Characteristics and Quality of Traditional Chinese Therapies and Integrative Medicine Clinical Practice Guidelines for Musculoskeletal Disorders Published in Mainland China
Article information
Abstract
Background
Musculoskeletal disorders are prevalent in adults. Traditional Chinese medicine (TCM) and integrative medicine (IM) are commonly used treatments which have clinical practice guidelines (CPGs). This study aimed to determine the characteristics and quality of these CPGs.
Methods
CPGs which recommended TCM/IM therapies in musculoskeletal conditions/diseases published in Chinese or English between January 2018 to December 2022 in mainland China were retrieved and analyzed for guideline classification, funding source, conflict of interest, and methodology. Appraisal of Guideline for Research and Evaluation II including 6 domains, was applied to assess CPG quality.
Results
Of the 50 CPGs included, there were 19 TCM, 5 IM, and 26 western conventional medicine (WCM) guidelines of which osteoporosis (13, 26%), osteoarthritis (11, 22%) and rheumatoid arthritis (6, 12%) were the most frequent diseases. The TCM therapies recommended by the CPGs successively were acupuncture and moxibustion, Chinese patent medicine, and TCM decoction based on syndrome differentiation. Nearly half of the CPGs reported funding source (52%) and conflict of interest (48%). Thirty-six CPGs used the Grading of Recommendations, Assessment, Development, and Evaluations method to present summaries of evidence, the remaining did not report the method. Based on Appraisal of Guideline for Research and Evaluation II scores, “clarity of presentation” scored the highest (55%), while “applicability” was the lowest (6%). No CPG was recommended without change, and 23 CPGs were not recommended.
Conclusion
The quality of CPGs for musculoskeletal conditions/diseases in China is generally low. Future CPGs should pay more attention to standardized developing procedures.
Introduction
The International Classification of Diseases 11th Revision (ICD-11)[1] for musculoskeletal conditions/diseases refers to a relatively common condition/disease in human which includes arthritis, spondylosis, and muscle strains. It is reported, by World Health Organization, that about 1.71 billion people in the world suffer from musculoskeletal disorders, and this is reported to be a leading cause of disability worldwide [2]. Musculoskeletal conditions/diseases severely limit mobility and flexibility, lead to an early retirement, a lower quality of life, and reduced ability to participate in social activities. However, due to population growth and aging in China, the number of people with musculoskeletal conditions/diseases is increasing rapidly, and the number of disabilities due to these conditions is also expected to continue to increase in the coming decades [2].
Clinical practice guidelines (CPGs) are developed using systematic review of evidence and an assessment of the benefits and harms of care options to provide statements of recommendations to optimize and assist practitioner and patient making decisions about the appropriate patient health care [3]. As statements are based on evidence-based medical evidence, normative guidelines recommendations are supposed to originate from the most optimized and comprehensive evidence [3]. Therapies from traditional Chinese medicine (TCM) and integrative medicine (IM) in China are widely used. Many publications of TCM/IM CPGs in musculoskeletal conditions/diseases have appeared in the past 5 years. Evidence-based guidelines should be the mainstream of TCM CPGs [4]. However, there is a lack of quality assessment in CPG methodology. Therefore, it is necessary to evaluate the quality of the guidelines using an international guideline evaluation instrument such as the Appraisal of Guideline for Research and Evaluation [5].
At present, quality assessment in most CPGs is aimed at individual musculoskeletal conditions/diseases, whilst there is a lack of overall characteristics and quality of CPGs in musculoskeletal disorders especially in the field of TCM/IM. This study aimed to describe the characteristics of CPGs of musculoskeletal conditions/diseases which were published in mainland China, and focus on the quality and the recommendations in the guidelines on TCM/IM to optimize clinical practice in these areas. The findings of our study may provide a basis for future guideline development and quality improvement.
Materials and Methods
1. Inclusion and exclusion criteria
1.1. The inclusion criteria
CPGs approved and issued by official associations in mainland China and published in Chinese and/or English between January 2018 to December 2022.
The guidelines focused on musculoskeletal conditions/diseases (named by the ICD-11), and the recommendations covered TCM/IM.
