This study was approved by the ethics committee of the institution involved, that is, the Local Health Unit of the Leiria Region (reference 10/2024, acta number 022024/01/16). Participant data were handled in accordance with applicable data protection regulations. Patients who agreed to participate in the study provided written informed consent before any data collection.

This was a single-group feasibility and acceptability study that investigated a 3-week PA coaching intervention implemented during a PR program. This study is part of the Effectiveness of Physical Activity Coaching using eHealth in patients with COPD (Physical Activity in COPD Using eHealth) study, which aims to assess the effectiveness of a personalized eHealth coaching intervention integrated into PR to increase the PA levels in patients with COPD. This feasibility study was reported in accordance with guidelines for reporting pilot and feasibility studies (adjusted for a single-arm study) [14-16]. Participant data were handled in accordance with applicable data protection regulations. The focus of this study was on the feasibility and acceptability of the intervention, specifically to test the PA goal-setting algorithm and assess the receptivity and acceptance of the intervention, including the digital platform, to identify any necessary adjustments for its implementation in a larger study. For this purpose, 3 weeks were needed for patients to interact with the platform, test its acceptability, and refine the algorithm.

Potential participants were recruited from the Local Health Unit of the Leiria Region (Portugal). The participant eligibility criteria are shown in Textbox 1.

The physician responsible for the PR program identified the eligible participants and briefly explained the study to them. Those interested in participating were then contacted by the researcher via telephone to inform them about the study procedures (data collection, dates, and intervention delivery), confirm their eligibility, and ask about their willingness to participate. Reasons for not participating were recorded. Patients who agreed to participate provided written informed consent before any data collection.

This study included 4 interaction time points with patients, as described in Figure 1. The assessments took place at baseline (T1) and 3 weeks after, corresponding to the end of the intervention (T4). In the baseline assessment (T1), participants completed a questionnaire with sociodemographic and health-related information (ie, age, sex, educational level, occupational status, smoking history, comorbidities using the Charlson Comorbidity Index [17], exacerbation history, medication, weight, and height), preferred physical activities (eg, walking, swimming, and dancing), and questions about the type and features or apps used on their smartphone. Lung function was assessed through spirometry [18] to further classify patients based on Global Initiative for Chronic Obstructive Lung Disease (GOLD) grades of airflow obstruction severity (GOLD 1: forced expiratory volume at 1 second [FEV₁]≥80% predicted, GOLD 2: 50%≤FEV₁<80% predicted, GOLD 3: 30%≤FEV₁<50% predicted, and GOLD 4: FEV₁<30% predicted) [1]. GOLD A, B, and E categories were determined according to the patients’ exacerbation history and symptoms [1]. The impact of COPD on daily living, dyspnea, fatigue, exercise tolerance, quality of life, and self-efficacy for PA was also assessed, as described in detail in the Secondary Outcome Measures (Health-Related Outcomes) section.

Then, patients received the PA mobile app paired with a smart band to use during the study period. In the first week, their baseline step count was established for goal setting in the following weeks. In the second time point (T2—end of the first week), patients received education on the importance of PA and recommendations and agreed on the first PA goal set. In the third time point (T3—end of the second week), the researcher reviewed the achievement of the previous PA goal, explored barriers to PA behavior and potential adverse events, and set the next goal with the patient through a telephone call. One week later (T4—end of the third week), patients returned the app and the smart band and were reassessed as in T1. A focus group with patients was also conducted to assess their feedback on the PA coaching intervention, including their experience with new technology; perspectives on the importance of PA promotion interventions; app functionality; and acceptability, barriers, and facilitators of its use. The intervention is described in detail below (refer to the PA Coaching Intervention section).

Descriptive statistics were used to characterize the sample using SPSS Statistics (version 28.0; IBM Corp). Continuous variables were summarized as mean (SD) when normally distributed and as median (IQR) when normality assumptions were not met. Categorical variables were described using absolute frequencies (n) and percentages (%).

