L Filleul1, I Baldi, J-F Dartigues, J-F Tessier. 1. Laboratoire Santé Travail Environnement, Université de Bordeaux 2, 146 rue Léo Saignat, 33076 Bordeaux, France. laurent.filleul@isped.u-bordeaux2.fr
Abstract
BACKGROUND: Air pollution has been linked to increased rates of mortality, but little is known about individual characteristics related to the increase in risk. AIMS: To examine short term effects of air pollution on daily mortality in elderly people in Bordeaux and compare characteristics of subjects > or =65 years old who died with levels of particulate air pollution. METHODS: Daily mortality data from Bordeaux were used to determine the core model of mortality for the period 1988-97. The air pollution indicator was regressed on the core model of mortality, allowing control of the main effect modifiers and confounding factors of air pollution on the same day. The residual series of this regression model was classified from the lowest to the highest to determine "low level days" and "high level days". A sample of 1469 elderly people in a French cohort study were studied. RESULTS: From 1988 to 1997, 543 subjects died; 55 deaths were during days with low air pollution and 51 during days with high air pollution. Only gender differed significantly according to both types of days, with a proportion of women significantly higher in high air pollution days than men. After adjustment for smoking habits, the odds ratio was 5.2 for women. CONCLUSION: The risk of mortality between women and men was determined on days with "high air pollution levels" above the 50-90th centiles compared to below the 10th centile. No clear pattern was observed between days with low levels of air pollution and the different centiles of exposure.
BACKGROUND: Air pollution has been linked to increased rates of mortality, but little is known about individual characteristics related to the increase in risk. AIMS: To examine short term effects of air pollution on daily mortality in elderly people in Bordeaux and compare characteristics of subjects > or =65 years old who died with levels of particulate air pollution. METHODS: Daily mortality data from Bordeaux were used to determine the core model of mortality for the period 1988-97. The air pollution indicator was regressed on the core model of mortality, allowing control of the main effect modifiers and confounding factors of air pollution on the same day. The residual series of this regression model was classified from the lowest to the highest to determine "low level days" and "high level days". A sample of 1469 elderly people in a French cohort study were studied. RESULTS: From 1988 to 1997, 543 subjects died; 55 deaths were during days with low air pollution and 51 during days with high air pollution. Only gender differed significantly according to both types of days, with a proportion of women significantly higher in high air pollution days than men. After adjustment for smoking habits, the odds ratio was 5.2 for women. CONCLUSION: The risk of mortality between women and men was determined on days with "high air pollution levels" above the 50-90th centiles compared to below the 10th centile. No clear pattern was observed between days with low levels of air pollution and the different centiles of exposure.