Protocol for a bandit-based response adaptive trial to evaluate the effectiveness of brief self-guided digital interventions for reducing psychological distress in university students: The Vibe Up Study

Abstract

Introduction: Meta-analytic evidence confirms a range of in- terventions, including mindfulness, physical activity and sleep hygiene, can reduce psychological distress in university stu- dents. However, it is unclear which intervention is most ef- fective. Artificial intelligence (AI) driven adaptive trials may be an efficient method to determine what works best and for whom. The primary purpose of the study is to rank the effec- tiveness of mindfulness, physical activity, sleep hygiene and an active control on reducing distress, using a multi-arm con- textual bandit-based AI-adaptive trial method. Furthermore, the study will explore which interventions have the largest ef- fect for students with different levels of baseline distress sever- ity. Methods and analysis: The Vibe Up study is a pragmatically-oriented, decentralised AI-adaptive group se- quential randomised controlled trial (RCT) comparing the ef- fectiveness of one of three brief, two week digital self-guided interventions (mindfulness, physical activity, or sleep hygiene) or active control (ecological momentary assessment) in reduc- ing self-reported psychological distress in Australian univer- sity students. The adaptive trial methodology involves up to 12 sequential mini-trials that allow for the optimisation of al- location ratios. The primary outcome is change in psycho- logical distress (DASS-21 total score) from pre-intervention to post-intervention. Secondary outcomes include change in depression, anxiety, and stress (measured by DASS-21 sub- scales) from pre-intervention to post-intervention. Planned contrasts will compare the four groups (i.e., the three interven- tion and control) using self-reported psychological distress at pre-specified time points for interim analyses. The study aims to determine the best performing intervention, as well as rank- ing of other interventions. Ethics and dissemination: Ethical approval was sought and obtained from the UNSW Sydney Human Research Ethics Committee (HREC A, HC200466). A trial protocol adhering to the requirements of the Guideline for Good Clinical Prac- tice [1] was prepared for and approved by the Sponsor, UNSW Sydney (Protocol number: HC200466_CTP). Registration details: The trial is registered with the Australian New Zealand Clinical Trials Registry (AC- TRN12621001223820).

Machine learning identifies a COVID-19-specific phenotype in university students using a mental health app

“…The present analysis was performed using participant screening data collected in the context of an adaptive trial of a smartphone-based program (‘Vibe Up’) that delivers brief mental health interventions to Australian-based university students ( Huckvale et al, 2023 ). Using an Artificial Intelligence (AI)-driven response adaptive randomization design, the trial used a series of sequential ‘mini trials’ to identify which of four, two-week smartphone interventions (mindfulness, physical activity, or sleep hygiene, with a mood monitoring intervention used as an active control condition) was the most effective for reducing symptoms of psychological distress in university students.…”

“…Prospective participants completed a battery of self-report questionnaires in the Vibe Up smartphone application as part of screening, and scores on these questionnaires were used as the features on which our machine learning models were trained. The questionnaires included the Abridged NIDA-Modified ASSIST Drug Screening Tool (AOD) ( National Institute on Drug Abuse, 2020 ), EuroQuol-5D 5-level version (EQ-5D-5L) ( Herdman et al, 2011 ), Kessler Psychological Distress Scale 10-item version (KTEN) ( Kessler et al, 2002 ), Use of Mental Health Care Services (MHS) ( Burgess et al, 2009 ), Perceived Stress Scale (PSS) ( Cohen et al, 1994 ), Physical and Mental Health (MED) ( Huckvale et al, 2023 ), Suicidal Ideation Attributes Scale (SIDAS) ( van Spijker et al, 2014 ), and Short Warwick Edinburgh Mental Well-being Scale (WBS) ( Stewart-Brown et al, 2009 ) (see Supplementary Methods for further details). Scores on all questionnaires were standardized by converting to z-scores prior to analyses.…”

Machine learning identifies a COVID-19-specific phenotype in university students using a mental health app

“…The present analysis was performed using participant screening data collected in the context of an adaptive trial of a smartphone-based program ('Vibe Up') that delivers brief mental health interventions to Australian-based university students 22 . Using an Artificial Intelligence (AI)-driven response adaptive randomization design, the trial used a series of sequential 'mini trials' to identify which of four, two-week smartphone interventions (mindfulness, physical activity, or sleep hygiene, with a mood monitoring intervention used as an active control condition) was the most effective for reducing symptoms of psychological distress in university students.…”

“…Prospective participants completed a battery of self-report questionnaires in the Vibe Up smartphone application as part of screening, and scores on these questionnaires were used as the features on which our machine learning models were trained. The questionnaires included the Abridged NIDA-Modified ASSIST Drug Screening Tool (AOD) 23 , EuroQuol-5D 5-level version (EQ-5D-5L) 24 , Kessler Psychological Distress Scale 10-item version (KTEN) 25 , Use of Mental Health Care Services (MHS) 26 , Perceived Stress Scale (PSS) 27 , Physical and Mental Health (MED) 22 , Suicidal Ideation Attributes Scale (SIDAS) 28 , and Short Warwick Edinburgh Mental Well-being Scale (WBS) 29 (see Supplementary Methods for further details). Scores on all questionnaires were standardized by converting to z-scores prior to analyses.…”