A signal across clinical and brain data

A study published in Neurology has renewed attention on a long-running question in women’s brain health: could the loss of estrogen around menopause influence later Alzheimer’s disease risk?

The researchers examined two existing datasets comprising 21,462 patient records. One included almost 3,000 people who had donated brain tissue after death; the other contained clinical and biomarker data from living participants. Across both sources, people recorded as using estrogen-only menopausal hormone therapy were less likely to have cognitive impairment or dementia diagnoses and showed fewer Alzheimer’s-associated changes.

In the autopsy dataset, estrogen-only therapy was associated with a lower burden of beta-amyloid plaques and tau tangles, the protein abnormalities central to Alzheimer’s disease. In the clinical dataset, estrogen users also had lower levels of Alzheimer’s-related biomarkers. The convergence of cognitive, diagnostic and neuropathological findings is the most notable aspect of the research: it suggests that the association is not limited to how patients perform on memory tests or to how clinicians assign diagnoses.

But the result is an association, not proof that estrogen therapy prevents Alzheimer’s disease. The study was retrospective, meaning that it looked back at health records and stored research data rather than randomly assigning participants to estrogen or placebo. That distinction is crucial when considering a treatment that carries meaningful benefits and risks depending on the individual, formulation and timing.

Why menopause is under scrutiny in Alzheimer’s research

Women account for nearly two-thirds of people living with Alzheimer’s disease in the United States. Longer average life expectancy explains part of that disparity, because age is the strongest known risk factor, but it is unlikely to explain the full difference. Researchers are therefore investigating how biology, cardiovascular health, genetics, education, social circumstances and reproductive ageing interact across the life course.

Menopause is a plausible period of interest because ovarian estrogen production falls substantially during the transition. Estrogen has wide-ranging effects beyond reproduction: it influences blood vessels, metabolism, inflammation and signalling in the brain. Laboratory and imaging studies have raised the possibility that estrogen loss may alter processes relevant to brain ageing, including glucose use, synaptic activity and vulnerability to amyloid and tau accumulation.

That biological rationale does not automatically make hormone therapy a dementia-prevention treatment. The effects of menopausal hormone therapy depend on variables that are often blurred in public discussion: the age at which therapy begins, the interval since menopause, dose, route of delivery, duration, and whether estrogen is used alone or with a progestogen.

What estrogen-only therapy means

Estrogen-only therapy is not interchangeable with all forms of menopausal hormone therapy. It is generally prescribed to people who no longer have a uterus, usually after hysterectomy. For people with an intact uterus, unopposed systemic estrogen can stimulate the uterine lining and raise the risk of endometrial cancer. A progestogen is ordinarily added to protect the endometrium.

This clinical difference also complicates interpretation of the new study. The datasets did not contain complete information on why participants received estrogen-only therapy or whether they had undergone hysterectomy. That matters because people who receive a particular therapy may differ from non-users in ways that affect dementia risk, including their medical history, contact with health services, socioeconomic circumstances and use of preventive care.

The researchers adjusted for several measured factors, including age, race and hypertension history. Such adjustments improve the analysis, but cannot remove biases arising from unmeasured or incompletely captured differences. For example, a group that accesses specialist menopause care may on average have other health advantages that reduce dementia risk independently of hormone use.

A mixed evidence base

The broader evidence on hormones and dementia has been inconsistent for decades. Observational studies have frequently suggested lower dementia or Alzheimer’s risk among some hormone-therapy users, especially when treatment began in midlife. A 2023 systematic review and meta-analysis found that observational evidence favoured estrogen-only therapy, while randomised trials in women aged 65 and older showed a higher overall dementia risk with hormone therapy, driven mainly by combined estrogen-progestogen treatment.

This apparent conflict may reflect genuine differences between therapies and the populations studied. It may also reflect the limitations of observational research: people who choose, can use or continue therapy may not be comparable with those who do not. Conversely, older randomised trials largely assessed specific oral preparations initiated relatively late in life, rather than the full range of therapies now used in menopause care.

Recent work has added further nuance rather than settling the question. A 2026 follow-up of the Women’s Health Initiative Memory Study found no significant difference between hormone-therapy and placebo groups in long-term change in blood biomarkers associated with Alzheimer’s pathology. Another 2026 systematic review of biomarker studies concluded that timing, formulation and hormonal composition may be important, but also called for better trials and more consistent measurement.

Taken together, the evidence supports a careful hypothesis: estrogen’s relationship with brain ageing may vary by treatment context. It does not yet support a general claim that menopausal hormone therapy reduces Alzheimer’s risk.

What research needs to answer next

The most informative next step would be prospective studies designed around brain outcomes from the outset. These should enrol participants near the menopause transition, document hysterectomy status and menopause timing, distinguish estrogen-only from combined therapy, and compare oral, transdermal and local treatments. They should also follow participants long enough to assess cognition, dementia diagnoses and validated Alzheimer’s biomarkers.

Randomised trials are particularly valuable because they can reduce the selection bias inherent in retrospective studies. However, trials involving hormone therapy must be designed with safety as a primary consideration. Potential harms vary with a person’s age, time since menopause, cardiovascular and clotting risk, cancer history, uterus status, and the medication regimen.

For now, estrogen-only therapy should not be started solely to prevent Alzheimer’s disease. Systemic menopausal hormone therapy remains principally a treatment for symptoms such as hot flashes and night sweats, with additional benefits and risks assessed individually. The new findings are scientifically important because they strengthen the case for targeted trials of menopause and brain health, not because they provide a preventive prescription.

A more precise public conversation

The renewed interest in estrogen and Alzheimer’s highlights a broader shift in dementia science. Rather than treating sex as a demographic variable to adjust away, researchers are increasingly examining reproductive ageing, hormone exposure and sex-specific biology as possible contributors to risk.

That approach may eventually improve prevention, but it also demands precision. “Hormone therapy” is not one intervention, and “Alzheimer’s risk” is not one outcome. The new study offers an encouraging signal for estrogen-only therapy and a reason to investigate it more rigorously. It does not overturn established caution around hormone prescribing or replace individual clinical decision-making.

Sources