Alzheimer's disease (AD) shortens life-expectancy, but the effects of pharmacological treatments for this disorder on mortality have not been studied. We compared two commonly prescribed medications, donepezil and memantine, with respect to the length of survival of veterans presumed to have AD. The Computerized Medical Records System at the Veterans Affairs Palo Alto Health Care System (VAPAHCS) was used to identify all patients prescribed these medications between 1997 and 2008. The VAPAHCS approved donepezil in 1997 and memantine in 2004. Kaplan-Meier and Cox regression analyses were used to test for chronological and drug-related associations with survival in 2,083 male veterans aged 55 years and older receiving prescriptions for donepezil, memantine, or both. Overall patient mortality decreased in the 2004 to 2008 era, compared with the 1997 to 2003 era, pre-memantine (HR: 0.75; 95% CI: 0.63, 0.89; p = 0.001). In analyses confined to the 2004 to 2008 era, patients prescribed memantine alone survived significantly longer than those prescribed donepezil alone (HR: 2.24; 95% CI: 1.53, 3.28; p < 0.001) or both donepezil and memantine (HR: 1.83; 95% CI: 1.14, 2.94; p = 0.012). While this study has several limitations, these findings suggest that memantine treatment is associated with an increased life-expectancy relative to donepezil treatment. Additional research is needed to replicate these unexpected findings and identify potential mechanisms to explain this apparent association, to establish if the relationship applies to other cholinesterase inhibitors, and to discover whether the findings generalize to women and patient populations with characteristics different from those of the veterans in this study.