Dataset

The study samples were formed by COPD patients that visited 8 health care facilities in cities of Hangzhou and Quzhou, Zhejiang province, including 3 hospitals and 5 primary health centers, from July 1, 2020 to October 30, 2020. Based on the evidence from the 2018 national cross-sectional COPD study in China, it was found that the prevalence of COPD among Chinese adults was significantly higher in individuals over 40 years old (13.7%) compared to those aged 20–39 (2.1%) [4]. In light of this, our study focused primarily on recruiting COPD patients aged 40 years and above to ensure the representativeness of the population, as early PR is recommended for all COPD patients following diagnosis [52, 53]. Detailed inclusion and exclusion criteria were summarized in Fig 1. Moreover, we recognized that the needs and perceived barriers of PR interventions may vary among patients with different ages or COPD severity levels. Therefore, we later conducted series of subgroup analyses to account for these potential differences. A total of 237 COPD patients were investigated by convenience sampling in this study.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow diagram of study.

https://doi.org/10.1371/journal.pone.0290828.g001

The study was conducted according to the guidelines of the Declaration of Helsinki, and was approved by the institutional review board of Hangzhou Normal University (No. 2020115). This study was guided by respiratory physicians and nurses from the investigated health care facilities. All participants were provided with a clear explanation of the study’s objectives and were assured that their data would be kept anonymous, confidential, and solely used for scientific purposes. Before the data collection, verbal informed consent was obtained from each participant.

Questionnaire design

The questionnaire for this study was designed in two sections. The first section consisted of a self-designed survey on patients’ intention to receive various PR interventions and its potentially relevant factors, covering 3 dimensions: personal, social, and environmental. The personal dimension further included 4 aspects: demographic factors, COPD status, personal awareness of PR, and health-related behaviors. The social dimension mainly involved PR-related social policies such as types of medical insurance enrollment, medical insurance reimbursement and the cost of PR services, while the environmental factors specifically encompassed the PR skills of medical staff, quality of PR provided by medical facilities, knowledge promotion or public education of PR in the community, transportation concerns and degree of support from family and friends. This section integrated several mature scales, such as the modified Medical Research Council dyspnea scale (mMRC) [33] and the social support rating scale (SSRS) [37]. These integrated scales have been all widely used and have demonstrated good reliability and validity among Chinese people. The second section was a demand survey of PR intervention services for COPD patients based on the Kano model. An extensive literature review and relevant guideline reference were conducted [54], and a total of 18 commonly defined PR services were included [8, 10, 11] (S1 File). The reliability of questionnaire was assessed using Cronbach’s α, which yielded a value of 0.800 for the intention and influencing factors questionnaire, 0.831 for functional questions and 0.826 for dysfunctional questions in demand questionnaire. As for validity, the Kaiser-Meyer-Olkin (KMO) measure was employed, yielding values of 0.677, 0.726 and 0.709, respectively. Both the validity and reliability of the questionnaire were acceptable.

Evaluation strategies according to Kano model

As shown in Table 1, following the steps of the Kano model, 18 attributes of PR services were transferred into a set of functional and dysfunctional questions [54]. Answer options for both types of questions were “Like it”, “Must-be”, “Neutral”, “Accept it” and “Dislike”. There were 5×5 possible results, each of which corresponded to a Kano category [46, 48]. M represented must-be attributes, O represented one-dimensional attributes, A represented attractive attributes, I represented indifferent attributes, R represented reverse attributes and Q represented questionable answers. Subsequently, the frequency of each assigned category was calculated, and the final category was determined to be the one reaching the highest frequency [55]. In detail, the implications of these 6 categories are shown in S2 File. Notably, must-be qualities(M) are the most important qualities and should be provided first, followed by one-dimensional qualities(O) and attractive qualities(A) whereas reverse qualities(R) should be removed from the service list [54, 56].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Kano evaluation table.

