Recruitment, participants, and setting
The study is being conducted at a single university-based site in Kansas City, Missouri. The study protocol is in compliance with the Helsinki Declaration and was reviewed and approved by the Institutional Review Board of the University of Missouri – Kansas City (#0978). Smokers are recruited community-wide through word of mouth, newspaper ads, flyers, billboards, internet advertising, and physician referral using printed cards. Recruitment materials invite participation in a study “for smokers” or “smokers not quite ready to quit” and provide a study phone number to call for more information. Potential participants who enquire about the study are informed that the goal is to learn how best to talk to smokers about their health. They are informed that although their smoking will be discussed they will not be required to quit. Eligibility criteria are being 18 years or older, English speaking, reporting smoking a minimum of 1 cigarette per day, having a mailing address and telephone number, willingness to participate in all study components, not currently pregnant or planning to become pregnant in the next 6 months, not currently using a smoking cessation medication, not planning to move from the metropolitan area in the next 6 months, and not currently motivated or ready to quit smoking (defined as scoring 6 or less on a 0–10 point scale of motivation to quit smoking and having no intention to quit in the next 7 days). Potential participants are pre-screened by phone and those likely to be eligible are scheduled for a baseline visit at which time they are re-screened and enrolled if eligible. After the first 24 participants were enrolled, confirmation of self-reported smoking status using a carbon monoxide monitor (Bedfont Scientific piCO + Smokerlyzer®) was added to the re-screen. Eligibility is determined by a CO level of 7 ppm or higher [19, 20].
MI consists of 4 sessions (baseline, week 6, week 12 and week 18) of approximately 20 minutes each conducted in-person (baseline and week 12) and over the phone (week 6 and week 18). The 4-session schedule is altered should participants set a quit date. All remaining sessions are scheduled on the day after the selected quit day and then every week after that. Although the optimal number of sessions of MI is not known [12, 17] four sessions were chosen as a practical compromise between having too many sessions for participants who do not progress toward making a quit attempt and having enough sessions to both motivate and then assist smokers who do progress to making a quit attempt.
In MI counselors assist participants to explore and resolve ambivalence regarding quitting smoking, consistent with MI principles and strategies described by Miller & Rollnick . MI emphasizes a collaborative, evocative, and autonomy supportive counseling approach, using specific communication skills (open-ended questions, affirmations, reflections, and summaries) to express empathy, develop discrepancy between client goals/values and current behavior, increase change talk, and “roll with” statements of resistance to change. MI sessions are characterized by a collaborative, interactive counseling style in which participants are engaged in the process of thinking and talking about their smoking behavior through the counselors’ use of strategic reflective listening and occasional open-ended questions and summaries. In MI, providing information on health-risks and the benefits of quitting is minimized and offered by counselors only when the participant asks for or appears clearly interested in the information.
Participants who express interest in quitting receive a self-help guide for quitting and, consistent with MI methods for strengthening commitment for change, are encouraged to develop a 5-step plan for quitting (set a quit date, change environmental triggers, prepare for triggers and urges, reward yourself, and use medication) based on the Clinical Practice Guideline . However, to remain consistent with MI this step is delivered in an MI style (i.e., maintaining the principles and specific communication skills that characterize the MI approach). For example, counselors continue to support patient autonomy, use reflective listening and encourage the participant to identify potential solutions before offering their own suggestions or giving advice. Participants who choose to set a quit date within the period of their planned participation in the study are offered a 12-week supply of free pharmacotherapy for smoking cessation (described in more detail below).
Meta-analyses have indicated that manualized MI interventions may produce poorer outcomes [14, 17] so MI is implemented without a structured guideline for the flow of the session (with the exception of being required to have the participant consider all 5 of the elements of the plan to quit should the participant express interest in quitting). Counselors are trained to use the various tools and strategies of MI as they see fit.
HE is designed to provide a plausible alternative intervention to MI that is equivalent in contact time and that can be delivered with fidelity. To match MI there are 4 HE sessions of approximately 20 minutes conducted on the identical schedule and in the same format (phone versus in-person) as MI. For participants who set a quit date the schedule is altered in the same fashion as in the MI arm.
