Observational evidence of the seasonal and demographic variation in experienced temperature from 77 743 UK Biobank participants

Abstract Background Exposure to cold is known to be associated with severe health impacts. The primary epidemiological evidence for this is the seasonal variation in mortality. However, there is a paucity of directly measured data for personal cold temperature exposure. This paper develops the concept of experienced temperature, and reports how it varies with season, demographics and housing factors. Methods This study uses data from 77 743 UK Biobank participants. A novel method to directly measure participant’s exposure to low temperatures using a thermistor in a wrist-worn activity monitor is described. These readings are combined with demographic and housing factor variables in a multiple regression model to understand underlying relationships. Results The study reveals a significant difference in experienced temperature of ~1.8°C between the periods of coldest and hottest external temperature. A number of demographic differences were also observed—such as people of Chinese ethnic background experiencing 0.65°C lower temperatures than other groups. Conclusions This paper presents primary evidence for a seasonal variation in experienced temperature. This variation likely contributes to cold related mortality and morbidity. It is hypothesized that this relationship would be less strong in countries which suffer fewer impacts of cold winter temperatures.