Daily myocardial infarct deaths for the period 1983 to 1991 were obtained from Statistics Canada for the conurbations of Montreal, Toronto, Winnipeg and Vancouver. The deaths were subdivided by gender and by age, above and below 65 years. Daily maximum and minimum temperatures were made availble for the same period.
There were relatively few heart attack deaths amongst those less than 65 years of age in all cities, and there was no relationship with temperature for that group. They were discarded from further analysis.
The conurbations ranged in population from 3,427,165 in Toronto to 625,305 for Winnipeg, while the proportion of those over age 65 varied from less than four percent in Toronto to over five percent in Vancouver. The range of mean daily temperature varied from 10C to 26C in Vancouver to -34C to 32C in Winnipeg. Likewise, the modal temperatures varied greatly from 7C in Vancouver to 22C in Toronto.
The immediate effect of temperature was determined by calculating the mean number of deaths per degree of average daily temperature. Longer term thermal stress was investigated through 15-day running means of death and temperature data. In all cities and at both levels of analysis, a minimum death temperature could be identified for each gender and numbers of deaths increased as temperature fell below that value. However, the relationships at extremes of both heat and cold differed between cities and genders.
Immediate effects. The temperatures at which the least deaths occurred are higher for men than for women in all cities. Male values range from 20C in Montreal to 28C in Toronto with comparable female values of 15C and 26C. At very low temperatures, the apparent avoidance of cold by women is strongly marked in Montreal, but is less evident elsewhere, where both the highest and the lowest death rates for both genders are found at low temperatures. Such extreme variation is attributed to small numbers of days with very low temperatures.
The longer term 15-day average values demonstrate the same increase in deaths as temperatures fall below optimum values. However, in extreme heat, female death rates rose, while male rates stayed low, except in Winnipeg where the gender pattern was reversed. In Vancouver and Toronto, cities without extremely cold winter conditions, the death rates for both genders were highest in the coldest conditions. In Montreal, by contrast, female death rates fell to their lowest level in extreme cold, while in Winnipeg, the coldest city, the rates for both genders fell on the coldest days.
Comparative figures are presented for cities in Australia, Iran and New Zealand.
The results of this study demonstrate a degree of adaption to the local thermal regime by the elderly that has implications for the possible results of global warming.