LAUSR

BACKGROUND Diabetes treatment includes very diverse drugs. It is essential to identify which drugs offer the best value for their costs. OBJECTIVES To estimate comparative cost effectiveness for treating diabetes mellitus with dulaglutide, liraglutide, or glargine in Colombia. METHODS A Markov model including diabetic microvascular and macrovascular complications was used to estimate cost-effectiveness. We used annual cycles, a 5-year time horizon, 5% discount rate, and third-party payer's perspective. Main outcomes were quality-adjusted life-years (QALYs) and incremental cost-effectiveness ratios (ICERs). Transition probabilities were obtained from primary studies and costs from local databases and studies. We used a threshold of 3 times the Colombian per capita gross domestic product (US $17,270 for 2015; US $1 = 2,743 Columbian pesos) to assess cost effectiveness. RESULTS Total costs related to dulaglutide, liraglutide, and glargine were US $8,633, US $10,756, and US $5,783, yielding 3.311 QALYs, 3.229 QALYs, and 3.156 QALYs, respectively. Dulaglutide dominated liraglutide given lower total costs and higher QALYs. The estimated ICER for dulaglutide compared with glargine was US $18,385, greater than the accepted threshold. Sensibility analysis shows that decreased dulaglutide cost, increased consumption of glargine, nondaily injection, and number and cost of glucometry could result in ICERs lower than the threshold. Probabilistic sensitivity analysis showed consistent results. CONCLUSIONS This estimation indicates that dulaglutide dominates liraglutide. Its ICER is, however, greater than the accepted threshold for Colombia in base case compared with glargine. By increasing population weight or glargine consumption, dulaglutide becomes cost effective compared with glargine, which could identify a niche where dulaglutide is the best option.