BACKGROUND: Early cardiopulmonary resuscitation (CPR) by bystanders prior to the arrival of the rescue team has been shown to be associated with increased survival after out-of-hospital cardiac arrest. The aim of this survey was to evaluate the impact on survival of no bystander CPR, lay bystander CPR and professional bystander CPR. METHODS: Patients suffering an out-of-hospital cardiac arrest in Sweden between 1990 and 2002 who were given CPR and were not witnessed by the ambulance crew were included. RESULTS: In all, 29,711 patients were included, 36% of whom received bystander CPR prior to the arrival of the rescue team. Among the latter, 72% received CPR from lay people and 28% from professionals. Survival to 1 month was 2.2% among those who received no bystander CPR, 4.9% among those who received bystander CPR from lay people (p<0.0001) and 9.2% among those who received bystander CPR from professionals (p<0.0001 compared with bystander CPR by lay people). In a multivariate analysis, lay bystander CPR was associated with improved survival compared to no bystander CPR (OR: 2.04; 95% CI: 1.72-2.42), and professional bystander CPR was associated with improved survival compared to lay bystander CPR (OR: 1.37; 95% CI: 1.12-1.67). CONCLUSION: Among patients suffering an out-of-hospital cardiac arrest, bystander CPR by lay persons (excluding health care professionals) is associated with an increased chance of survival. Furthermore, there is a distinction between lay persons and health care providers; survival is higher when the latter perform bystander CPR. However, these results may not be explained by differences in the quality of CPR.