|Purpose : The purpose of this study was to identify whether virtual reality-based exercise could improve on balance, gait and fall efficacy in patients with Parkinson`s disease. Methods : Ten patients with Parkinson`s disease were randomly divided into either an experimental or control group. The experimental subjects performed vertual reality-based exercise, whereas the control subjects performed conventional physical therapy for 4 weeks. The balance, gait and fall efficacy of all subjects were assessed by using the Measurement Training and Documentation (MTD) balance system, force platform system, Korean version of Berg Balance scale (K-BBS), 6 Minute Walk Test (6MWT), and Korean version of Fall efficacy scale (K-FES) at pre training and post training. Wilcoxon signed rank test was used to analyze change before and after intervention in intra-group. Mann Whitney U test was used to analyze changes of all variables in inter-groups. Results : Subjects in the experimental group showed significant improvements in difference of weight distribution, K-BBS scores, antero-posterior and medio-lateral sway length, ground reaction force (GRF), 6MWT, and step length following training. The changes of difference of weight distribution, K-BBS scores, AP Sway Length, GRF, 6MWT, step length and K-FES scores in the experimental group were significantly more than them of the control group. Conclusion : The result of this study suggest that virtual reality-based exercise training is an intervention to improve on balance, gait, and falls efficacy in patients with Parkinson`s disease.