Short-term effects of heat and cold in urban, sub-urban and rural municipalities of Italy - a national approach.

Abstract

Listen

Translate article

Background. The short-term effects of heat and cold on mortality are well known, however most evidence comes from studies conducted in urban areas and less is known on the effects in rural and suburban areas. The aim is to estimate the association between temperature and all-cause mortality at national level in Italy.Methods. Daily all-cause mortality for the 8,092 municipalities of Italy were collected for the period 2006 to 2015. A satellite-based spatiotemporal model was developed to estimate daily mean temperature at 1x1km resolution. Conditional Poisson regression models were fit to estimate the risk in mortality for variations (increase/decrease) in temperature (lag0-10) at province level. A DLNM approach was used to account for the non-linear association. Results were then pooled with a random-effects meta-analysis. Associations were estimated for increases (75-99th percentile) and for decreases (25th to the 1st percentile) in mean temperature and by age, sex and level of urbanization of the municipalities. Results. In the study period 6,552,257 deaths occurred in Italy (40% are aged 85+, 55% occur in non-urban municipalities). The risk in mortality for temperature increases was of 1.27 (IC95%: 1.23, 1.31) and of 1.19 (IC95%: 1.23-1.31) for temperature decreases. Some heterogeneity in the effects of heat and cold were observed across Italy; with strongest effects of heat in central-southern regions and no clear geographical pattern for cold effects. In summer, the elderly and women were most at risk. Effects estimates were similar in rural and urbanized municipalities for both heat and cold in Italy. Conclusions. The study confirms a significant risk of extreme temperatures (heat and cold) on all-cause mortality at national level and provides evidence of effects in both urban, sub-urban and rural municipalities.