Dr. Herman Weiss
Oct 14, 2026
Welcome back to The Pause Collective Science Review - the twice-a-week edition where I read the new menopause and PMOS research so you don't have to. This cycle's lead paper is one of the biggest menopause studies I have ever seen, and it comes with a built-in scientific argument I want you to hear both sides of. Three more papers round out a practical theme: what actually changes, and what actually helps.
1. Menopause is a cliff, not a slope - 300 million lab tests redraw the transition
A team led by Uri Alon at the Weizmann Institute (with Clalit Health Services) asked a deceptively hard question: when exactly does the body change at menopause? Longitudinal cohorts like SWAN say "gradually, over years" - but those studies are small. The problem with big datasets is that nobody records the date of a woman's final menstrual period, so every woman's timeline blurs into everyone else's. Their fix is a computational method borrowed from astronomy that un-blurs the data, anchoring every lab result to each woman's own final period. Applied to two independent cohorts - the US NHANES survey and Israel's Clalit - totaling more than 300 million lab tests from over a million women between 1999 and 2024.
The finding: at the final menstrual period, the body does not glide - it steps. Nearly every physiological system they measured showed an abrupt, jump-like shift: sex hormones (no surprise), but also bone metabolism, liver function, kidney function, lipids, and blood counts. Men show no such jumps at the same ages, so this is menopause, not generic aging. The two cohorts - American and Israeli - agreed with each other. Two more results matter to us. First, the dysregulation starts early: the drift begins more than a decade before the final period. Second, women on menopausal hormone therapy showed markedly smaller jumps - the bigger the jump, the larger the correction with MHT. And the clinical echoes: osteoporosis prevalence steepens right at the final period, and cardiovascular events accelerate after it.
What this means for you: if you have ever felt that your body changed "all at once" around menopause and been told it's just gradual aging - this study says your perception has data behind it. It also reframes prevention: if the drift starts ten-plus years out, then perimenopause is not "too early" to act - it is exactly when the acting happens. For my PMOS readers: your baseline hormonal pattern is already atypical, and we know from last issue's genetics update that the same gene regions help set both PMOS biology and menopause timing. How this cliff looks in PMOS is an open question I would love to see this team answer next. And a note of local pride: half of this dataset is Israel's own Clalit.
The limits, and the argument: the final menstrual period here is inferred by a model, not recorded; this is a clever reconstruction of cross-sectional data, not a follow-women-for-decades cohort. Lab values are not symptoms or diagnoses. And the MHT comparison is observational - women who take hormone therapy differ from women who don't. The debate is already live: Martha Hickey, head of menopause services at the Royal Women's Hospital in Melbourne, told ABC the "cliff" framing is "unnecessarily alarming" and not a medical term. My read: both things are true. The underlying drift is real and starts early (the gradualists' point), AND there is a genuine step at the final period that age-based averages have been hiding (the authors' point). Neither is a reason to panic; both are reasons to plan.
(Pridham G, Hayut Y, Lavi-Shoseyov N, Neeman M, Hovav N, Toledano Y, et al., Alon U. Population-scale reconstruction of menopause dynamics reveals a physiological cliff at the final menstrual period. Nat Med, 2026. doi:10.1038/s41591-026-04614-0)
2. Tai Chi for the menopause brain: a small trial with an unusually deep look
A randomized trial from Fuzhou, China took 40 postmenopausal women with mild cognitive impairment (MCI) and assigned them to 12 weeks of supervised 24-form Tai Chi or a health-education control group. What makes this exploratory trial worth your attention is everything they measured around the exercise: not just cognitive scores, but brain blood flow, vessel-function markers, and circulating Alzheimer's-related proteins.
The Tai Chi group's MoCA scores improved about 5 points more than the control group's (a clinically meaningful gap), with the biggest gains in delayed recall, visuospatial/executive function, and attention. Cerebral blood flow rose globally and in cognition-related regions. The bloodwork moved in the right direction too: nitric oxide (a vessel-relaxing signal) up, the vasoconstrictor-to-NO ratio down, and circulating T-tau, P-tau, and amyloid-beta 1-42 - proteins linked to Alzheimer's pathology - all significantly lower than control. No intervention-related adverse events.
