OBJECTIVE: To assess whether switching to insulin therapy in patients with type 2 diabetes mellitus is associated with medication refill compliance of oral hypoglycemic agents. RESEARCH DESIGN AND METHODS: The PHARMO Record Linkage System was useds data source for this study. Patients with newly treated type 2 diabetes mellitus were defined as subjects in whom oral hypoglycemic therapy was initiated between 1991 and 1998. We performed a matched case-control study in this cohort. Cases were patients who switched to insulin therapy. Date of switching in the case was defined as the index date. Controls were subjects still on oral therapy on the index date, matched on duration of diabetes and calendar time. We measured the medication refill compliance in the year starting 18 months before the index date and calculated various compliance indices. RESULTS: In total, 411 cases and 411 matched controls were identified. Cases suffered more often from more severe comorbidity and used a higher number of oral hypoglycemic agents and concomitant non-diabetic drugs. The overall compliance rate did not differ significantly between cases and controls, the adjusted odds ratio (OR) was 1.3 (CI 95% 0.6-2.8). After performing multivariate logistic regression modeling, age at onset of diabetes, gender, comedication, combination therapy, and daily dosage frequency, were independently related to switching. CONCLUSIONS: We were unable to confirm the hypothesis that noncompliance with treatment is more prevalent in patients with secondary failure. Other variables, like comorbidity and disease-related factors seem to play a more important role in switching to insulin therapy.