Type 2 diabetes mellitus is a chronic disease that requires lifelong therapy and high service costs so that the use of antidiabetic drugs affects the patient's therapeutic costs. The purpose of this study was to determine the cost of outpatient type 2 diabetes mellitus therapy for National Health Insurance (JKN) participants in Condong Catur Hospital Yogyakarta. The research method used is observational. The parameters measured are calculating the average direct medical costs (drug costs, doctor's examination and consultation fees, laboratory costs, medical treatment costs), and comparison between the costs of oral drugs, insulin, oral and insulin. The results of normality tests with Shapiro-Wilk (microvascular, microvascular and macrovascular), Kolmogorrov-Smirnov (macrovascular) p0.05 indicate that data are not normally distributed, the results of the Kruskal Wallis test of the three groups (microvascular, macrovascular, microvascular and microvascular) show that differences are not normally distributed significant (p = 0.001).The results showed that the microvascular group therapy costs Rp 3,164,733.58 per month, the macrovascular group Rp 9,984,566.85 per month, the microvascular and macrovascular groups Rp 11,260,372.39 per month. The cost of therapy is based on the microvascular group drug Rp. 408,567.78 per month, the macrovascular group Rp. 1,245,987.48 per month, the microvascular and macrovascular groups Rp. 1,786,270.68 per month so that statistically from the three complications groups show significant differences p = 0,001 (p 0,005). The difference in the difference between the total real costs and the total INA-CBG's rates, the microvascular group Rp -38,741.14 per patient, the macrovascular group Rp -10,914.03 per patient, while in the microvascular and macrovascular group Rp 3,272.90 per patient shows that from to three groups of complications were not significantly different p = 0.207 (p 0.005). Statistically it can be concluded that the cost of therapy based on antidiabetic drugs in the three groups of complications shows a significant difference. There is no significant difference in the difference between the real costs in the three groups of complications