App-based skills training to reduce problem drinking among adult internet help-seekers: a double-blinded randomized controlled trial

Problem drinking (PD), defined as hazardous or harmful drinking, constitutes a major health and economic burden worldwide. However, most individuals with PD typically remain untreated for many years as part of the “treatment gap” between prevalence and treatment access. The main objective of this study was to evaluate an unguided, app-based intervention for reducing alcohol consumption, compared to an information control app. Secondary analyses targeted motivation effects, subgroup differences in heavy drinking days, and the proportion of participants with excessive drinking. In this two-armed randomized controlled trial (RCT), Swedish-speaking adults seeking help for PD on the Internet were recruited via online ads (n = 576). Eligible participants, defined as those who reported PD without significant comorbid conditions, were randomly assigned to one of two unguided interventions, either the “TeleCoach” app (n = 293) or the information control app (n = 283), and followed up via online questionnaires at six-, 12- and 26-weeks post-recruitment. Participants and researchers were blinded to group assignment at recruitment but not for outcome analyses. The primary outcome was the number of standard drinks consumed per week. Participants reported drinking a mean of 28 (SD = 18) standard drinks per week at the baseline recruitment measure, and 65% of the participants reported a history of PD for over three years; only 7% of the participants reported less than a year of PD history. Only 26% of the participants had sought help before. A significant within-group reduction in alcohol consumption occurred in both conditions at the six-week follow-up (B = −10.57, SE = 0.83, 95% CI [−12.20, −8.94], p < 0.01; Cohen’s d = 0.69), sustained until the 26-week follow-up (B = −11.44, SE = 0.91, 95% CI [−13.23, −9.65], p < 0.01, Cohen’s d = 0.67), with no significant differences between the two groups. No adverse events were reported. The findings, although null, nonetheless supported app-based interventions’ feasibility and potential as a scalable digital approach for reaching individuals who have not previously sought help for PD. ClinicalTrials.gov as NCT03696888, Oct 5, 2018; analysis plan at https://osf.io/wxfqg, on Dec. 18, 2023 prior to analysis.