Background : One of the non-communicable diseases (PTM) is Diabetes Mellitus (DM). According to data from the International Diabetes Federation in 2013 the number of people with diabetes mellitus reached 382 million and will continue to increase to reach 592 million in 2035 (Simamora, 2016). In 2012 the prevalence of DM in Indonesia reached 371 million sufferers, this has increased by 57% from 2008. The proportion of type 2 DM is 95% when compared to type 1 DM (Fatimah, 2017). In 2015 the prevalence of DM in Central Java reached 99,646 cases (Hestiana, 2017). Method : The research method used was analytic observational (non-experimental) to explain the cost-effective comparison of giving Type 2 Diabetic Melitus scientific herbs and conventional Type 2 Diabetic Melitus therapy. The pharmacoeconomic approach is viewed using Cost Effectiveness Analysis (CEA) with retrospective data collection. Clinical outcomes were seen from interviews using the QALYs EQ 5D questionnaire. Result : Conventional therapy for diabetes mellitus has a therapeutic effectiveness of 8,7% and scientific herbal medicine for diabetes mellitus has a therapeutic effectiveness of 21,4%.The results of the cost-effectiveness analysis using the ICER and CEP methods concluded that scientific herbal medicine therapy for diabetes mellitus is not more cost effective than conventional diabetes mellitus therapy because it is in quadrant 1. The average direct medical cost at the Puskesmas is Rp. 70.134 and at Kalibakung Herbal Health Rp. 81.771.