download PDF
Introduction
While many rheumatic diseases are not immediately life-threatening, their symptoms and progression can significantly impair a patient's quality of life1,2. Because these patients first present to primary care settings, early recognition and appropriate initial assessment by generalists are essential. This is especially true in remote areas, where delayed diagnosis is common due to limited specialist access, often resulting in suboptimal patient care1. This issue will likely worsen in the coming years as the workforce shortage of rheumatologists coincides with an increasing number of musculoskeletal diseases in the aging population3. Addressing these challenges requires not only improving healthcare access but also ensuring that physicians working in such settings possess the competencies required to manage rheumatic diseases effectively.
Competency-based medical education (CBME) has long served as the framework for medical curricula. However, translating these theoretical competencies into reliable workplace assessments remains a significant hurdle4. To bridge this gap, the concept of entrustable professional activities (EPAs) has emerged. EPAs represent discrete units of professional practice that can be fully entrusted to a trainee once competence is demonstrated, thereby making clinical objectives both explicit and measurable. Crucially, EPA-based assessments offer a logistical advantage for remote or busy clinical settings, as they integrate into standard practice without requiring excessive time or additional personnel5.
Despite the growing adoption of EPA-based frameworks, their application in rheumatology remains limited. In the field of rheumatology, resources such as the Accreditation Council of Graduate Medical Education (ACGME) Milestones and The European Alliance of Associations for Rheumatology (EULAR) training curriculum exist6,7. However, these were developed specifically for the education of rheumatology fellows, and EPAs for general physicians managing rheumatic disease have not yet been defined8. In 2025, the Japan College of Rheumatology (JCR) recommended hybrid telemedicine models for rheumatic diseases, underscoring the need to define competencies for general physicians who coordinate care in remote areas9.
To address this educational void in remote health care, we aimed to develop a consensus-based list of essential tasks for young physicians, structured as EPAs, to enhance the quality of rheumatic care in underserved regions.
Methods
Study design
We employed a modified Delphi method to facilitate consensus building10. The Delphi method is a systematic, iterative process used to reach collective opinion among experts through rounds of anonymous feedback. Unlike the classical Delphi method, which begins with open-ended opinion gathering, the modified version utilizes a pre-developed initial list based on existing literature or preliminary research, which can improve the efficiency of the consensus process11(Supplementary table 1). In this study, the number of rounds was not predetermined, allowing for flexibility until a robust consensus was achieved.
The initial task list was synthesized through internal deliberations within the research team. We drew upon several key frameworks, including the Japan Model Core Curriculum for Medical Education (2022 edition), and the specialist learning objectives from JCR, ACGME, EULAR, and the Japanese Society of Internal Medicine6,7,12-15. The research team consisted of rheumatologists and medical education experts active at Okinawa Chubu Hospital. To ensure methodological rigor, the team included members with specific expertise in EPA development and research, addressing the known requirement for specialized knowledge in creating functional EPA frameworks16.
Context
In Japan, following 6 years of undergraduate study, physicians undergo a mandatory 2-year postgraduate residency12. Okinawa Chubu Hospital was originally established under the guidance of American general and academic physicians after World War 2 and has a longstanding mission to train professionals capable of serving on remote islands and in rural areas. The residency program provides comprehensive outpatient experience in the management of rheumatic diseases. Within this specific community-based hospital context, we identified the clinical tasks most relevant to remote practice and developed a set of EPAs targeted at internal medicine physicians in their fifth postgraduate year.
Expert panel selection and data collection
Twenty-five expert panel members were selected by purposive sampling between September 2023 and March 2024. While no universal standard exists for panel size, a minimum of 20 experts is generally considered sufficient for reliability; we recruited 25 to account for potential attrition17. Experts were defined as clinicians either actively engaged in rheumatology or experienced in primary care at a specialist level. Questionnaires were administered via Google Forms. Using a five-point Likert scale, panelists rated each task based on its importance for general practitioners in remote areas (1 = ‘definitely exclude’ to 5 = ‘definitely include’). We also provided space for qualitative feedback, allowing experts to suggest modifications or additional tasks. After each round, the research team refined the list by retaining items with high agreement and incorporating thematic suggestions. Rounds continued until all items met our pre-defined consensus criteria.
