Berkowitz and colleagues1 assert that there is little comparative effectiveness evidence to guide initial selection of therapy for diabetes mellitus. They therefore conducted this rigorous study to determine effects attributable to initial oral glucose-lowering agents. With a retrospective cohort design using 4 years of recent claims data from a large national insurer, they investigated outcomes from 4 classes of oral diabetes medications (metformin, sulfonylureas, thiazolidinediones, and dipeptidyl peptidase 4 inhibitors). In their adjusted models, initiation of therapy with a sulfonylurea, thiazolidinedione, or dipeptidyl phosphatase 4 inhibitor was associated with an increased hazard of treatment intensification (addition of another drug) relative to first therapy with metformin, without greater clinical benefits (and often more short-term adverse events). The authors noted that only 57.8% of individuals in their data set began pharmaceutical management of their diabetes with metformin. This study understandably omitted the newest class of medications—the sodium-dependent glucose transporter 2 inhibitors—because the first drug was not approved in the United States until March 2013. Less understandable is the exclusion of injectable medications from the glucagon-like peptide 1 agonist class despite their approval for use as monotherapy and availability since 2005.
Segal JB, Maruthur NM. Initial Therapy for Diabetes Mellitus. JAMA Intern Med. 2014;174(12):1962-1963. doi:10.1001/jamainternmed.2014.4296