Tobacco use continues to be a leading preventable cause of death and disease in the United States, accounting for >480,000 deaths each year, including 41,000 deaths from secondhand smoke exposure [1,2]. The annual economic cost of tobacco use in the United States was US $332 billion in 2014 and increases every year [2,3]. Moreover, tobacco use aligns with other systemic health inequities in the United States [1,4]. The prevalence of tobacco use among Medicaid beneficiaries and individuals of lower socioeconomic status is nearly twice that among the general population [1,4]. Importantly, however, tobacco cessation results in a nearly universal improvement in health.

Although evidence-based treatments for tobacco use are effective for many, there are multiple limitations, including a substantial variability in outcomes and limited reach. Current evidence-based approaches are not effective for every individual seeking to stop smoking [2,5]. Quitlines have made standard cognitive behavioral treatment widely accessible, but they reach fewer than 1% of the individuals who smoke cigarettes annually. In addition, evidence suggests that at least some of the individuals who smoke do not wish to engage in standard cognitive behavioral treatment and would prefer a more holistic or integrative approach (ie, combining complementary or alternative and standard treatments) [6-9]. New therapeutic approaches are needed to attract, engage, and retain individual tobacco users in treatment for smoking cessation and ultimately increase the number of individuals who achieve long-term abstinence from tobacco use.

A large proportion of the American population uses integrative health approaches, including mind-body techniques, in addition to, or instead of, conventional treatment [5]. Guided imagery (GI) is a mind-body technique that involves the visualization of mental images. GI is an alternative way to develop effective strategies to quit smoking, such as developing awareness and directing attention to one’s thoughts, feelings, situational cues or triggers, and urges. GI can effectively assist individuals to quit or reduce smoking [6-8]. A randomized controlled trial (RCT) with 779 adult smokers compared the use of bupropion versus brief psychotherapy that included GI intended to enhance self-management, decidedness, assertiveness, self-determination, and self-assurance [8]. Intent-to-treat analysis showed 12-month abstinence rates of 39.1% and 12.3% for the psychotherapy and bupropion groups, respectively [8]. In another RCT, 33 adults were taught mindfulness and GI and encouraged to practice at home [9]. The results showed that home practice predicted reduced cigarette use over 4 weeks [9]. In a pilot RCT, a GI program for smoking cessation was found to be feasible and improved intermediate abstinence measures [10]. In a small trial that compared education and counseling with education and counseling plus GI training, abstinence rates were 12% and 26%, respectively [7]. Finally, our teams’ preliminary evidence provides strong support for the use of GI for smoking cessation [11,12]. These findings support the potential efficacy of GI for smoking cessation. In addition, we are proposing to deliver a GI intervention in a low-cost highly scalable way (ie, by telephone). To date, no studies have established the efficacy of an integrative GI tobacco cessation treatment delivered by telephone.

Several meta-analytic reviews have shown that proactive telephone-based tobacco cessation services are an effective way to deliver treatment for tobacco dependence [13-15]. Telephone quitlines are efficient, centralized, and a highly scalable way for individuals to access tobacco treatment in all 50 states plus the District of Columbia, Guam, and Puerto Rico. Quitlines are the largest tobacco treatment network in the United States [16,17]. Researchers have demonstrated the cost-effectiveness of telephone quitlines in the United States and Europe over the past 25 years [3,13,15,18]. However, quitlines have limited reach into the population of people who use tobacco in the United States. According to a recent study, most US quitlines reach only 1.1% of this population [19]. This situation has prompted many state sponsors to try to improve their reach by increasing the types of evidence-based treatment offered to tobacco users. Evidence-based integrative treatment approaches that are appealing to those individuals not interested in standard behavioral approaches may increase the reach and effectiveness of quitlines. Using telephone technology enables access to a wide range of people who smoke because it does not rely on having an internet connection or competence in using more advanced technology. According to a 2019 survey [20], about 96% of Americans own a mobile phone of some kind; thus, our intervention is highly accessible and scalable.