In HE counselors deliver health education designed to persuade participants to quit. The method of motivating participants is based on providing a strong rationale for quitting based on the relevant risks of smoking, rewards of quitting, and addressing the roadblocks to quitting. These elements are consistent with the Clinical Practice Guideline, which recommends the use of the “5 R’s” (i.e., discuss patient relevant risks of smoking, rewards of quitting, roadblocks to quitting, and repeat at each visit), but excludes the Guideline’s recommendation to use MI principles in their delivery. To ensure HE is distinct from MI, counselors in this arm follow a semi-structured script and use printed slides (when counseling is in-person) to deliver the intervention.
The 5-part counseling protocol begins with an assessment of participants’ smoking and quitting history and their experience of common symptoms from smoking. Counselors point out the link between participants’ current symptoms and their smoking. In the absence of symptoms counselors underscore that the best thing that participants can do for their health is to quit smoking. Second, counselors describe major long-term (e.g., heart-disease) and short-term (e.g., slower wound healing) risks of smoking (4–5 at each session) and provide education on one of a variety of smoking related topics (i.e., content of cigarettes, costs of smoking, the addictive nature of tobacco, luring of smokers by tobacco companies, and the dangers of second-hand smoke). Third, counselors discuss the potential rewards of quitting including short and long-term health benefits. Fourth, counselors ask about participants’ roadblocks to quitting and provide suggestions or counterarguments for each. Finally, counselors provide personalized advice underscoring the importance of cessation and ask participants if they are interested in making a plan to quit. HE is designed to be warm and supportive (i.e., not confrontational) but differs from MI in that the primary focus is on giving information rather than eliciting participant engagement in considering their smoking behavior. Participants are engaged with a few specific questions that are used to guide counselors in providing appropriate information (e.g., suggestions for overcoming the specific roadblocks mentioned by the participant).
As with the MI arm, participants who express interest in quitting receive a self-help guide for quitting, and are encouraged to develop a 5-step plan for quitting based on the Clinical Practice Guideline . Participants who choose to set a quit date within the period of their planned participation in the study are offered a 12-week supply of free pharmacotherapy for smoking cessation.
BA is designed to mimic usual care and provide a comparison treatment that is consistent with that used in many prior MI studies for smokers [12, 22]. The design of BA is based on the recommendations of the Clinical Practice Guideline . Participants meet with a counselor for approximately 5 minutes during which they are asked about common smoking related symptoms and are provided with clear, strong, personalized advice to quit. Smokers are then asked if they are interested in quitting and, if so, are provided with a self-help guide to quitting and asked about their planned quit date. Participants who choose to set a quit date within the period of their planned participation in the study are offered a 12-week supply of free pharmacotherapy for smoking cessation.
Provision of pharmacotherapy
All participants who set a quit date are offered pharmacotherapy to assist with their quit attempt to meet standard of care. Due to evidence of its potentially superior efficacy [23–25], counselors encourage the use of varenicline (Chantix®) when not contra-indicated. Participants who cannot or who do not wish to use varenicline are offered nicotine replacement therapy in the form of the patch or the lozenge. To avoid free medication serving as the primary motivator for smokers to quit, medication is not mentioned or offered to participants until after they have demonstrated a commitment to make a quit attempt by setting a quit date.
Participants who decide to use varenicline are provided with a 12-week supply (a starter pack and 2 continuation packs). Participants who choose chose the patch receive an 8 week supply if they smoke less than 10 cigarettes per day (14 mg for 6 weeks, 7 mg for last 2 weeks), a 10 week supply if they smoke 10 to 39 cigarettes per day (21 mg for 6 weeks, 14 mg for 2 weeks, 7 mg for 2 weeks), and a 10 week supply if they smoke 40 or more cigarettes per day (42 mg for 6 weeks, 28 mg for 2 weeks, and 14 mg for 2 weeks). Participants who choose the lozenge receive a 12-week supply (2 mg if they smoke their first cigarette 30 minutes after waking and 4 mg if they smoke within 30 minutes of waking). Participants are given instructions for the proper use of their medications at the time that medications are dispensed. Participants who decide to use medications during a phone counseling session are scheduled for a brief medication-dispensing visit. Participants with cautions for using varenicline (e.g., history of depression) are monitored closely for adverse events by means of weekly phone calls.