What this means for you: I am not going to tell you Tai Chi prevents dementia - a 12-week exploratory trial cannot say that. But if you are a postmenopausal woman noticing your memory is not what it was, this is one more honest data point that structured, regular movement is the best-supported non-drug lever we have for the menopause brain, and Tai Chi is a low-impact, joint-friendly, genuinely pleasant way to pull that lever. The perfusion finding is the interesting part: it suggests the benefit is partly vascular - better blood flow, not just practice effects.
The limits: 40 women, one center, open-label (participants knew their group), 12 weeks, explicitly exploratory. Blood biomarkers of Alzheimer's are proxies, not outcomes; whether any of this touches actual dementia rates is unknown. Consider this a strong rationale for the bigger trial, and a reasonable nudge to start moving now.
(Zheng Y, Hong L, Wu L, Chen X, Hu Q, Chen L, Ni X, Zhuang X, Yin L. Effects of 12-week 24-form Tai Chi on cognitive function, cerebral perfusion, and vascular function in postmenopausal women with mild cognitive impairment: an exploratory randomized controlled trial. Front Aging Neurosci, 2026. doi:10.3389/fnagi.2026.1880915)
3. PMOS bone density looks better - but the strength does not follow
Here is a study that overturns an assumption. We have long thought PMOS might be somewhat protective for bones (higher androgens, higher average weight - both load the skeleton). A cross-sectional study used high-resolution peripheral quantitative CT - a scanner that sees bone microarchitecture, not just density - in 23 obese women with PMOS, 21 BMI-matched obese women without PMOS, and 24 normal-weight controls. Matching by BMI is the key design choice: it separates PMOS itself from the weight it often travels with.
The PMOS group did show higher volumetric bone density at both the radius and tibia, including denser cortical bone - so the assumption survives the density test. But here is the twist: estimated bone strength (stiffness and failure load, computed from the scans) was NO better than the BMI-matched controls. Density up, strength flat. And testosterone - the obvious suspect for explaining the difference - mediated nothing.
What this means for you: if you have PMOS, do not let a reassuring bone-density number talk you out of bone care. Density is quantity; strength is quality, and this study says the two can come apart. For my menopause readers, this is the mirror-image lesson of the week: the cliff paper shows bone metabolism jumping at the final period, and this one warns that a good baseline number is not armor. The boring levers - resistance training, protein, vitamin D adequacy, not smoking - work on the strength side of the equation. Women with PMOS heading into perimenopause: you need them as much as anyone.
The limits: small groups, cross-sectional (one snapshot, no fractures tracked), obese populations only at the PMOS end, and estimated strength from finite-element modeling rather than real-world fracture outcomes. The answer on actual fracture risk in PMOS needs a long longitudinal study we do not have yet.
(Zhao Y, Zhang Q, Gong F, Duan L, Zhang Y, Wang L, Zhu H, Xia W. Bone microarchitecture and strength in obese women with and without polycystic ovary syndrome (PCOS): a peripheral quantitative computed tomography (pQCT) study. Front Endocrinol (Lausanne), 2026. doi:10.3389/fendo.2026.1935187)
4. AMH tells you about the ovaries, not the metabolism - 1,147 women settle it
Anti-Mullerian hormone (AMH) is everywhere in PMOS care - it is part of how the condition is diagnosed and how ovarian reserve is gauged. A natural clinical question: since PMOS is so entangled with weight and metabolic dysfunction, does a woman's AMH also say something about her adiposity or metabolic risk? A Polish team measured AMH alongside detailed body composition (including visceral fat area by bioimpedance) and metabolic panels in 1,147 women newly diagnosed with PMOS.
