Barriers and facilitators to implementing core osteoarthritis treatments in China: a mixed-method study

Abstract

Objective

To understand current practices and identify barriers and facilitators to implementing guideline-recommended core osteoarthritis (OA) treatments in China.

Design

An exploratory mixed-methods design

Setting

Public and private clinical institutions across mainland China.

Participants

A total of 498 healthcare professionals participated. The qualitative phase included semistructured interviews (n=15) and a qualitative survey (n=181). The quantitative phase consisted of a survey with 302 respondents across 19 provinces, representing seven health professions.

Results

Five themes identified as barriers during the qualitative phase: misconceptions about OA; limitations in current medical insurance policies; insufficient multidisciplinary collaboration; lack of workplace support and low patient adherence to self-management. Three themes identified as facilitators: telehealth and community-based delivery pathways; professional training and patient education resources; and personalised services with positive feedback. Quantitative findings showed that physical agent therapy (56%) and traditional Chinese medicine (22%) were the most frequently used OA treatments, while exercise therapy was implemented in only 9% of cases. The average OA knowledge score was 31.2 (±8.9) out of 55, with the lowest self-rated confidence in interdisciplinary collaboration (3.4±0.1, ‘somewhat confident’). The most applicable factors impacting the implementation of core OA treatments included patient comorbidities, knowledge of pain science and exercise therapy, and financial support (all 2.8±0.8, ‘applicable’).

Conclusions

The uptake of core OA treatments in China remains suboptimal, constrained by limitations in insurance coverage, workforce capacity and interdisciplinary integration. Enhancing telehealth accessibility, strengthening professional training and refining policy incentives may help bridge this evidence-practice gap and improve OA management in China.

Strengths and limitations of this study.

• The study employs a mixed-methods design, integrating qualitative and quantitative approaches to provide comprehensive insights into the unique challenges within China’s healthcare system, which can inform locally relevant interventions.

• The inclusion of various healthcare professionals, including rehabilitation specialists, general practitioners and traditional Chinese medicine practitioners, enhances the generalisability of the findings and ensures diverse perspectives on osteoarthritis management.

• The study highlights China-specific healthcare system challenges, offering novel insights not covered in international literature.

• Limited representation of certain specialties, such as orthopaedic surgeons, may affect the comprehensiveness of the clinical perspectives captured.

Introduction

Osteoarthritis (OA) is a multifactorial joint disease that is highly prevalent worldwide, affecting 133 million people in China as of 2019.¹ It imposes both individual burden, such as pain and physical dysfunction, and socioeconomic burden, including healthcare costs and workforce loss.^{2 3} By 2044, the OA burden in China is projected to increase 1.5 times compared with 2019, driven primarily by an ageing population, rising obesity rates and the country’s large population size.^{4 5} Currently, there is no cure for OA.⁶ Addressing this major public health challenge in China is therefore crucial,⁷ with treatment priorities focusing on symptom management rather than pathophysiological joint changes.⁸ High-quality international clinical guidelines, including those from the USA, Australia, UK, Europe and leading research societies, consistently recommend exercise, education and weight management (where appropriate) as core treatments for all individuals with OA, regardless of age, pain severity or disease progression.9,14 However, even in high-income countries, OA care remains suboptimal,15,17 and implementing evidence-based OA care in low-income and middle-income countries presents even greater challenges.¹⁸

The successful translation of high-quality evidence into clinical practice requires a comprehensive understanding of current practice patterns, as well as the barriers and facilitators encountered by stakeholders in specific settings, to develop effective implementation strategies.¹⁹ In recent years, implementation research has also emerged as a key approach to enhancing the adoption of guideline-recommended OA treatments in real-world clinical settings based on certain frameworks,^{20 21} such as creating knowledge toolkits,^{22 23} identifying factors that influence implementation24,28 or developing evidence-based models of OA care.²⁹ While many countries have made progress in this area, China remains in an early, foundational stage. Addressing this gap is particularly important given the unique structure of the Chinese healthcare system and the controversial recommendations in its national OA clinical guidelines.³⁰ For example, the 2024 edition of the Chinese OA clinical guideline³¹ adopts a stage-based treatment approach, recommending those core treatments only for patients in the early stages of OA with mild symptoms, which contrasts with international guidelines. Additionally, China’s healthcare system differs from many Western systems, where specialist access typically requires a referral from a primary care provider. In China, patients can directly choose their first point of care, registering to see specialists at any hospital without a referral or opting to visit generalist doctors at community health centres (CHCs) for primary care.³² Furthermore, allied health services such as physiotherapy are not well-established or integrated into China’s healthcare system.^{33 34} For example, the term physiotherapist is often used interchangeably with rehabilitation therapist, creating confusion regarding their specific roles and qualifications. This lack of clarity and limited integration may hinder the widespread adoption of exercise-based interventions recommended in OA guidelines.³⁵

Given these challenges, this study aims to: (1) identify the barriers and facilitators experienced by Chinese healthcare professionals in implementing core OA treatments and (2) understand the current clinical practices in OA management within the Chinese clinical context.

