Secure Asynchronous Communication Between Smokers and Tobacco Treatment Specialists: Secondary Analysis of a Web-Assisted Tobacco Intervention in the QUIT-PRIMO and National Dental PBRN Networks.

BACKGROUND: Within a web-assisted tobacco intervention, we provided a function for smokers to asynchronously communicate with a trained tobacco treatment specialist (TTS). Previous studies have not attempted to isolate the effect of asynchronous counseling on smoking cessation. OBJECTIVE: This study aimed to conduct a semiquantitative analysis of TTS-smoker communication and evaluate its association with smoking cessation. METHODS: We conducted a secondary analysis of data on secure asynchronous communication between trained TTSs and a cohort of smokers during a 6-month period. Smokers were able to select their preferred TTS and message them using a secure web-based form. To evaluate whether the TTS used evidence-based practices, we coded messages using the Motivational Interviewing Self-Evaluation Checklist and Smoking Cessation Counseling (SCC) Scale. We assessed the content of messages initiated by the smokers by creating topical content codes. At 6 months, we assessed the association between smoking cessation and the amount of TTS use and created a multivariable model adjusting for demographic characteristics and smoking characteristics at baseline. RESULTS: Of the 725 smokers offered asynchronous counseling support, 33.8% (245/725) messaged the TTS at least once. A total of 1082 messages (TTSs: 565; smokers 517) were exchanged between the smokers and TTSs. The majority of motivational interviewing codes were those that supported client strengths (280/517, 54.1%) and promoted engagement (280/517, 54.1%). SCC code analysis showed that the TTS provided assistance to smokers if they were willing to quit (247/517, 47.8%) and helped smokers prepare to quit (206/517, 39.8%) and anticipate barriers (197/517, 38.1%). The majority of smokers' messages discussed motivations to quit (234/565, 41.4%) and current and past treatments (talking about their previous use of nicotine replacement therapy and medications; 201/565, 35.6%). The majority of TTS messages used behavioral strategies (233/517, 45.1%), offered advice on treatments (189/517, 36.5%), and highlighted motivations to quit (171/517, 33.1%). There was no association between the amount of TTS use and cessation. In the multivariable model, after adjusting for gender, age, race, education, readiness at baseline, number of cigarettes smoked per day at baseline, and the selected TTS, smokers messaging the TTS one or two times had a smoking cessation odds ratio (OR) of 0.8 (95% CI 0.4-1.4), and those that messaged the TTS more than two times had a smoking cessation OR of 1.0 (95% CI 0.4-2.3). CONCLUSIONS: Our study demonstrated the feasibility of using asynchronous counseling to deliver evidence-based counseling. Low participant engagement or a lack of power could be potential explanations for the nonassociation with smoking cessation. Future trials should explore approaches to increase participant engagement and test asynchronous counseling in combination with other approaches for improving the rates of smoking cessation.

Impact of a Collective Intelligence Tailored Messaging System on Smoking Cessation: The Perspect Randomized Experiment.

BACKGROUND: Outside health care, content tailoring is driven algorithmically using machine learning compared to the rule-based approach used in current implementations of computer-tailored health communication (CTHC) systems. A special class of machine learning systems ("recommender systems") are used to select messages by combining the collective intelligence of their users (ie, the observed and inferred preferences of users as they interact with the system) and their user profiles. However, this approach has not been adequately tested for CTHC. OBJECTIVE: Our aim was to compare, in a randomized experiment, a standard, evidence-based, rule-based CTHC (standard CTHC) to a novel machine learning CTHC: Patient Experience Recommender System for Persuasive Communication Tailoring (PERSPeCT). We hypothesized that PERSPeCT will select messages of higher influence than our standard CTHC system. This standard CTHC was proven effective in motivating smoking cessation in a prior randomized trial of 900 smokers (OR 1.70, 95% CI 1.03-2.81). METHODS: PERSPeCT is an innovative hybrid machine learning recommender system that selects and sends motivational messages using algorithms that learn from message ratings from 846 previous participants (explicit feedback), and the prior explicit ratings of each individual participant. Current smokers (N=120) aged 18 years or older, English speaking, with Internet access were eligible to participate. These smokers were randomized to receive either PERSPeCT (intervention, n=74) or standard CTHC tailored messages (n=46). The study was conducted between October 2014 and January 2015. By randomization, we compared daily message ratings (mean of smoker ratings each day). At 30 days, we assessed the intervention's perceived influence, 30-day cessation, and changes in readiness to quit from baseline. RESULTS: The proportion of days when smokers agreed/strongly agreed (daily rating ≥4) that the messages influenced them to quit was significantly higher for PERSPeCT (73%, 23/30) than standard CTHC (44%, 14/30, P=.02). Among less educated smokers (n=49), this difference was even more pronounced for days strongly agree (intervention: 77%, 23/30; comparison: 23%, 7/30, P<.001). There was no significant difference in the frequency which PERSPeCT randomized smokers agreed or strongly agreed that the intervention influenced them to quit smoking (P=.07) and use nicotine replacement therapy (P=.09). Among those who completed follow-up, 36% (20/55) of PERSPeCT smokers and 32% (11/34) of the standard CTHC group stopped smoking for one day or longer (P=.70). CONCLUSIONS: Compared to standard CTHC with proven effectiveness, PERSPeCT outperformed in terms of influence ratings and resulted in similar cessation rates. CLINICALTRIAL: Clinicaltrials.gov NCT02200432; https://clinicaltrials.gov/ct2/show/NCT02200432 (Archived by WebCite at http://www.webcitation.org/6lEJY1KEd).

