Perimenopause III
The Sentence Nobody Finishes
Perimenopause brain fog is real, measurable, and mostly temporary. What happens to women with ADHD in this window is a different question, and almost nobody has studied it.
There is a sentence women say in this transition, and they say it quietly, usually near the end of an appointment, usually while reaching for a coat.
I think there is something wrong with my brain.
Not: I am forgetful. Not: I am tired. Something is wrong with my brain. It comes with a specific fear attached, and most women will not name the fear out loud, because naming it makes it a real possibility instead of a bad thought at two in the afternoon.
Clinicians who work in this field write about it directly. Women arrive worried that what they are experiencing is the beginning of something serious, and a large part of the job is telling them what is actually known.[2]
So that is what this post is for.
The first post in this series was about the hormones and why they swing rather than fade. The second was about the hours nobody sees.
This one is about the part that frightens women most, and about a second question sitting underneath it that the research has barely touched.
What the testing actually found
Start with the good news, because it is real and it is specific.
SWAN followed 2,362 women for four years and tested them repeatedly on processing speed, verbal memory and working memory.[1] If you want to know whether the brain fog is measurable, this is the study that answers it.
The finding is subtler than "cognitive decline", and the subtlety is the whole point.
When you give someone the same cognitive test four times over four years, they get better at it. That is the practice effect, and in midlife it is expected. Premenopausal women improved. Early perimenopausal women improved. Postmenopausal women improved.
Women in late perimenopause did not.[1]
Their scores did not collapse. They flatlined. The learning curve went horizontal for the duration of the late transition, and then it came back.
Which happens to describe the experience precisely. Not a woman who cannot remember her own address. A woman who reads the same paragraph three times, who cannot get the new system at work to stick, who used to absorb things and now has to write them down.
Watch the shape of it.
The Curve That Went Flat
Cognitive test scores over repeated administrations. Stylised to show the pattern SWAN reported, not the exact published effect sizes.
Which functions, and how it feels
The changes cluster in specific places rather than spreading across everything.[2]
Verbal memory. Learning and recalling words, names, what someone told you on Tuesday. This is the domain that shows up most consistently across studies, including work following women in a separate community cohort.[5]
Processing speed. How fast you convert what you see into what you do. Not accuracy. Speed.
Verbal retrieval. The word is there. You know its shape, its first letter, the shelf it sits on. It will not come down. Then it arrives forty minutes later while you are doing something else.
Attention and working memory. Holding a thread while something interrupts it. Walking into a room and losing the reason.
Notice what is not on that list. Getting lost somewhere familiar. Not recognising people you know well. Losing the ability to manage money or medication. Those are different, and they belong in a different appointment, quickly.
Is it dementia
This is the question underneath the sentence nobody finishes, so let me answer it plainly.
What SWAN measured was a temporary flattening that rebounded after the transition.[1] The pattern of a neurodegenerative process is the opposite: it does not rebound, and it worsens.
The effect sizes in these studies are also small. Real, reproducible, and small. Large enough to be maddening in a job that depends on your memory, not large enough to look like impairment on clinical testing, which is exactly why so many women get told everything is fine while feeling that nothing is.
Two honest qualifications. The literature is not unanimous about whether every domain fully recovers; one longitudinal cohort found verbal learning difficulties persisting into postmenopause while memory recovered.[5] And "probably not dementia" is a statement about populations. A sudden change, a change other people notice before you do, or anything progressing steadily deserves a proper assessment rather than reassurance from an essay.
Why it happens
Estrogen is not only a reproductive hormone. Estrogen receptors are dense in the hippocampus and the prefrontal cortex, which are the structures handling exactly the functions above.
Now recall the first post. Estradiol in perimenopause does not fade smoothly. It surges and collapses, sometimes above the levels of a woman of twenty-five, then down within days. Those structures are being asked to run on a supply that keeps changing.
And the second post. Fragmented sleep degrades verbal learning and memory in anyone, at any age, which means some proportion of this is being manufactured at three in the morning rather than by estrogen directly.
Some, and not all. When SWAN researchers tested whether depressive symptoms, anxiety, sleep problems and hot flushes could account for the cognitive finding, those symptoms did not explain it away. Something was happening beyond them.
