Contrasting a Mobile App With a Conversational Chatbot for Reducing Alcohol Consumption: Randomized Controlled Pilot Trial
Bibliographic record
Abstract
Background Mobile apps have shown considerable promise for reducing alcohol consumption among problem drinkers, but like many mobile health apps, they frequently report low utilization, which is an important limitation, as research suggests that effectiveness is related to higher utilization. Interactive chatbots have the ability to provide a conversational interface with users and may be more engaging and result in higher utilization and effectiveness, but there is limited research into this possibility. Objective This study aimed to develop a chatbot alcohol intervention based on an empirically supported app (Step Away) for reducing drinking and to conduct a pilot trial of the 2 interventions. Included participants met the criteria for hazardous drinking and were interested in reducing alcohol consumption. The study assessed utilization patterns and alcohol outcomes across the 2 technology conditions, and a waitlist control group. Methods Participants were recruited using Facebook advertisements. Those who met the criteria for hazardous consumption and expressed an interest in changing their drinking habits were randomly assigned to three conditions: the Step Away app, Step Away chatbot, and waitlist control condition. Participants were assessed on the web using the Alcohol Use Disorders Identification Test, Adapted for Use in the United States, Readiness to Change Questionnaire, Short Inventory of Problems-Revised, and Timeline Followback at baseline and at 12 weeks follow-up. Results A total of 150 participants who completed the baseline and follow-up assessments were included in the final analysis. ANOVA results indicated that participants in the 3 conditions changed their drinking from baseline to follow-up, with large effect sizes noted (ie, η2=0.34 for change in drinks per day across conditions). However, the differences between groups were not significant across the alcohol outcome variables. The only significant difference between conditions was in the readiness to change variable, with the bot group showing the greatest improvement in readiness (F2,147=5.6; P=.004; η2=0.07). The results suggested that the app group used the app for a longer duration (mean 50.71, SD 49.02 days) than the bot group (mean 27.16, SD 30.54 days; P=.02). Use of the interventions was shown to predict reduced drinking in a multiple regression analysis (β=.25, 95% CI 0.00-0.01; P=.04). Conclusions Results indicated that all groups in this study reduced their drinking considerably from baseline to the 12-week follow-up, but no differences were found in the alcohol outcome variables between the groups, possibly because of a combination of small sample size and methodological issues. The app group reported greater use and slightly higher usability scores than the bot group, but the bot group demonstrated improved readiness to change scores over the app group. The strengths and limitations of the app and bot interventions as well as directions for future research are discussed. Trial Registration ClinicalTrials.gov NCT04447794; https://clinicaltrials.gov/ct2/show/NCT04447794
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.008 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".