[Impact of heat waves on daily mortality in distinct age groups]

To establish whether the daily maximum temperature during heat waves beyond which daily mortality increases differs in distinct age groups and to quantify the mortality increment by these age groups.

The dependent variable was daily mortality from all causes except for accidents in the city of Madrid from 1986 to 1997. The age groups analyzed were the following: 0-10 years, 18-44 years, 45-64 years, 65-74 years, and older than 75 years. Data on the daily maximum temperature in the summer months was used. Trend and seasonalities were adequately controlled for, as were air pollutant concentrations. Autoregressive moving average (ARIMA) and Poisson regression models were used.

In all age groups except that aged less than 10 years old, there was an association between extremely high temperatures and mortality. Independently of age group, the temperature threshold above which mortality increased coincided with the 95th percentile of the maximum temperature data series for the summer months, which in Madrid is 36.5 degrees C. The daily mortality increment for each degree above this threshold was practically constant between the ages of 18 to 64 years but markedly increased in persons aged more than 65 years old.

Mortality is associated with high temperatures in people younger than 64 years old and therefore prevention plans should not be focused exclusively on the elderly.