Neurorehabilitation including Virtual-Reality-Based Balance Therapy: Factors Associated with Training Response

Background: Virtual reality (VR) therapy is increasingly used and has shown encouraging effects. Yet, it is unknown which patients respond best to VR-based balance therapy as part of neurorehabilitation. Methods: Data from 30 persons with stroke and 51 persons with multiple sclerosis who performed three to four weeks of VR-based balance therapy during in-patient rehabilitation were analysed. Participants were divided into responders and nonresponders based on achievement of the minimal clinically important difference in functional balance post intervention. Measures of balance, trunk function, mobility, gait, motivation, and exergame parameters were compared between groups. Results: Post intervention, all clinical measurements significantly improved (p < 0.05; effect size: 0.45-0.59). Participants that achieved the minimal clinically important difference in functional balance (n = 49; 60%) had significantly lower preintervention functional and dynamic balance (median(IQR): 39(27-46) versus 45(37-50); p = 0.02 and 11(6-15) versus 16(11-18); p = 0.03). They spent less time on higher difficulty exercises (11(8-17) versus 14.5(10-12); p = 0.03) and demonstrated increased motivation over time compared with nonresponders (1(-1-5) versus -2(-7-3); p = 0.03). Conclusion: Lower baseline balance ability, spending more time on adequately challenging exercises, and increased motivation potentially influence response to therapy. These factors can support the personalisation of VR-based balance therapy.