We obtained pre-operative and six-month post-operative Oxford hip (OHS) and knee scores (OKS) for 1523 patients who underwent total hip replacement and 1784 patients who underwent total knee replacement. They all also completed a six-month satisfaction question. Scatter plots showed no relationship between pre-operative Oxford scores and six-month satisfaction scores. Spearman's rank correlation coefficients were -0.04 (95% confidence interval (CI) -0.09 to 0.01) between OHS and satisfaction and 0.04 (95% CI -0.01 to 0.08) between OKS and satisfaction. A receiver operating characteristic (ROC) curve analysis was used to identify a cut-off point for the pre-operative OHS/OKS that identifies whether or not a patient is satisfied with surgery. We obtained an area under the ROC curve of 0.51 (95% CI 0.45 to 0.56) for hip replacement and 0.56 (95% CI 0.51 to 0.60) for knee replacement, indicating that pre-operative Oxford scores have no predictive accuracy in distinguishing satisfied from dissatisfied patients. In the NHS widespread attempts are being made to use patient-reported outcome measures (PROMs) data for the purpose of prioritising patients for surgery. Oxford hip and knee scores have no predictive accuracy in relation to post-operative patient satisfaction. This evidence does not support their current use in prioritising access to care.