Introduction
Chronic obstructive pulmonary disease (COPD) patients engage in less daily physical activity (PA) than healthy individuals. Reduced PA levels are associated with clinical symptoms such as dyspnea and fatigue, which contribute to a decline in health-related quality of life, an increased risk of COPD-related hospitalization, and higher all-cause mortality [1,2]. Accumulated data indicate that promoting PA through pulmonary rehabilitation (PR) enhances exercise tolerance, improves quality of life, and delays disease progression [3-5]. However, challenges remain in optimizing PR, especially in customizing treatment plans for individual patients and ensuring their adherence [6,7].
While PA refers to the action itself, physical capacity (PC) represents the potential to perform that action. A dissociation between PA and PC is frequently observed in patients with COPD. Previous research has categorized COPD patients into quadrants, such as ‘can do, do do,’ based on their levels of PC and PA [8,9]. These studies have demonstrated that individuals in the ‘can do, don’t do’ and ‘can do, do do’ quadrants exhibit significantly lower mortality risk, underscoring the importance of PC. Although the prognostic significance of PC has been well-established, the impact of PA, particularly among those who are ‘can do’ but ‘don’t do,’ remains less clear. However, in real-world clinical settings, clinicians often struggle to understand, manage, and motivate ‘don’t do’ patients—those who exhibit low PA levels despite having adequate PC [10-12]. There is a pressing need for further research into strategies that enhance adherence to PR in this patient group.
Therefore, this study aimed to investigate the clinical characteristics of patients, including patient-reported outcomes (PROs), based on their PA, comparing those who ‘don’t do’ with those who ‘do do.’ By identifying the distinct features of these groups, we sought to develop more effective strategies for assessing and engaging patients with low PA levels in PR.
Materials and Methods
1. Study population and design
This study prospectively recruited patients aged 19 years or older with COPD who visited the respiratory clinic and were referred to the PR clinic at Samsung Medical Center between May 2022 and December 2023. COPD was defined as a post-bronchodilator forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) <70%, and patients with FEV1 <70% were included in the study. Participants were required to be ambulatory, independent in activities of daily living (Eastern Cooperative Oncology Group Performance Status [ECOG PS] <2), and capable of using a smartphone. Patients were excluded if they had a history of lung cancer surgery, severe bronchiectasis, severe tuberculous lung destruction, active tuberculosis, nontuberculous mycobacterial lung disease, or idiopathic pulmonary fibrosis. This study was conducted as a cross-sectional analysis using baseline data from the cohort, which included one month of continuous Fitbit activity monitoring (Fitbit Inc., San Francisco, CA, USA). Among the 105 patients who met these criteria, five withdrew consent, two were unable to use the Fitbit due to smartphone-related issues, and two did not complete the baseline measurements. A total of 96 patients were ultimately included in the analysis.
The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Samsung Medical Center (No. 2022-02-091). Written informed consent was obtained from all participants.
2. Measurements
Baseline patient characteristics, including age, sex, comorbidities, smoking status, and educational level, were collected. Smoking status was categorized as never, former, current smoker, or attempted cessation. Educational level was defined as having completed college or higher.
Measurements were taken using a bioelectrical impedance analysis device (Inbody S10, Inbody Co., Seoul, Korea) for height (cm), weight (kg), body mass index (BMI, kg/m²), muscle mass (kg), and skeletal muscle mass index (kg/m²). BMI was calculated in kg/m² and categorized as underweight, normal, overweight, or obese based on the cutoffs of 18.5, 23.0, and 25.0 [13]. A baseline 6-minute walk distance (6MWD, m) [14] and pulmonary function test (PFT) were conducted. Maximal inspiratory pressure (cmH₂O), maximal expiratory pressure (cmH₂O), and maximal phonation time (sec) were measured. Additionally, hand grip strength (kg), the five-time sit-to-stand test (STS, sec), and the 30-second STS (number of times/30 sec) were assessed.
