Sixteen studies (1,429 participants, range 21 to 242) were included in the review. These were all controlled studies although study design was not stated explicitly. Two studies reported using two experimental groups with one control group so there were 18 comparisons.
There was a small but significant reduction in cancer-related fatigue in favour of exercise compared to control (SMD 0.28, 95% CI 0.17 to 0.38; 18 comparisons). There was no significant heterogeneity.
There were significant reductions in cancer-related fatigue for aerobic exercise programmes (SMD 0.25, 95% CI 0.12 to 0.38; 13 comparisons). There were no significant differences between groups for resistance training programmes (three comparisons) or mixed training programmes (two comparisons).
Studies conducted in supervised settings reported significant reductions in cancer-related fatigue for exercise groups compared to control groups (15 comparisons). Studies conducted in unsupervised settings showed no significant differences between groups (three comparisons).
Significant improvements in cancer-related fatigue were found for exercise programmes conducted during or following treatment in patients with breast or other types of cancer and on aerobic and musculoskeletal physical fitness.
There was evidence of statistical heterogeneity for some subgroup analyses (aerobic-resistance programmes, exercise post-treatment, patients with breast cancer and aerobic physical fitness).