J Huber1, J Hüsler2, P Dieppe3, K P Günther4, K Dreinhöfer5, A Judge6. 1. Department of Orthopaedics, Stadtspital Triemli, Birmendsdorferstr. 497, CH-8063 Zurich, Switzerland. Electronic address: joerg.huber2@triemli.zuerich.ch. 2. Institute for Mathematical Statistics, University of Bern, Sidlerstr. 5, CH-3012 Bern, Switzerland. 3. Peninsula College of Medicine and Dentistry, C420 Portland Square, University of Plymouth Campus, Drake Circus, Plymouth PL4 8 AA, UK. 4. Department of Orthopaedics, Gustav Carus University of Dresden, Fetscherstr. 76, D-01307 Dresden, Germany. 5. Centre of Musculosceletal Surgery, Charité, University of Berlin, Charitéplatz 1, D-10117 Berlin, Germany; Department of Orthopaedics, Traumatology and Sports Medicine, Medical Park Berlin Humboldtmühle, An der Mühle 2-9, D-13507 Berlin, Germany. 6. Botnar Research Centre, Institute of Musculosceletal Sciences, Nuffield Department of Orthopaedics, Rheumatology and Musculosceletal Sciences, University of Oxford, Windmill Road, Headington, OX3 7LD, UK.
Abstract
OBJECTIVE: To validate a new method to identify responders (relative effect per patient (REPP) >0.2) using the OMERACT-OARSI criteria as gold standard in a large multicentre sample. METHOD: The REPP ([score before - after treatment]/score before treatment) was calculated for 845 patients of a large multicenter European cohort study for THR. The patients with a REPP >0.2 were defined as responders. The responder rate was compared to the gold standard (OMERACT-OARSI criteria) using receiver operator characteristic (ROC) curve analysis for sensitivity, specificity and percentage of appropriately classified patients. RESULTS: With the criterion REPP>0.2 85.4% of the patients were classified as responders, applying the OARSI-OMERACT criteria 85.7%. The new method had 98.8% sensitivity, 94.2% specificity and 98.1% of the patients were correctly classified compared to the gold standard. CONCLUSION: The external validation showed a high sensitivity and also specificity of a new criterion to identify a responder compared to the gold standard method. It is simple and has no uncertainties due to a single classification criterion.
OBJECTIVE: To validate a new method to identify responders (relative effect per patient (REPP) >0.2) using the OMERACT-OARSI criteria as gold standard in a large multicentre sample. METHOD: The REPP ([score before - after treatment]/score before treatment) was calculated for 845 patients of a large multicenter European cohort study for THR. The patients with a REPP >0.2 were defined as responders. The responder rate was compared to the gold standard (OMERACT-OARSI criteria) using receiver operator characteristic (ROC) curve analysis for sensitivity, specificity and percentage of appropriately classified patients. RESULTS: With the criterion REPP>0.2 85.4% of the patients were classified as responders, applying the OARSI-OMERACT criteria 85.7%. The new method had 98.8% sensitivity, 94.2% specificity and 98.1% of the patients were correctly classified compared to the gold standard. CONCLUSION: The external validation showed a high sensitivity and also specificity of a new criterion to identify a responder compared to the gold standard method. It is simple and has no uncertainties due to a single classification criterion.
Authors: Hasan Raza Mohammad; Rachael Gooberman-Hill; Antonella Delmestri; John Broomfield; Rita Patel; Joerg Huber; Cesar Garriga; Christopher Eccleston; Rafael Pinedo-Villanueva; Tamer T Malak; Nigel Arden; Andrew Price; Vikki Wylde; Tim J Peters; Ashley W Blom; Andrew Judge Journal: PLoS One Date: 2021-12-31 Impact factor: 3.240