Several PROs, defined as subjective self-reported data provided directly by patients about their health conditions, were collected. The severity of dyspnea experienced by patients was assessed using the modified Medical Research Council (mMRC) dyspnea scale and the COPD assessment test (CAT). The CAT consists of eight items—cough, sputum, chest tightness, dyspnea, activities, confidence, sleep, and energy—each rated on a scale from 0 to 5 [15].
The first four items were classified as ‘pulmonary items’ due to their direct relation to respiratory symptoms. In contrast, the remaining four items were categorized as ‘extra-pulmonary items’ to address broader health aspects. To assess the severity of depressive symptoms, we used the Patient Health Questionnaire-9 (PHQ-9), a widely recognized self-report tool.
Participants wore a Fitbit Charge 4 (Fitbit Inc.), a wearable accelerometer, continuously for 1 month after enrollment to monitor PA. The device recorded the number of steps taken and classified PA intensity and duration as either moderate or vigorous [16,17].
Non-wear time was defined as periods of 4 or more consecutive hours with zero step counts during daytime hours (9:00 AM to 4:00 PM), based on previously established methods [18]. The number of valid wear days per participant ranged from 1 to 11 (median, 7.2 days).
3. Group classification based on PA and PC
PA was assessed using Fitbit-derived moderate-to-vigorous physical activity (MVPA). A threshold of 30 minutes per day, corresponding to the cohort mean value, was applied. This cutoff was considered appropriate, as the American College of Sports Medicine recommends at least 30 minutes of moderate-intensity PA daily to maintain or improve fitness [19,20].
Accordingly, patients with MVPA ≥30 min/day were classified as the ‘Do do’ group, whereas those with MVPA <30 min/day were classified as the ‘Don’t do’ group.
PC was determined using the 6MWD. The mean 6MWD for the cohort was 455.7 m. Based on prior studies that use 450 m as a clinical benchmark [21], patients achieving a 6MWD of ≥450 m were classified as the ‘Can do’ group, while those with a 6MWD of <450 m were classified as the ‘Can’t do’ group (Supplementary Figure S1).
4. Statistical analysis
Mean and standard deviation or median and interquartile range were used to describe continuous variables. Chi-squared tests and Student’s t-tests were employed to compare categorical and continuous variables between the two groups, respectively.
To identify factors associated with low PA, we calculated odds ratios (ORs) with 95% confidence intervals (CIs) using univariable and multivariable logistic regression analysis, including the 6MWD, and FEV1% as covariates. A subgroup analysis was conducted for the ‘Can do’ group.
A p-value of less than 0.05 was considered significant for all analyses. Statistical analysis was performed using R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
1. Baseline characteristics
A total of 96 patients with a mean age of 68 years were included in the study, with the majority (93.8%) being male (Table 1). Among the 96 eligible patients, 44 were categorized as the ‘Do do’ group and the remaining 52 as the ‘Don’t do’ group. There were no significant differences between the two groups regarding BMI, muscle mass, skeletal muscle mass index, or socioeconomic factors such as smoking status and education level. However, patients in the ‘Do do’ group demonstrated a significantly longer 6MWD compared to the ‘Don’t do’ group (488 m vs. 424 m, p=0.005). Additionally, FEV1 and the FEV1/FVC ratio were higher in the ‘Do do’ group (54.48 vs. 46.73, p=0.005; 47.75 vs. 42.42, p=0.022, respectively). Furthermore, the mMRC dyspnea scale score was significantly lower in the ‘Do do’ group (1.30 vs. 1.77, p=0.019), indicating less severe dyspnea compared to the ‘Don’t do’ group. Although not statistically significant, a trend toward lower scores was observed in the ‘Do do’ group for both the CAT score and the PHQ-9 score, suggesting potentially improved health-related quality of life and fewer depressive symptoms in this group.
