Challenges to Guideline-Recommended Diabetes Care in the United States

Modern guidelines recommend novel, effective therapies such as sodium-glucose cotransporter 2 inhibitors (SGLT2i), glucagon-like peptide 1 receptor agonists (GLP1RA), and non-steroidal mineralocorticoid antagonists (MRA) in people with diabetes. However, there are significant gaps in adherence to guideline-recommended care. Some groups face a disproportionate burden of diabetes and likely have larger gaps in care in diabetes. The high cost of newer medications may further exacerbate these differences. However, this variability in quality of diabetes care and outcomes across age, socioeconomic status, and race/ethnicity and their complex interplay have not been well characterized. Challenges to engaging in care are also understudied. Little is known about what prevents physicians from prescribing guideline-recommended monitoring and treatment. The overarching goal of this proposed study is to address the challenges that prevent provision of guideline-recommended diabetes care in the United States. This study will leverage rich real-world datasets on persons with diabetes from OptumLabs Data Warehouse (OLDW, n ≈ 8.1 M), Medicare (n ≈ 3.1 M), and Medicaid (n ≈ 6.2 M) and survey a national sample of physicians (n ≈ 1000) with three physician focus groups (n ≈ 24) to achieve the following specific aims: describing differences in care receipt by individual-level, neighborhood-level, and insurance-level characteristics (Aim 1), examining differences in outcomes and the extent to which they associate with care receipt with traditional mediation analyses as well as novel causal decomposition methods (Aim 2), and investigating physician perspectives on multilevel challenges to care (Aim 3). The proposed work will represent the largest number of people with diabetes in the United States with different socioeconomic backgrounds. This research endeavor, informed by national priorities, seeks to advance knowledge, inform policies, and improve diabetes care.