Related Experiment Video
Updated: May 28, 2026
A Prediction Error-driven Retrieval Procedure for Destabilizing and Rewriting Maladaptive Reward Memories in Hazardous Drinkers
Published on: January 5, 2018
Effects, Acceptability, and Use of a Dynamically Tailored Mobile What Do You Drink Intervention to Reduce Excessive
Hilde van Keulen1, Carmen Voogt2, Marloes Kleinjan3
1Department of Child Health, Netherlands Organisation for Applied Scientific Research, Sylviusweg 71, Leiden, 2333 BE, The Netherlands, 31 652803631.
Background:
Excessive alcohol consumption among adolescents and young adults is a serious health problem. Dynamically tailored interventions could reduce their excessive drinking. We therefore developed "What Do You Drink" (WDYD), a 17-week dynamically tailored mHealth (mobile health) intervention providing personalized support on alcohol consumption.
Objective:
We aim to evaluate the effectiveness, acceptability, and use of WDYD in reducing alcohol consumption of adolescents and young adults at risk.
Methods:
We conducted a 2-arm, parallel-group randomized controlled trial using ecological momentary assessments. Recruitment was via an educational alcohol program, an online lifestyle monitor, social media advertisements, or news items on websites. Participants downloaded the standalone WDYD app, and when having given active informed consent, were randomized to the intervention or control group. Participants in the intervention group received dynamically tailored feedback sessions on alcohol consumption (wk 0-5, 7, 9, 13, and 17) and goal-monitoring reminders. Both groups completed an online baseline survey, 2 follow-up surveys (wk 9 and 33), and various ecological momentary assessments (7 daily assessments during wk 1, 7, 13, 19, 25, 31, and 33). Participants provided consent before randomization, in which they were informed that 2 study groups existed. After randomization, no disclosure of group assignment was provided, although participants could potentially infer it from receiving tailored sessions vs no tailored sessions. Primary outcomes were excessive drinking, binge drinking, and weekly alcohol consumption. Secondary outcomes were intrinsic motivation, self-confidence, and mood. Acceptability of WDYD was measured by survey questions; use was tracked via app data logs.
Results:
Analyses were based on data from 1767 participants; 720 in the intervention group and 1047 participants in the control group. Almost half of them were female (2276/4795, 47.5%), and most (3471/4595, 72.4%) participants were aged 18-24 (median 19.40, IQR 2.92) years. The dropout rate was high, up to 96% (4603/4795) in the final 33rd week. No significant effect of WDYD was found on primary outcomes and mood, except for week 1 (excessive drinking: standardized β=-0.35, SE 0.15; 95% CI -0.64 to -0.05; binge drinking: standardized β=-0.36, SE 0.16; 95% CI -0.68 to -0.04; mood: standardized β=0.20, SE 0.06, 95% CI 0.08 to 0.32). Both groups reduced their alcohol consumption. Significant positive effects were found for intrinsic motivation and self-confidence up to 25 weeks (wk 25: standardized β=0.54, SE 0.24; 95% CI 0.06 to 1.02 for motivation; standardized β=0.72, SE 0.26; 95% CI 0.22 to 1.23 for self-confidence). Participants evaluated WDYD as acceptable and usable.
Conclusions:
WDYD did not significantly reduce excessive drinking compared to control, but improved motivation and self-confidence. High dropout rates highlight challenges in sustaining engagement in long-term mHealth interventions. Future research should explore strategies to enhance retention and optimize dynamic tailoring.
