Learning in a Virtual Environment to Improve Type 2 Diabetes Outcomes: Randomized Controlled Trial

Abstract

Background Given the importance of self-management in type 2 diabetes mellitus (T2DM), a major aspect of health is providing diabetes self-management education and support. Known barriers include access, availability, and the lack of follow through on referral to education programs. Virtual education and support have increased in use over the last few years. Objective The purpose of the Diabetes Learning in a Virtual Environment (LIVE) study was to compare the effects of the LIVE intervention (educational 3D world) to a diabetes self-management education and support control website on diet and physical activity behaviors and behavioral and metabolic outcomes in adults with T2DM over 12 months. Methods The LIVE study was a 52-week multisite randomized controlled trial with longitudinal repeated measures. Participants were randomized to LIVE (n=102) or a control website (n=109). Both contained the same educational materials, but the virtual environment was synchronous and interactive, whereas the control was a flat website. Data were collected at baseline and 3, 6, and 12 months using surveys and clinical, laboratory, and Fitbit measures. Descriptive statistics included baseline characteristics and demographics. The effects of the intervention were initially examined by comparing the means and SDs of the outcomes across the 4 time points between study arms, followed by multilevel modeling on trajectories of the outcomes over the 12 months. Results This trial included 211 participants who consented. The mean age was 58.85 (SD 10.1) years, and a majority were White (127/211, 60.2%), non-Hispanic (198/211, 93.8%), married (107/190, 56.3%), and female (125/211, 59.2%). Mean hemoglobin A1c (HbA1c) level at baseline was 7.64% (SD 1.79%) and mean BMI was 33.51 (SD 7.25). We examined weight loss status versus randomized group, where data with no weight change were eliminated, and the LIVE group experienced significantly more weight loss than the control group (P=.04). There were no significant differences between groups in changes in physical activity and dietary outcomes (all P>.05), but each group showed an increase in physical activity. Both groups experienced a decrease in mean HbA1c level, systolic and diastolic blood pressure, cholesterol, and triglycerides over the course of 12 months of study participation, including those participants whose baseline HbA1c level was 8.6% or higher. Conclusions This study confirmed that there were minor positive changes on glycemic targets in both groups over the 12-month study period; however, the majority of the participants began with optimal HbA1c levels. We did find clinically relevant metabolic changes in those who began with an HbA1c level >8.6% in both groups. This study provided a variety of resources to our participants in both study groups, and we conclude that a toolkit with a variety of services would be helpful to improving self-care in the future for persons with T2DM. Trial Registration ClinicalTrials.gov NCT02040038; https://clinicaltrials.gov/ct2/show/NCT02040038