Abstract: This paper merges weekly average temperature data from the National Oceanic Atmospheric Administration’s National Climatic Data Center (NDCC) with Medicare claims data in order to analyze the impact of high and low temperatures on mortality, onset of new chronic conditions, and hospital spending. We find a U-shaped pattern to mortality, with high and low temperature weeks exhibiting higher mortality than a 70 degree reference week. The marginal deaths in extreme weeks are healthier than the typical person who dies in the reference week, but less healthy than the population as a whole. We find some evidence of short-term mortality displacement at moderately high temperatures, but not at extremely high temperatures and not for low temperatures, where the impact tends to grow over time. High temperatures are associated with increased onset of new chronic conditions, while low temperatures are associated with lower onset, although this result may be driven by differences in the propensity to access the health care system. In the short run, high temperatures are associated with increased Medicare hospital spending and lower temperatures are associated with decreased hospital spending, although over a one month period both high- and low-temperature weeks are associated with increased hospital spending. Using conventional figures for the value of a life year lost, we find the additional healthcare spending induced by a hot week to be about 3-6% of the mortality cost.
Heutel, Garth; Miller, Nolan H.; and Molitor, David, "The Impact of Air Temperature on Mortality, Morbidity, and Healthcare Cost in the Medicare Population" (2014). Economics Faculty Publications. 10.