The Use of Web-Based Support Groups Versus Usual Quit-Smoking Care for Men and Women Aged 21-59 Years: Protocol for a Randomized Controlled Trial

Background Existing smoking cessation treatments are challenged by low engagement and high relapse rates, suggesting the need for more innovative, accessible, and interactive treatment strategies. Twitter is a Web-based platform that allows people to communicate with each other throughout the day using their phone. Objective This study aims to leverage the social media platform of Twitter for fostering peer-to-peer support to decrease relapse with quitting smoking. Furthermore, the study will compare the effects of coed versus women-only groups on women’s success with quitting smoking. Methods The study design is a Web-based, three-arm randomized controlled trial with two treatment arms (a coed or women-only Twitter support group) and a control arm. Participants are recruited online and are randomized to one of the conditions. All participants will receive 8 weeks of combination nicotine replacement therapy (patches plus their choice of gum or lozenges), serial emails with links to Smokefree.gov quit guides, and instructions to record their quit date online (and to quit smoking on that date) on a date falling within a week of initiation of the study. Participants randomized to a treatment arm are placed in a fully automated Twitter support group (coed or women-only), paired with a buddy (matched on age, gender, location, and education), and encouraged to communicate with the group and buddy via daily tweeted discussion topics and daily automated feedback texts (a positive tweet if they tweet and an encouraging tweet if they miss tweeting). Recruited online from across the continental United States, the sample consists of 215 male and 745 female current cigarette smokers wanting to quit, aged between 21 and 59 years. Self-assessed follow-up surveys are completed online at 1, 3, and 6 months after the date they selected to quit smoking, with salivary cotinine validation at 3 and 6 months. The primary outcome is sustained biochemically confirmed abstinence at the 6-month follow-up. Results From November 2016 to September 2018, 960 participants in 36 groups were recruited for the randomized controlled trial, in addition to 20 participants in an initial pilot group. Data analysis will commence soon for the randomized controlled trial based on data from 896 of the 960 participants (93.3%), with 56 participants lost to follow-up and 8 dropouts. Conclusions This study combines the mobile platform of Twitter with a support group for quitting smoking. Findings will inform the efficacy of virtual peer-to-peer support groups for quitting smoking and potentially elucidate gender differences in quit rates found in prior research. Trial Registration ClinicalTrials.gov NCT02823028; https://clinicaltrials.gov/ct2/show/NCT02823028 International Registered Report Identifier (IRRID) DERR1-10.2196/16417

1 R01 CA204356-01 2 IPTA PECHMANN, C; PROCHASKA, J 1R01CA204356-01 Pechmann, Cornelia ADMINISTRATIVE NOTE RESUME AND SUMMARY OF DISCUSSION: This application proposes to revise, pilot test, and evaluate in a randomized trial the -Tweet2Quit‖ social media intervention for smoking cessation. This intervention is one of the first real-time, fully automated, twitter-based interventions. Both the environment and investigators are excellent. Other strengths included: excellent pilot data from a previous R34 in support of the application and was useful for designing the current trial, careful basis in current literature, focus on gender differences in response to smoking cessation interventions, and sophisticated analysis of online social dynamics. In summary, the committee concluded that the potential overall impact was high.

