Investigating dose-response patterns in virtual reality rehabilitation: a pilot study of patient satisfaction in subacute stroke.

Background: Virtual reality (VR) rehabilitation shows promise for stroke recovery, but optimal dosage remains unclear. We examined the relationship between VR therapy intensity and patient satisfaction, while assessing methodological challenges in dose-response research. Objective: To investigate relationships between VR rehabilitation dosage (sessions, duration) and patient satisfaction in subacute stroke patients, and identify requirements for future definitive studies. Methods: We analyzed data from 19 subacute ischemic stroke patients who received VR rehabilitation using VR Vitalis® Pro system (January-December 2024). Patient satisfaction was measured with the User Satisfaction Evaluation Questionnaire (USEQ). We examined correlations between VR dosage variables and satisfaction, then conducted post-hoc power analysis and confounding assessment. Results: Patients averaged 25.0 ± 6.8 USEQ points, with 68% achieving high satisfaction. They completed 4.2 ± 4.1 VR sessions (range 1-13), but 58% received only 1-2 sessions due to clinical factors. No significant correlation emerged between sessions and satisfaction (r = 0.18, p = 0.47). Post-hoc analysis revealed only 11% statistical power for the observed effect. VR sessions strongly correlated with hospital stay (r = 0.664, p = 0.002), indicating confounding by clinical severity rather than research-controlled dosage. Conclusions: Our underpowered study (11% power) with substantial clinical confounding cannot determine dose-response relationships or inform practice. Future studies need larger samples (n ≥ 85) with randomized dosage allocation. Our main contribution is demonstrating methodological requirements for rigorous VR dose-response research.