Participants, Interventions, Outcomes
Trial design
The FreSH study is a phase 2 cluster randomized effectiveness trial. AAR + NRTS is compared to enhanced usual care (AAR + electric toothbrush; ET) in a target sample size of 1,200 patients who smoke combustible cigarettes recruited from approximately 50 dental practices in the National Dental Practice-Based Research Network (National Dental PBRN; See Fig. 1).(17, 18) The FreSH study is registered with ClinicalTrials.gov (NCT05627596) and has been approved by the University of Alabama at Birmingham IRB (#300010014), which serves as the central IRB for the National Dental PBRN. The Standard Protocol Items for Clinical Trials (SPIRIT) criteria guided the present protocol; see Appendix 1 for checklist.(19, 20)
Study Setting
The FreSH study is being conducted at 50 dental practices across the Midwest and Northeast Regions of the National Dental PBRN.(18) The National Dental PBRN is a community of participating dental practices and organizations, including private and group practices, public health clinics, community health centers, and federally qualified health centers.(21) The network was established in 2005 and includes over 7,500 participating dentists, hygienists, and other oral health professionals.(17, 22) The National Dental PBRN is funded by the National Institute of Dental and Craniofacial Research (NIDCR) and operates across six regional nodes and a specialty node; its National Coordinating Center (NCC) is based at Kaiser Permanente in Oregon and the Administrative Resource Center (ARC) is based at the University of Alabama at Birmingham (UAB). Practitioner-investigators provide input on the design, conduct, and/or analysis of studies to increase relevance to real-world care.
Eligibility criteria
Practitioner eligibility criteria include being a dentist of any specialty, dental therapist, or dental hygienist who is a member or who is willing to become a member of the National Dental PBRN; willing to consent patients to the study and provide enrollment and intervention procedures; and expects to remain in their current practice for the next year.
Patient eligibility criteria. Inclusion criteria are: 18 years old or older (except in Nebraska, where the age of consent is ≥ 19 years), and smoking at least 1 combustible cigarette per day, 25 days per month. Exclusion criteria are: does not own a smart phone with internet access; unwilling to receive study emails and text messages; past week use of tobacco cessation medication; or report of contraindications for NRT, including myocardial infarction or stroke in the past 3 months or current pregnancy/breastfeeding.
Interventions
Ask-Advise-Refer (AAR)
Practitioners in both study arms administer AAR to patients.(9) This involves asking patients about tobacco use, advising those using tobacco to quit, and referring the participant to the state quitline, electronically or via fax.
Enhanced Usual Care (Electric Toothbrush; ET)
Practices assigned to the enhanced usual care condition (AAR + ET) provide participants with a Philips Sonicare Electric Toothbrushes ($43 value). ETs were selected to incentivize participation in enhanced usual care practices. Prior to providing the ET to enrolled patients, practitioners spend 1–3 minutes instructing patients how to use the ET via a written script. Written instructions are also provided.
Nicotine Replacement Therapy Sampling (NRTS)
Practices assigned to the NRTS condition (AAR + NRTS) provide participants with a “starter pack” (2-week supply) of over-the-counter NRT, consisting of a box of 14 mg nicotine patches (14 count) and a box of 4 mg lozenges (81 count; $46 value). Prior to dispensing samples, practitioners spend 1–3 minutes reviewing instructions with the patient, via a script (Appendix 2). Written instructions are also provided.
Participant withdrawal or discontinuation of study interventions
Reasons for participant withdrawal are documented. Patient participants are instructed by study staff to discontinue use of the patch and/or lozenge entirely if a severe reaction (e.g., a reaction requiring medical attention) develops. A study physician is on call to make determinations regarding discontinuations of study interventions.
