Long-term effects of air temperature on cardiopulmonary outcomes in prospective European cohort studies

BACKGROUND AND AIM: Short-term health effects of air temperature have been well recognized. However, little is known about the effects of prolonged exposure to non-optimal air temperatures, particularly at the individual level. This study assessed long-term temperature effects on cardiopulmonary (CPD) outcomes using data collected in four prospective European cohorts. METHODS: Participants of the KORA (Augsburg, Germany), RoLS (Rome, Italy), CONOR (Norway), and UK Biobank (UK) cohorts were included in the current analysis. Their vital status and incident diseases were identified through death and disease registries as well as follow-up surveys. Annual mean and standard deviation (SD) of air temperature were calculated for each participant and follow-up year based on the daily mean air temperature at residences estimated by spatial-temporal models. We applied cohort-specific Cox proportional hazard models with adjustment for individual characteristics and time trends, using the distributed lag non-linear model approach, to assess the associations of long-term air temperature with CPD mortality and morbidity. Random-effects meta-analysis was used to pool the cohort-specific estimates. RESULTS: In the RoLS cohort, we observed a monotonically increasing risk for cardiovascular mortality associated with increasing annual mean and SD of temperature. For an increment from the 75th to the 95th percentile in the annual mean and SD, the hazard ratios for cardiovascular mortality were 1.055 (95% CI: 1.029, 1.081) and 1.151 (95% CI: 1.060, 1.251), respectively. Besides, increasing annual SD showed an indication of being associated with increased respiratory mortality risk. In the KORA cohort, no effects were observed for annual mean temperature or SD on CPD mortality and morbidity. Cohort-specific analyses in Norway and UK and the meta-analysis are ongoing. CONCLUSIONS: This study provides evidence for long-term air temperature effects on CPD outcomes but also indicates heterogeneity in such effects among four European cohorts. KEYWORDS: Air temperature, long-term, cardiopulmonary mortality, cohorts