Musculoskeletal disorders are common among musicians due to repetitive movements, muscle strain, poor posture, and intensive practice, affecting their ability to perform and sustain careers [1-3]. At least 50% of musicians experience performance-related musculoskeletal disorders (PRMDs), which cause pain and other symptoms that hinder performance [4-6]. However, targeted therapy is underused, and specialized assessment and treatment strategies are needed [7,8].

Musician-specific physiotherapy is beneficial for identifying and treating these issues, focusing on aspects of posture, movement behavior, and musical expression that affect health [8-10]. Effective treatment requires a thorough understanding of musician-specific characteristics and identification of functional disorders in the musculoskeletal system through clinical neuro-orthopedic examinations and analyses of the musician’s movement while performing [11].

In physiotherapy, clinical reasoning is essential for decision-making, often using a hypothetical-deductive method [12]. Accurate diagnostics are crucial to prevent misjudgments [13], and comprehensive information gathering is vital for effective therapy planning [14,15]. Technology-driven assessments can enhance diagnostic accuracy, moving beyond subjective methods and trial-and-error approaches [16].

Clinical movement analysis (CMA) provides objective, quantifiable data on functional disorders, supporting personalized medicine [11,17] and offering physiotherapy applications that conventional assessments cannot achieve [18,19]. As an apparatus-based biomechanical analysis for individualized functional diagnostics, CMA uses techniques such as motion capture and electromyography [20,21]. It helps evaluate complex movements during performance, allowing for the identification of aberrant postures and movements contributing to PRMDs in musicians [22-24]. Thus, CMA aids in diagnosing functional disorders, planning individualized therapy, and monitoring therapeutic measures [17,19,25], minimizing the risk of false results [26,27].

The RefLabPerform project [28] aims to develop a reference laboratory for PRMD assessment in musicians, integrating CMA with physiotherapeutic assessments, which will be implemented at the Institute for Applied Physiotherapy Osnabrück (INAPO). Advanced instruments will be used by specialists for musician-specific CMAs, with results provided to physiotherapists for further clinical evaluation.

However, interpreting CMA data requires expertise and time, presenting challenges for clinical practice, highlighting the need for efficient, intuitive presentation for analysis and decision-making [11,29,30]. Current tools for biomechanical analysis lack guidance for musicians’ physiotherapy, leading to concerns about accuracy and usability [30,31].

A tailored health technology for musicians’ physiotherapy is essential to efficiently record, process, provide, analyze, and interpret musician-specific CMA data for routine use. Hereby, an interactive dashboard with clinical decision support (CDS) can improve diagnostics by integrating CMA findings, facilitating comprehensive decision-making and effective therapies [32]. Adapting research approaches from health while considering physiotherapy-specific conditions will ensure the tool aligns with the clinical workflow, balancing perceived benefit, effort, and costs [16,33].

Human factors engineering (HFE) is crucial for designing interactive health technologies and CDS tools, focusing on usability and workflow integration for decision support [34-36]. Effective CDS requires speed, efficiency, meaningful alerts, consistency, and logical grouping with intuitive interface design [37-40]. Hence, designing such a technology must address user needs, working environments, cognitive demands, decision-making requirements, and advanced interactive visualizations for data analysis.

In total, 3 HFE approaches—design study methodology (DSM), user-centered design (UCD), and decision-centered design (DCD)—provide guidance for this. DSM is a problem-driven research paradigm focused on creating innovative visualization solutions through iterative collaboration with domain experts to address real-world problems [41]. UCD emphasizes understanding user tasks, goals, and needs by involving users from the beginning, which informs the specification of user requirements and guides the design process [42]. DCD aims to improve decision-making in critical scenarios, such as medical diagnosis, by enhancing human decision-making [43,44]. Cognitive task analysis identifies key cognitive demands and translates them into decision requirements, guiding the design process to improve clinical performance [45]. These approaches share a human-centered, iterative, and multistage methodology, leveraging in-depth analysis and expert contributions, while each addresses distinct aspects of the design process.

The study aimed to identify design requirements for an interactive tool, specifically a physiotherapist dashboard, that aids clinical decision-making for musculoskeletal problems in musicians within the RefLabPerform project. This involved considering the working environment, clinical procedures, and user needs from the beginning and gathering detailed information on how decision-making in musicians’ physiotherapy can be supported technically, especially with the integration of CMAs.

To achieve this, requirements elicitation methods from DSM, UCD, and DCD were followed to comprehensively describe the domain and the underlying problem, resulting in well-informed requirements and recommendations. This approach is expected to enhance usability and acceptance, improve user efficiency and accuracy, and reduce the cognitive workload of the future tool.

