An Evidence Summary of Health Education in Chronic Obstructive Pulmonary Disease: A Practice-Oriented Assessment with Inter-Rater Reliability

Introduction

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide, second only to ischemic heart disease and stroke. The global burden of this disease is becoming increasingly severe and is expected to continue growing. According to the Global Burden of Disease Database, the number of patients with COPD is projected to approach 600 million by 2050.¹ The increase in prevalence is highly correlated with the trend of population aging, and the majority of patients are elderly individuals. This group often has limited self-management ability due to factors such as low education level and cognitive decline, which poses more severe challenges for disease education and management.^2,3

Although drug treatment is the basis for alleviating COPD symptoms, the full clinical efficacy of this treatment largely depends on whether patients correctly master the inhalation technique and maintain good compliance. Systematic patient education plays an irreplaceable role in standardizing inhalation behavior and enhancing compliance. Multiple systematic reviews have confirmed that self-management education can effectively improve the quality of life and disease knowledge of patients, and significantly reduce hospitalizations and emergency visits due to respiratory problems.^4–6 Therefore, effective health education strategies have become an indispensable component of comprehensive management plans for patients.

Although the Global Strategy for Chronic Obstructive Pulmonary Disease and the guidelines and statements of the American Thoracic Society/European Respiratory Society on pulmonary rehabilitation emphasize the crucial role of education in COPD management interventions,^7,8 existing studies significantly lack homogeneity in terms of intervention content, implementation methods, and interdisciplinary personnel allocation. This is manifested as follows: (1) in terms of educational content, there are complex multidisciplinary plans covering the entire disease cycle, as well as one-sided publicity focusing only on the use of a single inhaler.^9,10 (2) in terms of educational methods, from traditional offline oral guidance and paper manuals to digital medical and remote follow-up, the forms are diverse but mostly one-way teaching^11,12 and (3) in terms of personnel allocation, the admission standards and responsibilities of multidisciplinary teams are also not clearly defined.¹³ This heterogeneity not only causes the research evidence for COPD health education to be scattered but also leads to a lack of standardized implementation plans in clinical practice.

Therefore, to compensate for the lack of scattered evidence and standardized implementation plans, this study adopts an evidence-based approach, systematically searches, evaluates, and integrates evidence in the field of COPD health education, aiming to provide a rational educational basis for clinical medical staff, thereby improving patients’ understanding of the disease and enhancing their self-management ability at home.

Materials and Methods

Scheme Registration

The protocol of this study has been registered with the Open Science Framework (OSF). The registration DOI is https://doi.org/10.17605/OSF.IO/UJYRX.

Formulation of the Evidence-Based Question

The PIPOST model was used to frame the evidence-based question:

Population (P): Patients with COPD.

Intervention (I): Health education strategies.

Professionals (P): Healthcare professionals (physicians, nurses, respiratory therapists, pharmacists, nutritionists, etc).

Outcome (O): Disease knowledge level, self-management ability, quality of life, medication adherence, inhalation technique mastery, and acute exacerbation rate.

Setting (S): Hospitals, community health centers, rehabilitation facilities, and home care settings.

Type of evidence (T): Clinical practice guidelines, evidence summaries, systematic reviews and expert consensuses.

Retrieval Strategy

Following the “6S” pyramid evidence model and a top-down search principle, a comprehensive search strategy using subject headings and free-text terms was employed. Databases and websites searched included:

Evidence-based resources: BMJ Best Practice, UpToDate, Cochrane Library.

Guideline development websites: National Institute for Health and Clinical Excellence (NICE), Registered Nurses’ Association of Ontario (RNAO), Medlive, Guidelines International Network (GIN).

Professional society websites: American Thoracic Society (ATS), European Respiratory Society (ERS), Global Initiative for Chronic Obstructive Lung Disease (GOLD).

General databases: China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP Database, Sinomed, PubMed, Embase, Web of Science.

The search period was from database inception to November 18, 2025. The specific search methods for all databases and websites are detailedin Appendix 1. Additionally, we manually searched the reference lists of all included guidelines, systematic reviews, evidence summaries, and expert consensus.

