A feasibility clinical trial funded by the National Center for Complementary and Integrative Health will be conducted. Ethical approval for the procedures of this trial was obtained from the Yale University Institutional Review Board. All participants will provide online informed consent.

The study will be conducted by smartband and smartphone, and be online and optimized for mobile phones. Treatment delivery will be smartphone- and smartband-based.

Participants (N=100) will be eligible if they are 18 years or older, smoke at least 5 cigarettes per day (CPD) for at least 2 years, own an iPhone or Android, are fluent in English as intervention content was only available in English at study onset, and are motivated to quit as indicated by a score of at least 4 of 5 on one item of the Action subscale of the Readiness to Change Questionnaire, “I am trying to smoke less than I used to” from 1 (strongly disagree) to 5 (strongly agree) [37].

Participants will be recruited using online advertising (eg, Facebook ads) and directed to a study website and online screening survey. This recruitment method has been used previously to enroll 505 participants and randomize 56 participants per month in a similar smartband-based clinical trial [27]. This approach should have broad reach. The study website will link to a short screening survey on Yale Qualtrics Survey Tool. The survey will notify an individual if they are eligible or not eligible. If eligible, they will be asked for their contact information and directed to an online informed consent form. Those who provide contact information and online informed consent will receive an email with a copy of the consent form. Participants will then be asked to respond by text to confirm their interest in enrolling in the study and to schedule a short video call with a researcher for onboarding to the study technology. To finalize enrollment, participants must complete the onboarding session.

The smartband used in this study is developed by Somatix, Inc [10] and uses a machine learning algorithm to identify the hand-to-mouth gestures that characterize smoking a cigarette and differentiate these from other similar hand gestures for a given individual. Briefly, raw data are collected from the accelerometer and gyroscope sensors, and following data stabilization and noise filtering, the algorithm determines which movements being performed by the individual signify smoking. Upon smoking detection (3-4 puffs) the user is asked to confirm or deny smoking on the associated smartphone app.

Upon SMS text message confirmation of interest in enrollment, participants will be shipped the study equipment (smartband, charger, brief instructions, prepaid return envelope). Upon receipt of study equipment, a researcher will conduct an onboarding session with the participant by video call. The onboarding session will walk the participant through using the smartband and smartphone app, including charging and setting up the smartband, pairing and using the smartband and app, and reviewing study instructions. At the end of the onboarding session, participants will be sent a reminder of the study instructions by email. That email will also include troubleshooting tips to use as needed during the trial. Additionally, the email will include a recommendation for quit smoking medications [6]. After onboarding, participants will be SMS text messaged a link to the baseline survey to begin the study. Treatment starters will be defined as completing onboarding and wearing the smartband for 1 day.

Treatment retention will be defined as wearing the smartband on 75% (45/60) of treatment days. Retention rates will be ensured by emphasizing the importance of follow-up at study initiation, onboarding, and at each study survey; paying US $10 for onboarding, US $20 for the survey at 21 days, US $10 each for surveys at 28 and 60 days, and US $1 per day for wearing the smartband for 12 waking hours; paying for surveys immediately upon survey completion; and paying a US $50 completion bonus at the end of the study for completing all parts of the study.

All participants will receive a smartband and smartphone app for the smartband (Figure 1) and take part in a 60-day intervention (Figure 2). For the first 21 days, the smartband will automatically monitor and detect smoking and notify the participant of smoking events in real time to bring awareness to smoking behavior and triggers. When smoking is detected, the smartband will vibrate and the participant will receive a notification on their smartphone app asking them to confirm or deny smoking. The notification will remain until they reply. Alternatively, if a participant smokes and it is not detected by the smartband, participants will be asked to manually report the smoking event on the smartphone app. They will be instructed to smoke as much or as little as they like [31] but will be asked to set a quit date in 30 days. Participants will be instructed to wear the smartband for 12 waking hours per day. They will be automatically sent a reminder notification from the app if they wear the smartband for <12 hours each day. Additional SMS text message reminders will be sent by the study team if they wear the smartband for <12 hours per day for 48 and 72 hours. If they do not wear the smartband at all, they will receive an additional SMS text message, email, or phone call at 48 and 72 hours. Data from this initial 21 days of smoking monitoring and detection will be used to predict individual patterns of smoking for RAIN mindfulness exercise delivery later in the intervention.

