OBJECTIVES: To compare healthcare resource utilization (HCRU) among empagliflozin versus dipeptidyl peptidase-4 inhibitor (DPP4i) users observed during the first two years of EMPRISE study (08/2014-09/2016) using two US commercial (Optum and MarketScan) and Medicare claims datasets.

METHODS: In this interim analysis, we identified a 1:1 propensity-score-matched cohort of T2D patients ≥18 years initiating either empagliflozin or a DPP4i, and assessed balance on ≥140 covariates including clinical, HCRU, and cost related covariates at baseline using absolute standardized differences (aSD). We compared risk of first hospitalization, risk of first emergency department (ED) visit, hospital length of stay (LOS), and number of hospital admissions, office visits, and ED visits in empagliflozin and DDP4i users.

RESULTS: We identified 17,549 matched pairs in the three data sets with mean follow-up of 5.4 months. All baseline characteristics were well balanced (with aSD < 0.1) after propensity matching. The mean age was 59 and 47% were female. Approximately 24% of patients had history of CV disease and mean HbA1C was 8.6%. Risk of first hospitalization [Hazard ratio (HR)=0.84; 95%CI: 0.76, 0.92] and ED visit (HR=0.83; 0.73, 0.94) was lower in empagliflozin than DDP4i users as was the number of all hospital admissions (12% and 15% per member per year (PMPY); Incidence rate ratio (IRR)= 0.78; 0.72, 0.86). LOS was 0.52 days PMPY in empagliflozin compared to 0.75 days in DDP4i group (Diff= -0.23, 95% CI = -0.38, -0.08). Number of ED visits (0.07 and 0.09 PMPY; IRR=0.80; 0.71, 0.89), and number of office visits (8.03 and 8.39; IRR=0.96; 0.95, 0.98) were also lower in empagliflozin than DPP4i initiators.

CONCLUSIONS: We observed significant lower HCRU in empagliflozin compared to DDP4i initiators in the pooled dataset within the first two years of EMPRISE.