DESCRIPTION (provided by applicant):
Social Media Technology for Treating Tobacco Addiction. The internet was built for communication. This research project, with its cutting-edge social network approaches and analyses, will examine the potential of the internet, and more specifically social media, to engage health communications for quitting smoking and preventing relapse for addiction treatment and cancer prevention. Further, the design tests the personalized benefits and treatment processes of a women-centered internet cessation intervention. With initial funding from an NIH R34 innovation award, we developed and evaluated a novel online social media intervention for tobacco cessation called Tweet2Quit. Tweet2Quit is a 100-day online intervention that uses a novel, low cost, automated approach to encourage and direct peer-to-peer social support for quitting smoking. Private Twitter groups of 24 smokers are set up, and engagement is encouraged and directed by daily auto feedback and daily auto messages that suggest discussion topics consistent with US treatment guidelines for tobacco use cessation. In a randomized controlled trial (N=160), all participants received advice to quit, referral to the NCI Smokefree.gov website, and 8 weeks of nicotine patch (-usual care‖); and half were then randomized to a Tweet2Quit peer support group to test the net benefit of the social media intervention. At 60-days follow-up, Tweet2Quit participants reported 40% sustained abstinence compared to 20% for controls (p=.017), and engagement in the intervention was related to abstinence (p<.001). Consistent with the tobacco treatment literature, in both study conditions, men were more likely to quit smoking than women, and the gender effect in Tweet2Quit was moderate (d=.55). Notably, women in Tweet2Quit tweeted as much as the men, though the content differed thematically. The proposed randomized controlled evaluation (N=960) of the Tweet2Quit intervention will biochemically verify abstinence out to 6-months follow-up and will test the personalized benefit for women of a women-only versus co-ed Tweet2Quit group. In a 3-arm design, we will compare: 1) usual care, 2) Tweet2Quit-coed, and 3) Tweet2Quit-women only. Each 24-person Tweet2Quit group will include 12 buddy pairs based on similarity in demographics and abstinence goals. Also automated pattern detection will identify dips in tweeting and trigger automated alerts and engagement strategies. Our primary aims test hypothesis 1: Relative to usual care (n=240), Tweet2Quit-coed groups (n=480) will achieve significantly greater bioconfirmed sustained abstinence out to 6-months follow-up, and hypothesis 2: Women in Tweet2Quit will achieve significantly greater bioconfirmed 6-months abstinence in woman-only groups (n=240) vs. coed groups (n=240 women). Our secondary aims are to test the same hypotheses based on 3-month (end of treatment) sustained abstinence and 7-day point prevalence at 1, 3, and 6 months. Exploratory aims will study the Tweet2Quit groups' social network structures with a focus on the identification of buddy pairs and social brokers and test if these relationships are stronger for women in women-only groups versus women in coed groups of Tweet2Quit.
PUBLIC HEALTH RELEVANCE: Social Media Technology for Treating Tobacco Addiction Tweet2Quit is a low cost, web-based, social media-enabled, peer-to-peer intervention for smoking cessation with fully automated setup and automated daily texts to encourage peer discussions. In a randomized 1 R01 CA204356-01 3 IPTA PECHMANN, C; PROCHASKA, J controlled trial, we aim to determine intervention efficacy out to 6 months follow-up with bioconfirmation of abstinence and to assess the benefit among women of a women-only versus coed social media peer-to-peer group. Findings will inform observed gender differences in cessation and the use of social media for low cost, scalable addiction interventions. 1

CRITIQUE 1
Significance: 1 Investigator(s): 1 Innovation: 1 Approach: 3 Environment: 1 Overall Impact: Following very promising data collected as part of an R34, the proposed RO1 randomized controlled trial will test the Tweet2Quit intervention efficacy, including biochemically verify abstinence out to 6-months, in a 3-arm design: usual care, Tweet2Quit-coed, and Tweet2Quit-women only. Impact is very high given low cost and sustainability of this intervention, with auto messaging as opposed to peer moderators. Innovation is high via delivery on Twitter, including matching of peer dyads via computer algorithm and examination of social brokers (who will be assigned to each group). The examination of a women only condition, as opposed to a mixed gender condition, increases significance given the reduced quit rates among women. The investigative team is well suited to achieve study aims. The approach is strong, using a comparative efficacy trial in which the active control condition will receive online smoking cessation website with patch/gum and biological confirmation of self-report. The weaknesses are minor, with the exception that outcomes are relatively short term (6 months). Nonetheless, this application is very strong, with high impact.

Strengths
Testing use of social media, specifically Twitter, to facilitate social support for smoking cessation has high significance by ease of delivery, low cost and high reach.
Convincing argument that could be scalable and replicated for other health behaviors (e.g., alcohol, other drugs), which increases significance of the approach.
Although unlikely that Twitter would disappear in the near future, the intervention could be delivered on other social media platform such as Facebook.
Instead of relying on peer leaders, the 100 day program uses auto messages to increase interaction which has high sustainability if efficacious.

