Objective: To test whether body weight-supported treadmill training (BWSTT) is effective in improving functional outcome of patients with Parkinson's disease. Design: Prospective crossover trial. Patients were randomized to receive either a 4-week program of BWSTT with up to 20% of their body weight supported followed by 4 weeks of conventional physical therapy (PT), or the same treatments in the opposite order. Medications for parkinsonism were not modified throughout the study. Setting: Inpatient rehabilitation unit for neurologic diseases. Subjects: Ten patients (5 men, 5 women) with Hoehn and Yahr stage 2.5 or 3 parkinsonism; mean age 67.6 years, mean duration of Parkinson's disease 4.2 years. Main Outcome Measures: The Unified Parkinson's Disease Rating Scale (UPDRS); ambulation endurance and speed (sec/10m), and number of steps for 10-meter walk. Results: The mean total UPDRS before/after BWSTT was 31.6/25.6, and before/after PT was 29.1/28.0. Analysis of covariance for improvement of UPDRS demonstrated a significant effect of type of therapy (F(1, 16) = 42.779, p < .0001) but not order of therapy (F(1, 16) = 0.157, p = .6971). Patients also had significantly greater improvement with BWSTT than with PT in ambulation speed (BWSTT, before/after = 10.0/8.3; PT, 9.5/8.9), and number of steps (BWSTT, 22.3/19.6; PT, 21.5/20.8). Conclusions: In persons with Parkinson's disease, treadmill training with body weight support produces greater improvement in activities of daily living, motor performance, and ambulation than does physical therapy.
