Twelve studies were included in the review (408 participants in the study tables). Three studies were described as Level I, five as Level II and as four as non-randomised studies. All studies had clear inclusion criteria and used reliable and valid outcomes measures. Ten studies described the intervention clearly enough that it could be replicated. Two studies used blinded outcomes assessment. Five studies accounted adequately for attrition.
Body composition (nine studies): All studies used group-based interventions. Three studies reported that the intervention significantly improved muscle thickness and one reported a significant decrease in body fat.
Physical fitness (11 studies): Exercise significantly improved at least one measure of physical fitness in all seven studies of group-based and two out of four home-based interventions.
Functional outcomes (four studies): All studies were group-based. All studies found that exercise improved at least one measure of functional performance.
Quality of life or fatigue (11 studies): Exercise significantly improved at least one measure of quality of life or fatigue in five out of six studies of group-based and two out of five studies of home-based interventions.
All types of exercise were associated with statistically significant benefits for one or more measures of each outcome. Other findings were non-significant (where statistical significance was reported in the review).