IMPORTANCE: High-intensity statin therapy is recommended for the secondary prevention of atherosclerotic cardiovascular disease (ASCVD). Nevertheless, statin therapy in general, and high-intensity statin therapy in particular, is underused in patients with established ASCVD. OBJECTIVE: To determine the association between all-cause mortality and intensity of statin therapy in the Veterans Affairs health care system. DESIGN, SETTING, AND PARTICIPANTS: A retrospective cohort analysis was conducted of patients aged 21 to 84 years with ASCVD treated in the Veterans Affairs health care system from April 1, 2013, to April 1, 2014. Patients who were included had 1 or more International Classification of Diseases, Ninth Revision codes for ASCVD on 2 or more different dates in the prior 2 years. EXPOSURES: Intensity of statin therapy was defined by the 2013 American College of Cardiology/American Heart Association guidelines, and use was defined as a filled prescription in the prior 6 months. Patients were excluded if they were taking a higher statin dose in the prior 5 years. MAIN OUTCOMES AND MEASURES: The primary outcome was death from all causes adjusted for the propensity to receive high-intensity statins. RESULTS: The study sample included 509 766 eligible adults with ASCVD at baseline (mean [SD] age, 68.5 [8.8] years; 499 598 men and 10 168 women), including 150 928 (29.6%) receiving high-intensity statin therapy, 232 293 (45.6%) receiving moderate-intensity statin therapy, 33 920 (6.7%) receiving low-intensity statin therapy, and 92 625 (18.2%) receiving no statins. During a mean follow-up of 492 days, there was a graded association between intensity of statin therapy and mortality, with 1-year mortality rates of 4.0% (5103 of 126 139) for those receiving high-intensity statin therapy, 4.8% (9703 of 200 709) for those receiving moderate-intensity statin therapy, 5.7% (1632 of 28 765) for those receiving low-intensity statin therapy, and 6.6% (4868 of 73 728) for those receiving no statin (P < .001). After adjusting for the propensity to receive high-intensity statins, the hazard ratio for mortality was 0.91 (95% CI, 0.88-0.93) for those receiving high- vs moderate-intensity statins. The magnitude of benefit of high- vs moderate-intensity statins was similar, for an incident cohort hazard ratio of 0.93 (95% CI, 0.85-1.01). For patients aged 76 to 84 years, the hazard ratio was 0.91 (95% CI, 0.87-0.95). Patients treated with maximal doses of high-intensity statins had lower mortality (hazard ratio, 0.90; 95% CI, 0.87-0.94) compared with those receiving submaximal doses. CONCLUSIONS AND RELEVANCE: We found a graded association between intensity of statin therapy and mortality in a national sample of patients with ASCVD. High-intensity statins were associated with a small but significant survival advantage compared with moderate-intensity statins, even among older adults. Maximal doses of high-intensity statins were associated with a further survival benefit.