The effect of flooding on low birthweight and preterm birth: a systematic review and meta-analysis

Abstract

Background: Numerous studies have examined pregnancy outcomes following flood events, with the majority focusing on two related outcomes: preterm birth (PTB) and low birthweight (LBW). Summarizing the results of these previous studies and determining remaining data gaps is the main objective of this systematic review and meta-analysis.

Methods: We included publications in English that examined birthweight and/or gestational length related to exposure to floods, or events typically causing flooding (e.g. tropical cyclones). Seven academic databases were searched: CAB Abstracts (CABI), Academic Search Complete and Environment Complete (EBSCOhost), Environmental Science Index & Database (ProQuest), PubMed, Scopus, and Web of Science Core Collection. Searches were updated on February 23, 2025. For inclusion in meta-analyses, quantitative estimates of effect size and variance were required, and quality was assessed using the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. Random effects regression was used for meta-analyses, and results are presented in forest plots, with potential for publication bias assessed in funnel plots and Egger’s test results.

Results: Overall, data from 34 studies were extracted, and 25 studies across 13 countries were included in meta-analyses. Most studies (N = 18) examined tropical cyclone exposure. Meta-analyses indicate increases in LBW (RR = 1.03, 95% CI: 1.00, 1.05) and PTB (RR = 1.10, 95% CI: 1.00, 1.22). The LBW result was not significantly influenced by quality rating, while the PTB result is non-significant when all studies, regardless of quality rating, were included in the meta-analysis (RR = 1.01, 95% CI: 0.97, 1.05). Additionally, the PTB estimate is strongly influenced by one study with a large and highly significant effect size. Additional sub-analyses suggest no decreasing effect following more recent events (after 2005).

Conclusions: Results are limited by the range of methods used across studies to estimate exposure to flooding and potential co-exposures related to events that caused the flooding (e.g. wind damage-related health outcomes during tropical cyclones). Regardless, results indicate that adverse pregnancy outcomes may increase following in utero exposure to flood events. Future studies incorporating finer spatiotemporally resolved estimates of exposure to flooding will improve estimates of effect. The study is registered in PROSPERO (CRD42024514540).