Patient Participant Study Assessments
Patient participant assessments occur at 5 timepoints: baseline (during routine dental visit), immediately following the intervention visit (within 72 hours) and 1, 3, and 6 months following the baseline visit (see Table 1). Participants are compensated $20 for each survey they complete, except the baseline survey ($80 for all surveys). The baseline survey is self-administered immediately following enrollment, using the study-provided tablet. Participants are emailed and/or texted (participant preference) a direct link to all other follow-up surveys on the first date of their follow-up window. If an assessment is not completed, study staff send the survey link every day, switching between their preferred and non-preferred contact method, for up to 5 days. Participants who do not respond via email or text will be called and offered the opportunity to complete the survey by phone.
Table 1
Schedule of enrollments, interventions, and assessments
Location | Study Task | Dental visit (Day 0) | After visit (Target date: day 1, allowable window days 0–7) | One-month follow-up (Target date: Day 30, allowable window days 28–44) | Three- month follow-up (Target day 90, allowable window 83- 104) | Six-month follow-up (Target day 180, allowable window 173–210) | CO test (Target day 180, allowable range 174–234) | Interview (Target day 180, allowable range 173- 240) |
Dental visit (Dental clinic) | Assessment of eligibility | X | | | | | | |
Baseline survey | X | | | | | | |
Written informed consent | X | | | | | | |
Dental clinic | Study intervention | X | | | | | | |
Research laboratory (remote) | Fidelity monitoring | | X | | | | | |
Assessment of adverse events | | | X | X | X | | |
Follow-up survey | | | X | X | X | | |
CO test to verify abstinence | | | | | | X | |
Interview | | | | | | | X |
Outcomes
Primary Outcome and Measure
The primary outcome is carbon monoxide (CO) confirmed 7-day point prevalence abstinence (PPA) from combustible tobacco at 6 months post-baseline, measured through a self-report item using a modified timeline follow-back procedure.(16, 23, 24) Biochemical verification is established using the iCO quit personal CO monitor.(25) Those who report abstinence on all 7 days prior to the 6 month follow-up and have a CO test result of < 6ppm are classified as abstinent. Those who report any combustible tobacco use in the past 7 days or have a CO result > 6ppm are classified as non-abstinent.
Secondary Outcomes and Measures
Self-reported 7-day PPA from combustible tobacco is also assessed at 1 and 3 months post-baseline, using the same modified timeline follow back procedure.(16, 23, 24)
Twenty-four-hour intentional quit attempts reflect intentional periods of abstinence due to a quit attempt (not abstinence due to other reasons such as hospitalization, etc.). This is assessed via self-report single survey item from the Phenx toolkit(26, 27) on the 1, 3 and 6 month follow-up surveys.
Smoking heaviness (cigarettes per day) is measured at baseline and at 1, 3 and 6 month follow-up surveys. (28, 29)
Treatment targets
Cessation medication utilization is measured via self-report survey items assessing use of all 7 FDA approved smoking cessation medications at 1, 3, and 6 months post-intervention (yes, no, prefer not to answer).(16)
Attitudes towards NRT (safety and efficacy) are measured via self-report survey items at baseline, 1, 3, and 6 months post-baseline. We assess participants’ perceptions of NRT efficacy and safety(16) on 5-point Likert-scales,(30) where higher scores indicate greater efficacy/safety.(13, 31–33)
Intention to use NRT is measured at baseline and at 1-, 3-, and 6-month follow-up. Intention to use the nicotine patch and nicotine lozenge is assessed using 2 items regarding intentions to use nicotine patch and nicotine lozenge in one of 5 categories (currently using, plan to use in the next 30 days, plan to use in the next 6 months, may use some day, will never use).
Intention to quit smoking is assessed at baseline, 1, 3 and 6 months using a self-report survey item from the PHENX toolkit where participants report if they are considering quitting in the next 30 days, the next 6 months, or not in the next 6 months(26, 27).
