Investigating Novel Smoking Cessation Primary Care Interventions in Rural Environments

Overview

The purpose of this research study is to evaluate an electronic visit (e-visit) for smoking cessation across rural primary care settings. Participants will be randomly assigned to receive either the smoking cessation e-visit or not. The e-visit will look similar to an online questionnaire asking about smoking history, motivation to quit, and preferences for medications for quitting smoking. Participants may receive a prescription for a smoking cessation medication as an outcome of the e-visit, if randomized to the e-visit group, but there is no requirement to take any medication. This study consists of questionnaires and breath samples provided at 4 separate time points throughout the study. Participation in this study will take about 24 weeks.

We will conduct a stepped-wedge, cluster-randomized clinical trial to comprehensively evaluate effectiveness of the smoking cessation e-visit within rural South Carolina primary care clinics. Consistent with a Type I Hybrid effectiveness-implementation design, we will assess implementation concurrently with effectiveness. Adult smokers will be recruited proactively across rural MUSC RHN clinics and assigned based on their clinic division's current step to either e-visit or TAU. Implementation will be assessed consistent with an adaptation of Proctor's framework proposed by Hermes et al. for digital interventions. Effectiveness Trial: Participants will be recruited from rural South Carolina primary care clinics affiliated with MUSC's Regional Health Network (RHN). The RHN is divided into four divisions: 1) Florence, 2) Marion, 3) Lancaster, and 4) Chester. Across divisions, the RHN includes 16 primary care clinics distributed across South Carolina. Seven clinics are located in rural areas, defined as RUCA codes of 4-10. Among these seven clinics, three are affiliated with the Florence division, three with Marion, and one with Lancaster (no rural clinics are affiliated with Chester). All seven rural clinics will participate in this trial. These clinics treat 3,262 adult smokers annually. RHN clinics utilize MUSC's EHR (i.e., Epic), which supports use of the e-visit developed in our pilot. As of April 2020, all PCPs affiliated with RHN clinics can provide services via e-visits. Drs. Zebian and McCutcheon are the Chief Medical Officers for the RHN, are clinical partners on this proposal, and support recruitment efforts (see letters of support). Drs. Diaz and Player are MUSC PCPs and have leadership roles as Medical Directors for Care Coordination and Primary Care Telehealth respectively. Drs. Diaz and Player, in partnership with Drs. Zebian and McCutcheon, will serve as primary liaisons between the study and clinics, ensuring that our partnering rural clinics will contribute meaningfully to recruitment, intervention delivery, and implementation evaluation. Study enrollment will begin in month 4 and will continue for a total of 44 months, ending at the end of Year 4. Final assessments will occur between months 48-54. With planned enrollment of 288 for the effectiveness RCT, we fully expect to enroll 6-7 participants per month (\~2 per week) and recruit our full sample within 44 months. In our prior work, 20% of study invitations resulted in an enrolled participant and engagement rates were similar across rural and urban patients. As such, we will send 35 study invitations per month (35 \* 0.2 = 7 enrolled participants) and 1,540 study invitations in total to meet recruitment milestones. Study invitations will be equally distributed across divisions with \~12 study invitations sent per division per month. Within divisions, study invitations will be sent proportional to the total patient volume of each individual clinic. Enrollment will be capped at 24 patients per division per step and each step will last 11 months. Recruitment will occur proactively and remotely via the EHR using the same procedures utilized within our e-visit pilot. We will conduct an automated EHR search for all patients treated in participating clinics during the past 12 months who: 1) smoke, 2) are age 18+, and 3) have activated MyChart accounts (this search was used for Research Strategy Table 2). These patients will be sent an e-mail via MyChart from the study on behalf of their primary care team inviting them to participate in a study. MUSC is an opt-out research institution. Thus, all MUSC patients, including RHN patients, are eligible to be contacted for research unless they have specifically opted out of research contact in MyChart. Less than 5% of MUSC patients have opted out of research contact, thus we expect to have access to nearly all smokers from the rural RHN clinics. Following the initial study invitation, if the patient does not complete the screening within 72-hours, our team will contact the patient via automated phone calls and/or text messages (based on preferences in the EHR; these procedures are currently IRB approved across three protocols). We do not view these repeated contacts as an impediment to scalability as organizations often send automated reminders to patients for a variety of reasons, and these reminders can be sent via patients' preferred communication channels. If interested, participants will complete an online screening within REDCap to determine eligibility. After determination of eligibility, a study team member will complete remote electronic informed consent (e-consent) with the participant via REDCap. Participants will receive a link to an electronic consent form, available via REDCap, that they can review and sign. Review of the consent form will be paired with a phone call with a member of the research team to ensure that all questions are answered prior to enrollment. This remote consent procedure is currently utilized by Dr. Dahne in both her K23 and R21 awards and has been used with success with smokers residing in rural areas. As smartphone ownership is an inclusion criterion (to provide remote CO), all participants will have internet access and thus access to the electronic consent form. This Hybrid Type I trial is designed to optimize external validity while assessing implementation. A stepped-wedge, cluster-randomized clinical trial (N=288) will test e-visit effectiveness vs. TAU. This trial will involve three clinical divisions (Florence, Marion, Lancaster) and thus three wedges. At trial outset, divisions will be randomized to active intervention (e-visit) start as first, second, or third. All divisions will begin the trial assigned to TAU and