OBJECTIVES

This study assessed the cost-effectiveness of real-world use of GLP-1 receptor agonist (GLP1-ra) versus dipeptidyl peptidase-4 inhibitor (DPP-4i), sulfonylurea, and insulin in patients with type 2 diabetes from the healthcare sector perspective.

METHODS

Study cohort was assembled from Taiwan’s National Health Insurance (NHI) Research Database in 2007-2016. The balance of all baseline patient characteristics was achieved between drug comparison groups through a rigorous matching algorithm. Effectiveness outcomes were the number needed to treat for preventing one case of clinical events of interest (e.g., composite cardiovascular disease [CVD] events) and the extrapolated life expectancy. Individual-level medical costs from the initiation of study drugs throughout the follow-up period were estimated with adjustment for each patient’s baseline costs. Incremental cost-effectiveness ratio was calculated as the incremental cost per case of clinical events prevented and compared with the willingness-to-pay threshold (1-3 times of Taiwan’s GDP: $25,932-$77,796). Costs were adjusted using medical consumer price index and presented in 2019 U.S. dollar.

RESULTS

Study cohort included 1,415, 1,642, and 976 GLP-1ra users 1:1 matched to DPP-4i, sulfonylurea, and insulin users, respectively. During a mean follow-up of 2.3 years, GLP-1ra users had lower incidence rates of all clinical events compared to their counterparts. Total medical costs in GLP-1ra users were higher compared to DPP-4i and sulfonylurea users but lower relative to insulin users (i.e., GLP-1ra vs. DPP-4i, sulfonylurea and insulin: $8,200 vs. $6,956, $8,952 vs. $7,748, and $7,996 vs. $8,021). Using GLP-1ra vs. insulin, DPP-4i, and sulfonylurea yielded from –$36,843 to $13,655 incremental costs per case with any CVDs prevented from the healthcare sector perspective. Results were robust in several sensitivity analyses.

CONCLUSIONS

Real-world use of GLP-1ra relative to DPP-4i, sulfonylurea, and insulin could be cost-saving or effective because its higher acquisition drug cost was compensated by lower downstream costs of developing diabetic complications.