资源环境科技发展态势分析平台

Health risk assessments for extreme heat and the design of corresponding interventions can be enhanced with more information regarding causal drivers of year-to-year variability in adverse outcomes. Summer 2016 was a record-setting year in terms of summer heat and its impacts on health in Maricopa County, Arizona, USA. The month of June was the warmest observed in the county and the six-month warm season spanning May through October was the fourth warmest. In the same year, a record number of heat-associated deaths was reported by the heat surveillance program run by the county health department. We analyzed the time series of heat-associated deaths to quantify the extent to which the unprecedented death count in 2016 was driven by anomalous weather. We first estimated the historical association between temperature and heat-associated deaths for the time period 2006-2015 using a time series regression model. Subsequently, we used the model to generate predictions of daily heat-associated deaths in 2016 based on the observed weather. We found no evidence that the unusually high number of heat-associated deaths observed in Maricopa County in 2016 was related to observed meteorological conditions. Regardless of the exposure variable or model parameterization chosen, the prediction for 2016 fell near or below the historical average number of heat-associated deaths. If the conventional methods for estimating the temperature-mortality association are reasonably approximating a causal relationship, factors other than the weather were mostly responsible for the surge in deaths in 2016. These findings highlight the importance of non-meteorological factors as drivers of temporal variability in the health burden associated with heat, which have generally not been included in quantitative retrospective or prospective studies. Further, they highlight a shortcoming in preparedness and response efforts for heat in the study setting that should be diagnosed and addressed as soon as possible.