This 2007 annotation by Murray et al. addresses a decade of confusion surrounding the Oxford Hip Score and Oxford Knee Score. It clarifies the correct scoring convention, missing data rules, statistical approach, and clinical interpretation thresholds. It serves as the definitive methodological reference for anyone using these instruments in research or clinical audit.
When you design a study or audit using Oxford scores, three rules govern everything: use the 0–48 scale, collect a baseline score, and analyze change rather than absolute post-operative values.
A 2-point difference between surgical approaches in one THR study challenges the 3–5 point MCID estimate — meaning a statistically significant result in a large registry study may represent a real patient-perceived difference even if it looks clinically trivial on paper. For power calculations, budget for at least 100 patients per group when Oxford scores are the primary outcome. And usually more.
The authors developed these scores specifically to minimize the influence of comorbidities on joint-specific outcomes, which is why the Oxford instruments outperform generic health measures (SF-12, WOMAC) for detecting change after arthroplasty.
This 2007 annotation by Murray et al. addresses a decade of confusion surrounding the Oxford Hip Score and Oxford Knee Score. It clarifies the correct scoring convention, missing data rules, statistical approach, and clinical interpretation thresholds. It serves as the definitive methodological reference for anyone using these instruments in research or clinical audit.
When you design a study or audit using Oxford scores, three rules govern everything: use the 0–48 scale, collect a baseline score, and analyze change rather than absolute post-operative values.
A 2-point difference between surgical approaches in one THR study challenges the 3–5 point MCID estimate — meaning a statistically significant result in a large registry study may represent a real patient-perceived difference even if it looks clinically trivial on paper. For power calculations, budget for at least 100 patients per group when Oxford scores are the primary outcome. And usually more.
The authors developed these scores specifically to minimize the influence of comorbidities on joint-specific outcomes, which is why the Oxford instruments outperform generic health measures (SF-12, WOMAC) for detecting change after arthroplasty.