Studies of stroke rehabilitation therapies should report blinding and rationalize use of outcome measurement instruments.