The objective of this RCT is to test the efficacy of Be Smoke Free, a telephone-delivered integrative GI treatment for smoking cessation compared with a standard, evidence-based behavioral treatment. The specific aims and hypotheses of the study are described in the following subsections.

The project has the potential for high overall impact. Should this approach become readily available to quitlines as an evidence-based treatment, it will provide a sustainable and powerful alternative to the traditional treatment options available throughout the United States. There have been few innovations in quitline treatment. Thus, increasing the treatment options might also serve to increase the reach of quitlines into the tobacco-using population.

This 5-year, 2-group RCT will test the efficacy of the integrative GI treatment versus an evidence-based behavioral treatment. Participants will be recruited from 3 states (Arizona, New York State, and West Virginia). We will randomize 1200 participants to either the GIC or the BC: 600 (50%) participants will be recruited through the quitlines (n=200, 33.3%, per quitline) and 600 (50%) through population-based methods (eg, social media, earned media, local organizations, and health care providers) to ensure that we obtain a representative sample of people who smoke. Assessments will occur at baseline (T0), 3 months after enrollment (T1), and T2. The research team will deliver the GIC and BC protocols, coordinate and conduct all assessments, and perform all data analyses at the University of Arizona. The primary outcome will be self-reported 30-day abstinence at T2. Secondary outcomes will be self-reported and biochemically verified (expired carbon monoxide [CO]) 7-day point prevalence abstinence at 6 months.

This study is being conducted in 2 phases: during phase 1, we refined protocols and procedures for the New York State and West Virginia sites for use in the RCT. During phase 2, we are conducting the RCT.

We worked with our collaborating quitlines to revise procedures for recruitment, screening, data transfer, and reporting for half of our sample. Each of our participating sites provides different services and uses different procedures for screening callers. This required us to tailor our study recruitment, screening protocols, and procedures for each site. We also worked with each quitline to identify existing networks (eg, health departments and community organizations) to add to our population-based recruitment methods for each state. In addition, we tailored the methods for transferring participant data from each quitline to our project staff in a Health Insurance Portability and Accountability Act (HIPAA)–compliant and secure manner. Finally, we worked with each site to determine the process for providing regular reports on client contacts and tracking callers transferred to study staff. Noneligible callers who are transferred by project staff to our study will be transferred back to their respective quitlines so that they may return for quitline services before study completion. We worked with staff at each site to learn about their unique structures and tailor all study processes and materials to fit both the needs of the quitline and our study. Half of our sample will be recruited through the quitlines.

We will use population-based methods (eg, social media paid advertising, earned media, and community contacts) for the other half of the sample. The materials and methods created during the feasibility trial were designed for use in Arizona [21]. Therefore, they were modified for use in the other states participating in this study. These states serve different demographics of tobacco users who may also need to be reached in different ways (eg, rural vs urban and print vs web based). We worked with our collaborators in each state to identify changes that needed to be made to the text and graphics in our recruitment advertisements and materials (eg, wording and images for advertisements placed on social media and different photographs representing tobacco users in each state for use in print materials). We convened a community advisory board (CAB) that included key informants from each state (eg, members of community health organizations identified by our collaborators) to obtain information using semistructured questionnaires. Interviews or group interviews were recorded and transcribed. The study team analyzed the comments and developed a plan for refining the recruitment methods and materials for each state.

In consultation with our collaborating quitlines, we made minor refinements to our study protocols and corresponding participant materials to address the specific needs of people who use tobacco in different states; for example, although all participants will identify their primary tobacco use as cigarette smoking, a proportion will also use other tobacco products (eg, e-cigarettes or vaping products and smokeless tobacco). The use of these other tobacco products varies widely by state. The prevalence of smokeless tobacco use is very low in Arizona but much higher in West Virginia (2.8% vs 8.9%, respectively) [22]. Therefore, we have included additional information in the protocols to address secondary smokeless tobacco use. On the basis of the information and feedback we received from each participating site, we made final changes to our protocols, which were programmed into REDCap (Research Electronic Data Capture; Vanderbilt University). Using REDCap allows our coaches to consistently deliver the protocol for each condition across participants and collect data on each session. For this RCT, we used the same REDCap system developed under the feasibility trial.