Consensus standards
Recognizing that the expert panel members might not be entirely familiar with the concept of EPAs, explanatory materials were distributed to aid their understanding. Consensus for each item was defined as meeting both of the following statistical thresholds according to the previous report for consensus criteria18:
- average score of ≥4.0 with a standard deviation of <1.0
- at least 75% of respondents rate the item as 4 or 5.
Ethics approval
This study was approved by the Ethics Committee of Okinawa Chubu Hospital (approval no. 2023-9). All participants provided informed consent prior to their involvement.
Results
Expert panel characteristics
Of the 25 expert panel members, 12 were rheumatology specialists, 15 were general internal medicine specialists, and 10 were family medicine specialists (Table 1). Twenty-one participants had direct experience in remote area medicine, while four had no such experience (Table 2).
Table 1: Characteristics of expert panel members
| Specialty | n (%)† |
|---|---|
| Rheumatology | 12 (48.0) |
| General internal medicine | 15 (60.0) |
| Family medicine | 10 (40.0) |
† 12 doctors had expertise in multiple fields.
Table 2: Remote area experience of expert panel members
| Experience of remote area practice | n (%)† |
|---|---|
| Remote area clinics | 17 (68.0) |
| Remote area community hospitals | 12 (48.0) |
| Remote area medical support organizations | 4 (16.0) |
| None | 4 (16.0) |
† 12 doctors had experience in multiple locations.
Delphi process and consensus building
The response rates for the Delphi method were 25/25 (100%) in rounds 1 and 4, and 24/25 (96%) in rounds 2 and 3 (Table 3). In round 1, participants emphasized several key skills necessitated by limited medical access in remote settings. These included the ability to perform initial assessments at symptom onset, basic diagnostic techniques, and joint aspiration. Experts also highlighted the importance of recognizing the appropriate timing for specialist collaboration and ensuring continuity of care for chronic conditions. However, concerns were raised regarding the difficulty of aspirating joints other than the knee, and some members argued that preconception care should remain the purview of specialists. Consequently, we narrowed the aspiration skill specifically to the knee joint and rephrased the item on preconception care to ‘consult with a specialist as needed’. In round 2, the panel proposed excluding the ‘utilization of social resources’ from the list, as this was deemed a general clinical skill rather than one specific to rheumatology; this suggestion was adopted. By round 3, most items had reached a consensus. A further refinement was made to ‘infectious disease screening’ to include ‘consult with a specialist as needed’. In round 4, the Delphi rounds were terminated as the criteria for agreement were met for all items.
As a result of the Delphi rounds, a list of 14 tasks was finalized (Table 4).
Expert feedback underscored several critical themes: the prevalence of undiagnosed common rheumatic symptoms; the necessity of maintaining a broad differential diagnosis during physical examinations in resource-limited settings; and the indispensable nature of early diagnosis, timely specialist referral, and structured handover skills.
