Background: The ability for backward walking is considered to be necessary for the neuromuscular control and maintenance of balance in daily ambulatory activity. This study aimed to determine the effect of backward walking training on the walking speed and balance control in patients with hemiplegia. Methods: Fourteen patients with hemiplegia were randomly allocated to an experimental and control groups of seven patients each. For the experimental group, we performed both conventional training and backward walking training, and conventional training only for the control group. The conventional training programs for the 2 groups were conducted for 30 min, twice a day, 5 times a week for 4 weeks, and backward walking training for the experimental group was conducted for 30 min, 3 times a week. The outcomes were assessed using the functional reach test (FRT), timed up-and-go (TUG) Test, and the 10 meter walk time test (10mWT). Result: A comparison of the FRT, TUG test, and 10mWT scores obtained before and after the 4-week treatment revealed statistically significant differences (p<.05) for the experimental group; however, there was no such difference in the case of the control group (p>.05). On assessment after the 4-week treatment, statistically significant differences were noted in the TUG test and 10mWT scores of the experimental group (p<.05). Conclusion: Our findings suggest that backward walking training is an effective clinical strategy for improving the walking speed and functional mobility of patients with hemiplegia.