To determine if there are differential treatment effects of second-generation cholinesterase inhibitors over one year, 130 patients (untreated=65, treated=65) meeting NINCDS-ADRDA criteria for mild or moderate probable AD underwent standardized cognitive testing at baseline and 12 months later at a university memory clinic. Patients were followed either prior to or after the availability of treatment and were matched on education and baseline Mini Mental State Examination (MMSE). A detailed medical history evaluation was conducted. In this well matched longitudinal observational cohort study, there were no differences in the prevalence of comorbid illnesses, concomitant medication use or vascular risk factors except for a greater number of treated patients with a previous history of smoking. Separate repeated measures MANCOVAs on the MMSE, Mattis Dementia Rating Scale (DRS), and its 5 subscores (attention, initiation/perseveration, conceptualization, construction and memory) (Bonferroni corrected), after covarying for the effects of smoking, and SSRI use, showed less decline over one year in the treated group in overall cognition and in all subscores of the DRS except for memory (effect sizes 0.5-0.7). Less decline was also seen in the treated group in function and in instrumental and basic activities of daily living as measured with the Disability Assessment for Dementia Scale (DAD) (effect sizes 0.4-0.8). Executive, language and visuospatial functions, rather than memory, appeared to be more amenable to stabilization over one year by cholinesterase inhibitors in AD.