Association between pregnancy loss and ambient PM2.5 using survey data in Africa: a longitudinal case-control study, 1998-2016

Background Increasing evidence from epidemiological studies suggests that maternal exposure to ambient PM2.5 can increase the risk of pregnancy loss. However, no studies have been done in low-income countries such as those in Africa, which have the highest incidences of pregnancy loss. In this study, we aimed to analyse the association between PM2.5 and pregnancy loss, including miscarriage and stillbirth. Methods This self-compared case control study collected data on mothers who reported at least two births (at least one case of pregnancy loss plus at least one normal delivery) in the African Demographic and Health Surveys. Gestational exposure to PM2.5 was assessed using a state-of-the-art estimator based on satellite measurements and chemical transport model outputs. Each case of pregnancy loss was compared with the successful control deliveries in the same mother. We estimated the association between gestational exposure to PM2.5 and pregnancy loss using a conditional logistical model with multiple adjustments, and assessed it using the odds ratio (OR) derived from the regression. Findings Between Jan 1, 1998, and Dec 31, 2016, 67 566 cases of pregnancy loss were reported across Africa in the DHS. After removal of mothers who did not report at least one successful delivery and those with missing spatial information, 42 952 were included in the study. Of these, 30 418 were categorised as miscarriages and 12 534 were stillbirths. Each increment of 10 mu g/m(3) PM 2.5 was associated with an adjusted OR of 1.122 (95% CI 1.107-1.137) for pregnancy loss, including miscarriage (1.125; 1.109-1.142) and stillbirth (1.094; 1.051-1.138). Interpretation Our findings from African DHS data showing that PM 2.5 exposure is significantly associated with increased risk of pregnancy loss add to the existing epidemiological evidence from middle-income and high-income countries on the health impacts of poor air quality. Our results support public health interventions for reducing ambient particulate matter to improve maternal health in Africa. Copyright (C) 2019 The Author(s). Published by Elsevier Ltd.