Evidence based medicine (EBM)
We undertake systematic reviews and meta-analyses in order to robustly interpret research evidence, to inform and influence our programme of research, the scientific community and clinical practice.
We have led the development of the European evidence based recommendations for the diagnosis and treatment of osteoarthritis, gout and CPPD related arthritis, and the global evidence based treatment guidelines for hip and knee osteoarthritis (OA).
We have a track record of systematic review and meta-analysis of randomised controlled trials to assess clinical effectiveness and cost-effectiveness of therapeutic interventions. Our studies for paracetamol (Ann Rheum Dis 2004;63:901) and topical NSAIDs (BMJ 2004;329:324) have been widely used in clinical practice to guide the treatments in osteoarthritis. They were part of the evidence base for the development of the national (NICE 2008) and international (Ann Rheum Dis 2005;64:669) treatment guidelines for osteoarthritis. Our recent meta-analysis on the placebo effect (Ann Rheum Dis 2008;67:1716) confirmed that placebo is effective to relieve osteoarthritic pain and its effect increases when placebo is delivered through needles instead of pills. We are currently extending this work to compare the magnitude and determinents of placebo effects in fibromyalgia with those in osteoarthritis. The Evidence Based Osteoarthritis Research Database (eBOARD) is developing a comprehensive, coherent research database containing clinical effectiveness and cost-effectiveness data for all available therapies in osteoarthritis.
Diagnosis and classification
Our EBM research has been extended to the assessment of the validity and reliability of diagnostic tests in OA and other musculoskeletal conditions. The results have been used to develop the European EBM recommendations for the diagnosis of OA, gout and CPPD. More recently we are involved in the OARSI-FDA Working Group to evaluate the performance of MRI in the early diagnosis of knee osteoarthritis.
A systematic review is currently undertaken to assess the performance and application of quantitative sensory testing (QST) in the identification of pain phenotypes in osteoarthritis.
Three major modifiable risk factors (body mass index, occupational risk and knee injury) have been identified and their relative risks associated with the development of knee OA have been systematically reviewed. This has led to the development of the Nottingham Knee OA Risk Prediction Models and the estimates of the potential risk reduction in different countries due to the control of these risk factors (Ann Rheum Dis 2011;70:1599). In contrast, smoking has been confirmed no "protective effect" for OA. The negative association found from previous epidemiological studies is more likely false due to the selection bias of the hospital controls associated with smoking (Ann Rheum Dis 2011;70:1231).
[Back to top]