Weaknesses
Daily use of Facebook (eligibility criteria) limits future impact; however 70% use Facebook daily.

Investigator(s):
Strengths MPI: Dr. Pechmann is a professor of Marketing at Vanderbilt with training in psychology and has conducted research on tobacco control issues for two decades. Although she has never been the PI on a R01, she was the PI of the R34 that developed and piloted the Tweet2Quit intervention which increases confidence in her ability to lead this study. 1 R01 CA204356-01 4 IPTA PECHMANN, C; PROCHASKA, J MPI: Dr. Prochaska at UCSF has experience with intervention studies and will lead the team in social media recruitment; also, as a clinical psychologist, she will monitor group interactions.
MPI plan has clear roles, with Dr. Pechmann leading the technology pieces and Dr. Prochaska leading the intervention, recruitment and retention pieces.
Roles are mostly distinct with appropriate expertise: Dr. Lakon (Co-I) will lead social network analysis, Dr. Ramo, (consultant) will assist with the social medial recruitment; Dr. Delucchi (consultant) will prepare the randomization scheme and analytic strategy; and Dr. Sanders-Jackson (co-I) will conduct social network analytics.

Strengths
Using Twitter (social media) to deliver low-cost and high reach interventions by private social support groups with technology matched peers (using a prior tested computer algorithm that increases dyad ties) is highly innovative.
Inclusion of mixed gender group and a women only group is highly innovative and important given lower quit rates among women.
Prescreening will result in identification of -social brokers‖ (personality trait of selfmonitoring; greater centrality in social network analyses) who in the pilot posted more tweets and had great abstinence; these participants will be balanced by groups (24 per group) to increase likelihood of interaction and success across groups. This approach is highly innovative.
Exploratory social network analyses will compare peers, social brokers, and examine interactions in the Twitter group is interesting and novel.
Use of pilot data regarding types of Tweets in relation to engagement and outcome will improve the proposed revised intervention.

Strengths
Self-report of quitting augmented with biochemical data collection via the mail with Skype verification.
Careful plan for participant identify verification via IP addresses as well as video/skype with participants at enrollment and during saliva tests.
Promising pilot data currently in review (e.g., doubled sustained reported abstinence relative to usual care at 2-month follow-up, with high engagement which was related to abstinence); however, the quit rates were lower in women than men. Enhancement of prior intervention by linkage with a peer buddy and auto messages to increase engagement; although this approach has face validity and appears to be a strength, these enhancements are untested.
Detailed theory driven hypotheses for models testing mediation via social network analyses.

Weaknesses
Short term follow-up (1, 3, 6 months) is moderate limitation in that data will be lacking on sustainability of quitting.
90% 2 month follow-up (with abbreviated assessments for hard to reach) is good, with planned monetary incentives at 6 months; however, demonstration of ability to retain participants with the full assessment at 6 months is unclear.
Prior study resulted in few minority participants; thus, the team now includes a consultant with expertise in Facebook recruitment including increasing minority participation.
Because there will not be a therapist moderator, daily marijuana use is described as an exclusion criterion (only 21 smokers excluded); this rationale is not convincing.
The lack of measurement of alcohol, marijuana, and other substances seems like a missed opportunity.

Strengths
Facilities at University of California Irvine are sufficient to conduct the proposed research in terms of space and computing etc., with access to research staff through the undergraduate research opportunity program.
Letter of support with detailed intervention costs from the computer development firm Web Advanced, increases confidence in ability to develop the proposed intervention.
Letter of support from NCI Smokefree.gov, with ability to share data on visits to the site, which is a strength.

Protections for Human Subjects:
Acceptable Risks and/or Adequate Protections New names/Twitter accounts will be created for confidentiality.