Study design

This study employed a mixed-methods design to investigate the barriers and facilitators experienced by Chinese healthcare professionals in implementing core OA treatments and to understand their current clinical practices in OA management. Mixed-methods research is particularly suited to these aims, as it provides both depth and breadth of understanding.³⁶ An exploratory sequential approach³⁷ was selected, considering the limited literature on OA care delivery in the Chinese clinical context and the approach’s comparatively more robust validity.³⁸ This design also allowed for the iterative exploration of key themes and the validation of findings across a broader population. This study proceeded in three phases: initial semi-structured interviews, a qualitative survey with open-ended questions, and a final quantitative survey evaluation. A visual summary of the sequential study design is presented in figure 1. This study follows the Good Reporting of A Mixed Methods Study Checklist.³⁹

Figure 1. Flow chart of the sequential mixed-methods study design. GPs, general practitioners; OA, osteoarthritis; OAKS, Osteoarthritis Knowledge Scale; TDF, Theoretical Domains Framework.

Participants

Participants in this study were selected based on the healthcare professionals targeted by Chinese clinical guidelines for OA. To ensure representation of diverse specialties contributing to OA care, healthcare professionals managing OA across mainland China were recruited without limitations on specific disciplines or institutions. Demographic variables of participants were collected for all phases, including the type of medical institution (public or private), professional specialty, years of professional experience, OA-related certifications or qualifications, and other relevant information. The detailed selection process for participants in each phase is summarised below.

Qualitative phase 1

This phase aimed to initially explore the barriers and facilitators to implementing exercise therapy for patients with OA.

Participants: Participants at this phase were recruited through professional networks, including direct email invitations, messages shared in professional WeChat groups, and personal contacts, between September 2023 and November 2023. Inclusion criteria required participants to have a minimum of 2 years of clinical experience in OA-related care and without conflict of interests with the research group. Potential participants were screened via phone call to confirm their professional background and the frequency with which they managed OA patients. During the recruitment, we confirmed that all participants regularly treated OA patients, with most reporting a caseload of at least one OA patient per week. We purposively recruited healthcare professionals with diverse roles, geographical locations and institutional affiliations to ensure variation in clinical experience and practice settings. There were no further exclusion criteria.

Data collection: For phase 1, we approached 20 clinicians and of these, 15 agreed to participate, 5 declined due to scheduling conflicts. Data saturation was achieved after the 13th interview, with two additional interviews confirming no new themes. The interviews were guided by the Theoretical Domains Framework (TDF), which offers a structured approach to examining behavioural determinants in healthcare,⁴⁰ with minor adaptations to question wording to better reflect the context of OA management in China. Interview questions were designed to elicit in-depth responses on participants’ experiences with implementing evidence-based OA practices, barriers encountered and strategies perceived as facilitators in their clinical settings. The interview guide was included in the online supplemental material 1 and was reviewed by two OA experts and one qualitative research specialist to ensure content validity, cultural appropriateness and alignment with the TDF. Individual interviews were conducted by a trained qualitative researcher (ZW), either in person or via online meetings depending on availability. Interviews were recorded and transcribed verbatim by ZW. Field notes were taken during and after the interviews to capture contextual details. Each interview lasted approximately 40 min.

Data analysis: We analysed the data using framework analysis,⁴¹ facilitating the systematic organisation of data into themes that reflected participants’ clinical experiences and perspectives. Two researchers (XC and SD) independently coded the transcripts using NVivo software (V.14). Coding consistency was maintained through iterative comparison of codes, and any discrepancies were resolved through discussion until agreement was reached, with arbitration by a third researcher (ZW) when necessary. ZW mapped the finalised themes deductively to the domains of the TDF. Recruitment continued until data saturation was achieved, defined as the point at which no new themes were identified in subsequent interviews. To minimise potential bias from the research team’s clinical backgrounds, all coding decisions and theme interpretations were discussed collaboratively to ensure interpretations remained grounded in participants’ narratives.

