A Faith-Based Mobile Application for Mental Health and Well-Being Among University Students: A Single-Arm Feasibility and Preliminary Effectiveness Trial

This study is a single-arm feasibility and preliminary effectiveness trial conducted with currently enrolled students at Regent University. Participants are recruited via a university-wide email listserv and are asked to use the PRAY.COM app, a faith-based mobile application delivering Daily Prayer, Meditative Prayers, Scripture Readings, Bedtime Bible Stories, Podcasts, Sermons, a Bible in a Year program, and Pray AI (an AI-powered faith companion), for at least 10 minutes per day over 8 weeks. No instructions are given regarding which specific features or content to engage with. Upon completing the baseline survey, participants receive a complimentary premium PRAY.COM subscription for the duration of the study.

Feasibility is assessed across demand and acceptability domains. Demand is operationalized as the percentage of participants reporting at least weekly app use with a benchmark of ≥70%. Acceptability is operationalized as the percentage of participants reporting satisfaction (satisfied or very satisfied) with a benchmark of ≥75%. Additional feasibility indicators include perceived appropriateness for students and intent to continue use. Objective app usage data are obtained directly from PRAY.COM to complement self-reported engagement.

Preliminary effectiveness outcomes are assessed using validated measures at baseline, 4 weeks, and 8 weeks. Mental health outcomes include depressive symptoms, anxiety, and stress (DASS-9; 3 items each, 0-3 scale), and loneliness (UCLA-3; 3 items, 1-3 scale). Well-being outcomes include sense of purpose and meaning (2 items adapted from the Claremont Purpose Scale, 1-5 scale). Academic outcomes include time management confidence, school-life balance satisfaction, confidence managing academic challenges, feeling overwhelmed by academic responsibilities, classes missed due to mental health, difficulty concentrating, frequency of mental health negatively impacting academic performance, and difficulty completing coursework (8 investigator-developed items).

Statistical approach: Pre-post changes are analyzed using Wilcoxon signed-rank tests with normality of change scores assessed via Shapiro-Wilk tests. Effect sizes are calculated as Cohen's d (mean change / SD of change scores). Three analysis groups are examined: 1) the primary per-protocol sample of 8-week completers with confirmed objective app engagement (i.e., having at least one recorded app session exceeding one second); 2) a secondary analysis of all 8-week completers regardless of objective engagement status; and 3) a tertiary full-enrollment analysis using last observation carried forward (LOCF) imputation as a conservative lower-bound estimate, with multiple imputation as a sensitivity check. Additional analyses examine dose-response relationships, subgroup differences (age, sex, race/ethnicity, student type), and baseline severity effects.