Background: Asthma is associated with fixed airflow obstruction and an increased risk of death. The aim of this study was to relate asthma mortality and airflow obstruction to severity of asthma in a cohort of patients with chronic asthma studied over a 17-year period.Methods: In 1980, asthma severity based on symptoms, asthma duration, treatment and smoking habits were assessed and lung function was measured after maximal therapy in 89 patients. In 1997, mortality was recorded and 44 of 60 patients known to be alive were restudied.Results: Eighteen patients had died: eight deaths were associated with asthma (seven occurred before and one after 1990). The risk of death was higher with decreased forced expiratory volume in 1 s (FEV1), increased FEV1 variability, age and treatment requirements but not symptom severity, at initial study. In the patients restudied, asthma severity and FEV1 variability decreased whereas the dose of inhaled corticosteriods increased 2.8-fold. Highest FEV1 was negatively related to treatment score and smoking history at initial study, but not at follow up and was < 80% predicted in 19 (43%) patients, 11 of whom had never smoked. The mean (+/- standard deviation) decrement in FEV1 was 32 +/- 24 mL/year and correlated positively with FEV1 at initial study, smoking history, age of onset of asthma and treatment requirements at follow up.Conclusion: In this cohort study, asthma was associated with chronic airflow obstruction and that with increased risk of mortality. Symptoms and mortality risk improved in association with increased reported use of inhaled corticosteroids; however, there was ongoing chronic airflow obstruction.