OBJECTIVE: To evaluate if home-based rehabilitation of inpatients improved outcome compared to standard care.

DESIGN: Interventional, randomised, safety/efficacy open-label trial.

SETTING: University hospital stroke unit in collaboration with three municipalities.

SUBJECTS: Seventy-one eligible stroke patients (41 women) with focal neurological deficits hospitalised in a stroke unit for more than three days and in need of rehabilitation.

INTERVENTIONS: Thirty-eight patients were randomised to home-based rehabilitation during hospitalization and for up to four weeks after discharge to replace part of usual treatment and rehabilitation services. Thirty-three control patients received treatment and rehabilitation following usual guidelines for the treatment of stroke patients.

MAIN MEASURES: Ninety days post-stroke the modified Rankin Scale score was the primary endpoint. Other outcome measures were the modified Barthel-100 Index, Motor Assessment Scale, CT-50 Cognitive Test, EuroQol-5D, Body Mass Index and treatment-associated economy.

RESULTS: Thirty-one intervention and 30 control patients completed the study. Patients in the intervention group achieved better modified Rankin Scale score (Intervention median = 2, IQR = 2-3; Control median = 3, IQR = 2-4; P=0.04). EuroQol-5D quality of life median scores were improved in intervention patients (Intervention median = 0.77, IQR = 0.66-0.79; Control median = 0.66, IQR = 0.56 - 0.72; P=0.03). The total amount of home-based training in minutes highly correlated with mRS, Barthel, Motor Assessment Scale and EuroQol-5D™ scores (P-values ranging from P<0.00001 to P=0.01). Economical estimations of intervention costs were lower than total costs of standard treatment.

CONCLUSION: Early home-based rehabilitation reduced disability and increased quality of life. Compared to standard care, home-based stroke rehabilitation was more cost-effective.