Effectiveness of engagement promoting strategies compared to no strategies in a brief intervention app targeting harmful drinking in young adults: Randomised controlled trial (Preprint)

BACKGROUND Electronic screening and brief intervention (eSBI) has demonstrated some effectiveness in reducing harmful alcohol consumption. However, a challenge remains as to how to sufficiently engage users with app and online based eSBI, as trials demonstrate low engagement rates. Young adults are an under-identified harmful drinking population who prefer eSBI delivered alcohol interventions. Engagement has been associated with improved clinical outcomes in the digital behaviour change literature. Therefore, identifying novel methods to engage users is warranted to improve the effectiveness of app-based eSBI interventions. OBJECTIVE (i) To conduct a Randomised Controlled Trial (RCT) to determine whether eSBI plus novel Engagement Promoting Strategies (EPS) compared to eSBI only was associated with higher levels of user engagement in an app targeting harmful drinking in young adults (ii) To examine the relationship between the inclusion of EPS and changes in harmful alcohol consumption and, (iii) To determine the relationship between participants who engaged and those who did not engage with the app and change in AUDIT-C score. METHODS An app called ‘BRANCH’ was developed, including EPS of tailoring, social features, in-app reminders and gamification. A two-arm RCT examined whether participants in the eSBI plus EPS arm (intervention) logged into BRANCH more times than the eSBI only arm (control). Participants were young adults (>=18 years old and <=30 years old) who drank alcohol at a harmful level (scored >=7 on the AUDIT-C), and lived in the UK. The primary outcome was number of logins at 28 days. Secondary outcomes included AUDIT-C at six months and engagement measured by treatment compliance (logged in > 1). RESULTS The addition of EPS to eSBI in an app targeting harmful drinking in young adults did not result in increased use, measured by number of logins compared, to the same app including only eSBI components 28 days post randomisation (P=.60). Users of the eSBI plus EPS app did not show a significant reduction in harmful drinking score as measured by the AUDIT-C score compared to users of the eSBI only app at 6-month post randomisation (P=.17). App engagers (logged in > 1) did not show a significant reduction in harmful drinking compared to non-engagers (logged in once) (P=.58). CONCLUSIONS An eSBI app optimised with tailored, social and gamification features did not promote increased usage compared to eSBI only. Engagement with eSBI in young adults continues to be a barrier for designing effective eSBI apps. The associations between EPS, engagement rates and drinking outcomes remain unclear. The low engagement rates with some eSBI apps may be leading to a lack of exposure to the active behaviour change components, resulting in limited effects. CLINICALTRIAL ISRCTN: 70980706