Heatwave and elderly mortality: An evaluation of death burden and health costs considering short-term mortality displacement

BACKGROUND: A heatwave can be a devastating natural disaster to human health, and elderly people are particularly vulnerable. With the continuing rise in earth's surface temperature alongside the world's aging population, research on the mortality burden of heatwave for the older population remains relatively sparse. The potential magnitude of benefits of averting such deaths may be considerable. OBJECTIVES: This paper examined the short-term mortality displacement (or "harvesting") of heatwave, characterized the heatwave-mortality relationship, and estimated death burden and health costs attributable to heatwave among the elderly in Australia. METHODS: We collected daily data on the temperature and deaths of people aged ≥75 years in the five largest cities of Australia (Sydney, Melbourne, Brisbane, Perth and Adelaide), totaling 368,767 deaths in different periods between 1988 and 2011. A total of 15-tiered heatwave definitions, based on intensity (95th to 99th percentiles of temperature distribution) and duration (two or more consecutive days), were used to quantify heatwave effects, using time-series regression and random-effects meta-analysis. We calculated attributable deaths for each city and by different types of heatwave. Potential economic benefits in monetary terms were also estimated, considering that heat-related deaths are avoidable. RESULTS: Among the Australian elderly population, we found significant associations between heatwave and deaths, with raised mortality immediately in the first few days followed by lower-than-expected mortality. In general, heatwave was associated with an average death increase of 28% (95% confidence interval: 15% to 42%), and greater increases were mostly observed for more intense heatwaves across multiple megacities. During the study period, there were dozens to hundreds of deaths attributable to heatwave for each city, equating to an economic loss of several million Australian dollars every year. Although the estimated attributable deaths varied by heatwave intensity and duration, the pattern was not consistent across cities. CONCLUSIONS: Heatwave caused harvesting effects on mortality in the elderly population of Australia, and contributed to a substantial amount of death burden and indirect financial costs. To lessen the health impacts of heatwave in the affected regions, effective heatwave early warning systems and interventions targeted at the elderly population could be beneficial, both now and in the future.