Background. It is widely believed that exercise improves mobility in people with Parkinson's disease (PD). However, it is difficult to determine whether a specific type of exercise is the most effective. The purpose of this study was to determine which outcome measures were sensitive to exercise intervention and to explore the effects of two different exercise programs for improving mobility in patients with PD. Methods. Participants were randomized into either the Agility Boot Camp (ABC) or treadmill training; 4x/week for 4 weeks. Outcome measures were grouped by the International Classification of Function/Disability (ICF). To determine the responsiveness to exercise, we calculated the standardized response means. t-tests were used to compare the relative benefits of each exercise program. Results. Four of five variables at the structure/function level changed after exercise: turn duration (P = 0.03), stride velocity (P = 0.001), peak arm speed (P = 0.001), and horizontal trunk ROM during gait (P = 0.02). Most measures improved similarly for both interventions. The only variable that detected a difference between groups was postural sway in ABC group (F = 4.95; P = 0.03). Conclusion. Outcome measures at ICF body structure/function level were most effective at detecting change after exercise and revealing differences in improvement between interventions.