OBJECTIVES: To identify predictors associated with mail order pharmacy use and investigate whether mail order pharmacy use produces cost savings for patients and payers in diabetes care. METHODS: We conducted a longitudinal cross sectional study covering the years 2006-2010 among Medical Expenditure Panel Survey (MEPS) household component participants who were 18 years or older, diagnosed with diabetes and took antidiabetic medications for treatment. The types of pharmacy from which the medications were purchased were measured, including mail-order and community pharmacies. Potential predictors such as socioeconomic and health-related variables associated with mail order use were identified using multivariable logistic regression. Annualized average health care expenses per patient including prescription drug and medical care expenses during the 2 year period were measured from patient and payer perspectives respectively. The associations between various health care expenses and pharmacy dispensing channels were assessed by multivariable linear regression. RESULTS: We identified a total of 3,668 eligible subjects in our study, representing more than 66 million individuals during 2006-2010 in the United States. Nearly 20% of the subjects filled at least 2 antidiabetic drug prescriptions via mail order pharmacies. The mail order pharmacy users were older, had high school or college degrees and higher incomes, and were more likely to be covered by private insurance. No significant difference in diabetes-related drug expenses paid by the patient themselves was found between mail order and community pharmacy users after adjustment. From the payer perspective, diabetes-related drug expenses and total health care expenses were 35% and 17% higher in mail order users, respectively. CONCLUSIONS: Besides pharmacy benefit design, socioeconomic status plays a role to influence patient preference of prescription drug dispensing channels in diabetes care, which may affect patient medication use behavior. Mail order use might not produce cost savings on diabetes-related medication use for patients and payers.