BACKGROUND AD, the most prevalent cause of dementia, affects twice as many women as men. Therapeutic options are limited, but results of prior studies support the hypothesis that estrogen treatment may improve symptoms of women with this disorder. METHODS Forty-two women with mild-to-moderate dementia due to AD were enrolled into a randomized, double-blind, placebo-controlled, parallel-group trial of unopposed conjugated equine estrogens (1.25 mg/day) for 16 weeks. RESULTS Outcome data were available for 40 women at 4 weeks and 36 women at 16 weeks. At both 4 and 16 weeks, there were no significant differences or statistical trends between treatment groups on the primary outcome measure (the cognitive subscale of the Alzheimer's Disease Assessment Scale), clinician-rated global impression of change, or caregiver-rated functional status. Exploratory analyses of mood and specific aspects of cognitive performance also failed to demonstrate substantial group differences. CONCLUSIONS Although conclusions are limited by small sample size and the possibility of a type II error, results suggest that short-term estrogen therapy does not improve symptoms of most women with AD. These findings do not address possible long-term effects of estrogen in AD, possible interactions between estrogen and other treatment modalities, or putative effects of estrogen in preventing or delaying onset of this disorder.