https://doi.org/10.1371/journal.pone.0290828.t001

The coefficient of patient satisfaction shows how strongly a service influences satisfaction or dissatisfaction, which helps medical staff prioritize patient demands [45]. The coefficient consists of better (satisfaction coefficient, SI) and worse (dissatisfaction coefficient, DSI) values, and the absolute value of the coefficient (ranged from 0 to 1) reflects the level of satisfaction or dissatisfaction with the presence or absence of a service. Specifically, the closer the SI value was to 1, the more satisfied patients were when the target attribute is provided. And the closer the absolute value of DSI was to 1, the more dissatisfied patients were when the target attribute is not provided. If the absolute value is greater than 0.5, an attribute is assumed to be important. The coefficient is calculated as follows [57, 58]:

After that, the matrix graph is plotted with worse (DSI) absolute value as x-axis and better (SI) value as y-axis and divided into 4 areas. Concretely, the attributes located in the first quadrant (upper right area), with high absolute values of both SI and DSI, deserve more attention and should be offered with priority to improve satisfaction and reduce dissatisfaction. The attributes located in the second quadrant (upper left area), with higher SI values but lower DSI absolute values, would have a greater impact on satisfaction but less on dissatisfaction. The attributes located in the fourth quadrant (lower right area), with lower SI values but higher DSI absolute values, would have a greater impact on patient dissatisfaction but less on patient satisfaction. The attributes located in the third quadrant (lower left area), with lower absolute values of both SI and DSI, show that they are not important and do not need to improve [56].

Statistics analysis

Epidata was used for survey data entry, check and error correction. All statistical analysis was carried out using SPSS 22.0 and R software v.4.2.0. Firstly, univariant comparison was conducted in 237 COPD patients. Patients’ intention to receive PR services was treated as the outcome variable, and 23 factors were used as independent variables (S3 File). Quantitative variables were showed as mean ± sd, and categorical variables were showed as n (%). The differences among groups were analyzed by ANOVA (for quantitative variables) and the chi-square/fisher test (for categorical variables). After that, statistically significant variables in the univariant analysis were delivered for the following multiple linear regression analysis, where the patients’ intention was treated as a continuous variable (i.e., coded as 1, 2, 3, 4, and 5 for very low, low, moderate, high, and very high intention, respectively). P < 0.05 was considered statistically significant.

Subjects’ characteristics and their awareness of PR interventions

Among 237 COPD patients, 156 were residents of Hangzhou, and 81 resided in Quzhou. The male-to-female proportion was 78.5% and 21.5%, respectively. Regarding age distribution, the participants’ ages ranged from 45 to 97, with the majority being over 60 years old (92.5%). The majority (78.5%) achieved educational levels of primary school or junior secondary school, and more than half (51.5%) of the patients had monthly household income between 4000 and 6999 RMB. About 60% investigated patients were in the stage of stable COPD. When using mMRC dyspnea scale to measure the severity of breathlessness, 59.5% of the patients had the mMRC grade of 2 (i.e., walks slower than people of the same age on level ground because of dyspnea or has to stop for breath when walking at own pace). More details could be found in S3 File.

When investigating the overall awareness of PR, 62.9% of the investigated COPD patients were aware of PR interventions (S3 File), mainly informed by medical staff in hospitals (62.4%), followed by primary health centers (12.2%), as shown in Fig 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Sources of awareness on PR interventions in COPD patients.

https://doi.org/10.1371/journal.pone.0290828.g002

Analysis of PR intervention intention in COPD patients

When investigating the intention to receive PR interventions, 75.1% of COPD patients reported that they were willing to accept services related to PR interventions. From the patients’ self-reported perspective, it showed that “the cost of PR services” (96.2%), “the ratio of medical insurance reimbursement in PR services” (96.2%), and “the PR skills of medical staff” (95.4%) were recognized as the most important self-reported factors that may influence patients’ intention to accept PR interventions (S3 File).

