Factors Associated with Initiation of Glucose-lowering Agents among Medicare Beneficiaries with Newly Diagnosed Type 2 Diabetes, 2007 – 2017

Patients with type 2 diabetes (T2D) endure significant clinical and economic burdens. Early initiation of a pharmacologic agent can help achieve better glycemic control and reduce the risk of diabetes-related complications. Professional society guidelines recommend initiating metformin at the time of diagnosis, while others recommended newer drug classes depending on patients’ clinical characteristics. Little is known about the proportion of and timing of initiation of glucose-lowering agents among Medicare beneficiaries newly diagnosed with T2D. As the initiation and the selection of glucose-lowering agents are based on a myriad of clinical characteristics and patient preferences, it also remains unclear which factors are associated with treatment initiation and therapeutic class selection. In this study, we sought to understand the patterns and factors associated with the initiation of pharmacological treatment after T2D diagnosis and the selection of therapeutic class. Using 2006-2018 Medicare claims data, we identified patients newly diagnosed with T2D in 2007-2017 and assessed the initiation of a glucose-lowering agent within the first year of diagnosis and time from diagnosis to initiation. Independent variables included patient demographics, social factors, clinical characteristics, and healthcare utilization. Adjusted Cox Proportional Hazard models were constructed to identify factors associated with time to treatment initiation. We found that only 13.7% of Medicare beneficiaries with newly diagnosed T2D initiated a glucose-lowering agent within the first year after diagnosis, and this remained relatively constant from 2007-2017. In the adjusted model, increasing age (HR 0.92, 95%CI 0.91-0.93 for 10-year increase) and female gender (HR 0.89, 95% CI 0.87-0.91) were associated with a lower likelihood of initiation within one year of diagnosis. Black race was associated with a lower hazard of initiation than White race (HR 0.92, 95% CI 0.89-0.96) in the adjusted model. Antihypertensive drug use (HR 1.23, 95% CI 1.20-1.26) and statin use (HR 1.17, 95% CI 1.14-1.20) were associated with increased hazards of glucose-lowering agent initiation, whereas a history of CVD (HR 0.86, 95% CI 0.84-0.88) and chronic kidney disease (HR 0.84, 95% CI 0.82-0.87) were associated with a reduced likelihood of initiation. There was a large variation in T2D treatment initiation across regions and states, being lowest in Hawaii (6.35 cases per 100 persons-year) and highest in North Dakota (20.17 cases per 100 persons-year). Of those who initiated an antidiabetic drug within the year of T2D diagnosis, 54.2% and 21.5% of patients received metformin and sulfonylureas in 2007, while 84.4% and 6.0% initiated metformin and sulfonylureas in 2017, respectively. After adjustment, older age (OR 0.67, 95% CI 0.65-0.70 for ten years of increase), Black race (OR 0.80, 95% CI 0.72-0.90), chronic kidney disease (OR 0.44, 95% CI 0.40-0.48) and CDV disease (OR 0.76, 95% CI 0.71-0.82) were associated with lower odds of initiating metformin compared to sulfonylureas, while female gender (OR 1.15 95% CI 1.07-1.23) was associated with higher odds of receiving metformin compared to sulfonylureas. Our findings describe the real-world initiation of glucose-lowering agents following the first T2D diagnosis in older adults and carry important implications for quality prescribing initiatives. Further investigation is needed to investigate the barriers of the low initiation rate and nonconformity with the guidelines.