Adequate persistence to antidiabetic treatment is highly important to achieve proper glycemic control. Data on persistence with new antidiabetic drugs are limited. The aim of the study was to evaluate the persistence to treatment with DPP-4-inhibitors, SGLT2-inhibitors and GLP-1-RAs in comparison to insulin (INS) in a nationwide cohort of T2DM patients. Using the database of the Hungarian National Health Insurance Fund, the persistence to the treatment with DPP-4-inhibitors (n=59,900), SGLT2-inhibitors (n=26,052), GLP-1-RAs (n=17,332) and INS (n=51,781) was analyzed between 2014 and 2016. We captured the antihyperglycemic agents either in monotherapy or in add-on combinations after failure with first antidiabetic drug. Persistence was defined as the duration of time from initiation to discontinuation of the particular therapy. A permissible gap of 180 days (grace period) was allowed. The results are reported as the rate of persistent patients at year 1 and year 2. The persistence of DPP-4-inhibitors and SGLT2-inhibitors was similar at year 1 (68%, 67%), both of which decreased by year 2 (55% and 56%, respectively). The persistence of GLP-1-RAs was 66% at year 1, which decreased remarkably to year 2 (51%). DPP-4-inhibitors + metformin in fix combination had higher persistence (64% and 50%) than that of free add-on combinations (43% and 29%, at year 1 and year 2, respectively). Similar trend was observed regarding SGLT2-inhibitors + metformin fix vs. free add-on combinations at year 1 (75% vs. 55%). Treatment with INS had a relatively high persistence (71% and 61% at year 1 and year 2, respectively). Persistence to treatment with innovative drugs was the highest in patients with age of 40-60 years. Persistence data of different antihyperglycemic agents provide useful information and should be considered as one of components that help to guide the proper decision at initiation of antidiabetic treatment.


G. Jermendy: None. G. Rokszin: None. Z. Abonyi-Tóth: None. I. Wittmann: None. P. Kempler: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at