What the Lancet’s Menopause-Dementia Study Actually Tells Us, and What It Doesn’t
Women develop Alzheimer’s disease at nearly twice the rate of men. Two out of every three Alzheimer’s cases are in women. And yet for most of the past two decades, women were told that hormone therapy, one of the most biologically logical interventions we have for the estrogen-brain connection, was too risky to consider.
So when a major Lancet meta-analysis reviewed the evidence on menopause hormone therapy and dementia risk across more than a million women, the headlines ran with a clean conclusion: HRT doesn’t increase or decrease dementia risk.
Here’s what I want you to understand about that headline: it is not a clean bill of health. It is a statement about the limits of the evidence we currently have. If you read those two things as the same, you’ll make a very different decision than the biology actually warrants.
What the Study Found, and What It Couldn’t Say
The December 2025 review by Melville and colleagues at University College London analyzed studies from January 2000 through October 2025, covering more than a million women’s health records.
Their conclusion: the available evidence does not confirm whether hormone therapy has a positive, negative, or null effect on the risk of dementia or mild cognitive impairment.
Read that carefully. They are not saying HRT has no effect. They are saying the current evidence base cannot tell us what the effect is. That is a very different statement.
Why the Data Can’t Give a Clean Answer
So why can’t the evidence tell us? Because the data is genuinely messy.
The vast majority of studies reviewed were observational, meaning researchers watched women who happened to take or not take hormones, rather than randomly assigning treatment. That creates enormous confounding. Specifically, women who seek hormone therapy tend to be healthier, more proactive about their care, and more engaged in their health overall. You can’t separate that from the effect of the hormones themselves.
On top of that, the studies combined radically different approaches under one label: oral versus transdermal, estrogen-only versus combined, bioidentical versus synthetic, started at age 48 versus started at age 65. Combining all of those into “MHT” and expecting a clean answer is like combining aspirin, chemotherapy, and blood pressure drugs and asking whether “medication” prevents heart attacks. The signal gets buried in the noise.
What the Biology Actually Tells Us
Here’s where I part ways with a purely agnostic reading of the data.
The mechanistic case for estrogen’s role in brain health is not ambiguous. In fact, it is solid and it has not changed.
Estrogen receptors sit throughout the hippocampus, the prefrontal cortex, and the brainstem. Estrogen promotes the brain’s ability to form and hold new connections, which is the core mechanism of learning and memory. Brain cells also depend on estrogen to generate energy well. Additionally, amyloid plaque buildup, the hallmark of Alzheimer’s pathology, is reduced when estrogen is present. And the blood vessels feeding the brain stay healthier as a result.
When estrogen drops at menopause, all of that protection is withdrawn at once. The National Institute on Aging acknowledges that sex hormones play a role in Alzheimer’s risk differences between men and women. Menopause reshapes the brain through hormonal shifts and the loss of estrogen affecting neural pathways, mood, and memory. This is not controversial. Rather, this is textbook neuroscience.
The question that observational data cannot cleanly answer is whether hormone therapy, given the wide range of forms and timing windows studied, translates this biological protection into lower dementia rates at the population level. That’s a harder question. Still, the absence of a clean population-level answer does not mean the underlying biology has changed.
Why Timing Still Matters
The Lancet review found no clear evidence for a meaningful timing window. I read that finding against a broader body of research that suggests otherwise, including studies showing midlife estrogen use tied to meaningfully lower dementia risk.
The “healthy cell” idea gives us a mechanistic reason to believe timing matters: estrogen is brain-protective only when neurons are still metabolically intact. If you wait until years of estrogen loss have already impaired how cells generate energy and allowed amyloid buildup to begin, the hormone may no longer be working in a healthy environment. In other words, the protective effect may be missed not because it doesn’t exist, but because you waited too long to introduce it.
This is why I keep returning to the early window. Not because the trial data is definitive. Because the biology is.
What Brain Health During Menopause Actually Requires
First, hormonal support, timed early when possible, delivered through the skin when possible, and bioidentical when possible. Because the cell-level biology of estrogen in the brain is real and it matters clinically.
However, this is not a single-intervention problem. Insulin resistance is independently tied to Alzheimer’s disease. Some researchers now call it Type 3 diabetes. Controlling blood sugar, reducing systemic inflammation, and supporting how brain cells generate energy are not optional extras. They are core to brain protection during the menopause transition.
Sleep is not optional either. During deep sleep, the brain runs its own waste-clearing system, flushing out amyloid and other metabolic debris. Research from the National Institutes of Health confirms that this overnight clearing process is disrupted by poor sleep quality. Menopause wrecks sleep. So treating that disruption, whether through hormonal support, sleep hygiene, or targeted functional medicine tools, is one of the most important things a woman can do for long-term brain health.
And finally, resistance training. According to the Alzheimer’s Association, regular physical activity is among the most evidence-backed strategies for reducing dementia risk available. Building and keeping muscle through the menopause transition is not about how you look. It is one of the most brain-protective things you can do.
