CHICAGO – Only 31% of patients with type 2 diabetes who were placed on a glucagonlike peptide-1 agonist persisted on the medication for at least 6 months in a very large national retrospective cohort study.
That’s significantly lower than the 39% rate of 6-month treatment persistence in patients placed on a dipeptidyl-peptidase-4 inhibitor, Carol E. Koro, Ph.D., reported at the annual scientific sessions of the American Diabetes Association.
The 12-month treatment discontinuation rate was 89% in patients placed on a GLP-1 agonist, compared with 82% for patients on a DPP-4 inhibitor and 84% among those placed on other antidiabetic agents, added Dr. Koro, an epidemiologist at GlaxoSmithKline in Research Triangle Park, N.C.
She presented a retrospective study examining treatment utilization patterns among 134,696 commercially insured type 2 diabetes patients placed on a GLP-1 agonist, 202,269 given a DPP-4 inhibitor, and 1,014,630 on other antidiabetic agents. The data source was the Truven Health Analytics MarketScan database for 2005-2011.
Nonadherence was defined by a prescription refill pattern indicating that less than 80% of a prescribed medication was being taken. The nonadherence rate was 9.6% in patients whose physicians prescribed a GLP-1 agonist, 5.4% in those placed on a DPP-4 inhibitor, and 4.4% in patients on other antidiabetic agents.
The median time to treatment discontinuation was 90 days in the GLP-1 agonist cohort, 120 days for the DPP-4 inhibitor cohort, and 90 days for patients assigned to other antidiabetic agents. The mean time to discontinuation was 178 days for the GLP-1 agonist group, 226 days for the DPP-4 inhibitor cohort, and 196 days for the group on other antidiabetic medications.
This was not a randomized trial, and the GLP-1 agonist cohort differed from the other two groups in important ways.
For example, they were more likely to be female: 57% of the GLP-1 agonist cohort were women, compared with 44% in the DPP-4 inhibitor cohort and 47% of patients on other antidiabetic agents. The GLP-1 agonist group also had a greater prevalence of obesity: 12%, compared with 8% in the DPP-4 inhibitor cohort and a similar figure among those on other antidiabetic agents. In addition, the GLP-1 agonist cohort had a higher baseline prevalence of microvascular complications.
Only 8% of patients placed on a GLP-1 agonist initiated the drug as new therapy; 78% started it as add-on therapy. In contrast, 17% of patients initiated a DPP-4 inhibitor as new therapy, while 59% employed the drug as add-on therapy.
The subgroup of the GLP-1 agonist cohort who initiated the drug as an add-on to basal insulin had a higher baseline prevalence of microvascular complications and greater use of antilipidemic drugs, antihypertensive agents, and antiplatelet agents than did those using the drug as an add-on to oral antidiabetic therapy. Nevertheless, the 6-month treatment persistence rates in the two subgroups were nearly identical.
The take-away message from this study is clear, Dr. Koro said: "These results demonstrate the need for improved persistence with GLP-1 agonist treatment."
She noted that other investigators have estimated that reducing the rate of nonadherence to antidiabetes medications could avoid 700,000 emergency department visits and 341,000 hospitalizations annually, with a resultant $4.7 billion annual savings in health care costs (Health Aff. [Millwood] 2012;31:1836-46).
GlaxoSmithKline funded Dr. Koro’s study. The company’s investigational GLP-1 agonist albiglutide is under review by the Food and Drug Administration.