Three studies were included in the review: one randomised controlled trial (n=6), and two pre-test post-test clinical trials (n=40).
The median PEDro score for quality was 4 out of 10 (range: 3 to 5).
Three studies assessed impairment. Significant increases in strength were noted in studies focusing on progressive resistance programmes, in a circuit strength training programme without the use of weights, and in a 25-week treadmill walking programme. Effect sizes ranged from 3.08 (95% CI: 2.16, 4.00) to 7.52 (95% CI: 6.46, 8.58) for the progressive resistance programmes, from 1.94 (95% CI: 1.02, 2.86) to 5.06 (95% CI: 4.14, 5.98) for the circuit strength training programme, and from 2.68 (95% CI: 1.62, 3.73) to 11.20 (95% CI: 10.48, 11.92) for the treadmill walking programme.
Comparisons of strength changes between different programmes favoured progressive resistance programmes over circuit training. There were no reported adverse effects.
One study assessed activity. Large effect sizes were reported for the 25-week treadmill walking programme in terms of improved walking speed (2.95, 95% CI: 2.42, 3.48), distance (4.67, 95% CI: 4.14, 5.20) and duration (3.98, 95% CI: 3.45, 4.51) from baseline. Compared with a non-intervention control group, the timed sit up-and-go test produced a favourable performance in the treatment group (effect size 1.03, 95% CI: 0.19, 1.87).
There was no measurement of participation restriction in any of the included studies.