Falling asleep isn't the problem. It's staying there. You're out by 10:30, and then somewhere between 2 and 3 AM your eyes open, your mind is suddenly, unhelpfully alert, and an hour later you're still lying there doing math on how many hours of sleep you have left. By morning, "just stress" is the explanation that gets offered, usually by someone who hasn't looked at what's actually changing in your hormones this decade.
Stress can absolutely fragment sleep. But if this is a new pattern that showed up sometime in your 40s, there's a more specific, more measurable explanation, and it has a name: the loss of one of your brain's own natural sedatives.
The Sedative You're Already Making, Until You're Not
Progesterone doesn't just regulate your cycle. Once it reaches your brain, it gets converted into a metabolite called allopregnanolone, and allopregnanolone acts as a positive allosteric modulator at the GABA-A receptor, the same receptor benzodiazepines like lorazepam target. In plain terms: it turns up the volume on your brain's primary calming signal. This isn't a theory. A randomized, double-blind, crossover trial gave postmenopausal women oral micronized progesterone for 21 days and recorded sleep EEG before and after. The result was a measurable decrease in time spent awake during the night, along with an increase in REM sleep in the first third of the night (Schüssler P et al., Psychoneuroendocrinology, 2008 — PMID: 18676087).
During your reproductive years, progesterone (and the allopregnanolone made from it) rises through the second half of your cycle, which is part of why the luteal phase can feel more sedating. In perimenopause, progesterone is usually the first hormone to decline, often years before estrogen drops meaningfully. Less progesterone means less allopregnanolone, which means less of the GABA-A support that was quietly holding your sleep together through the night. You're not imagining a change. Something you were making every night for decades is measurably gone.
Why 2 to 3 AM Specifically
Sleep isn't a flat line of the same depth all night. It cycles, and non-REM sleep, particularly deep slow-wave sleep, is concentrated earlier in the night, tapering off as the night goes on. Whatever is holding your sleep together through the thinner, more fragile stretches later in the night is going to matter most exactly when your natural sleep architecture is offering you the least protection. That's the early morning hours. If your GABA-A support has quietly declined, that's where the crack shows up first.
The Second, Overlapping Driver: Estradiol Instability, Not Just Decline
Progesterone isn't the only hormone in play, but the common assumption about the other one is backwards. Perimenopausal estradiol doesn't decline in a steady line the way progesterone does. Daily hormone sampling in perimenopausal women found marked swings, including genuine episodes of estradiol running higher than in women of reproductive age, right alongside the lows (Santoro N et al., Journal of Clinical Endocrinology & Metabolism, 1996 — PMID: 8636357). The mechanism traces back to inhibin B: as the number of remaining ovarian follicles shrinks, inhibin B production falls, which removes a brake on FSH. Rising FSH then overstimulates whatever follicles are left, and estradiol can spike higher than it did in your 30s well before it ever settles into a consistently low postmenopausal range (Prior JC, Endocrine, 2005 — PMID: 16034185). This pattern is sometimes described as endogenous ovarian hyperstimulation, and it's a genuinely different physiological state from the steady, low estradiol of true menopause.
This distinction has real clinical stakes, not just accuracy. Hot flashes and night sweats do track with estradiol instability, but instability isn't the same problem as deficiency, and treating it as if it were can backfire. Adding exogenous estradiol into a system that's already prone to unpredictable highs doesn't just correct the low points, it can push the high points further up, which shows up as breast tenderness, heavy or irregular bleeding, bloating, and mood swings, sometimes called estrogen dominance. That's a meaningfully different risk calculation than giving estradiol to a woman whose ovaries have already gone quiet.
The two mechanisms in this post aren't competing explanations, they're often happening in the same woman at the same time: less GABA-A support from declining progesterone making sleep more fragile overall, and estradiol instability making an actual thermal awakening more likely to get triggered on a given night. That's part of why "just manage your stress better" so often misses the mark. It's treating a hormonal architecture problem as a psychological one.
A Familiar Pattern
A patient in her late 40s came in describing herself as a lifelong good sleeper who, over about a year, had started waking reliably between 2 and 3 AM, wide awake, unable to fall back asleep for an hour or more. She'd already tried magnesium, cutting off caffeine by noon, and a stricter bedtime, all reasonable moves that hadn't touched the pattern. Her cycles had become irregular over the same stretch, and lab work showed progesterone at the low end for a woman still cycling, with estradiol still relatively preserved. Addressing the progesterone specifically, alongside her existing sleep habits rather than replacing them, brought the 2 AM waking down from nearly every night to occasional within about six weeks.
