OBJECTIVES: The benefits of initiating statins in the elderly remains debated. We evaluated the effects of initiating statins in the elderly, according to cardiovascular risk. METHODS: This population-based cohort study used data of the representative sample of the French healthcare system database for the 2008-2015 period. New users of statins, aged 75 years and above were dynamically included in the cohort and matched 1:1 to statin non-users on age, gender, numbers of different drugs dispensed and medical consultations, and cardiovascular history. Patients were classified in three cardiovascular risk groups: secondary prevention (history of coronary heart disease), primary prevention with modifiable risk factors (diabetes or cardiovascular medications), and primary prevention without modifiable risk factor (none of the above). Effect of cumulative use of statins on occurrence of acute coronary syndrome or all-cause death was analyzed by using multivariable time-dependent Cox models stratified on cardiovascular risk at inclusion. RESULTS: Among the 7284 patients included, median follow-up was 4.7 years. Cumulative use of statins was associated with a lower risk of outcomes in the primary prevention with modifiable risk factors group (adjusted hazard ratio 0.93 per year of use; 95% confidence interval 0.89-0.96; p<0.01), and in the secondary prevention group (0.75; 0.63-0.90; p<0.01) but not in the primary prevention without modifiable risk factor group (1.01; 0.86-1.18; p=0.92). CONCLUSIONS: Statin treatment was not associated with a reduction in acute coronary syndrome or all-cause death in elderly without modifiable cardiovascular risk factor treated in primary prevention.< Réduire