What is known and Objective: Out-of-pocket expenses of drug therapy may negatively affect adherence. We aimed to analyse 1-year discontinuation rates between cohorts initiating therapy with either generic simvastatin or non-generic atorvastatin. Methods: Statin-naìve initiators of atorvastatin and generic simvastatin in April-June 2003, and corresponding cohorts in 2005, were identified through the nationwide Finnish prescription register. Persistence with statin therapy was followed for 365 days, considering the treatment to have been discontinued when the tablet-free gap between two consecutive refills exceeded 90 days. Using multivariate-adjusted logistic regression, odds ratios (OR) for discontinuation associated with initiating with simvastatin vs. atorvastatin were estimated separately for each year. Results and Discussion: In the year 2003, 5838 persons initiated treatment with atorvastatin and 5644 with generic simvastatin. In the year 2005, the respective numbers were 5228 and 10 987. Soon after the introduction of generic substitution in 2003, there was no difference in the risk of discontinuation between the comparator groups [OR 0·97, 95% confidence interval (CI) 0·89-1·05]. Two years later, persons initiating with generic simvastatin were 20% less likely to discontinue statin therapy (OR 0·80; 95% CI 0·74-0·83). Among persons whose medicinal costs were almost completely reimbursed towards the end of the initiation year, the OR was 1·14 (95% CI 0·76-1·64, P = 0·033 for interaction). What is new and Conclusions: We found that lower out-of-pocket expenses associated with the initiating statin had a positive impact on persistence with therapy. The finding does not seem to apply to persons with minor copayments towards the end of the initiation year.