Based on our experience with mind–body interventions and a review of the literature on facilitators and barriers of physical activity in COPD and HF, we developed an initial conceptual framework (Figure 1) that combined the existing Moy and Richardson intervention with elements of a mind–body exercise program to impact long-term behavior change.

We chose self-efficacy as a key outcome of interest. According to social cognitive theory, self-efficacy is one’s confidence in the ability to perform certain health behaviors and can influence the engagement in and actual performance of those behaviors, affecting health outcomes thereby [80]. Self-efficacy is a modifiable characteristic, and exercise self-efficacy is one of the strongest predictors of physical activity behavior [81]. Within our conceptual model, there are multiple characteristics of mind–body exercise that reinforce and enhance constructs of self-regulation from the existing intervention and additionally provide new constructs and potential mediating factors toward self-efficacy [82-84].

Experience/goal attainment, based on traditional self-efficacy theory [80], refers to the notion that prior successes boost confidence and facilitate self-efficacy. For example, mind–body exercise is often described as gentle, accessible, and nonthreatening, which may allow participants to achieve goals, even if in small increments. Similarly, small individual successes in iterative step-count goals further enhance that model with repeated success. Modeling describes when participants gain from seeing others succeed and modeling others’ behaviors. Prior qualitative data, including our studies of tai chi and HF, have reinforced the value of seeing others, perhaps even in a poorer condition than oneself, achieve success, which feeds confidence and motivation for one’s own abilities [82,83]. Self-awareness and mindfulness of physiologic cues and emotional states are inherently cultivated through mind–body exercise and are at the foundation of mind–body practices. Body signals such as fast heart rate or breathing due to anxiety or fear can negatively impact one’s perception of a particular task, such as exercise or walking. This may be illustrated by the anxiety–breathlessness cycle in COPD [85]. Strategies for mindfulness, body- and self-awareness, and concomitant acceptance and nonreactivity can mitigate this effect. Mind–body exercises can also address pain cycles and related exercise barriers [86-89] and impact fear of falling, which is correlated with self-efficacy and decreased falls [90,91]. Intrinsic motivation refers to motivation that is driven, not because of external factors (eg, a reward through an intervention website, feedback, or praise) but by internal factors, such as personal satisfaction and enjoyment. The inherent likeability of mind–body therapies is not well studied; however, numerous studies have reported qualitatively on enjoyment of mind–body exercise [82,84,88,92]. Certainly, increased mood with physical activity is well established. Finally, each of these potentially mediating factors are held within a larger domain of self-reflection and transformation that is a part of self-determination theory [93]. There is a growing literature that describes personal transformative experiences that can occur with mind–body practices [44–46].

In the modified Mindful Steps intervention, we retain the original constructs of goal setting, iterative feedback, extrinsic motivation, and social support that have appeared effective for short-term adherence and additionally introduce constructs that facilitate self-reflection and personal transformation through heightened, mindful, self-awareness and intrinsic motivation that can impact overall self-efficacy and facilitate long-term physical activity behavior [80,93,94].

We assembled an expert panel to inform the creation and integration of a mind–body curriculum within the existing web-based platform. The panel included those with clinical and research expertise in mind–body exercise and tai chi, chronic cardiopulmonary disease (COPD and HF specialists), behavioral psychology, exercise physiology, and physical activity. We first began with open feedback from the core investigator team to develop the initial proposal for the curriculum. Using a modified Delphi process [95,96], we then solicited input from the larger expert panel on issues from practical implementation (eg, optimum ratio of and integration of in-person class contact with online video to decrease participant burden, modifications for safety, use of technology) to content-specific recommendations (eg, essential components of the mind–body curriculum, how best to integrate it with existing content, modifying disease-specific educational content for HF and already existing information for COPD). After a first round expert panel group meeting, individual interviews took place in an iterative fashion to gather further input and refine the protocol. After 2 subsequent group meetings with the expert panel, we arrived at a consensus protocol outlined in the following sections.

The final Mindful Steps Intervention to be pilot-tested contains 7 primary components (Table 1). Each of the components is described below, with specific information on how it was modified from the prior Moy and Richardson intervention.

