Extreme temperatures and mortality in 326 Latin American cities

Background: Climate change and urbanization are rapidly increasing human exposure to extreme ambient temperatures, yet few studies have examined the impact of temperature on mortality across Latin America, where 80% of residents live in urban areas. Methods: We used distributed lag nonlinear conditional Poisson models to estimate city-specific associations between daily temperatures above (“heat”) and below (“cold”) each city-specific minimum mortality temperature and all-cause mortality, overall and stratified by age and cause of death. We estimated the percentage of total deaths attributable to heat or cold (excess death fraction [EDF]) and the difference in mortality per 1°C higher daily mean temperature above the 95th percentile of observed daily temperature. Results: We analyzed data from 326 cities in nine Latin American countries between 2002-2015, representing 15,431,532 deaths from 249 million residents. The EDF of total deaths from heat was 0.67% (95% confidence interval [CI] 0.58%, 0.74%), and from cold was 5.09% (95% CI 4.64%, 5.47%). EDFs from heat and cold were particularly high among individuals aged 65+ years (0.81% [95% CI 0.75%, 0.86%] and 6.82% [95% CI 6.41%, 7.18%], respectively). The relative risk of death per 1°C increase above the city-specific 95th percentile daily temperature was 1.057 (95% CI 1.046, 1.067). Conclusions: In Latin American cities, a substantial proportion of deaths are attributable to nonoptimal ambient temperatures. Older populations are particularly vulnerable. Marginal increases in observed hot temperatures are associated with steep increases in mortality risk. Projected increases in temperature from climate change may have a substantial impact on mortality.