Abstract
We investigate the effect of extreme heat on home births in India, proposing that such extreme weather events may impede access to health facilities for childbirth. Utilizing geocoded data from the 2019-2021 Demographic and Health Survey for India, we identified the place of delivery of 208,368 births as home versus health facility. We incorporated maximum values for gridded wet-bulb globe temperatures (WBGT(max)) and dry-bulb temperatures (DBT(max)) corresponding to delivery dates and maternal residences. We defined context-specific extreme heat events using several percentile-based thresholds (between 80th and 95th) over varying durations (1-5 days). We used Generalized Estimating Equations (GEE) with inverse probability of treatment weighting, incorporating socioeconomic factors and state-level fixed effects, and adjusted for seasonality. We tested for effect-measure-modification by socio-economic factors (e.g., caste, wealth), healthcare access factors (e.g., rural/urban place of residence, difficultly in accessing healthcare), and contextual factors (e.g., long-term mean temperature, prevalence of institutional delivery). Both WBGT(max) and DBT(max)-based heatwave exposures were associated with increased likelihood of home births, with WBGT exposures demonstrating an earlier onset of significant associations at lower percentile thresholds while DBT showed stronger associations at higher thresholds and longer durations. Effect modification analyses revealed heightened impacts in warmer regions, states not designated as high-focus under the Janani Suraksha Yojana program, and non-Hindu populations. We find that extreme heat may compromise delivery at health facilities in India. Findings call for improved health system preparedness via early warning systems and advanced resource allocation to mitigate some of these effects.