Trends in Diabetes Treatment and Monitoring among Medicare Beneficiaries

Abstract

Background: Diabetes is a costly and common condition, but little is known about recent trends in diabetes management among Medicare beneficiaries.

Objective: To evaluate the use of diabetes medications and testing supplies among Medicare beneficiaries.

Design/setting: Retrospective cohort analysis of Medicare claims from 2007 to 2014.

Participants: Traditional Medicare beneficiaries with a diagnosis of diabetes in the current or any prior year.

Main measures: We analyzed choices of first diabetes medication for those new to medication and patterns of adding medications. We also examined the use of testing supplies, use of statins and ACE inhibitors/angiotensin receptor blockers, and spending.

Key results: Diagnosed diabetes increased from 28.7% to 30.2% of beneficiaries from 2007 to 2014. The use of metformin as the most commonly prescribed first medication increased from 50.2% in 2007 to 70.2% in 2014, whereas long-acting sulfonylureas decreased from 16.6% to 8.2%. The use of thiazolidinediones fell considerably, while the use of new diabetes medication classes increased. Among patients prescribed insulin, long-acting insulin as the first choice increased substantially, from 38.9% to 56.8%, but short-acting or combination regimens remained common, particularly among older or sicker beneficiaries. Prescriptions of testing supplies for more than once-daily testing were also common. The mean total cost of diabetes medications per patient increased over the period due to the increasing use of high-cost drugs, particularly by those patients with costs above the 90th percentile of spending, although the median costs decreased for both medications and testing supplies.

Conclusions: The use of metformin and long-acting insulin have increased substantially among elderly Medicare patients with diabetes, but a substantial subgroup continues to receive costly and complex treatment regimens.

Keywords: Medicare; diabetes; elderly.

Figure 1

Choice of initial medication for diabetes treatment, 2007–2014, FFS.

Figure 2

Choice of second medication based on first medication, 2007–2014. First medication: any sulfonylurea.

Figure 3

Choice of first insulin for diabetes treatment, insulin cohort 2007–2014.