In clinical trials, short-term galantamine treatment produces consistent positive effects on global ratings, cognitive tests, and assessments of activities of daily living and behavior in patients with mild-to-moderate Alzheimer's disease (AD), providing the rationale for longer-term, open-label treatment. In this continuation trial following enrollment in previous 12-month trials, patients received galantamine 24 mg/day for a total of 24 months (total exposure up to 36 months). Primary efficacy measures were the ADAS-cog/11 and DAD. Adverse events (AEs) were coded to WHO preferred terms, including AEs begun in previous trials. Initial improvement in cognitive function was followed by a gradual decline, as measured by increased ADAS-cog/11 scores. At 36 months, ADAS-cog/11 scores increased by a mean (SEM) of 12.4 (0.80) points (P < 0.001) versus a projected 22-point increase for untreated patients. Functional abilities, as measured by the DAD, had decreased significantly at each time point versus baseline (P < 0.001). The most common treatment-emergent AEs were agitation (16.1%), insomnia (12.4%), fall (11.2%), and urinary tract infection (10.2%). AEs were mainly mild to moderate, appropriate for an elderly population, with few judged treatment related. Galantamine 24 mg/day is safe and effective for long-term treatment of mild-to-moderate AD. Potential exists for prolonged benefit with galantamine therapy versus lack of treatment for the long-term.