Related Experiment Video
Updated: Aug 6, 2026

Breakfast Habits among Schoolchildren in the City of Uruguaiana, Brazil
Published on: July 29, 2020
Association between drought and adverse birth outcomes: a nationwide cohort study with 29.8 million live births in
Aline Dos Santos Rocha1,2, Leydson Galvíncio Dantas1,3, Andrêa J F Ferreira1
1Center of Data and Knowledge Integration for Health (CIDACS), Oswaldo Cruz Foundation, Salvador, Brazil.
Background:
Evidence on the association between drought and adverse pregnancy outcomes is inconsistent, particularly for preterm birth and small for gestational age (SGA). This study evaluated the association between drought exposure and adverse birth outcomes in Brazil and assessed whether this relationship varies by biome and level of human development.
Methods:
We conducted a longitudinal study using routinely collected socioeconomic, birth, and climate data in Brazil from 2001 to 2020. Drought exposure was measured using the Standardized Precipitation Evapotranspiration Index (SPEI) at 3-, 6-, and 9-month timescales. Logistic regression models were used to estimate associations between drought exposure and preterm birth, term low birth weight (LBW), and SGA. Analyses were stratified by biome and level of municipal Human Development Index (HDI).
Findings:
The study included approximately 29.8 million live births, of which 8.0% were preterm, 3.9% were term LBW, and 9.3% were SGA. Across all exposure windows, drought was associated with higher odds of preterm birth (ORs ≈1.02). For SPEI-6, a dose-response relationship was observed, with higher odds of preterm birth from moderate drought (OR = 1.01; 95% CI: 1.01-1.02) to severe (OR = 1.03; 95% CI: 1.03-1.04) and extreme drought (OR = 1.05; 95% CI: 1.03-1.06). Drought exposure was also associated with SGA at the SPEI-6 (OR = 1.01; 95% CI: 1.01-1.02) and SPEI-9 (OR = 1.02; 95% CI: 1.01-1.02) scales, while no consistent association was observed for term LBW. Associations varied across biomes and HDI levels.
Interpretation:
Drought exposure was associated with higher odds of preterm birth and SGA, particularly during severe events, but showed no association with term LBW.
Funding:
Decit, Wellcome Trust.
Related Concept Videos
Bias in Epidemiological Studies
Teratogenicity
Regression Toward the Mean
Development of the Oral Microbiota
Confounding in Epidemiological Studies