Management of acute stroke

SIR,-The limitations inherent in conducting controlled trials of methods of providing health services, which we refer to in our paper, mean that we have no information on the criteria which were used to transfer a selected group of patients from medical units during the acute phase of rehabilitation. But we have been able to examine the results of standard clinical tests which were designed to predict outcome following acute stroke, and which will form the subject of a future publication. No difference was found in the pattern of neurological impairment present in the transferred group compared with those patients remaining in medical units throughout. Therefore it may not be surprising that there was no difference in outcome between the two groups. Transferring stroke patients from medical units for further rehabilitation appears to be a common practice in Britain, and on the basis of our results these transfers should be arranged earlier rather than later if the full benefits of intensive therapy are to be realised. But, as Dr Powell correctly infers, conclusive evidence to support this statement can come only from further randomised controlled trials. W M GARRAWAY A J AKHTAR R J PRESCOTT L HOCKEY