A short-term cost-of-treatment model for type 2 diabetes: comparison of glipizide gastrointestinal therapeutic system, metformin, and acarbose.

Author(s): Ramsdell JW, Grossman JA, Stephens JM, Botteman MF, Arocho R

Affiliation(s): Department of Medicine, University of California, San Diego, USA.

Publication date & source: 1999-08, Am J Manag Care., 5(8):1007-24.

Publication type:

OBJECTIVE: To compare, from a managed care perspective, the 3-year costs of 3 first-line monotherapy strategies in type 2 diabetes patients: glipizide gastrointestinal therapeutic system (GITS), metformin, and acarbose. STUDY DESIGN: A Markov model, with a Monte Carlo simulation, was developed to compare the costs to achieve full glycemic control (hemoglobin A1c of < or = 7%) with each first-line strategy. PATIENTS AND METHODS: The patient population for the model was assumed to be all newly diagnosed type 2 diabetes patients eligible for monotherapy with an oral agent. Each monotherapy could be succeeded by add-on treatments. The model included the costs of routine medical care and supplies, medication, adverse events, and treatment failures. RESULTS: Using a Monte Carlo simulation, the mean 3-year cumulative costs per patient were $4971, $5273, and $5311 for glipizide GITS, metformin, and acarbose first-line strategies, respectively. The main cost drivers were drug prices. Mean 3-year cost savings for first-line glipizide GITS were $301 over metformin and $340 over acarbose. Between 83% and 85% of all simulations showed cost savings with glipizide GITS compared with the other agents. CONCLUSIONS: The model suggests first-line monotherapy with glipizide GITS should result in desirable short-term economic benefits for managed care. Because the model incorporates recommended glycemic goals and performed well in sensitivity analyses, it should be applicable to a variety of clinical practices and useful for economic assessments of new therapies. Results of this model should be verified prospectively in typical care settings.