Sample characteristic variables
Demographic variables are collected at baseline, including biological sex, gender, ethnicity, race, age, income, health and dental insurance status, and geographic region.(34–38)
Smoking history is assessed at baseline. We assess smoking heaviness (i.e., total cigarettes per day)(28, 29), average time (minutes) to first cigarette in the morning,(39–41) count of previous lifetime quit attempts,(27, 42) past cessation treatment utilization,(16) interest in quitting and confidence to quit,(43, 44) interest in quitting and confidence to quit.(43, 44)
Oral health is assessed at baseline and all follow-ups and includes assessing toothbrushing habits (frequency of brushing,(45, 46) type of toothbrush), and perceived oral health status.(47, 48)
Process evaluation measures
Feasibility and acceptability of NRTS are measured 3 ways:
(1) Surveys: to assess acceptability, patients complete the client satisfaction questionnaire(49) regarding the FreSH study interventions during the after-visit survey.
Readiness to implement NRTS is measured for NRTS practitioners using the Organizational Readiness to Implement Change measure(50) and the Acceptability of Intervention Measure, the Appropriateness of Intervention measure, and Feasibility of Intervention measure(51) prior to and following site data collection. To allow examination of implementation variables by practitioner characteristics, we assess provider sex, gender identity, age, race, ethnicity,(34–38) job role, years in practice, and years at current practice.
(2) Interviews: 10–12 patients, 10–12 dentists and 10–12 other staff (e.g., hygienists, dental therapists) in study sites assigned to the NRTS condition participate interviews at the end of site data collection regarding the feasibility and acceptability of the NRTS intervention.
(3) Administrative data: We assess acceptability by the number of patient participants offered the intervention who accept it. We also collect the costs of the intervention including staff time, costs of printed materials, and costs of NRT to assess cost effectiveness.
Fidelity
Practitioner fidelity to the AAR intervention is assessed during the patient after-visit survey with 7 items that map on to AAR including asking if the participant smokes, asking about readiness to quit smoking, discussing cessation medications, recommending cessation medications, providing medication samples, and referring to the state quitline.(13, 16) Practitioners also attest to having provided the interventions (AAR plus sample distribution) on the tablet as part of the intervention data collection.
Sample size
The sample size and power calculations pertain to the primary outcome biochemically verified abstinence from combustible tobacco at 6 months post quit. With a sample size of 1200 and a cluster size of 24, the study was designed to have acceptable (> 80%) power to detect a difference in CO-verified quit rates of 5.5 percentage points, and good power (> 90%) for effect sizes of 6.5 percentage points or higher. Sample size estimates account for 11% lost to follow-up/withdrawal among patients. The calculations were informed by a study of nicotine replacement therapy sampling in primary care and assume an intraclass correlation coefficient of .013 and a CO-confirmed quit rate of 4.7% in the ET arm.(13, 16) This is somewhat lower than in the usual care condition from the primary care trial because it is more common for primary care physicians than for oral health professionals to prescribe NRT.(5)
Recruitment
Practitioner recruitment
Network practitioners in the Midwest and Northeast regional nodes are approached for interest in study participation by node via webinars, newsletters, announcements at network meetings, email blasts, direct mail campaigns, and directly targeting practitioners with existing relationships with node coordinators.
Practitioner training
Interested and eligible practitioners complete human subjects training and good clinical practice training, as well as a live or pre-recorded training about AAR adapted from Rx for Change(52). They then complete a live training regarding condition-specific study procedures and are given quick guides (which include an outline of study procedures) as well as a manual of operating procedures. Practitioners deliver the study interventions via a script provided on study tablets.
Patient recruitment
Practitioners or practice staff approach consecutive adult patients with a chart-documented smoking history regardless of interest in cessation and invite them to participate. Patients interested in the study undergo a self-screening process, which takes place on a tablet computer. Patients read a study description and indicate interest. Then they complete an eligibility screen. Interested and eligible participants complete an electronic written informed consent via REDCap(53, 54) electronic data capture tools hosted at the NCC. A Health Information Portability and Accountability Act (HIPAA) authorization form is completed on paper or electronically (depending on the regional IRB requirements). Then participants complete the baseline assessment. Upon completion, instructions on the screen direct patients to return the tablet to clinic staff who then provide the study intervention.
