OBJECTIVES: Despite evidence from randomized controlled trials which show that treatment with high-intensity statins can result in an additional 16% reduction in CVD events compared to moderate-dose statins, only 20-25% of very high risk patients receive high-intensity statins.  Geographic variation in statin prescribing, in addition to the existence of a “treatment risk paradox” in which complex patients with apparently more to gain from treatment are less likely treated, may be associated with this low rate. Our objective is to determine whether there is an association between patient complexity and the intensity of statin treatment following acute myocardial infarction (AMI) by geographical region. METHODS: This was a retrospective cohort study of 124,397 Medicare patients with fee-for service Part A and B, and the Part D benefit who were hospitalized with AMI in 2008 or 2009. Patient complexity was defined by the presence or absence of diabetes, heart failure, and chronic kidney disease in the year prior to admission. Treatment with atorvastatin 40,80mg; and rosuvastatin 20,40mg in the first thirty days post discharge was defined as high-intensity; other statin treatment as low/moderate intensity, and the absence of a fill as no treatment. Area treatment rates (ATRs) adjusted for baseline patient characteristics were calculated and grouped into quintiles of “no treatment”, “low/moderate”, and “high” treatment areas. Tables were generated calculating statin intensity treatment rates by comorbidity-based AMI patient complexity. RESULTS: Those patients with no evidence of any complex conditions have the lowest "no treatment" rates (36.4%), while the highest "no treatment" rates occur among those with all three conditions (54.2%). Substantial geographic variation in the intensity of statin treatment within patient complexity subgroups is revealed using the ATR method. CONCLUSIONS: Despite guidelines promoting the use of high-intensity statins, more complex patients are less likely to be treated and treatment discretion is revealed in the geographic variation.