This qualitative user research study followed a problem-driven research paradigm and formed part of the initial domain characterization and analysis phase within a larger design study [46]. It systematically integrated elements from DSM, UCD, and DCD to elicit and structure design requirements for an interactive dashboard with CDS functionality in physiotherapy (see Figure 1). The requirements elicitation process drew on the first 4 steps of DSM (learn, winnow, cast, and discover) [41], the initial 2 phases of UCD (understanding context of use and specifying user requirements) [42], and the first 3 stages of DCD (preparation, knowledge elicitation, and analysis and representation) [44]. These approaches were merged and operationalized into three stages: (1) precondition and preparation, (2) knowledge acquisition and characterization, and (3) analysis and representation, with subsequent design and evaluation phases planned for future work. DSM ensured methodological rigor and problem characterization through iterative domain expert engagement; UCD emphasized contextual inquiry and stakeholder needs; and DCD supported the analysis of complex cognitive processes in physiotherapy decision-making. Together, they provided a robust foundation for deriving clinically relevant, user-informed, and cognitively aligned design requirements. A detailed mapping of their contributions is presented in Table 1.

The main author (EW) conducted the workflow observations and moderated the focus groups in German in March and April 2021. His researcher characteristics are the following: “gender: male,” “experience: 6 years research experience in health informatics, main focus in physiotherapy and movement analysis projects,” “degree: M.Sc. in Computer Science,” and “occupation: Research assistant.”

The study was conducted in accordance with the Declaration of Helsinki and approved by the institutional review board (or Ethics Committee) of Osnabrueck University of Applied Sciences (protocol code HSOS/2021/1/1, date of approval March 17, 2021). Signed informed consent for publication was obtained from all participants after providing them with a written information sheet and verbal explanation of the study context, the intended procedure, and data usage. All participants were given a copy of the signed informed consent as well as confirmation of anonymization of study data. The study data were anonymized, so participants cannot be identified. Participants did not receive any financial compensation for their participation.

To describe the domain, the characteristics were categorized under the following themes: domain problem; goals, skills, and competences, work equipment, and barriers; and tasks, routines, and challenges.

The findings from the prior domain characterization were further examined and discussed with the domain experts. This process aimed to address the integration of musician-specific CMA into physiotherapy, identify user requirements, and determine key cognitive tasks and decision requirements.

The domain characterization and analysis informed concrete design requirements for a physiotherapist dashboard, specifying key features and CDS recommendations based on user needs for decision support. The use case defined technical prerequisites for implementation.

We proposed and applied a novel HFE process model to inform the design of a physiotherapist dashboard with CDS for CMA of musicians with musculoskeletal problems, using methods, and addressing aspects from DSM, UCD, and DCD. Following this model, we first identified and thoroughly described the characteristics of the domain and problem area. We explored user needs, cognitively challenging tasks, information needs, and decision-making challenges faced by physiotherapists managing musicians with musculoskeletal conditions. We then proposed design seeds for CDS, translating user and decision requirements into concrete design requirements, including key features, technical prerequisites, and recommendations for decision support.

Key features include efficient integration of biomechanical data, combining heterogeneous data types, and providing adaptive overviews for quick data inspection. The tool should present patient data in a suitable form and facilitate interactive visual analyses using familiar techniques like filtering, highlighting, and annotating. Summarizing and exporting findings for communication are also essential. User and decision requirements led to specific decision-support recommendations: CDS should indicate if musician-specific diagnostics are possible, remind physiotherapists of physical exams, suggest efficient configuration parameters, preselect essential data, highlight significant changes, integrate normative values, support interactive annotations, compare baseline and final findings, and provide tools for reviewing patient goals and preparing reports. To enable the tool, technical prerequisites must be met, including access to patient data (anamnesis, physical exams, treatments, and problem- or musician-related questionnaires), biomechanical data from musician-specific CMAs, support for electronic diagnostic orders, and management of these orders and CMA statuses. The tool should use standardized protocols, integrate with existing tools, and automate data processing to present biomechanical findings.

Additionally, our study contributes to the existing literature by detailing a methodology for requirements elicitation, comprehensively describing the domain of musicians’ physiotherapy, and identifying requirements for the successful use of the tool. Our findings support previous research indicating that HFE is crucial for designing CDS technologies [34,35], emphasizing aspects such as speed, efficiency, user needs anticipation, workflow integration, meaningful alerts, consistency, logical grouping, interface design, and design simplicity [37-40].