To comprehensively and objectively explore factors influencing COPD patients’ intention to receive PR services, we also conducted the intention analysis from 3 dimensions: patients’ personal factors, social policies and perceived environmental barriers. The intention to receive PR services was treated as the outcome variable, and factors from the above 3 dimensions were used as independent variables. During the univariate analysis (S3 File), 14 variables were recognized as significant factors associated with the intention, including patients’ personal awareness of the PR interventions, 4 demographic factors (i.e., region, gender, educational level and monthly household income), 4 health-related behaviors (i.e., the drinking habit, emotion management, healthy diet and exercise) and 5 perceived environmental barriers (i.e., PR skills of medical staff, quality of PR services provided by medical facilities, lack of knowledge promotion and public education of PR in the community, transportation concerns and degree of support from family and friends).

Subsequently, the 14 significant factors screened out from the univariate analysis were further introduced into the multiple linear regression analysis. We checked the assumptions of normality, multicollinearity, and autocorrelation before performing multiple linear regression analysis on our data. The residuals were normally distributed, as shown by the Shapiro-Wilk test and S4 File. The Durbin-Watson statistic was 1.824, indicating no autocorrelation. The tolerance values ranged from 0.525 to 0.898, and the VIF values ranged from 1.114 to 1.904, indicating no multicollinearity. These results confirmed that multiple linear regression was appropriate for our data. Ultimately, 6 variables were survived in this step. Specifically, higher PR awareness of patients and better PR skills of medical staff would increase patients’ intention. Patients who had a healthier lifestyle (i.e., no drinking) and received more support from family and friends would have stronger intention. On the contrary, transportation inconvenience and lack of knowledge promotion or public education in the community would reduce the patients’ intention to receive PR interventions (Table 2). Also, multiclass ordinal logistic regression analysis was adopted for comparison. As a result, the significant influencing factors survived in the multiclass ordinal logistic regression analysis were all consistent with those of the multiple linear regression analysis, validating and reinforcing our findings (S5 File).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Multiple linear regression analysis of COPD patients’ intention to accept PR services.

https://doi.org/10.1371/journal.pone.0290828.t002

Considering potential difference in disease progression and symptom severity, we classified patients into two groups: those with less breathlessness (mMRC grade:0–1) and those more breathlessness (mMRC grade:2–4). The intention to receive PR services, as well as personal, social and environmental influencing factors were further examined between the two groups. The results, as shown in S6 File, indicated that there was no significant difference in awareness and intention to receive PR between the two groups. In terms of personal, social and environmental factors, patients with more breathlessness were older, retired, had lower level of education, and showed a higher preference for no smoking. In contrast, patients with less breathlessness placed greater emphasis on maintaining regular exercise and the quality of PR services provided by medical facilities. As our data were collected from two distinct regions of Zhejiang province in China: Hangzhou and Quzhou, we also compared intention to receive PR services, as well as personal, social and environmental influencing factors, between the two regions. The results showed that, a significantly higher proportion of patients living in Hangzhou had stronger intention to receive PR than those living in Quzhou. Notably, there were no significant differences in most patients’ personal influencing factors between the two regions, except for personal awareness. However, there were significant disparities in all of the investigated environmental barriers (S7 File).

