Oestrogen use by postmenopausal women has many health benefits, but findings on the effect of oestrogen in Alzheimer's disease are conflicting. Oestrogen promotes the growth and survival of cholinergic neurons and could decrease cerebral amyloid deposition, both of which may delay the onset or prevent Alzheimer's disease. To investigate whether use of oestrogen during the postmenopausal period affects the risk of Alzheimer's disease, we studied 1124 elderly women who were initially free of Alzheimer's disease, Parkinson's disease, and stroke, and who were taking part in a longitudinal study of ageing and health in a New York City community. Relative risks and age-at-onset distributions were calculated from simple and adjusted Cox proportional hazards models. Standard annual clinical assessments and criterion-based diagnoses were used in follow-up (range 1-5 years). Overall, 156 (12.5%) women reported taking oestrogen after onset of menopause. The age at onset of Alzheimer's disease was significantly later in women who had taken oestrogen than in those who did not and the relative risk of the disease was significantly reduced (9/156 [5.8%] oestrogen users vs 158/968 [16.3%] nonusers; 0.40 [95% Cl 0.22-0.85], p < 0.01), even after adjustment for differences in education, ethnic origin, and apolipoprotein-E genotype. Women who had used oestrogen for longer than 1 year had a greater reduction in risk; none of 23 women who were taking oestrogen at study enrolment has developed Alzheimer's disease. Oestrogen use in postmenopausal women may delay the onset and decrease the risk of Alzheimer's disease. Prospective studies are needed to establish the dose and duration of oestrogen required to provide this benefit and to assess its safety in elderly postmenopausal women.