Data collected through the focus group were analyzed using a qualitative methodological approach, namely, thematic analysis [34]. Participants were anonymized using alphanumeric codes (P1, P2, etc) during data analysis. Themes and categories were developed in an Excel spreadsheet, enabling the identification of the main topics discussed by the participants. The analysis was primarily conducted independently by 2 researchers. Subsequently, both researchers met to systematically compare codes and themes and to reach consensus. Themes were generated and analyzed question by question, following the focus group guide. Once no new categories were identified, the information was organized into a descriptive framework [34].

The interviews were conducted by 2 female physiotherapists with a master’s degree in the respiratory field and experience in PA assessment and promotion. Both researchers made efforts to ensure neutrality throughout the analysis process, complying with established qualitative research rigor guidelines. The research team also included a senior member with experience in COPD management, PA promotion, and qualitative research.

Patients had a mean age of 67.0 (SD 8.9) years and a mean predicted FEV₁ of 49% (SD 22.8%), indicating a population with moderate to severe airflow limitation. The sample was predominantly female (n=4, 80%), with most patients classified as GOLD E (n=4, 80%), suggesting a high risk of exacerbations and symptom burden. Regarding educational level, 60% (n=3) completed 10 to 12 years of formal education, and all patients were retired (n=5, 100%). The mean COPD Assessment Test score was 13.0 (SD 9.0). Fatigue levels, assessed using the Portuguese version of the Checklist of Individual Strength, showed a mean score of 78.4(SD 28.9). Dyspnea severity, measured by the modified Medical Research Council Dyspnea Scale, had a median score of 3.0 (IQR 0.5-3.5). On average, each assessment session lasted approximately 40 minutes. Characteristics of participants are presented in Table 1.

Regarding the experience with new technologies, all patients used smartphones, performing various tasks on them, such as browsing the internet and listening to music (n=5, 100%); sending or receiving SMS text messages (n=4, 80%); taking photos, playing games, and using social media (n=3, 60%); and using GPS (n=2, 40%).

The recruitment rate was 83% (5/6) with 100% retention and no dropouts.

During the intervention, no adverse events or unplanned health care use were reported. Only 1 technical issue with the app was reported, involving difficulties connecting to the app; the researcher restarted the app, and the problem was resolved. Adherence was 100% (5/5), with patients syncing the mobile app daily. The call for goal progression (T2) lasted an average of 10 (SD 3.13) minutes. The summary of goal progression and diary of telephone calls (to report technical issues from the app, goal setting, and PA promotion) was recorded, and it can be found in Multimedia Appendix 2.

1. The focus group lasted 1 hour and 6 minutes, and the several themes were identified.

Value of the PA coaching intervention using eHealth and its impact on PA levels—all patients considered this type of intervention an added value, both for themselves to increase their perception of PA and for health care professionals (HCPs) to help them monitor patients’ PA levels:

2. Barriers and facilitators of the use of Fitrockr Hub App to promote PA—the focus only on step counts to define PA goals, the lack of personalization (excluding PA activities other than walking), occasionally step miscounts (through the smart band), and not having the app for free in the real context were considered barriers. Participants identified a user-friendly layout, acceptable time demands, and the prescription of PA goals as facilitators:

3. Goal setting, progression, PA monitoring, and promotion—patients considered goal setting to be a collaborative and personalized process, highlighting the importance of negotiated PA goals with their HCPs. Patients considered the adjustment of their PA goals (based on performance and willingness to increase) adequate and emphasized the identification of barriers to PA and strategies to overcome them every week as being beneficial for PA engagement. The main barriers to not reaching the PA goals by patients were bad weather, exacerbations or worsening of symptoms, and variability in response to PA. The importance of maintaining contact with HCPs during the intervention, either via video call or telephone, was also highlighted by all patients:

Overall, this PA coaching intervention using eHealth appears to be feasible and well accepted by patients. Patients’ feedback about the intervention was positive, as they considered the app easy to use and valuable for PA self-management and remote monitoring by HCPs. Patients underlined the importance of weekly goal adjustments, frequent app notifications, and weekly contact with HCPs through telephone or video calls. However, patients identified some limitations regarding the use of the app, including a lack of personalization, the exclusive focus on step counts to define PA goals, and occasional step miscounts.

