2025 Southeastern Residency Conference

Title: Medication Adherence for Diabetes (MAD) Quality Measure and Correlation to A1c and Other Diabetes Measures in a Medicare Advantage Population 
Authors: Rachel Winters, Amanda Moyer, Sarah Blandy, Victoria Hetherington, and Casey Penland 
Background: The Centers for Medicare and Medicaid Services (CMS) provides financial incentives for insurance plans demonstrating high-quality care to improve patient outcomes. Medicare Advantage (MA) plan performance is assessed by nearly 40 quality measures, with diabetes care heavily represented by 5 of those measures. In accordance with ADA clinical practice guidelines, holistic care is evaluated using Diabetes Care – Eye Exams, Kidney Health Evaluation in Patients with Diabetes, and Statin Use in Persons with Diabetes. Further, Medication Adherence for Diabetes (MAD) and Diabetes Care – Blood Sugar Controlled are emphasized as triple-weighted quality measures impacting the overall star rating. The national average ratings for Medicare Advantage Prescription Drug (MA-PD) plans from 2022-2025 show a flat average pass rate of 86% four years in a row for the MAD quality measure but a steadily increasing pass rate each year (78%, 78%, 79%, 83%) for controlled blood sugar.  This leads to speculation that the MAD measure may not correlate with other positive health outcomes in diabetic patients.  
Methods: This is a retrospective cohort study including Aetna Medicare Advantage beneficiaries attributed to a Prisma Health primary care provider under contract with the inVio Health Network who indexed into the MAD and subsequent diabetes control quality measures in the 2023 measurement year. Data will be collected from January 1, 2023 to December 31, 2023. Notable exclusion criteria are those excluded from the MAD quality measure by definition (end stage renal disease, insulin use, or hospice patients). The primary endpoint is to identify the relationship between MAD performance and A1c control, comparing both the Diabetes Care – Blood Sugar Controlled quality measure and patients’ actual A1c. Secondary endpoints include the correlation between MAD and additional diabetes care quality measures including eye exams, kidney health evaluation, and statin use. Tertiary endpoints include factors affecting MAD measure performance such as low-income subsidy, dual eligibility, indexing medication class, and number of classes prescribed. 
Results: There was no statistically significant difference found between MAD pass/fail rates and the clinical definition of A1c control (<7%) (p=0.98). Statistical significance was observed between MAD and annual eye exams (EED) (p<0.05) as well as annual kidney health evaluation (KED) (p< 0.05). However, statistical significance was not observed between MAD and the HBD (p=0.31) or SUPD (p=0.30) quality measures. Most patients in this study were well-controlled on 1-2 diabetes medications with an A1c <7%.
Conclusion: In this study population, MAD performance does not impact actual A1c, the triple-weighted HBD quality measure, or the clinical definition of A1c control.