On the basis of changes to the study protocols, we made corresponding refinements to the coach training program and coach materials. Information was added regarding regional differences in smoking and other tobacco use patterns, culturally responsive counseling for specific racial and ethnic groups, issues specific to rural versus urban tobacco users, and other state-specific differences that the coaches may encounter when working with participants across our participating states.

We hired 4 coaches (2 for each condition) to deliver one of the study protocols at the University of Arizona. The training of the intervention coaches included a 2-part workshop, individual and group exercises designed to provide practical experience that resulted in competency to deliver the protocols and achieve a positive therapeutic alliance, and ongoing supervision designed to solve problems and maintain implementation fidelity. The workshops for both conditions involved the cognitive and motivational underpinnings of each program. The BC workshop focused solely on behavioral strategies for motivating and assisting tobacco users to quit and the use of nicotine replacement therapy for minimizing withdrawal symptoms. The GIC workshop included all content in the BC workshop plus the use of GI, including how to create vivid and evocative GI scripts and record them as audio files for participants assigned to this condition.

This study has been reviewed and approved by the University of Arizona institutional review board (2103633455; principal investigator [PI]: JG). All participants complete an informed consent document either through REDCap or on the telephone with trained study staff. Potential participants may ask questions of study staff before signing the consent document. The consent document describes the study, expectations, risks, and benefits of study participation. All study data are housed in REDCap and University of Arizona Box Health. Both systems are encrypted and HIPAA compliant. Only trained study staff have access to participant data, and data are restricted to study staff by role on the project. Participants may receive up to US $125 for participating in all study activities. Participants receive US $10 for completing the T0 survey, US $15 for completing the 3-month survey, and US $25 for completing the 6-month survey. All participants who report 7-day abstinence at the 6-month assessment are asked to complete an expired CO test to biochemically verify abstinence. Participants who complete biochemical verification receive US $50. Some participants will be selected based on their responses to the 3-month survey and invited to participate in a semistructured interview. Participants who complete this interview will receive US $25.

Phase 1 of the study was conducted between October 1, 2021, and July 30, 2022, and completed successfully. The study team met weekly and updated the design, content, and formatting of the intervention quit booklets that were created for the pilot study. We also created recruitment materials, including digital and print media, featuring images representing various communities. We recruited 8 stakeholders from 3 states to form a CAB. We held general and state-specific CAB meetings to obtain feedback on recruitment and intervention materials, as well as recruitment plans and methods. We also updated and refined coach training materials, revised data collection materials from the pilot study, and created new data collection materials. We developed protocols for screening, contact attempts, biochemical verification, and other data collection procedures and programmed and tested them all in the REDCap database. We developed procedures for recording and storing coaching calls using new Zoom (Zoom Video Communications, Inc) technology and hired and trained 4 quit coaches and all research staff. We received institutional review board approval for all human participants study procedures and registered the study with ClinicalTrials.gov.

We began phase 2, conducting the RCT, in July 2022. We have experienced both challenges and successes in the first 6 months of the RCT. To date, we have recruited participants through the West Virginia and New York State quitlines and via study social media posts and posts from community partners in Arizona, New York State, and West Virginia. We designed and sent recruitment postcards to former New York State quitline clients, used display ads on Meta sites, advertised on radio stations in New York State and Arizona, and distributed rack cards and flyers to our CAB members and other contacts in the community (eg, physicians’ offices). We recruited our first participant in Arizona in July 2022, then opened recruitment in West Virginia in August 2022, and in New York State in September 2022. In September 2022, the ASHLine contract was transferred from the University of Arizona to National Jewish Health, and the Arizona Department of Health Services declined to collaborate on the rest of the study. The New York State Smokers’ Quitline agreed to increase recruitment to compensate for the loss of the ASHLine.

Potential participants who were referred by a quitline or found out about the study from community-based methods were contacted via telephone calls, emails, and SMS text messages and screened by study staff. Potential participants who met the eligibility criteria completed consent and T0 assessments on the web in REDCap and were randomized to 1 of 2 study conditions.

