FRI653 Hemoglobin A1C-testing Inertia: A Barrier Observed In A Digital Diabetes Ecosystem Study

Abstract

Abstract Disclosure: N. Thingalaya: Employee; Self; Sanofi. D. Kerr: Advisory Board Member; Self; Sanofi, Novo Nordisk, Abbott Diabetes Care. Research Investigator; Self; Novo Nordisk, Abbott Diabetes Care. Other; Self; Glooko, Hi.Health, SNAQ. F. Lee: Employee; Self; Sanofi. K.M. Pantalone: Consulting Fee; Self; AstraZeneca, Bayer, Inc., Corcept Therapeutics, Diasome, Eli Lilly & Company, Merck, Novo Nordisk, Sanofi. Research Investigator; Self; Bayer, Inc., Merck, Novo Nordisk, Twin Health. Speaker; Self; AstraZeneca, Corcept Therapeutics, Merck, Novo Nordisk. P. Potukuchi: Employee; Self; Sanofi. J. Unger: None. G. Grunberger: Speaker; Self; Eli Lilly & Company, Abbott Diabetes. Introduction: For people living with type 2 diabetes (T2D), digital health interventions show promise in improving clinical outcomes. Dario Diabetes Solution (DDS) is an application that combines remote self-blood-glucose monitoring, data visualization, and disease education to facilitate behavior change. We designed a retrospective study (RS) to assess clinical effectiveness of DDS on patients with uncontrolled T2D. As clinicians continue to rely on monitoring hemoglobin A1C (A1C) to assess disease progression, this study design gave us insights into the prevalence of A1C-testing inertia in T2D. Methods: Data from DDS users, their A1C and electronic medical records (EMR) were analyzed. Adults with T2D receiving ≥1 diabetes medication, with A1C ≥7.0%, and not using a continuous glucose monitor between 01/01/2017 and 10/31/2021, were eligible for inclusion. Baseline (BL) period was 1 year before the index date, defined as first registration for DDS users or first claim date in the quarter for nonusers. The follow-up period was index date +180 days. Severity of illness was compared using Charlson Comorbidity Index (CCI) score, diabetes medication use, and other comorbidities, including mental health burden such as depression and antidepressant use. We identified test-strip use as a proxy for degree of self-care. By applying sequential 1:3 exact and propensity-score (PS) match methods, we identified 1,005 users and 1,160,825 nonusers at BL from a database linked to claims and EMR. Results: The exact and PS match methods used in the RS generated a well-matched external cohort, with CCI scores of 1.5, 51% of oral antidiabetic-only users, and 6% insulin-only users. There was a well-matched distribution of payer mix, depression, and use of antidepressants. There were 7% more users of test-strips in the nonuser groups (p=0.0026) compared with DDS users at BL. 43.3% of users (n=570) and 45.9% of nonusers (n=628,167; p=0.098) did not have A1C results at 6 months. Conclusion: These findings suggest that the frequency of A1C testing remains sub-standard in the United States. Over 40% of a population with uncontrolled T2D did not have A1C values at 6 months, regardless of their degree of self-care behavior. While not all sources of A1C values were captured, this study ensured comparable disease severity and access to care. A1C was not performed to clinical guideline recommendations in either group. Testing inertia should be further studied to improve implementation of routine A1C testing. Despite use of glycemic targets as measures by payers, frequency of A1C testing remains sub-standard, even among digital health users. Identifying specific barriers to A1C testing could help inform future strategies for digital health solutions, which may help to reduce the existing therapeutic inertia. Presentation: Friday, June 16, 2023