Key Takeaways
- A December 2025 Lancet meta-analysis found that current evidence cannot confirm whether hormone therapy increases, decreases, or has no effect on dementia risk. This is different from a finding of “no effect.”
- Most studies analyzed were observational and combined vastly different forms, timing windows, and delivery routes under one label. The data is too noisy for a clear signal.
- The biological case for estrogen’s brain-protective role, covering memory formation, cellular energy, amyloid clearance, and brain blood vessel health, remains strong and unchanged.
- The early initiation window within 10 years of menopause is important based on the biology, even where the population-level data is not yet conclusive.
- Brain health during menopause requires a full strategy: hormonal support, metabolic health, sleep quality, and resistance training.
Frequently Asked Questions
Does hormone therapy prevent dementia? The current evidence cannot confirm that HRT prevents dementia at the population level, primarily because the studies available are observational and inconsistent in their methods. However, the biological case for estrogen’s brain-protective role is well established. Estrogen supports the brain’s ability to form new connections, reduces amyloid plaque buildup, and helps brain cells generate energy. The absence of definitive trial data does not mean the biology is wrong. It means we don’t yet have the right studies to measure it cleanly.
What did the Lancet study actually find about HRT and dementia? The December 2025 Lancet systematic review analyzed over a million women’s health records and concluded that the available evidence does not confirm whether hormone therapy has a positive, negative, or null effect on dementia risk. Crucially, that is not the same as finding no effect. The authors pointed to observational study bias, inconsistent formulations, and mixed timing windows as reasons the evidence cannot produce a clear answer. For a deeper look at how sex hormones relate to cognitive aging, the National Institute on Aging provides useful background.
Is there a window of time when hormone therapy is most protective for the brain? The biology strongly suggests yes, even though the Lancet review did not find statistical confirmation. The “healthy cell” concept proposes that estrogen is most protective when neurons are still metabolically healthy. Starting hormone therapy within 10 years of menopause, before significant amyloid buildup or cellular energy impairment has occurred, is when the biological environment is most likely to respond favorably. Waiting until later may mean the protective conditions are no longer in place.
Why do women get Alzheimer’s more than men? The sex difference in Alzheimer’s risk is significant. According to the Alzheimer’s Association, women account for nearly two-thirds of all Alzheimer’s cases in the United States. Estrogen loss at menopause is one of the leading proposed explanations, given estrogen’s well-documented roles in supporting brain function, energy production in neurons, and clearance of amyloid deposits. Longevity is a factor too, as women live longer on average. But the biology of estrogen withdrawal is increasingly viewed as a major contributor.
What else can I do to protect my brain during menopause? Several interventions have strong evidence behind them. Controlling insulin resistance is one of the most important, since insulin resistance is independently tied to Alzheimer’s risk. Prioritizing sleep matters enormously, because the brain’s overnight waste-clearing process removes amyloid buildup. Resistance training is among the most evidence-backed preventive strategies available, according to the Alzheimer’s Association. Reducing systemic inflammation through diet and stress management rounds out a functional medicine approach to brain protection. Hormone therapy, used appropriately and timed well, works alongside all of these strategies.
Should I take hormone therapy specifically for brain protection? That’s a conversation that needs to happen with a provider who knows your full picture. However, given what we know about estrogen’s role in brain function and the fact that doing nothing is also a choice with real biological consequences, many women in early menopause have good reasons to consider this seriously. The best approach is a comprehensive evaluation: your hormonal status, metabolic health, sleep quality, family history, and cardiovascular profile. Your decision should be based on your biology, not on a headline.
Dr. Betty’s Bottom Line
I want to give you the honest read on this study, not the comfortable one.
The headline says “HRT doesn’t help or hurt dementia risk.” What the authors actually said is: we cannot tell from this evidence. That distinction matters for how you think about your own choices.
The biology of estrogen in the brain has not changed. The cell-level case for estrogen’s role in brain protection is as strong as it has ever been. What we don’t have yet is the long-term, well-designed trial with the right formulations, the right timing, and decades of follow-up. We may never have it.
What we do have is biology. We have clinical observation across 20 years of practice. And we have the clear knowledge that doing nothing, leaving your brain without estrogen support during its most vulnerable transition, is also a choice. It just doesn’t come with a warning label.
At Living Well Dallas and through Menrva Health, my job is to give you the full picture and build a plan based on your biology. Not just a prescription. And definitely not “the evidence is unclear, good luck.” You deserve better than that.
Ready to find YOUR root cause? Visit getmenrva.com for telehealth nationwide, or livingwelldallas.com for in-person care in Dallas.
Source: Melville M, He L, Desai R, et al. Menopause hormone therapy and risk of mild cognitive impairment or dementia: a systematic review and meta-analysis. The Lancet Healthy Longevity. Published online December 22, 2025. PMID: 41448220.