Technology-Assisted Motivational Interviewing: Developing a Scalable Framework for Promoting Engagement with Tobacco Cessation Using NLP and Machine Learning

Saiyed, Ahson; Layton, John F.; Borsari, Brian; Cheng, Jing; Kanzaveli, Tatyana; Tsvetovat, Maksim; Satterfield, Jason M.

doi:10.1016/j.procs.2022.09.091

Cited by 10 publications

(5 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A complex reflection attempts to infer something relevant to the prior conversation or guess something that might be relevant to a recent utterance, which also invites continued contemplation on the part of the patient. MI has been shown to be a successful therapeutic tool for motivating many behavior changes [18], including smoking cessation [19,20].…”

Section: Prior Related Workmentioning

confidence: 99%

“…There have been several prior attempts to automate MI using a chatbot across different domains, including stress management, sex health education, and smoking cessation [19,20,[28][29][30][31][32]; for example, Park et al [28] designed a conversational sequence using MI to aid in stress management. This conversation was deployed to 30 graduate students to compare its efficacy with that of human-to-human MI.…”

Section: Chatbotsmentioning

confidence: 99%

“…MI has been shown to be a successful therapeutic tool for motivating many behavior changes [ 18 ], including smoking cessation [ 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

“…Although these reflections were used to train practitioners and were not patient facing, they highlight the capacity of transformer-based models to produce good-quality patient-specific reflections. Similarly, Saiyed et al [ 20 ] created a technology-assisted MI chatbot for smoking cessation. The chatbot was designed to onboard participants, use MI, and refer participants to human-to-human treatment.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Motivational Interviewing Chatbot With Generative Reflections for Increasing Readiness to Quit Smoking: Iterative Development Study

Brown,

Kumar,

Melamed

et al. 2023

JMIR Ment Health

View full text Add to dashboard Cite

Background The motivational interviewing (MI) approach has been shown to help move ambivalent smokers toward the decision to quit smoking. There have been several attempts to broaden access to MI through text-based chatbots. These typically use scripted responses to client statements, but such nonspecific responses have been shown to reduce effectiveness. Recent advances in natural language processing provide a new way to create responses that are specific to a client’s statements, using a generative language model. Objective This study aimed to design, evolve, and measure the effectiveness of a chatbot system that can guide ambivalent people who smoke toward the decision to quit smoking with MI-style generative reflections. Methods Over time, 4 different MI chatbot versions were evolved, and each version was tested with a separate group of ambivalent smokers. A total of 349 smokers were recruited through a web-based recruitment platform. The first chatbot version only asked questions without reflections on the answers. The second version asked the questions and provided reflections with an initial version of the reflection generator. The third version used an improved reflection generator, and the fourth version added extended interaction on some of the questions. Participants’ readiness to quit was measured before the conversation and 1 week later using an 11-point scale that measured 3 attributes related to smoking cessation: readiness, confidence, and importance. The number of quit attempts made in the week before the conversation and the week after was surveyed; in addition, participants rated the perceived empathy of the chatbot. The main body of the conversation consists of 5 scripted questions, responses from participants, and (for 3 of the 4 versions) generated reflections. A pretrained transformer-based neural network was fine-tuned on examples of high-quality reflections to generate MI reflections. Results The increase in average confidence using the nongenerative version was 1.0 (SD 2.0; P=.001), whereas for the 3 generative versions, the increases ranged from 1.2 to 1.3 (SD 2.0-2.3; P<.001). The extended conversation with improved generative reflections was the only version associated with a significant increase in average importance (0.7, SD 2.0; P<.001) and readiness (0.4, SD 1.7; P=.01). The enhanced reflection and extended conversations exhibited significantly better perceived empathy than the nongenerative conversation (P=.02 and P=.004, respectively). The number of quit attempts did not significantly change between the week before the conversation and the week after across all 4 conversations. Conclusions The results suggest that generative reflections increase the impact of a conversation on readiness to quit smoking 1 week later, although a significant portion of the impact seen so far can be achieved by only asking questions without the reflections. These results support further evolution of the chatbot conversation and can serve as a basis for comparison against more advanced versions.

show abstract

Section: Prior Related Workmentioning

confidence: 99%

Section: Chatbotsmentioning

confidence: 99%

“…MI has been shown to be a successful therapeutic tool for motivating many behavior changes [ 18 ], including smoking cessation [ 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Motivational Interviewing Chatbot With Generative Reflections for Increasing Readiness to Quit Smoking: Iterative Development Study