The answer is clean: no. AMH's correlations with every adiposity marker - waist-to-hip ratio, visceral fat area, lipid accumulation product, cardiometabolic index - were negligible (correlation coefficients around 0.05-0.1, statistical noise dressed as significance thanks to the huge sample). Meanwhile the metabolic markers themselves were tightly interlocked exactly as expected (visceral-fat indices tracked fasting insulin and insulin resistance at r around 0.67-0.76). When they ran a principal-component analysis, the metabolic dimensions explained 68.8% of the variance - and AMH correlated with none of them.
What this means for you: two women with PMOS can have identical AMH levels and completely different metabolic risk profiles. AMH is a reproductive signal, not a metabolic one - it should be read against your age and reproductive picture, and it cannot substitute for actually measuring insulin, lipids, and waist circumference. If your AMH is "good," that says nothing about your metabolic risk; if it is high (as PMOS often makes it), that says nothing protective either. For my menopause readers: AMH is also the number marketed to you as a "menopause countdown clock" - same lesson, different stage. It reflects the ovaries' bookkeeping, and it was never a whole-body health score.
The limits: cross-sectional, bioimpedance rather than imaging for visceral fat, and a single-ethnicity cohort. The correlations were so weak, though, that the core message - AMH is not a metabolic marker - is unlikely to flip.
(Kowalczyk K, Galczynska D, Herman J, Czyz M, Kowalczyk D, Drosdzol-Cop A, Franik G. Anti-Mullerian hormone in relation to adiposity and metabolic indices in polyendocrine metabolic ovarian syndrome. Front Endocrinol (Lausanne), 2026. doi:10.3389/fendo.2026.1975667)
What is trending
The Nature Medicine "cliff" paper is the menopause story of the week, and the argument about it is healthy. Coverage ranges from breathless ("Bones, Liver, And Blood Fats All Shift Abruptly at Menopause" - ScienceAlert) to openly skeptical (ABC's piece quoting Australian specialists who call the framing alarmist). My annotation above is the middle path. If your feed shows you the scary version, remember: the same dataset shows hormone therapy markedly softens the jumps - the cliff is real, and it is also modifiable.
The Menopause Society's Annual Meeting opens October 20 in San Diego, and the pre-meeting press releases keep coming. The one getting traction this week: a Mayo Clinic retrospective cohort (about 2,400 women on semaglutide, 1,500 on tirzepatide) finding women on hormone therapy lost significantly more weight on semaglutide - with no significant interaction for tirzepatide. The Society's own caveats are correct: observational, retrospective, needs prospective confirmation, and weight loss is not an indication for MHT. Also queued for the meeting: a study linking combined estrogen-plus-testosterone therapy to lumbar spine bone density gains, and AbCellera's first phase 1/2 data on ABCL635, a non-hormonal drug candidate for hot flashes. Expect a full meeting-news cycle over the next two weeks; I will sort signal from noise as it lands.
On the PMOS side, the policy conversation is catching up with the science. Australia's PMOS association (POSAA) is publicly pressing for GLP-1 weight-loss drugs to be subsidized for PMOS on the national Pharmaceutical Benefits Scheme - currently they are prescribed off-label and cost patients hundreds of dollars a month (The Guardian, Oct 9). The Royal Australasian College of Physicians has not taken a position. Watch this space: if PMOS becomes a reimbursed indication anywhere, it will set the template everywhere.
A transparency note, as always: the above is grounded in news coverage and press releases I could read directly. I did not directly browse TikTok or Reddit threads this cycle, so I am not claiming to represent what those feeds showed today.
One cross-reference: Issue 9, published Sunday, covered the new PMOS gene map - 120 regions, several of which also help set menopause timing. It pairs naturally with this week's cliff paper.
Until the next one,
Dr. Herman Weiss
Medical Disclaimer
The information provided in this blog post and newsletter is for educational and informational purposes only. It does not constitute medical advice or professional services and should not be used to diagnose or treat any health problem or disease. Always seek the advice of your physician or other qualified health-care provider regarding a medical condition. Never disregard professional medical advice or delay seeking it because of something you have read here. Use of this content does not create a doctor-patient relationship. Individual responses to treatments and lifestyle changes can vary, and only your healthcare provider can evaluate your specific circumstances. If you are experiencing a medical emergency, call your local emergency services immediately.