A Personalized Approach Bias Modification Smartphone App (“SWiPE”) to Reduce Alcohol Use: Open-Label Feasibility, Acceptability, and Preliminary Effectiveness Study
Bibliographic record
Abstract
Background Approach bias modification (ApBM), a computerized cognitive intervention that trains people to “avoid” alcohol-related images and “approach” nonalcohol images, reduces the likelihood of relapse when administered during residential alcohol treatment. However, most individuals experiencing alcohol problems do not require, do not seek, or have difficulty accessing residential treatment. Smartphone-delivered ApBM could offer an easily accessible intervention to reduce alcohol consumption that can be personalized (eg, allowing selection of personally relevant alcohol and positive nonalcohol training images) and gamified to optimize engagement. Objective We examined the feasibility, acceptability, and preliminary effectiveness of “SWiPE,” a gamified, personalized alcohol ApBM smartphone app, and explored alcohol consumption and craving outcomes in people drinking at hazardous levels or above (Alcohol Use Disorders Identification Test [AUDIT] score ≥8) who wanted to reduce their alcohol use. Methods In this open-label trial, frequency and quantity of alcohol consumption, alcohol dependence severity, and craving were measured prior to participants downloading SWiPE. Participants (n=1309) were instructed to complete at least 2 sessions per week for 4 weeks. Recruitment and completion rates were indicators of feasibility. Functionality, aesthetics, and quality ratings were indicators of acceptability. Participants were prompted to report frequency and quantity of alcohol consumption weekly during training and 1 month after training. They completed measures of craving and dependence after 4 weeks of training. Results We recruited 1309 participants (mean age 47.0, SD 10.0 years; 758/1309, 57.9% female; mean AUDIT score 21.8, SD 6.5) over 6 months. Participants completed a median of 5 sessions (IQR 2-9); 31.2% (409/1309) completed ≥8 sessions; and 34.8% (455/1309) completed the posttraining survey. Mean Mobile Application Rating Scale scores indicated good acceptability for functionality and aesthetics and fair acceptability for subjective quality. Among those who completed the posttraining assessment, mean past-week drinking days reduced from 5.1 (SD 2.0) pre-training to 4.2 (SD 2.3) in week 4 (t454=7.87; P<.001), and mean past-week standard drinks reduced from 32.8 (SD 22.1) to 24.7 (SD 20.1; t454=8.58; P<.001). Mean Craving Experience Questionnaire frequency scores reduced from 4.5 (SD 2.0) to 2.8 (SD 1.8; t435=19.39; P<.001). Severity of Dependence scores reduced from 7.7 (SD 3.0) to 6.0 (SD 3.2; t435=12.44; P<.001). For the 19.4% (254/1309) of participants who completed a 1-month follow-up, mean past-week drinking days and standard drinks were 3.9 (SD 2.5) and 23.9 (SD 20.7), respectively, both significantly lower than at baseline (P<.001). Conclusions The findings suggest SWiPE is feasible and acceptable and may be effective at reducing alcohol consumption and craving in a predominantly nontreatment-seeking sample of adult Australians drinking at hazardous levels. SWiPE’s efficacy, relative to a control condition, now needs establishing in a randomized controlled trial. Smartphone-delivered personalized ApBM could be a highly scalable, widely accessible support tool for reducing alcohol use. Trial Registration Australian New Zealand Clinical Trials Registry ACTRN12620000638932; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?ACTRN=12620000638932p International Registered Report Identifier (IRRID) RR2-10.2196/21278
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".