Background
Arthritis and musculoskeletal conditions are more prevalent in Australia than any other National Health Priority area, including cancer, diabetes and obesity [
1]. Osteoarthritis (OA) in particular is a leading cause of pain, disability and early exit from the workforce in Australia [
2], with the knee commonly affected. Arthritis leads to a substantial loss of income [
3] (with a resultant increase in welfare dependency) [
4] and reduction in taxation revenue, and a significantly increased risk of falling into poverty [
5].
Current care of people with OA in Australia is inconsistent with clinical guidelines, with 57% of people not receiving appropriate care according to evidence-based quality indicators [
6]. As a result, 68% of Australians with arthritis report “doing badly” or “fairly badly” with respect to how their lives are affected by arthritis [
7]. Knee OA in Australia is mostly managed in general medical practice. A recent systematic review has highlighted that general practitioners (GPs) are hampered in their treatment of this chronic condition by lack of knowledge of non-surgical management options, and limited access to services that support the key recommended options such as lifestyle and behavioural changes [
8]. To address this gap, GPs and other health professionals [
9‐
11] have called for new models for OA primary care that provide clear clinical pathways and support networks to allow multi-disciplinary input and lifestyle counselling for ongoing self-management of OA. This trial aims to address a major evidence-to-practice gap in primary care management of OA by evaluating a new service delivery model implemented with an intervention targeting GP practice behaviours to improve the health of people with knee OA. Importantly, the model of service delivery aligns with key recommendations of established models of care in Australia [
10] and will provide important policy-relevant data to support implementation and scalability of these models.
Current clinical guidelines emphasise non-surgical treatments, coupled with appropriate pharmacologic care, as the cornerstone of OA management [
12,
13]. In particular, education and advice, exercise and physical activity, and weight management are the gold standards. Benefits of exercise were well-established in a 2015 Cochrane Review [
14] with effect sizes higher than, or comparable to, those of simple analgesics and oral non-steroidal anti-inflammatories [
15]. Patients with knee OA often report preferring exercise over drug treatments [
16], due to a lower risk profile. For those who are overweight or obese, weight loss is critical to improving overall health and joint symptoms [
17]. Meta-analysis suggests patients should reduce body weight by at least 5% to gain improvement in pain and function [
18], while a large RCT [
19] has provided further evidence of the benefits of ~ 10% weight loss in OA populations, particularly when diet is combined with exercise.
Analysis of Australian BEACH data [
20] from 487,000 GP consultations for OA found rates of drug prescription were much higher than rates of lifestyle management (79 vs 21/100 knee OA contacts). Most referrals were directly to orthopaedic surgeons (68%) with few to physiotherapists (18%). Among people with hip/knee OA referred for orthopaedic management at a major Australian tertiary hospital, 80% felt they had not been sufficiently educated about OA and 33% had not engaged in any core non-drug conservative treatment [
21]. A meta-analysis assessing OA care [
22] found that quality indicator pass rates were suboptimal particularly for non-drug, non-surgical treatment, demonstrating that this is a worldwide problem. Such gaps in care are highly relevant as poorer access to adequate information about OA, poorer perceived quality of care and poorer perceived GP knowledge about treatment options are associated with worse patient outcomes [
7].
Undertaking regular exercise and losing weight is difficult for many people with knee OA and requires long-term behaviour change, coupled with appropriate support. A scoping review highlighted many barriers to undertaking exercise, including lack of knowledge and/or incorrect beliefs about capabilities and consequences [
23]. Similarly, 89% of obese patients with knee OA consider lack of motivation to be their greatest barrier to weight loss [
24]. Effective communication and support from health professionals are vital for self-belief and sustained motivation [
25]. However, many clinicians typically practice within a biomedical framework that inadequately considers psychosocial factors that are important in disease control [
26]. In addition, time constraints in consultations and a lack of knowledge, skills and confidence in behavioural counselling are reported as barriers to optimal OA care delivery by GPs [
8].
The need for new, effective primary care models was identified as the research priority most likely to alleviate Australian OA burden at the 2012 Australian OA Summit [
27]. It was also identified in a White Paper by Arthritis Australia following stakeholder consultation. [
28] There is also a wealth of evidence to support the system-level benefits of the development and evaluation of models of care and their models of service delivery [
29], including a broad acceptance of this approach in Australia [
30]. As such we have performed extensive work to develop a new model of service delivery: the PARTNER model, and have designed an implementation plan for delivering the model in the current Australian primary care context. PARTNER is underpinned by the Chronic Care model [
31], evidence-based clinical practice guidelines [
12,
13], and informed by broad stakeholder input (consumers, GPs, physiotherapists, rheumatologists, nurses, behaviour change experts, policy makers, health insurers and consumer advocates) and the knowledge and experience of the Osteoarthritis Healthy Weight For Life Program (OAHWFL) [
32]. The implementation plan was designed using the ‘Behaviour Change Wheel’ and informed by the Theoretical Domains Framework [
33]. The implementation plan targets GPs via GP professional development modules and provision of a desktop EMR support tool, and the PARTNER model targets their patients who are referred to a centralised, remotely-delivered, multi-disciplinary Care Support Team (CST) for proven exercise, weight loss and pain management interventions. Further details on the theory underpinning the development of the PARTNER model of service delivery and the implementation plan will be presented in a separate paper.
The aim of this project is to implement the PARTNER model for people with knee OA in an Australian primary care setting and to evaluate the effectiveness and cost-effectiveness of the PARTNER model compared with usual care. We hypothesise the PARTNER model will be superior. We will also conduct a process evaluation to assess success of the implementation plan, PARTNER model fidelity, identify contextual influences on scalability and sustainability and identify cost considerations for scaling up the GP-level intervention and CST service throughout Australian primary care.
Discussion
Given knee OA is one of the most prevalent and disabling chronic diseases, even small reductions in ineffective practices with small improvements in care may lead to marked cost savings and reductions in the individual and societal burden of the disease. If the PARTNER intervention rationale is correct and implementation is successful, the results will have major significance for Australia with potential implications internationally, including potential relevance to low- and middle-income settings in the context of remotely-delivered care. The intervention is intentionally complex and targets both GPs and patients. By improving the capacity of GPs to offer their patients support for behavioural change through the CST, we hope to ensure uptake and maintenance of exercise and weight loss and better satisfaction with the care they are provided. We also hope to improve access to high-quality self-management support and behavioural counselling for patients, and by so doing improve adherence to effective conservative non-drug treatment including exercise and weight loss, thus reducing the OA burden.
We have clear intentions for subsequent implementation by planning to ensure the intervention is acceptable, feasible, aligned with current Australian models of care and to assess its cost-effectiveness. We will assess the barriers and facilitators to more widespread implementation and work closely with stakeholders including consumers, government and insurers to ensure policy recommendations stem from this work if found to be effective and cost-effective.
The PARTNER CST: Hayley Morey, Joanne Bolton, Kim Allison, Kelly Woosnam, Kirsty Tindal, Jane Evans, Liz Dixon, Chris Yeomans, Heidi Williams. Thank you to Birgit Schnelle for her assistance with the PARTNER pilot study.
The PARTNER Study Team: Karen Shuck, Charlotte Marshall, Michelle King, and Anna Wood.