Effects of Weather on Mortality from Major Causes Using Monthly Number of Deaths for a Thirty Year Period

Abstract

It is likely that effects of weather on mortality show variation according to some factors such as cause of death, place, period of year, age group and others. In this report, data on monthly mortalities from hypertensive disease (HTD), heart disease (HD), cerebrovascular disease (CD), pneumonia-bronchitis (PB), liver cirrhosis (LC) and senility (SN) were collected as well as data on the weather conditions in the Tokyo Metropolis during the period from 1952 to 1981. Data on monthly mortalities from liver disease (LD) including LC, ischemic heart disease (IHD), cerebral hemorrhage (CH), cerebral infarction (CI) were added to the previous mortalities for both sexes in two age groups (0-69 and 70-yr.) in 1980 and 1981.
As the results from correlation and regression analyses between mortality and weather conditions in the fifteen two-year periods, mortalities from HTD, HD, CD and SN were closely associated with air temperature in every period. However, correlations of mortality from PB with air temperature clearly decreased and slopes in the linear regressions were also less steep in the later periods. On the other hand, trends of the correlations and regressions for mortality from LC in the fifteen periods brought about results contrary to those from PB. According to a comparison between average death rates during periods from March to June with a rise in air temperature and average death rates from September to December with a fall in air temperature using the death rates for the ten years from 1972 to 1981, HTD, PB and SN for the former months showed higher death rates than for the latter months. “Winter risk index” was defined as an index of quantification of risk of death in winter and the following results were obtained. First, the risks of many of the previous mortalities have distinctly decreased during the period from 1952 to 1981. Second, a higher risk was generally observed in the older than in the younger age group and for mortalities from HTD and SN among these causes, but the sex with the higher risk was different for each cause of death.