Temperature-Related Mortality in Latin America: A Multi-City Multi-Country Study

Abstract

BackgroundAttributable mortality risk for heat and cold has been investigated in Europe and North America. However, none so far has provided estimates for Latin American, with a whole temperature range in different climates. We aimed to estimate the mortality attributable to heat and cold in the Latin American countries participating in the Multi-City Multi-Country Collaborative Research Network, and study their heterogeneity of effects by climatic zones.MethodsWe collected data for 40 cities in Argentina, Brazil, Chile, Colombia and Mexico, between 1997-2015, totalling over 9 million deaths. We conducted a two-stage approach. Firstly, we estimated city-specific temperature-mortality associations using quasi-Poisson regression with distributed lag non-linear models with 21 days of lag. We calculated attributable deaths for heat and cold above and below the city-specific minimum mortality temperature using cutoffs at the 2.5th and 97.5th temperature percentiles. Next, we combined city-specific estimates using meta-regression including indicator variables for the country and climatic zone.ResultsMore temperature-attributable deaths were caused by cold (4.1%, 95%CI=[3.2 to 5.0%]) than by heat (0.7%, 95%CI=[0.4 to 0.9%]). This was observed consistently in all countries. Larger mortality was attributable to cold in dry and temperate than in tropical climates. While for heat, attributable mortality was larger in dry than in tropical and temperate climates.DiscussionMost of the temperature-related mortality burden in Latin American is attributable to the contribution of cold and mainly observed in dry climates. This could be of relevance to developing public-health interventions in the region to minimize the adverse health effects of temperatures.On behalf of the MCC Collaborative Research Network.