Demand analysis of PR interventions in COPD patients based on Kano model

By using the Kano model, we conducted the demand analysis of PR interventions in COPD patients from 18 attributes, categorized in 8 dimensions (Table 3). As a result, 9 attributes were classified as being of one-dimensional quality, that is, the greater the level to which these PR services were provided, the higher the level of patient satisfaction, and vice versa. These 9 attributes were “provide lower limb interventions”, “inform about prohibited foods”, “education on how to arrange diet properly”, “regularly assess patients’ psychological condition”, “regularly communicate with patients and their family to motivate them for rehabilitation”, “popularize the pathophysiology and clinical basic knowledge of COPD”, “medication instruction”, “personalized guidance on daily life”, and “provide sputum expectoration guidance”. On the other hand, the remaining 9 attributes were classified as being of indifferent quality, in which the quality is not important to COPD patients. These indifferent attributes were services to “provide upper limb exercise interventions”, “provide systemic exercise interventions”, “provide chest breathing training”, “provide pursed lip breathing training”, “provide abdominal breathing training”, “provide long-term oxygen therapy interventions”, “provide non-invasive ventilation interventions”, “dietary guidance according to TCM syndrome differentiation theory”, and “provide TCM non-drug therapy”.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Evaluation of PR intervention need attributes based on the Kano model.

https://doi.org/10.1371/journal.pone.0290828.t003

To further clarify which attributes should be provided with the highest priority, the Kano analysis also provided an overview of the degree of satisfaction or dissatisfaction among COPD patients when a certain service was or was not offered (Fig 3). As shown in Fig 3A, the same 9 attributes which were recognized as being of one-dimensional quality above also attained the higher SI, ranged between 0.73 and 0.88, indicating a high degree of satisfaction if they were offered. The same 9 attributes also displayed relatively high absolute value of DSI (all above 0.50), indicating a high level of dissatisfaction if these services were not provided. Similarly, these 9 attributes all exactly located in the upper right section of the matrix graph presented in Fig 3B, which further proved that these services should be provided with the highest priority.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. The SI and DSI (A) and their matrix graph (B) of PR needs of COPD patients.

https://doi.org/10.1371/journal.pone.0290828.g003

Comparison of the demands of PR services in various subgroups of COPD patients

To compare Kano quality categories between various subgroups, the chi-square test was performed (Table 4). First, COPD patients were divided into two subgroups according to their awareness of PR (yes or no), and the findings revealed that 6 attributes were assigned to distinct demand categories (“O” and “I”) under these two subgroups, suggesting that these 6 services were in high demand (i.e., one-dimensional quality) for COPD patients who were aware of PR but resulted in indifferent quality for the rest of patients. The identified 6 distinct PR interventions were lower limb exercise interventions, systemic exercise interventions, chest breathing training, pursed lip breathing training, abdominal breathing training and expectoration guidance. Second, based on patients’ intention to receive PR services, we further classified patients into willingness (with “very high” or “high” level of intention) and unwillingness (with “moderate”, “low” or “very low” level of intention) subgroups. As a result, 5 attributes were of greater demand in the willingness subgroup (categorized as “O”) but were of no demand in the unwillingness subgroup (categorized as “I”). All of the 5 attributes were overlapped with the findings from the awareness-subgroup analysis. The details were summarized in Table 4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Demand differences of PR interventions in COPD patients in various subgroups.

https://doi.org/10.1371/journal.pone.0290828.t004

In addition, we divided the investigated COPD patients into subgroups based on several influencing factors, including breathlessness level, age, household monthly income, educational level and region. The results showed that, patients with less breathlessness had a higher demand for systemic exercise interventions, which were categorized as one-dimensional qualities. This was consistent with the findings from the influencing factors section. On the other hand, patients with more breathlessness had a higher demand for the following 3 services: long-term oxygen therapy interventions, non-invasive ventilation interventions, and expectation guidance, all of which were classified as one-dimensional qualities. Likewise, COPD patients with older age (≥75 years old), higher levels of household monthly income(≥7000RMB) and education (Senior secondary school and above) were also found to have high demand for these 3 PR services. There were no substantial disparities in the demand for most PR services between the two regions, except for expectoration guidance. Patients from Hangzhou manifested a stronger need for this service, which was identified as one-dimensional quality, whereas patients in Quzhou considered it an indifferent quality (Table 4). The SI and DSI results also confirmed these findings (S8 File).

