OBJECTIVES: To compare recent trends in treatment pattern for diabetes between 2002 and 2010. METHODS: A cross-sectional study of expenditures was carried for representative sample of civilian, noninstitutionalized U.S individuals with diabetes from the Medical Expenditure (MEPS) 2002-2010. Expenditures include all sources of payment for oral anti-diabetic medications, insulin, and non-insulin injectables.  We inflated 2002 dollar values to 2010 values using the consumer price index. RESULTS: From 2002 to 2010, the estimated number of persons reporting medication therapy for diabetes increased from 11.8 to 17.6 million, but the percentage of patients receiving prescription medication for diabetes dropped from 90.5% to 87.5% (p<.001). The mean number of physician’s office visits for treating diabetes increased from 45.9 million in 2002 to 55.9 million in 2010, while the mean number of visits per treated patient declined from 4.7 to 3.9 during the same period. Among the patients used prescription medications, the proportion of those using both insulin and oral anti-diabetic medications increased from 14.9% in 2002 to 18.4% in 2010. For specific classes of oral anti-diabetic medications, the percentage using biguanides increased from 40.1% to 60.5%, in the contrast, use of sulfonylureas and thiazolidinediones dropped from 50.1% to 36.1% and 20.7% to 12.8% respectively. The annual expenditures per user for medications for treating diabetes increased from $889 in 2002 to $1026 in 2010. Increasing use of newer insulin and thiazoliones was the main driver of recent increases in the medication expenditures ($10.5 billion in 2002 to $18.1 billion in 2007). CONCLUSIONS: Increasing costly medication therapy for diabetes is being applied to an increasing population. However, the percentage of diabetes patients received medication treatment slightly dropped. This suggests that much more attention should be directed in reducing the gap between treatment need and use as well as evaluation of more costly medication therapy for diabetes.