The associations of cold spells with human mortality in Shanghai, China

Abstract

Current studies of climate change impacts on human health primarily focus on extreme heat. It is equally imperative to understand the substantial effects of extreme cold on human well-being. We defined 15 distinct cold spells based on various temperature percentiles and durations observed in Shanghai between 2016 and 2020. To estimate the association between cold spells and human mortality, we employed a quasi-Poisson regression with a distributed non-linear model (DLNM). We quantified the mortality burden attributed to cold spells. Our observations showed that the relative risks of cold spells on mortality varied depending on cold spell definitions and sub-population groups. People were more susceptible to cold spells that spanned at least two consecutive days with a daily mean temperature falling below the 5th percentile of temperature records. The prolonged lag effect of cold spells on mortality lasted for two weeks. Cold spells positively associated with total non-accidental deaths and cardiovascular disease. Sub-population groups, including females, the elderly, and individuals with low education, exhibited heightened vulnerability to cold spells compared to their respective counterparts. Our findings are expected to provide guidance for the development of an early warning system aimed at mitigating the adverse impacts of cold spells on human mortality.