Abstract Background Dual antiplatelet therapy (DAPT), aspirin plus a P2Y12-i (clopidorel, prasugrel or ticagrelor), is recommended for one year after myocardial infarction (MI) for secondary prevention of cardiovascular disease (SP-CVD). Beyond one year maintaining DAPT is controversial. Purpose To compare the 3-year risk of a composite of MI, ischemic stroke (IS), major bleeding (MB) and death between DAPT and single antiplatelet therapy with aspirin (SAPT) beyond one year after MI. Methods All adults hospitalized in 2013 or 2014 for acute MI (trigger event) with intensive care unit stay were identified in the French SDNS nationwide claims database. Patients who survived at least one year without MI or MB, and with a DAPT medication possession ratio (MPR) ≥80% were included in a cohort study. All patients were followed for 3 years after the index date (defined 365 days after the MI trigger event), except right-censored observations for those who died or discontinued aspirin with a 60-day grace period. The 3-year hazard ratios (HR [95% CI]) were estimated using Cox proportional hazards risk model for outcomes including death, and Fine and Gray competing risks model for non-fatal outcomes, with a time-dependent variable for DAPT-SAPT exposure, and adjusted on a high-dimensional disease risk score (hdDRS) plus time dependent variables for SP-CVD drugs, oral antidiabetics, insulin, anticoagulants, NSAIDs, corticoids and proton pump inhibitors. HdDRS were estimated for the composite outcome, a composite of ischemic outcomes, and MB alone, and variables were selected using a combination of Principal Component Analysis and Lasso regression. Results From the 105,080 adults admitted in intensive care units for acute MI in 2013 or 2014, 53,399 were included in the cohort. The most common reasons for non-inclusion were death (n=12,012) and a DAPT MPR <80% (n=25,000). At index date, mean age was 65 years, with 74.6% men, 21.8% diabetes, 9.4% heart failure, 5.6% peripheral arterial disease, 72.2% with DAPT score ≤2, 61.9% Charlson index ≤1; 79.2% had a STEMI trigger event and 82.6% had cardiac revascularization (PCI 98.6%). P2Y12-i used at least once from the trigger event to the index date were clopidogrel (41.5%), ticagrelor (41.1%) and prasugrel (26.2%). Follow-up was 111,770 person-years and 4,268 composite outcomes were recorded. The 3-year HR of DAPT compared to SAPT was 1.21 [1.13–1.30] for the composite of MI, IS, MB and death, 1.22 [1.07–1.38] for MI, 0.98 [0.80–1.20] for IS, 1.89 [1.55–2.30] for MB and 1.16 [1.06–1.27] for death. Conclusions In this nationwide real-life population-based study in France, DAPT maintained beyond one year after MI is significantly associated with increased harm compared to SAPT with increased risks of 21% (IC95% [13–30]) for the composite of MI, IS, MB and death (net clinical benefit), 22% [7–38] for MI, 89% [55–130] for MB, 16% [6–27] for death, and no difference for IS. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): French Ministry of Health (PHRCN-18-0745)