Seasonal Temperature Variability and Mortality Across the United States

Background: Seasonal temperature variability remains understudied and is set to be modified by climate change. Most temperature studies consist of short-term exposures using time series data. These studies are limited by an inability to observe longer-term trends in temperature and mortality along with regional adaptation and short-term mortality displacement. Cohort studies can address these limitations, but no countrywide cohort study has been conducted to study these long-term temperature trends, especially in a cohort with differing climatic zones. Aim: We aimed to examine the association between seasonal temperature variability and mortality across the contiguous US with its five defined regional climate zones. Further, we aimed to investigate effect modification by race, poverty, population density, and green space. Methods: We carried out this analysis in the US Medicare cohort (age 65+) from 2000-2016 with 622,427,230 years of person time. Our exposure was seasonal temperature variability or the standard deviation of warm (April to September) and cold (October to March) season temperature. We used adapted difference-in-difference, quasi-experimental methods to reduce unobserved confounding at the ZIP-code level. We employed a clustering approach based on climate zone and socioeconomic factors, combined with a meta-analysis. Results: We showed that for every 1 °C increase in warm and cold season temperature variability, the mortality rate increased by 1.54% and 0.69% in over 65s, respectively. Over 270 million Americans are exposed to at least one form of seasonal temperature variability. These effects were most pronounced in low-income individuals, low green space residents, and those in lower population density areas such as rural locations. Further, these effects were not modified by one's climate zone of residency. Conclusion: Seasonal temperature variability was associated with an increased risk of mortality in over 65s across the US. Keywords: Mortality; Temperature Variability; Climate Change; Medicare Cohort; Race; Poverty; Green Space.