The Effectiveness of Personalized Robot-Assisted Rehabilitation on Fall Risk: A Retrospective Controlled Study with a...

Falls represent a severe public health issue and a primary cause of impairment across both geriatric and adults clinical populations, underscoring the importance of early and tailored prevention. Research evidence shows that technology-assisted rehabilitation may enhance balance and reduce fall risk; however, long-term real-world evidence remains limited. This retrospective study evaluated the impact of personalized balance training by hunova^® robotic platform on fall risk in a heterogeneous clinical cohort (n = 355; mean age 58.24 ± 19.63 years) comprising neurological and orthopedic conditions. Fall risk was assessed using the Silver Index, a sensor-based robotic posturographic score to predict 12-month fall probability at baseline (T0), after 6 weeks (T1), and at 6 months (T6). Robotic rehabilitation was provided to 162 patients (TREAT-G), while 193 served as a control group (NOTREAT-G). TREAT-G demonstrated a significant reduction in fall risk over time (p = 0.036), with improvements from T0-T1 (p < 0.001) and T0-T6 (p = 0.009). Domain-specific analysis showed significant increases in limits of stability, sit-to-stand, and gait speed (p < 0.001). Significant time × group interactions were observed for limits of stability (p = 0.004) and sit-to-stand (p < 0.001). Based on the occurrence of falls in the previous six months, the sample was then divided into 'Fallers' (at least one fall) and 'No-Fallers' (no falls). Subgroup analysis revealed significant improvements in 'Fallers' (p = 0.016) and 'No-Fallers' (p = 0.034) following treatment in contrast to untreated patients-both 'Fallers' and 'No-Fallers'. These results indicate that tailored robot-assisted therapy with hunova^® significantly reduces risk of fall and improves long-term dynamic balance. This study supports the clinical integration of robotic assessment and therapy as a robust tool for proactive fall prevention in diverse clinical settings.