OBJECTIVE: To examine the impact of pre-existing depression on utilization patterns and adherence to oral hypoglycemic agents (OHAs) in patients newly diagnosed with type-2 diabetes. METHODS: Newly diagnosed type-2 diabetes patients during the three-year period (1998-2000) were identified from a Medicaid claims database. Presence of pre-existing depression was determined on the basis of ICD-9 CM codes for depression. Utilization patterns (switching, augmentation) and adherence to OHAs were computed for a 12-month follow up period from the date of the index OHA prescription. A multivariate framework was used to estimate the impact of depression on utilization patterns and adherence, controlling for confounders such as demographics, co-morbidity, diabetes severity, regimen complexity, and interaction with health care providers. RESULTS: A total of 1326 newly diagnosed type-2 diabetes patients were identified (depressed=471; non-depressed=855). A significantly higher number of depressed patients (23.3%) switched or augmented therapy as compared to non-depressed patients (16.2%). Results of a multinomial logit model indicated that controlling for covariates, patients with depression were 1.7 times more likely to switch (p=0.046) and two times more likely to augment therapy (p=0.003) as compared to non-depressed patients. Results of an extended-Cox proportional hazard model indicated that the hazard to switch/augment therapy was 2.4 times more for depressed patients as compared to non-depressed patients in the latter six-months of the follow-up period (p=0.0005). Depression was consistently found to be a significant predictor of adherence, with depressed patients being 3-6% less adherent to their OHAs than non-depressed patients. CONCLUSION: Depression significantly impacts utilization patterns and adherence to OHAs in patients with type-2 diabetes. This lack of adherence may affect glycemic control and consequently incidence of diabetes related complications. The study results imply that depression screening and treatment may be included in the protocol for management of type-2 diabetes patients.