The association of adverse birth outcomes with flood exposure before and during pregnancy in Australia: a cohort study
Zhengyu Yang,Wenzhong Huang,Joanne E McKenzie,Pei Yu,Yao Wu,Rongbin Xu,Tingting Ye,Ke Ju,Yiwen Zhang,Yujia Huang,Yuming Guo,Shanshan Li
DOI: https://doi.org/10.1016/S2542-5196(24)00142-6
Abstract:Background: Exposure to floods might increase the risks of adverse birth outcomes. However, the current evidence is scarce, inconsistent, and has knowledge gaps. This study aims to estimate the associations of flood exposure before and during pregnancy with adverse birth outcomes and to identify susceptible exposure windows and effect modifiers. Methods: In this cohort study, we obtained all the birth records occurring in Greater Sydney, Australia, from Jan 1, 2001, to Dec 31, 2020, from the New South Wales Midwives Data Collection and in the Brisbane metropolitan region, Australia, from Jan 1, 1995, to Dec 31, 2014, from the Queensland Health Perinatal Data Collection. For each birth, residential address and historical flood information from the Dartmouth Flood Observatory were used to estimate the numbers of days with floods during five exposure windows (Pre-1 was defined as 13-24 weeks before the last menstrual period [LMP], Pre-2 was 0-12 weeks before the LMP, trimester 1 [Tri-1] was 0-12 weeks after the LMP, trimester 2 [Tri-2] was 13-28 weeks after the LMP, and trimester 3 [Tri-3] was ≥29 weeks after the LMP). We estimated the hazard ratios (HRs) of adverse birth outcomes (preterm births, stillbirths, term low birthweight [TLBW], and small for gestational age [SGA]) associated with flood exposures in the five exposure windows using Cox proportional hazards regression models. Findings: 1 338 314 birth records were included in our analyses, which included 91 851 (6·9%) preterm births, 9831 (0·7%) stillbirths, 25 567 (1·9%) TLBW, and 108 658 (8·1%) SGA. Flood exposure in Pre-1 was associated with increased risks of TLBW (HR 1·06 [95% CI 1·01-1·12]) and SGA (1·04 [1·01-1·06]); flood exposure during Tri-1 was associated with increased risks of preterm births (1·03 [1·002-1·05]), stillbirth (1·11 [1·03-1·20]), and SGA (1·03 [1·01-1·06]). In contrast, flood exposures during Pre-2 and Tri-3 were associated with reduced risks. Interpretation: Exposures to floods in Pre-1 and Tri-1 are both associated with increased risks of adverse birth outcomes, and the risks increase with a higher exposure. Upon planning for conception and prenatal care, individuals and health practitioners should raise awareness of the increased risks of adverse birth outcomes after experiencing floods. Funding: The Australian Research Council and the Australian National Health and Medical Research Council.