This mixed methods study involved 3 stages (development, implementation, and evaluation), and followed the New South Wales Translational Research Framework [42] with outcomes assessed according to the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework [50]. Both frameworks provide structured approaches to ensure that research is relevant and scalable and are widely used for research in the Australian health care context.

Ethics approval was obtained from the Human Research Ethics Committees at The University of Sydney (2018/926, 2021/951), The University of Melbourne (1954239), The University of Queensland (2019000700), Curtin University (HRE2019-0263, HRE2020-0562) and Western Sydney Local Health District (2021/ETH11872). Participants were informed about the study before providing their written consent (either on paper or via a web-based form). Participation was voluntary and participants were informed about their right to withdraw from the study at any point without any consequences. Collected data were coded and deidentified.

Each stage involved different recruitment procedures. During the development stage, researchers, consumer, industry and clinical partners were identified by members of the research team and invited to participate via direct contact. These stakeholders were those with recognized expertise in musculoskeletal conditions and who had previously collaborated with members of the research team or been involved in previous studies. HCPs were involved in the PACE-MSK trial or had been involved in previous trials and had consented to be contacted for future trials. HCPs were working in either primary or tertiary (hospital) settings. People with MSK conditions had been involved in previous trials and had consented to be contacted for future trials. Contact was made with these groups via email or direct phone call. Students allied to HCPs were enrolled in a unit of study at a participating university and had consented to participate.

In the implementation and evaluation phases, HCPs and patients were involved in the PACE-MSK trial or had been involved in previous trials and had consented to be contacted for future trials. In addition, HCPs working in occupational health settings known to the research team or who had been involved in previous studies were invited to participate either by email or direct phone call. Those HCPs working in our clinical partners’ tertiary care setting (Western Sydney Local Health District, [WSLHD]) were directly invited by their peers and managers. Patients with MSK conditions were also invited to participate by their treating clinician.

The aim of the development stage was to design a website where patients and HCPs could access relevant, up-to-date, high-quality evidence-based information on common musculoskeletal conditions (LBP, neck pain, WAD and knee osteoarthritis). Given the expected commencement date of the PACE-MSK trial, the development stage incorporated comprehensive stakeholder engagement rather than a formal co-design process. The problem identified by the research team was that multiple resources exist, with neither a central nor coordinated location for people to access credible evidence-based information [51].

The first step of the New South Wales Research Translation Framework is Idea Generation, where key stakeholders are engaged to design a solution to the problem. Our core team comprised clinician- researchers from 4 Australian Universities (The University of Sydney, The University of Queensland, The University of Melbourne and Curtin University) with research expertise in musculoskeletal conditions. The team determined that key stakeholders likely to benefit from this resource were allied HCPs working in both primary and tertiary care, researchers involved in creating evidence for musculoskeletal conditions, specialist musculoskeletal clinicians, specialist medical professionals, people with musculoskeletal disorders, consumer organizations and industry partners. We then determined the level of influence and interest of each stakeholder using a stakeholder engagement matrix and considered appropriate strategies for managing their involvement and expectations [52]. Engagement methods ranged from being involved in co-design and implementation (manage) to piloting of resources (meet needs) and meetings (keep informed; Table 1).

We held work-in-progress meetings with key stakeholders between July 2018 and October 2022. Discussions included design concepts, content, functionality, website maintenance and support and how this could be implemented. To ensure optimal functionality and ease of use, a digital agency (Vivo Group, Brisbane, Queensland, Australia) with experience in development of user-friendly websites was selected to design the website following a robust tendering process. As the content was developed, each stakeholder group shared their views and perspectives and provided feedback about additional key messages and resources that should be included or omitted. Before implementation, the website went through final checks by the research team and the digital agency to test features, functionality and resolve any issues.