Literature Inclusion and Exclusion Criteria

The inclusion criteria were as follows: (1) patients with COPD; (2) focus on COPD health education; (3) study types: clinical practice guidelines, evidence summaries, expert consensus, or systematic reviews (with the most recent version included for guidelines or expert consensus); and (4) published in Chinese or English.

The exclusion criteria were as follows: (1) not in Chinese or English; (2) duplicate publications; and (3) low methodological quality.

Evaluation and Screening Criteria for Literature Quality

Two researchers (MZ and JC) who had received systematic evidence-based medicine training independently conducted literature screening and quality evaluation. Using NoteExpress, they independently evaluated the titles and abstracts based on preset inclusion and exclusion criteria. For the screened literature, full texts were obtained and independently evaluated by the two researchers. Subsequently, the AGREE II tool was used to evaluate the quality of the guidelines, and the Australian JBI standards (2016) were used to evaluate the quality of systematic reviews and expert consensus.

The inter-rater consistency of the title/abstract screening, full-text eligibility assessment, and quality evaluation stages was tested using Cohen’s Kappa coefficient,¹⁴ with the Kappa value serving as the effect size. A Kappa value of ≥ 0.81 indicates excellent consistency, 0.61 to 0.80 indicates good consistency, 0.41 to 0.60 indicates moderate consistency, 0.21 to 0.40 indicates average consistency, and ≤ 0.20 indicates poor consistency. Any disagreements that arose during this process were first resolved through consensus discussions; if a consensus could not be reached, the final decision was made by the third senior researcher (HYZ).

Data Synthesis and Classification

During evidence extraction, when the two researchers (MZ and JC) encountered conflicting recommendations, they followed the principle of “prioritizing the most recent, high-quality, evidence-based evidence.”

Subsequently, the Australian JBI Evidence Pre-grading System (2014)¹⁵ was used to classify the levels of evidence and determine the strength of recommendations. According to the study design, the evidence was classified into five main levels: Level 1 (experimental designs, including randomized controlled trials), Level 2 (quasi-experimental designs), Level 3 (observational-analytic designs), Level 4 (observational-descriptive designs), and Level 5 (expert opinion or basic research). For each piece of evidence, a corresponding sub-level (eg, 1a, 1b, 5b) was assigned based on the specific document type and original research method. Following the JBI recommendations, after a structured assessment of effectiveness, feasibility, appropriateness, and clinical significance, the strength of recommendation was rated as Grade A (strong recommendation) or Grade B (weak recommendation).

Results

Consistency Test

The Cohen’s Kappa coefficient between the two researchers (MZ and JC) was 0.83 (95% CI: 0.79–0.87) for the title/abstract screening stage, and 0.88 (95% CI: 0.82–0.94) for the full-text eligibility assessment and quality appraisal stage, indicating excellent agreement between the two reviewers.

General Characteristics of the Included Literature

A total of 9,825 records were initially identified through the database searches: CNKI (n=96), Wanfang Data (n=332), VIP (n=659), PubMed (n=1,455), Web of Science (n=1,758), Cochrane Library (n=322), SinoMed (n=310), Embase (n=3,392), BMJ Best Practice (n=272), UpToDate (n=87), NICE (n=60), RNAO (n=5), GIN (n=0), Medlive (n=450), ATS (n=587), ERS (n=20), and GOLD (n=20). Furthermore, all references included in the guidelines, systematic reviews, evidence summaries, and expert consensuses were manually searched, and no other eligible literature was found.

After removing 3,385 duplicate records using NoteExpress, 6,440 records remained eligible for screening. Following title and abstract screening, 6,267 records were excluded, leaving 173 records for full text assessment. After a full text review, 144 records that did not meet the inclusion criteria were excluded, 5 records could not be obtained in full text, and 3 records were excluded because they were not published in Chinese or English. Ultimately, 21 records were included (Figure 1), comprising eight systematic reviews,^5,6,16–21 nine clinical practice guidelines,^22–30 and four expert consensuses (Table 1).^31–34

Table 1 Basic Characteristics of Included Articles (n=21)

Figure 1 Flow diagram of the article selection process.

