She joined me for a telemedicine visit last week - sharp, accomplished, and frustrated.

"It's embarrassing, but I feel like I'm losing my mind," she said. "I'll be mid-sentence and the word just... disappears. I walk into a room and forget why I'm there. I read something and five minutes later I can't tell you what it said. My previous doctor said my hormone levels are fine."

She's not alone. Nearly half of postmenopausal women report forgetfulness. Many describe the same pattern: brain fog, slower thinking, trouble making decisions, words that vanish mid-thought. These aren't vague complaints. Research has shown that the brain skills most affected during and after menopause are exactly the ones these women are describing - finding words, processing information quickly, and holding onto recent memories.

I hear some version of this story on repeat in my practice each week. And I ask every patient about it - not just "any trouble remembering things?" but the specific things that matter: Are you losing words mid-sentence? Does your thinking feel slower? Are decisions harder than they used to be? Is it taking longer to process what someone just said to you?

These questions matter. Because if you're on hormone therapy (HT) and still experiencing these symptoms, the answer isn't that HT doesn't work. The answer may be that your treatment hasn't yet been fine-tuned, and new research is helping us understand why.

New research out of Northwestern Medicine changes how we should think about estrogen, memory, and Alzheimer's risk

In May 2026, researchers at Northwestern University Feinberg School of Medicine published a study in the journal Aging Cell that identified something previously unstudied: a specific process that helps explain why women lose memory function after menopause, and potentially why nearly two-thirds of Americans with Alzheimer's disease are women.

The answer isn't simply "low estrogen." It's about what happens to the physical structure of your brain when estrogen has been missing.

What the researchers found

Between every brain cell, there's a network of molecules called the extracellular matrix, or ECM. Think of it as the mortar between bricks. It fills the spaces between your brain cells, makes up about 20% of your brain's total volume, and helps brain cells send signals to each other. This scaffolding is especially concentrated in the hippocampus- the part of your brain responsible for learning and memory, and also where Alzheimer's disease begins.

The Northwestern team showed something for the first time: when the brain loses its ability to make estrogen on its own, this scaffolding changes in ways that could interfere with how brain cells talk to each other, and this only happened in older females.

Male mice with the same estrogen loss didn't show the same changes.

Why does this matter? Prior research has shown that women who develop Alzheimer's have lower estrogen levels in their brain tissue than women without the disease. This study adds a possible explanation: the scaffolding between brain cells may be part of the story.

A note about this study: This research was done in mice, not people. The scientists created mice whose bodies couldn't make estrogen- some couldn't make it anywhere in the body, and some couldn't make it only in the brain. That let them see exactly what happens when the brain loses its own estrogen supply. These results don't change how doctors should prescribe HT today, but they point in an important direction, and they support what other research has been suggesting: the type of estrogen, how it's delivered, and when it's started all seem to matter.

What this means if you're already on HT

Here's where the science meets what I see in my practice every week.

A patient comes to me on estradiol. Her levels look good. She's been on HT for two years. She's doing everything she was told to do. And she still can't find words mid-sentence. She still walks into rooms and forgets why she's there. She still leaves a conversation and can't recall what was said.

Her previous provider told her it was stress. Stress does play a role, but that answer still falls short.

This study helps explain why. Your brain doesn't just use estrogen from your bloodstream. Before menopause, your brain also makes its own estrogen on-site, using an enzyme called aromatase. After menopause, that local production drops. HT puts estrogen back into your blood, but the brain's own estrogen supply works somewhat independently. Scientists don't yet know whether HT fully restores what's happening inside the brain itself - that question hasn't been directly tested. But the biology suggests that a blood level alone may not tell the whole story.

This is likely one reason why studies on HT and brain function have produced mixed results. The Northwestern researchers said as much: differences in study outcomes may depend on the type of hormone used, the age when treatment begins, and how the studies were designed.

When you start matters, but it's not the whole story

The scaffolding changes in this study showed up in older females, not younger ones. This fits with something researchers call the "critical window,"  the idea that estrogen protects the brain best when it's started closer to menopause, before the brain's structure has had time to change without it.

Large reviews of existing research support this idea. Women who started estrogen-only therapy around midlife had roughly a 32% lower risk of developing dementia. That's a meaningful number.

But it's not that simple. The ELITE trial, the only study specifically designed to test whether starting earlier makes a difference- found that women who started estradiol pills within 6 years of menopause didn't score better on memory or thinking tests than women who started 10 or more years later. That study used oral estradiol (a pill), which may be part of the explanation- because how estrogen gets into your body affects how much reaches your brain.

This is where route of delivery comes in. A large Canadian study found that women using estradiol through the skin (a patch or gel) performed better on memory tasks that depend on the hippocampus- the same brain region affected in the Northwestern study. Women taking estradiol as a pill performed better on a different type of memory- remembering to do things in the future. Neither form was linked to worse thinking compared to women not on HT.

What is clearer is the risk of waiting too long. In the Women's Health Initiative Memory Study, women over 65 who were started on a specific combination, a type of estrogen derived from horse urine (Premarin) plus a synthetic progesterone called medroxyprogesterone acetate, had roughly double the risk of dementia compared to women who took a sugar pill. This combination and this age group are not what most women are prescribed today, but the finding matters: starting late, with the wrong formulation, may cause harm.

The bottom line: when you start HT likely matters. What type you use matters. How it gets into your body matters. And these factors interact with each other in ways science is still working to fully understand.

What I do with this in my practice

I don't wait for brain fog to become disabling before I start asking the right questions. Every patient gets asked about word-finding, processing speed, decision-making, and whether her thinking feels slower than it used to be.

Here's what the evidence tells us about the details that matter:

Route of delivery matters. A patch or gel delivers estradiol through the skin, bypassing the liver. This keeps a healthier balance of estrogen types in your blood and has been linked to better performance on the kind of memory that depends on the hippocampus. A pill works differently- it goes through the liver first, which changes how the estrogen is processed.

The type of progesterone matters too. Studies have shown that the specific synthetic progesterone used in the WHI study (medroxyprogesterone acetate) had a negative effect on verbal memory. Other types of progesterone did not show this problem. This is an underappreciated detail.

Sleep quality matters. Your hippocampus locks in memories while you sleep. If sleep is disrupted, whether from hot flashes, stress, or other causes, memory suffers no matter what your estrogen level says.

The full picture matters.  Estrogen levels alone don't tell the whole story. Cortisol, thyroid function, the type of progesterone you use, and how well you sleep all interact with how your brain works. No single lab number can capture all of that.

Your next step

If you've been told your levels are fine and that should be enough - I'd push back on that.

Getting your levels into range is a good first step. But it's not where the work ends.

There's a difference between a provider who starts HT and one who fine-tunes it. If you're still foggy, still forgetting, still not feeling like yourself-  that's worth investigating, not accepting.

That's not a failure on your part. It's a gap in your care, and it's exactly what I help my patients close.

Resources