Randomized participants who scheduled and completed the first study visit (session 1) were considered enrolled participants in the study.

To ensure that we are meeting our goals and maintaining fidelity of all procedures, we review enrollment and recruitment progress at weekly study team meetings, monitor intervention fidelity in REDCap, and review call recordings on a monthly basis. Any discrepancies in fidelity are immediately corrected, and study personnel receive weekly feedback. Coaches attend individual monthly supervision sessions with the PI.

We are currently retaining >80% of the sample at 3 months and >75% at 6 months. This is similar to our pilot study [12] and compares favorably with other studies that are conducted entirely remotely [50]. If participants miss some data collection visits or drop out from the study after session 1, partial data will be captured for them. As indicated previously, mixed effects models will be used to model smoking outcomes. Mixed effects models have an inherent way to handle ignorable missing data mechanisms (ie, missing completely at random and missing at random), which can be accounted for based on the observed data. In addition, extensive efforts will also be made to maximize retention as well as minimize missing data, and reasons for dropping out and missing data will be recorded. If the missing data mechanism is suspected to be nonignorable (ie, missing not at random), a multiple imputation–based sensitivity analysis approach [51] will be used to evaluate the impact of missing data mechanisms on the intervention effect, where the sensitivity analysis parameter controls the missing mechanism, and the multiply imputed data will be analyzed using standard methods based on the established rules in Rubin [52] that account for uncertainty owing to missingness.

We experienced several challenges during the first year of the RCT. First, we lost one of our quitline partners. Despite many attempts to replace the ASHLine, we were unable to do so. Both the New York State and West Virginia quitlines agreed to compensate for our loss of the ASHLine. However, West Virginia experienced a drop-off in call volume and staffing shortages. Therefore, the New York State Smokers’ Quitline increased their recruitment to maintain accrual goals. Although the demographics of New York State are different from those of Arizona, we will continue to conduct community recruitment from Arizona so that we can examine treatment outcomes by state.

Second, the social media advertising and earned media (eg, interviews with news outlets) strategies were not as successful as in our pilot study. Therefore, we attempted to recruit using radio advertising. However, radio advertising was not effective either. We have engaged a recruiting firm to assist with community recruitment in Arizona. Despite these setbacks, we have consistently maintained our accrual goals. If needed, we have built in an additional 3 months in year 5 for continued recruitment.

Third, biochemical verification has been difficult for many participants because of inexperience using technology. Although study staff schedule Zoom sessions to walk participants through the data collection procedure, many of the participants find it difficult or impossible to complete the expired CO test. Therefore, self-reported 30-day abstinence is our primary outcome, and we will also examine 7-day self-reported and biochemically validated abstinence.

Finally, because the intervention is conducted entirely remotely with no person contact, we experience high attrition between randomization and the first treatment session. Therefore, we classify people as participants only after they have scheduled and completed their first treatment session, and all metrics regarding outcomes are reported for enrolled participants. Participants do not know to which condition they have been randomized until the first session. Therefore, we do not expect differential attrition before session 1 to cause bias while evaluating the intervention effects. Even if the differential attrition rate is not expected to relate to the intervention conditions, an inverse probability of attrition weighting approach [53] will be used to handle differential attrition rates and account for potential bias if attrition is indeed selective.

The Be Smoke Free study is a first-of-its-kind RCT that incorporates GI into telephone-based tobacco cessation treatment. This study uses rigorous, transparent, and reproducible methods. The study protocols are based on strong theoretical frameworks and empirical evidence. We use best practices to address challenges in large studies with remote program delivery and data collection, well-established measures to reduce errors associated with self-report, successful recruitment and retention methods to reduce selection and nonresponse biases, and both intent-to-treat analysis and other methods to handle missing data.

Delivering a GI tobacco cessation intervention via telephone is a highly scalable novel integrative approach that may appeal to a broad range of smokers and increase the use of evidence-based tobacco treatment quitlines. This rigorous research has the potential for high overall impact. If successful, the Be Smoke Free program will have substantial benefits for the long-term health of people who use tobacco across the United States.