Purpose: To explore whether newly diagnosed type 2 diabetes patients without previous cardiovascular disease (CVD) initiate preventive statin therapy regardless of ethnic background. Methods: Using nationwide individual-level registers, we followed a cohort of Danish-born residents and immigrants from Turkey, Pakistan and Ex-Yugoslavia, all without previous diabetes or CVD, during the period 2000-2008 for first dispensing of oral glucose-lowering medication (GLM), first dispensing of statins and register-markers of CVD (N = 3,764,620). Logistic regression analyses were used to test whether the odds ratios (ORs) of early statin therapy initiation (within 180 days after first GLM dispensing) are the same regardless of ethnic background. While age and gender were included as confounders in the basic model, income was included in the second model as a potential mediating variable. Results: Compared to native Danes, the ORs for early statin therapy were 0.68 (95 % confidence interval 0.50-0.92], 0.67 (0.56-0.81) and 0.56 (0.44-0.71) for Ex-Yugoslavians, Turks and Pakistanis, respectively. The differences remained largely unchanged after adjusting for income and tended to be accentuated when the threshold period was extended. The ORs of women initiating therapy (compared to native Danes) were 0.56 (0.35-0.90), 0.60 (0.46-0.78) and 0.48 (0.32-0.72) for Ex-Yugoslavians, Turks and Pakistanis, respectively, and those for men were 0.78 (0.52-1.17), 0.74 (0.58-0.95) and 0.60 (0.44-0.83), respectively. Conclusions: Immigrants from Turkey, Pakistan and Ex-Yugoslavia with type 2 diabetes were less likely to initiate statin therapy than Danish-born residents - despite a similar or even higher risk of CVD. The treatment inequities associated with ethnicity were more pronounced in women than men.