Call for comparative effectiveness research: lowering A1c with sitagliptin, saxagliptin, or cinnamon.

rosiglitazone within 3 months after the publication of the Nissen and Wolski meta-analysis, 13% had no prescription claims for any oral antidiabetic agent in the 90-day period following the final rosiglitazone prescription.12 Providing further insight into what actually happens following the delivery of unexpected news to the medical community, Starner et al. (2008) retrospectively analyzed administrative claims for 9 million members enrolled in 9 health plans, finding that use of rosiglitazone dropped by approximately 50% from April 2007 through December 2007, following 5 safety warnings regarding rosiglitazone.13 However, in May 2008, 12 months after publication of Nissen and Wolski’s findings, more than 20% of the remaining rosiglitazone users had clinical characteristics putting them at high cardiovascular risk, such as congestive heart failure, use of nitrates or insulin, or ischemic heart disease. Starner et al.’s findings highlight the conundrum facing many managed care organizations—the process of translating new information into routine clinical practice is nearly always difficult and sometimes slower than ideal.