SAN FRANCISCO
Ask most women what age their periods stopped and they can tell you within a year or two. Ask them whether they have ever mentioned that number to the doctor treating their memory, and you will get a blank look.
That is about to change.
A study led by researchers at the University of California, San Francisco, published in JAMA Network Open on August 25 and now rippling through neurology and women’s health coverage on both sides of the Atlantic, has produced the clearest evidence yet that the timing of menopause leaves a physical signature on the ageing brain — one that is still visible on an MRI scanner thirty years after the last period.
The women in this study were followed from an average age of 78. For up to 18 years. Some of them until they died and donated their brains to science. And across all that data, one unglamorous number on a medical history form kept predicting how the rest of the story went.
The age menopause began.
The number: 15 per cent
The headline finding is deceptively simple, and its implications are anything but.
Researchers analysed data from roughly 2,600 women enrolled in two of the most respected long-term ageing studies in the world: the Rush Memory and Aging Project and the Religious Orders Study, both led out of Rush University Medical Center in Chicago and running for decades. Participants — a cohort that includes a large number of nuns, priests and brothers alongside community-dwelling older adults — agreed to annual clinical evaluations, annual neuropsychological testing, brain imaging, and brain donation after death. It is, in research terms, an extraordinary resource: real people, tracked in obsessive detail, all the way to the end.
When the UCSF team lined menopause age up against what the MRI scanners were showing decades later, they found this: over a ten-year window, a woman whose menopause occurred earlier might accumulate roughly 15 per cent more damaged white matter than a woman whose menopause occurred five years later.
Fifteen per cent more injury, in the wiring of the brain, purely as a function of when a hormonal transition happened in midlife.
The researchers deliberately did not impose an arbitrary age cutoff for what counts as “earlier.” They simply compared women against each other. Among the women who went through natural menopause, the average age was 50. Among the roughly one third who experienced surgical menopause — an operation that removes the ovaries and ends the menstrual cycle overnight — the average age was 43.
What white matter damage actually does to you
“White matter hyperintensities” is a phrase that means very little to most people, which is unfortunate, because they are among the most common findings on brain scans in older adults and among the most consequential.
White matter is the brain’s cabling — the insulated fibres that carry signals between regions. Hyperintensities are bright patches on an MRI where that cabling has been damaged, usually by small-vessel disease: the slow, quiet degradation of the tiniest blood vessels feeding the brain.
They do not typically announce themselves with a dramatic event. There is no white-matter equivalent of a stroke you can feel. Instead they show up as the things people in their sixties and seventies describe with embarrassment and vagueness: losing the thread mid-sentence, needing the room quiet to concentrate, taking three times as long to work through a form, forgetting why you walked into the kitchen. Formally, the burden of white matter hyperintensities has been linked to difficulties in attention, focus, executive functioning and information processing.
And they accumulate. Slowly, silently, for years before anyone runs a cognitive test.
“Injury in the white matter becomes increasingly common as we age,” said senior author Dr Riley Bove, director of the UCSF Sex and Gender Enriched Program and the Rosenberg Ach Family Endowed Professor in Neuroimmunology at the UCSF Weill Institute for Neurosciences.
“Along with aging itself, other conditions like hypertension, elevated blood sugar and cholesterol, and obesity can also contribute to this risk. Estrogen appears to provide some protection against these conditions, or against underlying brain vulnerability to injury, until it declines with menopause.”
It was not just the scans
If the study had stopped at brain imaging it would still be notable. It did not stop there.
Earlier menopause was also associated with a slightly faster decline in overall thinking ability and in episodic memory — the brain’s filing system for specific personal events, the kind of memory that lets you retrieve a conversation from last Tuesday or a holiday from 1994. Episodic memory is characteristically among the first things Alzheimer’s disease takes.
For every year earlier that menopause occurred, the researchers observed a small but statistically significant increase in the rate of global cognitive decline and episodic memory decline.
And the team found an association between earlier menopause and earlier clinical onset of Alzheimer’s disease itself.
Not necessarily more Alzheimer’s. Earlier Alzheimer’s. Which, for a disease whose stolen years are counted in the ability to recognise your own children, is not a small distinction.
“These findings show that timing of menopause can influence how the brain ages, even decades later,” said first author Dr M. Pia Campagna, a postdoctoral fellow in the UCSF Department of Neurology and the Weill Institute for Neurosciences. “There is a major effort to study the menopause window to identify interventions that could optimize brain health as women age.”
The study’s formal conclusion is blunt about what it thinks this is for: menopause age, the authors write, is a midlife-identifiable marker for risk stratification — offering a window for targeted prevention before structural or cognitive changes become apparent.
Translated: a number every woman already knows, available decades before the first symptom, that nobody is currently using.
The signal was in the wiring, not the size
One nuance is worth being straight about, because early coverage has blurred it.
The clearest and most consistent signal in this study was in white matter damage — the wiring — and in the cognitive and Alzheimer’s-timing outcomes. The picture on overall brain volume, the straightforward shrinkage that people usually mean when they say “brain atrophy,” was considerably less clean; independent summaries of the paper noted that the researchers did not find a link between menopause timing and brain volume in this cohort.
That is not a weakness of the study. It is arguably its most interesting detail. It suggests that whatever oestrogen is doing for the brain, it may be doing it primarily through the blood vessels and the wiring rather than by simply preserving bulk tissue — which points investigation squarely at vascular protection as the mechanism to chase.
