Objectives. To survey and compare secondary prevention measures in diabetic and non-diabetic patients following myocardial infarction (MI), Design. Follow-up of a cohort of patients who suffered their first: MI 1 year previously. Setting. Three district general hospitals. Main outcome measures. Review 1 year post-MI for signs of left ventricular failure (LVF), serum cholesterol, smelting status, weight, blood pressure and glycaemic control. Assessment of appropriate treatment with aspirin, beta-blockers, angiotensin-converting enzyme (ACE) inhibitors and lipid-lowering therapy before discharge and at least 1 year post-MI. Results, A total of 189 non-diabetic and 86 diabetic patients were studied. Most patients received beta-blockers and aspirin appropriately, and most gave up smoking. In non-diabetic subjects, cholesterol fell significantly (P < 0.05), as did the proportion of patients with cholesterol > 5.5 mmol L-1 (P < 0.05), whereas cholesterol did not fall significantly in diabetic subjects, due to a lower proportion of patients being on lipid-lowering therapy (27.5 vs. 37.9%). A higher proportion of non-diabetic patients with LVF were treated with ACE inhibitors compared with diabetic subjects (73.6 vs. 61.0%), Glycaemic control did not improve in the diabetic subjects. Conclusions. Patients with diabetes do not receive optimal secondary prevention measures compared with their non-diabetic counterparts. This issue needs to be addressed by all units dealing with patients with diabetes in order to reduce the mortality and morbidity of MI in such patients.