Collective-Intelligence Recommender Systems: Advancing Computer Tailoring for Health Behavior Change Into the 21st Century.

BACKGROUND: What is the next frontier for computer-tailored health communication (CTHC) research? In current CTHC systems, study designers who have expertise in behavioral theory and mapping theory into CTHC systems select the variables and develop the rules that specify how the content should be tailored, based on their knowledge of the targeted population, the literature, and health behavior theories. In collective-intelligence recommender systems (hereafter recommender systems) used by Web 2.0 companies (eg, Netflix and Amazon), machine learning algorithms combine user profiles and continuous feedback ratings of content (from themselves and other users) to empirically tailor content. Augmenting current theory-based CTHC with empirical recommender systems could be evaluated as the next frontier for CTHC. OBJECTIVE: The objective of our study was to uncover barriers and challenges to using recommender systems in health promotion. METHODS: We conducted a focused literature review, interviewed subject experts (n=8), and synthesized the results. RESULTS: We describe (1) limitations of current CTHC systems, (2) advantages of incorporating recommender systems to move CTHC forward, and (3) challenges to incorporating recommender systems into CTHC. Based on the evidence presented, we propose a future research agenda for CTHC systems. CONCLUSIONS: We promote discussion of ways to move CTHC into the 21st century by incorporation of recommender systems.

An observational study of social and emotional support in smoking cessation Twitter accounts: content analysis of tweets.

BACKGROUND: Smoking continues to be the number one preventable cause of premature death in the United States. While evidence for the effectiveness of smoking cessation interventions has increased rapidly, questions remain on how to effectively disseminate these findings. Twitter, the second largest online social network, provides a natural way of disseminating information. Health communicators can use Twitter to inform smokers, provide social support, and attract them to other interventions. A key challenge for health researchers is how to frame their communications to maximize the engagement of smokers. OBJECTIVE: Our aim was to examine current Twitter activity for smoking cessation. METHODS: Active smoking cessation related Twitter accounts (N=18) were identified. Their 50 most recent tweets were content coded using a schema adapted from the Roter Interaction Analysis System (RIAS), a theory-based, validated coding method. Using negative binomial regression, the association of number of followers and frequency of individual tweet content at baseline was assessed. The difference in followership at 6 months (compared to baseline) to the frequency of tweet content was compared using linear regression. Both analyses were adjusted by account type (organizational or not organizational). RESULTS: The 18 accounts had 60,609 followers at baseline and 68,167 at 6 months. A total of 24% of tweets were socioemotional support (mean 11.8, SD 9.8), 14% (mean 7, SD 8.4) were encouraging/engagement, and 62% (mean 31.2, SD 15.2) were informational. At baseline, higher frequency of socioemotional support and encouraging/engaging tweets was significantly associated with higher number of followers (socioemotional: incident rate ratio [IRR] 1.09, 95% CI 1.02-1.20; encouraging/engaging: IRR 1.06, 95% CI 1.00-1.12). Conversely, higher frequency of informational tweets was significantly associated with lower number of followers (IRR 0.95, 95% CI 0.92-0.98). At 6 months, for every increase by 1 in socioemotional tweets, the change in followership significantly increased by 43.94 (P=.027); the association was slightly attenuated after adjusting by account type and was not significant (P=.064). CONCLUSIONS: Smoking cessation activity does exist on Twitter. Preliminary findings suggest that certain content strategies can be used to encourage followership, and this needs to be further investigated.

Who participates in Web-assisted tobacco interventions? The QUIT-PRIMO and National Dental Practice-Based Research Network Hi-Quit studies.

INTRODUCTION: Smoking is the most preventable cause of death. Although effective, Web-assisted tobacco interventions are underutilized and recruitment is challenging. Understanding who participates in Web-assisted tobacco interventions may help in improving recruitment. OBJECTIVES: To understand characteristics of smokers participating in a Web-assisted tobacco intervention (Decide2Quit.org). METHODS: In addition to the typical Google advertisements, we expanded Decide2Quit.org recruitment to include referrals from medical and dental providers. We assessed how the expanded recruitment of smokers changed the users' characteristics, including comparison with a population-based sample of smokers from the national Behavioral Risk Factors Surveillance Survey (BRFSS). Using a negative binomial regression, we compared demographic and smoking characteristics by recruitment source, in particular readiness to quit and association with subsequent Decide2Quit.org use. RESULTS: The Decide2Quit.org cohort included 605 smokers; the 2010 BRFSS dataset included 69,992. Compared to BRFSS smokers, a higher proportion of Decide2Quit.org smokers were female (65.2% vs 45.7%, P=.001), over age 35 (80.8% vs 67.0%, P=.001), and had some college or were college graduates (65.7% vs 45.9%, P=.001). Demographic and smoking characteristics varied by recruitment; for example, a lower proportion of medical- (22.1%) and dental-referred (18.9%) smokers had set a quit date or had already quit than Google smokers (40.1%, P<.001). Medical- and dental-referred smokers were less likely to use Decide2Quit.org functions; in adjusted analysis, Google smokers (predicted count 17.04, 95% CI 14.97-19.11) had higher predicted counts of Web page visits than medical-referred (predicted count 12.73, 95% CI 11.42-14.04) and dental-referred (predicted count 11.97, 95% CI 10.13-13.82) smokers, and were more likely to contact tobacco treatment specialists. CONCLUSIONS: Recruitment from clinical practices complimented Google recruitment attracting smokers less motivated to quit and less experienced with Web-assisted tobacco interventions.