Climate and mortality in Australia: retrospective study, 1979-1990, and predicted impacts in five major cities in 2030

Abstract

Quantitative assessment of climatic and environmental health risks is necessary because changes in climate are expected. We therefore aimed to quantify the relationship between climatic extremes and mortality in the 5 largest Australian cities during the period [1979][1980][1981][1982][1983][1984][1985][1986][1987][1988][1989][1990]. We then applied the relationship determined between recent climatic conditions and mortality to scenarios for climate and demographic change, to predict potential impacts on public health in the cities in the year 2030. Data on mortality, denominator population and climate were obtained. The expected numbers of deaths per day in each city were calculated. Observed daily deaths were compared with expected rates according to temperature thresholds. Mortality was also examined in association with temporal synoptic indices (TSI) of climate, developed by principal component and cluster analysis. According to observed-expected threshold analyses, for the 5 cities combined, the annual mean excess of deaths attributable to temperature over the period [1979][1980][1981][1982][1983][1984][1985][1986][1987][1988][1989][1990] was 175 for the 28°C threshold. This sum of statistically significant differences from the 5 cities was the greatest excess found in association with any threshold considered in the range of temperatures that occur. Excess mortality for the hottest days in summer was greater than for the coldest days in winter. Temperature-mortality relationships were little modified by socio-economic status. TSI analyses produced similar results: using this method, the climate-attributable mortality in the 5 cities was approximately 160 deaths yr -1 , although this number was evenly distributed across summer and winter. Persons in the group aged 65 yr and older were the most vulnerable. After allowing for increases in population, and combining all age groups, the synoptic method showed a 10% reduction in mortality in the year 2030. We conclude that the 5 largest Australian cities exhibit climate-attributable mortality in both summer and winter. Given the scenarios of regional warming during the next 3 decades, the expected changes in mortality due to direct climatic effects in these major coastal Australian cities are minor.

Long-term projections and acclimatization scenarios of temperature-related mortality in Europe

“…Because of the extremely large differences in population between regions, they were grouped in 54 larger and more homogeneously populated areas. A transfer function (also referred here as temperature/mortality or T/M relationship) was estimated for the period with available observations (1998)(1999)(2000)(2001)(2002)(2003), and used to convert simulated daily temperatures from other periods into daily mortality projections 15,26,27 . The transfer function is separately adjusted to data above (warm tail) and below (cold tail) the temperature of minimum mortality, or comfort temperature 15 .…”

Seasonality of climate-human mortality relationships in US cities and impacts of climate change

“…For example, Guest et al (1999) noted that future mortality impacts in Australia's 5 largest cities would be small by 2030. McGregor et al (2004) found that years with strong temperature seasonality also exhibited elevated mortality from ischaemic heart disease in 5 English counties.…”

Relationships between the seasonality of temperature and ischaemic heart disease mortality: implications for climate based health forecasting

“…Given the general sensitivity of death and illness to temperature (West 1973, Bull & Morton 1975, Auliciems & Frost 1989, Auliciems & Skinner 1989, McKee 1990, Frost et al 1992, Frost 1993, Guest et al 1999, Laschewski & Jendritzky 2002, diseases such as IHD may be potentially predictable. The purpose of this paper is therefore to undertake an exploratory analysis of the statistical association between the seasonality of temperature and IHD mortality for 5 English counties as a basis for investigating the potential predictability of the magnitude and timing of the annual IHD mortality peak.…”