The aim of the implementation stage was to determine strategies that would help ensure effective and scalable implementation of the website before its widespread public release. Groups thought likely to source information about musculoskeletal conditions were identified and included allied health students and graduate allied HCPs working in health care settings where people with musculoskeletal conditions present. Settings included primary care, occupational health, and tertiary care settings. Finally, allied HCPs working in a specialist capacity were engaged. This process ensured implementation occurred with clinicians with a range of clinical experiences and in different care settings. We chose implementation strategies ranging from passive (dissemination) to more active strategies (facilitated and discussion-based education with opinion leaders) based on strategies that had been successful in our previous work (Table 1) [33,43,44]. During this stage, the website was not publicly accessible.

Patients with musculoskeletal conditions were invited by their treating clinician to use the resources on the website and then provided consent to participate in qualitative studies associated with both our primary and tertiary care implementation initiatives. After using the website, patients participated in a semistructured interview (via a web platform) or focus group (in person), depending on what was feasible in the particular health care setting and on the patient’s preference and location, to provide their opinions on the resource.

The aim of the evaluation stage was to investigate both effectiveness, acceptability, and adoption of the website (RE-AIM framework) by allied HCPs (students, allied HCPs, and specialist HCPs) and people with musculoskeletal conditions. Mixed methods involving quantitative (web-based survey) and qualitative (focus groups or semistructured interviews) were used.

The total number of participants involved in each stage is provided in Table 2.

The stakeholder consultation generated ideas that informed the website design and implementation. This process informed the development of the website both for “Patients” and “Clinicians.” Key features and content were identified and then modified iteratively as a result of the consultation process. Modifications did not follow a strict, linear order occurring instead in parallel.

Implementation of the website took place in 4 Australian states (New South Wales, Queensland, Western Australia, and Victoria) from February 2018 to December 2023 as a component of PACE-MSK, the clinical trial for which the website was built [49]. As part of PACE-MSK, 767 people with musculoskeletal conditions had consented to be part of the trial and were therefore required to register on MyPainHub. A total of 693 HCPs were involved in the PACE-MSK trial but registering on MyPainHub, whilst possible, was not a requirement of being involved in the trial. In addition to the PACE-MSK trial, the website was implemented within WSLHD (both patients and HCPs), with allied HCPs in occupational health settings and with students (Table 3).

The implementation strategy adopted influenced the number of participants who, despite consenting to participate, registered on the website (Table 3). As a result of the education via webinars and workshops, 100% of allied HCPs working in occupational settings, tertiary care settings and specialist HCPs who consented to participate registered on the website. However, despite being embedded in a unit of study in a university physiotherapy program, only 16% of students registered on the site. Passive strategies, such as phone calls or reminders, were ineffective with only 29% of HCPs working in primary care registering (Table 3).

Website data for a 12-month period (June 2021-June 2022) are presented in Table 4. The total page views of the website during the period were 23025 of which 12149 were unique page views. The pages with the highest unique views were general information about knee osteoarthritis, neck pain assessment (clinicians), frequently asked questions, LBP assessment (clinicians), general information about neck pain (general information), and exercise for knee osteoarthritis (Table 4). Users viewed on average 4 pages during a visit and stayed approximately 4 minutes on each page.

This paper describes the process undertaken to develop, implement and evaluate a website designed to support patients with, and HCPs who manage, musculoskeletal conditions, specifically LBP, neck pain, WAD, and knee osteoarthritis. The findings suggest that following a research translation framework and incorporating extensive and iterative consultation with multiple key stakeholders resulted in MyPainHub being acceptable, credible, and easy to navigate with participants reporting they are likely to revisit and recommend the website to others. Embedding prognostic tools (the SF-ÖMPSQ and the Keele STarT MSK tool) allowed information to be tailored to a person’s risk of poor outcome which may help both patients and HCPs better adopt guideline-based management. However, training for HCPs that includes more specific strategies on how to incorporate MyPainHub into clinical practice would be beneficial for more successful implementation.