Data and Safety Monitoring Plan (Applicable for Clinical Trials Only):
Acceptable Overall acceptable; however, the timeframe for reporting SAEs is needed. Minorities will be oversampled (than represented on Facebook) to 35%, which is strength.

Vertebrate Animals:
Not Applicable (No Vertebrate Animals)

Resource Sharing Plans:
Acceptable Budget and Period of Support:

CRITIQUE 2
Significance: 2 Investigator(s): 1 Innovation: 2 Approach: 3 Environment: 1 Overall Impact: This is very well-written grant application proposing to further test a peer-to-peer, twitter group intervention based on highly informative data from a previous R34 grant. The investigators are highly experienced and knowledgeable as evidenced by the detailed and thoughtful application. The pilot data are rich and have led to a separate aim to improve the gender inequity in post-treatment smoking cessation rates, as well as innovative mechanisms related to matching dyad pairs and social leaders emerging within each twitter group. The main weakness that somewhat lessens the impact potential is the higher educated/SES sample that will result from the inclusion/exclusion criteria and the nature of the intervention itself. It is also unfortunate that the biochemical validation was not tested during the R34 study. Weaknesses do not outweigh strengths of this excellent application, however. The intervention, if efficacious, has the potential for a relatively large impact among smoking populations who regularly use mobile phones and social media.

Strengths
Social media is increasingly popular and is an important platform on which to reach and engage smokers.
The intervention is fully automated and low cost which increases potential for implementation/dissemination. An NCI Program Director has offered support for disseminating the program on the Smokefree.gov website if found to be efficacious.
Determining ways in which to reduce gender inequity in quit rates is highly significant.

Weaknesses
Targets educated, middle/upper class populations with smartphone and/or computer for on-line access, e.g., 30 -40% of pilot participants had a college degree. Most smoking is concentrated in lower SES populations that may not have access to such resources.

Strengths
An outstanding team of investigators with diverse areas of expertise, such as marketing, health communication, smoking cessation, mobile and web-based interventions, network analyses, psychology, has been assembled.
Multiple PIs are proposed and justified; each with distinct yet complementary backgrounds and experience.

Weaknesses
None noted.

Strengths
The proposed Tweet2quit intervention is one of the first real-time, fully automated, twitter-based interventions The intervention explores novel mechanisms to promote change, including the utility of matched buddy pairs (dyadic tie strength) and -social brokers‖ (betweenness centrality) in increasing sustained cessation.

Weaknesses
Social network-based interventions are not completely novel but research in this area is in its infancy making this an important study.

Strengths
Excellent pilot data from a previous R34 is presented in support of the application and was highly informative for designing the current trial.
The design allows a test of the overall intervention in a male and female sample vs. usual care, as well as a third group for testing a manipulation targeting women (women only groups vs. coed twitter groups).
Both discussion topic and engagement feedback messages will be sent to participants with early detection of nonparticipation via website-based pattern detection processes, which will trigger additional prompts to encourage re-engagment.
Data assessing engagement from the smokefree.gov website will also be collected.
Outcomes follow recommended guidelines for smoking cessation trials and include multiple measures.
Biochemical validation is proposed which is challenging (but necessary) for large media-based interventions.
False reporting rates are taken into account in the power analysis.

Weaknesses
Study methods require participants to have a smartphone and/or a computer, reducing diversity in SES/level of functioning and potential for generalization to lower income populations who tend to smoke at higher rates.
In the pilot data analysis, nonrespondents were counted as missing but the % of nonrespondents was not reported.
Biochemical data collection via web or phone camera and mailed cotinine test strips has not been fully tested, although there is a plan to gain experience in an initial pilot phase.
Requiring daily Facebook use seems a little heavy-handed and is based on correlational data that daily Facebook users had more tweets. It could be that these participants were more interested in social media via multiple platforms and not that the use of Facebook caused more tweeting.
Effect size estimates for sustained abstinence at 6 months in the women only twitter group are proposed to be equal to the sustained abstinence rates of men at 60 days in the co-ed developmental study. This seems overly optimistic.

Strengths
Outstanding resources are available to support the proposed study at the institutions of both PIs.

