An official website of the United States government
BackgroundTobacco use remains the leading cause of preventable mortality in the United States; yet, evidence-based cessation services remain underused due to staffing constraints, limited access to counseling, and competing clinical priorities. Generative artificial intelligence (GenAI) chatbots may address these barriers by delivering personalized, guideline-aligned counseling through naturalistic dialogue. However, little is known about how GenAI chatbots support smoking cessation at both outcome and communication process levels. ObjectiveThis feasibility study evaluated the implementation of an evidence-based smoking cessation counseling session delivered by a GenAI-powered chatbot, Aipaca. We examined (1) pre-post changes in cessation preparedness, (2) communication dynamics during counseling sessions, and (3) user perceptions of the chatbot’s value, limitations, and design needs. MethodsWe conducted an observational, single-arm, mixed methods study with 29 adult smokers. Participants completed pre-post surveys measuring knowledge of smoking-related health risks and cessation methods, self-efficacy, and readiness to quit. Each engaged in a 30-minute text-based counseling session with Aipaca, powered by GPT-4 and structured using the 5A’s framework (Ask, Advise, Assess, Assist, Arrange). Sessions were transcribed for microsequential conversation analysis. Twenty-five participants completed semistructured interviews exploring perceived value, challenges, and design suggestions. Quantitative data were analyzed with paired-samples t tests, qualitative data were thematically analyzed, and transcripts were analyzed for interactional practices. The methodological strength of this study lies in its triangulated approach, which combines quantitative measurement of intervention effectiveness, qualitative analysis of user interviews, and conversational analysis of counseling transcripts to generate a comprehensive understanding of both outcomes and underlying mechanisms. ResultsParticipants demonstrated significant improvements in all preparedness indicators: knowledge of health risks, knowledge of cessation methods, self-efficacy, and readiness to quit. Conversation analysis identified three recurrent patterns enabling counseling-relevant dynamics: (1) contextual referencing and continuity, (2) formulations with elaboration prompts, and (3) narrative progression toward collaborative planning. Interview themes underscored Aipaca’s perceived value as an accessible, nonjudgmental, and motivating resource, capable of delivering personalized and interactive support. Criticisms included limited accountability, reduced cultural resonance, and overly goal-directed style. Participants emphasized design needs such as proactive engagement, gamified progress tracking, empathetic or anthropomorphic personas, and safeguards for accuracy. ConclusionsThis mixed methods feasibility study demonstrates that GenAI can deliver evidence-based smoking cessation counseling with measurable short-term gains in cessation preparedness and process-level communication patterns consistent with motivational interviewing. Users valued Aipaca’s accessibility, empathy, and personalization, while also articulating expectations for richer social roles and long-term accountability. Findings highlight both the promise and challenges of integrating GenAI into digital health: pairing adaptive language generation with human-centered design, embedding accuracy safeguards, and ensuring integration into multilevel cessation infrastructures will be essential for future clinical deployment.
more »
« less
Communicating about chemicals in cigarette smoke: impact on knowledge and misunderstanding
Background The USA must publicly share information about harmful and potentially harmful constituents (chemicals) in tobacco products. We sought to understand whether webpages with chemical information are “understandable and not misleading to a lay person.” Methods Participants were a national probability sample of US adults and adolescents ( n =1441, 18% smokers). In an online experiment, we randomly assigned participants to view one of the developed webpages (chemical names only, names with quantity ranges, names with visual risk indicators) or no webpage in phase one (between subjects). Participants completed a survey assessing knowledge, misunderstanding, perceived likelihood, perceived severity of health effects from smoking and quit intentions (smokers only). In phase two (within subjects), participants viewed all three webpage formats and reported webpage perceptions (clarity, usability, usefulness) and perceived impact (affect, elaboration, perceived effectiveness). Results In phase one, viewing any webpage led to more knowledge of chemicals (48%–54% vs 28% no webpage, p s<0.001) and health harms (77% vs 67% no webpage, p s<0.001). When exposed to any webpage, 5%–23% endorsed misunderstandings that some cigarettes are safer than others. Webpage format did not affect knowledge or reduce misunderstandings. Viewing any webpage led to higher perceived likelihood of experiencing health effects from smoking ( p < 0.001) and, among smokers, greater intentions to quit smoking ( p =0.04). In phase two, where participants viewed all formats, a visual risk indicator led to the highest perceived impact. Conclusions Knowledge of chemicals and health effects can increase after viewing a website. Yet, websites may not correct the misunderstanding that some cigarettes are safer.