The guidelines published in the form of a journal or monograph. TCM CPGs must meet the following criteria: (1) The title of the CPG contains TCM-related elements (1 or more), such as “traditional Chinese medicine,” “Chinese medicine,” “Chinese patent medicine,” “acupuncture,” and “combination of disease and pattern;” and (2) CPGs issued by the China Association of Chinese Medicine and/or its affiliated branches. IM CPGs must meet the following criteria: (1) The title of the guideline contains “integrative medicine” or “traditional Chinese medicine and Western medicine;” and (2) CPGs were issued by Chinese Association of the Integration of Traditional and Western Medicine and/or its affiliated branches. Western conventional medicine (WCM) CPGs must meet the following criteria: (1) There are no obvious words in the title that fall into TCM/IM CPGs categories (secondary criteria); and (2) CPGs were issued by official societies or institutions related to musculoskeletal conditions/diseases.
1.2. The exclusion criteria
WCM guidelines which did not recommend any TCM therapies. For guidelines with different versions, the older version(s) were excluded. Experts’ consensus, unpublished guidelines, or interpretation of CPGs were excluded.
2. Search strategy
Guidelines approved by official associations in mainland China were retrieved from January 2018 to December 2022 (SinoMed, China National Knowledge Infrastructure, Wanfang, China Science and Technology Journal Database, PubMed, Embase, and Google Scholar). The Chinese and English keywords for the search were: guideline, osteoarthritis, infectious arthritis, rheumatoid arthritis, psoriatic arthritis, polymyalgia rheumatica, adult-onset still disease, juvenile idiopathic arthritis, gout, joint disorders, deformities of limbs, effusion of joint, osteoporosis, spine deformities, isthmic spondylolisthesis, intervertebral disc degeneration, spondylolysis, spinal stenosis, infection of vertebra, infection of intervertebral disc, inflammatory spondyloarthritis, spondylopathies, cervical pain, thoracic pain, low back pain, sciatica, infectious myositis, calcification or ossification of muscle, contracture of muscle, muscle strain, sprain, ischemic infarction of muscle, tenosynovitis, ganglion, fracture, chondropathies, osteonecrosis, osteomyelitis, osteitis, low bone mass disorders, traditional Chinese medicine, integrated traditional and Western medicine, Chinese medicine, Chinese patent medicine, herbal medicine, ethnic medicine, acupuncture, acupoint, needling, needle, moxibustion, massage, tuina, cupping, qigong, and scraping therapy. Detailed search strategies were shown in Appendix A.
Additional searches were conducted in: (1) 2 guideline websites [Guidelines International Network, and Medlive (www.medlive.cn)]; (2) 3 major Chinese associations websites (the Chinese Medical Association, the Chinese Society of Traditional Chinese Medicine, and the Chinese Association of Integrative Chinese and Western Medicine); and (3) 2 online websites where books can be purchased: Amazon.com and Dangdang.com. Search terms contained “condition/disease name + guideline.”
3. Guideline selection and data extraction
Guidelines were screened independently by 2 authors (XFW and JLZ) and cross-checked. If there was a disagreement, a 3rd author (JPL) helped to reach a consensus. Reference software NoteExpress 3.8 was used to remove duplicates and select studies. Two authors (XFW and JLZ) read the title and abstract for the initial screening, and the full texts were retrieved for further screening. Four authors (LJL, XZL, LDX, JLZ) extracted the data in pairs from the eligible guidelines according to a predesigned information sheet, and the extracted data included published year, funding source, conflict of interest, methods, guideline classification, scope, focused condition/disease and recommended TCM therapy.
4. Quality evaluation
The Appraisal Guideline for Research and Evaluation II (AGREE II) tool was used for quality assessment [5]. AGREE II has 6 domains (scope and purpose, stakeholder involvement, rigor of development, clarity of presentation, applicability, and editorial independence) of which there are 23 items. The domain “scope and purpose” is concerned with the overall aim of the guideline, the specific health questions, and the target population. “Stakeholder involvement” focuses on the extent to which the guideline was developed by the appropriate stakeholders and represents the views of its intended users. “Rigor of development,” relates to the process used to gather and synthesize the evidence, the methods to formulate the recommendations, and to update them. “Clarity of presentation” deals with the language, structure, and format of the guideline. “Applicability” pertains to the likely barriers and facilitators to implementation, strategies to improve uptake, and resource implications of applying the guideline. The last domain is “editorial independence,” which is concerned with the formulation of recommendations not being unduly biased with competing interests. Assessors responded to the 23 items under 6 domains by giving 1 for “strongly disagree” to 7 for “strongly agree.” Each domain score was calculated as follows: (the actual score - the lowest possible score) / (the highest possible score - the lowest possible score) × 100%. The higher the score, the more normative the domain. The highest score rate is 100% and the lowest score rate is 0%. Guidelines where all domains had a standardized score of 50% or greater were rated as “A” which indicated recommendation without change. Guidelines that owned at least 3 domains with less than 30% were rated as “C,” which indicated not to be recommended. The rest were rated as “B,” which indicated that following revision it would be recommended.