Summary

In this study, firstly, we revealed a low acceptance of PR among Chinese COPD patients and thoroughly investigated the underlying barriers from personal, social, and environmental perspectives. The recognized barriers included limited awareness of PR, concerns about financial burden, transportation difficulties, and lack of family support. Secondly, we applied the Kano model to COPD patients for the first time and identified the 9 most-requisite PR intervention items, categorized into 4 dimensions: lower limb exercise interventions, diet guidance, psychological interventions and education interventions. Meanwhile, our subgroup analyses highlighted the significant demand for certain PR items, including exercise interventions, respiratory muscle training, oxygen therapy, and expectoration, among patients with higher levels of breathlessness, older age and higher income and education levels, this could provide a theoretical basis for designing and delivering tailored PR interventions for specific patients. Based on the findings of the Kano model and the roles/functions of different medical institutions, we proposed here a differentiated hierarchical PR intervention mode, which clarifies the responsibilities of medical institutions at various levels and enables medical staff to promote tailored PR services, forming a whole-cycle continuous PR intervention for COPD patients at different stages.

Patients’ intention and perceived influencing factors

In our study, 75.1% of COPD patients were willing to participate in PR interventions while the intention still needs to be improved. A cross-sectional survey in China reported the similar findings and further revealed that Chinese COPD patients were not sufficiently aware of the knowledge and importance in PR [24]. This may explain the low willingness and participation rate of PR interventions in China. Another study also reported that lack of knowledge of PR and its benefits was the most frequently identified barriers for low referral to PR [59]. Likewise, our study showed that only 62.9% of the investigated COPD patients were aware of PR interventions, which were mainly informed by hospital medical staff, instead of workers in community-level health centers.

Additionally, over 90% of patients in our study cited the cost of PR services and the ratio of medical insurance reimbursement as major concerns for their intention to participate in PR programs, which indicated that financial status and out-of-pocket costs of PR were the significant factors influencing patients’ perceptions, acceptances and participation in PR [22, 24, 60]. Meanwhile, as observed in our study, the PR skills of medical staff, knowledge promotion or public education of PR in the community, transportation concerns, and degree of support from family and friends, were significantly associated with patients’ intention to accept PR. Our subgroup analysis of different regions also revealed that, compared to patients in Hangzhou, COPD patients residing in the less economically-developed city of Quzhou had significantly lower levels of awareness and willingness to accept PR. They also perceived more environmental barriers, indicating regional disparities in the accessibility and utilization of PR. In summary, these results confirmed that limited environmental resources such as an inadequate medical insurance system [60, 61], inconvenient transportation [20, 61], lack of knowledge regarding the benefits of PR [24, 62], as well as variations in the availability and quality of PR services [24, 62], were major obstacles for the promotion of PR.

Therefore, addressing these environmental barriers should be a priority to increase access to PR for patients living in deprived areas who are at higher risk for hospital admission and poorer health outcomes [63]. For instance, some participants who did not complete the PR program expressed a preference for home-based programs due to transportation burdens and a greater sense of security at home [20]. It was also found that patients and family members who received a family-based PR program had significantly greater improvements in family coping and lower stress levels. This could potentially enhance patients’ beliefs about the consequences and expectations of rehabilitation outcomes and promoting attendance in PR [18, 64]. On the other hand, evidence from other studies suggested that PR could reduce subsequent health resource utilization, such as days spent in the hospital [65, 66], highlighting the importance for education about the potential benefits of PR for COPD patients [20]. Therefore, further efforts should prioritize knowledge promotion and public education of PR in the community for patients and their family. Simultaneously, community-based and family-based PR programs should also be emphasized to facilitate COPD patients’ perceptions, acceptances and participations in PR.

