Online Interventions to Prevent Alcohol Use Disorders

Plain-language summary

Background: Internet-based interventions can improve access to non-treatment-seeking populations, preventing the onset or progression of alcohol use disorder (AUD). Stepped-care guidelines for face-to-face AUD interventions recommend internet-based Brief Intervention (iBI) or unguided Cognitive Behavioural Therapy (iCBT) for no or mild AUD, and guided iCBT for moderate to severe AUD. However, no large-scale superiority trial has compared the effectiveness of these interventions among non-treatment-seeking individuals across the full spectrum of problematic alcohol use. Aims: 1) Compare the effectiveness of iBI, unguided, and guided iCBT in reducing alcohol consumption in non-treatment-seeking individuals with sub-threshold or full AUD; 2) develop models via machine learning for personalized AUD prevention and progression management. Methods: A nationwide sample of 3519 individuals will be stratified by sub-threshold/mild AUD and moderate/severe AUD and randomized to: 1) online assessment (OA)+ iBI; 2) OA+ unguided iCBT; or 3) OA+ guided iCBT. The iCBT sessions will address problematic alcohol use and co-occuring externalizing and internalizing psychiatric symptoms. Data will be collected from OA, interventions, and Danish registries at baseline and 3-, 6-, 12-, and 24-month follow-ups, with registry follow-up over 10 years. Perspectives: Findings will compare stepped-care and machine learning-driven personalized approaches to inform guidelines for non-treatment-seeking populations. Internet-based assessment and interventions support continuous data collection, enabling ongoing improvements and personalized prevention. This large-scale dissemination targeting non-treatment-seeking populations across the full spectrum of problematic alcohol use will pave the way for future initiatives and may refine prevention strategies if the stepped-care model proves insufficient for this group. Key words: Alcohol Use Disorder, Internet-Based Interventions, Machine Learning, Non-treatment Seekers, Stepped-Care

Who can participate

What they're measuring

Trial details