Researchers analyzed data from 25 randomized controlled trials involving middle-aged and older adults with stroke. The study looked at how the total amount of exercise and the specific type of training influenced lower-limb functional mobility. The primary measure used was the Timed Up and Go test, which assesses how quickly a person can stand, walk, turn, and sit down.
The analysis found a non-linear relationship between exercise dose and improvement. Overall, the greatest benefit occurred at approximately 1400 MET-minutes per week. However, the ideal amount of exercise depended on the specific modality used. For example, resistance and gait training showed strong effects at lower doses around 560 MET-minutes per week.
Other types of training also showed benefits at different levels. Motor control training achieved significant effects at moderate doses near 890 MET-minutes per week. Integrated aerobic and combined cognitive-exercise training peaked at higher doses between 1700 and 1800 MET-minutes per week. In contrast, virtual reality training showed wide credible intervals and non-significant effects, indicating the evidence for this method is currently inconclusive.
No safety concerns or adverse events were reported in the included trials. The study provides quantitative guidance for creating individualized exercise prescriptions in neurorehabilitation. However, readers should be cautious about generalizing these specific dose zones to populations or exercise types not studied. The findings are based on Bayesian modeling of aggregated data rather than direct causation.