Perimenopause II
The Hours Nobody Sees
Perimenopause sleep problems are not a side effect of the transition. For a great many women they are the transition, and the mechanism has a name.
The last post was about a woman who threw her jacket on the floor in front of four men who were laughing at their wives.
This one is about the hours nobody sees.
I have watched my mother move through this. I have watched other women do it. And what none of them led with was hot flushes.
What they mentioned, always as an aside, always as if it were the boring part, was that they were not sleeping.
Not tired. Not run down.
Awake. At three, at four, with a heart going faster than the situation called for, doing arithmetic about how many hours were left.
Then getting up at seven and going to work.
We file that under side effect. Something the hormones do on the way past, unpleasant but secondary, the price of the main event.
It is not secondary. A great deal of what gets called perimenopause is manufactured in those hours. The flatness. The word that will not arrive. The blood sugar that behaves differently than it did last year. The mood that has no subject. Take a woman's sleep apart for three years and you will produce most of that list in a woman of any age.
So this post is about the night. What is happening in it, why three in the morning specifically, and what the evidence says helps.
How common this is
Between forty and fifty-six percent of women report insomnia symptoms during the menopause transition. Around twenty-six percent meet full diagnostic criteria for insomnia disorder, meaning symptoms severe enough to damage daytime functioning.[1]
One in four.
And the increase tracks the transition itself rather than age. Women moving through perimenopause report more trouble falling asleep, more waking through the night, and more waking earlier than they meant to, compared with where they started.[1]
The complaint that dominates is not getting to sleep. It is staying there.
Why three in the morning
Several things are happening at once, and they are separate mechanisms with separate answers. This matters, because a woman handed one solution for a problem she does not have concludes that nothing works.
The molecule you stopped making
In the first post I wrote that progesterone leaves first, because it needs ovulation to exist at all. Here is what that costs you at night.
Progesterone crosses into the brain and gets converted, through two enzymatic steps, into allopregnanolone. Allopregnanolone is a positive allosteric modulator of the GABA-A receptor: the same receptor targeted by benzodiazepines and by zolpidem.
Remove ovulation from that chain and the whole thing goes.
Which means a woman in perimenopause is in withdrawal from a sedative she was producing herself, every month, for thirty years. Nobody told her she was taking it. Nobody mentions that she stopped.
Work tracking sleep continuity against measured reproductive hormone profiles across the perimenopause found exactly this kind of link between hormonal patterns and the fragmentation of sleep, rather than between hormones and total sleep time.[2] It is continuity that goes.
The heat
The second mechanism is the obvious one. A vasomotor event during sleep brings an arousal with it, and a substantial share of recorded night-time flushes are accompanied by waking.[1]
What makes this hard to see from the outside is that the arousals are brief. Total time in bed can look almost untouched on a sleep tracker while the architecture underneath has been shredded. The woman knows. The device does not.
And the part that is neither
Here is where I want to be careful, because the honest version is more useful than the tidy one.
Sleep disruption in perimenopause is not fully explained by hot flushes. Women without significant vasomotor symptoms develop insomnia in this window too. Depressive symptoms, anxiety, the stress axis, and sleep-disordered breathing all contribute, and they contribute independently.[1]
Which is why "treat the flushes and the sleep will follow" fails for so many women. Sometimes the flushes were never the thing.
Below: the same eight hours, three ways.
The Same Night, Three Ways
A hypnogram maps which stage of sleep you are in across the night. Stylised from published patterns, not a recording of one woman.
What the recordings show, and what they miss
Something uncomfortable sits in this literature and I would rather you heard it from me.
When researchers put midlife women in a lab and record them overnight, the differences between menopausal stages are often smaller than the differences in how women describe their nights.[1] Some studies find little change in deep sleep at all.
You can see how that gets used. A woman says her sleep has fallen apart. A study says the architecture looks broadly intact. The gap becomes a reason to doubt her.
Read it the other way. A single night in an unfamiliar lab, wired to equipment, captures one night out of a transition that runs for years and whose defining feature is variability. It also measures the wrong thing. The complaint is not duration. It is fragmentation, arousals, and the felt quality of the night, and standard staging was never built to capture those.
Where researchers have looked at women whose insomnia began in the transition, the deficits show up plainly: shorter measured sleep, more time awake after falling asleep, worse efficiency.[1]
The instrument was wrong. Not the woman.
What is waking you
Different mechanisms, different answers. Tap what fits your nights.
Which Night Is Yours?
Orientation, not diagnosis. Several of these run together, and the combination is the useful information to bring to an appointment.
The one that gets missed
One item on that list deserves its own paragraph, because it is the thing most likely to be overlooked in a woman of this age.
