OBJECTIVES: SGLT2 and GLP1 are increasingly used as second-line augmentation therapies added after metformin. We conducted an observational study to examine and compare the real-world performance of SGLT2 and GLP1 in terms of HbA1c reduction and target achievement.

METHODS: Using claims from Optum’s Clinformatics® Data Mart, we selected metformin users who initiated SGLT2 or GLP1 between 01/01/2013 and 12/31/2019, in which the initiation date was the index date, with a 12-month baseline and follow-up period, respectively. Ordinary least square and logistic regressions were conducted to examine HbA1c reduction and target achievement between the two cohorts, controlling for baseline patient demographics and health status.

RESULTS: In a preliminary (1%) sample, a total of 150 and 118 patients were identified in the SGLT2 and GLP1 cohorts. The average HbA1c before and closest to the index date was 8.33% for the SGLT2 cohort and 8.66% for the GLP1 cohort. SGLT2 users were more likely to attain the American Diabetes Association (ADA) target of HbA1c < 7% [OR: 2.524, CI: (1.398, 4.555), p = 0.0021], slightly more likely to achieve the Healthcare Effectiveness Data and Information Set (HEDIS) target of HbA1c < 8% [OR: 1.086, CI: (0.585, 2.017), p= 0.7931], and less likely to fall within the HEDIS standard of poor control of HbA1c > 9% [OR: 0.409, CI: (0.168, 0.999), p = 0.0497], compared with GLP1 takers by the end of the study period. Less HbA1c reduction was estimated for the SGLT2 cohort, compared with the GLP1 cohort (0.277, p = 0.1253), where 0.277 was the estimated difference of reduction level (follow-up HbA1c minus index HbA1c) by being in the SGLT2 cohort.

CONCLUSIONS: The study results suggested that SGLT2 showed better achievement in three HbA1c targets by the end of the study period, while GLP1 was more efficacious in HbA1c reduction.