Combined analysis of two randomized trials of community physiotherapy for patients more than one year post stroke

Background: Some patients continue to have mobility problems as a long-term consequence of stroke and it is unclear whether routine, further contact with a physiotherapy service is beneficial. Two single-centre, randomized controlled trials of physiotherapy for patients more than one year post stroke have been undertaken in Oxford and Bradford in the UK and the results from these two trials have been combined to give a more precise estimate of effectiveness.

Method: The computerized databases from both trials were combined for a joint analysis. Outcome measures common to both trials were: Rivermead Mobility Index; gait speed measured over 10 metres; Barthel Index; Frenchay Activities Index; Hospital Anxiety and Depression Scale.

Results: There were 264 patients available for the combined analysis (Oxford= 94; Bradford= 170). There was a significant but clinically small improvement in mobility at three months in the combined treatment group measured by the Rivermead Mobility Index (median of the differences= 0 (95% confidence interval (CI) 0, 1); interpolated values= 0.43 (95% CI 0.08, 0.80)) and gait speed (treatment effect 2.7 m/min (95% CI 0.94, 4.46)). There were no other significant differences. Intervention given in both studies was at the discretion of the physiotherapists and was of similar and low intensity (mean visits Oxford= 4 (SD 2.5); Bradford= 5 (SD 4.5)).

Conclusion: A more effective physiotherapy intervention is required for stroke patients with persisting mobility problems after stroke.