OBJECTIVES Few studies have assessed the real-life impact of secondary prevention drugs on all-cause mortality post-myocardial infarction (MI), especially in countries with low incidence of MI. The objective of this interim analysis after 3.5-year of follow-up was to assess the real-life all-cause mortality impact of drugs reimbursed for MI secondary prevention in France: acetylsalicylic acid (ASA), anti-platelet agents (APA), beta-blockers (ß-), angiotensin converting enzyme inhibitors (ACEI), statins, and omega-3 supplementation (Om3). METHODS Cohort study of patients with recent (≤3 months) acute MI included by hospital and non-hospital cardiologists, with 6-year follow-up. Vital status was obtained from the National death registry, and failing that by patient/relatives/physicians investigation. Drug exposure was defined using both physician and patient reports at inclusion. Cox proportional hazard model was used to estimate for each drug, mortality hazard ratio (HR) of exposed versus non exposed patients, adjusted for gender, age, cardiovascular risk factors, other MI prevention drugs, and propensity score to be exposed at inclusion. RESULTS Between May 2006 and June 2009, 596 physicians included 5538 patients: mean age 62.1 years, 77.6% male, 9.6% current smokers, 14.5% diabetic, 44.6% hypercholesterolemic, 43.6% hypertensive, 8.2% with LVEF <40%. At inclusion, 97.5% were exposed to ASA, 91.0% to APA, 89.7% to ß-, 71.1% to ACEI, 92.0% to statins, and 15.7% to Om3. The 3.5-year mortality was 7.8% (95%CI [7.1%-8.5%]) with an incidence rate of 23.2 per 1000 patient-years. Adjusted HR were: 0.98 [0.60-1.61] for ASA, 0.86 [0.60-1.24] for APA, 0.84 [0.63-1.11] for ß-, 0.80 [0.61-1.03] for ACEI, 0.67 [0.45-1.00] for statins, and 0.82 [0.58-1.16] for Om3. CONCLUSIONS The 3.5 year interim all-cause real-life death reduction point estimates were close to those of large randomized controlled trials, except for ASA, for which almost all patients were exposed. The study's statistical power will be sufficient to confirm or not these trends at the final 6-year analysis.