In moderate-to-severe Alzheimer disease (AD), there are significant losses of activities of daily living (ADL). In a recent prospective, randomized, placebo-controlled trial, memantine treatment lessened the overall functional decline in AD patients already on stable donepezil therapy. In this trial, patients (n=404) with Mini-Mental State Examination scores of 5 to 14 receiving stable donepezil treatment were randomized to double-blind treatment with memantine (10 mg b.i.d.; n=203) or placebo (n=201). A primary outcome measure was the 19-item Alzheimer's Disease Cooperative Study--Activities of Daily Living Inventory (ADCS-ADL(19)). To further evaluate the treatment effects of memantine on function, we performed post hoc analyses of ADCS-ADL(19) data from this trial, including ADL items and new subscales derived from factor analysis. Using mixed model analyses, patients receiving memantine had statistically significant less decline in total ADCS-ADL(19) scores compared with placebo. An item analysis revealed statistically significant benefits of memantine on grooming, toileting, conversing, watching television, and being left alone. Statistically significant improvements were noted in subscales evaluating higher-level functions and connectedness/autonomy with memantine compared with placebo. These post hoc analyses in moderate-to-severe AD patients receiving stable donepezil treatment suggest that memantine may impact overall functional levels, and some of the cognitive processing underlying ADL performance.