Does an early increased-intensity interdisciplinary upper limb therapy programme following acute stroke improve outcome?

Abstract

Objective: To determine whether an early increased-intensity upper limb therapy programme following acute stroke improves outcome. Design: A randomized controlled trial. Setting: A stroke unit which provides acute care and rehabilitation for all stroke admissions. Subjects: One hundred and twenty-three patients who had had a stroke causing upper limb impairment within the previous 10 days. Intervention: The intervention group received stroke unit care plus enhanced upper limb rehabilitation provided jointly by a physiotherapist and occupational therapist, commencing within 10 days of stroke, and available up to 30 minutes/day, five days/week for six weeks. The control group received stroke unit care. Main outcome measures: The primary outcome measure was the Action Research Arm Test (ARAT) three months after stroke. Secondary outcome measures: Motricity Index; Frenchay Arm Test; upper limb pain; Barthel ADL Index; Nottingham E-ADL Scale; and costs to health and social services at three and six months after stroke. Results: There were no differences in outcomes between the intervention and control groups three and six months after stroke. During the intervention period the intervention group received a median of 29 minutes of enhanced upper limb therapy per working day as inpatients. The total amount of inpatient physiotherapy and occupational therapy received by the intervention group was a median of 52 minutes per working day during the intervention period and 38 minutes per working day for the control group ( p = 0.001). There were no differences in service costs. Conclusions: An early increased-intensity interdisciplinary upper limb therapy programme jointly provided by a physiotherapist and occupational therapist did not improve outcome after stroke. The actual difference in the amount of therapy received by intervention and control groups was less than planned due to a competitive therapy bias.