User Perspectives on Conversational Agents in Preventive Alcohol Self-Help: A Qualitative Exploration (Preprint)
Bibliographic record
Abstract
Abstract Background Alcohol use remains a major public health concern, and although preventive alcohol self-help interventions aim to support individuals in need of indicated prevention, they continue to face challenges related to low engagement and high attrition. Chatbots, also known as conversational agents (CAs), powered by artificial intelligence, may enhance engagement by offering personalized guidance and 24/7 availability. Yet, user perspectives on CAs in preventive alcohol self-help care remain largely unexplored. Objective This paper aimed to gain insight into the diverse factors shaping the adoption of CAs in preventive alcohol self-help by qualitatively exploring users’ expectations and concerns, complemented by descriptive questionnaire data on perceived usefulness, intention to use, comfort, and trust. Methods This qualitative, exploratory study with a small purposive sample (N=12) included adults recruited from a Dutch digital alcohol self-help intervention and a client advisory board. Data were collected during an online focus group (n=8, 66.7%), 4 semistructured interviews (n=4, 33.3%), and a brief questionnaire administered to all participants in March 2025 and April 2025. The qualitative data were analyzed inductively using thematic analysis. Results Qualitative findings showed that participants appreciated the 24/7 availability, practical guidance, and nonjudgmental stance of CAs but doubted their capacity for genuine empathy and authenticity. Participants emphasized that the conversational tone was important: CAs should be supportive without being overly agreeable and should be confrontational once trust is established. Concerns about undue influence arose occasionally, while privacy was viewed by 1 (8.3%) participant as a major issue. Descriptive questionnaire results showed that 11 (92%) participants had prior experience with CAs, whereas only 6 (50%) had experience with digital health interventions. Most participants considered CAs potentially useful for reducing alcohol use, yet few expressed strong intentions to use one themselves. Comfort and trust ratings were largely neutral, reflecting a cautious but generally open attitude toward CA use. Conclusions CAs appear to hold promise as complementary tools for preventive alcohol self-help, although they are not perceived as replacements for human contact. Future design efforts should emphasize personalized interactions; an appropriate conversational tone, supportive yet not overly agreeable; and transparent communication about data use to enhance engagement and acceptance. The findings are descriptive and are not intended to be broadly generalizable.
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 | 0.015 | 0.020 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.005 | 0.006 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 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".