Real-world study of ethnic differences in glycaemic control and clinical characteristics among insulin-naïve people with type 2 diabetes initiating biphasic insulin aspart 30/70: A retrospective, observational cohort study in England.

Abstract

This study investigated the ethnic differences in glycaemic levels and clinical characteristics among insulin-naïve people with type 2 diabetes (T2D) initiating biphasic insulin aspart 30/70 (BIAsp 30) in primary practice in England. Retrospective, observational cohort study utilizing data from the Clinical Practice Research Datalink Aurum database, including White, South Asian, Black and Chinese insulin-naïve adults with T2D, initiating BIAsp 30. The index date was that of the first BIAsp 30 prescription. Endpoints included change in glycated haemoglobin (HbA1c) and body mass index (BMI) 6 months post index. In total, 11 186 eligible people were selected (9443 White, 1116 South Asian, 594 Black, 33 Chinese). HbA1c decreased across all subgroups 6 months post index: estimated %-point changes [95% CI of -2.32 (-2.36; -2.28) (White); -1.91 (-2.02; -1.80) (South Asian); -2.55 (-2.69; -2.40) (Black); and -2.64 (-3.24; -2.04) (Chinese)]. The BMI increased modestly 6 months post index in all subgroups [estimated changes (95% CI) kg/m2 : White, 0.92 (0.86; 0.99); South Asian, 0.60 (0.41; 0.78); Black, 1.41 (1.16; 1.65); and Chinese, 0.32 (-0.67; 1.30)]. In the overall population, hypoglycaemic event rates increased from 0.92 events per 100 patient-years before the index to 3.37 events per 100 patient-years post index; event numbers were too low to be analysed by subgroup. Among insulin-naïve people with T2D initiating BIAsp 30, clinically meaningful HbA1c reductions in all ethnicities were observed. Some ethnic groups had larger reductions than others, but differences were small. In all groups, small BMI increases were seen, with small differences observed between groups. Hypoglycaemia rates were low.