The population of Budapest is exposed to high ambient temperatures and recurrent heat-waves. In 2000, a severe heat-wave affected many countries in the Balkans. We have investigated the impact of temperature and heat episodes on mortality in Budapest in order to identify the population groups that are the most vulnerable.
Methods
Daily time-series data were obtained for all deaths of Budapest residents between 1993 to 2000 inclusive. The relationship of mortality with temperature was examined using auto-regressive Poisson models with adjustment for season, relative humidity, ozone and particulate air pollution, day of week, and public holidays. This was conducted separately by cause of death, age-group, sex, and area of residence. In addition, we identified specific heat-wave events (defined as a minimum three-day period when mean temperature was above the 99th percentile) and estimated the attributable mortality for each heat episode as the difference between the observed deaths and the "expected" mortality from the regression models.
Results
An increase in mortality was observed once temperatures reached a certain threshold level. This threshold, and the subsequent linear heat effect, differed by age, sex and area of residence. Six specific heat episodes were identified: 28 June to 1 July 1994 (4 days); 30 July to 8 August 1994 (10 days); 22 July to 25 July 1988 (4 days), 3 August to 5 August 1998 (3 days); 13 June to 15 June 2000 (3 days); and 20 August to 22 August 2000 (3 days). During the heat-waves, daily maximum temperatures reached, on average, 34ÂșC. Excess mortality (all cause) was observed during all the episodes: 22% (June 94), 12% (Aug 94), 24% (Jul 98), 26% (Aug 98), 52% (Jun 00), and 14% (Aug 2000). In general, the impact was strongest during episodes occurring early on in the summer months rather than later. Different approaches in the definition of a heat-wave will be discussed.
Conclusions
Hot weather and heat-waves contribute a significant burden on mortality in Budapest, particularly events that occur early on in summer. The impact of subsequent heat-waves is likely to have been diminished by both the loss of susceptibles and the increased acclimatisation of the population to hot weather. In the light of changes in the worlds climate, there is a need to improve public health advice in order to reduce the burden of heat-waves on human health.