OBJECTIVE: To evaluate healthcare and outcomes of disease management programmes (DMPs) for patients with coronary heart disease (CHD) in primary care, and to assess selection of enrollment for these programmes. METHODS: A cross-sectional survey of 2330 statutorily insured patients with a history of acute myocardial infarction (AMI) was performed in 2006 by the population-based KORA Myocardial Infarction Register from the region of Augsburg, Germany. Patients enrolled in DMP-CHDs receive evidence-based care, with patients not enrolled receiving standard care. To control for selection bias, a propensity score approach was used. RESULTS: Main factors influencing DMP participation were age (OR 0.98, 95% CI 0.96-0.99), diabetes (OR 1.56, CI 1.25-1.95) and time since last heart attack (OR 0.98, CI 0.95-0.99). Significantly more patients enrolled in DMP-CHDs stated that they received medical counselling for smoking (OR 3.77, CI 1.07-13.34), nutrition (OR 2.15, 1.69-2.74) and for physical activity (OR 2.58, 1.99-3.35). Furthermore, prescription of statins (OR 1.58, CI 1.24-2.00), antiplatelets (OR 1.96, CI 1.43-2.69) and beta-blockers (not significant) were higher in the DMP group. With respect to outcomes, we did not see relevant differences in quality of life and body mass index, and only a minor reduction in smoking. CONCLUSIONS: Enrollment into DMPs for CHD exhibits systematic selection effects. Participants tend to experience--at least on a short to medium term and for AMI patients--better quality of healthcare services. However, since DMP-CHDs were initiated only 2 years ago, we were unable to identify significant improvements in health outcomes. Only the reduction in smoking provides a first indication of better quality outcomes following DMP-CHD. Thus, policy-makers must provide appropriate incentives to sickness funds and physicians in order to ensure initiation and continuation of high quality DMPs.