SAT135 Prospective Analysis For Improvement For Glycemic Control Using A Patient Portal And Automated Messages

Disclosure: C.C. Cristancho: None. M. Brandao: None. A. Lema Morocho: None. T. Amorim: None. Background: Type II diabetes is a complex metabolic disorder which is influenced in part by patient dietary choices, medication adherence, and environmental factors. Behavioral health interventions offer an opportunity to potentially affect patient outcomes in a cost-effective manner. Purpose: To study whether automated clinical messaging via the electronic medical chart would be associated with improvement in diabetes-related health measures. Methods: We created an automated clinical messaging system (January 2020) via our electronic medical record. Patients who opted-in for this service received bi-weekly messages which included reminders about diabetes medication adherence as well as exercise and dietary recommendations. We then performed a retrospective longitudinal analysis of patients followed in our diabetes clinic, comparing those who opted-in for messing vs those who did not, and assessed diabetes-related biomarkers including baseline and follow-up hemoglobin A1c, lipid markers, and BMI using linear mixed-effects regression models. Results: 69 patients who had at least 2 measures of hemoglobin A1c were followed longitudinally, of whom 37 (54%) had opted to receive automated messaging. Patients who opted-in were younger (53±12 vs 62±11 years old, p<0.1), but not different in terms of sex (60% F vs 61% F), BMI (31±6 vs 30±7, p=0.84), or A1c (9.4±2.3 vs 8.8±2.7, p=0.34). In the group that received messages, we did not find any significant pre-post differences in A1c, LDL, or BMI over time. In the opt-in group, mean change (over average follow-up of 256 days) in HgbA1c was -0.4 (p=0.42) compared to a change of 0.0 in the non-opt-in group (p=0.99). Discussion: Patients who opted-in to electronic messaging were younger on average. We did not observe any significant longitudinal change in diabetes-related biomarkers within or between groups as related to an automated clinical messaging service, although there was a directional trend toward improvement. Given the cost-effectiveness and ease of this program, consideration should be given toward a trial with a larger sample size and longer follow-up. Presentation: Saturday, June 17, 2023