Assignment of interventions
Allocation
The NCC Data Manager, who has no contact with participants or practices, conducts randomization. Practices are randomized to NRTS or ET with a 1:1 allocation (approximately 25 practices per condition). Condition assignment is stratified by practice-level patient demographic data (highest percentage minority race/ethnicity vs. lowest) and practice payer mix (highest vs. lowest public insurance/free care) reported when the practitioner enrolls in the National Dental PBRN. We use block randomization with permutated block sizes of 2 and 4. Randomization occurs once the practice has completed all activities required for onboarding.
Blinding
Blinding of practitioners, participants, and study staff engaged in data collection is not possible, due to evident differences between the conditions. However, surveys are conducted primarily electronically. In a minority of cases, study staff may administer the survey by phone. An external researcher with no relationship with the study practices or participants will perform the statistical analysis.
Data collection, management, and analysis
Data collection methods
Participant data collection is centralized. Survey data are collected using the REDCap(53, 54) platform administered by the NCC. CO data are collected via the iCOquit personal CO monitor. Upon completion of the 6-month survey, participants who report abstinence are overnight mailed the iCOquit device. Along with the device, participants receive instructions on how to download the Bedfont Scientific iCOquit app(25) onto their smartphone, a deidentified username and password, instructions on how to self-administer the CO breath test, and an information sheet explaining the meaning of their result. Participants are encouraged to complete the test within 24 hours of receiving the CO monitor. Participants who do not complete the test receive daily reminders via text, email, or phone. Participants are paid $50 for the completion of the CO test. Study staff log into the iCOquit app to receive the result and enter it into the study database. Interviews are completed by phone and audio recorded for later transcription. Practitioner time use for cost analysis is automatically logged as they work through the study procedures on the study tablet.
Data management
All data are managed centrally at the NCC. Survey data are managed in REDCap. The NCC generates reports on data completeness and accuracy as well as protocol compliance on an ongoing basis. Data quality is assessed using measures such as time from study visit to data entry, time to resolution of data queries, number of missing forms, and proportion of all study variables queried. Interview data is transcribed and uploaded on an electronic file share system.
Statistical methods
To test the primary hypothesis, that NRTS will produce more 6-month abstinence than ET, we compare treatment groups with respect to the primary outcome (i.e., biologically verified 7-day PPA from use of combusted tobacco at 6 months) using generalized hierarchical linear modeling(55–57) with a binomial distribution and logit link (aka mixed effects logistic regression) that account for the clustering of individual observations within dental practices, while adjusting for various patient and practice related factors.(58, 59) This same model framework is used to examine the secondary outcomes of abstinence at 1 month and abstinence at 3 months and whether there was a quit attempt at 1, 3, or 6 months. To determine whether the intervention had an effect on secondary outcomes, (e.g., smoking heaviness) over time, we use hierarchical linear modeling.
We test whether NRT utilization, perceptions of NRT effectiveness and safety, intent to use NRT, and intent to quit mediate the relationship between study group (ET vs NRTS) and smoking abstinence using a cross-lagged panel model (CLPM) in a structural equation modeling (SEM) framework.(60, 61)
With regard to survey data relating to implementation, we will report on descriptive statistics of the full sample and as they relate to demographic variables.
With regard to missing data, for the primary outcome, participants who do not complete a follow-up assessment or CO test will be coded as non-abstinent. However, we will also consider sensitivity analyses using multiple imputation and complete case analysis. For longitudinal outcomes, we rely on the direct maximum likelihood approach used in the hierarchical linear model to handle missing data.