What Actually Gets Checked
This is worth a real look if the waking is new within the past one to three years, if better sleep hygiene alone hasn't moved the needle, if it's mostly absent in the two weeks after ovulation but present the rest of the cycle, or if it's become nightly as your cycles have gotten irregular. Progesterone rises after ovulation and stays elevated through most of the luteal phase, the one real high point for GABA-A support all cycle. It drops again as the corpus luteum regresses in the days before your period, then stays low through the entire follicular phase until the next ovulation restarts the rise. The vulnerable window, in other words, is wider than just the premenstrual days, it's that late-luteal drop plus the follicular phase that follows.
A useful evaluation looks at progesterone and estradiol relative to where you are in your cycle (or the absence of one), along with TSH and free T3, since thyroid dysfunction can independently disrupt sleep continuity and deserves to be ruled out rather than assumed away. A single estradiol draw is a snapshot, not the full picture, given how much it can move day to day at this stage, so it's read alongside your symptom pattern and cycle history rather than as a stand-alone number. Night sweats or a subjective sense of temperature swings at night point more toward the estradiol-instability side of this; waking with a racing or overly alert mind but no temperature complaint points more toward the GABA-A side. Progesterone is the better-studied first approach for hot flashes and night sweats in this population, and some patients also report additional benefit from testosterone optimization, though that's a clinical observation rather than something backed by controlled trial evidence.
What a Realistic Timeline Looks Like
Oral micronized progesterone taken at night is the most studied approach for this specific mechanism, though individual response and appropriate use depend on your full hormonal picture, not any single lab value in isolation. The trial evidence for progesterone improving vasomotor symptoms specifically is stronger after menopause than during the perimenopausal transition, where a large randomized trial didn't reach statistical significance on its primary hot-flash outcome, even though participants did report a meaningful improvement in night sweats and symptom intensity. The sleep-continuity effect described earlier in this post is a separate, more directly demonstrated mechanism. Most women who respond notice a change in sleep continuity within two to four weeks, with the fuller picture usually clear by six to eight weeks. This isn't a one-time fix and forget: cycles and hormone levels shift as perimenopause progresses, so periodic reassessment matters more here than in a lot of hormone therapy, since what worked at 44 may need adjusting at 47.
If "it's just stress" hasn't matched what 2 AM actually feels like, there's a specific, measurable mechanism behind it, and it's worth a real look rather than another round of sleep hygiene tips that were never going to touch it.
Has your sleep changed in a way that started around the same time your cycle did?
A free consultation is a good place to start that conversation, no commitment required, just a chance to look at the full picture together. To request one, email drwilcox@precisionhormoneconsulting.com.
— Darrell Wilcox, MD | @wellnessdoc_4everyoung on Instagram
References
Schüssler P, Kluge M, Yassouridis A, Dresler M, Held K, Zihl J, Steiger A. Progesterone reduces wakefulness in sleep EEG and has no effect on cognition in healthy postmenopausal women. Psychoneuroendocrinology. 2008;33(8):1124-1131 — PMID: 18676087.
Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. Journal of Clinical Endocrinology & Metabolism. 1996;81(4):1495-1501 — PMID: 8636357.
Prior JC. Ovarian aging and the perimenopausal transition: the paradox of endogenous ovarian hyperstimulation. Endocrine. 2005;26(3):297-300 — PMID: 16034185.
Hitchcock CL, Prior JC. Oral micronized progesterone for vasomotor symptoms: a placebo-controlled randomized trial in healthy postmenopausal women. Menopause. 2012;19(8):886-893 — PMID: 22453200.
Prior JC, Cameron A, Fung M, Hitchcock CL, Janssen P, Lee T, Singer J. Oral micronized progesterone for perimenopausal night sweats and hot flushes: a Phase III Canada-wide randomized placebo-controlled 4 month trial. Scientific Reports. 2023;13(1):9082 — PMID: 37277418.
This content is for educational purposes only and does not constitute medical advice. Sleep disruption has multiple possible causes, including thyroid dysfunction, and an accurate evaluation requires individualized lab work and clinical history. Hormone therapy, including progesterone, requires physician supervision and periodic monitoring, and is not appropriate for everyone. Individual results vary, and no outcome is guaranteed. Dr. Wilcox is licensed to practice in Texas and Arizona.
