The feasibility of self-measurement (SM) of the blood pressure (BP) is well demonstrated in hypertensive patients (pts) but, in elderly pts physical and intellectual limitations can restrict the use of SM. The aim was to evaluate the feasibility of SM as a function of autonomic psychomotricity in pts aged more than 75 years and to estimate the reproducibility of SM in comparison with office's measurements (OM). We initially proposed the use of SM devices of BP to 53 pts aged > 75 years. One-third of pts refused to participate in the study. In 32 pts we did a mini-mental-state score (MMS) and an evaluation of autonomic functions (IADL). Two patients with a MMS < 20 were excluded. The OM of BP was done in sitting position with a mercury sphygmomanometer (4 measures) and the SM by a Omron device during 5 consecutive days (3 measures morning and night). Eighteen (60%) pts did more or equal to 15 measures (good measurer) and 12 less than 15 (bad measurer). The pts of the group "good measurer" were significantly younger, were all hypertensive treated patients and had a higher MMS and a lower IADL than those of the group "bad measurer". Considering the differences (OM-SM), 55% of patients showed values superior to 10 mmHg for systolic blood pressure, and 64% of subjects had values superior to 5 mmHg for diastolic blood pressure. We conclude that the SM is acceptable by patients older than 75 years and that feasibility is optimal in those patients where the autonomic and cognitive functions are preserved and when the patient is hypertensive. Also, as the concordance between OM and SM of patients older than 75 years is below 50%, diagnostic and therapeutic decisions based on OM is hazardous in these patients.