The identification of preliminary prognostic indicators that predict treatment response for exercise therapy in patients with nonspecific chronic low back pain: A multiple-arm cohort study design

Abstract

BACKGROUND: For patients with nonspecific chronic low back pain (CLBP), exercise therapy is stated to be the most effective intervention strategy but it is unclear which kind of exercise therapy is most beneficial. OBJECTIVE: To identify preliminary prognostic indicators that predict outcome for exercise therapy in patients with nonspecific CLBP. METHODS: Patients were recruited in two hospitals and received 18 intervention sessions: stabilization therapy, isometric training therapy or a combination therapy. The primary outcome measure was the change in the Modified Low Back Pain Disability Questionnaire after nine weeks. RESULTS: A total of 59 patients completed the study which represents a statistical power of 90%. In total, 30 patients were categorized as having treatment success and 29 as treatment failure. After using regression analyses to determine the association between standardized examination variables and treatment response status, prognostic indicators were identified for predicting therapy success (positive likelihood ratio [LR], 3.8) and failure (negative LR, 0.19). CONCLUSIONS: The most important variables were the prone instability test, pelvic tilt test, straight leg raise, body weight, visual analogue scale and the short form 36 health survey.