OBJECTIVES: Following a commitment to an intensive glucose-lowering regimen that includes mealtime insulin is eventually required by patients taking basal insulin to maintain good glycemic control. The objective of this study was to characterize and examine factors associated with persistence to mealtime insulin. METHODS: Patients with diagnosed type-2 diabetes, with at least 2 prescriptions for mealtime insulin (index medication) between July 2001 and Sept 2006, were identified from a US managed care claims database. Patients were required to have 6 months pre- and 15 months post-index continuous eligibility and at least 2 basal insulin prescriptions in pre-index period. Persistence measure #1 was defined by the absence of prescription gap of greater than 90 days, with non-persistence effective the date of the last prescription prior to the 90 day gap. Persistence measure #2 required one prescription per quarter to be persistent at 12 months; persistence at 3 and 6 months were defined similarly. Logistic regression models were used to examine predictors of persistence to mealtime insulin at 12 months. RESULTS: Patients adding mealtime insulin to their basal regimen (n=4,752; 51% male, mean age=60.3 years) mostly used insulin analogs (60%) and vial/syringe (87%). The median number of mealtime insulin prescription claims filled per patient was 2, 3 and 4 at 3, 6 and 12 months respectively, with the median time between refills being 75.5 days. Persistence to mealtime insulin was 40.7%, 30.2% and 19.1% using measure #1 and 74.3%, 55.3% and 42.2% using measure # 2 at 3, 6 and 12 months, respectively.  Patients initiating with human insulin were less likely to be persistent across measures of persistence (OR <0.80, p<0.01). Additional predictors of persistence at 12 months included age, copayment, mental health comorbidity and polypharmacy. CONCLUSIONS: Persistence to insulin therapy is poorer than one would anticipate and is significantly lower for human insulin compared to analogs.