Effectiveness of a Conversational Chatbot (Dejal@bot) for the Adult Population to Quit Smoking: Pragmatic, Multicenter, Controlled, Randomized Clinical Trial in Primary Care
Bibliographic record
Abstract
Background Tobacco addiction is the leading cause of preventable morbidity and mortality worldwide, but only 1 in 20 cessation attempts is supervised by a health professional. The potential advantages of mobile health (mHealth) can circumvent this problem and facilitate tobacco cessation interventions for public health systems. Given its easy scalability to large populations and great potential, chatbots are a potentially useful complement to usual treatment. Objective This study aims to assess the effectiveness of an evidence-based intervention to quit smoking via a chatbot in smartphones compared with usual clinical practice in primary care. Methods This is a pragmatic, multicenter, controlled, and randomized clinical trial involving 34 primary health care centers within the Madrid Health Service (Spain). Smokers over the age of 18 years who attended on-site consultation and accepted help to quit tobacco were recruited by their doctor or nurse and randomly allocated to receive usual care (control group [CG]) or an evidence-based chatbot intervention (intervention group [IG]). The interventions in both arms were based on the 5A’s (ie, Ask, Advise, Assess, Assist, and Arrange) in the US Clinical Practice Guideline, which combines behavioral and pharmacological treatments and is structured in several follow-up appointments. The primary outcome was continuous abstinence from smoking that was biochemically validated after 6 months by the collaborators. The outcome analysis was blinded to allocation of patients, although participants were unblinded to group assignment. An intention-to-treat analysis, using the baseline-observation-carried-forward approach for missing data, and logistic regression models with robust estimators were employed for assessing the primary outcomes. Results The trial was conducted between October 1, 2018, and March 31, 2019. The sample included 513 patients (242 in the IG and 271 in the CG), with an average age of 49.8 (SD 10.82) years and gender ratio of 59.3% (304/513) women and 40.7% (209/513) men. Of them, 232 patients (45.2%) completed the follow-up, 104/242 (42.9%) in the IG and 128/271 (47.2%) in the CG. In the intention-to-treat analysis, the biochemically validated abstinence rate at 6 months was higher in the IG (63/242, 26%) compared with that in the CG (51/271, 18.8%; odds ratio 1.52, 95% CI 1.00-2.31; P=.05). After adjusting for basal CO-oximetry and bupropion intake, no substantial changes were observed (odds ratio 1.52, 95% CI 0.99-2.33; P=.05; pseudo-R2=0.045). In the IG, 61.2% (148/242) of users accessed the chatbot, average chatbot-patient interaction time was 121 (95% CI 121.1-140.0) minutes, and average number of contacts was 45.56 (SD 36.32). Conclusions A treatment including a chatbot for helping with tobacco cessation was more effective than usual clinical practice in primary care. However, this outcome was at the limit of statistical significance, and therefore these promising results must be interpreted with caution. Trial Registration Clinicaltrials.gov NCT 03445507; https://tinyurl.com/mrnfcmtd International Registered Report Identifier (IRRID) RR2-10.1186/s12911-019-0972-z
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 | 0.008 | 0.010 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.004 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.004 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 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".