Seasonality of climate-human mortality relationships in US cities and impacts of climate change

Human mortality in US cities is highest on extremely hot, humid summer days, but in general, winter-mortality rates are significantly higher than summer rates. The observed winter-dominant warming pattern, which has been linked to increasing greenhouse-gas concentrations, has led some researchers to propose future mortality decreases, while others contend that increasing heat-related mortality in summer will more than offset any winter-mortality reductions. Because winter mortality is only weakly linked to daily weather, we examine the seasonality of mortality using monthly data for 28 major US cities from 1964 to 1998. Daily all-causes mortality counts are age-standardized, aggregated monthly, and related to mean monthly 07:00 h local standard time (LST) air temperature in each city. The climate-mortality seasonality patterns are examined for spatial and temporal (decadal-scale) variability, and the impact of climate change on mortality rates is investigated after an approximation of the inherent technology/adaptation trend is removed from the monthly time series. Mortality seasonality varies little between most US cities with comparable climates. By the 1990s, monthly mortality anomalies were similar between all cities regardless of climate, suggesting there is no net mortality benefit to be derived from a location's climate. After removing the impact of long-term declining mortality rates, some statistically significant monthly climate-mortality relationships remain in most cities, with generally positive temperature-mortality relationships in summer and negative relationships in winter. Future mortality could be reduced with a winter-dominant warming but increase with pronounced summer warming. In each case, however, net future climate-related mortality rates are very low relative to the baseline death rate, indicating that climate change will have little impact in defining future mortality patterns in US cities.