202-OR: Basal Insulin Digital Titration App vs. Enhanced Paper Titration Tool: A Randomized Control Study

Self-adjusting insulin is challenging for people with type 2 diabetes (PWT2D). Mobile health technologies (mHealth) can help PWT2D track blood glucose more easily, but it remains unclear whether an mHealth-enabled digital therapeutic tool can effectively support basal insulin (BI) titration in a large multispecialty health care setting. We randomized 242 PWT2D on BI (baseline HbA1c 7.5% - 12.5%) to use either an app-based self-titration tool [Mobile Insulin Dosing System (MIDS)] or an enhanced paper titration tool based on a stepped algorithm with diabetes educator support (control) for 16 weeks. Cohort characteristics for the intent to treat sample (MIDS n = 117; control n = 120) were: median age 61 years [IQR: 53 - 69], 41.7% female, 75.9% non-Hispanic White, 11.4% new to insulin, median duration of diabetes 11 years [IQR: 7 - 18], and median baseline HbA1c of 8.7% [IQR: 8.0 - 9.6]. Improvements in HbA1c (-1.3% [IQR: -2.2 - -.5] and -1.2% [IQR: -1.9 - -.5]; both Ps < .05) with increased median insulin dose (+8 IU [IQR: 2 - 22] and +10 IU [IQR: 2 - 25.5]; both Ps < .05) and no change in hypoglycemia were observed at 16 weeks in both the MIDS and control groups, respectively. No differences in HbA1c (P = .48) or insulin dose change (P = .34) were observed between the groups. Among SMBG readings recorded during the study, the mean proportion of readings between 70-180 mg/dL (76.8% ± 22.0; 70.0% ± 26.7; p250 mg/dL (4.9% ± 9.2; 9.2% ± 18.4; p