OBJECTIVES: Assess associations between the Deyo-Charlson index (DCI) score and annual cardiovascular disease (CVD) event risk and attributable total health care costs (THC) between patients initiating niacin extended-release (NER) plus simvastatin (NER/S) and simvastatin plus ezetimibe (S/E) fixed-dose therapy among patients with prior CVD. METHODS: A retrospective analysis of patients aged ≥ 18 years newly initiating S/E or NER/S therapy (initial therapy of NER added to existing simvastatin therapy) between January 1, 2001-June 30, 2006 (index date) was performed using a large integrated managed care research database. Patients with a minimum of 12 months pre- and post-index date follow-up and a diagnosis of CVD at some time during 12 months prior to index date were included. Associations between pre-index date DCI score, CVD event risk, and THC [sum of inpatient, emergency room, and outpatient visit costs] were estimated using Cox proportional hazards regression model and multivariate generalized linear model, respectively. RESULTS: A total of 7065 study patients were identified initiating S/E (n=6513) or NER/S (n=552). NER/S patients were significantly younger (58.5±9.2 years vs. 61.3±10.2 years; p<0.0001) and more likely to be male (85.1% vs. 67.9%; p<0.0001). Pre-index date DCI score (1.3±1.3 vs. 1.4±1.6; p=0.1018) was comparable between the two groups. Cox regression showed that every one unit increase in pre-index date DCI score was associated with a 27% [Hazard Ratio (HR): 1.27 (1.22-1.32); p<0.05] higher likelihood to experience a post-index CVD event. Multivariate regression showed that every one unit increase in pre-index date DCI score was associated with a 38% (Coefficient: 1.38, 95% CI: 1.30-1.47; p<0.0001) increase in mean annual CVD THC. CONCLUSIONS: High-risk patients with prior CVD and with an increasing DCI score were associated with a higher CVD event risk and total annual CVD-attributable THC. Further studies on dyslipidemia treatment strategies on higher risk patients are warranted.