.

- Penalized splines showed inverse relationships between LC omega-3 intake and both outcomes up to thresholds: ~900–1100 mg/day for MDD and ~500–600 mg/day for PTB. - Below 1000 mg/day LC omega-3 (n=177), a 1 SD increase (380 mg/day) associated with a 0.50 decrease in MDD cases per 100 people (95% CI: 0.13, 0.88) after adjusting for income and PTB rate; above 1000 mg/day (n=5), no significant association. - Below 550 mg/day LC omega-3 (n=157), a 1 SD increase associated with a 1.51 decrease in PTB per 100 live births (95% CI: 0.15, 2.88) after adjusting for income and MDD; above 550 mg/day (n=25), no significant association. - Country income positively associated with MDD prevalence and inversely with PTB rate; PTB rate positively associated with MDD prevalence in adjusted models. - Correlations: LC omega-3 intake correlated with country income (Spearman 0.44) and inversely with PTB rate (-0.32). - Geographic patterns: Low intakes concentrated in Africa, South-Central Asia, and Central America; higher intakes in Southeast Asia and North Atlantic Europe. - Sufficiency status: 6 countries above 1000 mg/day (MDD threshold); 26 above 550 mg/day (PTB threshold); 158 countries below 550 mg/day; 53 countries below 170 mg/day (≥1 SD below PTB threshold). - Policy-relevant estimates: For countries well below threshold, increasing LC omega-3 to 550 mg/day could reduce PTB by ≥15 per 1000 births and MDD by ≥5 per 1000 people; increasing to 1000 mg/day would further reduce MDD by ≥11 per 1000 people, with PTB already at threshold. - Sensitivity analyses varying ALA conversion (0–25%) yielded similar spline shapes with predictable threshold shifts; separate plant vs seafood analyses supported thresholds (seafood: ~400–550 mg/day for PTB; ~850–1100 mg/day for MDD; plant-based: PTB ~2500–3500 mg/day threshold; no MDD threshold detected within observed range).

The findings support the hypothesis that low LC omega-3 PUFA intake at the country level is associated with higher MDD prevalence and PTB rates, with clear sufficiency thresholds beyond which additional intake confers no further benefit. These ecological results accord with individual-level observational and biomarker studies and meta-analyses demonstrating inverse associations and threshold effects, enhancing biological plausibility. Adjusted models indicate that associations are not explained by country income or the reciprocal outcome (MDD or PTB), and the positive correlation between PTB and MDD at the country level aligns with individual-level evidence linking depressive states and adverse perinatal outcomes. The work underscores the importance of accounting for baseline intakes, sufficiency thresholds, and endogenous ALA-to-EPA/DHA conversion when evaluating omega-3 relationships with health outcomes. It suggests that food systems and habitual dietary patterns may drive population-level burdens of MDD and PTB, indicating potential for population-level nutritional interventions to reduce these burdens.

This study integrates country-level data to show inverse, threshold-limited associations between LC omega-3 intake and both MDD prevalence (~1000 mg/day threshold) and PTB rate (~550 mg/day threshold). Given alignment with extensive individual-level evidence, results suggest that low omega-3 intake may elevate MDD and PTB rates in about 85% of countries analyzed. The work highlights the necessity of considering baseline intake, sufficiency thresholds, and ALA conversion in research and practice. Future research should refine threshold estimates using internal biomarkers, characterize who benefits most from increased omega-3, account for genetic and nutritional modifiers of conversion, and design interventions and policies that safely and sustainably increase LC omega-3 levels in populations with low intake.

- Ecological, cross-sectional design limits causal inference and is susceptible to ecological fallacy and reverse causation. - Limited covariate data at the country level; potential confounding by unmeasured factors (e.g., other nutrients like folate or vitamin D, environmental exposures, reproductive histories) could bias estimates. - No country-level data on omega-3 supplement use; income used as a proxy, but residual confounding likely. - Exposure metric relies on dietary intake estimates and an assumed average ALA-to-EPA/DHA conversion rate (15%); true conversion varies by genetics, nutrition, sex, and other factors. - Could not directly analyze internal biomarker levels or distinguish effects of EPA vs DHA. - Outlier handling (Afghanistan, Maldives) may affect generalizability, though results robust to exclusions. - PTB and MDD are heterogeneous conditions with multiple determinants; analyses focus on one nutritional component. - Country-level analyses cannot determine individual-level sufficiency requirements or clinical efficacy in specific subgroups.