In this study, 4 authors (XFW, JLZ, LJL, LDX) assessed the guideline quality and they all received training on the use of the AGREE II tool before the assessment. Then interclass correlation coefficient (ICC) of results from 4 authors for 6 guidelines (randomly selected by SPSS Version 26.0) was calculated by SPSS Version 26.0, only ICC values between 0.75–0.9 were used, and a formal evaluation of the other guidelines was conducted. Scores from 23 items under 6 domains for all guidelines were summarized in Excel 2020 and further analyzed using SPSS Version 26.0. Descriptive approach was used to analyze guidelines for their characteristics and quality components.
Results
1. Guideline selection
A total of 50 guidelines met the inclusion criteria, among which 49 were published in journals, and 1 was published as monograph. The selection process was shown in the PRISMA format (Fig. 1).

Flow chart of guideline selection.
* Number of articles retrieved from different databases and registers.
† Number of non-related records excluded by a human.
CACM = Chinese Association of Chinese Medicine; CAIM = Chinese Association of Integrative Medicine; CMA = Chinese Medical Association; CNKI = China National Knowledge Infrastructure; GIN = Guidelines International Network; TCM = traditional Chinese medicine; VIP = China Science and Technology Journal Database.
2. Characteristics of included guidelines
Detailed characteristics of the 50 CPGs is described in Table 1 [6–56]. These guidelines were published between 2018 and 2022, with the highest number being published in 2020. Of the 50 guidelines, 19 were classified as TCM, 5 as IM, and 26 as WCM. The study also noted that among all 74 retrieved WCM guidelines on musculoskeletal conditions/diseases, only 26 (35%) involved TCM/IM therapies while the remaining 48 guidelines were excluded because they did not meet the inclusion criteria. Only 1 guideline was published in English, while the remaining 49 were published in Chinese.

Characteristics of the Clinical Practice Guidelines of TCM, IM, and WM for Musculoskeletal Conditions/Diseases
In terms of individual conditions/diseases, 13 guidelines (26%) focused on osteoporosis, followed by osteoarthritis (11, 22%), including knee osteoarthritis (7, 14%) and osteoarthritis in general (4, 8%). Rheumatoid arthritis had 6 guidelines (12%), lumbar disc herniation had 5 guidelines (10%), osteoporotic fracture had 3 guidelines (6%), hyperuricemia/gout had 2 guidelines (4%), and low back pain had 2 guidelines (4%). The remaining 8 guidelines (16%) covered other musculoskeletal conditions/diseases. Overall, there were more guidelines addressing bone disorders than muscle disorders.
The most frequently recommended TCM therapies in the 50 guidelines were acupuncture and moxibustion. This was followed by Chinese patent medicine and Chinese herbal decoction based on syndrome differentiation. Additionally, non-pharmaceutical therapies including Tai chi/Tai chi chuan (typical therapeutic exercise), external application of Chinese medicine, and herbal fumigation were commonly recommended for musculoskeletal conditions/diseases. Other therapies mentioned were tuina, acupotomology, and Baduanjin exercise, and 7 out of 50 guidelines did not recommend any one specific intervention in TCM, but just mentioned the general term “TCM” (Fig. 2).
Regarding the methodology employed in the development of the guidelines, a majority of 36 guidelines (72%) incorporated the Grading of Recommendations, Assessment, Development, and Evaluations method for making recommendations. Conversely, in the remaining 14 guidelines (28%), no specific methodology was mentioned.
In terms of funding sources and potential conflicts of interest, 26 guidelines (52%) disclosed their funding origins (with most funding originating from national, provincial, or municipal projects), and 24 guidelines (48%) reported conflicts of interest.
The scope of the guidelines varied among the different domains. Approximately 68% (34 guidelines) focused on the aspects of diagnosis and treatment, while 8 guidelines (16%) covered diagnosis, treatment, and rehabilitation. The remaining guidelines primarily centered on single treatment modalities (6, 12%), prevention and treatment (1, 2%), or techniques related to condition/disease diagnosis (1, 2%).