Which is worth holding onto, because "you are just tired" and "you are just stressed" are the two most common things a woman gets handed here, and the data does not support either as a complete answer.
The part where it stops being about brain fog
Now the harder question, and the reason I wanted this post in the series.
There are two groups of women arriving at this transition with something that looks like a cognitive problem and is not the thing described above.
The first group has had ADHD their whole lives and has managed it. Not with medication necessarily. With systems. Lists, alarms, rituals, over-preparation, a job chosen to fit, a partner who handles the admin. Compensation built over decades and running quietly in the background.
In perimenopause, for a lot of these women, the systems stop holding.
The second group has never been diagnosed with anything. They arrive in their late forties with new executive dysfunction: cannot start things, cannot finish things, cannot hold a plan together, emotionally raw in a way that is new. It looks like ADHD arriving. What it may be is ADHD that was always there, masked by decades of compensation and adequate estrogen, becoming visible when one of those supports goes.
Both groups tend to hear the same two sentences. It is just hormones. Or it is just your age.
Neither of those is a clinical response.
The mechanism people reach for
The explanation offered is dopamine. ADHD involves dopaminergic signalling, estrogen modulates dopamine, so falling and fluctuating estrogen should destabilise the system further.[4]
It is a reasonable mechanism. It is coherent with what we know about estrogen in the prefrontal cortex. It fits what women describe with uncomfortable precision.
And now I have to tell you what happens when you go looking for the studies.
What the research contains
In 2025 a team published a systematic review of ADHD and sex hormones across the female lifespan, searching for empirical studies covering puberty, the menstrual cycle, pregnancy, the postpartum period and menopause.
On ADHD during menopause, their search returned no empirical studies at all.[3]
Zero.
Not a small literature. Not a mixed literature. An absence, in the life stage where women with ADHD report the most difficulty, in a condition already known to be underdiagnosed in women.
The little work that does exist points in different directions. A separate 2025 study examining ADHD symptoms against menopausal experience did not find that women with ADHD report more severe menopause symptoms than women without.[4] Which is a genuine finding and sits awkwardly against what women say, and I would rather show you the awkwardness than hide it.
What that means practically: a woman whose life is coming apart in perimenopause, who has ADHD, walks into an appointment where nobody, including a scrupulous clinician who reads the literature, can tell her what is known. Because so little is.
Why the overlap makes this so hard
There is a second problem, and it is structural.
Look at the two symptom lists side by side. Inattention. Poor working memory. Emotional dysregulation. Disorganisation. Trouble starting tasks. Sleep problems. Restlessness. Irritability.
That list works for both. Almost every prominent feature of ADHD in adult women also appears on the perimenopause list, which means symptoms alone cannot separate them.
So the answer has to come from the history. Whether this is new or lifelong. Whether it tracks the cycle. Whether it arrived with the bleeding changes. Whether the systems that used to work stopped working, or were never there.
Which is a long conversation, and long conversations are the one thing a ten-minute appointment cannot hold.
Which List Does It Belong To?
Tap what applies to you. Orientation, not diagnosis, and the result is meant to show you why this is hard rather than to tell you what you have.
Why nobody studied this
Three failures stacked on top of each other, and each one has been documented separately.
ADHD in women was missed for decades, because the diagnostic picture was built on hyperactive boys and inattentive girls do not disrupt a classroom. So a generation of women reached midlife undiagnosed.
Perimenopause was under-researched, for the reasons in the first post of this series.
And the intersection of the two belongs to nobody. It falls between psychiatry and gynaecology, and each assumes the other is handling it. Neither is.
Which gives you an empty cell in a table where a great many women are currently living.
What to do inside an evidence gap
You cannot wait for the research. You are in it now. So this is the honest sorting of what is established, what is reasonable, and what is being oversold.
Sorted By How Much We Actually Know
Three tiers, honestly labelled. The middle tier is where most of this sits, and pretending otherwise would be the easy thing to do.
The practical part
Fix the nights before you conclude anything about your brain. Fragmented sleep produces exactly this symptom list in anyone. It is the largest confounder and the most treatable one, and it makes no sense to assess cognition through the fog of two years of broken sleep. The whole of the second post is about how.
Take the timeline, not the complaint. Whether this is lifelong or new is the single most useful piece of information for a clinician, and it is information only you hold. School reports. What you were like at university. Whether you built systems, when, and whether they stopped working. Write it before the appointment.
Ask for the assessment separately. If you suspect ADHD, ask for an ADHD assessment as its own referral rather than folding it into a menopause conversation. The overlap means it will otherwise be absorbed into the hormonal explanation and never examined.
Say the timeline out loud in both rooms. The gynaecologist should know you are exploring ADHD. The psychiatrist should know you are in perimenopause. Neither will ask.
Expect the honest answer to be uncertain. A clinician who tells you the evidence on ADHD in perimenopause is thin is being accurate, not dismissive. The problem is not that they cannot answer. It is that nobody has done the work that would let them.
And take the reassurance where it is earned. For the general cognitive change, the measured finding is a temporary flattening that recovered.[1] That is worth carrying around with you at two in the afternoon.
Three posts, one pattern
I set out to write about hormones and ended up writing the same paragraph three times in different vocabulary.
In the first post, a staging system with seven research priorities written into its founding document. In the second, a treatment that works over the telephone and is almost never offered. In this one, a systematic review that went looking for studies and came back with none.
Every time, the same shape. Women reporting something consistently, for decades. A plausible mechanism sitting right there. And an evidence base that thins out precisely where the women are standing.
I keep returning to the woman with the jacket, because she is what this looks like at ground level. She had to explain her own endocrinology to a room, badly served by science, holding a job together on no sleep, and doing it without the words that would have made it legible.
The words exist now. Not all of them. Enough to start with.
If you are in this and you have been carrying the sentence nobody finishes, put it down. What you have is a measurable, time-limited, mechanistically explainable change in a brain that is recalibrating to a signal that will not sit still. And if there is something else underneath it, you are allowed to go looking, even where the map runs out.
Especially there.
The perimenopause series
- Perimenopause hormones explained. Why they swing rather than fade, and why one blood test cannot see it.
- Perimenopause and sleep. The allopregnanolone withdrawal nobody mentions, and the treatment that works by telephone.
- Brain fog, cognition and ADHD. You are here.
For the women reading
If you took one thing from this series, take this: the difficulty you are having is not evidence of your decline. In most cases it is evidence of a transition that was measured late, explained badly, and researched last.
For the cognitive part specifically, the reassurance is real. What was measured was temporary and it recovered. Bring anything sudden, progressive, or noticed by other people to a doctor properly, because that is a different question and deserves a real answer.
And if you have spent your life quietly compensating for something nobody named, and the compensation has stopped working, that is worth investigating in its own right, as its own referral, in its own appointment. The absence of research is not the same as the absence of a problem. It only means nobody looked.
None of this is medical advice, and none of it knows your history.
References
- Greendale, G. A., Huang, M.-H., Wight, R. G., Seeman, T., Luetters, C., Avis, N. E., Johnston, J., & Karlamangla, A. S. (2009). Effects of the menopause transition and hormone use on cognitive performance in midlife women. Neurology, 72(21), 1850–1857. https://doi.org/10.1212/WNL.0b013e3181a71193
- Maki, P. M., & Jaff, N. G. (2022). Brain fog in menopause: A health-care professional's guide for decision-making and counseling on cognition. Climacteric, 25(6), 570–578. https://doi.org/10.1080/13697137.2022.2122792
- Osianlis, E., Thomas, E. H. X., Jenkins, L. M., & Gurvich, C. (2025). ADHD and sex hormones in females: A systematic review. Journal of Attention Disorders, 29(9), 706–723. https://doi.org/10.1177/10870547251332319
- Chapman, L., Gupta, K., Hunter, M. S., & Dommett, E. J. (2025). Examining the link between ADHD symptoms and menopausal experiences. Journal of Attention Disorders. Advance online publication. https://doi.org/10.1177/10870547251355006
- Epperson, C. N., Sammel, M. D., & Freeman, E. W. (2013). Menopause effects on verbal memory: Findings from a longitudinal community cohort. The Journal of Clinical Endocrinology & Metabolism, 98(9), 3829–3838. https://doi.org/10.1210/jc.2013-1808