Before the recommendations for more aggressive treatment of type 2 diabetes that have developed in the past decade, treatment was often a slow, stepwise approach. This involved months, or even years, of lifestyle interventions before initiation of pharmacologic therapy.
This began to change in 2006 when the American Diabetes Association and European Association for the Study of Diabetes came to a consensus and recommended that metformin be initiated at diagnosis, along with lifestyle interventions. Although there are many appropriate exceptions, one would think that most patients diagnosed with type 2 diabetes since then have been prescribed metformin at diagnosis.
Raebel and colleagues recently published the results of their analysis of initial treatment choice in more than 240,000 patients newly diagnosed with type 2 diabetes. They used data from SUPREME-DM, a registry that includes a wealth of information (demographics, health care utilization, medications, procedures, laboratory data, etc) for more than 1 million patients with diabetes. They identified those with new-onset diabetes from 2005 to 2010 and documented what, if any, treatment was started at diagnosis, determined predictors of treatment choice and what factors affected use of metformin.
James R. Taylor
Their results showed that only 40.3% of the patients were started on any antihyperglycemic therapy within 6 months of diagnosis. The rates were significantly better in 2010 vs. 2005 (44.3% vs. 36.5%, P<.001). Of those who started therapy within 6 months, 60% started in the first week and 93% within the first 3 months. An average of 75.2% of those who received therapy within first 6 months were prescribed metformin (alone or combined with another agent). The most common initial treatment regimen not involving metformin was a sulfonylurea (28.4%), and metformin as the initial choice increased from 62.5% in 2005 to 84.6% in 2010.
The baseline HbA1c was found to be the strongest predictor of choice to initiate any agent within the first 6 months, with higher HbA1c levels obviously corresponding to increased likelihood of initiating treatment. Other predictors of initiating any agent within 6 months included being female, black or Asian, current or former smoker, and diagnosis of depression. Not surprisingly, the No. 1 predictor of starting a sulfonylurea rather than metformin was elevated serum creatinine. Other predictors of sulfonylurea use as initial therapy included age, black or Asian race, and male sex.
The overall relatively low use of therapy early after diagnosis is surprising, but can be partially explained by the low mean HbA1c (6.6%) of the study population who did not start therapy. However, most guidelines recommend starting therapy at diagnosis, even if HbA1c is relatively low. Based on the results of this study, it seems that we still have room for being more aggressive in treating diabetes, and perhaps that would improve the percentage of patients who are controlled.
Raebel MA. Ann Pharmacother. 2013;47:1280-1291.
James R. Taylor, PharmD, CDE, is a clinical associate professor in the department of pharmacy practice at the University of Florida, Gainesville. He can be reached at University of Florida, College of Pharmacy, P.O. Box 100486, Gainesville, FL 32610-0486; email: firstname.lastname@example.org. He reports no relevant financial disclosures.