Weaknesses
None noted.

Acceptable Risks and/or Adequate Protections
Risks are minimal and addressed appropriately. Appropriate precautions are in place.

Data and Safety Monitoring Plan (Applicable for Clinical Trials Only):
Acceptable o A satisfactory plan has been developed.

Inclusion of Women, Minorities and Children:
Sex/Gender: Distribution justified scientifically Race/Ethnicity: Distribution justified scientifically Inclusion/Exclusion of Children under 21: Including ages < 21 justified scientifically Men and women, all races/ethnicities, and children between the ages of 18 -21 are included.

CRITIQUE 3
Significance: 1 Investigator(s): 3 Innovation: 1 Approach: 2 Environment: 1 Overall Impact: This application seeks to revise, pilot test, and evaluate-in a fully powered randomized trial-the fully automated -Tweet2Quit‖ social media (Twitter-based) intervention for smoking cessation. If successful, this trial could validate the efficacy of a highly replicable, accessible, high-reach intervention for smoking cessation that offers a unique new option for those wishing to quit. Significance is therefore high. This application has many significant strengths, including its focus on broad implementability, basis in a strong pilot (R34), encouragingly strong pilot outcomes, thoughtful adjustments in response to those findings, careful basis in existing literature, solid and much-needed focus on gender differences in response to smoking cessation interventions, and sophisticated analysis of online social dynamics as exploratory aims. Weaknesses are present but are relatively minor, and include unclear rationale for some changes (for example, making the previously organic -buddy‖ process artificial and investigator-directed) when the pilot results were so strong; failure to provide data regarding the proportion of participants who are likely to provide cotinine data in the manner proposed; and a PI with a unique background who-although certainly not lacking in overall experience-has limited experience leading large-scale NIH studies. Although the PI did not include a senior colleague from her home institution on the proposed team, concerns in this regard are largely attenuated by Dr. Prochaska's involvement as an MPI at 30% time throughout the award period.

Acceptable Risks and/or Adequate Protections
The minimal risks in this study are adequately acknowleged and addressed.

Data and Safety Monitoring Plan (Applicable for Clinical Trials Only):
Acceptable o The investigators propose a DSM plan, under which MPIs Pechmann and Prochaska will be responsible for plan administration.

Inclusion of Women, Minorities and Children:
Sex The investigators propose targeting their internet ads such that approximately 35% of participants will be of minority status. By design, 63% of participants will be women. Children age 18-20 will be included.

Vertebrate Animals:
Not Applicable (No Vertebrate Animals)

Resource Sharing Plans:
Acceptable De-identified data will be made available to other researchers.

Additional Comments to Applicant (Optional):
Please be sure to only include relevant information in the Protection of Human Subjects section, and avoid including general methodological information.

ADMINISTRATIVE NOTE:
During the review of this application, reviewers noted that: 1) page limits for one or more sections of the application may have been circumvented by including excess text in one or more application sections that do not have specified page limits (e.g., Protection of Human Subjects), and/or 2) the appendix may have been used to circumvent the page limits of the Research Plan. In egregious cases, the NIH has the authority to withdraw such applications from review or consideration for funding . See concerns of Reviewer 3. 1 R01 CA204356-01 11 IPTA PECHMANN, C; PROCHASKA, J NIH has modified its policy regarding the receipt of resubmissions (amended applications). See Guide Notice NOT-OD-14-074 at http://grants.nih.gov/grants/guide/notice-files/NOT-OD-14-074.html. The impact/priority score is calculated after discussion of an application by averaging the overall scores (1-9) given by all voting reviewers on the committee and multiplying by 10. The criterion scores are submitted prior to the meeting by the individual reviewers assigned to an application, and are not discussed specifically at the review meeting or calculated into the overall impact score. Some applications also receive a percentile ranking. For details on the review process, see http://grants.nih.gov/grants/peer_review_process.htm#scoring. Consultants are required to absent themselves from the room during the review of any application if their presence would constitute or appear to constitute a conflict of interest.