more »
« less
- Award ID(s):
- 2001000
- PAR ID:
- 10277535
- Date Published:
- Journal Name:
- Tobacco Control
- ISSN:
- 0964-4563
- Page Range / eLocation ID:
- tobaccocontrol-2018-054863
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Although electronic cigarette (e-cigarette) aerosol contains similar toxicants to combustible cigarettes, few studies have examined their influence on fecundability. We assessed the association between e-cigarette use and fecundability, overall and according to combustible cigarette smoking history, in a cohort of 4,586 North American women (aged 21–45 years) enrolled during 2017–2020 in Pregnancy Study Online, a Web-based prospective preconception study. Women reported current and former e-cigarette use on baseline and follow-up questionnaires, and they completed bimonthly follow-up questionnaires until self-reported pregnancy or censoring. Fecundability ratios and 95% confidence intervals were calculated using proportional probabilities models, controlling for potential confounders. Overall, 17% of women had ever used e-cigarettes and 4% were current users. Compared with never use of e-cigarettes, current e-cigarette use was associated with slightly lower fecundability (fecundability ratio = 0.84, 95% confidence interval (CI): 0.67, 1.06). Compared with current nonusers of e-cigarettes and combustible cigarettes, fecundability ratios were 0.83 (95% CI: 0.54, 1.29) for current dual users of e-cigarettes and combustible cigarettes, 0.91 (95% CI: 0.70, 1.18) for current e-cigarette users who were nonsmokers of combustible cigarettes, and 1.01 (95% CI: 0.85, 1.20) for nonusers of e-cigarettes who were current smokers of combustible cigarettes. Current e-cigarette use was associated with slightly reduced fecundability, but estimates of its independent and joint associations with combustible cigarette smoking were inconsistent and imprecise.more » « less
-
Abstract Background: Smoking has not been an established risk factor for prostate cancer (PCa), and has not been emphasized in PCa prevention. However, recent studies have shown increasing evidence that there is a higher risk of biochemical recurrence, PCa mortality, and metastasis among current smokers, presenting an urgent need in re-evaluating the association between smoking and aggressive PCa. This study aimed to determine whether smoking increase the likelihood of developing a more aggressive prostate cancer. Methods: Equal numbers of African Americans (AAs) and European Americans (EAs) by smoking status (never/former/current) matched with PCa aggressiveness, BMI, 5-year age group, and year of baseline recruitment, totaling 480 participants, were included in the metabolomics study. For metabolomics analysis, fold change and BH-adjusted p-value from t-test adjusted for age for univariate analysis, and PCA adjusted for age and PLS-DA supervised statistical analysis for multivariate analysis were employed to decipher the underlying metabolomic patterns, and identify significantly dysregulated metabolites for the variables of interest. Results: AA participants were significantly younger (mean=61.4, SD=7.7) compared with EAs (mean=63.5, SD=7.5). Current smokers had a 2.4 times higher risk of high aggressive PCa. When stratified by race, the risk diminished for EAs but increased for AAs. Global metabolic profiles detected a total of 1,487 compounds of known identity. After excluding metabolites with missing values in more than 20% of the samples and with small standard variation, we observed a distinct cluster of participants from AA aggressive PCa patients and current smokers that were separated from EAs and never smokers. With BH-adjusted p-value < 0.05 and fold change > 2, we identified 10 significantly dysregulated metabolites between AA and EA among high aggressive PCa and current smokers. Further, 36 metabolites between current and never smokers among AA high aggressive PCa were significantly dysregulated, but none of them are annotated as tobacco metabolites. Conclusion: Our study presented distinctive metabolomics profiles specific to AA current smokers who had high aggressive PCa. Furthermore, the distinctive patterns were not driven by the tobacco metabolites, with the potential to identify metabolites that might help to understand the relationships between smoking and aggressive PCa in AA. Citation Format: Se-Ran Jun, L. Joseph Su, Eryn Matich, Ping-Ching Hsu. Distinctive metabolomics profiles associated with African American current smokers who have high aggressive prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3680.more » « less
-
