Characteristics of insulin prescriptions and their association with glycemic control in adults with type 2 diabetes mellitus

Abstract

Objective: We assessed the association between insulin prescription characteristics and glycemic control. Methods: Electronic health records (2017-2023) were used to identify adults with type 2 diabetes prescribed insulin from outpatient visits on the same day as a hemoglobin A1c (HbA1c) result in a Midwest US region. Insulin prescription characteristics were total daily dose, providing a maximum dose, and instruction types (dosing frequency, carbohydrate (carb) counting, correction, and sliding scale). Mixed-effects linear and logistic regression modeled HbA1c and meeting HbA1c target (<7 for age18-64 and <8 for age ≥ 65). Results: Among 11,179 subjects, mean age was 53; 49 % were male, and 58 % White. Overall cohort glycemic control was low, median HbA1c was 9.3; 16 % met HbA1c target; mean HbA1c at last follow-up was 8.7 %. Instruction types were 88 % dosing frequency, 7 % correction, 3 % sliding scale, and 1 % carb counting. HbA1c reduction was associated with carb counting (-0.4, p = 0.009), correction (-0.2, p < 0.001), sliding scale (-0.1, p = 0.014), and providing a maximum daily insulin dose (-0.02, p < 0.001). An increase in total daily insulin dose by 10 units was associated with a 0.001 increase in HbA1c (p = 0.003). Correction instruction was 1.4 times more likely than dosing frequency to meet HbA1c target (p < 0.001). Conclusion: Insulin instruction type was associated with glycemic control, but control was low.