Quality Evaluation Results of the Included Literature

Guideline

A total of nine clinical practice guidelines were included, among which two^22,23 were from Web of Science, four^24–26,28 from Wanfang Data, one²⁷ from Medlive, one²⁹ from GOLD, and one³⁰ from BMJ Best Practice The overall quality was good. In addition to Chen et al²³ and Criner et al²² were grade B recommendations. The rest were recommended at grade A and included. (Table 2).

Table 2 Results of the Quality Evaluation of the Guidelines (n=9)

Systematic Review

A total of eight systematic reviews were included, among which six^{5,6,16–18,20} were from PubMed and two^19,21 from CNKI. Among them, all entries in the three articles^19–21 were marked as “yes.” Wang et al,⁵ Smith et al,¹⁶ Stoilkova et al,¹⁷ Liu et al,²¹ and Roberts et al¹⁸ only had the entry “Was the possibility of publication bias evaluated?”, which were evaluated as “no” or “unclear”, the rest of the items were all “yes”, and the overall quality was acceptable Therefore, all high-quality systematic reviews were included (Table 3).

Table 3 Quality Appraisal results of Systematic Reviews (n = 8)

Expert Consensus

A total of four expert consensuses were included, among which one³³ was from PubMed, one³¹ from CNKI, and two^32,34 from Wanfang Data. The evaluation results of all entries were “yes,” indicating high quality of the literature, and all were included(Table 4).

Table 4 Quality Appraisal Results of Expert Consensuses (n = 4)

Evidence Summary and Description

A total of 31 evidence items related to COPD health education were extracted in this study (Table 5) and categorized into seven themes: (1) principles, (2) providers, (3) recipients, (4) methods, (5) objectives, (6) core content, and (7) outcome assessment. The Cohen’s Kappa coefficient for the two researchers’ agreement (MZ and JC) on theme assignment was 0.82 (95% CI: 0.73–0.91), indicating a good inter-rater agreement.

Table 5 Summary of Best Evidence for Health Education for Patients with COPD

Discussion

Through systematic search and quality evaluation, this study ultimately included 21 articles, extracted 31 pieces of evidence, and summarized them into seven themes. The following section discusses this evidence and analyzes the challenges and prospects of clinical application.

Consensus and Differences in Core Topics of Health Education for COPD

Effective health education themes should be determined based on the available evidence, and individualized strategies should be developed by integrating patients’ preferences and specific circumstances.

Evidence items 6–30  (Table 5) systematically synthesized the consensus reached in the included literature, clarifying the core content of basic COPD health education, including disease basics, medication and inhalation techniques, recognition and management of acute exacerbations, pulmonary rehabilitation, smoking cessation, vaccination, nutrition, psychological support, and oxygen therapy. However, there are differences in the coverage and emphasis on educational themes across regions.

In high-income countries and international guidelines, while retaining the above basic themes, the scope of health education has been further extended to more advanced and strategic topics, such as advance care directives and end-of-life care,²⁹ travel advice,³⁰ and environmental exposure reduction.²⁷ These themes reflect the relatively well-developed healthcare infrastructure in these regions, as well as an emphasis on holistic care, patient dignity, and quality of life. In contrast, in low- and middle-income countries, the selection of health education themes is often constrained by practical conditions such as healthcare resource allocation and patients’ basic health literacy,^19,26,28 with a greater focus on fundamental and accessible content, such as basic disease concepts and correct use of inhalation devices. This heterogeneity may not reflect a flaw in the study quality but rather differences in healthcare systems, cultural backgrounds, resource accessibility, and disease management philosophies across countries. Taking China as an example, incorporating traditional exercises such as Tai Chi and Ba Duan Jin into education programs as low-cost, culturally adapted alternatives within the pulmonary rehabilitation theme demonstrates a pragmatic orientation under resource-limited conditions.

Therefore, future health education strategies should, on an evidence-based foundation, fully consider regional characteristics, healthcare resource allocation, and individual patient preferences to achieve truly individualized education.

