Feedback in Rehabilitation Following Stroke

Abstract

Stroke is a major cause of disability worldwide. Following stroke, patients need to re-learn the ability to perform activities such as sitting, standing up, standing, walking, and reaching and grasping objects, since independence in these activities has significant influence on quality of life and the ability to participate in activities of daily living. Information feedback, that is, specific content about successful or unsuccessful components of the attempt in order to enhance the next attempt, received during or after practice, is important for effective re-learning of activities following stroke. The studies included in this thesis aimed to investigate the nature of feedback provided to patients in stroke rehabilitation; and increase evidence of the effect of information feedback from equipment (biofeedback) to improve performance and learning of activities in people following stroke. Study 1 was an observational study of people in stroke rehabilitation that investigated usual therapist feedback during practice of activities. This study found that therapists provided motivational statements more often than information feedback. Presence of the therapist during practice of activities positively influenced the feedback received by patients. However, when the therapist left the patients to practice alone, the therapist’s verbal feedback was not replaced by another source of information feedback. Equipment as a source of information feedback, whether the therapist was present or the patient practicing alone, was used rarely. Study 2 was a systematic review and meta-analysis that investigated the effectiveness of biofeedback in improving activities of the lower limb in people following stroke. Biofeedback had a moderate effect (10 high quality trials, 241 participants; SMD = 0.49, 95% CI 0.22 to 0.75) in the short term compared with usual therapy/placebo. Furthermore, the benefits were still present beyond intervention, although slightly diminished (5 high quality trials, 138 participants; SMD = 0.41, 95% CI 0.06 to 0.75), suggesting learning had occurred. Importantly, this beneficial effect of biofeedback was present even when the amount of practice of the activity was equivalent between the two groups, with the only difference being the substitution of biofeedback for therapist feedback in the experimental group, with an analysis of those trials showing a similar effect (8 trials, 170 participants; SMD 0.51, 95% CI 0.20 to 0.83). This suggested that the mode of feedback is important, and that biofeedback is superior to therapist feedback. Study 3 investigated the feasibility and potential effectiveness of biofeedback from customised equipment during training of standing up in stroke rehabilitation. The intervention was feasible and safe, progressions implemented effectively, compliance excellent, and the participants’ rating of satisfaction was high. The results suggested that the intervention had the potential to improve the ability to stand up in people following stroke, as participants were able to stand up more effectively, i.e. with increased quality and speed. Importantly, these improvements carried over into their activities of daily living, suggesting learning had occurred. Biofeedback appears to be a good source of information feedback, and is more effective in improving activities than usual therapist feedback. This is likely explained by the observation that therapists are more motivational than informative. Biofeedback should be used more often in clinical practice as it is currently used rarely.