A total of seven RCTs were included in the review (n=3,194, range 30 to 2,832); three studies met more than 12 of the quality criteria. Not all included studies represented individual patient cohorts. Duration of follow-up ranged from three to 72 months.
Four of the seven studies showed statistically significant effects. Compared with wait-and-see control groups, cognitive exercise interventions yielded significantly improved neuropsychological performance (WMD 1.07, 95% CI 0.32 to 1.83; seven studies). Studies with long-term follow-up (WMD 1.02, 95% CI 0.14 to 1.89; five studies) did not differ from those with medium-term follow-up (WMD 1.16, 95% CI 0.37 to 1.96; two studies).