Severe obesity (class II and III; BMI ≥35 kg/m²) affects approximately 9.5% of Canadians [1]. Metabolic and bariatric surgery (MBS) is generally effective in inducing significant and lasting weight loss [2] while reducing the costs related to severe obesity and associated complications [3,4]. MBS offers significant health benefits and improves the quality of life [4,5]. The number of procedures performed worldwide has increased during the last decade in particular, reaching almost 700,000 surgeries performed in 2018 [6]. However, approximately half of the patients who have undergone MBS regain at least 10% and 20% of their maximum weight loss 1 year and 5 years after reaching their nadir weight, respectively [7].

In the context of MBS, physical activity (PA) is recommended given its multiple benefits [8,9]. Studies have shown that pre- and post-MBS PA can improve physical fitness, cardiometabolic health, quality of life, and depressive symptoms [10-17]. However, 70% to 95% of adults are inactive before MBS and remain so after MBS [18,19]. Thus, there is a need to develop interventions with the aim of increasing pre- and post-MBS PA. Introducing PA behavior change interventions before MBS with additional support offered after MBS seems to be an ideal approach to optimize MBS benefits and provide transitional support toward post-MBS lifestyle [20]. The most compelling evidence to support this assertion comes from studies that have shown that PA interventions delivered before MBS can increase PA until 1 year after MBS [21,22]. Furthermore, pre-MBS PA can improve physical fitness, which may help reduce perioperative complications and length of hospital stay [23]. Finally, participants’ motivation and readiness for PA increased close to their surgical date [24].

Supervised exercise within the context of clinical- and research-based interventions, although effective in conferring health benefits, does not increase long-term PA in adults undergoing MBS [17,25,26], likely because this type of intervention almost never includes behavioral change methods (ie, no target of PA determinants such as self-determination and self-efficacy that may help with maintenance of PA behavior). Interventions that apply methods specifically targeting PA determinants are needed to equip adults with the knowledge, skills, and confidence to engage in regular PA on their own, as resources are not infinite. Therefore, PA behavior change interventions should be considered, which can be effective in increasing PA in adults with obesity [27]. In adults undergoing MBS, Bond et al [22,28,29] have shown that BariActiv, a 6-week intervention delivered before MBS, which included weekly one-on-one meetings and focused on behavior change techniques (BCTs), increased moderate-to-vigorous intensity PA (MVPA) and step counts after the intervention, and these changes were sustained 6 months after MBS. Thus, PA behavior change interventions seem promising to increase MVPA in the midterm and, in turn, limit weight regain to maintain health improvements. However, additional studies investigating the effects of behavior change interventions on MVPA are required owing to the heterogeneity and small number of studies available and the divergent results across studies with adults undergoing MBS [30].

PA behavior change interventions for adults with obesity include center-, community-, and home-based programs. However, alternative delivery formats, including email, phone, or videoconferencing counseling, have also been shown to be effective [58]. Videoconferencing counseling, otherwise referred to as telehealth interventions, have been introduced to address key issues and patients’ preferences (eg, lack of time and resources, supervision, face-to-face, accessibility, and simplicity) while considering current contextual challenges (eg, limited financial and human resources, travel to and from MBS or research centers, pandemic lockdown, and environmental concerns) and maintaining opportunities for meetings with health professionals as per patient preferences. Therefore, it is important to develop and implement behavior change interventions using telehealth.

