Observed effect of ambient temperature on human health in Japan

Yasushi Honda, University of Tsukuba

Objective: To estimate the effect of global warming, it is necessary to evaluate the relation between ambient temperature and health. For this purpose, we analyzed Japanese data for 1972-1995.

Methods: The computerized data for population, meteorological factors and mortality were obtained from the relevant governmental offices. We have 47 prefectures in Japan , and we made this as a unit of observation. For each prefecture, we set temperatures measured in a capital city as representative. Mortality rate was calculated for each prefecture by age-category, sex, chronological era, and cause of death.

Results and Discussion: Our analysis revealed the following: (1) the majority of heat related deaths occurred among elderly subjects; (2) cardiovascular diseases and respiratory diseases were temperature dependent, whereas cancers were not; (3) the relation was shown to be V-shaped, i.e., the mortality rate was the lowest at a certain temperature (called "optimum temperature"), and it rose as the temperature became higher or lower; (4) the optimum temperature was higher for warmer prefectures; (5) for children, accidents were also related to temperature, but some of this effect was due to "opportunity difference;" during hot summer, schools are out and children experience outdoor activities such as swimming more often than in other seasons.

The difficulties in estimating the health effect of global warming in Japan includes the following: we recognized adaptation to heat, but we have no idea how fast the adaptation occurs; the V-shaped curve has gone down for the last 30 years and it is unpredictable when this tendency stops or changes.