Physiotherapy for the arm and hand after stroke

Summary Evaluations of physiotherapy for stroke patients have been criticised for their lack of description of the content of treatments. The purpose of this paper is to describe in detail the physiotherapy approach employed in a recent trial (Lincoln et al, 1999). The trial evaluated effects of early additional physiotherapy for the arm. The main outcome measures were the Rivermead Motor Assessment Arm Scale and the Action Research Arm Test. The trial included a comparison between treatments administered by a qualified physiotherapist and treatments delegated to a trained and closely supervised physiotherapy assistant. A post hoc subgroup analysis of the data from this trial suggested benefits in a small group of less severely impaired patients who had completed additional treatment in the assistant group (Parry et al, 1999). This paper will discuss this finding in the light of differences in the content of treatment which was applied by the qualified and assistant therapists. Design The study was a randomised controlled trial. Outcome was compared between additional and routine amounts of physiotherapy, and between treatment given by a qualified physiotherapist and treatment given by a trained and supervised assistant. During the trial, descriptive data concerning the content of treatments were recorded. Following post hoc analysis which subdivided the patients into those who were more and less impaired, content of the treatments applied by the qualified therapist and by the trained supervised assistant were compared. Finding Treatment content differed between the assistant and qualified groups. Treatment in the less severe assistant subgroup included a greater proportion of repetitive supervised practice of movements and functional tasks. For all but one of the 93 patients randomised to the assistant's group, it was possible to construct a programme of treatment activities suitable for administration by a trained supervised assistant. There was no significant difference between the assistant and qualified groups in the number of patients who completed the treatment. Conclusions The findings from the subgroup analysis were the result of post hoc analysis of small groups and must therefore be viewed as speculative. However, previous research supports an argument that benefits in the milder patients treated by the assistant may have resulted from the emphasis on repetitive supervised practice in their therapy. We argue that it is appropriate to delegate this sort of treatment to trained and supervised physiotherapy assistants