The reliability and validity of the self-reported patient-specific index for total hip arthroplasty

Abstract

Background: The Patient-Specific Index is unique in that it reflects how individual patients weigh concerns in rating the outcome of total hip arthroplasty. The Patient-Specific Index was originally administered by an interviewer, which is not always feasible and can be costly. The purposes of the present study were (1) to create a self-reported version of the Patient-Specific Index, (2) to determine the reliability of this new self-reported version, and (3) to determine the relationship between the scores on the new self-reported version and those on the original interviewer-administered version.

Methods: A self-reported version of the Patient-Specific Index was developed, and a pilot test was performed on ten patients. Patients who were scheduled for a total hip arthroplasty or who had recently had a total hip arthroplasty were eligible for the reliability and validity testing. A copy of the new self-reported Patient-Specific Index was mailed to the patients, and they completed it independently. The patients' ratings of the importance and severity of twenty-four concerns prior to total hip arthroplasty were added together to create a summary Patient-Specific Index score. To determine test-retest reliability, patients completed the self-reported Patient-Specific Index a second time, two weeks later. To determine criterion validity, participants also completed the interviewer-administered Patient-Specific Index.

Results: Fifty-five patients completed the study. The random-effects intraclass correlation test-retest coefficient was 0.79 (greater than 0.75 represents excellent reliability). The mean Patient-Specific Index scores on the self-reported version and on the interviewer-administered version were 173 and 165 points, respectively (Student t test, p = 0.45). The self-reported Patient-Specific Index was concordant with the interviewer-administered Patient-Specific Index (intraclass correlation coefficient, 0.78).

Conclusions: We concluded that a self-reported version of the Patient-Specific Index, which focuses on the concerns of individuals, is reliable and has criterion validity compared with an interviewer-administered version.