Establishing Patient-Centered COPD Health Education

However, a personalized theme list is only the first step in the educational process. Translating this content into patients’ actual actions requires targeted strategies, which is also an inevitable requirement for practicing a “patient-centered” approach.

Evidence item 4 (Table 5) indicates that common methods of health education include verbal instruction, printed materials, video demonstrations, health lectures, and Internet-based remote education.³⁵ In clinical practice, verbal instructions and printed handouts still dominate. However, Giguère et al³⁶ conducted a systematic review of 23 high-quality studies and concluded that printed educational materials as a standalone intervention did not significantly improve patient clinical outcomes. Muijsenberg et al³⁷ clearly pointed out that traditional didactic education struggles to meet patients’ real needs, providing a strong rationale for shifting health education toward personalized and multimodal approaches. Furthermore, approximately 40–80% of medical information is forgotten by patients immediately after consultation; during a 5-minute routine education session, the effective information retention rate is only 20%, and nearly half of that is misunderstood.³⁸ This phenomenon is known as the “knowledge transmission inverted triangle,” where information continuously attenuates as it is transferred from healthcare providers to patients, highlighting the fundamental limitation of traditional one-way educational models.

Therefore, based on patient empowerment theory, the core goal of health education should not remain at the level of knowledge transmission but should shift toward promoting patient behavioral change and ultimately enhancing patients’ self-management abilities. The teach-back method has been confirmed by multiple studies to be an effective two-way interactive strategy.^39–41 Notably, although the teach‑back method does not appear as an independent item in the evidence presented in this study (Table 3), its concept is highly consistent with Evidence item 5 (Table 5) and can serve as an important supplement for clinical translation. Its core advantage lies in transforming one-way information delivery into a two-way interactive cycle of “explain - repeat - correct” between healthcare providers and patients, ensuring that patients truly understand and master the information. This method not only effectively identifies and corrects patient misunderstandings, reducing the risk of medication errors or disease deterioration caused by incorrect understanding, but also enhances patient engagement and adherence to treatment. Simultaneously, it provides healthcare professionals with immediate feedback to evaluate educational effectiveness. Theoretically, the effectiveness of the teach-back method can be supported by the Knowledge-Attitude-Practice (KAP) model and the Health Belief Model (HBM). The KAP model points out that knowledge is the foundation of behavioral change, attitude is the driving force, and behavior is the ultimate goal. By repeatedly verifying patients’ understanding of information, the teach-back method ensures the accuracy of knowledge acquisition, lays the premise for forming correct attitudes, and ultimately promotes stable behavioral establishment. The HBM emphasizes that whether an individual adopts a health behavior depends on their perception of disease threat, recognition of behavioral benefits, and level of self-efficacy. During the “repeat-correct” process, the teach-back method helps patients clearly understand the specific risks of non-adherent behavior, reinforces the benefits of correct performance, and builds the confidence of “I can do it” through repeated successful repetition, thereby effectively translating health beliefs into daily health actions.

With “Internet + Healthcare” becoming a hot topic in research and practice, inefficient educational models are undergoing profound transformations.⁴² The teach-back method focuses on interaction quality, while digital technology focuses on resource accessibility, and the two can complement one another. Increasingly, researchers are leveraging remote platforms to deliver online education to patients, driving health education toward digitalization, personalization, and systematization.^43,44 Many platforms have achieved Internet interoperability, supporting patients in autonomously retrieving customized health knowledge.⁴⁵ These technologies not only expand the accessibility of educational resources but also empower patients with greater confidence to actively participate in health management processes.