Telehealth refers to the use of the internet and connected technologies to provide care [59], and adherence rates as acceptability of such interventions are high (≥75%) [60-64]. The feasibility of telehealth interventions is very promising given the ubiquitous access to the internet and technologies in Canada, with 94% of residents having access to the internet and 88% owning a smartphone [65]. Moreover, telehealth is cost-effective and has already been shown to improve health outcomes [60,62,66] and PA among adults with or without obesity [67,68]. Although many studies are available on telehealth interventions with patients who had undergone MBS [69-74], to our knowledge, only 3 of them included a PA behavior change intervention: 1 feasibility study before MBS (before-after study design) [72]; 1 feasibility study after MBS (before-after study design) [73]; and 1 ongoing randomized controlled study after MBS [74]. However, PA was generally self-reported and was only measured in the short term, introducing measurement bias and preventing the assessment of PA maintenance [71]. Robust and innovative study designs, such as a single-case experimental design with long-term objective PA measures, should be used in future studies to assess the effects of telehealth PA behavior change interventions in the short and long term. Therefore, Bari-Activ, having already shown its efficacy [22,28,29], was revised to reinforce theory-based content, reflect intended users’ perspectives and the context in which the intervention will be conducted, and provide a means to increase reach (via telehealth). It will be tested within a single-case experimental design to ensure that it is feasible and applicable to intended users and implementers and generate data to inform a large-scale trial to test whether it is effective.

This manuscript reports the protocol used in this study. The first objective is to assess the feasibility and acceptability of the protocol or methods and the Telehealth Bariatric Behavioral Intervention (TELE-BariACTIV), an intervention designed to increase MVPA that is guided by two perspectives: (1) a multitheory approach, thus targeting SDT and SCT constructs and integrating related MBCTs, and (2) a patient perspective. The second objective is to generate an estimate of the effect of the TELE-BariACTIV intervention on presurgical MVPA (the primary outcome for a future large-scale trial) to determine the appropriate sample size for a multicenter large-scale trial.

The study uses a phased approach to behavior change intervention development supported by the Obesity Related Behavioral Intervention Trials model [31].

The study results will be reported according to the recommendations for single-case protocols in behavioral interventions [75]. Owing to the objectives of the study (feasibility, acceptability, and effect size exploration) and recommendations [31,76], this study is a multicenter study with a single-case experimental design with multiple base levels (ABAB’A). It is an open-label trial in which research staff, researchers, and participants are not blinded to the study objectives, intervention, and allocation, given the nature of the intervention and measures.

Approximately 2 weeks before the baseline assessment, the study will be explained by phone to potential participants and the selection criteria, checked. Eligible and interested participants will be sent an accelerometer and a A370 Polar (Polar Electro) watch with cables to recharge and information sheet by mail with a prepaid return envelope. Participants will be instructed to initiate the baseline assessment on the web (Limesurvey) once they receive their packages, including an informed consent form. Then, participants will be randomly allocated to a 1- or 2-week baseline A1 phase using a list of random numbers generated electronically by a person outside the project, and they will start the 5 phases of the study (A1-B1-A2-B2-A3; Figure 1). The A phases are observational phases without intervention, and the B phases are intervention phases with PA counseling delivered via Zoom technology platform. Participants had different dates of starting the study owing to the variability in their MBS dates. Phase A2 may also vary in length because the exact MBS date is usually not known when participants are to start the study.

To meet the design standards that recommend at least 3 measurements per phase [75,77], MVPA is assessed continuously over 7 to 14 days during each phase. The storage limit of the accelerometer is 180 days, which ensures the feasibility of the measurement. Therefore, participants have to download the free Actisync application on a computer, connect the device with a cable, and then upload their anonymized accelerometry data using the Actigraph Centerpoint minimum every 180 days. Quality of life, anxiety and depressive symptoms, motivational regulations, basic psychological needs satisfaction and frustration, self-efficacy to overcome barriers to PA, PA enjoyment, and social support for exercise are assessed using a series of validated questionnaires 6 to 10 times via Limesurvey (Table 1). In addition, participants are asked to respond to questions regarding the acceptability of the protocol and intervention in web-based questionnaires before and after the intervention (phase B1), as well as during a web-based semistructured individual interview 1 to 4 weeks after phase B1. Medical records will be consulted at the end of the study with the participants’ authorization to validate the MBS date, medical conditions, and weight loss.

