脊柱与神经外科杂志

Responsiveness of the Balanced Inventory for Spinal Disorders, A Questionnaire for Evaluation of Outcomes in Patients with various Spinal Disorders

Elisabeth Svensson, Birgitta Schillberg, Xin Zhao and Bo Nystrm

Responsiveness of the Balanced Inventory for Spinal Disorders, A Questionnaire for Evaluation of Outcomes in Patients with various Spinal Disorders

Background: The Balanced Inventory for Spinal disorders, BIS, is a questionnaire for evaluating the extent to which spinal disorders affect perceived pain, physical and mental health, social life and quality of life. The validity and the test-retest reliability of assessments have previously been reported.

Objective: To evaluate the responsiveness of the BIS items and of comparable items of the Short-Form-36 (SF-36), the Oswestry Disability Index (ODI), and the European Quality-of-Life Scale (EQ-5D). The patient-reported levels of change by transition scales were also studied.

Design: Pre- and post-surgery assessments were completed by 101 patients before and four months after spinal surgery. The postsurgery BIS had additional transition item scales.

Methods: The statistical method used takes account of the properties of paired ordered categorical data from scale assessments, and provides measures of the systematic group change separately from a measure of additional individual variations. Classical test of change in paired assessments was also performed.

Results: The pre-and-post-assessments on the BIS items showed a high level of responsiveness. Between 55 and 72 percentage units more patients rated lower than higher levels of pain or affections in the core items of pain, physical health, social life, the way of living, and of the quality of life four months after surgery.

Conclusions: The responsiveness of the BIS was confirmed by the significant measures of systematic group change. One advantage of the use of transition scale over analysis of pre-and post-treatment data is that the risk of calculating differences and of applying parametric statistical methods is avoided. The responsiveness of items of the SF-36 and the ODI were similar regarding the systematic group changes, but showed higher levels of individual variations. The EQ-5D is not recommended for evaluation of change because of few scale categories.