A Personalized Data Dashboard to Improve Compliance with Ecological Momentary Assessments: Protocol for a Study to Compare EMA as Usual to EMA with a Data Dashboard Deployed in a Micro-Randomized Trial (Preprint)

Abstract

BACKGROUND Background: Ecological momentary assessments (EMA) are ideal for capturing the dynamic nature of young adult substance use behavior in daily life and identifying contextual risk factors that signal higher-risk episodes. These methods could provide a signal to trigger real-time intervention delivery. Study compliance and engagement are common barriers to participation, but may be improved by personalizing messages. This study compares compliance outcomes between one group of young adults receiving standard (generic) prompts at each assessment and another group that received additional personalization and an updated data dashboard (DD) showing study progress to date at one randomly selected prompt per day. OBJECTIVE Objective: The primary objectives are to (a) develop a real-time DD for giving participants personalized updates on their progress in the study and (b) examine its preliminary overall effects on study compliance and experiences. Secondary objectives are to identify person-, day-, and moment-level characteristics associated with study compliance and person-level characteristics associated with perceived usefulness of the DD. METHODS Methods: This is a protocol for Project ENGAGE, a 2-arm randomized controlled trial. Arm 1 (EMA Group) is engaged in a standard EMA protocol, and Arm 2 (EMA + DD Group) is engaged in the same study but with additional personalization and feedback. Inclusion criteria are (1) previous participation in a recent college student survey about health behavior and mental health who indicated willingness to participate in future research studies and (2) indicated past-month alcohol use and/or lifetime marijuana/hashish/delta-8 THC use on that survey. All participants in the current study completed a baseline survey; EMA at 11am, 2pm, 5pm, and 8pm each day for 21 days; and an exit survey. Participants in Arm 2 engaged in a micro-randomized trial, receiving a personalized DD at one randomly selected prompt per day. Primary outcomes include whether a survey was completed, time to complete a survey, and subjective experiences in the study. Primary analyses will compare groups on overall study compliance and, for Arm 2, use marginal models to assess the momentary effect of receiving one updated DD per day. RESULTS Results: Approval was granted by the university’s Institutional Review Board on 2/8/23. Recruitment via direct email occurred on March 30 and April 6, 2023; data collection was completed by April 29, 2023. A total of 91 individuals participated in the study. CONCLUSIONS Conclusions: Results from the evaluation of this study will indicate whether providing (at randomly selected prompts) real-time, personalized feedback on a participant’s progress in an EMA study improves study compliance. Overall, this study will inform whether a simple, automated DD presenting study compliance and incentives earned to date may improve young adults’ compliance and engagement in intensive longitudinal studies. CLINICALTRIAL ENGAGE was not considered a clinical trial, and thus was not registered anywhere.It was a feasibility/pilot study.