(C-103) Treatment of Type 2 Diabetes Among Medicare Beneficiaries With and Without Alzheimer’s Disease: A Retrospective Cohort Study

Background: While many individuals with Alzheimer’s Disease (AD) have Type 2 diabetes (T2DM), little is known regarding whether and how the treatment of Type 2 diabetes varies by AD status.

Objectives: To compare and contrast T2DM treatment among individuals with and without AD.

Methods: We conducted a retrospective cohort study using 20% Medicare Fee-for-Service claims data from January 2016 through December 2020. We included beneficiaries aged ≥66 years with newly diagnosed T2DM and at least 12 months of continuous enrollment prior to diagnosis. Individuals with prior diagnoses of type 1 diabetes or T2DM were excluded. We used the 12 months preceding T2DM diagnosis to identify individuals with and without AD. The primary outcome was initiation of any antidiabetic medication within one year after T2DM diagnosis, and secondary outcomes examining different drug classes. Multivariable logistic regression estimated adjusted odds ratios (ORs) for the association between AD status and treatment initiation, controlling for demographic, socioeconomic, regional, and clinical characteristics.

Results: Among 388,359 Medicare beneficiaries newly diagnosed with T2DM, 9,584 (2.5%) had AD. Patients with AD were older (83.2 ± 7.4 years vs. 75.8 ± 7.3 years) and more frequently female (64.4% vs. 53.4%). Compared to individuals without AD, those with AD were less likely to initiate antidiabetic treatment within one year (41.7% vs. 64.4%). At treatment initiation, patients with AD were more likely to receive insulin (32.3% vs. 18.6%) and less likely to initiate metformin (41.9% vs. 53.4%), SGLT2 inhibitors (0.6% vs. 2.1%), or GLP-1 receptor agonists (0.8% vs. 2.4%). In adjusted models, AD was associated with lower odds of overall treatment initiation (OR 0.55, 95% CI 0.53–0.57) and initiation of newer agents like GLP-1 receptor agonists (OR 0.49, 95% CI 0.33–0.69) and SGLT2 inhibitors (OR 0.47, 95% CI 0.31–0.68).

Conclusions: Among this large, diverse cohort of Medicare beneficiaries in the U.S., there were notable differences in diabetes treatment among older adults with and without AD. Beneficiaries with AD are less likely to initiate antidiabetic therapy, particularly newer agents including SGLT-2 inhibitors and GLP-1 receptor agonists. Future work should explore the basis for these differences, which may reflect a combination of patient, provider and system-level factors.