The Efficacy of "Inner Peak" an AI-enabled Digital Wellness App on Mental Wellness Among a Sample of University of West Indies (UWI) Mona Students

Brief Summary

This study aims to test the efficacy of therapist-created video content delivered by an Artificial Intelligence (AI) conversational agent and recommendation algorithms (as in the Inner Peak app) among a sample of University of the West Indies, Mona (UWI) students in Jamaica for four weeks. Additionally, we aim to assess if relational accountability and increased users' re-engagement over time will improve mental wellbeing in comparison to a no intervention control group with self-report psychological measures on wellbeing, stress, anxiety and depression. Using a randomized controlled trial on a sample of consented university student volunteers 18 years and older, we sought to ascertain if engagement with an AI-enabled, video content-based mental wellness app will improve mental well-being and other psychological outcomes. In the study, a control group will receive email reminders about existing university mental health resources bi-weekly for 4 weeks. The treatment group will receive access to the Inner Peak mobile application, as well as instructions for how to use it. At the end of the study all participants will be provided with optional access to Inner Peak (waitlist control condition). Using a power analysis a sample of 400 was calculated. Participants' responses will be anonymized using a unique encrypted key in a data protected account. Data collection will extend from December 2025 - July 2026. The encrypted data will be stored on a password-protected Google Drive account, with access limited to designated researchers. Data will be destroyed 5 years after it has been collected. Outcome metrics will include increased re-engagement with the app and improvement in wellbeing, stress, anxiety, and depression. The study will use two clinically validated scales: the 14-item Warwick Edinburgh Mental Wellbeing Scale and the 21-item Depression, Anxiety, and Stress Scale. Descriptive statistics will be used to describe variables: means and standard deviations for continuous variables; frequencies and proportions for categorical variables. T-tests, correlation, and regression analyses will be used to determine the nature and strength of association between specific factors and outcome variables.