Short-term heat effects on cardiopulmonary morbidity using national small area data – Results of the EXHAUSTION project

Abstract

BACKGROUND AND AIM: Short-term associations between heat and cardiopulmonary morbidity have been examined mostly in major cities. Given potential differences in the effects of heat between urban and rural areas, it is important to analyze areas of various degrees of urbanization to better understand the health effects of heat. This study investigated heat effects on cardiopulmonary hospital admissions in the warm season (May-September) in Italy and Germany at the small-area level. METHODS: Daily counts of cardiovascular and respiratory hospital admissions were obtained for each municipality in Italy (2001–2015) and Germany (2000–2016). Spatio-temporal models estimated daily mean air temperatures at the municipality level. In each municipality-aggregated district, we first assessed heat effects at lag 0-1 days using district-specific Poisson regression with adjustment for a 4-way interaction between municipality, year, month, and day of the week, using the distributed lag non-linear model approach. Multivariate meta-analysis was then applied to pool the district-specific estimates in each country. Effect modification by age and sex was examined in subgroup analyses. RESULTS: For an increment in daily temperature from the 75th to the 99th percentile, the risk for respiratory hospital admissions increased in both Italy [pooled risk ratio (RR): 1.112, 95% CI: 1.098-1.126] and Germany (pooled RR: 1.074, 95% CI: 1.067-1.081). We did not observe adverse heat effects on cardiovascular hospital admissions in either country. The heat effects on respiratory hospital admissions were stronger in the elderly (75+ years) and females. CONCLUSIONS: Our study provides evidence for an adverse heat effect on respiratory morbidity, which was stronger in Italy with lower latitude. In both countries, the observed heat effects were more prominent in the elderly and females. Further analyses are ongoing to disentangle effects between urban and rural areas. KEYWORDS: air temperature, heat, cardiopulmonary hospital admissions, small area