The inclusion criteria include the following: age over >40 years; clinical diagnosis of COPD, defined as either a ratio of forced expiratory volume in one second (FEV₁) to forced vital capacity <0.70 or chest computed tomography evidence of emphysema, and/or clinical diagnosis of HF syndrome (with left ventricular systolic dysfunction or preserved ejection fraction, and New York Heart Association Class 1-3); medical clearance from a provider to participate in an exercise program; an active email account with the ability to check email at least weekly; and access to a computer with an internet connection, USB port, and Windows. Meanwhile, the exclusion criteria include the following: self-reported COPD or HF exacerbation in the previous 2 weeks; inability to ambulate; clinical signs of unstable cardiovascular disease; hypoxemia during 6-minute walk test (oxygen saturation <85% with supplemental oxygen); inability to collect at least 7 of 14 days of baseline step counts; current participation in a cardiac or pulmonary rehabilitation program.

To identify potential participants, we searched hospital registries, screened physician schedules, solicited direct physician referrals through email, and used targeted patient advertising (paper flyers, digital screen ads, in-person presentations) in primary care or specialty clinics at Beth Israel Deaconess Medical Center and pulmonary rehabilitation clinics in the community. This was followed by opt-in mailing and telephone outreach conducted by our research assistants.

As in prior studies, we used a 14-day run-in period of pedometer (Fitbit) use requiring at least 7 days of valid data (>200 steps per day) as a criterion for participants to be eligible for randomization. As the sample size is relatively small, we will use blocked randomization with varying block sizes and a computer-generated sequence of random numbers with a concealed and unpredictable allocation scheme.

Participants in the Mindful Steps intervention will receive the pedometer, access to the Mindful Steps website, and a schedule of in-person mind–body exercise classes as described in detail previously in this paper. All participants will continue pharmacological treatment and receive care through their usual providers. Participants in both groups will also receive an education handout from Harvard Health Publishing and Harvard Medical School titled “Walking for Health,” which includes information on the health benefits of walking, how to get started on a walking program, and specific walking workouts [104].

Table 2 outlines study feasibility and intervention acceptability and adherence measures that are being tracked to inform a future larger trial.

In addition, all participants will undergo testing to collect exploratory outcome measures at baseline, month 3 and 6 in-person, and month 9 and month 12 by phone. Table 3 outlines exploratory patient-centered outcomes including cognitive-behavioral and psychosocial measures, overall physical activity, disease-specific HRQL and symptoms, and physical function and exercise capacity.

At each time point, we will administer questionnaires and collect physical activity data. At in-person visits only, we will additionally perform the 6-minute walk test. All tests will be conducted by study staff who will be blinded to treatment assignment. We will also track potential adverse events and COPD- and HF-related exacerbations and hospitalizations at each testing visit to assess safety.

Qualitative interviews at 6 months and 12 months will further assess facilitators and barriers to participation and acceptability of the intervention and its multiple components. Additional questions will be guided by our conceptual model and explore themes of motivation, goal attainment, social support, self-reflection, and personal transformation.

For study feasibility and intervention adherence, we will use descriptive statistics to evaluate recruitment and retention rates, attendance at mind–body exercise classes, and adherence with the online intervention (web-platform usage). We will consider study retention successful if the retention rate is at least 80% and if attendance at in-person classes is at least 70% (17/25 classes). Mean scores on the Physical Activity Enjoyment Scale at each time point will be reported [105].

To further evaluate intervention acceptability, 6-month and 12-month qualitative interviews will be recorded and transcribed verbatim. Thematic analysis will be informed by grounded theory methods based on our semistructured questions. Transcripts will be independently coded by at least 2 separate investigators to identify emergent themes in an iterative process until thematic saturation is reached. We will particularly search for themes that may give insights into factors mediating a shift in self-efficacy towards longer-term maintenance of walking and overall physical activity.

For each patient-centered outcome, repeated measures analyses will be performed using SAS PROC MIXED to evaluate trajectories over the course of the 4 time points for each group and the differences between the groups in these trajectories at each time point. Planned paired comparisons and Cohen’s d effect sizes will also be computed to evaluate the magnitude of improvement between time points and group differences in change. These data will provide valuable information including what outcome domains and measures may demonstrate clinically meaningful changes in future models, the appropriate number of time points to use in a future clinical trial, and measures of variability within and between groups to inform the design of future studies. Measures considered sensitive to change will be those that indicate trends in improvement over time.