Study Design. Analysis of longitudinal data collected prospectively from patients seen in 27 National Spine Network member centers across the United States.
Objective. To evaluate the responsiveness of the Oswestry Disability Index, MODEMS scales, and all scales and summary scales of the MOS Short-Form 36 (SF-36) for patients with low back pain/leg symptoms.
Summary of Background Data. The responsiveness of general and condition-specific health status instruments is a key concept for clinicians and scientists. Various authors have explored responsiveness in common surveys used to assess spine patients. Although it is generally believed that condition-specific measures are more responsive to change in the condition under study, in the case of low back pain, most authors agree that further exploration is necessary.
Methods. Patients with diagnoses of herniated disc, spinal stenosis, and spondylosis from the National Spine Network database who completed baseline and 3-month follow-up surveys were analyzed. Patient-provider consensus regarding improvement, worsening, or no change in the condition was selected as the external criterion. Responsiveness was evaluated using ROC curve analysis and effect size calculations.
Results. Nine hundred and seventy patients had complete data at baseline and 3 months. At follow-up, 68% of the patients had consensus improvement. Based on ROC analysis, scales assessing pain were significantly more responsive than scales assessing function. There were no significant differences between the condition-specific scales and their equivalent general-health counterpart. The scales with the highest probabilities of correctly identifying patient’s improvement were: the condition-specific pain scale from MODEMS (PAIN, ROC = 0.758); the combined pain and function scale from MODEMS (MPDL, ROC = 0.755); the general pain scale from the SF-36 (BP, ROC = 0.753); the combined pain and function scale from the SF-36 (PCS, ROC = 0.745); the condition-specific function measure from the Oswestry (ODI, ROC = 0.723); and the physical function measure from the SF-36 (PF, ROC = 0.721). A similar rank order was typically maintained with effect size calculations. Results were nearly identical in patients with multiple non-spine-related comorbidities and in patients with high degrees of perceived disability. The BP scale was most responsive to worsening of symptoms.
Conclusion. For studies of patients with low back problems, the general SF-36 may be a sufficient measure of health status and patient function, without the need for additional condition-specific instruments. Pain scales appear to be the most responsive measures in patients with low back pain.
From *The Spine Center
and †Section of Orthopedics at Dartmouth Hitchcock Medical Center, Lebanon,
‡Department of Community and Family Medicine, Dartmouth Medical School, Hanover,
and §The Center for the Evaluative Clinical Sciences at
∥Dartmouth Medical School, Hanover, New Hampshire.
Supported in part by: The National Institute of Arthritis and Musculoskeletal and Skin Diseases and the Office of Research on Women’s Health, the National Institutes of Health, and the National Institute of Occupational Safety and Health, the Centers for Disease Control and Prevention NIAMS # AR45444-01A1, and the members of the National Spine Network.
Acknowledgment date: March 25, 2002.
First revision date: July 29, 2002.
Acceptance date: August 19, 2002.
The manuscript submitted does not contain information about medical device(s)/drug(s).
No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.
Address correspondence and reprint requests to Thomas Walsh, MS, PT, OCS, Dip MDT, The Spine Center, Dartmouth Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756, USA; E-mail: Thom@Hitchcock.org