Qualitative phase 2

This phase aimed to validate and complement the findings from individual interviews, and to inform the design of further quantitative survey.

Participants: Participants for this phase were recruited through flyers with QR code linking to the survey, which were posted in professional WeChat groups for healthcare practitioners, public accounts of national professional societies (eg, rehabilitation, geriatric, orthopaedics), and networks across hospitals, community and private sectors from November 2023 to January 2024. The sample size for this phase was determined using the principle of data saturation, whereby recruitment ceased when no new themes or patterns were identified from the open-ended responses.

Data collection: An online qualitative survey with open-ended questions was conducted to expand the sample size and capture perspectives from a broader group of practitioners across various disciplines in China. Exercise-focused items were prioritised in the survey because phase 1 identified exercise implementation as the most prominent gap in core OA treatments. In addition to demographic questions, participants were asked to respond to several key questions: (1) ‘What factors hinder your delivery of exercise-based therapies for your OA patients?’; (2) ‘What factors facilitate your delivery of exercise-based therapies for your OA patients?’ These were followed by a non-mandatory question: (3) ‘If you do not agree with exercise-based therapies, please provide the reasons behind your decision’. Lastly, participants were asked: (4) ‘What are your usual treatment options for your OA patients?’ and (5) ‘What are your main treatment principles/goals when selecting treatments for your OA patients?’ Participants were required to provide their answers in written text. In phase 2, around 300 practitioners received the qualitative survey invitation, with 181 completing it (response rate 60%).

Data analysis: Two researchers (XC and SD) independently conducted the analysis using summative and deductive content analysis. Summative content analysis was used to quantify the frequency of key terms and patterns within responses, while deductive content analysis involved categorising responses into predefined themes derived from the research questions and theoretical frameworks. This combined approach ensured both thematic frequency and theoretical depth, which are suitable for analysing large amounts of open-ended survey data.^{42 43} Summative content analysis identified and quantified key terms and patterns related to barriers, facilitators and treatment options for OA. Deductive content analysis then categorised responses into predefined themes derived from the research questions and theoretical frameworks. NVivo software was used to manage and organise data, and discrepancies were resolved through discussion. A senior researcher (ZW) reviewed the final themes to ensure validity and reliability.

Quantitative phase

This phase aimed to evaluate the applicability of the themes and common practices identified in the qualitative phase by quantifying the most frequently mentioned themes and the most common OA practices in clinical settings.

Participants: Participants in this phase were recruited using a combination of convenience and snowball sampling from February 2024 to May 2024. The sample size was determined based on descriptive results reported in a similar international study,²⁵ which reported barriers related to knowledge and skills with proportions ranging from 37% to 88%. At a 95% confidence level, a minimum sample size of 141 was calculated for adequate precision. To allow for subgroup analyses and greater accuracy, a more conservative estimate with a maximum variability of 50% and a 5% margin of error resulted in a required sample size of up to 384 participants.

Data collection: A structured survey was developed based on the themes and subthemes identified in the qualitative phases and piloted with 20 healthcare professionals who previously participated in a local in-person workshop. Each survey item corresponded to a domain or theme from phases 1 and 2, ensuring that the quantitative data would validate and expand on the qualitative findings. Feedback from this pilot phase led to minor revisions, including rewording three items for clarity, adapting terminology to align with Chinese clinical practice and adjusting examples to improve cultural relevance (eg, references to community healthcare settings). Internal consistency for key multi-item scales, such as the Chinese Osteoarthritis Knowledge Scale (OAKS), was acceptable (Cronbach’s α=0.70).⁴⁴ For phase 3, the quantitative survey was sent to about 380 clinicians, with 302 valid responses collected (response rate ~79%). No exclusion criteria were applied beyond incomplete responses. The final survey (see online supplemental material 2) was administered via the ‘Wenjuanxing’ online platform. The survey primarily included the following measures:

1. Bespoke questions about core OA treatment usage: Respondents answered four questions using a 5-point Likert scale to indicate the frequency of implementing exercise, education, weight management and support for self-health management. The response options were: ‘never’, ‘occasionally’, ‘half of the time’, ‘often’ and ‘always’. Higher scores for each question indicate a higher frequency of implementation, with scores calculated separately for each question.

2. Factors impacting the implementation of best practices: Respondents evaluated the applicability of factors impacting best practice implementation using a matrix question with four response options: ‘not applicable’, ‘somewhat applicable’, ‘applicable’ and ‘highly applicable’. These factors included time allocation, team-based care, resource availability, guideline accessibility and patient-related factors, among others.

