OSWESTRY HIP SCORE: A PATIENT ASSESSED TOOL FOR COMPLETE EVALUATION OF HIP FUNCTION

Abstract

Objective: Assessment of hip function is done by surgeon and few patient based tools. These patient assessed scores do not measure range of motion. The American Academy of Orthopaedic Surgeons has outlined pain, mobility and range of motion as three fundamental aspects of joint assessment. We aimed to validate Oswestry hip score which was developed as a patient-completed self-assessment to provide both Harris hip score and Merle d’Aubigne hip score with added content to estimate hip range of motion.

Methods: 144 patients completed the Oswestry hip score, WOMAC Index and the Oxford hip score at two different occasions. The patients were seen two weeks later and filled the Oswestry hip score and a surgeon filled the Harris hip score. The study included 80 females and 64 males with a mean age of 62 years range (32–91). We assessed the reliability, validity and responsiveness of this new tool by comparing the individual domains of the Oswestry hip score to similar domains of the WOMAC, Harris hip score and Oxford hip score.

Results: The reliability of this new score was established by the Intraclass Correlation Coefficient. Internal consistency was measured by Cronbach’s alpha. The Cronbach’s alpha was 0.7, which is considered a good measure of internal consistency. Content validity of the Oswestry hip score was established by the validated domains of pain, function and range of motion of the Harris hip score. Analysis of frequency of response distribution showed normal floor and ceiling effect for any of the domains of the Oswestry hip score. Multi-method multitrait matrix analysis was used to establish the construct validity of the Oswestry hip score. There was good correlation between pain and function domains (p< 0.001). Moderate correlation was found among clinical assessment of hip movement and movement domains of Oswestry hip score (Pearson’s r=0.55; p0.001). The responsiveness of the Oswestry hip score was measured with Cohen’s effect size. An effect size of > 0.8 is considered large. The mean effect size was 2.06 (1.36–2.97).

Conclusion: A positive construct validity and high correlation with WOMAC, Oxford Hip Score and Harris hip score shows that the Oswestry hip score can give an adequate measure of hip joint function. An effect size of 2.0 shows good responsiveness of the Oswestry hip score. The Oswestry hip score can be completed by patients themselves and is therefore ideal for long-term and large scale collection of clinical outcome data.