When the number of deaths (excluding deaths due to external causes) of Kyoto citizens which occurred in Kyoto city every day from July 18 to September 10, 1971 and from January 7 to March 1, 1972 were investigated, it was found that the number of deaths fluctuated considerably from day to day both in summer and in winter (Fig. 1)
As variables indicating meteorological conditions,
X1: mean of average air temperature on the day and the previous day of the death (mean of two days),
X2: mean of average air temperature on the day and the preceeding six days of the death (mean of seven days),
X3,
X4: average humidity (mean of two days and mean of seven days),
X5,
X6: average wind velocity (mean of two days and mean of seven days),
X7,
X8: average SO
2 concentration in air (mean of two days and mean of seven days),
X9,
X10: maximum (in summer) or minimum (in winter) air temperature (mean of two days and mean of seven days),
X11: difference of maximum and minimum air temperature (mean of two days),
X12: difference of mean air temperature between the day and the previous day of the death were adopted, and the correlations between the daily number of deaths and each of the above variables were calculated (Table 4) In summer
X1,
X2,
X9,
X10 and
X11 were positively and
X3 was negatively correlated with the daily number of deaths, but in winter no variables significantly correlated with the daily number of deaths. Furthermore, the multiple regression equation of the daily number of deaths on eight predictor variables (
X1-
X8) was obtained (table 5). However, the multiple correlation coefficient in summer was only slightly greater than the correlation coefficient between the daily number of deaths and average air temperature (mean of two days) and that in winter was not statistically significant.
Based on these findings, the extent and the mode of the effect of meteorological conditions, especially of air temperature on death or human health in summer and in winter were discussed.
View full abstract