Abstract
All patients undergoing knee arthroplasty at our institution complete Oxford Knee Scoring (OKS) at nurse-led pre-assessment and again at an admission physiotherapy visit on the ward which may be up to 2 months later.
The aim of this study is to establish the extended reproducibility of the OKS by statistical analysis of scores taken at these intervals.
44 patients were required to achieve a 90 % probability to detect a difference at a two-sided 5 % significance level with a minimum clinical difference of 3 points, a cut off used in previous works regarding the use of OKS.
Both the overall population means and the differences between individual questions were analysed by a paired samples t test and a Wilcoxon Signed Ranks Test respectively.
Mean interval between attendance for pre-assessment and admission visit was 16 days (7–60).
A statistically significant result at the 5% level was observed for the t test t= 2.197 (44df), p= 0.03. OKS at pre-assessment was lower than at admission to the ward by 1.1 point. (−2.1 – 0.9 95% CI).
Analysis of difference between individual questions revealed only three of the pairs achieved statistical significance and in each case, the difference was less than 3 scoring units.
This study demonstrates that although there is a difference in total scoring using the OKS between two patient episodes prior to arthroplasty, a clinically relevant difference is not detected, and neither is a statistically significant difference detected when all scoring steps are analysed.
The original validation of the OKS was obtained using test-retest reproducibility over a 24 hour period. This work shows that the OKS is robust to violations in reproducibility at duration much greater than this and for practical purposes is valid if taken at any point during the pre-admission phase of care.
Correspondence should be addressed to: BASK c/o BOA, at the Royal College of Surgeons, 35–43 Lincoln’s Inn Fields, London, WC2A 3PE, England.