Sleep-disordered breathing becomes markedly more common after the menopause transition. Before it, obstructive sleep apnoea is largely thought of as a condition affecting men, and the diagnostic picture most clinicians carry was built on male presentation: loud snoring, witnessed pauses, daytime sleepiness.
Women present differently. Insomnia. Fatigue rather than sleepiness. Low mood. Morning headache. Waking unrefreshed after adequate hours.
Which is a list that maps almost exactly onto what everyone, including the woman herself, will attribute to hormones.[1] So she gets an antidepressant, or a sleeping tablet, and the breathing goes unexamined for years.
If you sleep long hours and wake wrecked, that is worth naming out loud to a doctor, in those words.
Why nobody fixes this
A woman with three years of broken sleep walks into an appointment. What she is most likely to leave with is a prescription for a hypnotic, or advice about sleep hygiene.
Both are the wrong answer, for different reasons.
Sleep hygiene, delivered on its own, performs poorly as a treatment for insomnia disorder. It is the intervention non-specialists reach for most, and the one with the least behind it.
Hypnotics work in the short term and were never designed for a problem lasting seven years.
And the treatment with the strongest evidence, which I will come to next, requires a trained therapist and a referral pathway that in most countries does not exist for this population. So it does not get offered, and its absence gets read as evidence that nothing helps.
Underneath sits the same pattern as everywhere else in this series. A symptom reported overwhelmingly by women in midlife, taken less seriously than the equivalent complaint in almost any other group, and researched late.
What actually works
The MsFLASH network ran a set of randomised trials in peri- and postmenopausal women and then pooled the individual participant data, which let them line up seven different interventions against each other in over five hundred women with insomnia symptoms and bothersome hot flushes.[3]
Escitalopram. Yoga. Aerobic exercise. Omega-3. Low-dose oral estradiol. Venlafaxine. And cognitive behavioural therapy for insomnia.
CBT-I won, and it was not close.[3]
The trial underneath that result is worth knowing about in its own right, because of how it was delivered. Six sessions. By telephone. Women with vasomotor symptoms and insomnia, randomised against menopause education, with substantial and durable reductions in insomnia severity in the CBT-I group.[4]
By telephone. Which means the barrier was never that this treatment is difficult to deliver.
Sorted By What The Evidence Says
For sleep specifically. Post one covered the same question for hot flushes, and the answers are not identical.
On progesterone specifically
Because the mechanism is so clean, this is the question I get asked most, so here is where the evidence sits.
A systematic review and meta-analysis of randomised trial data on micronized progesterone found a small improvement in sleep, most visible where sleep had been disturbed rather than in women sleeping normally.[5] Small. Real. Not the transformation the mechanism might lead you to expect.
I include the size honestly because the gap between a beautiful mechanism and a modest clinical effect is where most wellness content quietly cheats. Timing matters here too, and dose and formulation are a clinical conversation rather than a blog one.
What the missing hours do to everything else
Return to that list from the opening, because this is where the series joins up.
In the first post I wrote about what happens to glucose in the transition, and how post-meal control slips before fasting numbers move. Short and fragmented sleep pushes in the same direction, independently. So a woman in perimenopause is running a metabolic change and a sleep change at once, and each one worsens the other.
The same is true of mood. Insomnia in midlife women travels with depressive symptoms so reliably that the two get treated as one thing, and the direction of the arrow is usually assumed rather than established.[1] A woman who has not slept properly in two years and feels flat is not necessarily depressed. She may be exhausted in a way that is indistinguishable from it at the ten-minute mark.
And cognition, which is the subject of the next post. Working memory, word retrieval, the ability to hold a thread through a meeting: all of it degrades on broken sleep, in anyone, at any age. Some of what women in this window experience as brain fog and fear as something worse is the predictable output of a nervous system that has not consolidated properly in months.
Not all of it. Estrogen has its own direct effects on the brain, and I will get to those. But sleep is the variable most likely to be dismissed and most likely to be treatable, which makes it the wrong one to leave until last.
The practical part
Ask for CBT-I by name. Not sleep hygiene, not a sleep clinic referral in general terms. The specific thing, the acronym. It works by telephone and increasingly through digital programmes, which matters when no local service exists.
Protect the front half of the night. Deep sleep clusters early. Alcohol in the evening suppresses exactly that portion and then fragments the second half through withdrawal, which lands on top of a night that is already fragmenting. This is the change with the largest return for most women in this window.
Keep the wake window boring. When you wake at three, the instinct is to solve something. What consolidates the pattern is lying in bed awake and alert for long stretches, which teaches your brain that bed is a place for thinking. Getting up, staying dim, doing something dull, and returning when sleepy is the core behavioural move inside CBT-I.