3. Quality evaluation of guidelines for musculoskeletal conditions/diseases using AGREE II
The ICC value of 6 selected guidelines were demonstrated (Table 2 [11,18,25,26,45,54]), and concerning all 50 guidelines, the ICC value of the AGREE II official evaluation was 0.796 [95% CI (0.774, 0.817), p < 0.001], which indicated 4 evaluators could achieve a better consistency in the evaluation of each item. Domain scores of the AGREE II quality assessment are illustrated in Table 3.
As to the domain “scope and purpose,” most of the CPGs included in this review only reported the overall purpose and intended population for the guideline, and it was not clear what health issues were covered by the reporting guidelines, particularly, the key recommendations. In this study, CPGs scored 37.92% in the domain “stakeholder involvement” and most CPGs only provided the names of the main drafting experts, the working bodies, and the division of responsibilities in the preparation process, and did not clearly indicate important information such as their geographical distribution, subject categories, and technical titles. Neither patient nor methodological involvement was reported in CPGs. None of the CPGs reported patient preference and values. In the domain “rigor of development,” CPGs included in this review provided recommendations and references when providing evidence. However, imperfections such as search strategies, evidence-to-recommendation formation process, and solutions to differences in the consensus process were observed (Appendix A). Most of the CPGs did not describe complete search strategies, selection criteria, methods of evidence appraisal, expert external review process, timings for guideline updates, and update procedures. Regarding the domain “clarity of presentation,” CPGs included in this review clearly presented different recommendations for different populations experiencing the same condition and were highlighted by fonts, paragraph settings or charts. However, most of the recommendations did not report in detail the number of times that a recommended drug was taken, the time, the dosage, and the recommended exercise time. The domain “applicability” scored lowest (6.04%). None of the CPGs included in this review reported on the facilitators and disincentives for the use of the guidelines, did not provide recommendations and tools for applying recommendations, and did not describe resources for the application of all recommendations. Most CPGs did not explicitly provide monitoring and auditing criteria for the application of recommendations. For the domain “editorial independence,” most of the CPGs included in this review did not declare whether the sponsors had influence in the content of the guidelines. Most CPGs did not report management of conflict of interest.
No CPG was evaluated as a Level A (directly recommended without any changes), 27 CPGs were evaluated as Level B (modified and improved before a recommendation), and 23 CPGs were evaluated as Level C (not recommended).
Discussion
Musculoskeletal conditions/diseases, including arthritis, spondylosis, muscle strain, and other related disorders, are relatively common in the human population. According to World Health Organization, billions of people are afflicted by musculoskeletal disorders, making it the leading cause of disability on a global scale. The implications of musculoskeletal conditions/diseases extend beyond the physical realm and affect individuals’ overall wellbeing and societal engagement. As a traditional but effective complementary alternative medicine, TCM-related therapies should be included and recommended alone or jointly, in high-quality CPGs for the diagnosis and treatment of musculoskeletal conditions/diseases. The AGREE II is an authoritative tool that can objectively assess and comprehensively evaluate the quality of guidelines from multiple dimensions and items.
For the 50 CPGs included in this study, the AGREE II tool evaluation suggested that methodological quality varied largely. The highest scoring domain was “clarity of presentation” (55.36%), and in the process of evaluating, we determined that majority of CPGs had understandable language, clear structure, and obvious format. While the lowest scoring domain was “applicability” (6.04%), which indicated few CPGs pertained to the current barriers and potential facilitators to implementation, strategies to improve uptake, and resource implications of applying the guideline. In other domains, scores generally ranged from 30% to 40%. The inclusion of CPGs was mostly without external review or updated information. Most CPGs based on expert consensus did not explicitly report management of conflict of interest. In addition, CPGs did not mention involvement of patient or public views. Overall, no CPGs were evaluated as a Level A (directly recommended without any changes), 27 CPGs were evaluated as Level B (modified and improved before recommendation), and 23 CPGs were evaluated as a Level C (not recommended). This neglect leads to the opacity of key information, which will make it difficult to implement CPGs in the clinical setting, and ultimately affect the pragmatic effectiveness of TCM for musculoskeletal conditions/diseases.
