Women who used estrogen-only menopausal hormone therapy had substantially lower odds of showing Alzheimer's disease pathology in autopsied brains, according to a study published this week, but the researchers who conducted it are clear that the finding cannot establish that the therapy caused the difference.
The study, published on August 12, 2026, in Neurology, the medical journal of the American Academy of Neurology, analyzed data from 21,462 women. Estrogen-only hormone therapy use was associated with 39 percent lower odds of a clinical dementia diagnosis and 35 percent lower odds of Alzheimer's pathology at autopsy. Users also showed biomarker profiles consistent with less amyloid accumulation in the brain.
The reason to lead with the limitation rather than the number is that this specific question has a history of whiplash. Earlier studies indicated that menopausal hormone therapy raises dementia risk, and women have made decisions based on those findings. A retrospective observational study pointing the other way is worth reporting carefully, not celebrating.
"More research needs to be done before we can make recommendations to women," said study author Jennifer Bruno, an instructor in psychiatry and behavioral sciences at Stanford Medicine, in a statement released through the American Academy of Neurology. She added that the women studied were using hormone therapy decades ago, with timing and type of use differing from current practice, so the results may not apply to today's standards.
The Design That Makes This Study Unusual
What sets this analysis apart is that it did not rely solely on diagnosis codes. It looked at brain tissue.
The researchers drew on two large datasets, the National Alzheimer's Coordinating Center and the Alzheimer's Disease Neuroimaging Initiative. In one, 728 participants completed brain scans or biomarker tests while living. In the other, 2,959 participants underwent autopsy after death, at an average age of 82. Participants were followed for roughly 3 to 5 years, starting at an average age of 71. Of the total, 1,953 took hormone therapy, and 19,509 did not.
The autopsy component is the reason the finding carries weight. Investigators compared 258 brains of women who reported estrogen-only therapy against roughly 2,701 brains of women who reported no hormone therapy of any kind, scoring the levels and locations of amyloid plaques and tau tangles along with plaque density. According to Stanford Medicine, this is the first study to measure the connection between menopausal hormone therapy and reductions in Alzheimer's-related pathological deposits in autopsied brains.
The raw proportions are easier to hold onto than the odds ratios. Among hormone therapy users, 18 percent had no signs of Alzheimer's disease at autopsy, compared with 10 percent of non-users. Forty percent of users had all three hallmarks, compared with 51 percent of non-users.
The researchers examined estrogen-only therapy specifically because earlier work suggested estrogen plus progestin combination therapy may increase dementia risk. Those are different treatments and should not be treated as interchangeable when reading this result.
The Selection Problem Sitting Underneath the Numbers
Observational studies of hormone therapy have a well-known vulnerability, and it applies here in an unusually concrete form.
In current practice, estrogen-only therapy is prescribed only for women who have had a hysterectomy, because of the risk of endometrial cancer in women with a uterus. That means most users in this study presumably had hysterectomies, which is not a random subset of women. It reflects a distinct surgical and medical history, a distinct pattern of engagement with the health system, and often distinct socioeconomic circumstances.
Healthy-user bias compounds it. Women who receive and continue a prescription therapy into their seventies tend to have regular medical care, higher health literacy, and fewer competing conditions than women who do not. Any of those factors can independently affect dementia risk. The researchers adjusted for age, education, genetics, race, and hypertension, which addresses part of the problem but cannot address unmeasured differences.
Senior author Hadi Hosseini, a Stanford neuroscientist, told Scientific American that other factors may be at play, including the possibility that women who took estrogen-only therapy were healthier overall or had better access to health care. He also noted the team did not have information about why individual patients had been prescribed the therapy.
Timing adds another layer. Participants who used hormone therapy started, on average, after age 70. Current practice generally starts therapy in the late forties to early fifties and stops it before age 60, which limits how directly the findings apply to women making the decision today. Reporting on the study has noted that starting estrogen-only therapy before age 60 was linked to lower odds of Alzheimer's pathology, but that this particular association did not reach statistical significance.
The Decision This Does Not Change Today
Menopausal hormone therapy has established uses, and this study does not add a new one.
Hormone therapy is prescribed to manage menopausal symptoms, including hot flashes, night sweats, and genitourinary symptoms, and decisions about it involve individual factors, including age, time since menopause, cardiovascular history, breast cancer risk, and whether a woman has a uterus. Nothing in this study changes that framework, and no clinical guideline has been revised in response to it.
Women currently taking hormone therapy should not stop based on this or any single study, and women not taking it should not start in hopes of preventing dementia. The realistic use of this finding is as a question to raise at an appointment: whether hormone therapy makes sense for a specific person's symptoms and risk profile, with brain health as one consideration among several rather than the deciding one.
Alzheimer's affects women disproportionately, with about two-thirds of patients being women, and understanding why remains unsettled. Whether that reflects longer life expectancy, the hormonal changes of menopause, or social and genetic factors remains unresolved. Access to unhurried appointments where these trade-offs get discussed is itself uneven, a problem MedicalDaily has covered in reporting on adults postponing care until Medicare eligibility.
What would settle the question is a randomized trial testing hormone therapy against placebo with dementia outcomes, which no one has completed for this specific comparison. The study was supported by the National Institute on Aging. Until better evidence arrives, readers should expect further observational analyses and should treat any single result, in either direction, as one input rather than an answer.
Key Questions Answered
What did the study find? Among 21,462 women, estrogen-only menopausal hormone therapy was associated with 39 percent lower odds of a dementia diagnosis and 35 percent lower odds of Alzheimer's pathology at autopsy.
Does it prove hormone therapy prevents Alzheimer's? No. It is a retrospective observational study showing an association. The authors state it cannot establish cause.
What is healthy-user bias? The tendency for people who receive and continue a prescription therapy to be healthier in ways that are not measured can make the therapy look protective when other factors explain the difference.
Why only estrogen-only therapy? Previous research suggested that estrogen-plus-progestin combination therapy may increase dementia risk, so the researchers examined estrogen-only use separately.
What did the autopsies actually show? Among hormone therapy users, 18 percent had no signs of Alzheimer's disease compared with 10 percent of non-users, and 40 percent had all three hallmarks compared with 51 percent.
Should women start or stop hormone therapy because of this? No. No guideline has changed. Decisions about hormone therapy should be made with a clinician based on symptoms and individual risk.
When did participants start therapy? On average, after age 70, which differs from current prescribing practice and limits how directly the findings apply today.