Table 3: Response rate for Delphi rounds
| Round 1 | Round 2 | Round 3 | Round 4 | |
|---|---|---|---|---|
| Response rate n (%) | 25/25 (100) | 24/25 (96) | 24/25 (96) | 25/25 (100) |
Table 4: List of entrustable professional activities
| Task | Category | Entrustable professional activities |
|---|---|---|
| 1 | History taking and physical exam | Can take medical histories of patients complaining of symptoms that may be related to rheumatic diseases (including finding abnormalities in tests during medical check-ups). |
| 2 | Can perform a thorough physical examination including joint, skin and nail examinations. | |
| 3 | Diagnostic skills | Can perform screening tests related to the diagnosis of rheumatic diseases. |
| 4 | Can perform puncture of knee joint. | |
| 5 | Manage emergencies | Can determine the urgency of rheumatic diseases and, if the urgency is high, refer to a specialist. |
| 6 | Diagnosis of new onset rheumatic diseases | Can provisionally diagnose typical cases of newly developed rheumatoid arthritis, crystal-induced arthritis, osteoporosis, osteoarthritis, and polymyalgia rheumatica. |
| 7 | Can refer patients suspected of developing a new rheumatic disease to a specialist as necessary. | |
| 8 | Initial management of new onset rheumatic diseases | Can provide initial treatment for newly developed rheumatoid arthritis, crystal-induced arthritis, osteoporosis, osteoarthritis, and rheumatic polymyalgia. |
| 9 | Can refer newly developed rheumatoid arthritis, crystal-induced arthritis, osteoporosis, osteoarthritis, and polymyalgia rheumatica to a specialist if the patient does not respond well to initial treatment. | |
| 10 | Continue managing chronic phase rheumatic disease | Can evaluate disease activity, daily living ability, and degree of organ damage of patients with chronic rheumatic diseases in outpatient settings; and continue/adjust treatment. |
| 11 | Can deal with or refer to specialists for the following problems among patients on rheumatic diseases: relapse of the primary disease, problems during treatment including infection, and side effects of medication. | |
| 12 | Can work with specialists to implement infection prevention measures (HBV, HCV, tuberculosis, PCP, VZV) necessary for patients with rheumatic diseases. | |
| 13 | Can handover patients with chronic rheumatic diseases to successor. | |
| 14 | Preconception care and pregnant patient care | Can refer rheumatic disease patients who need preconception care and pregnant patients with rheumatic diseases to a specialist. |
† Clinical skill decay advisory: In rural settings with low patient volumes, clinicians will face risk of clinical skill decay. If a clinician suspects skill decay, they should leverage external support through specialist referral or direct in-person supervision before proceeding. This principle applies regardless of prior entrustment level and is fundamental to patient safety.
Discussion
We have successfully developed a list of rheumatology-related tasks for young generalists assigned to remote areas. The uniqueness of this list lies in its comprehensive scope that, in addition to covering clinical skills, emergency response, initial treatment, and chronic rheumatic disease management, it addresses how to collaborate with specialists and ensure continuity of care.
The expert panel’s contributions reflected the practical realities of remote medicine. They prioritized simplifying technical requirements, emphasizing the timing of referrals over high-level diagnostic precision, and identifying urgent clinical situations. The inclusion of infection prevention and the focus on common geriatric conditions such as osteoarthritis, osteoporosis, crystalline arthritis, and polymyalgia rheumatica reflect the demands of Japan's super-aged society.
ACGME and EULAR have developed task lists comparable to EPAs for training rheumatology fellows, which served as reference materials for our initial list6,7. However, these existing lists are specifically designed for specialists, not generalists. Consequently, they require a relatively advanced level of knowledge and procedural skills, and do not address collaboration with specialists.
Our list is more specialized in the rheumatology field than the general training EPAs for initial postgraduate training. Because these EPAs were intended for generalists in early residency, particular emphasis is placed on collaboration with specialists. This list is specifically intended for physicians who will be assigned to remote areas. The expert panel underscored the necessity of providing optimal care within limited resources and collaborating appropriately with specialists19.
Key challenges in remote medical areas include a lack of medical resources, poor access to higher-level medical institutions, and issues with continuity of care20. While urban generalists have easier access to specialists, the ability to determine when to consult and how to provide initial management during the waiting period is a universal requirement. Therefore, this EPA list has the potential to be used for training generalists not only in remote areas but also in aging urban areas.
