Associations between extreme temperature and all-cause, and cause-specific mortality in elderly German women

Le résumé fourni par la source

BackgroundClimate change and extreme temperature significantly impact public health by influencing morbidity and mortality. However, the cause-effect relationship remains unclear due to limited data on individual characteristics.ObjectiveThis study investigates the association between extreme temperature and all-cause, and cause-specific mortality in elderly German women, using data from the SALIA cohort (4,756 women, with 718 deaths recorded from a mortality follow-up in 2008).MethodsWe used a time-stratified case-crossover study design with a distributed lag nonlinear model (DLNM) with up to 5-day lags. Effects of temperature on mortality were expressed as odds ratios (OR), which were the result of a conditional logistic regression.ResultsA nonlinear association between mean ambient temperature (Tmean) and all-cause and cause-specific mortality was found. Higher temperatures (15°C; 75th percentile & 21°C; 95th percentile) had a significant delayed association with all-cause mortality, with a significant increase in the OR above 1 starting at lag 4. Lower temperatures (-1°C; 5th percentile & 5°C; 25th percentile) showed no significant effects on mortality. Moreover, no significant associations were observed for cause-specific mortality such as mortality due to cardiorespiratory disease (CRD), cardiovascular disease (CVD), or ischemic heart disease (IHD). Sensitivity analyses further confirmed the robustness of the findings.ConclusionHigher temperatures had significant delayed effects on all-cause mortality, while lower temperatures and cause-specific mortality showed no significant associations. Cohort studies are essential for elucidating temperature-mortality relationships and guiding preventive strategies, though further research is needed to understand the complex interplay between temperature and health outcomes, considering various individual factors.

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