Multidisciplinary Collaboration is the Structural Prerequisite for Achieving Personalized Education

Given that COPD is a chronic disease with multiple comorbidities, patients often suffer from various physical and psychological complications, such as cardiovascular disease, diabetes, osteoporosis, anxiety, and depression. This pathological characteristic determines that the self‑management skills that patients need to master are not limited to respiratory rehabilitation or medication adherence but rather span a complex knowledge system across multiple domains, including respiratory, cardiovascular, metabolic, musculoskeletal, and psychological health. This inherent complexity requires a multidisciplinary team (MDT) to provide leadership and support for individualized education.^46,47

As Anne stated,⁴⁸ MDT is not simply a collection of individuals but a collaborative network that organically integrates the roles of physicians, respiratory nurses, physiotherapists, dietitians, psychological counselors, and pharmacists. However, multiple studies have consistently indicated that the clinical implementation rate of MDT is far lower than expected.⁴⁹ Even in the context of strong guideline recommendations for MDT effectiveness,³⁰ real-world data remain discouraging. Henoch et al,⁵⁰ based on data from 7,810 patients with COPD in the Swedish National Airway Registry, found that less than 20% of patients had ever been in contact with a physiotherapist or dietitian. This means that although guidelines advocate for multidimensional, interdisciplinary comprehensive care, the vast majority of patients with COPD do not receive these core support services during routine clinical visits.

The reasons for this are as follows: First, there is a shortage of interdisciplinary staff. Effective MDT operation for COPD requires the joint participation of respiratory physicians, nurses, physiotherapists, pharmacists, and other professionals. Such professionals are severely lacking in most healthcare institutions, even in developed countries (eg, Ireland), and especially in low- and middle-income countries. Lim et al⁵¹ conducted a cross‑sectional survey of consultant respiratory physicians in the Republic of Ireland and found that only 31% of respondents reported having a dedicated COPD multidisciplinary team in their institution, and approximately 30% felt that adequate pulmonary rehabilitation resources were unavailable. Moreover, up to 71.4% of nurses reported a lack of specialized health-education training.⁵² Nurses are the main executors of education, and their insufficient knowledge base and lack of training opportunities are the deeper reasons why MDT implementation is difficult.

Therefore, we strongly recommend: (1) Strengthen the core staffing of MDT. Each tertiary hospital respiratory department should be equipped with at least one respiratory nurse, one physiotherapist, and one pharmacist as permanent core members of the MDT. For institutions with staffing shortages, assistance should be provided to lower-level hospitals through tertiary hospital Internet platforms. Simultaneously, systematically trained respiratory nurses should be allowed to take on multiple expanded roles (eg, basic pulmonary rehabilitation guidance, rapid nutritional risk assessment, and psychological support) so that limited staffing can cover the core functions of the MDT. (2) Improving the education and certification systems for respiratory nurses. Pre-service and in-service education for all respiratory nurses should adopt a model combining simulated cases and standardized patients, and MDT qualification should be certified through objective structured clinical examinations (OSCEs). Healthcare institutions should make passing this certification a mandatory requirement for nurses responsible for COPD patient follow-up, and regular re-evaluations should be conducted.

Limitations

This study has the following limitations: (1) the literature search was limited to Chinese and English databases, which may introduce language bias; (2) some of the included literature was not specifically focused on COPD health education, so its applicability may be limited; and (3) the existing evidence originates from studies conducted in multiple countries, with insufficient consideration given to specific cultural backgrounds, health literacy levels, and social support systems. Therefore, when health education programs are applied in local contexts, adaptive adjustments and innovations are needed; (4) the researchers performed the extraction, grading, and synthesis of evidence. Although we used independent assessments by two reviewers and Kappa consistency tests, some subjective judgments may still exist, potentially affecting the robustness of the evidence synthesis.

Conclusion

This study systematically retrieved and synthesized 31 evidence items on health education in COPD, covering seven themes: educational principles, providers, recipients, methods, objectives, core content, and outcome assessment. The evidence indicates that effective health education should shift from traditional one-way knowledge transmission to an interactive model centered on patient empowerment. Educational content should be tailored based on regional resources, cultural backgrounds, and patients’ health literacy levels, and a closed loop of “education – assessment – feedback” should be established to continuously optimize outcomes. Relying on multidisciplinary teams and intelligent technologies, health education can achieve systematic, personalized, and interactive delivery. Future research should focus on the development and validation of localized education programs in low‑resource settings, implementation strategies for multidisciplinary team collaboration, and long‑term intervention outcome evaluations supported by digital technologies.