Exploring the interaction between ethnicity, deprivation and the use of CGM on diabetes outcomes—A study from the Association of British Clinical Diabetologists

Abstract

Aims: Ethnic and socioeconomic disparities in diabetes outcomes persist despite widespread adoption of technologies such as continuous glucose monitoring (CGM). We examined the independent and interactive effects of ethnicity and area-level deprivation on baseline glycaemic control and CGM effectiveness in a large UK diabetes cohort. Materials and Methods: This retrospective observational study used data from the Association of British Clinical Diabetologists (ABCD) national audit, including 18,139 adults with type 1 or type 2 diabetes initiating FreeStyle Libre. Ethnicity was categorised into White, Black, Indian, Pakistani/Bangladeshi, and Mixed/Other groups. Deprivation was assessed using hospital-level Index of Multiple Deprivation (IMD) rankings. Baseline and follow-up HbA1c and diabetes distress scores (DDS) were analysed. Linear regression assessed associations between predictors and HbA1c, and interaction effects were evaluated using both regression and Gradient Boosting Machine (GBM) models, with Friedman's H-statistic quantifying interaction strength. Results: Compared to White individuals, Black participants had significantly higher baseline HbA1c (β = 6.19, SE = 1.87, p < 0.001), while Indian participants had lower levels (β = −3.50, SE = 1.37, p = 0.01). Compared to those in low deprivation areas, baseline HbA1c was significantly higher in high (β = 3.38, p < 0.001), moderate (β = 2.90, p < 0.001), and very high (β = 4.54, p < 0.001) deprivation categories. The interaction between ethnicity and deprivation was not statistically significant (p > 0.05); GBM analysis supported this with a low H-statistic of 0.088. CGM use resulted in significant HbA1c reductions across all IMD groups, most notably in very high deprivation areas (−8.54%, p < 0.001). Reduction was similar across ethnicities (all p = 1.0). Conclusions: Ethnicity and deprivation independently influence glycaemic control. However, CGM yields equitable HbA1c improvements across all subgroups.