Gouveia et al. (2003) derived daily numbers of deaths from all causes, excepting violent deaths and deaths of babies up to one month old, from Sao Paulo, Brazil's mortality information system for the period 1991-1994, after which they analyzed them for three groups of people - those less than 15 years of age (children), those 15-64 years of age (adults), and those 65 and above (the elderly) - with respect to the effects of air temperature relative to "change points" at which warming and cooling begin to influence death rates. This work revealed that the change points for heat- and cold-induced deaths were identical, i.e., 20°C; and for each 1°C increase above this value for a given and prior day's mean air temperature, they observed a 2.6% increase in deaths from all causes in children, a 1.5% increase in deaths from all causes in adults, and a 2.5% increase in deaths from all causes in the elderly. For each 1°C decrease below the 20°C change point, however, the cold effect was much greater, with increases in deaths from all causes in children, adults and the elderly being 4.0%, 2.6% and 5.5%, respectively, which cooling-induced death rates are 54%, 73% and 120% greater than those attributable to warming for people in these three respective age groups.

A similar analysis with respect to cardiovascular-related deaths actually found no evidence of heat-induced deaths in adults but the familiar 2.6% increase for each 1°C drop in temperature below 20°C. In the elderly, however, a 1°C warming above 20°C led to a 2% increase in deaths; but a 1°C cooling below 20°C led to a 6.3% increase in deaths in this age group, or more than three times as many cardiovascular-related deaths due to cooling than to warming in the elderly.

Findings with respect to respiratory-induced deaths were similar. Death rates due to a 1°C cooling were twice as great as death rates due to a 1°C warming in adults, and 2.8 times greater in the elderly. Hence, when it comes to the bottom-line reality of living or dying, it is clear that a modest warming of the climate would be much preferred to a modest cooling or even no change at all, not only in Sao Paulo, Brazil, but in all the many other places of the world where similar studies have found essentially the same results as those described in this study.

In another study from Sao Paulo, Brazil, Sharovsky et al. (2004) investigated associations between weather (air temperature, humidity and barometric pressure), air pollution (sulfur dioxide, carbon monoxide, and inhalable particulates), and the number of daily deaths attributed to myocardial infarction from 1996 to 1998, when 12,007 fatalities were recorded. In doing so, they found "a significant association of daily temperature with deaths due to myocardial infarction (P<0.001), with the lowest mortality being observed at temperatures between 21.6 and 22.6°C." For all practical purposes, however, Sharovsky et al.'s data show little variation in death rates from 18°C to just over 25°C, the latter of which values represents the typical upper limit of observed temperature in Sao Paulo, which is located on the Tropic of Capricorn at an altitude of 800 m. As mean daily temperature drops below 18°C, however, death rates rise in essentially linear fashion to attain a value at 12°C (the typical lower limit of observed temperature in Sao Paulo) that is more than 35% greater than the minimum baseline value registered between 21.6 and 22.6°C.

Sharovsky et al. say their analysis demonstrates "a strong association between daily temperature and myocardial infarction in Sao Paulo, Brazil," which suggests that "an acclimatization of the population to the local climate occurs and that myocardial infarction deaths peak in winter not only because of absolute low temperature but possibly secondary to a decrease relative to the average annual temperature," which indeed must be true, for deaths due to heart attacks are consistently greater in winter than in summer, as they note, "across many regions of the world (Marshall et al., 1998; Douglas et al., 1991; Seto et al., 1998; Sheth et al., 1999)." Hence, it can be appreciated that the global warming of the past century, which increased minimum temperatures considerably more than maximum temperatures almost everywhere, likely prevented - or significantly forestalled - the deaths of many people around the world who otherwise would have succumbed to this deadly scourge of the human race, i.e., relatively cold temperatures.

In a final intriguing study, Hajat et al. (2005) analyzed the history of heat-wave related deaths in three cites of contrasting wealth (defined as gross national income per capita) - Delhi (India), Sao Paulo (Brazil) and London (England) - based on daily numbers of non-violent deaths derived from mortality registries for the four-year period January 1991 to December 1994, examining "time-series of daily mortality data in relation to daily ambient temperature using Poisson models and adjusting for season, relative humidity, rainfall, particulate air pollution, day of the week and public holidays," and using "unconstrained distributed lag models to identify the extent to which heat-related [death] excesses were followed by deficits," which phenomenon (mortality displacement) arises when people who die during heat waves would normally have died shortly thereafter even in the absence of the elevated warmth.

For each city, an increase in all-cause mortality was observed for same- and previous-day temperatures greater than 20°C, with excess deaths being greatest in Delhi and smallest in London. In Delhi, the excess of deaths persisted for three weeks, while in London they prevailed for only two days and were followed by deficits that led to the sum of the two effects being zero by day eleven. In Sao Paulo, as might have been expected, the pattern of deaths was intermediate between these two extremes. Summed over the course of 28 days, therefore, the risk of death associated with heat stress was 2.4% per degree greater than 20°C in Delhi, 0.8% in Sao Paulo and a negative 1.6% in London.

These findings led Hajat et al. to conclude that "populations in low-income countries where life-threatening infections are still common may have the greatest vulnerability to the effects of heat," and they say that "those most susceptible to heat are likely to remain susceptible if there is not due attention paid to infectious disease, diarrheal illness, and other major causes of early mortality in these poor populations." We agree with these conclusions and suggest that one way to provide the attention that Hajat et al. say is lacking is to not hamper the abilities of underdeveloped countries to grow their economies in the same way those of the developed world grew theirs, i.e., via the unrestrained use of fossil fuels, which can now be done much more cleanly than was the case when the world's developed countries did it.

In conclusion, it would appear that (1) unseasonable cold temperatures kill far more people than do unseasonable warm temperatures, (2) death associated with unseasonable warmth is largely but a hastening of the demise of people who would normally have died just a short time later, and (3) highly developed national economies tend to mitigate deaths at both ends of the temperature spectrum.

References
Douglas, A.S., Al-Sayer, H., Rawles, M.M. and Allan, T.M. 1991. Seasonality of disease in Kuwait. Lancet 337: 1393-1397.

Gouveia, N., Hajat, S. and Armstrong, B. 2003. Socioeconomic differentials in the temperature-mortality relationship in Sao Paulo, Brazil. International Journal of Epidemiology 32: 390-397.

Hajat, S., Armstrong, B.G., Gouveia, N. and Wilkinson, P. 2005. Mortality displacement of heat-related deaths: A comparison of Delhi, Sao Paulo, and London. Epidemiology 16: 613-620.

Marshall, R.J., Scragg, R. and Bourke, P. 1988. An analysis of the seasonal variation of coronary heart disease and respiratory disease mortality in New Zealand. International Journal of Epidemiology 17: 325-331.

Seto, T.B., Mittleman, M.A., Davis, R.B., Taira, D.A. and Kawachi, I. 1998. Seasonal variations in coronary artery disease mortality in Hawaii: observational study. British Medical Journal 316: 1946-1947.

Sharovsky, R., Cesar, L.A.M. and Ramires, J.A.F. 2004. Temperature, air pollution, and mortality from myocardial infarction in Sao Paulo, Brazil. Brazilian Journal of Medical and Biological Research 37: 1651-1657.

Sheth, T., Nair, C., Muller, J. and Yusuf, S. 1999. Increased winter mortality from acute myocardial infarction and stroke: the effect of age. Journal of the American College of Cardiology 33: 1916-1919.