At 21 days, participants will be SMS text messaged a link to the 21-day survey. At the end of the survey, they will take part in a brief mindfulness training adapted from Craving to Quit, a smartphone app for mindfulness training for smoking cessation [27,38]. The mindfulness training uses videos and animation to introduce mindfulness, introduces habit formation using an animation, and provides a guided mindful smoking exercise, which helps participants bring awareness to the present moment effects of smoking to gain a clear recognition of the effects of smoking. After the mindfulness training, participants will be instructed that, for the next week, the smartband will continue to automatically monitor and detect smoking, notify them of smoking events in real time, and ask them to confirm or deny smoking. They should continue to manually report any smoking events that are not detected by the smartband. In addition, anytime smoking is detected by the smartband or manually reported, they will automatically receive an SMS text message with a link to the mindful smoking exercise. They can also access mindful smoking anytime by clicking the link in their text messages. They will be instructed to try to use the mindful smoking exercise any time they smoke. After the mindful smoking exercise, they will be asked to rate how helpful they found it. The link will direct the participant to a 2-minute audio-guided mindful smoking exercise with subtitles, followed by an ecological momentary assessment (EMA) rating, “How helpful did you find this exercise?” (visual analog scale [VAS]: “Not at all helpful” to “Very helpful”).

At 28 days, participants will be SMS text messaged a link to the 28-day survey. At the end of the survey, they will take part in another brief mindfulness training adapted from Craving to Quit. The mindfulness training uses videos and animation to introduce the concept of craving, including using an animation with the metaphor of craving as a tantrum toddler (ie, let the toddler cry it out, introduces the concept of urge surfing, and provides a guided mindfulness exercise, RAIN) [21]. This exercise teaches smokers how to work mindfully with triggers to smoke using RAIN. Smokers learn to recognize cues, cravings, or affective states that trigger smoking; allow and accept their experience, however pleasant or unpleasant, without trying to change it; investigate what the experience feels like in the body, emotions, and thoughts; and nonidentification with what is happening from moment-to-moment (RAIN). By openly investigating one’s present moment experiences of craving in a nonjudgmental way, one may learn that cravings are physical sensations, thoughts, and emotions that will subside. After the mindfulness training, participants will be reminded of their quit date, and instructed that, for 30 days after their quit date, the RAIN exercise will be sent to them by SMS text message at the times they typically smoke. They can also access the RAIN exercise anytime by clicking the link in their SMS text messages. They will be instructed to try to use the RAIN exercise any time they have a craving to smoke. RAIN will be delivered according to individualized smoking profiles generated from smoking data from the first 21 days of the intervention. Briefly, a prediction algorithm is run every 5 minutes and if >1 smoking event is detected within 15 minutes of the current time on any day in the 21 days, the answer is true. A text cooldown period is set at 45 minutes such that RAIN will not be sent twice within 45 minutes. These parameters were determined by an initial pilot study (n=8 completers) to provide support but minimize text burden. Before and after the RAIN exercise, they will be asked to rate their craving and affect [39]. After the RAIN exercise, they will also be asked to rate how helpful and timely they found it. The link will direct participants to a 2-minute audio-guided RAIN exercise with subtitles, preceded by rating, “How much are you craving a cigarette right now?” (VAS: “Not at all” to “Very much”) and “How are you feeling right now?” (VAS: “Very bad” to “Very good”), and followed by the same ratings plus “How helpful did you find this exercise?” (VAS: “Not at all helpful” to “Very helpful”) and “How was the timing of this exercise?” (VAS: “Too early” to “Too late”). Additionally, participants will be instructed that the smartband will continue to automatically monitor and detect smoking, notify them of smoking events in real time, and ask them to confirm or deny smoking. They should continue to manually report any smoking events that are not detected by the smartband. Anytime smoking is detected by the smartband or manually reported, they will continue to automatically receive an SMS text message with a link to the mindful smoking exercise.

