OBJECTIVES : To assess the short-term cost-effectiveness of switching patients with type 2 diabetes from dipeptidyl peptidase-4 inhibitors (DPP-4i) to either sodium-glucose cotransporter 2 inhibitors (SGLT2i) or glucagon-like peptide-1 receptor agonists (GLP-1RA) on inpatient visits

METHODS : A TreeAge decision tree-based economic model, over a one-year horizon, was developed to assess the cost-effectiveness of switching patients from DPP-4i to either SGLT2i or GLP-1RA on inpatient visits averted. We used real-world patient data for those on DPP-4i and switched to either SGLT2i or GLP-1RA in the 2016-2017 period and compared them to patients who did not switch. Administrative claims data from a large national pharmacy benefit manager was the source of the input variables. The model was analyzed from a third-party payer’s perspective. Propensity score matching was used to estimate the total health care costs (i.e., direct medical cost (including inpatient, outpatient, and other medical services) and pharmacy cost (DPP-4i vs. GLP-1RA and DPP-4i vs. SGLT2i)). Outcome of interest was the incremental cost effectiveness of switching per inpatient visit averted.

RESULTS : Of the 47,953 patients in the study, 531 and 847 switched to SGLT2i and GLP-1RA respectively. There were more males in the switchers to SGLT2i (61%) than switchers to GLP-1RA (48%). Compared to SGLT2i, a higher proportion of the GLP-1RA switchers had a cardiovascular disease-related medical claim (15% vs. 13%). Those who remained on DPP-4i had a higher proportion of inpatient visits compared to those who switched. The incremental cost (savings) of switching to GLP-1RA was (-$556) and to SGLT2i was (-$422). Switching presented a cost saving outcome for both DPP-4i to GLP-1RA and DPP-4i to SGLT2i cohorts.

CONCLUSIONS : We found that although the pharmacy costs of SGLT2i and GLP-1RA were higher than DPP-4i, their favorable total direct medical cost accounted for the cost-saving estimation. The findings could be useful to prescribers.