Patients’ demand

By using the Kano model, we identified most-requisite 9 PR services (i.e., one-dimensional qualities) from patients’ perspective. These services could be classified into 4 categories: diet guidance, education interventions, psychological interventions and lower limb exercise interventions. Our survey found that over 70% of COPD patients had a high level of understanding of these services (i.e., 82.7% for "diet guidance", 78.5% for "psychological interventions" and 70.9% for "exercise interventions"). Malnutrition is a prevalent condition among COPD patients with an incidence ranging from 25% to 65%, and has been identified as an independent risk factor for poor prognosis in COPD, resulting in various adverse consequences, such as prolonged hospitalization, increased family financial burden as well as heightened anxiety [67–69]. Evidence suggests that diet plays a vital role in obstructive lung diseases, including COPD [70, 71]. For instance, a mediterranean-like diet (characterized by high intake of poultry, eggs, fish, vegetables, legumes, potatoes, dairy desserts, fruits, nuts, and dried fruit) may protect lung function [72], while cured meat and processed red meat intake may increase the risk of COPD readmission [73, 74]. Therefore, how to choose an appropriate diet is a common concern and a crucial aspect of daily life for COPD patients, and this may lead to a high demand for dietary guidance in PR interventions. Consistent with the findings in the influencing factors section, health education and psychological guidance offering to COPD patients, especially to their family, have been further proven vital for raising patients’ awareness of PR and strengthening belief of PR effectiveness to promote patients’ willingness to engage in PR programs [61, 64, 75]. In terms of exercise interventions, a prospective cohort study suggested that physical activity is the strongest predictor of all-cause mortality in COPD patients and showed that active participants have a higher probability of survival in comparison to those who are sedentary or very inactive [76]. Dyspnea due to impaired function of ambulation muscles is one of the common symptoms of COPD, and better lower limb muscle function represents the physiological basis of better exercise training in COPD [8, 77]. Therefore, both the above studies and our findings emphasized the importance of lower extremities training.

Meanwhile, the remaining 9 items related to respiratory muscle training, oxygen therapy, TCM-based non-pharmacological therapy and provide upper limb and systemic exercise interventions, were identified as being of indifferent attributes in our study, implying that patients didn’t want to engaged in these services at this stage. However, such low demand may be attributed to patients’ insufficient knowledge of these services, as our findings showing that less than 30% of investigated patients rated themselves as having high-level of cognition of respiratory muscle training and TCM-based non-pharmacological therapy. Based on this finding, we performed subgroup analysis of awareness and intention. The results further proved that higher awareness and willingness would increase patients’ need for lower limb exercise interventions, respiratory muscle training and expectoration guidance, whereas poor perception resulted in the low attendance rate and poor adherence in PR [24]. Therefore, we anticipated that such low demand could be improved through targeted PR education. In addition, our subgroup analysis revealed that a higher level of breathlessness, older age and higher income and education levels would specifically increase patients’ demand for long-term oxygen therapy interventions, non-invasive ventilation interventions, and expectoration guidance. Previous evidence suggested that oxygen therapy was mainly used for exacerbation COPD patients with severe dyspnea symptoms, mostly among older patients, while some of them may also require ventilatory support if they need immediate admission to the respiratory care or intensive care unit (ICU) [78, 79]. However, in our study, there was no significant difference in awareness and intention to receive the overall PR services among patients with different levels of breathlessness, implying the importance of individualized PR strategies. The 2015 official policy statement of enhancing implementation, use, and delivery of PR issued by ATS/ERS stated that patients entering PR programs exhibit heterogeneity regarding disease state, symptoms, as well as functional limitations, and further emphasized giving weight on patient assessments conducted at the start of PR to properly characterize patients and enable delivery of individualized rehabilitation according to each patient’s needs [21]. Furthermore, it has been noted that highly educated patients might have a better understanding of COPD and PR, promoting higher demand [24], while patients with low income tend to have a poor awareness of disease and be vulnerable to adverse circumstances, resulting in lower demand [22, 24, 63]. Therefore, for patients with a higher level of breathlessness, older age, and lower income and education levels, medical staff should specifically emphasize on the severe outcomes of COPD, including respiratory limitations or acute exacerbation. This can help patients understand the importance of long-term oxygen therapy interventions, non-invasive ventilation interventions, and expectoration guidance, so as to improve their participation in PR [23].

