Mirror-Based Visual Feedback Versus Conventional Sit-To-Stand Training in Sub-Acute Stroke: A Quasi-Experimental Study

Abstract

Background: Adults with stroke rise from a chair with more weight on the non-paretic leg and cannot sense the imbalance, so verbal correction may leave it unchanged. A full-length mirror gives a visual signal without instruments. We tested whether mirror feedback added to dose-matched sit-to-stand training improves performance in sub-acute stroke.
Methods: In this quasi-experimental study, 68 adults 1 to 6 months after a first stroke were allocated by alternating enrolment to mirror feedback (n = 34) or identical training with the mirror covered (n = 34), 5 days a week for 4 weeks, with feedback faded in weeks 3 and 4. A blinded assessor measured the Five-Times Sit-to-Stand test (5-STS, primary), a symmetry index, Timed Up and Go (TUG), walking speed and balance confidence at baseline, week 4 and week 8. Analysis used ANCOVA with multiple imputation.
Results: Sixty-three participants (93%) completed follow-up. At week 4 the mirror group was 1.8 s faster on 5-STS (95% CI 0.6 to 3.0; p = 0.003; d = 0.36), below the 3 s clinical threshold, and the benefit grew with baseline slowness (interaction p = 0.001). Symmetry (2.5 percentage points) and TUG (2.6 s) also favoured mirror feedback. These differences were not clear at week 8, whereas balance confidence differed more (5.6 points; p = 0.026). Adverse events were similar (4 versus 5 participants).
Conclusion: Mirror feedback gave a modest early gain, concentrated in slower participants and mostly gone by week 8. A randomised trial with baseline stratification is needed.