Virtual reality (VR) may be useful during rehabilitation of service members with persistent vestibular impairment following concussion. Thirty-eight active duty US military service members with persistent balance impairment resulting from concussion were randomized into three groups [Conventional Vestibular Physical Therapy (CVPT, n = 13), Virtual Reality Vestibular Physical Therapy (VRVPT, n = 12), and Hybrid Virtual Reality and Conventional Vestibular Physical Therapy (HybridVPT, n = 13)] and were treated twice weekly for 6 weeks. Changes in clinical measures such as Activities-specific Balance Confidence (ABC) Scale, Dizziness Handicap Inventory (DHI), Functional Gait Assessment (FGA), and Sensory Organization Test (SOT) were assessed from pre-, mid-, and post-treatment scores. A significant main time effect was observed demonstrating clinical improvement over time (ABC: p < .001, η2p = .54; DHI: p < .001, η2p = .57; FGA: p < .001, η2p = .74; SOT: p < .001, η2p = .35). Both CVPT and HybridVPT groups demonstrated significant improvements in patient-reported confidence and function earlier in the treatment course (p < .005). FGA significantly and incrementally improved at each assessment time point in all treatment groups. The SOT significantly improved early in treatment in the CVPT group only and pre-to-post-treatment in the CVPT and VRVPT groups only. The HybridVPT group did not demonstrate any significant improvement with time in the instrumented SOT measure. In the comparison of pre-to-post-effects of VRVPT and HybridVPT effects compared to CVPT, there was no clear superiority or inferiority observed in either of the experimental treatments. This preliminary work shows initial efficacy of using VR-based therapy for concussed individuals allowing future work to personalize treatment that may improve adherence and engagement to therapy.