Participants are adults residing in or around 3 cities (ie, Montreal, Quebec City, and Chicoutimi) in the province of Quebec (Canada) and receiving care at either one of 2 tertiary care centers (Institut Universitaire de Cardiologie et de Pneumologie de Québec-Université Laval and the Centre Intégré Universitaire de Soins et de Services de Santé [CIUSSS] du Nord-de-l’Île-de-Montréal) or at a regional center (CIUSSS du Saguenay–Lac-Saint-Jean). Recruited strategies include (1) placing posters in waiting rooms at the target centers and (2) asking health professionals from hospitals at each site to refer to potentially eligible patients. For the latter, health professionals will provide the research team with the name and contact information of patients who have a planned sleeve gastrectomy in >12 weeks and who have provided verbal consent to be contacted by the research team for information on the study. The target sample size will be 12 (approximately 4 per site), which is considered acceptable for a single-case experimental study with multiple base levels, as participants serve as their own control, and outcomes will be measured repeatedly across observational and interventional phases [75,77].

The inclusion criteria are (1) ≥18 years of age; (2) scheduled to undergo sleeve gastrectomy in >12 weeks in one of the 3 associated hospitals (University Institute of Cardiology and Pulmonology of Quebec, Quebec; Sacré Cœur Hospital, Montreal [CIUSSS du Nord-de-l’Île-de-Montréal]; Chicoutimi Hospital, Chicoutimi [CIUSSS Saguenay–Lac-Saint-Jean]) Qc, Canada); (3) self-report ≤150 minutes of MVPA per week; and (4) access to a computer with internet and an interface with a camera. Exclusion criteria are (1) having a contraindication to PA without medical clearance (the Get Active questionnaire from the Canadian Society for Exercise Physiology [78] is used to determine if medical clearance is needed); (2) already enrolled in a supervised exercise intervention or PA behavior change intervention; (3) inability to speak and understand French; and (4) need a wheelchair, cane, walker, or other support or supports to move.

During phase B1, participants will receive 6 real-time, face-to-face sessions with the same PA counselor (to the extent possible) over a 6-week period before MBS. The sessions will last approximately 45 minutes and will be conducted via videoconferencing using Zoom. The TELE-BariACTIV intervention is a multicomponent intervention derived from a previous intervention by Bond et al [22,28,29], which was developed based on multiple theories, including SDT and SCT, and was revised by the current research team. The intervention targets SDT and SCT constructs, focusing specifically on (1) providing autonomy support, structure, and interpersonal involvement; (2) increasing perceptions of autonomy (ie, perceived control over one’s actions by providing a rationale, structure, and emphasizing responsibility), competence (ie, perceived mastery of tasks and skills by providing support and encouragement, information feedback, and support barrier identification), and relatedness (ie, perceived belonging and connection to others by encouraging to seek social support); (3) increasing autonomous motivation for PA (ie, acting through self-endorsement and volition because the activity holds inherent interest or personal value); and (4) fostering perceived self-efficacy to overcome PA barriers and engage in PA. Accordingly, the intervention integrates several content- and relational-based techniques hypothesized to target the theoretical constructs. The techniques are selected on the basis of the theoretical construct or constructs they are proposed to target, wherein the research team drew on published literature [105-108] to determine which techniques target specific SDT and SCT constructs to guide the selection. The MBCTs used to guide the content of the intervention sessions during phase B1 are presented in Table 2.

In addition, the PA counselor uses motivational interviewing techniques [108] (ie, empathic or reflective listening, asking open-ended questions, prompting the participant to ask questions, prompting participants to offer solutions, seeking permission to provide information and advice, shifting focus, supporting change or persistence, and showing unconditional regard) that align with SDT and SCT for the relational component of the intervention to support the delivery of the content. The goal of the intervention is to address participants’ mindset about PA, as well as their motivation to engage in PA, with the goal of supporting PA uptake and maintenance. To this end, participants are encouraged to progressively increase their daily PA to achieve the current PA guidelines of 150 minutes per week of MVPA by the end of phase B1 (before MBS) and to maintain or increase PA levels during phases B2 (after MBS) and A3 (at the 1-year follow-up). Owing to the goal of helping participants increase and maintain PA after MBS, participants are offered 3 booster sessions to discuss progress or relapses and work on PA psychosocial and cognitive processes already addressed during phase B1. These sessions are one-on-one sessions, occur at 1-month intervals, and are conducted via videoconferencing using Zoom. They last 45 minutes, though they can be shorter as the topics are not pre-established; rather, the participants are free to choose the topics according to their needs. Length and topics discussed are recorded by the PA counselor.

