Derivation of a preliminary clinical prediction rule for identifying a subgroup of patients with low back pain likely to benefit from Pilates-based exercise.

STUDY DESIGN Prospective cohort study. OBJECTIVE To derive a preliminary clinical prediction rule for identifying a subgroup of patients with low back pain (LBP) likely to benefit from Pilates-based exercise. BACKGROUND Pilates-based exercise has been shown to be effective for patients with LBP. However, no previous work has characterized patient attributes for those most likely to have a successful outcome from treatment. METHODS Ninety-six individuals with nonspecific LBP participated in the study. Treatment response was categorized based on changes in the Oswestry Disability Questionnaire scores after 8 weeks. An improvement of 50% or greater was categorized as achieving a successful outcome. Thirty-seven variables measured at baseline were analyzed with univariate and multivariate methods to derive a clinical prediction rule for successful outcome with Pilates exercise. Accuracy statistics, receiver-operator curves, and regression analyses were used to determine the association between standardized examination variables and treatment response status. RESULTS Ninety-five of 96 participants completed the study, with 51 (53.7%) achieving a successful outcome. A preliminary clinical prediction rule with 5 variables was identified: total trunk flexion range of motion of 70° or less, duration of current symptoms of 6 months or less, no leg symptoms in the last week, body mass index of 25 kg/m2 or greater, and left or right hip average rotation range of motion of 25° or greater. If 3 or more of the 5 attributes were present (positive likelihood ratio, 10.64), the probability of experiencing a successful outcome increased from 54% to 93%. CONCLUSION These data provide preliminary evidence to suggest that the response to Pilates-based exercise in patients with LBP can be predicted from variables collected from the clinical examination. If subsequently validated in a randomized clinical trial, this prediction rule may be useful to improve clinical decision making in determining which patients are most likely to benefit from Pilates-based exercise.