Purpose: To undertake a randomised controlled trial to test the benefits of cognitive training (CT) vs. cognitive stimulation (CS) on general cognitive function and memory in mild Alzheimer's disease (mAD).

Method: A consecutive series of 55 mAD outpatients was randomly allocated to one of the three conditions, namely two experimental interventions (CT and CS; 17 participants each) and a control dementia (no intervention) group (CD; 21 participants). Participants in the experimental conditions received a 4-month individual intervention programme.

Results: The participants in the CT condition demonstrated a large improvement in general cognitive state with moderate gains in areas of memory targeted by the intervention (semantic memory, naming and retrieval ability) 12 months after baseline measurement compared to the other two conditions. In addition, CT seemed to generalise the other cognitive abilities that were not directly targeted with small gains noted in other domains of cognition (prospective memory, face recognition, name a person) compared to CS and CD. Participants in both the CS and CD conditions showed significant general cognitive decline, even though CS was found to ameliorate the trajectory of this decline.

Background: There is inadequate information regarding the neuropsychiatric aspect of Mild Cognitive Impairment (MCI).

Objective: To determine the neuropsychiatric profile of MCI, and compare this with normal controls and patients with mild Alzheimer's Disease (AD).

Design: Cross-sectional assessment of psychiatric symptoms in subjects that are enrolled in Mayo Clinic's longitudinal study of normal aging, MCI and dementia.

Methods and Participants: The Neuropsychiatric Inventory (NPI) was administered to normal control subjects, MCI subjects and patients with early AD. Individual NPI domain scores and total NPI scores were compared among the three groups after controlling for age, educational status, Dementia Rating Scale (DRS) and Mini-Mental State Examination (MMSE) scores. Statistical analysis was performed by utilizing ANOVA, χ2 and Fisher's exact test.

Results: Data were analyzed on 514 normal controls, 54 MCI subjects, and 87 subjects with mild AD (CDR of 0.5 or 1); females consisted of 60.3%, 53.7% and 57.5%; and, the average ages (SD) were 77.8 (1.95), 79 (4.6), 80.5 (14.6) respectively. ANOVA pair-wise comparison revealed that both MMSE and DRS differences among the three groups were significantly different at (p=0.05). The total NPI scores were significantly different (p=0.0001, F=107.93) among the three groups using ANOVA. Pair-wise comparison of individual behavioral domain of NPI showed statistically significant differences between MCI and normals; and MCI and AD (p=0.001). Group differences on NPI remained after controlling for age and education at p=0.0375 and p=0.0050 respectively.

Conclusion: The neuropsychiatric pattern is reminiscent of the clinical, neuroimaging and neuropsychological profile of MCI. It gives further credence to the view that MCI is indeed the gray zone, with overlap on both ends of the pole.