AIMS: To determine whether survival after discharge following pre-hospital cardiopulmonary arrest has improved. METHODS AND RESULTS: The Heartstart Register was used to identify all 1659 patients discharged alive from Scottish hospitals during 1991-01 following pre-hospital arrest due to cardiac aetiology. The cohort was split into tertiles using year of arrest. A Cox proportional hazards model was used to determine risk of death relative to 1991-93. Patients who survived cardiopulmonary arrest in 1997-01 were less likely to die from any cause (unadjusted HR 0.60, 95% CI 0.48-0.75, P<0.001) or cardiac disease (unadjusted HR 0.50, 95% CI 0.38-0.65, P<0.001). After adjustment for case-mix, there remained significant declines in all-cause (adjusted HR 0.62, 95% CI 0.50-0.78, P<0.001) and cardiac death (adjusted HR 0.52, 95% CI 0.39-0.68, P<0.001). Clinical management had improved, with increased use of thrombolysis (47-63%, chi2 trend, P<0.001), beta-blockers (28-53%, chi2 trend, P<0.001), ACE-inhibitors (48-69%, chi2 trend, P<0.001), and anti-thrombotics (79-88%, chi2 trend, P<001). Adjustment for recorded changes in management attenuated the decline in all-cause death (adjusted HR 0.77, 95% CI 0.60-0.98, P=0.03). CONCLUSION: Survival following cardiopulmonary arrest has improved after adjusting for changes in case-mix. Better clinical management has contributed to this improvement.