713: Oral hypoglycemic agents in the treatment of gestational diabetes mellitus: a cost-effectiveness analysis

Abstract

Recent studies have shown statistically significant adverse neonatal outcomes associated with the use of oral hypoglycemic agents in pregnancy for the treatment of GDM. Due to robust patient preference for oral agents over subcutaneous insulin, we sought to explore the cost-effectiveness and outcomes associated with both glyburide and metformin. A cost-effectiveness model using TreeAge software was designed to compare outcomes in neonates of women treated for GDM with oral hypoglycemic agents. Oral agents were compared through insulin in the absence of adequately powered randomized controlled trials for direct comparison. We used a theoretical cohort of 195,000 women and assumed complete medication adherence. Outcomes included those statistically significant in meta-analyses of the three pharmaceuticals: macrosomia, with shoulder dystocia and permanent branchial plexus injury, neonatal hypoglycemia and NICU admission, in addition to cost and quality-adjusted life years (QALY) for both woman and neonate. Probabilities were derived from current literature and a cost-effectiveness threshold was defined as $100,000/QALY. In our theoretical cohort of 195,000 women with GDM, metformin treatment resulted in 20,163 fewer cases of macrosomia, 19,457 fewer cases of neonatal hypoglycemia and 2,140 fewer NICU admissions than glyburide (Table 1). Metformin decreased cost of oral pharmacotherapy by $2.3 million, despite the two-fold increased risk of requiring supplemental insulin during treatment. Additionally, we found that both oral agents dominated insulin as less costly and more efficacious when defining utility of subcutaneous insulin use as suboptimal (0.9) for the duration of gestation (Table 2). Our model was robust over a wide range of assumptions. In our theoretical cohort, metformin resulted in improved outcomes and decreased costs compared to glyburide for the oral treatment of GDM. These results should be interpreted in the context of insulin as the preferred treatment for GDM with long-term consequences of in-utero exposure to oral agents yet to be fully elucidated.View Large Image Figure ViewerDownload Hi-res image Download (PPT)