Changing Trends in Type 2 Diabetes Mellitus Treatment Intensification, 2002-2010

Objectives: Glycemic control can lower the risk of diabetes-related complications, and delayed treatment intensification can impede optimal diabetes care. This study examines trends in hyperglycemia treatment intensification between 2002 and 2010. Study Design: Retrospective secondary data analysis of a large national administrative data set of privately insured individuals across the United States. Methods: Adults 18 years or older with diabetes, initiated on metformin monotherapy between 2002 and 2007, were studied, stratified by date of first metformin prescription (2002-2003, 2004-2005, 2006-2007). Time to treatment intensification between 2002 and 2010, defined by the addition of >= 1 agents to metformin, was estimated using Kaplan-Meier and Cox proportional hazards regression analysis. Results: There were 75,069 treatment-naive adults with diabetes first initiated on metformin between 2002 and 2007; mean age was 60 years (SD = 11.5), 49.7% were women, and 63.1% were non-Hispanic white. Diabetes therapy was intensified in 26,169 individuals (34.6%). Treatment intensification became increasingly more likely with time for the 2004-2005 cohort (hazard ratio [HR], 1.07; 95% CI, 1.04-1.10) and for the 2006-2007 cohort (HR, 1.11; 95% CI, 1.07-1.14) compared with the 2002-2003 cohort (P <. 001), after adjustment for significant confounders including sex, income level, education level, and comorbidity burden. Sulfonylureas were the most commonly used agents, though their use declined over time; thiazolidinedione use decreased; and incretin use increased (all P <. 001). Conclusions: There was a significant increase in diabetes treatment intensification between 2002 and 2010. Choice of second-line agents changed as well, with decreasing prevalence of thiazolidinedione and sulfonylurea use and rising prevalence of incretin use.