Beitragstitel | The long and short of it: outcome assessment using the SRS-instrument in patients with adult spinal deformity |
---|---|
Beitragscode | P022 |
Autoren | |
Präsentationsform | Poster |
Themengebiete |
|
Abstract |
Introduction Designed for patients with adolescent idiopathic scoliosis, the SRS-22 is also widely used as an outcome instrument in patients with adult spinal deformity (ASD), although its validity in ASD has not been adequately evaluated. A recent confirmatory factor analyses of its 20 non-management items in ASD patients revealed some consistently weak item-loadings. It was suggested that 4 items (3,14,15,17) be removed, to provide an improved, shorter version of the instrument. This study aims to evaluate the responsiveness of that version. Methods The analysis involved the data of 214 patients with ASD (45±20 years; 168 F, 46 M; 58 degenerative, 156 idiopathic; 130 surgical, 84 conservative) who had completed the following questionnaires within the European Spine Study Group’s (ESSG) multicentre prospective study: the SRS-22 and the Core Outcome Measures Index (COMI) upon inclusion and at 12 mo follow-up (FU); and a transition question indicating how the back problem had changed relative to 1 year ago. The latter used a Likert-scale (-7 to +7) and was the external criterion for treatment success in a Receiver Operating Characteristics (ROC) curve analysis, with ≥+4 (i.e., at least “moderately better”) being a “good” outcome. Results The SRS 16-item and SRS 20-item scores showed similarly strong correlations with the COMI scores (r=-0.79 and r= 0.78, respectively), indicating construct validity for both. The correlations between "change in the back problem" and SRS change scores (baseline to 12mo FU) were r=0.57 for the 16-item SRS and r=0.55 for the 20-item SRS. The standardized response means (change score/SD) for patients with a "good" outcome were 1.13 and 1.11 for the 16-item and 20-item instruments, respectively (= large effect sizes); the areas under the ROC curve were 0.85 and 0.83 respectively (16 v 20, p=0.053). The 16-item version discriminated slightly better between good and poor outcomes for the SRS sub-domains. Conclusion With the “change in back problem” serving as external criterion, the 16-item instrument showed as good if not better external responsiveness compared with the 20-item. The 16-item instrument seems to be well able to detect important change in adult spinal deformity. Since it is shorter, more responsive, and has already been shown to have better structural validity than the 20-item version, we recommend its use in future studies of patients with ASD. Together with the 2 management items, this should form a new SRS-18 instrument. |