In addition, participants are encouraged to use the provided PA monitor (A370 Polar watch) for the study duration to track their PA behavior (not for research data collection), as this has been shown to be an effective way to increase goal setting, self-monitoring, and overall PA adherence [109,110].

PA counselors are persons with at least a bachelor’s degree in kinesiology or a related field (eg, nursing care and physiotherapy). They receive training and supervision from AB and JB. As part of their training, they are required to participate in a 2-hour training course facilitated by JB, AB, and Jenson Price on how to deliver the intervention. In addition, they are observed performing the mock counseling encounters and take part in a bracketing interview with JP to reflect on their own background, perceptions, and interests and to explore how these could impact them in their role. PA counselors also receive an intervention manual created by JB, AB, and JP and reviewed by PB, AJR, and DB, which details the intervention. Specifically, the manual provides session plans and suggested activities, as well as relevant worksheets and documents that are shared with participants via mail and email. Finally, notes for each session with duration, topics discussed, accomplishment of session objectives, participants’ reactions to the content, next steps with the participants, and reflection as a PA counselor (eg, problems arising during the sessions, overall experience delivering the session, suggestion for intervention or personal improvements, and session deviations) are taken by the PA counselor using a standardized form and discussed with AB, if needed. This information will be used to track fidelity to the intervention protocol and explore avenues for intervention improvement in future large-scale trials.

During the individual interviews, the participants are asked about all life events or factors that occurred during the study period (eg, divorce, injuries, illness, lockdown, and new employment) that may have impacted the study results. This strategy allows for greater certainty in confirming whether any changes observed are related to the intervention [111].

Descriptive statistics will be computed for baseline sociodemographic, clinical, feasibility, and acceptability data derived from the numerical responses and close-ended questions.

All statistical analyses will be carried out using R and the following packages: scan, SCRT [112], metafor, ggplot2, and tydiverse. For the primary outcomes of a future large-scale trial, a comparison of daily MVPA based on accelerometer data between phases A1 and B1 will be carried out using a randomization test for a single-case experimental design [113] for each participant. Subsequently, individual effect sizes will be aggregated in a meta-analysis to obtain a group-based effect size. This will provide the necessary information for computing the required sample size for a future large-scale trial. A sensitivity analysis will also be performed to compare A1 and B1 using a Tau-U test [114]. Then, a second meta-analysis will be performed using the same approach. The global effect of the intervention on daily MVPA will be tested using another set of randomization tests comparing the ABABA phases. This test makes it possible to compare 2 short time series and to provide an effect size in case of a significant difference. This analysis will be performed on an individual scale.

Generalization measures (secondary outcomes for a future large-scale trial: health-related quality of life, anxiety, and depression symptoms) will be analyzed using the Reliable Change Index or cutoff score for descriptive purposes; the differences in individual scores across phases will be compared [115].

Theoretical construct measures (secondary outcomes for a future large-scale trial: PA enjoyment, motivational regulations for PA, basic psychological needs satisfaction and frustration, self-efficacy to overcome PA barriers, and social support for exercise) and self-reported MVPA will be analyzed using visual analysis. As recommended [116], stability, mean level change, and range of data by phase will be examined.

Content analysis of participants’ responses to open-ended questions during the interviews and of the PA counselors’ notes completed after the session will be conducted to identify themes and see if or what changes to the protocol or methods or intervention are warranted [117]. This will involve five steps, which will be followed by 2 research staff members: (1) familiarize themselves with the data by reading each transcribed interview several times before coding and looking for patterns, (2) code the transcripts and compile the initial codes, (3) put together similar codes to form potential themes and subthemes, (4) review the main themes and subthemes in relation to the data, and (5) define and name themes, select illustrative quotes from the interviews and produce a report.