3. Chinese version of the OAKS⁴⁴: This scale assesses respondents’ knowledge of OA management. It consists of 11 items, each rated on a 5-point Likert scale: ‘false’, ‘possibly false’, ‘unsure’, ‘possibly true’ and ‘true’. The total score ranges from 11 to 55, with higher scores indicating greater knowledge about OA.

4. Items evaluating self-reported confidence in OA care: Based on the OA core capability framework,⁴⁵ respondents rated their confidence in 13 specific capabilities using a 5-point Likert scale: ‘not confident at all’ (1 point), ‘not very confident’ (2 points), ‘somewhat confident’ (3 points), ‘confident’ (4 points) and ‘very confident’ (5 points).

Data analysis: Data were analysed using RStudio (V.4.2.0). Descriptive statistics were applied to summarise demographic characteristics, frequencies and proportions of the quantitative responses. All data were anonymised, with informed consent provided on the opening page, and securely stored, and accessible only to the research team.

Results

The detailed demographic characteristics, including their professional backgrounds of the participants for all three phases, are summarised in table 1. Detailed individual-level demographic characteristics for phase 1 participants (semistructured interviews, n=15) are presented in online supplemental material 3.

Table 1. Characteristics of participants.

Demographic characteristics	Semistructured interview (N=15)	Qualitative survey (N=181)	Quantitative survey (N=302)
Average age, years (SD)	42.5 (6.3)	36.5 (7.8)	35.3 (9.4)
Gender, n (%)	Female: 6 (40%) Male: 9 (60%)	Female: 93 (51%) Male: 88 (49%)	Female: 156 (52%) Male: 146 (48%)
Profession, n (%)	Rehabilitation medicine specialists: 6 (40%) General practitioners: 3 (20%) Traditional Chinese medicine specialists: 3 (20%) Sport medicine specialists: 3 (20%)	Rehabilitation medicine specialists: 94 (52%) Traditional Chinese medicine specialists: 29 (16 %) Sport medicine specialists: 19 (11%) Health managers: 14 (8%) Orthopaedist: 14 (8%) General practitioners: 11 (6%)	Rehabilitation medicine specialists: 150 (50%) General practitioners: 66 (22%) Traditional Chinese medicine specialists: 37 (12%) Health managers: 14 (5%) Sports and fitness practitioner: 14 (5%) Orthopaedist: 11 (4%) Sport medicine specialists: 10 (3%)
Primary site of clinical practice, n (%)	Public: 8 (53%) Private: 7 (47%)	Public: 160 (88%) Private: 16 (9%) Both: 5 (3%)	Public: 274 (91%) Private: 23 (8%) Both: 5 (2%)
Have qualification certificate, n (%)	15 (100%)	39 (22%)	98 (33%)
Year range of clinical OA practice, n (%)	1–5 years: 8 (53%) 6–10 years: 7 (47%)	Less than 1 year: 18 (10%) 1–5 years: 53 (30%) 6–10 years: 34 (19%) >10 years: 76 (42%)	Less than 1 year: 82 (27%) 1–5 years: 81 (27%) 6–10 years: 47 (16%) >10 years: 92 (31%)
Education level, n (%)	Bachelor degree: 6 (40%)	College diploma: 13 (7%)	College diploma: 49 (16%)
	Master degree: 7 (47%)	Bachelor degree: 114 (63%)	Bachelor degree: 169 (56%)
	Doctoral degree: 2 (13%)	Master degree: 50 (28%)	Master degree: 77 (26%)
		Doctoral degree: 4 (2%)	Doctoral degree: 7 (2%)

OA, osteoarthritis.

Qualitative phase 1

In this initial phase, four themes were identified as barriers and five as facilitators using framework analysis, with the detailed findings presented according to the major TDF domains.

Barriers

1. Knowledge gaps and limited awareness of updated evidence

Participants frequently reported challenges in accessing and integrating evidence-based guidelines. Many described a reliance on traditional practices or personal experiences due to a lack of formal training and regular updates. This issue was particularly pronounced in resource-limited settings.

Guidelines are helpful, but in real practice, I rely more on my own experience and what patients respond to. (GP1)

In our practice, we seldom refer to modern OA guidelines; our training is based more on traditional theories and senior teacher mentoring. (TCM1)

• 2 iResource constraints and systemic limitations

Insufficient time, funding and infrastructure were recurring themes that limited the delivery of evidence-based care. Participants highlighted issues such as low reimbursement rates for exercise therapies and a lack of dedicated rehabilitation facilities.

