BACKGROUND: Alzheimer’s disease (AD) is the leading cause of dementia and its course renders patients functionally disabled. Memantine is the first drug to demonstrate a clinical benefit in the treatment of patients with moderately-severe to severe AD. OBJECTIVES: Our objective was to illustrate the benefits of memantine on functional disability. METHODS: We classified 252 patients from a randomised 28-week clinical trial of memantine vs placebo according to their Activities of Daily Living capabilities measured by the ADCS-ADLsev scale. The scale was divided into two sub-scores: basic and instrumental. The relevance of this classification was validated by comparing clinical and socio-demographic parameters between the different autonomy classes (autonomous and dependent). The effect of memantine was estimated by using a logistic regression model on the autonomy status of patients at week 28, controlling for confounding factors (Observed Cases analysis). RESULTS: Our results showed that dependent patients (n = 106) had significantly longer disease duration, poorer cognition, greater severity, more behavioural alterations and higher total societal costs compared with autonomous patients (n = 146). When controlling for autonomy and severity at baseline, memantine-treated patients were three times more likely [Odds Ratio (OR) = 3.03; 95% Confidence Intervals (CI) = (1.38, 6.66)] to remain autonomous after 28 weeks. Analysis of the Treated Per Protocol set and the use of Last Observation Carried Forward analyses confirmed this finding. CONCLUSIONS: Memantine enhances autonomy in patients with moderately-severe to severe AD by increasing the probability of their remaining autonomous, therefore delaying transition to the dependent stage.