A key component of the development of MyPainHub was the extensive involvement of a wide range of key stakeholders with different levels of experience and from different health care settings over a 4-year period. The initial development phase focused on gathering perspectives from people with musculoskeletal conditions and HCPs from primary care settings, researchers, key opinion leaders, professional associations, consumer groups and website developers. The latter development phase involved seeking perspectives of HCPs and patients from tertiary care settings, allied professionals from occupational settings and from allied health students. This broad stakeholder engagement reflects contemporary user-centered design and was undertaken to facilitate future wider implementation of MyPainHub across a range of settings and populations [61]. Whilst active implementation strategies (eg, interactive education and workshops) were generally more effective than passive strategies (eg, phone calls and emails), implementing MyPainHub in a unit of study was not successful amongst allied health students. In our study, students were introduced to MyPainHub during a lecture and were then responsible for designing a treatment program using MyPainHub but no specific tutorial time was allocated for this task. A previous study [43] that had implemented an innovative, guideline-based website, My Whiplash Navigator, found student HCPs had a higher uptake of the website than primary HCPs but the students were allocated time during both a lecture and a tutorial to using the resource which may explain the difference in the level of engagement.

The main purpose of this project was to develop a resource that would support guideline-based care for patients at low and high risk of poor outcome. Tailoring information to a person’s risk of poor outcome afforded the opportunity to provide different information on likely recovery pathways and interventions that may be required. In this way, MyPainHub reinforces key guideline-based messages, that people at low risk should do well with minimal care but that people at high risk may require more comprehensive assessment of multiple domains and a multidisciplinary approach to optimize health outcomes. Greater adoption and wider implementation of MyPainHub may improve the efficiency of health service delivery by helping to educate people about musculoskeletal conditions and influence care decisions but requires further research.

Despite the opinions of stakeholders that MyPainHub is acceptable, credible, and easy to navigate, few used the tools with patients, and additional training that focuses on how best to integrate the website into clinical practice in different health care settings is likely to be required for greater adoption and usage. Active engagement strategies were more effective than passive strategies and taking a multifaceted approach, for example embedding MyPainHub in existing educational opportunities (eg, workshops, curriculum, and training programs) and using key opinion leaders to reinforce the credibility of information, is likely to be important in future implementation [62,63]. The poor response rate and level of engagement of students could be overcome by integrating MyPainHub as a supplementary resource into the physiotherapy program curriculum and ensuring the information provided aligns with the learning objectives of the specific unit of study with which it is associated. Effective incorporation of MyPainHub for students may be achieved through thoughtful planning and implementation, integrating into specific lesson plans including activities such as group discussions followed by monitoring of student engagement via Learning Management Systems or linked to assessable content [64].

As with any web-based resource, updates to MyPainHub are ongoing with the maintenance of the website being the responsibility of the research team and the digital agency. Addition of information about other conditions (eg, hip pain) is underway and incorporating additional features such as more dynamic elements (eg, carousels and expandable sections) and gamification to increase engagement [65] are likely to be important features of future updates. Limitations of this study include that we did not control the time between participants viewing the website and asking for their feedback. Despite efforts to engage general practitioners and other medical professionals, the majority of those involved in the development of MyPainHub were allied HCPs. Further work is needed to determine whether MyPainHub, or specific sections of MyPainHub, would be useful for general practitioners. Further work is also required to implement MyPainHub in other primary (eg, community-based care) and tertiary care settings.

This study details the development, implementation, and evaluation of MyPainHub, a web-based resource designed to support guideline-based care for musculoskeletal conditions. Stakeholder engagement throughout the process contributed to its acceptability, credibility, and ease of use. While MyPainHub effectively tailors information based on prognostic tools to support both patients and health care providers, further efforts are needed to facilitate its integration into clinical practice. Training for health care professionals, particularly in how to apply MyPainHub in diverse settings, is likely essential for greater adoption.