Beyond geographic remoteness, these EPAs are valuable in any setting with limited specialist access. Digital tools and telemedicine can support generalists, but the physician’s initial response such as differential diagnosis and early treatment remains crucial. The 2025 JCR recommendation on telemedicine for rheumatic diseases envisions hybrid care models in which specialists provide remote consultations while local physicians manage patients on site9. Our EPAs, by defining the competencies required of local physicians, directly complement this vision. Specifically, EPAs such as initial clinical assessment, recognition of indications for urgent specialist consultation, and structured handover represent the foundational skills that enable effective doctor-to-patient-with-doctor telemedicine. Our Delphi data were collected in 2023-2024; subsequent publication of the JCR recommendation reflects a convergent national recognition of the same care gap that our study sought to address at the competency level.
Given the severe human resource constraints in rural practice, we propose that the 14 EPAs need not be fully applied for all clinical encounters. Instead, we suggest a scenario-based fractional utilization strategy, in which relevant EPAs are selected according to the clinical context. For example, in routine outpatient encounters, EPAs related to chronic disease monitoring and medical management would be most relevant. In acute or emergency presentations, EPAs addressing initial assessment and recognition of urgent referral indications would take precedence. During care transitions and specialist handovers, EPAs focused on structured communication, continuity of care, and infection screening would be prioritized. This modular approach reduces the per-encounter assessment burden on supervisors. To ensure feasibility within busy rural settings, we plan to implement this framework using a programmatic assessment approach, in which each EPA is evaluated incrementally across multiple clinical encounters over time, rather than assessed comprehensively in a single observation. Individual case-based assessments will capture partial evidence of competence within specific clinical scenarios, and this cumulative evidence will be aggregated and reviewed periodically. Final entrustment decisions will not rest on any single assessment but will instead be determined by a clinical competency committee that will synthesize longitudinal performance data to render defensible, holistic judgements on each trainee's readiness for unsupervised practice. Moving forward, we have designed a structured three-phase implementation roadmap. In phase 1, the framework will be introduced to a small cohort of frontline generalists and supervisors at rural facilities who were not involved in the Delphi process. This phase will systematically measure supervisor assessment burden. In phase 2, feedback from phase 1 will be used to refine the EPA list. In phase 3, the refined EPA list will be embedded into structured outpatient training rotations, linked to a competency matrix, and evaluated for its impact on trainee performance and, ultimately, patient outcomes. This phased approach mirrors the methodology reported in previous publications 16,21.
Limitations
A limitation of this article is the definition of remote areas. There is no single, universally accepted definition of remote areas. While remoteness is often defined by distance or population size, it can also be characterized by service availability and geography. In this study, we defined remote areas as regions with limited human and material resources where emergency access to specialists or advanced intensive care is difficult22.
Additionally, while IT tools and telemedicine could significantly mitigate the challenges of remote medicine, they were not the focus of this study. Given that telemedicine is still in the process of being integrated into the local context, we did not include specific telehealth competencies. This list will need to be updated as these technologies become more widespread.
A further and significant limitation is the absence of preliminary field testing prior to publication. As noted by foundational EPA development methodology, a consensus-derived EPA list requires real-world clinical piloting before it can function as an operationalized assessment tool15,21. The present work should be regarded as a validated prototype framework rather than a deployment-ready curriculum tool, and the implementation roadmap described above is designed to address this limitation systematically.
Conclusion
This study establishes a framework for managing rheumatic diseases within the constraints of limited resources, focusing on the synergy between generalists and specialists. We hope this work serves as a catalyst for broader discussions on which aspects of specialty care can be safely and confidently entrusted to generalists, thereby improving the quality of care in rural and remote communities.
Acknowledgements
This work was significantly enriched by the 25 expert panel members who participated in our Delphi rounds. We are profoundly grateful for their time and the rigorous expertise they brought to our discussions, which directly shaped the final EPA framework.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflicts of interest
The authors declare no conflict of interest.
AI disclosure statement
During the preparation of this work, the authors used Google Gemini 3.1 Pro for English language editing purposes. The authors reviewed and edited the content as necessary and take full responsibility for the content of the publication.
Data availability
Data is available on reasonable request.