Smoking cigarettes during pregnancy is associated with adverse effects on infants including low birth weight, defective lung development, and skeletal abnormalities. Pregnant women are increasingly turning to vaping [use of electronic (e)-cigarettes] as a perceived safer alternative to cigarettes. However, nicotine disrupts fetal development, suggesting that like cigarette smoking, nicotine vaping may be detrimental to the fetus. To test the impact of maternal vaping on fetal lung and skeletal development in mice, pregnant dams were exposed to e-cigarette vapor throughout gestation. At embryonic day (E)18.5, vape exposed litter sizes were reduced and some embryos exhibited growth restriction compared to air exposed controls. Fetal lungs were collected for histology and whole transcriptome sequencing. Maternally nicotine vaped embryos exhibited histological and transcriptional changes consistent with impaired distal lung development. Embryonic lung gene expression changes mimicked transcriptional changes observed in adult mouse lungs exposed to cigarette smoke, suggesting that the developmental defects may be due to direct nicotine exposure. Fetal skeletons were analyzed for craniofacial and long bone lengths. Nicotine directly binds and inhibits the Kcnj2 potassium channel which is important for bone development. The length of the maxilla, palatal shelves, humerus, and femur were reduced in vaped embryos, which was further exacerbated by loss of one copy of the Kcnj2 gene. Nicotine vapor exposed Kcnj2KO/+ embryos also had significantly lower birth weights than unexposed animals of either genotype. Kcnj2 mutants had severely defective lungs with and without vape exposure, suggesting that potassium channels may be broadly involved in mediating the detrimental developmental effects of nicotine vaping. These data indicate that intrauterine nicotine exposure disrupts fetal lung and skeletal development likely through inhibition of Kcnj2.more » « less
-
BackgroundRisk-based screening for lung cancer is currently being considered in several countries; however, the optimal approach to determine eligibility remains unclear. Ensemble machine learning could support the development of highly parsimonious prediction models that maintain the performance of more complex models while maximising simplicity and generalisability, supporting the widespread adoption of personalised screening. In this work, we aimed to develop and validate ensemble machine learning models to determine eligibility for risk-based lung cancer screening. Methods and findingsFor model development, we used data from 216,714 ever-smokers recruited between 2006 and 2010 to the UK Biobank prospective cohort and 26,616 high-risk ever-smokers recruited between 2002 and 2004 to the control arm of the US National Lung Screening (NLST) randomised controlled trial. The NLST trial randomised high-risk smokers from 33 US centres with at least a 30 pack-year smoking history and fewer than 15 quit-years to annual CT or chest radiography screening for lung cancer. We externally validated our models among 49,593 participants in the chest radiography arm and all 80,659 ever-smoking participants in the US Prostate, Lung, Colorectal and Ovarian (PLCO) Screening Trial. The PLCO trial, recruiting from 1993 to 2001, analysed the impact of chest radiography or no chest radiography for lung cancer screening. We primarily validated in the PLCO chest radiography arm such that we could benchmark against comparator models developed within the PLCO control arm. Models were developed to predict the risk of 2 outcomes within 5 years from baseline: diagnosis of lung cancer and death from lung cancer. We assessed model discrimination (area under the receiver operating curve, AUC), calibration (calibration curves and expected/observed ratio), overall performance (Brier scores), and net benefit with decision curve analysis.Models predicting lung cancer death (UCL-D) and incidence (UCL-I) using 3 variables—age, smoking duration, and pack-years—achieved or exceeded parity in discrimination, overall performance, and net benefit with comparators currently in use, despite requiring only one-quarter of the predictors. In external validation in the PLCO trial, UCL-D had an AUC of 0.803 (95% CI: 0.783, 0.824) and was well calibrated with an expected/observed (E/O) ratio of 1.05 (95% CI: 0.95, 1.19). UCL-I had an AUC of 0.787 (95% CI: 0.771, 0.802), an E/O ratio of 1.0 (95% CI: 0.92, 1.07). The sensitivity of UCL-D was 85.5% and UCL-I was 83.9%, at 5-year risk thresholds of 0.68% and 1.17%, respectively, 7.9% and 6.2% higher than the USPSTF-2021 criteria at the same specificity. The main limitation of this study is that the models have not been validated outside of UK and US cohorts. ConclusionsWe present parsimonious ensemble machine learning models to predict the risk of lung cancer in ever-smokers, demonstrating a novel approach that could simplify the implementation of risk-based lung cancer screening in multiple settings.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1136/tobaccocontrol-2018-054863