Why this lands on two-thirds of Alzheimer’s patients
The context here is stark. Roughly two thirds of people living with Alzheimer’s disease are women, most of them postmenopausal. That imbalance has been known for years and explained badly — usually with a shrug about women living longer, which the data has never fully supported.
A growing body of research points instead at midlife, and at the hormonal cliff-edge that only half the population goes over.
A meta-analysis pooling 22 studies and close to five million women previously found that later menopause — after age 55 — was associated with a 33 per cent reduced risk of Alzheimer’s disease, a 13 per cent reduced risk of all-cause dementia and an 18 per cent reduced risk of cognitive impairment. Premature menopause, defined as occurring before age 40, and early menopause, before 45, have both been repeatedly linked to higher dementia risk.
What the UCSF study adds is the thing those earlier papers could not deliver: an actual physical substrate, visible on a scanner, measured in the same women over almost two decades, in a cohort where the brains were later examined directly.
The hormone therapy question everybody will ask
The obvious next question — if oestrogen is protective, does replacing it protect the brain? — does not have a clean answer, and anyone who tells you otherwise is selling something.
Evidence that menopausal hormone therapy prevents cognitive decline or dementia remains inconclusive. Related recent work has found that hormone therapy did not significantly modify the association between menopause timing and brain volume trajectories. The picture is further complicated by timing: the age at which hormone therapy is started appears to matter, and starting it late may not deliver benefits observed when it is started early.
Where the evidence is stronger is at the extremes. For women who experience premature menopause before age 40 — and potentially early menopause before 45 — current guidance supports hormone therapy up to the typical age of natural menopause, precisely because those women are otherwise living an extra decade or more without hormones their bodies would ordinarily still be producing.
That is a specific clinical recommendation for a specific group. It is not a general licence, and this study does not turn into one.
The more immediately actionable implication is the unglamorous one Dr Bove gestured at: hypertension, blood sugar, cholesterol and obesity all independently damage white matter. Those are modifiable. A woman who went through menopause at 43 cannot change that fact — but she can change how hard her small blood vessels are being worked for the next thirty years.
What the study cannot tell you
The honest limitations matter.
This is an observational cohort study. It establishes association, not causation. It is entirely possible that whatever caused a woman’s ovaries to stop early — genetics, autoimmune disease, smoking, socioeconomic hardship, chemotherapy — is also independently damaging her brain, with menopause timing serving as a flag rather than a cause.
The cohort is also unusual. The Religious Orders Study and the Rush Memory and Aging Project skew towards highly educated participants, include a large religious-order population, and are not demographically representative of women worldwide. Menopause age in older cohorts is frequently self-reported and recalled across many decades.
And the effect sizes, while real and statistically significant, are described by the researchers themselves as small on a per-year basis. Fifteen per cent more white matter damage over ten years is meaningful at a population level; it does not mean any individual woman who finished menopause at 46 is destined for dementia. Most will not develop it.
The study was funded by the National Institutes of Health and the National Institute of Neurological Disorders and Stroke, with additional support from the McKnight Brain Research Foundation. Co-authors included Dr Julie A. Schneider, Dr Lisa L. Barnes, Dr Konstantinos Arfanakis and Dr David A. Bennett of Rush University Medical Center, and Dr Galit Levi Dunietz of the University of Michigan.
What this means for you
If you are a woman, or you love one, four things follow.
Know the number and say it out loud. The age your periods stopped — or the date of the surgery that stopped them — belongs in your medical history alongside your blood pressure and your family history of heart disease. If it happened before 45, that is worth raising specifically. Most doctors are not yet asking.
Treat the years after menopause as a prevention window, not a wind-down. The study’s authors frame menopause age as a marker that identifies risk decades before symptoms appear. The entire value of an early marker is the time it buys you — and the interventions that protect white matter are the deeply boring ones: blood pressure control, blood sugar control, cholesterol management, weight management, not smoking, moving your body, sleeping.
If you had surgical menopause, pay closer attention. A third of the women in this study did, at an average age of 43 — seven years earlier than the natural-menopause group. Removal of the ovaries is sometimes clinically unavoidable and sometimes done as a precaution alongside a hysterectomy. Anyone facing that decision now has one more thing to weigh, and one more question to ask their surgeon.
Do not panic-order hormone therapy off the internet. The evidence on hormone therapy and dementia prevention is genuinely unresolved. It is a conversation to have with a clinician who knows your history, your cardiovascular risk and your cancer risk — not a conclusion to draw from a headline.
The bottom line
For most of medical history, menopause was filed under gynaecology, discussed for about five years either side of the event, and then dropped.
This study, tracking women into their nineties and in many cases into the pathology lab, argues that this filing was wrong. Menopause is not an episode that ends. It is a hinge — and the position of that hinge appears to still be exerting force on the brain three decades later, in white matter that is quietly fraying, in memory tests that slip a little faster, in an Alzheimer’s diagnosis that arrives a little sooner.
The consolation is that a hinge you can see is a hinge you can plan around. Every woman reading this already knows her number. The medical system is only now working out what to do with it.
Which means the most useful thing to come out of a study that took eighteen years might be a question that takes eighteen seconds to ask at your next appointment.