After evaluating CPGs for musculoskeletal conditions/diseases, focus was expanded to quality evaluation of the guidelines for other conditions/diseases. Through literature research, it was observed that compared with the Quality evaluation of guidelines for endometriosis using AGREE II [57], the CPGs for TCM/IM for conditions/diseases of musculoskeletal scored lower in all domains, which may be because endometriosis is a common condition/disease in clinical practice, and research funding on this conditions/disease is more available or has been carried out earlier, with a more comprehensive and thorough understanding. The quality of the guideline was also relatively high. The range of musculoskeletal conditions/diseases was large, and the quality of guidelines varied greatly, so the average score in each domain was relatively low. Compared with Assessing the Quality of Global Clinical Practice Guidelines on Gout Using AGREE II Instrument [58], the scores of the CPGs included in this current study were low except for “editorial independence,” which indicated that the CPGs involved in traditional Chinese therapies generally exhibited a lower level of quality compared with CPGs that covered a broader range of health care practices. This might be related to the fact that the inclusion criteria of the gout guideline, 1 of musculoskeletal conditions/diseases and with a relatively narrow range, clearly defined the objective and target population. Additionally, the development of western medicine guidelines is relatively mature, the overall quality, therefore, is better. In the future, TCM-related CPGs should aspire to bring transparency to the guideline development process, emphasize the utilization of methodological approaches, and strive for continuous enhancement of guideline quality.
In China, the healthcare landscape encompasses 3 distinct systems: TCM, IM, and western medicine. As a result, multiple sets of guidelines have emerged, and have been utilized in parallel in clinical practice. Through demonstrating the characteristics and evaluating the quality of 50 CPGs, this study delved into the level of cross-reference observed amongst these 3 sets of guidelines during clinical practice.
In terms of the distinctive features, the TCM CPGs primarily embody the TCM principles or related elements through title, which, explicitly reference terms such as “TCM” “combination of condition/disease and pattern” or “acupuncture.” Furthermore, the CPGs’ core content, focus, and recommendations revolve around the TCM concepts of syndrome differentiation and treatment, as well as the utilization of decoctions. In contrast, the IM CPGs, while indicated through their title, do not truly reflect the essence of “integration.” The majority of IM CPGs simply present a parallel listing of traditional Chinese and western medicine approaches without offering a comprehensive amalgamation of their respective practices. The western medicine CPGs typically lack explicit characteristics in title, and when presenting recommendations related to TCM, they merely encapsulate CPGs under the general term of “TCM” but do not recommend a specific TCM therapy or provide a convincing body of evidence.
There are several limitations in this study: (1) Most of the guidelines evaluated were based on publications in journals and some information regarding the development of guidelines may have been missed; (2) In this study, the quality between TCM, IM and WCM guidelines for musculoskeletal conditions/diseases were not compared.. The recommended therapies of TCM in WCM guidelines were mainly focused on non-pharmaceutical guidelines such as tuina, and therapeutic exercises, and these recommendations often lacked support from the evidence. (3) Although AGREE II is a well-recognized tool for evaluation of guidelines, it focuses more on the reporting quality of the guidelines, and the scores cannot fully reflect the methodological quality of CPGs development.
The implications of the findings of this review could be taken as references for future improvement of guidelines of musculoskeletal conditions/diseases for TCM/IM: (1) Before formulating or updating CPGs, the right clinical questions to ask should be clarified so the “scope and purpose” can be clearly defined. (2) A multidisciplinary team should be established, and patient value and preference should be considered in the process of clinical questions, formation of recommendations and external review before release of updated CPGs; (3) Recommendations in CPGs should be supported by best available evidence. Thus, systematic searches and reviews are needed and the use of the Grading of Recommendations, Assessment, Development, and Evaluations method is encouraged for certainty of evidence and recommendations. (4) Recommendations in CPGs for TCM and IM should be in detail to allow implementation by users since TCM/IM are more complex and sometimes require training and qualification to practice. (5) In the process of guideline development, the contents of users and context, external review, funding source, conflict of interest, formation of the recommendations as well as the updating of guidelines should be transparently reported, and with the data being openly accessible.
Supplementary Material
Supplementary material is available at doi: https://doi.org/10.56986/pim.2024.02.002.
Notes
Author Contributions
Conceptualization: JPL. Methodology: JPL and SSM. Study screening and selection: XFW and JLZ. Data extraction: JLZ, LDX, LJL, and XZL. Data analysis: XZL. Quality assessment: XFW, JLZ, LDX, and LJL. Manuscript drafting: XFW and JLZ. Manuscript revision and editing: SSM, JPL, XFW, and JLZ.
Conflicts of Interest
Authors declared no conflict of interest.
Funding
National Natural Science Foundation of China “Key techniques and outcome research for therapeutic effect of TCM as complex intervention based on holistic system and pattern differentiation and prescription” (no.: 81830115).
Ethical Statement
No ethics approval is needed in this study.