The study protocol was approved by the Institutional Review Boards at the CIUSSS du Saguenay–Lac-Saint-Jean (approval MP-25-2021-354; 2021-6-11), the Institut Universitaire de Cardiologie et de Pneumologie de Québec-Université Laval (approval MP-25-2021-354; 2021-8-2), and the CIUSSS du Nord-de-l’Île-de-Montréal (approval MEO-25-2022-2264; 2021-10-22).

Increasing MVPA uptake and maintenance in adults undergoing MBS is a public health priority. This study will provide critical evidence pertaining to the feasibility and acceptability of the TELE-BariACTIV aimed at promoting MVPA before and after MBS while addressing several barriers to MVPA and intervention engagement previously identified, as well as the protocol or methods proposed to evaluate it. The data collected in the individual interviews will help optimize the intervention and protocol or methods for a larger study. In addition, an estimate of the effect of the TELE-BariACTIV intervention on MVPA will be useful for calculating the required sample size for future multicenter large-scale trials.

This paper describes the protocol and intervention used in the TELE-BariACTIV study. To our knowledge, only 2 studies investigating PA behavior change interventions among adults undergoing MBS have been published [72,73]. The TELE-BariACTIV study extends these previous works by prolonging the duration of the intervention (to cover both before and after MBS) and accordingly integrates longer follow-up assessments to assess initial changes in PA and maintenance of PA. Furthermore, the intervention and protocol or methods were designed to be as accessible and simple as possible (ie, distance-based), with no need for in-person assessments or intervention sessions while still using an objective PA measure.

A single-case experimental study with multiple base levels (ABAB’A) design, in which a small group of participants is used to test causal relationships between variables of interest, is often used in behavior research because it allows rigorous experimental manipulation of independent variables and repeated measures of dependent variables over time [119]. This design has been used with various clinical populations to study different behaviors, including PA [87,120,121] in patients who have undergone MBS [122]. In addition, observational phases improve internal validity by controlling for maturation and historical variables, thus performing a function similar to that of the control group without intervention [75].

Although the TELE-BariACTIV intervention was developed according to patients’ preferences, some patients will like other options (eg, in-person intervention and supervised exercise training), and some will be excluded because they have no internet access or interface with a camera and need supportive devices to move, leading to potential enrollment difficulties and selection bias. The feasibility of the intervention in the general MBS population will not be assessed. However, this could be the goal of another larger study. In addition, recruitment began during the COVID-19 pandemic period. The COVID-19 pandemic has led to an overload on the health care system, which presumably slowed recruitment. Thus, the recruitment rate that will be estimated after the study is likely to be conservative. MBCTs are grouped together, so there is no way to know which technique has more or less impact on MVPA change. Nevertheless, the main aim of this initial study is to explore the effect of the intervention to calculate the sample size for a future large-scale trial and not to isolate the most efficacious intervention components. Finally, several outcome measures used are not specific to patients who have undergone MBS and may not be sensitive enough to capture slight changes that may be clinically meaningful for this population. Therefore, participants are offered an opportunity to share any additional thoughts they have, which can relate to any change in the outcomes measured during the interview.

To conclude, the TELE-BariACTIV study is innovative in its approach and design [76]. Feasibility and acceptability evidence would support proceeding with a larger trial that is sufficiently powered to test its effectiveness. Moreover, the TELE-BariACTIV intervention has strong potential for sustainability in the current context (pandemic, environmental, and budgetary restrictions). Indeed, given the ubiquity of the internet, TELE-BariACTIV intervention could be implemented across Canada to support cost-effective efforts to promote PA in this growing, mostly inactive, segment of the population. In addition, it would offer adults access to supportive care and trained professionals regardless of where they live. Finally, use of distance-based means to collect data to encourage participation of adults near and far from urban centers and use of technology to deliver the intervention to address participants’ lack of time and unwillingness to travel for in-person sessions are strengths of this study.