Aim We investigated the impact of ambulatory physiotherapy (AP) on motor abilities in elderly subjects with Alzheimer's disease (AD). Methods Subjects with mild to moderate AD were included and divided into “physiotherapy group” (PG) and “no physiotherapy group” (NPG) according to whether or not they received AP between inclusion (T0) and the second time of assessment, between 15 and 36 months after inclusion (T1). The follow‐up duration, Mini‐Mental State Examination, Tinetti and mini motor test (MMT) scores, Timed Up & Go test (TUG), gait speed (GS), one‐leg balance (OLB), history of falls within the last 6 months (HF), ability to rise from the floor (RFF) and the use of a walking aid (UWA) were recorded at T0, and after at least 15 months of follow up (T1). Results A total of 50 subjects were included in the NPG and 20 in the PG. At baseline, these groups were not significantly different for all the parameters recorded. The anova showed a progression of cognitive disorders in the two groups between T0 and T1 (P < 0.001), which was similar in the two groups (P = 0.83). For each postural and motor quantitative test (Tinetti, MMT, TUG, GS) the anova showed a main effect of time of assessment (All P < 0.05) associated with a group × time of assessment interaction (All P < 0.05). The comparison between the two groups with regard to the evolution of qualitative parameters showed a significant difference for the OLB test only. No significant difference was found for RFF, HF and UWA. Conclusions There was a significant improvement or stability of motor abilities in the PG; while these abilities decreased in the NPG. Geriatr Gerontol Int 2013; ●●: ●●–●●.