Mortality of elderly subjects with self-reported asthma in a French cohort, 1991-1996.

The objective of this study is to describe the mortality of subjects with self-reported asthma aged 65 and over and to determine risk factors. PAQUID (Personnes Agées QUID) is a prospective cohort of 3777 elderly people, living at home in the South-west of France. The study of subjects with self-reported asthma started at 3 year follow-up by using three epidemiological questions: 'Did you have an attack of wheezing that made you feel short of breath in the last 12 past months?', 'Have you ever had asthma?', 'Did you have at least one asthma attack in the previous 12 months?'. Data on mortality were available at 8-year follow-up. Among the 2348 subjects who responded to these questions, 206 reported symptoms of asthma. Two groups were defined as 'pure asthma' and 'associated asthma' taking into account chronic bronchitis and smoking. Besides a gender difference, there was no statistical difference between those two groups so they were combined in a single group for further analysis. During the 5-year study period, 29.1% of subjects with self-reported asthma vs. 23.8% nonasthmatics died. A Cox model with delayed entry was used to calculate mortality rates. Interaction terms between each of the factors related to mortality and asthmatic status were analyzed to determine risk factors of mortality. Subjects with self-reported asthma had a higher risk of death than nonasthmatics (relative risk (RR): 1.49; 95% confidence interval (CI) 1.1-1.9; p = 0.009). This risk was significant (RR: 1.4; 95% CI: 1.05-1.8; p = 0.02) even after adjustment on morbidity variables. However no specific mortality risk factors were found for subjects with self-reported asthma.