Comparison of Diabetes Medications Used by Adults With Commercial Insurance vs Medicare Advantage, 2016 to 2019
Rozalina G McCoy,Holly K Van Houten,Yihong Deng,Pinar Karaca Mandic,Joseph S Ross,Victor M Montori,Nilay D Shah
DOI: https://doi.org/10.1001/jamanetworkopen.2020.35792
2021-02-01
Abstract:Importance: Glucagonlike peptide-1 receptor agonists (GLP-1RA), sodium-glucose cotransporter-2 inhibitors (SGLT2i), and dipeptidyl peptidase-4 inhibitors (DPP-4i) are associated with low rates of hypoglycemia, and postmarketing trials of GLP-1RA and SGLT2i demonstrated that these medications improved cardiovascular and kidney outcomes. Objective: To compare trends in initiation of treatment with GLP-1RA, SGLT2i, and DPP-4i by older adults with type 2 diabetes insured by Medicare Advantage vs commercial health plans. Design, setting, and participants: This retrospective cohort study used administrative claims data from a deidentified database of commercially insured and Medicare Advantage beneficiaries. Adults aged 58 to 66 years with type 2 diabetes who filled any medication prescription to lower glucose levels from January 1, 2016, to December 31, 2019, were compared between groups. Exposure: Enrollment in a Medicare Advantage or commercial health insurance plan. Main outcomes and measures: The odds of initiating GLP-1RA, SGLT2i, and DPP-4i treatment were examined for Medicare Advantage vs commercial insurance beneficiaries using 3 separate logistic regression models adjusted for year and demographic and clinical factors. These models were used to calculate adjusted annual rates of medication initiation by health plan. Results: A total of 382 574 adults with pharmacologically treated type 2 diabetes (52.9% men; mean [SD] age, 62.4 [2.7] years) were identified, including 172 180 Medicare Advantage and 210 394 commercial beneficiaries. From 2016 to 2019, adjusted rates of initiation of GLP-1RA, SGLT2i, and DPP-4i treatment increased among all beneficiaries, from 2.14% to 20.02% for GLP-1RA among commercial insurance beneficiaries and from 1.50% to 11.44% among Medicare Advantage beneficiaries; from 2.74% to 18.15% for SGLT2i among commercial insurance beneficiaries and from 1.57% to 8.51% among Medicare Advantage beneficiaries; and from 3.30% to 11.71% for DPP-4i among commercial insurance beneficiaries and from 2.44% to 7.68% among Medicare Advantage beneficiaries. Initiation rates for all 3 drug classes were consistently lower among Medicare Advantage than among commercial insurance beneficiaries. Within each calendar year, the odds of initiating GLP-1RA treatment ranged from 0.28 (95% CI, 0.26-0.29) to 0.70 (95% CI, 0.65-0.75) for Medicare Advantage and commercial insurance beneficiaries, respectively; SGLT2i, from 0.21 (95% CI, 0.20-0.22) to 0.57 (95% CI, 0.53-0.61), respectively; and DPP-4i, from 0.37 (95% CI, 0.34-0.39) to 0.73 (95% CI, 0.69-0.78), respectively (P < .001 for all). The odds of starting GLP-1RA and SGLT2i increased with income; for an income of $200 000 and higher vs less than $40 000, the odds ratio for GLP-1RA was 1.23 (95% CI, 1.15-1.32) and for SGLT2i was 1.16 (95% CI, 1.09-1.24). Conclusions and relevance: These findings suggest that Medicare Advantage beneficiaries may be less likely than commercially insured beneficiaries to be treated with newer medications to lower glucose levels, with greater disparities among lower-income patients. Better understanding of nonclinical factors contributing to treatment decisions and efforts to promote greater equity in diabetes management appear to be needed.