Future Mortality due to extreme heat in an aging population of Japan

摘要

Background Population aging is anticipated to have great influence on the future estimation of heat-related illnesses since the proportion of elderly susceptible to heat will expand. In assessing the health impact of climate change, very few studies had looked into the uncertainty arising from such demographic change. Aims In this study, we estimated the number of deaths attributed to heat in the Kanto region of Japan under different population growth assumptions. Methods Daily weather data for August of current (2004) and future period (2030) were derived from the Nonhydrostatic Icosahedral Atmospheric Model (NICAM). Extreme heat was defined by the 95th percentile of annual daily mean temperature for the Kanto region covering Saitama, Chiba, Tokyo, and Kanagawa. The following population forecast scenarios for 2030 were assumed: (A) population and age distribution remained constant as in 2005; (B) population projection with similar age distribution as 2005; (C) population projection with aging. The number of deaths due to heat for the 65 year-old and above was estimated for daily temperature changes above the percentile cut-off in each population grid of about 1×1 kilometer. Results When aging was factored in the 2030 population forecast, the estimated proportion of heat-related deaths among the elderly aged 65 years and over tripled from 0.6 to 1.8% of non-injury deaths for August 2004 and 2030 respectively. Assuming no adaptation, estimated crude number of deaths due to extreme heat under aging scenario C were 629 compared to the estimates under naieve scenario A and B of 232 and 212, respectively for August 2030. Conclusions While aging varies by country, the future health impact of weather can be underestimated without taking into account the increasing vulnerability of population due to aging.