Treat the room as thermal management. Vasomotor arousals depend on the gap between core temperature and the environment. A cooler room and layers you can shed without fully waking reduce the cost of each event. This will not stop the flushes, and the evidence review is clear that cooling does not treat them[6], but it changes what each one does to your night.
Name the breathing question. If you sleep long and wake unrefreshed, say that in those words and ask directly whether sleep-disordered breathing has been considered.
Bring the pattern, not the feeling. Two weeks of rough notes. What time you woke, whether you were hot, whether your heart was going, how long you were up. That converts an unwinnable ten-minute appointment into a clinical picture.
The hours nobody sees
The thing about this symptom is that it happens where there are no witnesses.
A hot flush in a meeting is public. Everyone in the room watches a woman go scarlet and reach for her collar, and she knows they are watching, which is its own separate cost. But the three in the morning is unwitnessed. Nobody is awake for it. Nobody counts it. There is no equivalent of a jacket on the floor for the fourth hour of lying in the dark.
She turns up at nine having lost most of a night, and the only visible evidence is that she is slightly slower than usual, which everybody, including her, will read as a character problem.
This is why I put a number near the top of this post. One in four women in the transition meets the criteria for a diagnosable sleep disorder. Not tiredness. A disorder, with treatment that works, that most of them will never be offered.
You are not failing at sleep. Sleep is being taken from you by a mechanism with a name, and the name is worth having, because you cannot ask for help with a thing you cannot describe.
The perimenopause series
- Perimenopause hormones explained. Why they swing rather than fade, and why one blood test cannot see it.
- Perimenopause and sleep. You are here.
- Perimenopause, brain fog and ADHD: when the strategies that always worked stop working. Coming next.
For the women reading
If you are awake at three and have been for months, the most useful thing you can do is stop treating it as the background noise of something else. It has its own mechanisms, its own evidence base, and its own treatment, and that treatment is not the sleeping tablet you are most likely to be offered.
Two weeks of notes will do more for you in an appointment than any amount of explaining how tired you are. Times, temperature, heart rate, how long you were awake. And the sentence worth saying out loud: I would like to be referred for CBT for insomnia.
None of this is medical advice, and none of it knows your history. It is here so that the hours nobody sees stop being invisible to you too.
References
- Baker, F. C., de Zambotti, M., Colrain, I. M., & Bei, B. (2018). Sleep problems during the menopausal transition: Prevalence, impact, and management challenges. Nature and Science of Sleep, 10, 73–95. https://doi.org/10.2147/NSS.S125807
- Coborn, J., de Wit, A., Crawford, S., Nathan, M., Rahman, S., Finkelstein, L., Wiley, A., & Joffe, H. (2022). Disruption of sleep continuity during the perimenopause: Associations with female reproductive hormone profiles. The Journal of Clinical Endocrinology & Metabolism, 107(10), e4144–e4153. https://doi.org/10.1210/clinem/dgac447
- Guthrie, K. A., Larson, J. C., Ensrud, K. E., Anderson, G. L., Carpenter, J. S., Freeman, E. W., Joffe, H., LaCroix, A. Z., Manson, J. E., Morin, C. M., Newton, K. M., Otte, J., Reed, S. D., & McCurry, S. M. (2018). Effects of pharmacologic and nonpharmacologic interventions on insomnia symptoms and self-reported sleep quality in women with hot flashes: A pooled analysis of individual participant data from four MsFLASH trials. Sleep, 41(1), zsx190. https://doi.org/10.1093/sleep/zsx190
- McCurry, S. M., Guthrie, K. A., Morin, C. M., Woods, N. F., Landis, C. A., Ensrud, K. E., Larson, J. C., Joffe, H., Cohen, L. S., Hunt, J. R., Newton, K. M., Otte, J. L., Reed, S. D., Sternfeld, B., Tinker, L. F., & LaCroix, A. Z. (2016). Telephone-based cognitive behavioral therapy for insomnia in perimenopausal and postmenopausal women with vasomotor symptoms: A MsFLASH randomized clinical trial. JAMA Internal Medicine, 176(7), 913–920. https://doi.org/10.1001/jamainternmed.2016.1795
- Nolan, B. J., Liang, B., & Cheung, A. S. (2021). Efficacy of micronized progesterone for sleep: A systematic review and meta-analysis of randomized controlled trial data. The Journal of Clinical Endocrinology & Metabolism, 106(4), 942–951. https://doi.org/10.1210/clinem/dgaa873
- The North American Menopause Society. (2023). The 2023 nonhormone therapy position statement of The North American Menopause Society. Menopause, 30(6), 573–590. https://doi.org/10.1097/GME.0000000000002200