BACKGROUND: Poor health status has been associated with morbidity and mortality in patients with COPD. To date, the impact of changes in health status on these outcomes remains unknown. AIMS: To explore the relationship of clinically relevant changes in health status with exacerbation, hospitalisation or death in patients with COPD. METHODS: Characteristics and health status (St George's Respiratory Questionnaire, SGRQ) were assessed over a period of 3 years in 2138 patients with COPD enrolled in the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) study: a longitudinal, prospective, observational study. Associations between change in health status (=4 units in SGRQ score) during year 1 and time to first exacerbation, hospitalisation and death during 2-year follow-up were assessed using Kaplan-Meier plots and log-rank test. RESULTS: 1832 (85.7%) patients (age 63.4±7.0 years, 65.4% male, FEV1 48.7±15.6% predicted) underwent assessment at baseline and 1 year. Compared with those who deteriorated, patients with improved or stable health status in year 1 have a lower likelihood of exacerbation (HR 0.78 (95% CI 0.67 to 0.89), p<0.001 and 0.84 (0.73 to 0.97), p=0.016, respectively), hospitalisation (0.72 (0.58 to 0.90), p=0.004 and 0.77 (0.62 to 0.96), p=0.023, respectively) or dying (0.61 (0.39 to 0.95), p=0.027 and 0.58 (0.37 to 0.92), p=0.019, respectively) during 2-year follow-up. This effect persisted after stratification for age and the number of exacerbations and hospitalisations during the first year of the study. CONCLUSIONS: Patients with stable or improved health status during year 1 of ECLIPSE had a lower likelihood of exacerbation, hospitalisation or dying during 2-year follow-up. Interventions that stabilise and improve health status may also improve outcomes in patients with COPD.