2. PROs: CAT score and PHQ9-score
To evaluate and quantify patients’ symptoms, we utilized the CAT score for patient-reported symptoms and the PHQ-9 score for psychological factors. The ‘Don’t do’ group reported significantly higher levels of breathlessness in the pulmonary items of the CAT score compared to the ‘Do do’ group (Figure 1 and Supplementary Table S1). Additionally, a trend toward higher levels of energy deficiency was observed in the ‘Don’t do’ group, although no statistically significant differences were found in the extra-pulmonary items of the CAT score. Furthermore, the PHQ-9 score indicated that the ‘Don’t do’ group experienced more difficulties with sleep (Figure 2 and Supplementary Table S2).
3. Factors associated with ‘Don’t do’ group
Table 2 presents the adjusted ORs for each component of the CAT and PHQ-9 scores associated with patients in the ‘Don’t do’ group. These results are adjusted for 6MWD (≤450 m vs. >450 m) and FEV1 (≤50% vs. >50%). Neither 6MWD (adjusted OR, 1.23; 95% CI, 0.99 to 1.52) nor FEV₁% (adjusted OR, 1.20; 95% CI, 0.96 to 1.49) showed significant associations with the ‘Don’t do’ group. In contrast, breathlessness as measured by the CAT score was significantly associated with the ‘Don’t do’ group (OR, 1.31; 95% CI, 1.06 to 1.62). Additionally, patients in the ‘Don’t do’ group were more likely to report higher levels of fatigue (OR, 1.24; 95% CI, 1.01 to 1.51). Furthermore, within the ‘Can do’ group, breathlessness as measured by the CAT score was also associated with the ‘Don’t do’ group (OR, 1.40; 95% CI, 1.08 to 1.82).
Discussion
Patients classified as ‘Don’t do’ represent a challenging group for clinicians, as they have the capacity to participate in PR but often lack the motivation to do so. This study aimed to stratify COPD patients into ‘Don’t do’ and ‘Do do’ groups based on their PA levels and to analyze the clinical characteristics and PROs of the ‘Don’t do’ group. As anticipated, the ‘Don’t do’ group exhibited reduced physical and pulmonary function. Additionally, analysis of PROs revealed that this group experienced greater dyspnea, a finding that remained significant even after adjusting for physical and pulmonary function and in subgroup analyses of patients with preserved PC (the ‘Can do’ group).
In our study, the ‘Don’t do’ group exhibited significantly lower physical and pulmonary function compared to the ‘Do do’ group, as represented by shorter 6MWD and decreased pulmonary function parameters (FEV1, FEV1/FVC). These results align with previous research, which indicates that PC and lung function are key determinants of PA levels in COPD patients. Previous studies have demonstrated that patients with better PC ultimately show improvements in PA following PR, highlighting the importance of PC [22,23]. However, in our study, we found that the ‘Don’t do’ group experienced significantly higher levels of breathlessness, even after adjusting for the 6MWD and PFT. Notably, this trend continued when we further divided the ‘Can do’ group, the patients with higher PC, into ‘Do do’ and ‘Don’t do’ subgroups, with breathlessness remaining a significant barrier to PA. This indicates that breathlessness is not solely a consequence of reduced PC but may also be influenced by other factors such as deconditioning, psychological distress, and heightened symptom perception. These findings underscore breathlessness as a crucial target for intervention, particularly in patients with low PA levels, regardless of their baseline PC. Although traditional measures like PC (e.g., 6MWD) and PFT (e.g., FEV1, FVC) provide valuable insights, they fail to interpret the subjective experiences that significantly impact a patient’s ability to engage in PA. Symptoms such as breathlessness and fatigue, captured through PROs like the CAT and the PHQ-9, may be the key drivers of low PA levels. Addressing these PRO-related barriers is essential for optimizing PR engagement and outcomes.