To evaluate readiness for future implementation, we conduct a qualitative analysis of stakeholder interviews. Deductive codes are drawn from the interview questions; inductive codes capture concepts that emerge from the interviews. Two coders will independently code the transcripts and meet to resolve discrepancies. A qualitative expert will conduct a framework matrix analysis(62) to identify the most effective ways to organize the content and logistics of conducting the trial to help guide our trial protocol.(62, 63)
Finally, we conduct a cost effectiveness analysis (CEA) to calculate the incremental cost-per-quit and cost-per-quit.(64) The incremental cost-per-quit of NRTS vs. ET is estimated as: (Total costs at follow-up for NRTS participants - Total costs at follow-up for ET participants)/(Total successful quits at follow-up for NRTS participants - Total successful quits at follow-up for ET participants).
Monitoring
NIDCR determined this study did not require a Data Safety and Monitoring Board or a medical monitor due to the minimal risk of the intervention and the streamlined electronic data collection.
Safety monitoring for this study will focus on Unanticipated Problems (UPs) involving risks to participants, including unanticipated problems that meet the definition of a Serious Adverse Event (SAE). Practitioners are trained to direct any participants reporting AEs to Hennepin Healthcare Research Institute study staff and will alert study staff of any reports of AEs. Adverse events (AEs) to study medication are queried in the 1-, 3-, and 6-month follow-up assessments. AEs, including SAEs, are recorded, and the principal investigator (PI) will review these events to grade severity, relationship to the study intervention, and assess whether the nature, severity, or frequency is unexpected. Safety events are recorded and reported into the National Dental PBRN safety event reporting system maintained by the NCC and to the IRB if the events are determined to be both unexpected and related to the study intervention.
Harms
The potential risks of the study are: (1) risk of negative reaction to study procedures, (2) risk of breach of confidentiality or loss of privacy, (3) risk of nicotine replacement therapy side effects, (4) risk of negative reactions to the electric toothbrush.
Auditing
The NCC developed a data management system for study collection and safety event reporting. Study progress and safety is reviewed monthly by the PI via reports provided by the NCC.
Questionnaire data is obtained electronically through the NCC-managed study-specific electronic data system (REDCap). The RedCAP provides: 1) an intuitive interface for validated data entry; 2) audit trails for tracking data manipulation; 3) access to study datasets that can be imported into common statistical packages; and 4) procedures for importing data from external sources.
Ethics and dissemination
Research ethics approval
The protocol, informed consent form, recruitment materials, and all participant materials have been approved by the sIRB (University of Alabama at Birmingham; sIRB #300010014). Any modification to the protocol will be approved by the sIRB before implementation (See Table 2 for major protocol amendments; current protocol version 3.0; 8/18/23).
Table 2
Major protocol amendments
Date | Description |
01/06/2023 | Original submission. |
02/17/2023 | Language was added to record the amount of time practitioners used to deliver the intervention, which will be used to conduct cost-effectiveness analyses. |
08/18/2023 | The after visit survey window was extended to two weeks, to increase the timeframe for data collection. It was clarified that surveys completed outside the assessment window will be accepted, to reduce administrative work from protocol deviations. References to recruitment and screening logs were removed, to minimize practitioner labor. |
Consent
For participating patients, an electronic written, informed consent form is completed via REDCap (Appendix 3). HIPAA information as required by state law is shared with the patient. The practitioner or appropriate trained office staff explain the research study to the patient and answer any questions that may arise. The rights and welfare of the patients is protected by emphasizing to them that the quality of their clinical care will not be adversely affected if they decline to participate in this study.
Providers who agree to be part of the key informant process evaluation receive a study fact sheet and completion of study surveys and/or interviews. A waiver of documentation of informed consent has been obtained.
Confidentiality
No information is given to anyone without permission from the subject. Confidentiality is assured by use of identification codes linked to the subject. HIPAA guidelines of all Network clinical sites are followed. A certificate of confidentiality has been issued for this study.
Availability of data
The NCC is responsible for creating publicly available data and resources following the completion of the study. The protocol, data dictionaries, a de-identified dataset(s), and key study publications will be posted on the National Dental PBRN public website.