Short-term association between hot nights and mortality: a multicountry analysis in 178 locations considering hourly ambient temperature

Le résumé fourni par la source

BACKGROUND: The rise in hot nights over recent decades and projections of further increases due to climate change underscores the critical need to understand their impact. This knowledge is essential for shaping public health strategies and guiding adaptation efforts. Despite their significance, research on the implications of hot nights remains limited. OBJECTIVE: This study estimated the association between hot-night excess (the sum of excess heat during the nighttime above a threshold) and duration (the percent of nighttime with a positive excess) based on hourly ambient temperatures and daily mortality in the warm season over multiple locations worldwide. METHODS: We fitted time series regression models to mortality in 178 locations across 44 countries using a distributed lag non-linear model over lags of 0-3 days, controlling for daily maximum temperature and daily mean absolute humidity. Next, we used a multivariate meta-regression model to pool results and estimated attributable burdens. RESULTS: We found a positive, increasing mortality risk with hot-night excess and duration. Assuming 0 as a reference, the pooled relative risks of death associated with extreme excess and duration, defined as the 90th percentile in each index, were both similar at 1.026 (95 % CI, 1.017; 1.036) and 1.026 (95 % CI, 1.013; 1.040). The overall estimated attributable fractions were also observed to be closely similar at 0.60 % (95 % CI, 0.09; 1.10 %) and 0.62 % (95 % CI, 0.00; 1.23 %), respectively. DISCUSSION: This study provides new evidence that hot nights have a specific contribution to heat-related mortality risk. Modeling thermal characteristics' sub-hourly impact on mortality during the night could improve decision-making for long-term adaptions and preventive public health strategies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.