ABSTRACT
OBJECTIVE: This study evaluates the efficacy at 12 months of a computer-based cognitive training (CBCT) program, adjunctive to traditional cognitive training (TCT), on the basis of pen-and-paper exercises. METHODS: Sixty patients with multi-domain mild cognitive impairment and mild Alzheimer's disease who were already receiving cognitive training, recruited from a day hospital, were assigned into two groups following a simple randomization procedure (computerized random numbers): (i) a group that received CBCT during 3 months and TCT (CBCT + TCT), n = 37, and (ii) a group that received only TCT, n = 23. Patients were assessed at baseline and after 3 and 12 months of treatment by a neuropsychologist blinded to group assignment, with a neuropsychological battery (primary outcomes) and measures of decision making, memory complaints, and emotional disturbances. RESULTS: With the use of repeated-measures analyses of covariance, the CBCT + TCT group showed less anxiety symptoms (F = 5.13, p = 0.03, d = 1.12) and less disadvantageous choices (F = 4.70, p = 0.04, d = 0.89) in decision making than the TCT group at 12 months. No significant improvement or worsening was observed in the other measures examined. However, positive effect sizes favoring the CBCT + TCT group were observed in all variables. CONCLUSIONS: The addition of a CBCT program was effective in anxiety and decision making but had no significant effects on outcomes in basic cognitive functions in patients who were already receiving cognitive training, possibly due to a ceiling effect. Future studies should compare the efficacy of CBCT with TCT in naïve patients.