The mediating role of acute respiratory infections in temperature-mortality associations in the Czech Republic, 1982–2019

Numerous epidemiological studies have shown that non-optimal temperatures and acute respiratory infections (ARIs) contribute to increased mortality. The effect of temperature on mortality is typically estimated using splines of time, which are designed to account for seasonal and longer-term trends, and account for the potential unmeasured confounding, such as the seasonal effect of ARIs. However, the distinct impact of ARIs on mortality, as well as the direct effect of temperature on mortality that is not mediated via ARIs, remains insufficiently understood. This study investigates and quantifies the impact of ARIs on mortality, and the impact of temperature on mortality that is not mediated via ARIs. We used 38-year time-series data from the Czech Republic and applied a distributed lag non-linear model (DLNM) with one and multiple cross-bases to isolate the direct pathways of temperature and ARIs. Our approach allowed us to estimate the fraction of mortality attributable to each factor, providing a clearer understanding of their respective contributions to seasonal mortality patterns. Our findings indicate that ARI activity is a significant mediator of the relationship between temperature and mortality. Since low temperatures increase the risk of both cold-related mortality and ARI incidence, our results indicate that approximately 12% of cold-related deaths during the study period could be attributed to ARI activity. Additionally, an analysis of temporal changes in the combined effect of ARIs and temperature suggested that while the mediating role of ARIs on temperature-related mortality has weakened, the proportion of cold-attributable mortality has remained constant throughout the study period. Our findings are important for understanding both historical trends and future projections of seasonal mortality patterns. They highlight the need for further research on the roles of climatic and individual risk factors in long-term changes in temperature-related mortality, particularly considering the mediating role of ARIs.