will transition to e-visit according to randomization order. Individual clinics (seven total) will be assigned to treatments based on their divisional affiliation. Participants will be recruited within clinics. After completing consent, participants will complete baseline assessments and receive the intervention currently assigned to their clinic/division (based on their last primary care visit). All participants will complete follow-up research assessments at 1-, 3-, and 6-months post-enrollment. We will require that participants complete follow-ups via their smartphone so that CO collection is seamlessly integrated with assessments. Assessments are estimated at 20 minutes. Participants will be compensated $20 in electronic gift codes for completion of each, $20 for submission of CO at each follow-up timepoint, and will receive a $100 bonus if all follow-up assessments are completed. Procedures for remote remuneration are well-established through our prior trials. At baseline, participants will self-report basic demographics including home address which will be used to determine degree of rurality. Experience using technology and internet access (home broadband, access via mobile device) will be assessed via questions from Pew Research Center's technology adoption survey. Digital literacy will be assessed at baseline via the Mobile Device Proficiency Questionnaire (MDPQ-16) and the Computer Proficiency Questionnaire (CPQ-12). Both questionnaires are valid, reliable measures of device (mobile, computer) proficiency and have been used to facilitate digital literacy training within research contexts. Cigarette smoking, use of other tobacco products (e.g., e-cigarettes), and quit attempts/quit duration will be assessed at each follow-up using a timeline followback for the last 6-months at baseline and since prior follow-up for each subsequent assessment. Nicotine dependence will be assessed at baseline via the Fagerström Test of Nicotine Dependence. Participants will report motivation to quit and confidence in quitting using a modified Contemplation Ladder. Self-reported smoking will be biochemically verified via breath CO, with abstinence defined as CO of ≤ 4ppm. Self-report and CO data will be utilized together to determine 7-day PPA. Treatment utilization will be assessed via self-report and EHR data. At each follow-up, all participants regardless of intervention will be queried for: 1) use of a cessation treatment (medication or psychosocial counseling) since the last assessment, 2) how the medication was obtained, and 3) receipt of the 5As from their PCP. Self-report data will be supplemented with treatment utilization data pulled from the EHR coinciding with each follow-up. Specifically, we will capture: 1) cessation medication prescriptions, 2) if prescribed, whether cessation medications were filled, 3) whether the participant was referred to counseling, and 4) whether the participant attended a counseling session. Confounders of CO including combustible cannabis use, secondhand smoke exposure, and environmental CO exposure within the last 24 hours will be assessed at all timepoints to account for factors that may falsely inflate CO. Additional data from the EHR will be captured to describe the sample including information on: 1) medical and psychiatric comorbidities, 2) medications, 3) tobacco-related billing codes, and 4) insurance type. Because the e-visit will be delivered remotely and the trial will be conducted remotely, biochemical verification of smoking must also be completed remotely for all participants. Following enrollment, participants will be mailed an iCO™ Smokerlyzer (personal breath CO monitor). Prior to mailing, all iCO™ devices will be tested against a fixed concentration CO cannister and only devices that test within the manufacturer's stated accuracy range (within 15%) will be sent to participants. All participants will receive their iCO™ prior to their 1-month follow-up, and we anticipate having CO readings for all follow-ups. To capture CO, after completing self-reports, participants will be instructed to sync their iCO™ via Bluetooth with their smartphone and provide CO (all in REDCap). These procedures have been developed and refined in Dr. Dahne's NCI R21. Identity will be video confirmed, and all videos will be stored in REDCap with date and time stamps. Implementation Evaluation: We will use mixed methods to assess implementation during our effectiveness trial at patient, provider, and organizational levels. Our framework is guided by the Consolidated Framework for Implementation Research (CFIR), which provides a comprehensive, pragmatic approach to understand implementation barriers, facilitators and processes. The goal is to provide an in-depth understanding of implementation acceptability, adoption, and capacity for sustainability. Specific implementation outcomes will be assessed according to Proctor's guidance, which has recently been adapted by Hermes et al. for digital intervention evaluation. These models suggest the evaluation of key implementation factors including: acceptability, adoption, fidelity, cost, penetration, and sustainability. All self-report assessments will be administered to patients in the e-visit condition during the 3-month research assessment, following completion of baseline and 1-month e-visits. Provider questionnaires will be administered via REDCap to MUSC RHN PCPs affiliated with the rural clinics involved in the trial who have at least one patient enrolled in the e-visit condition. Provider questionnaires will be sent at 6 weeks following each site's start in the e-visit arm and again at the end of Year 4. Drs. Zebian and McCutcheon, CMOs for the RHN, will aid our team to ensure high response (see support letters). Systems-level evaluation will utilize aggregate analytics supplemented with qualitative data. After the implementation period, a set of key informant interviews will be conducted with patients, PCPs, and stakeholders to enhance quantitative data. No studies to our knowledge have specifically examined implementation outcomes of proactive EHR-facilitated cessation treatments. As such, for each implementation factor, we have identified benchmarks that we believe would be indicative of meaningful uptake. These benchmarks have been selected based on prior documented rates of cessation treatment acceptance and medication receipt within primary care, Healthy People 2020's goals for cessation treatment in ambulatory settings, and prior uptake rates in response to proactive, automated cessation intervention delivery in primary care.