Despite a shorter intervention period compared with similar eHealth-based PA interventions in COPD [35,36], this study achieved high recruitment and retention rates (>80%) supporting the feasibility of this study design [15]. These rates are comparable to or higher than those reported in previous feasibility studies, suggesting that integrating PA coaching alongside PR is practicable within this clinical context. However, these findings should be interpreted with caution due to the small and relatively homogeneous sample, which was predominantly female, retired, and familiar with smartphone use. Therefore, the generalizability of these results is limited. Consistent with previous literature, future studies should recruit more diverse samples, including individuals with lower digital literacy [37], varying levels of motivation for PA [38], and higher multimorbidity burden [39], as these factors are known to influence engagement with eHealth interventions.

Regarding intervention feasibility, no major technical or PA barriers were identified during implementation. This contrasts with previous reports highlighting poor digital literacy and technical challenges as key barriers to eHealth adoption in COPD [40] and may be partly explained by the short intervention duration and the provision of technical support. Over time, factors such as comorbidities, environmental challenges, and psychological barriers (eg, fear of breathlessness) can become more pronounced, potentially impairing adherence to the intervention [41].

Regarding the feasibility of PA goal-setting algorithm, although participants did not consistently achieve their weekly goals, they reported that the algorithm was appropriate, with a period of revision and incremental increases being acceptable. These findings align with previous literature recommending weekly goal adjustments, with step count increases ranging from 5% to 15%, tailored to individual tolerance and adherence levels [42-44].

The intervention was proven to be acceptable—patients demonstrated full adherence by synchronizing the smart band with the app daily, regardless of goal achievement. Most patients found the PA app easy to use and well integrated into daily life, supporting evidence that the simplicity and usability of technology can be more decisive than personalization for engagement, as technical issues often lead to participant’s dropout [35]. Patients also highly valued the weekly contact with HCPs, which aligns with evidence highlighting patient-professional interaction as a key facilitator of PA promotion [45], particularly in eHealth interventions [40]. Consistent with the literature, personalized goal setting, tailored to individual needs, and feedback notifications were identified as essential components of these interventions [46,47]. However, important limitations were identified, including the exclusive use of step-based goals and limited app personalization, which may reduce patients’ motivation, engagement with the intervention, and sustained behavior change. While the eHealth platform could not be modified, as it was already developed and PA coaching may not have been its primary focus, future studies should consider greater personalization to better meet patients’ needs. Patients reported PA barriers consistent with the literature, particularly health-related issues (eg, symptom exacerbations), individual variability, and environmental factors such as weather [48]. Addressing these barriers is essential for developing effective and individualized PA interventions. In this study, several evidence-based behavioral strategies, such as goal setting, feedback and monitoring, repetition, substitution, and self-belief, were applied and have since been shown to improve PA in individuals with COPD by supporting barrier identification and problem-solving [49-51].

The findings of this feasibility study suggest that the goal-setting algorithm is appropriate for patients with COPD. Weekly contact between HCPs and patients to establish PA goals, identify potential barriers and corresponding strategies to overcome them, and provide feedback on performance emerged as essential components to retain in future larger-scale studies.

Although PA coaching using eHealth may represent a promising approach for self-managing COPD, the findings of this study should be interpreted with caution due to the small sample size and the predominance of participants with prior digital literacy. This limits the generalizability of the results, as previous literature has indicated that individuals with greater interest and experience in technology are more likely to benefit from eHealth interventions [52]. Given the short duration of the intervention, significant changes in PA behavior were not expected. Instead, collecting health-related outcome measures at baseline and the end of the study primarily served to assess the feasibility of data collection procedures and inform refinements for future large-scale trials.

This feasibility and acceptability study showed that the PA coaching intervention using eHealth is feasible and well accepted by the patients. However, some barriers still need to be addressed in future studies to ensure adherence to the intervention. Regarding feasibility, while the remote monitoring procedures were successful, future research must prioritize the robustness of the digital infrastructure to eliminate technical errors and data discrepancies, which can undermine clinical reliability. To ensure acceptability and long-term adherence, it is critical that future interventions move toward a more comprehensive model of personalized coaching. This approach should be tailored to each patient’s unique clinical characteristics and PA goals.