Clinical trials have demonstrated the efficacy of cholinesterase inhibitors (ChEI) in improving cognitive status and disability in subjects with mild to moderate Alzheimer's disease (AD). However, little is known about the effectiveness of ChEI in clinical practice, and no large clinical trials comparing different ChEI are available at present. Aim of this study was to evaluate safety and effectiveness of ChEI in a sample of elderly outpatients diagnosed with mild to moderate AD. We selected 407 subjects for ChEI treatment (donepezil, rivastigmine or galantamine). Their cognitive function was evaluated by means of the mini mental state examination (MMSE), and the global functional status was estimated by using the activities of daily living (ADL) and the instrumental activities of daily living (IADL) scales at baseline (T-0), then after 1 (T-1), 3 (T-2) and 9 months (T-3), respectively. T-3 follow-up was completed by 212 subjects. The patients were considered as responders (R), if the MMSE score at T-2 was unchanged or improved, if compared to that of To. In 35 patients (8.6%) treatment was withdrawn because of mostly gastrointestinal adverse events. Compared to the other drugs, donepezil was associated with a lower incidence of withdrawals due to adverse events. Subjects who completed T-3 follow-up (age 78 +/- 6 years, MMSE scores 18.8 +/- 3.9) showed an increase at T-2 of 0.7 +/- 2.7 (p = 0.001) and a decrease at T-3 of -0.6 +/- 3.4 (p = 0.008) in the MMSE scores, as compared to T-0. The ADL and IADL scores did not show significant changes at T-2; however, both decreased significantly at T-3. The patients R-at-T-2 showed a better cognitive and functional outcome at T-3, compared to the non-responders (NR-at-T-2), displaying values of MMSE R-at-T-2 0.4 +/- 3.1 vs. NR-at-T-2 -3.0 +/- 2.5,-p = 0.001, and ADL values of -0.3 +/- 1.2 vs. -0.7 +/- 1.3, p = 0.03, respectively. No significant difference was found in the changes of MMSE scores between donepezil and rivastigmine (galantamine was not included in the comparison due to the small number of treated subjects). In conclusion, in this sample of elderly subjects with mild to moderate AD, treated with ChEl, a small but significant decline in cognitive and functional status was observed after 9 months. Subjects who showed a good response to treatment after 3 months, had a better cognitive and functional outcome at 9 months. No significant difference in cognitive outcome was found between drugs, while donepezil was better tolerated.