This supplement was written by William T. Elliott, MD, FACP, Chair, Formulary Committee, Kaiser Permanente, California Division; Assistant Clinical Professor of Medicine, University of California-San Francisco. In order to reveal any potential bias in this publication, we disclose that Dr. Elliott reports no consultant, stockholder, speaker’s bureau, research, or other financial relationships with companies having ties to this field of study. For questions and comments, please e-mail:

Diabetics on metformin who need additional therapy may be better off with a sulfonylurea than insulin, according to a new study. In a large, retrospective cohort study from the Veterans Health Administration, researchers reviewed the records of nearly 180,000 diabetic veterans on metformin of whom (after propensity score matching) 2436 added insulin and 12,180 added a sulfonylurea. After a mean follow-up of 14 months, patients were assessed for a composite of acute myocardial infarction (MI), stroke, or all-cause mortality. There were 172 events in the insulin group vs 634 in the sulfonylurea group (42.7 insulin vs 32.8 events/1000 person-years, adjusted hazard ratio [HR], 1.30; 95% confidence interval [CI], 1.07-1.58; P = 0.009). The rate of MI and stroke were similar; however all-cause mortality was significantly higher in the insulin group (33.7 vs 22.7 deaths/1000 person-years, HR, 1.44; CI 1.15-1.79; P = 0.001). The authors suggest that the addition of insulin vs a sulfonylurea to metformin is associated with an increase in a composite of nonfatal outcomes as well as increased mortality. They also suggest that more research is needed to help understand the risks associated with insulin, but these findings corroborate other studies that have shown no cardiovascular benefit associated with insulin and increased mortality associated with tighter diabetic control. The authors state the “intensification of metformin with insulin among patients who could add a sulfonylurea (HbAc level less than ≈ 10%) offers no advantage in regard to risk of cardiovascular events and is associated with some risk” (JAMA 2014;311:2288-2296).