Our rehabilitation staff are insufficient, facilities and equipment are lacking, communication time with patients is limited, and everyone is exhausted (RM1)

Rehabilitation requires investment in time and equipment, but financial constraints mean we often have to rely on quick fixes, like medication or injections, that are reimbursed (RM2)

• 3 iScepticism and resistance to change among colleagues

Resistance to exercise-based interventions, particularly among surgical and pharmaceutical-focused health professionals, was identified as a significant barrier. This scepticism often stemmed from a preference for quicker or more invasive solutions.

Convincing other doctors of the value of exercise is tough; surgery is often seen as the ultimate solution and patients referred to our department are mostly at the end-stage (GP2)

Challenging traditional concepts takes time, and inherent habits create obstacles during implementation—for both doctors and the general public (TCM2)

• 4 iPatient misconceptions and adherence challenges

Misunderstanding the role of exercise in OA management was a significant patient-level barrier. Participants noted that many patients associated exercise with potential joint damage or pain, leading to poor adherence to prescribed therapies.

Patients often stop exercising as soon as they feel discomfort, thinking they’re doing more harm than good. (GP1)

Discomfort during exercise is often misunderstood; people believe that lying still and avoiding movement is better because pain means something is wrong (SM2)

Facilitators

1. Continuous education and professional training

Structured training programmes and peer mentoring were recognised as essential for building confidence and competence in implementing evidence-based practices.

Attending workshops and learning from colleagues has helped me incorporate evidence-based strategies into my daily practice. (GP2)

Most TCM practitioners rarely get systematic training on OA rehabilitation, and I feel more inspired after attending some joint sessions with other doctors. (TCM3)

• 2 iTechnology and digital tools

The integration of telehealth platforms, mobile applications and exercise videos was highlighted as a valuable facilitator, particularly in under-resourced settings. These tools enhanced patient education and adherence by providing accessible and visual guidance to improve patient understanding of exercise programmes.

Apps and demo-videos make it much easier for me to explain exercises and track patients progress remotely and efficiently. (RM3)

When I use WeChat to follow-up patients, patients seem more willing to adhere to the plan. (SM3)

• 3 iInstitutional and policy support

Support from healthcare institutions, such as prioritising rehabilitation services and providing financial incentives for non-surgical interventions, was seen as a crucial facilitator.

When our institution supports exercise therapy, especially when leadership decision-making groups recognize the value of rehabilitation, it makes our job easier. (GP3)

Government policy on OA care have encouraged hospitals to allocate more resources to rehabilitation services. (RM4)

• 4 iCollaborative teamwork

Multidisciplinary collaboration was frequently cited as a facilitator for implementing OA care. Teams that included physicians, therapists and support staff were more likely to achieve consistent and effective care delivery.

Working as a team ensures we can align our goals and provide better care to patients. (GP1)

When nurses, doctors, and rehab specialists all share feedback, the patient’s rehabilitation pathway becomes smoother. (RM6)

• 5 iPersonalised communication and good clinician-patient alliance

Effective communication strategies are considered key drivers in improving patient adherence and treatment outcomes. Participants emphasised the importance of tailoring the content of information according to the needs of individual patients, using simple, empowering language and success stories to motivate patients and strengthen the doctor-patient alliance.

I always tell patients that having arthritis doesn’t mean they have to stop being active; instead, they need to find the right way to stay active. When they realize that moderate exercise can improve their quality of life, not just reduce pain, they are more willing to stick with the treatment. (RM5)

Patients trust you more when you listen to their concerns and show them real cases where small lifestyle changes made a big difference. (TCM3)

Mapping findings to TDF domains

The identified barriers and facilitators were organised into themes based on the relevant TDF domains to ensure a structured analysis. Table 2 presents examples of barriers and facilitators mapped to each domain.

Table 2. Examples of barriers and facilitators mapped to the TDF from semistructured interview responses.

