OBJECTIVE: To compare effectiveness of rosuvastatin (RSV) with other statins on lowering low-density lipoprotein cholesterol (LDL-C) and LDL-C goal attainment among patients with type 1 or type 2 diabetes mellitus. METHODS: A retrospective study using US General Electric Medical Systems (GEMS) database of patients with diabetes mellitus (ICD9 code = 250, prescription for anti-diabetic medication or fasting blood glucose level > or = 126 mg/dL in the 12 months preceding statin therapy) treated across clinical practices in the US, who were newly prescribed statin therapy during August 2003-March 2006, was conducted. Multivariate linear and logistic regression models were used for analyzing prescription data with baseline LDL-C, age, gender, smoking, very high CHD risk, systolic blood pressure, and statin duration as covariates. RESULTS: Of 4754 diabetes mellitus patients, 5% were prescribed RSV, 59% atorvastatin (ATV), 21% simvastatin (SMV), 5% pravastatin (PRV), 2% fluvastatin (FLV), and 7% lovastatin (LOV). RSV patients had significantly higher (p < 0.05) baseline mean LDL-C levels (138 vs. 117-131 mg/dL), lower average starting dose (11.7 vs. 17.0-63.7 mg) and were younger (p < 0.005) than patients on other statins (mean age 61 vs. 63-69 years). Percent LDL-C reduction was significantly greater (p < 0.0001) with RSV (28.4%) compared to ATV (22.5%), SMV (20.1%), PRV (13.7%), FLV (15.8%), and LOV (17.3%). A greater (p < 0.05) proportion of RSV diabetes patients attained LDL-C goal < 100 mg/dL (72.8%) vs. diabetes mellitus patients on other statins (36.8-67.4%). CONCLUSIONS: Rosuvastatin was more effective in lowering LDL-C and achieving LDL-C treatment goals in the diabetes mellitus population as compared to other statins in real-world clinical practice setting. Validating study results in a different diabetes population with dispensed statin prescriptions will help increase generalizability of study findings.