Our study highlights the relationship between PROs and PA, underscoring the importance of considering various factors for objective measures. While previous research has primarily focused on PC and pulmonary function as determinants of PA, our findings indicate that subjective experiences such as breathlessness and fatigue serve as significant barriers to PA, even after adjusting for objective measures. This suggests that PROs offer unique and valuable insights into patients’ activity levels. The significance of PROs lies in their ability to provide a comprehensive understanding of a patient’s condition by assessing aspects beyond measurable physical parameters [24,25]. PROs also capture the patient’s subjective experiences, including symptoms, quality of life, and overall well-being. Previous research suggests that while PROs and exercise test outcomes are significantly correlated, they assess different aspects of COPD. Therefore, supplementing PROs with functional measurements is necessary for a more comprehensive evaluation of PR program effectiveness [26]. Additionally, the relationship between PROs for dyspnea and the 6MWD has yielded inconsistent findings in previous research [27]. Some studies have reported moderate to strong positive correlations, suggesting that improvements in exercise capacity are associated with reductions in dyspnea. In contrast, other studies have found negative correlations or no significant association. These variations underscore the complex interplay between objective performance measures and subjective symptoms, indicating that one cannot always reliably predict the other. Furthermore, it is well-documented that even when PR successfully enhances exercise performance, this does not always lead to a corresponding decrease in dyspnea [27]. This discrepancy highlights the need for separate monitoring and management of exercise capacity and symptoms like dyspnea. Recognizing and addressing these relationships can optimize patient care and ensure that both physical and subjective outcomes are adequately considered.
Our results have several clinical implications for PR design and patient engagement strategies. While current PR programs primarily rely on objective PC parameters as primary assessment measures, our findings suggest that this approach may be insufficient. PROs can provide additional insights into patients who might be reluctant to participate, and healthcare providers should systematically evaluate these outcomes to gain a more comprehensive understanding of patient needs. Moreover, the significant association between PROs and PA levels indicates that rehabilitation programs should incorporate systematic assessment of PROs to better understand the subjective barriers patients face. Adopting a multidimensional approach using tools such as the CAT, mMRC, and PHQ-9 may facilitate a more holistic understanding of patients’ physical, psychological, and symptom-related challenges, enabling tailored interventions. This approach can support the development of individualized strategies that address both objective physical limitations and subjective symptoms.
There are several limitations to this study. A key limitation is the conceptual and methodological overlap between PA and PC. Although we attempted to account for this overlap in our analysis, fully disentangling the influence of PC on PA levels is challenging, particularly in ‘Don’t do’ patients who inherently exhibit lower PA. This underscores the importance of interpreting PC and PA as interdependent constructs within a comprehensive clinical framework, rather than as isolated parameters. Additionally, the relatively small sample size restricts the generalizability of our findings. While large-scale studies in the field of PR are inherently challenging, future multicenter and longitudinal studies will be essential to provide more robust evidence. Furthermore, we did not use energy expenditure (EE) to quantify PA. Although EE is a commonly used parameter to assess PA, we chose not to include it in our analysis because algorithm-based EE estimations from wearable devices are known to have limited accuracy and reliability for clinical research purposes [28]. We focused on MVPA, which is directly supported and recommended by the World Health Organization PA guidelines. Although this was a prospective study, the ‘Do do’ and ‘Don’t do’ groups were based on objectively observed PA behaviors rather than predetermined or randomized assignments. Therefore, baseline differences in PC and pulmonary function naturally existed between the groups. We addressed these factors through multivariable adjustment; however, the possibility of residual confounding remains. We did not apply alternative balancing methods, as such approaches could distort the real-world behavioral heterogeneity that the study aimed to capture, although this design choice may limit causal interpretation. Future studies with larger cohorts and causal inference methods could further validate our findings.
In conclusion, this study underscores the significance of PROs in understanding PA levels in COPD patients. The observed reductions in PA are not solely due to impaired PC but are also influenced by PROs such as breathlessness and fatigue. Incorporating PROs alongside traditional physiological measures may enable a more comprehensive understanding of patients and contribute to the development of more personalized PR programs.
PDF Links
PubReader
ePub Link
XML Download
Data Sharing Statement
Full text via DOI
Supplement1
Print
Download Citation