Domain	Barriers	Facilitators
Knowledge	Limited awareness of updated evidence-based guidelines among practitioners.	Access to continuous education and workshops to bridge gaps in guideline knowledge
	Patients misunderstand the role of exercise, associating it with harm or ineffectiveness	Use of visual tools like videos and diagrams to simplify patient education
Skills	Lack of practical training opportunities in multidisciplinary settings	Leveraging digital tools (eg, apps) to enhance skill acquisition and patient guidance
Social/professional role	Disjointed collaboration between physicians and therapists; differing views on care priorities	Regular team meetings and aligned goals within multidisciplinary teams
Beliefs about capabilities	Low confidence in addressing comorbid conditions and tailoring interventions to individuals	Peer mentorship and case discussions to boost confidence in applying evidence-based interventions
Optimism	Patient scepticism about exercise’s effectiveness, especially with slow progress or initial pain	Clear communication of realistic benefits and managing patient expectations effectively
Beliefs about consequences	Fear of patient injury or exacerbation of symptoms due to inappropriate exercise protocols	Provision of scientific evidence showing the long-term benefits of exercise in improving joint function and reducing surgical interventions
Reinforcement	Reliance on traditional treatments (eg, massage or medication) over evidence-based practices	Other patients’ success stories and improved clinical outcomes to motivate patients
Intentions	Challenges in integrating exercise-based strategies into existing workflows	Seamless incorporation of exercise into multidisciplinary care plans
Environmental context	Limited infrastructure and resource allocation for rehabilitation, including insufficient funding and staff	Investment in telehealth platforms and advanced rehabilitation equipment
Social influences	Cultural norms favouring passive treatments like acupuncture over active rehabilitation approaches	Public awareness campaigns and education to shift societal perceptions of exercise
Emotion	Practitioners’ fear of patient complaints due to perceived lack of improvement or discomfort during exercise	Open dialogue with patients to address concerns and build trust, reducing resistance
Behavioural regulation	Inconsistent application of monitoring strategies to ensure patient adherence	Structured follow-up systems to maintain adherence to exercise plans

TDF, Theoretical Domains Framework.

Qualitative phase 2

Identified barriers and facilitators

The qualitative survey (n=181) largely validated the barriers and facilitators identified in phase 1, while also providing additional insights. For instance, within the ‘Beliefs About Consequences’ domain, participants highlighted that patients’ unrealistic high expectations regarding the speed of symptom relief often led to frustration and early discontinuation of exercise. This finding extended the earlier observations about misconceptions related to exercise. For the ‘Environmental Context’ domain, participants identified a lack of localised easily understood educational resources tailored to rural populations with low education levels, which exacerbated existing disparities in access to evidence-based care. Within the ‘Social/Professional Role’ domain, participants emphasised the importance of clearer task/responsibility allocation and more defined roles within multidisciplinary teams as a facilitator. Additionally, within the ‘Reinforcement’ domain, respondents highlighted the value of incorporating real-world examples of successful delivery to motivate and inspire both patients and healthcare professionals.

Synthesising the findings from phase 1 and phase 2, five themes were identified as key barriers based on their frequency and relevance. These included widespread misconceptions about OA among patients, healthcare professionals and society at large; limitations in medical insurance policies for non-pharmaceutical services alongside broader economic constraints; insufficient multidisciplinary collaboration with unclear task allocation; a lack of workplace support; and low patient adherence to self-management strategies. Similarly, three themes were identified as facilitators. These included the potential of telehealth and community-based delivery models, the availability of professional training and patient education resources, and the implementation of personalised services supported by positive delivery feedback.

Themes of OA treatment goals and usual care practices

Analysis of the text-based responses identified five frequently mentioned themes representing the goals of OA treatment. These themes include pain management, aimed at alleviating patient discomfort; functional recovery and muscle enhancement to improve joint mobility and strength; repair and nourishment techniques focused on cartilage and joint health; anti-inflammatory and swelling reduction strategies; and maintaining overall health through behavioural and lifestyle interventions.

Usual care for OA was also categorised into five main treatment approaches based on the responses. These included physical modality therapies, such as electrotherapy, ultrasound and similar modalities; TCM practices, including acupuncture and herbal treatments; exercise therapies, emphasising joint mobilisation, functional training and strengthening exercises; pharmacological treatments, such as non-steroidal anti-inflammatory drugs and glucosamine; and injection therapies, including corticosteroids and platelet-rich plasma.

Quantitative findings

Practice of OA management

Participants from 19 provinces in China completed the quantitative survey, with Fujian province accounting for half of the total respondents. As shown in table 1, only 33% of participants completing the survey held a relevant professional healthcare qualification. Among these, 70% held a specialist certification, while the remaining 30% held a physician’s licence. The most commonly used OA treatments were physical agent therapy (56%) and TCM (22%), while exercise therapy was used in only 9% of cases. Pharmaceutical treatments and injections accounted for 8% and 4%, respectively.