This study is in the early stages of design, focusing on establishing a structured foundation for the tool’s design and domain-specific clinical workflows, relying primarily on qualitative methods like workflow observations and focus group discussions. While the findings provided in-depth qualitative insights from a small group of domain experts, generalizability is limited, and broader validation is needed. To address this, we plan to conduct a Delphi study involving a larger and more diverse panel of physiotherapists, movement scientists, and other relevant stakeholders [64]. This structured, iterative process will facilitate expert consensus on the tool’s design requirements, ensuring their robustness and applicability across various physiotherapy contexts. The Delphi study will also help refine the identified requirements based on collective expert feedback, enhancing the tool’s validity before further prototyping, usability testing, and implementation in clinical practice.

At this stage, no functional prototype exists, and technical features or implementation details, such as specific visualizations, interaction methods, and decision logic, were not addressed. These aspects will be defined and iteratively refined during future stages of prototyping and evaluation. Furthermore, quantitative evaluation of metrics such as diagnostic accuracy, efficiency, or patient outcomes is not yet feasible. Future work will involve the development of a prototype, followed by iterative usability testing and structured evaluations. These evaluations will incorporate quantitative measures—such as time-to-decision, task completion rates, diagnostic decision support effectiveness, efficiency gains, and user satisfaction scores—to assess the tool’s impact and clinical utility in real-world physiotherapy settings.

In addition, artificial intelligence (AI)–based CDS features were overlooked. Integrating machine learning and predictive analytics could significantly enhance precision and relevance [65]. In future development stages, machine learning algorithms can be incorporated to support early diagnosis, pattern recognition, as well as personalized treatment planning and outcome prediction considering individual patient characteristics. For example, supervised learning models could be trained on annotated CMA data to automatically identify movement anomalies or classify injury risks [65]. Predictive analytics could also be used to forecast recovery progression based on patient-specific parameters such as movement patterns, therapy progress, and contextual factors (eg, instrument type or workload) [66]. These enhancements would augment the therapist’s decision-making with advanced data-driven insights while maintaining transparency and clinician control. Overall, this highlights the tool’s extensibility and long-term innovation potential.

Moreover, while the proposed CDS approach focuses on integrating CMA with gold-standard methods, AI-driven approaches like computer vision-based pose estimation (eg, OpenPose [67]) should be explored. Sensorless systems could provide nonintrusive, real-time movement tracking, making assessments more accessible, especially in nonlaboratory settings. AI pattern recognition could automate impairment classification and suggest diagnostic hypotheses, aiding physiotherapists and improving decision-making. It could also enable remote physiotherapy, allowing musicians to exercise at home with real-time feedback. However, challenges in accuracy, occlusion, and clinical validation must be addressed before full integration into physiotherapy workflows.

The integration of CMA into physiotherapy and incorporating clinical and biomechanical data with CDS within an interactive dashboard present both opportunities and challenges. Our study emphasizes the need to tailor CDS to real clinical environments and therapists’ workflows. Ensuring these technologies are perceived as useful is crucial, as the balance between perceived benefits and implementation effort significantly affects adoption. Developing an interactive tool for supporting diagnosis and treatment of musculoskeletal disorders in musicians requires a multidisciplinary approach, combining health informatics, biomechanics, and clinical practice. Collaboration with domain experts ensures the tool meets practical needs. Requirements were elicited through problem-driven research, ensuring that real-world problems were addressed, as well as UCD and DCD approaches, considering user needs, cognitive demands, and decision requirements, aiming to enhance the usability and usefulness of the solution.

Most existing tools in physiotherapy focus on general clinical documentation [68] or basic measurement instruments like goniometers and numeric rating scales [18], with limited quantitative and objective significance. Advanced biomechanical analysis systems (eg, Qualisys [69]) remain primarily research tools—technically complex and disconnected from everyday clinical decision-making, lacking user-friendliness, tailored visualizations, and integration with clinical reasoning. CDS tools in physiotherapy are still emerging, with AI-based approaches showing promise for personalizing treatment and predicting patient-specific outcomes [70-72]. However, they lack integration of CMA into clinical workflows and provide limited support for context-specific decision-making, particularly for niche populations like musicians with PRMDs. Existing visualization tools support rehabilitation or monitoring but do not incorporate CMA data [73,74], and those that do focus on clinical gait analysis or specific disorders [75,76]. Currently, no tool integrates CMA data into the decision-making process for physiotherapists managing musicians with PRMDs.

An interactive tool based on the identified design requirements would fill this gap by directly integrating CMA into clinical workflows, aligning with physiotherapists’ reasoning. It would enable interactive visualizations tailored to the needs of diagnosing and treating musicians with PRMDs, offering flexibility for different instruments and playing styles. Unlike existing tools, it bridges the gap between biomechanical analysis and clinical practice, providing musician-focused support and a platform for integrating advanced analytics to enhance diagnostic precision and longitudinal tracking. Such a system aims to integrate CMA into real-world physiotherapy workflows, providing context-aware support and decision facilitation across all therapy stages, particularly for underserved populations like performing artists.

