Predictors of Hemoglobin A1c Testing in Adults with Type 2 Diabetes Covered by Alabama Medicaid

Background: ADA Standards of Care recommend A1C tests at least twice per year for patients with diabetes. We evaluated predictors of A1C testing among adults with type 2 diabetes covered by Alabama Medicaid. Methods: We conducted a retrospective analysis of Medicaid claims data between 2011-2020 using person-year observations as the unit of analysis. Inclusion criteria were ages 19-64 years, type 2 diabetes diagnosis, continuous Medicaid enrollment for the full calendar year and the year prior. The primary outcomes were receiving ≥1 and ≥2 A1C test (s) during a calendar year. We conducted multivariable Poisson regression stratified by eligibility (i.e. disability, poverty) to evaluate the relationships of study year, demographics, clinical factors, and healthcare utilization with A1C tests using generalized estimating equations to account for individuals contributing >1 observations. Results: We analyzed 288,379 person-year observations; 69% were female; 49% were Black; 47% were 35-54 years old, 44% 55-64 years old; 51% were eligible based on disability, 49% based on poverty. Overall, 57% observations had ≥1 A1C test per year, 35% had ≥2 A1C tests. More observations eligible by disability had ≥1 A1C test (76% vs. 38%) and ≥2 A1C tests (49% vs. 20%) than those eligible by poverty. For observations eligible by disability, Black race and older age were associated with higher likelihood of having ≥1 A1C test. For observations eligible by poverty, year after 2011, female sex, and younger age were associated with higher likelihood of having ≥1 A1C test. For both groups, rurality, insulin use, endocrinology care, diabetes complications, and ambulatory care visits were associated with higher likelihood of having ≥1 A1C test. Conclusions: Just over one-third of adults with diabetes covered by Alabama Medicaid had at least 2 A1C tests per year. A1C testing frequency differed by Medicaid eligibility; poverty-based eligibility was associated with lower likelihood of having recommended A1C testing. Disclosure C.A.Presley: None. C.R.Howell: None. Y.Khodneva: None. K.R.Riggs: n/a. L.Huang: None. E.B.Levitan: Research Support; Amgen Inc. A.Cherrington: Consultant; Bayer AG, Other Relationship; Novo Nordisk. Funding National Institute of Diabetes and Digestive and Kidney Diseases R18DK109501PI: CherringtonUAB Diabetes Research Center P30 DK079626PI: Cherrington