The most frequently reported frequency of managing OA patients in clinical practice was reported as ‘often/at least once per week’. More than half of the participants reported a delivery frequency of no more than ‘half of the time’ for exercise/physical activity (73.2%), patient education (65.5%) and self-management support (68.2%). The average OA knowledge score was 31.2 (SD=8.9) out of 55, with 96% of the participants believing that joints wear down with daily use (mean score=2.1, out of 5), while the role of scans in diagnosing OA was rated the lowest (mean score=2.0). Among the 13 domains of the OARSI core capability framework, self-rated confidence was lowest for ‘referrals and interdisciplinary collaboration’ (mean score=3.4, ‘somewhat confident’), with none of the domains reaching the level of 4 (‘confident’).

Most common factors impacting implementation

Guided by themes and codes from the qualitative results, we identified the most relevant factors that participants believe impact the implementation of core OA care in their practice, using a 4-point Likert scale (ranging from ‘not at all applicable’ to ‘highly applicable’). Table 3 presents the average rating of each factor, sorted in descending order from high to low.

Table 3. Mean scores of factors influencing the implementation of guideline-recommended core OA treatments.

Factors	Mean	SD
Patient comorbidities	2.78	0.81
Patient understanding of pain management	2.76	0.79
Patient beliefs and attitudes towards exercise therapy	2.75	0.81
Financial incentives for best practices at work	2.75	0.84
Health insurance support for patients	2.74	0.88
Referral systems in the multidiscipline networks	2.74	0.84
Patient accessibility to core OA care	2.72	0.80
Patient social support	2.72	0.78
Information consistency among colleagues	2.65	0.82
Consultation time allocation	2.69	0.85
Institutional value of core treatments	2.58	0.91
Working environment resources availability	2.58	0.87
Professional training and knowledge updates	2.58	0.85
Availability of both paper and online educational resources	2.52	0.80
Patient personal treatment preferences	2.50	0.82
Teamwork in the working environment	2.39	0.89
Skill application for implementing guideline recommended care	2.39	0.81
Guideline accessibility	2.37	0.93
Alignment of professional training with guidelines	2.25	0.85
Understanding of the guideline	2.15	0.89
Practical relevance of guidelines	2.11	0.97
Colleague practices	1.84	0.93

OA, osteoarthritis.

Many factors were rated highly including: patient comorbidities (such as obesity, mental health problems and other chronic diseases) (2.8±0.8); patient knowledge of pain science (2.8±0.8) and exercise therapy (2.8±0.8); financial support for both practitioners (2.8±0.8) and patients (2.7±0.9); multidiscipline networks and referral systems (2.7); patient access to appropriate care and support (2.7) and the time available for each consultation (2.7). Personal bias towards guidelines (1.9) and practices of other working colleagues (1.8) was rated as less applicable.

The quantitative results demonstrated strong alignment with the qualitative results. For example, patient misconceptions, resource constraints and insufficient multidisciplinary collaboration—initially identified in interviews—were also among the highest-rated factors in the quantitative survey. Similarly, facilitators such as professional training and telehealth solutions were consistently supported by both qualitative and quantitative findings.

Discussion

This study is the first to identify Chinese healthcare professionals’ views on the barriers and facilitators to implementing core OA treatments (exercise, patient education and weight management) through a mixed-methods design. It also explored usual OA clinical practices in Chinese healthcare settings and professionals’ confidence levels in relevant core capabilities. The findings provide a solid basis for informing further interventions to optimise high-value OA health service delivery in China.

Gaps in current OA care practices

Similar to findings from other countries worldwide,^{16 46 47} OA management in China remains suboptimal, with few healthcare professionals actively delivering first-line core treatments. Instead, physical modality therapy (eg, electrotherapy) and TCM were the most commonly used approaches among the participants in our study. A report published in 2013⁴⁸ highlighted that physical therapy in China was dominated by these two treatments at that time, suggesting little change in OA clinical practice over this time. A similar trend of little change over a decade was observed among Australian general practitioners (GPs).⁴⁶ Our findings also revealed that most participants, including many rehabilitation specialists, lacked confidence in their ability to deliver recommended OA treatments. In contrast, a multinational (Australia, New Zealand, Canada) interprofessional study⁴⁹ found that all disciplines, except nurses, reported moderate levels of confidence. This discrepancy may highlight a gap in confidence among our participants. A validated OA Knowledge questionnaire revealed a low level of knowledge in our sample of healthcare professionals in China, showing certain bias and misconceptions towards OA concepts and management that have also been observed in other countries.⁵⁰ Given existing evidence-practice gaps in current OA practices in China, it is important to develop strategies to optimise the timely use of core treatments, by addressing the barriers and facilitators identified.