Brown,

Kumar,

Melamed

et al. 2023

JMIR Ment Health

View full text Add to dashboard Cite

show abstract

“…Recently, MI has been used in multiple settings and with multiple populations including inpatient, emergency rooms, rural settings [48], and other ethnicities to reduce alcohol/substance use [49][50][51] and MI via text messages has been well received by adults [52,53]. MI has also been applied through digital format using text messaging to address a variety of health behaviors including tuberculosis [54], smoking [55], and alcohol use [56]. Most recently, chatbots, which are conversational tools that frequently employ artificial intelligence to provide a more engaging approach, have been used in apps (SoberGrid and CHESS) to support motivational interviewing to encourage action towards recovering from substance use [57][58][59], but as yet no apps exist that provide screening for SUD and alcohol and related risk factors to encourage change in risky behaviors.…”

Section: Introductionmentioning

confidence: 99%

Women Empowered to Connect with Addiction Resources and Engage in Evidence-based Treatment (WE-CARE): A Usability and Feasibility Study of a Mobile Health Application for the Universal Screening of Alcohol, Substance Use, Depression and Anxiety. (Preprint)

Isaacs,

Shifflett,

Patel

et al. 2024

Preprint

View full text Add to dashboard Cite

BACKGROUND Women of childbearing age (WOCA,18-44 years of age) face multiple barriers to receiving screening and treatment for unhealthy alcohol and substance use, depression, and anxiety, including lack of screening in the primary care setting and lack of support in accessing care. The Women Empowered to Connect with Addiction Resources and Engage in Evidence-based Treatment (WE-CARE) mobile app was developed to test universal screening with WOCA and linkage to care after an assessment for substance use disorder (SUD). OBJECTIVE In this study, we aimed to investigate the feasibility and acceptability of providing anonymous screening instruments for alcohol and substance use, as well as depression and anxiety, for women of childbearing age through mobile phones. METHODS We used Agile development principles based on prior formative research to create an alpha version of a WE-CARE mobile health application that was piloted with WOCA (N = 30) who resided in one of six counties in Central Florida. WE-CARE included screening instruments (for substance use, depression, and anxiety), a moderated discussion forum, educational micro-learning videos, an FAQ section, and resources for linkage to treatment. Individuals were recruited using flyers, academic listservs, and a commercial human subject survey company (Prolific). Upon completion of the screening instruments, women explored the educational and linkage to care features of the app and filled out a System Usability Scale (SUS) to evaluate the mobile health app’s usability and acceptability. Post-pilot semi-structured interviews (n = 4) were conducted to further explore the women’s reactions to the application. RESULTS Seventy-seven women downloaded the application and 30 completed testing. WOCA gave the WE-CARE app an excellent SUS score of 86.7 + 12.43 (SD). Out of the 30 participants (African American (N=1, 3.3%), Hispanic (N=6, 20%), more than one race (N=5,16.7%), White (N=14, 46.7%), Unknown (N=7, 23.3%)), 60% (n=18) had results indicating elevated risk for substance use, 50% of these 18 (n=9/18) indicated elevated risk also for anxiety or depression, and 61% (n=11/18) indicated elevated risk for substance use and/or anxiety or depression. Participants reported that WE-CARE was easy to navigate and use but they would like to see more screening questions and more educational content. However, linkage to care was an issue, as none of the women identified as ‘at-risk’ for SUDs contacted the free treatment clinic for further evaluation. CONCLUSIONS The mobile health application was highly rated for acceptability and usability, but participants were not receptive to seeking help at a treatment center after only a few brief encounters with the application. The linkage to care design features was likely insufficient to encourage them to seek treatment. The beta version of WE-CARE will include normative scores for participants to self-evaluate their screening status compared to their age- and gender-matched peers and enhanced linkages to care features. Future development will focus on improving engagement to change behaviors and assess readiness for change.

show abstract

Motivational Interviewing for the Prevention and Treatment of Pediatric Obesity

Woolford,

Villegas,

Resnicow

2024

Pediatric Clinics of North America

View full text Add to dashboard Cite

Technology-Assisted Motivational Interviewing: Developing a Scalable Framework for Promoting Engagement with Tobacco Cessation Using NLP and Machine Learning

Cited by 10 publications

References 27 publications

A Motivational Interviewing Chatbot With Generative Reflections for Increasing Readiness to Quit Smoking: Iterative Development Study

A Motivational Interviewing Chatbot With Generative Reflections for Increasing Readiness to Quit Smoking: Iterative Development Study

Women Empowered to Connect with Addiction Resources and Engage in Evidence-based Treatment (WE-CARE): A Usability and Feasibility Study of a Mobile Health Application for the Universal Screening of Alcohol, Substance Use, Depression and Anxiety. (Preprint)

Motivational Interviewing for the Prevention and Treatment of Pediatric Obesity

Contact Info

Product

Resources

About