Evidence of validity for the hip outcome score in hip arthroscopy

Abstract

Purpose: The purpose of this study was to provide evidence of validity for the Hip Outcome Score (HOS) as an outcome instrument in hip arthroscopy.

Methods: We mailed the short form 36 (SF-36) and HOS questionnaires, as well as questionnaires regarding self-reported current activity level (normal, nearly normal, abnormal, or severely abnormal) and self-reported surgical outcome (excellent/good or fair/poor), to 337 subjects. Medical records were used to collect surgical and demographic information. Two groups were formed for those above and below the median age. Pearson correlation coefficients were used to assess the relation between the SF-36 scores and HOS scores. One-way analysis of variance was used to determine whether HOS scores differed according to current level of function, surgical outcome, and age.

Results: Of the subjects, 116 (34%) returned the questionnaire material. Nine of these subjects reported having surgery after August 2003 and were excluded. The data analysis was therefore performed on 107 subjects (32%). Within this group, there were 56 female patients (52%) and 51 male patients (48%), with a mean age of 42 years (median, 44.2 years; range, 14 to 79 years; SD, 14) and mean time to follow-up of 3.1 years (range, 2 to 4.6 years; SD, 0.49). The HOS activities of daily living (ADL) and sports subscales had a high correlation to the SF-36 physical function subscale (r = 0.86 and r = 0.84, respectively) and physical component summary score (r = 0.80 and r = 0.81, respectively) and a significantly (P < .005) lower correlation to the mental health subscale (r = 0.41 and r = 0.43, respectively) and mental component summary score (r = 0.17 and r = 0.18, respectively). HOS ADL and sports subscale scores were significantly different based on current activity level, surgical outcome, and age (P < .002).

Conclusions: This study provides evidence of validity for the HOS in a sample of subjects at a mean of 3 years after hip arthroscopy. As hypothesized, the HOS scores had a high correlation to measures of physical function and a low correlation to measures of mental health. The HOS scores were different based on subjects' reported current activity level, reported surgical outcome, and age. The results of this study support the use of the HOS ADL and sports subscales as a self-report outcome instrument for hip arthroscopy.

Level of evidence: Level III, development of diagnostic criteria in a study of nonconsecutive patients.