Implications of identified barriers and facilitators

Our exploratory sequential design demonstrated convergence across phases. Themes from the qualitative phases informed the development of the quantitative survey items, while quantitative results provided back-validation and a measure of the prevalence of these barriers and facilitators. This iterative approach ensured that the findings were both grounded in real-world experiences and generalisable across a larger sample. The barriers and facilitators identified in this study are mainly consistent with those observed in other countries, which typically fall into three meta-themes: personal factors (including both patients and healthcare professionals, knowledge/skills), workforce environment (resources and training), and system-level factors (society and policy).^{24 25 51} For example, the most frequently mentioned and common barrier identified in this study was the widespread misconceptions about OA and exercise therapy among society, patients and healthcare professionals, which is a common issue worldwide. Therefore, China should develop knowledge dissemination initiatives, such as a toolkit⁵² with essential components and practical resources for healthcare professionals and patients.

Environment-level barriers identified in this study are shaped by China’s unique healthcare context. First, restrictions in medical insurance policies for non-pharmaceutical services have led to insufficient funding support for rehabilitation within the healthcare system. The identified issue of multidisciplinary collaboration by our participants is also tied to the limited influence of rehabilitation departments in public hospitals, making referral systems and the role of first assessors/first contact practitioners focused on conservative treatments less effective. This aligns somewhat with a study⁵³ in Australia, which found that the likelihood of GPs referring patients to physiotherapists is low for certain health conditions. As the first assessor often plays a central role in managing OA,³⁵ optimising referrals to rehabilitation-led care could be an important way to promote better patient health outcomes. In public hospitals, exercise-based services bring minimal financial incentives, thus rehabilitation departments focus more on TCM than on exercise interventions, which makes organisational leadership support for exercise-based therapies essential.

Facilitators identified in this study highlight the potential importance of telehealth and CHCs. As CHCs have gradually become major primary care providers in China, and OA is a complex chronic health condition requiring long-term self-management support, developing a community-based model of OA care in the future could help alleviate outpatient workload stress in public hospitals.⁵⁴ Regarding the use of telehealth in OA care, recent evidence⁵⁵ has shown that telerehabilitation with a physiotherapist can be as effective as in-person care for the non-surgical management of OA. Improving access to core treatments is particularly important, as there is a persistent shortage of rehabilitation specialists in China, and the uneven distribution of rehabilitation resources remains a significant challenge.³³ Most rehabilitation resources are concentrated in large cities and hospitals, while rural and underdeveloped areas lack sufficient rehabilitation services, even though symptomatic knee OA is more common in rural China.⁵⁶ Telehealth, therefore, has the potential to bridge this gap and make core treatments more accessible to underserved populations. To facilitate its implementation, relevant training programmes should be continuously provided for Chinese to upskill their telehealth knowledge and practices.

This study has several limitations. First, due to the network recruitment process, participants in the quantitative phase were primarily rehabilitation doctors and therapists from Fujian province, which may introduce certain perspective biases, as surgeons or orthopaedic doctors might hold different views. Additionally, Fujian, particularly its capital city Xiamen, is a sports-friendly city with strong government support for exercise, which could further influence responses. However, in the preceding qualitative phase, participants were more evenly distributed. Second, different healthcare work environments may present distinct barriers, potentially influencing participants’ beliefs. Since we did not collect detailed information about their specific workplaces, we cannot ensure that the participants’ work settings were evenly represented. Third, this study focused primarily on the management of OA, without addressing other chronic diseases or cases with multiple comorbidities. Finally, we did not collect cost-related economic data in this study, as most participants lacked access to their institutions’ health insurance systems. This aspect will be targeted in future interventional studies.

Conclusions

Core OA treatments are infrequently implemented in China, reflecting significant evidence-practice gaps. Targeted efforts are needed to strengthen interdisciplinary collaboration, develop adaptive Chinese OA educational resources for all stakeholders, and provide systematic workplace support for practitioners. Improving accessibility through innovative approaches such as telehealth and community-based care is particularly critical for addressing disparities in resource-limited areas. Future interventions addressing these gaps will optimise evidence-based OA care delivery and improve patient outcomes in China.

Data availability statement

Data are available on reasonable request.