Study objectives: To determine the effectiveness of influenza vaccination on influenza-related acute respiratory illness (ARI) and overall ARI in patients with COPD, and its relationship to the degree of airflow obstruction.Design: Stratified, randomized, double-blind, placebo-controlled trial.Setting: From June 1997 to November 1998 at a single university hospital.Patients and interventions: One hundred twenty-five patients with COPD were stratified based on their FEV1 as having mild, moderate, and severe COPD. Within each group, they were randomized to the vaccine group (62 patients who received purified, trivalent, split-virus vaccine) or the placebo group (63 patients).Measurements: The number of episodes and severity of total ARI, classified as outpatient treatment, hospitalization, and requirement of mechanical ventilation; and the number of episodes and severity of influenza-related ARI. Results: The incidence of influenza-related ARI was 28.1 per 100 person-years and 6.8 per 100 person-years in the placebo group and vaccine group, respectively (relative risk [RR], 0.24 [p = 0.005]; vaccine effectiveness, 76%). The incidences were 28.2, 23.8, and 31.2 per 100 person-years in the patients with mild, moderate, and severe COPD, respectively, in the placebo group, and 4.5, 13.2, and 4.6 per 100 person-years in the patients with mild, moderate, and severe COPD, respectively, in the vaccine group (RR, 0.16 [p = 0.06]; vaccine effectiveness, 84%; RR, 0.55 [p = 0.5]; vaccine effectiveness, 45%; and RR, 0.15 [p = 0.04]; vaccine effectiveness, 85%, in the patients with mild, moderate, and severe COPD, respectively). Bivariate analysis revealed that the effectiveness of influenza vaccination was not modified by the severity of COPD,comorbid diseases, age, gender, or current smoking status. There was no difference in the incidence or severity of total ARI between the placebo group and the vaccine group.Conclusions: Influenza vaccination is highly effective in the prevention of influenza-related ARI regardless of the severity of COPD. Influenza vaccination does not prevent other ARIs unrelated to influenza. The effectiveness of influenza vaccination in the prevention of overall ARI in patients with COPD will depend on how much the proportion of influenza-related ARI contributes to the incidence of total ARI. Influenza vaccination should be recommended to all patients with COPD.