OBJECTIVES: Little progress has been made in creating new treatments and donepezil is still one of the most common treatments for Alzheimer’s Disease (AD). In the light of recommendations of the Second Panel and newly available data, we analyzed the cost-effectiveness of donepezil in the treatment of patients with mild, moderate and severe AD from a societal perspective. METHODS: A Markov model with 10-year time horizon compared the cost-effectiveness of treating AD patients (mild, moderate, severe) with donepezil versus no treatment from a U.S. societal perspective. We used the transition probabilities reported by Spackman et al (2012) and the proxy-reported quality of life years (QALYs) from Linus et al (2006). Relative risk values were noted by a 24-week randomized clinical trial by Rogers et al (1998). Cost data were extracted from Neumann et al (1999), Knopman et al (2012), Small et al (2002), and the American Medical Association database. Costs (2017 USD) and QALYs were discounted at 3% annually. We measured results by calculating the incremental cost-effectiveness ratio (ICER) and comparing it to the willingness-to-pay (WTP) threshold of $100,000/QALY. Multivariate probabilistic sensitivity analysis tested model uncertainty. RESULTS: Compared to no treatment for Alzheimer’s patients ($266,432; 2.81 QALYs), treatment with donepezil is costlier, but also more effective ($302,139; 3.42 QALYs). Donepezil is cost-effective with an ICER of $58,194/QALY which is below our WTP. Our estimate is robust to sensitivity analysis. Out of 10,000 Monte Carlo simulations, donepezil was cost-effective in 8937 trials. CONCLUSIONS: From a societal perspective, donepezil treatment for AD patients is cost-effective compared to no treatment. Our analysis provides an important update to the current literature by considering technological advances, lifestyle changes, and a growing aging population.