Do Patient Predicted Oxford Scores Correlate with Actual Outcomes


Patient reported outcome measures (PROMs) are utilised as a means of assessing the subjective experience of the patient. The Oxford Hip (OHS) and Knee Scores (OKS) are joint specific instruments designed to quantify the patient’s experience before and after their surgery. Uniquely, we asked patients attending pre-operatively to estimate a predicted Oxford Score (pOS). To our knowledge this is the first study to utilise a pOS and the relationship to actual outcome 12 months post joint replacement.


To assess the relationship between the predicted OHS and OKS collected pre- operatively and compare the patient’s score reported 12 months post-operatively.

Study Design & Methods

This is a prospective cohort study of a database of 3853 patients over 6 years undergoing hip and knee arthroplasty (November 2012 – March 2018). Pre and post- operative variables were available for comparison for 389 patients (10%). A Pearson Correlation Co-Efficient was calculated to assess the relationship between the absolute values of these scores.


Comparing the predictive Oxford Scores to the 12 month scores yielded an r value of 0.14 (Pearson Correlation Co-Efficient). There was no relationship between the patient pOS and the combined OHS and OKS at 12 months. The mean pOS for total hip replacement was 47 compared to an actual mean score of 39 at 12 months. The mean pOS for total knee replacement was 42 compared to an actual mean score of 37 at 12 months.


Our study demonstrates there is no relationship between the patients’ perception of their future outcome following arthroplasty surgery using a predicted Oxford score and further self-evaluation at 12 months. The majority of patients over estimate long term clinical outcomes. These predicted scores may be a means of flagging patients who over estimate outcomes so that expectation may be moderated pre-operatively.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *