A Study on Hormones and Dementia Just Made Headlines. Here Is What It Actually Found.
Somewhere this week a post told you that hormone therapy cuts Alzheimer's by a third. It came with a graphic, a lot of shares, and not much small print.
The study behind it is real. It was published in a serious neurology journal, it came out of Stanford, and it's genuinely one of the more interesting pieces of work in this area for years.
It also doesn't show what most of the posts said it showed. Both of those things are true at once, and the gap between them is worth 5 minutes of your time, because it's the same gap that has caused real harm to women before.
What the study actually did
Researchers used 2 large research databases holding records for 21,462 participants. The core of the work compared women who had used oestrogen-only hormone therapy with women who had used none.
The part that makes it unusual is what they looked at. Most research on hormones and the brain uses memory tests and questionnaires. This looked at donated brain tissue, at the actual physical changes that define Alzheimer's disease. A few hundred women had donated their brains to research after death, and that's a rare and generous thing that makes a study like this possible at all.
What it found
Women who had used oestrogen-only hormone therapy had around 35% lower odds of Alzheimer's changes in that tissue. They also had around 39% lower odds of having received a dementia diagnosis during their lifetime, and they did better on memory and independence measures.
That's a real result. I'm not going to talk you out of finding it interesting, because I find it interesting.
Why it isn't proof, and this is the whole point
Nobody in this study was randomly given hormones.
The researchers looked back at women who had already ended up on hormone therapy, for their own reasons, through their own doctors, in their own lives. Nobody was assigned to a group and followed forwards.
That matters more than it sounds. Women who start hormone therapy and stay on it for years aren't a random slice of the population. They're more likely to have a doctor who listens. They're more likely to be able to afford to keep going back. They're more likely to be engaged with their health in general, and to be better off. Every one of those things has its own effect on brain health, entirely separate from any hormone.
Researchers call this healthy user bias, and it isn't a small technical footnote. It's the thing that makes observational studies exciting and unreliable in the same breath.
Related reading
Brain fog in perimenopause is real, and it's one of the first things women are told to ignore. Perimenopause: the early signs and what changes first →
We have been here before with this exact hormone
Decades ago, studies built almost exactly like this one were the reason we all believed hormone therapy protected women's hearts. It wasn't a fringe view. It was in the textbooks and it shaped how a generation of women were treated.
Then a large randomised trial was run, where women genuinely were assigned to treatment or not. It didn't agree.
What followed was worse than the finding itself. The reversal was reported in a way that frightened women off hormone therapy almost overnight. Prescribing collapsed. Women who were doing well came off treatment, and many of them never went back on. Much of that trial has since been re-read and qualified, and the picture today is far more nuanced. But the fear outlasted the correction by about a generation.
All of that started with a strong observational signal that everybody treated as settled.
So when another strong observational signal arrives, about the brain this time, I want the randomised trial before I want the headline. Not because the finding isn't exciting. Because I've seen what it costs when we get ahead of ourselves with this particular hormone.
The small print worth knowing
Even taken at face value, this study covers a narrower slice than the posts suggest.
It looked at oestrogen-only therapy. Most women who still have a womb are given oestrogen with a progestogen, because oestrogen alone isn't suitable for them. The researchers couldn't draw a conclusion about combined therapy, because too few women on it had donated brain tissue.
It looked at tablets. Women using patches or gels weren't counted.
The women averaged around 70 years old, which is late to be starting or continuing hormone therapy. The authors suggest that might actually understate any effect, since the thinking is that timing matters, but it does mean this isn't a study of women starting in their 40s.
The brain tissue comparison rested on a few hundred donations, not thousands.
So where does this leave you
Hormone therapy isn't a dementia insurance policy. No honest clinician will tell you it is, and that hasn't changed this week.
What has changed is that the question is more interesting than it was, and better evidence is being built to answer it. Researchers are actively investigating whether hormone therapy started early affects long term brain health. That's a genuine open question, and open is the honest word for it.
There's also something quietly reassuring in here, as long as it isn't oversold. The frightening old headlines came from studies of older women taking older, combined, synthetic hormones. Study after study since has painted a calmer picture on safety, and this adds to it.
But reassuring about safety and protects your brain are 2 very different sentences. Be careful of anyone who slides from one to the other without telling you they have.
What to do with this
If you're on hormone therapy, this isn't a reason to change anything. If you're not, this isn't a reason to start. It's a reason to be a bit more sceptical of the next graphic that crosses your feed, including the ones that agree with you.
Take hormone therapy for what it genuinely does, which for the right woman is a great deal. Don't take it as a bet on your memory in 30 years, because that bet hasn't been priced yet.
And if you want somebody to look at your own situation rather than a headline, that's what a consultation is for.
A few things worth having alongside the conversation
These are the guides people ask about most when they're trying to work out what's actually going on. They're free.
FREE High Cortisol Guide What sustained stress looks like day to day, and where it overlaps with perimenopause. Get it free → |
FREE Iron Deficiency Guide Why a normal blood count doesn't rule out low iron stores, and what to ask for. Get it free → |
FREE Blood Pressure & Bloating Guide Two things that change quietly in your 40s and rarely get connected. Get it free → |
Book a Women's Health Consultation
One appointment that takes your cycle, your symptoms, your sleep and your bloods together rather than one at a time. You leave with a written plan and a date to review it.
I send one email a week, on Sunday mornings. It's called Beyond Prescriptions, and it's the same voice as this. Sign up here.
This is general information, not advice about your own health.
Source: Bruno J, Shaw JS, Hosseini SMH et al. Association Between Menopausal Hormone Therapy and Alzheimer Disease Neuropathology. Neurology, 107(5):e218413. DOI 10.1212/WNL.0000000000218413
Keep reading
If you've ever been told your bloods were fine and still felt awful, this one goes through what a normal result does and doesn't rule out. Your thyroid test was normal. Here is what that does and does not rule out →