The successful deployment of a future CDS tool will require addressing key technical and ethical challenges. Integration into existing health care IT systems involves overcoming data heterogeneity, ensuring interoperability, safeguarding data privacy, and achieving clinician adoption. Many institutions rely on proprietary electronic health records [77], complicating seamless data exchange. Standardized protocols such as HL7 (Health Level Seven) and FHIR (Fast Healthcare Interoperability Resources) can help address interoperability [78], while compliance with regulations like GDPR (General Data Protection Regulation) will require robust encryption, anonymization when applicable, and role-based access controls. Diverse IT environments further necessitate system compatibility; a modular, cloud-based architecture could offer the flexibility and scalability needed for broad adoption [79]. Clinician adoption will depend on smooth integration into existing workflows, user-friendly interfaces, and comprehensive training [80].

Future work must standardize and integrate diverse data sources—such as motion capture outputs, clinical notes, and patient records—into a unified platform compatible with physiotherapy workflows. Ethical considerations are equally critical. The envisioned tool is designed to assist, not replace, physiotherapists by providing objective biomechanical data to enhance—not automate—clinical decision-making. The system will maintain clinician autonomy, ensure transparency and interpretability, and be developed in adherence to data protection regulations. Although risks related to automation or bias are minimal at this stage, they will remain important considerations in later development phases, particularly as data handling, user control, and AI components evolve. Finally, the evolving nature of clinical practices and data types presents ongoing challenges for maintaining a CDS tool’s relevance. Continuous updates are needed to keep pace with advancements in clinical knowledge and technology, which can be resource-intensive. Addressing these challenges in future research will enhance such a tool’s usability, adaptability, and impact on clinical decision-making.

In conclusion, successful implementation of the physiotherapist dashboard with CDS based on musician-specific CMA depends on alignment with clinical workflows, ease of use, successful integration in existing IT infrastructure, maintaining data protection and clinician autonomy, and demonstrable benefits in diagnostic accuracy and treatment efficiency. Insights from this study provide a solid foundation for designing an interactive tool for physiotherapists, incorporating CMA for musicians with musculoskeletal demands, aiming to improve patient outcomes and support clinical decision-making. Iterative feedback from domain experts during design and development should ensure the tool addresses real-world problems, is validated through practical application, and is continuously refined based on user insights. The involvement of physiotherapists and movement scientists from initial design to validation is pivotal.

While this study focuses on the niche context of musicians with PRMDs, the identified design requirements, domain-specific decision-making insights, and integration of CMA hold potential for broader application. The underlying framework and methodology can be adapted to other physiotherapy contexts that involve complex, movement-related diagnostic and therapeutic challenges—such as in athletes, dancers, or individuals with work-related repetitive strain or posture-related injuries. These populations similarly require tailored functional assessments and context-specific CDS [81-84]. Modular, task-oriented design supports adaptation across these groups by enabling customization of movement analysis protocols, clinical decision parameters, and visualization formats to meet domain-specific needs. This enhances both the generalizability and scalability of the proposed framework. Future work may explore how the proposed process model and design requirements can be transferred and tailored to other physiotherapy contexts, extending the broader relevance and impact of this research beyond musicians.

This work contributes to HFE, physiotherapy, CMA, and CDS for musicians with PRMDs by combining rigorous research methods with expert insights. The proposed process model helps stakeholders identify user needs and decision-making requirements at the point of care, while the design requirements lay the groundwork for developing a specialized interactive physiotherapist dashboard integrating CMA with CDS. This tool can enhance diagnostic accuracy and therapy effectiveness through interactive, visual data presentation, streamline workflows, and maintain clinical relevance. Ultimately, it aims to improve patient outcomes, reduce injury recurrence, and support musicians’ long-term careers.

In the next phase, we plan to translate the design requirements into a functional prototype as part of a physiotherapy-integrated diagnostic system [85]. To ensure practicality and effectiveness, we will conduct rigorous usability testing and evaluations with physiotherapists. This will involve iterative feedback loops through think-aloud protocols and task-based evaluations, ensuring that the tool aligns with clinical workflows and cognitive demands. Structured feedback will be used to refine interface features, data presentation, and interaction flows, providing empirical evidence of feasibility, usability, and alignment with the identified requirements.

Future research should validate usability, decision support, and clinical effectiveness; explore advanced analytics; adapt to various CMA systems (eg, sensorless tracking); and expand applications across diverse musicians, instruments, and therapeutic domains to enhance its impact.