OBJECTIVES: Japan has universal health coverage, and, regulations concerning medical reimbursements are loose in the prescription, with physicians having a high degree of discretion in prescriptions. There is no upper limit or reference pricing, and as both physicians and patients prefer brand drugs, there has been a lack of progress in the use of generic drugs.It is possible that even patients who have continually been prescribed generic drugs may switch to Protected Brand Drugs when they reimburse. The purpose of this study was to clarify this switch of patients to whom generic drugs had been prescribed to Brand Drugs using nationwide administrative data. METHODS: Using dispensation  data of pharmacies on an annual basis from April 2011 to December 2014, we identified all prescriptions for DM, Hypertension and Hyperlipidemia.１) We targeted patients prescribed only generic drugs between April and September each year, and who have dispensation records for three months or longer. 2) Where there was a change to the brand drug from a patient from Oct to March each year, this was defined as a drug switch. In terms of data analysis, we used SQL Server 2008 R2 for data handling and R for data analysis. RESULTS: From among the patients with diabetes  (N=181,378), hypertension (343,188) and hyperlipidemia (343,188), those matching condition 1) in 2012 were patients with diabetes (n=44,533), hypertension (132,165) and hyperlipidemia (81,455). 31% of all patients were continually prescribed only generic. The switch from generic to Protected Brand Drugs was higher in the younger 40-50 age group (p=0.05) than in the elderly generation, and there was a higher rate of males (p=0.05)。CONCLUSIONS: The results of this study showed that when promoting drug substitution to generic drugs as a policy for reducing medical costs, it is also necessary to consider measures to counteract the shift to brand drugs.