Physiotherapy intervention late after stroke and mobility.BMJ 1992; 304 doi: http://dx.doi.org/10.1136/bmj.304.6827.609 (Published 07 March 1992) Cite this as: BMJ 1992;304:609
OBJECTIVE--To determine whether the intervention of a physiotherapist improved mobility in patients seen more than one year after stroke. DESIGN--Randomised crossover trial comparing two groups offered intervention by a physiotherapist, one immediately after entry into the trial and the other after a delay of three months. The intervention consisted of identifying problems and offering advice and help to solve the problems. SETTING--Patients' homes in Oxfordshire. SUBJECTS--Patients who had reduced mobility due to a stroke more than one year before entry; 60 were recruited from a community stroke register and 34 in other ways. MAIN OUTCOME MEASURES--Standard measures of mobility including gait speed, functional ambulation categories, the Nottingham extended activities of daily living index, and individual items from the Barthel activities of daily living index and the Frenchay activities index. Measures of manual dexterity, depression, and anxiety were used as controls. RESULTS--94 patients entered the trial and 49 were randomised to immediate and 45 to delayed physiotherapy; 89 were compared at the crossover point. At randomisation the two groups were comparable. At three months the group given early therapy showed an improvement in gait speed whereas the untreated group had declined (differences of -3.9 v 6.4 s to walk 10 m; p less than 0.01); between three and six months the group given delayed therapy showed improvement and the previously treated group declined (differences of 6.5 v -3.9 s to walk 10 m; p less than 0.01). A 9% (95% confidence interval 0% to 18%) decrease in time taken to walk 10 m was associated with treatment and a 12% (2% to 19%) increase when patients were untreated. Other measures did not change significantly. CONCLUSION--Intervention of an experienced physiotherapist late after stroke specifically improves mobility, albeit by a small amount, but the effects do not seem to be maintained, perhaps because there is an underlying decline in mobility in these patients. Gait speed offers a simple and sensitive measure of outcome.