Throughout the intervention, participants can delay or skip mindfulness exercises by ignoring the SMS text message. They can also access the mindfulness exercises at any time by clicking the link in their SMS text messages. Completion of mindfulness exercises will be time-stamped including onset and offset times for the 2-minute guided audio. To encourage completion of mindfulness exercises, participants will be sent encouraging SMS text messages indicating the number of exercises they completed on the previous day with congratulations or a reminder to use the mindfulness exercises if none were completed. These texts will be sent daily for mindful smoking (week 4) and for the first week of RAIN (week 5), and weekly thereafter for RAIN (weeks 6-8).

At 60 days, participants will be SMS text messaged a link to the 60-day (end of treatment) survey. At the end of the survey, individuals who report 1-week point prevalence abstinence from smoking (up to n=30) will take part in carbon monoxide (CO) monitoring. CO levels will be measured using piCO+ Smokerlyzer breath CO monitors (Bedfont Scientific Ltd). Participants will be shipped the CO monitor, brief instructions, and a prepaid return envelope. They will set up a video call with a researcher (recorded), during which they will use the CO monitor and display the output [27,40,41]. Once a participant has completed the 60-day survey and returned any study equipment, they will be paid their final study payment.

Descriptive statistics of primary and secondary outcome measures will be reported prior to statistical analysis. For treatment fidelity, adherence, and acceptability, we will estimate proportions with 95% CIs. Change in smoking will be assessed using mixed regression models to compare self-reported and smartband-detected CPD, using all available data with focused contrasts on changes from baseline to 60 days. Random effects or structured variance-covariance matrix of the errors will be used to take into account correlations of repeated measures within an individual, and the best-fitting model will be selected based on Bayesian information criterion. We will perform exploratory mediation analysis of the exploratory outcomes to test mechanistic hypotheses related to smoking, craving, affect and mindfulness, and other psychological variables.

This trial is designed to test the feasibility of using a smartband and smartphone to automatically monitor and detect smoking, notify smokers of smoking events in real time, and deliver brief mindfulness interventions to help daily smokers in a smoking quit attempt. We anticipate collecting important data and information regarding innovative features of the intervention including using wearable technology (smartband) to monitor and detect (track) smoking, automatically detecting smoking rather than relying on self-report, using data from wearables to identify patterns of individual smoking behavior over time, delivering interventions contingent on smoking events and targeted to predicted smoking events, and combining novel mHealth technology with an evidence-based mindfulness intervention for smoking cessation.

There are several potential limitations of this study. First, it is possible that data from the initial 21 days of tracking smoking by smartband may not be sufficient to obtain predictable individualized smoking profiles. This is an empirical question—the trial will provide useful data and information on our ability to predict smoking based on 21 days of smartband data. Second, smoking patterns may change over time in response to being monitored for smoking and other factors. Averaging smoking patterns for an individual across 21 days allows us to account for these changes but stay in line with clinical guidelines for setting a quit date at 30 days [56]. Third, additional psychological mechanisms other than craving and affect may impact smoking. Once the feasibility of the approach is established, future studies can include additional EMA ratings to evaluate other key psychosocial factors (eg, anxiety, stress, or activity). Fourth, an alternative design would be to combine the smartband technology with another full-scale treatment for smoking such as the Craving to Quit app [27]. However, it is a challenge in clinical trials of smartphone apps to determine efficacious components of multifeatured apps. This study takes a component approach by testing mindful smoking and RAIN exercises, which also builds upon prior work including our own demonstrating that brief mindfulness interventions can reduce craving and smoking. Last, individuals may use the technology inappropriately, such as wearing the smartband on the hand not used for smoking or not wearing the smartband upon waking up. Although we cannot completely control for missed cigarettes due to these or other factors, tracking smoking by smartband vastly improves smoking estimates beyond self-report, and this study will be one of the first to test how smartband-based smoking detection relates to self-reported CPD. These activities and other factors will also be queried on user experience questionnaires at each time point.

This study will provide data and information on whether smartphone- and smartband-based smoking monitoring and detection, notification, and delivery of brief mindfulness interventions aid daily smokers in a quit attempt. Feasibility outcomes will be used to inform future randomized clinical trials of the tested technology and intervention.