Current issues of ACP Journal Club are published in Annals of Internal Medicine


Physiotherapy offered late after stroke increased mobility but improvements were not maintained

ACP J Club. 1992 Sept-Oct;117:35. doi:10.7326/ACPJC-1992-117-2-035

Source Citation

Wade DT, Collen FM, Robb GF, Warlow CP. Physiotherapy intervention late after stroke and mobility. BMJ. 1992 Mar 7;304:609-13.



To determine the effectiveness of physiotherapy in improving mobility in patients seen > 1 year after stroke and to determine if any observed improvement is maintained.


Randomized, single-blind, crossover trial.


Oxfordshire, United Kingdom.


Patients were included if they had mobility problems > 1 year after stroke and used a walking or mobility aid; had had a fall within 3 months; were unable to manage stairs, slopes, or uneven surfaces independently; or had a slow gait speed. Patients were excluded if there was no impairment related to stroke that was likely to reduce mobility and if another cause for impaired mobility was identified. 94 patients (mean age, 72 y; 47 men) were randomized, and 75 patients (80%) completed the trial.


Patients received 3 months of physiotherapy at home either immediately after randomization (n = 49) or after a 3-month delay (n = 45). The intervention consisted of identifying mobility problems and offering advice and help to solve these problems.

Main outcome measures

The primary outcome measure was gait speed measured over 10 meters indoors. Other outcomes were measures of motor function including walking independently (from the Barthel index) and walking outdoors (Frenchay activities index). Tests for depression anxiety and manual dexterity were also done. Patients were assessed at 3-, 6-, and 9-month follow-ups.

Main results

The mean time taken for all randomized patients to walk 10 meters was 44.5 seconds at baseline. At 3 months the patients given early treatment showed an improvement in gait speed, reducing the mean time to walk 10 meters by 3.9 seconds, whereas the untreated group (patients randomized to delayed treatment) took 6.4 seconds longer (P < 0.01). Between 3 and 6 months the patients receiving delayed treatment (physiotherapy at 3 months) showed an improvement in gait speed, reducing the mean time to walk 10 meters by 3.9 seconds, whereas the group receiving early treatment declined by requiring 6.5 seconds longer (P < 0.01). For all patients during treatment the time taken to walk 10 meters was reduced by 4.2 seconds (95% CI 0 to 8.4 s) and in the no-treatment period the time taken to walk 10 meters increased by 4.9 seconds (CI 1.1 to 8.7 s, P < 0.05).


Physiotherapy offered > 1 year after stroke improved gait speed slightly, but the improvement was not maintained 3 months after physiotherapy stopped.

Source of funding: Medical Research Council, United Kingdom.

Address for article reprint: Dr. D.T. Wade, Rivermead Rehabilitation Centre, Oxford OX1 4XD, United Kingdom.


Stroke is a leading cause of long-term disability in developed countries, largely because of its effect on patient mobility. Any intervention that improves mobility after stroke is likely to reduce the burden of stroke on patients, caregivers, and the community. Physiotherapy has long been advocated for the management of patients with impaired mobility after stroke but its efficacy has not been established.

The study by Wade and colleagues is a landmark in rehabilitation research for its methodologic rigor and sets a standard for future trials of rehabilitation programs in the management of patients after stroke. The study provides unbiased evidence of a favorable, but nonsustained, effect of physiotherapy on patient mobility (time to walk 10 m) among 1-year survivors of stroke with residual mobility problems. It fails to identify an effect on other measures of mobility, activities of daily living, anxiety and depression, or manual dexterity. This may be in part because of the small number of patients studied (type II error).

The immediate question that this study raises is whether improvement in gait speed is clinically important. Gait speed is a useful surrogate for other measures of mobility because it is valid, reliable, easy to measure and communicate to others, and sensitive to change. It is likely that quicker walking means fewer falls, fewer aids needed, and hence a better quality of life.

The implication of the finding that the effects of physiotherapy are not sustained is that ongoing, regular physiotherapy is required by patients with long-term mobility problems after stroke. Health care professionals must address whether continuing physiotherapy is cost effective.

Graeme J. Hankey, MBBS
Royal Perth HospitalPerth, Western Australia, Australia