OBJECTIVE: COPD increases morbidity and decreases life expectancy. Airway obstruction as measured by the forced expiratory volume in one second (FEV1) is used most commonly to measure the extent of airways obstruction but the marginal impact of FEV1 in decreasing survival is unclear. METHODS: This was a retrospective record linkage study using data from a region of Wales, UK, with a population of around 425,000 people. Data were probability matched from a number of data sources: hospital utilisation data from local hospitals, lung function data from the Respiratory Medicine Department and mortality data from the Office of National Statistics. Individual data sources were available spanning a period of 14 years to the present. Survival was characterised from the first recorded FEV1 measurement. Multivariate analysis techniques were used to standardise for potentially confounding factors such as age, sex, morbidity, BMI and smoking status. RESULTS: Among 33,357 subjects with data describing their FEV1 status, we identified 2,326 patients with COPD. Mean and median (SD and IQR) FEV1 values at first measurement were 1.04 litres and 0.89 litres (0.64 and 1.15-2.62), respectively. Mean survival at one year and at five years was 95.5% and 74.7%, respectively. In univariate analysis by FEV1 quartile (1st quartile was the lowest FEV1 value) survival at five years was 71.2%, 82.9%, 80.4% and 85.0%. After standardisation for age, sex, morbidity, smoking status and BMI using a Cox's model, the hazard ratios for the 1st, 2nd and 3rd quartile by comparison to the 4th were 2.3, 1.7 and 1.4, respectively. CONCLUSION: Airways obstruction (FEV1) was an important determinant of survival in COPD following standardisation for other potentially confounding factors.