Effectiveness of interventions in increasing physical activity of inpatients after stroke: A systematic review and meta-analysis.

Most pooled findings were graded very low certainty because the included trials had missing outcome data, measurement problems and large differences between studies. The confidence intervals for the main physical activity results included the possibility of no effect, so the observed increases are not certain. Much of the activity feedback evidence came from people who could walk without assistance, so it may not apply to stroke inpatients who cannot walk independently. Secondary outcomes were only available from studies that also measured physical activity, so they do not represent the wider evidence on therapy dose or function. Interventions and activity measures varied widely, and the review selected only one activity measure per study, which may have changed the pooled estimates. Adverse event definitions and reporting varied between studies, so confidence in the safety conclusions is limited. The patient-directed programme finding came from one study and had low certainty.

Funding was disclosed for Peter Hartley from Homerton College and the Health Foundation grant to the University of Cambridge for THIS Institute, and from the NIHR Cambridge Biomedical Research Centre. The authors declared no potential conflicts of interest.

Do not conclude that extra physiotherapy should be added to stroke inpatient care to improve function. The pooled physical activity increases had confidence intervals including no effect, function did not improve, and non-serious adverse events including falls may increase.