Promotion strategies for PR management

In this study, we found that the awareness rate of PR interventions among COPD patients is considerably low, with the primary source of PR knowledge acquisition being limited to hospitals. However, it is unrealistic for patients to fully understand the importance of PR interventions, develop the necessary skills for PR, and effectively engage in self-management through brief consultations and interactions with healthcare professionals in hospitals alone. As such, there is an urgent need to provide COPD and PR-related health education or knowledge promotion to patients and their family through community-based programs, to help raise their awareness of COPD treatment strategies, as well as the benefits of PR.

Currently, there is a lack of specific guidelines addressing PR promotion for COPD patients. The "Technical Plan for Graded Diagnosis and Treatment Services for Chronic Obstructive Pulmonary Diseases" issued by the General Office of the National Health and Family Planning Commission primarily focuses on the responsibilities of medical institutions based on the complexity of patients’ conditions, without providing a systematic guide for specific PR projects [80]. As a result, most COPD treatments and interventions are still performed by hospitals in China. However, due to the characteristics of chronic respiratory diseases and the need for long-term and continuous rehabilitation, it is difficult for hospitals that mainly treat severe patients, to manage a large number of stable COPD patients simultaneously. According to previous study and our findings, cost and transportation concerns have been identified as major barriers that prevent individuals from PR participation [20, 61]. A more cost-effective community-based mode may be the solution for PR interventions in the near future. However, primary healthcare institutions currently lack the capabilities to provide comprehensive PR services, which may be a great challenge for both community health centers and their medical staff [21]. We proposed here a differentiated hierarchical PR intervention mode based on our findings of the Kano model and the roles/functions of different medical institutions, aiming to not only alleviate the burden on medical staff in all level medical institutions by clarifying their distinct scope of services, but also help to provide tailored PR service packages to improve patient satisfaction and quality of life. Under this hierarchical mode, hospitals mainly provide PR interventions related to long-term oxygen therapy and non-invasive ventilation for severe or exacerbated COPD patients, while stable patients are referred to primary care institutions for continuous PR services involving personalized COPD education, psychological interventions and dietary guidance. Ultimately, this creates a whole-cycle continuous PR intervention for COPD patients at different stages, integrating the resources of both hospitals and primary medical institutions. Notably, novel techniques, such as telerehabilitation, can be incorporated into the community or home-based PR services in primary healthcare facilities as convenient and cost-effective ways to serve COPD patients [81, 82].

It has been demonstrated elsewhere that not all educational and care management programs are appropriate for all patients [83]. To increase the effectiveness and efficiency of PR programs, we suggested that diet guidance, education interventions, psychological interventions and lower limb exercise interventions should be provided to every COPD patient with high priority. Meanwhile, long-term oxygen therapy interventions, non-invasive ventilation interventions, and expectoration guidance may be more important for patients with severe breathlessness or older age. Notably, along with the changing environment and disease progress, the demand attributes identified by the Kano model exhibit a progressive developmental pattern, changing from indifferent demand to attractive demand, one-dimensional demand, and must-be demand, and have a certain life cycle [84]. Consequently, it is crucial to continuously monitor and adjust the proposed strategies of PR services in a dynamic manner, so as to strengthen COPD patients’ motivation to engage in PR interventions, enhance their satisfaction, and improve their health condition and quality of life. Further studies with larger samples are necessary to validate our findings and optimize the individualized PR plans for different COPD patients.

Limitations

There are some limitations in this study. Firstly, potential bias may exist due to the small sample size, as well as the convenient sampling of COPD patients in hospitals of Zhejiang province of China. Secondly, we explored intention and demand from the level of patients’ perception, while it is also essential to explore current PR services from medical staff’s perspective, and ultimately match the supply and demand from both sides.