This study aimed to investigate the treatment effects of a hybrid home-based rehabilitation program combining mirror therapy and task-specific training compared with clinic-based rehabilitation in stroke patients.

This was a single-blinded, 2-sequence, 2-period, crossover-designed study. The participants were randomly allocated to 1 of 2 order conditions: home-based rehabilitation first or clinic-based rehabilitation first. During each period, each participant received 12 treatment sessions, with a 4-week washout phase between the 2 periods. Both the home-based and clinic-based rehabilitation groups received 30 to 45minutes of mirror therapy, followed by 45 to 60minutes of functional task training, at each 90-min session. Before the interventions began, the client-centered Canadian Occupational Performance Measure was administered to help identify individual tasks and goals of patients. Outcome measures were selected based on the ICF. Outcomes of impairment level were the Fugl–Meyer Assessment, Box and Block Test, and Revised Nottingham Sensory Assessment. Outcomes of activity and participation levels included the Motor Activity Log, Canadian Occupational Performance Measure, and EuroQoL-5D Questionnaire.

A total of 24 and 18 patients with stroke completed the study during the intervention period 1 and 2, respectively. Pretest analyses showed that there was no significant evidence of carryover effect. Home-based rehabilitation resulted in significantly greater improvements on the amount of use subscale of Motor Activity Log (P=0.01) than clinic-based rehabilitation. Clinic-based rehabilitation group had better benefits on the health index measured by the EuroQoL-5D Questionnaire (P=0.02) than home-based rehabilitation. There were no statistically significant differences between the 2 groups on the other outcomes.

This study is unique in developing a hybrid home-based stroke rehabilitation intervention with patient-oriented goals. The home-based and clinic-based rehabilitation groups had comparable benefits in the outcomes of impairment level but showed differential effects in the outcomes of activity and participation levels.