Effects of Weather on Mortality from Major Causes Using Daily Number of Deaths

Abstract

Effects of weather conditions on daily number of deaths for nine major causes in the Tokyo Metropolis during the period of 1 January in 1980 to 31 December in 1981 were studied in this report.
1. According to correlation analyses by season, the deaths for almost all causes correlated significantly with Tmin (minimum air temperature) negatively in spring and positively in summer, particularly in an elder age group (70-yr) . However, these deaths did not correlate significantly with weather conditions in winter, the coldest seasons as described by some other reports.
2. In spring, number of deaths for cerebrovascular diseases in the elder group increased with a rise in Tmin in the daily average Tmin for the previous seven days. This result indicated that the aged patients for circulatory diseases should be carefully attended in spring in which acute and large weather changes occur.
3. Daily numbers of deaths compared between perio ds from March to June with a rise in temperature and from September to December with a fall in temperature were not equivalent for the same temperature. Particularly for pneumonia-bronchitis, more deaths were observed during the former period than the latter period and significant differences between number of deaths during the two periods were found in the wide range.
4. In order to examine whether there was a time lag between Tmin and daily number of deaths, a correlation analysis was conducted for the seven days which followed the death. Deaths in summer had the most significant positive correlation in the day, i.e. there was no time lag, but with respect to deaths in the other seasons, it was estimated that there was a time lag of some days. Although death in winter were inferred to be unlikely to be associated with Tmin above mentioned, it was concluded in this study that they had significant association with Tmin in the two or three days before. In addition, cumulative effects of hot temperature on increased number of deaths were found for female, while cumulative effects of cold temperature on deaths were slightly found for male.
5. Although the proportion of home-deaths, which may be affected by weather conditions near open air, to all death is not so large, it was estimated that data on all death should be used to analyze relationship between daily number of deaths and weather conditions, because daily variations of both deaths were closely associated with each other and the number of deaths in all places has smaller random flucturations, than those of the home-deaths.