Osteoarthritis (OA) is the leading cause of musculoskeletal pain and disability and is the third leading cause of life-years lost due to disability in Australia, only behind depression and dementia . The annual total cost of arthritic disease in Australia is estimated at $24 billion , with the knee joint contributing substantially to this overall cost. The prevalence of OA in people aged over 55 years is 20–26% and rising, with arthritis rates expected to increase by 30% over the next 40 years . The pain and suffering endured by patients as a result of OA decreases their quality of life, with the annual burden of disease costs ($12 billion in Australia) being half the total costs associated with this condition . Pain associated with daily activities such as walking and stair-climbing ultimately leads to profoundly reduced functional independence .
The patellofemoral joint (PFJ) is one of the three knee joint compartments. Awareness of its importance in the OA process has been raised by the increasing use of lateral and skyline x-rays in recent times. Research has revealed that PFJ OA is more common than previously thought. In a community-based study of knee OA (N = 218), the frequency of radiographic osteophytes was greater in the PFJ (65% knees) than in the tibiofemoral joint (TFJ) (55% knees) . Furthermore, in people with knee pain (N = 777), the most common compartmental distribution of radiographic OA was a combination of TFJ and PFJ disease (40%), followed by isolated PFJ OA (24%), and isolated TFJ disease (4%) . Within the PFJ, the lateral compartment is more frequently affected by the OA process than the medial [5, 6]. Importantly, the presence of baseline PFJ OA predicts structural deterioration in the TFJ compartment over 30 months (OR 2.31, 95% CI 1.37, 3.88) .
The PFJ is an important source of symptoms associated with knee OA . Knee pain has been found to be significantly associated with PFJ osteophytes (OR 2.25, 95%CI 1.06, 4.77), but not TFJ osteophytes (OR 1.19, 95% CI 0.46, 3.09) , suggesting that the PFJ may be a more important source of knee pain than the TFJ. Hunter et al  noted that increased pain and poorer function was associated with reduced cartilage volume in the patella, but not in the femur nor the tibia. Other authors have confirmed the relationship between radiographic PFJ OA and knee pain [11–13].
Management strategies for knee OA have traditionally focussed on alleviating symptoms, primarily using drug therapies or surgery. A meta-analysis of OA trials highlights this, with most trials evaluating drug treatments (60%) or surgical procedures (26%) . OA experts have highlighted the overall dearth of quality evidence to support the use of non-pharmacological interventions such as physiotherapy. Despite this, knee OA clinical guidelines recommend that conservative treatments be included as a first line strategy for the optimal management of the disease [15, 16]. Physiotherapy is a conservative intervention, which is non-toxic, inexpensive and promotes physical activity and self management through exercise. Therefore, rigorous randomised clinical trials (RCTs) that evaluate the efficacy of physiotherapy are clearly needed, to better guide clinical decision-making.
Given the heterogeneity of knee OA with regard to aetiology, clinical presentation and natural history, guidelines also recommend the tailoring of knee OA treatments to the location of joint damage in order to optimise treatment outcomes [15, 16]. However, most trials of physiotherapy for knee OA have not been targeted to disease subgroups, with participant selection typically based on the presence of non-specific knee pain and radiographic changes anywhere on an anteroposterior radiograph. While a plethora of evidence attests to the benefits of exercise for patients with predominant TFJ OA  there is no level I evidence and only one RCT  specifically addressing the problem of PFJ OA. The dearth of evidence for a compartment-specific treatment for PFJ OA necessitates our proposed study to establish the efficacy of a compartment-specific physiotherapy treatment using the rigour of a RCT.
While there is little known about the physical impairments associated with PFJ OA, there are several RCTs that have evaluated physical interventions for PFJ pain in younger adults (patellofemoral pain syndrome, or anterior knee pain). We have previously conducted a double blind, placebo-controlled RCT , which demonstrated the efficacy of a targeted physiotherapy program for this patient population. The targeted treatment involved (i) quadriceps muscle retraining; (ii) patellar taping; (iii) manual PFJ and soft tissue mobilisation; and (iv) hip muscle retraining. We have recently confirmed the beneficial effects of this targeted physiotherapy approach on pain and physical function in another population of young adults with PFJ pain . Therefore, we are proposing to evaluate a similar, targeted physiotherapy intervention for people with PFJ OA.
This project aims to evaluate whether a physiotherapy treatment, targeted to the PFJ and based on successful treatment for PFJ pain in younger populations, results in greater improvements in pain and physical function than a physiotherapy education intervention in participants with symptomatic and radiographic PFJ OA.