OBJECTIVES: To assess the relationship between statin therapy on mortality risk in usual care management of elderly. METHODS: The present study is a population-based, observational, longitudinal study performed on 59,398 subjects aged 70 years or older, between 2000 and 2005 in Northern France (266,071 person-years). The data was extracted from a French administrative managed care database (CANSSM). Survival of incident persistent statin users was compared to survival of non lipid-lowering agent (LLA) users, thanks to multivariable Cox proportional models stratified on five-year birth cohorts adjusted on gender and hospitalizations occurring in 1999. In the same way, impact of fibrate therapy was explored. Survival of antiglaucoma treatment (AG) users versus non AG users was also assessed as a neutral comparator model. RESULTS: As of January 1, 2000, 59,398 subjects over 70 years of age were included in the cohort.. The median age at baseline was 76 years of age [min: 70 / max: 107]. Overall survival at the end of follow-up was 79.3% for persistent statin users versus 61.0% for non LLA users. In the global multivariate Cox model, stratified on five-year birth cohorts and adjusted on gender and hospitalizations occurring in 1999, the HR for mortality associated with persistence of statins was 0.71 (95% CI 0.63 – 0.81; p<0.0001). No statistic association was found for the survival of subjects persistent to fibrate versus non LLA users, nor for persistent AG users versus non AG users, with respectively HR = 1.05 (95% CI 0.75 - 1.48, adjusted; p=0.77) and HR = 0.81 (95% CI 0.63 - 1.05, adjusted; p=0.12). CONCLUSIONS: Although further investigation must be conducted, in particular on adjustment variables, the main results of our study suggest a beneficial impact of persistent statin therapy in reducing all cause mortality by 29%, in a large cohort of elderly.