Interventionists and training
Three Master’s level health professionals with prior training and experience using MI are trained to deliver all 3 interventions to eliminate any potential of confounding across arms due to counselor specific effects. The use of dedicated, specifically trained staff (as opposed to nurse or physician providers for example) is necessary because of the complexity of delivering all three of these treatments with high fidelity. While this diminishes generalizability of the results to primary care, meta-analyses suggest that the effects of MI are at least similar , if not improved, when delivered by general practitioners .
Training for MI consisted of refresher training in the practice and principles of MI through reading materials, video demonstrations, and a half-day training workshop; training in the MI protocol for the study (i.e., procedures and requirements for implementing MI in this study); and completion of role-play practice sessions with feedback from the MI arm supervisor in a group setting (i.e., with all counselors present). Counselors then practiced the MI protocol with pilot participants and received group supervision until they met criterion on rating scales designed for fidelity monitoring (described below) for 3 consecutive sessions.
Training for HE and BA was similar, including training in the HE and BA protocols for the study (i.e., procedures and content for implementing HE and BA in this study) and completion of role-play and pilot participant practice sessions accompanied by group supervision until 3 consecutive sessions met criterion.
Fidelity assurance procedures
All counseling sessions in the study are digitally recorded. Counselors alternate in receiving regular group supervision by separate expert supervisors (weekly for MI, every other week for HE, and monthly for BA). For each supervision session an audio-file is either randomly selected from those completed by the counselor since their last supervision session, or the counselor identifies a challenging session that they wish to receive feedback on. The supervisor provides verbal feedback as the group listens to the session and rates adherence to the respective protocols using rating scales adapted for this study. The rating scales assess procedural and content requirements for each arm. For example, the MI rating scales assess completion of basic procedures (e.g., remind the participant that the session is being audio recorded) and fidelity on 13 elements of MI (e.g., uses reflective listening, rolls with resistance vs. confronting, affirms and builds efficacy) and 1 global rating (how well did the counselor conduct the session). HE and BA rating scales similarly assess fidelity on basic procedures and key counseling elements (e.g., providing personalized advice, recommending quitting, linking participants’ symptoms to smoking). To externally verify that MI and HE sessions differ as expected, 10% of the MI and HE sessions (excluding those sessions that involve making quit plans which are expected to be quite similar) are randomly selected for coding on adherence to MI principles using the Motivational Interviewing Treatment Integrity code  by an expert independent coding group. Coders are blind to the study arm of the session and the study hypotheses. We anticipate that MI sessions will receive scores of 4 or higher on the 1–5 global rating and be significantly higher than HE sessions on ratings of evocation, collaboration, empathy, and autonomy support. With respect to frequency measures of counselor behavior we expect MI sessions to have a higher reflection to question ratio and significantly fewer instances of giving information.
Participants complete computer-administered assessments at baseline, week 12, and 6-month follow-up. Baseline variables include socio-demographic characteristics (gender, age, education, employment), smoking characteristics (cigarettes smoked per day, number of years smoking, number of prior quit attempts), level of nicotine dependence using the Severity of Dependence Scale , and single item measures of motivation and confidence to quit [28–30]. The primary outcome measure is the occurrence of any quit attempt defined as a serious quit attempt of at least 24 hours  by 6-month follow-up. The secondary outcome is biochemically verified 7-day point-prevalence abstinence at 6-month follow-up [32, 33]. Biochemical verification is conducted by research staff using cotinine test strips for saliva . Hypothesized mediators of MI’s effect on quit attempts are autonomous motivation and autonomy support which are assessed using the Treatment Self-Regulation Questionnaire  and the Health Care Climate questionnaire [36, 37], respectively. For those who decide to make a quit attempt, weeks of pharmacotherapy use is assessed with a self-report checklist  and will be examined as a mediator of MI’s effect on cessation.