African Americans with type 2 diabetes (T2D) have higher average A1c levels than White patients. However, few studies have examined racial disparities in diabetes management in primary care, particularly provider-level variability. Study goals were to analyze racial differences for patients with any/2 or more elevated A1cs, explore patterns of visits/providers seen in patients with ≥1 elevated A1c, and explore the contributions of provider variability in patient A1c. A retrospective secondary analysis of electronic medical record data from a large urban health system was conducted, involving adult African American or White patients (ages18-65 years) with ≥2 measured A1cs between January 1, 2017-February 1, 2018. Descriptive statistics were calculated for demographic variables; paired t tests evaluated changes in A1c levels across the 2 most recent measurements, and a repeated measures ANOVA evaluated the impact of race on A1c changes. Logistic regression analyses examined the relationship of race with any elevated A1c levels and persistent A1c levels (≥ 2 consecutive A1c measurements ≥8.5). The intraclass correlation coefficient (ICC) estimated clustering of A1c by provider. A total of 1764 patients were included. African Americans were more likely to have any (odds ratio [OR] = 1.48, P < .001) and persistently elevated A1c (OR = 1.75, P = .0003). ICC was .27 for any elevated A1c and .32 for persistently elevated A1c. In primary care patients with T2D, African Americans were more likely than Whites to have any/persistently elevated A1c, with substantial variability attributable to the provider. Further research is needed to better understand patient- and provider-level contributors to A1c disparities.
Keywords: diabetes; disparities; hemoglobin A1c; primary care.