Predicting patient satisfaction using the Oxford knee score: where do we draw the line?

Abstract

Objectives: The aim of this study was to identify threshold values in the pre- and post-operative Oxford knee score (OKS), and change in the score for differing levels of patient satisfaction with their total knee replacement (TKR).

Methods: We prospectively collected pre-operative and 1-year post-operative OKS for 2392 patients undergoing a TKR. Patient satisfaction was categorically assessed, according to whether they were: very satisfied, satisfied, neutral, and unsatisfied. Receiver operating characteristic curve analysis was used to identify thresholds in the OKS score that identified each level of patient satisfaction.

Results: The post-operative OKS was the most accurate predictor of the level of patient satisfaction (area under the curve = 0.86). Very satisfied patients had a threshold value in the post-operative OKS of ≥36, which decreased to ≥27 points for satisfied patients, and further still to ≤25 for unsatisfied patients.

Conclusion: The threshold values, we have identified for the different levels of satisfaction using the post-operative OKS, which is the most accurate predictor, can be used to predict level of patient satisfaction and give quantification of the OKS.