The effect of extreme temperatures on healthcare utilization during pregnancy: Findings from a nationally representative survey in India.

Abstract

Background: Extreme temperatures are increasing in frequency and severity, posing growing risks to maternal and child health through adverse pregnancy and birth outcomes. This challenge is especially pronounced in countries like India, that lack adequate protections against climate-related health risks. While socio-economic and accessibility barriers already limit healthcare utilization during pregnancy, climatic factors such as extreme temperatures can create additional obstacles to accessing these essential services, thereby influencing pregnancy and birth outcomes both directly and indirectly.

Methods: We conducted a cross-sectional analysis using data from the National Family Health Survey (NFHS-5), a nationally representative survey for India. We examined recent healthcare utilization reported by women in their fifth month of pregnancy or later (N = 10,606). Temperature exposure was measured using daily Wet Bulb Globe Temperature (WBGT), Heat Index, and Dry Bulb Temperature (DBT) matched to respondents' geo-coded residential clusters. We calculated the number of days during the study period when temperature values exceeded or fell below extreme temperature thresholds. Generalized mixed-effects models with Inverse Propensity of Treatment Weighting (IPTW) examined associations between extreme temperature and no healthcare contact, with effect modification tested across socioeconomic variables.

Findings: Extreme heat exposure significantly increased the likelihood of healthcare non-utilization among pregnant women, with the strongest associations observed for WBGT and Heat Index measures. Each additional day above the 90th percentile WBGT threshold increased odds of no healthcare contact by 11%, with similar patterns across the 85th and 95th percentiles. Heat Index showed consistent positive associations across all thresholds (8-11% increases), while DBT trends were positive but not statistically significant. For extreme cold exposure, temperature indicators showed some association with healthcare non-utilization, though results were not consistent across exposure definitions. We did not find evidence of effect modification across socioeconomic groups, with only a few isolated exposure definitions showing significant difference between subgroups.