Perimenopause Diet: What to Eat, What to Limit, and What the Evidence Says

Quick answer: No health body publishes a list of foods to avoid in perimenopause, and the two that address the question disagree with each other. The NHS suggests avoiding or reducing potential triggers. The Menopause Society reviewed that same advice and found that nobody has ever tested whether avoiding triggers helps. Alcohol has a firmer reason to cut down, and it has nothing to do with hot flashes. Where food genuinely changes something measurable in these years is bone, cholesterol and muscle.

Most pages on this subject pick one of those halves and run with it. This one holds both, because the evidence does.

Foods and drinks you are told to avoid

There is no authoritative list. The advice most women meet first — cut the caffeine, the alcohol, the spicy food, the sugar — traces back to one of two sources, and those two sources do not agree about whether it works.

Here is the disagreement in full, because it is the whole answer to this question:

• The NHS includes, among the things you can try for hot flushes and night sweats, "to avoid or reduce potential triggers, such as spicy food, caffeine, hot drinks, smoking and alcohol" (NHS, Things you can do, page last reviewed 19 May 2026).
• The Menopause Society reviewed that same advice and concluded: "There are no clinical trials assessing the effects of avoiding triggers for the alleviation of VMS." It grades trigger avoidance Level II and places it on its not-recommended list (The Menopause Society, 2023).

Both statements are accurate, and they are not really answering the same question. The NHS is listing things that are free, safe and worth a try. The Menopause Society is asking a narrower question — has this been tested as a treatment? — and reporting that it has not. Neither body claims trigger avoidance is proven. One suggests trying it anyway; the other declines to recommend what nobody has measured.

What little observational evidence exists is thin and inconsistent. The position statement cites a single cross-sectional study of 4,595 Chinese women that found a positive association between alcohol intake and vasomotor symptoms, then notes the finding was not reproduced elsewhere, including in the Melbourne Women's Midlife Health Project. SWAN, which followed a multi-ethnic cohort through five visits rather than surveying it once, found no significant association between vasomotor symptoms and alcohol or caffeine consumption at all (Gold et al., 2006).

So the practical resolution is not a list. It is a diary. If a glass of wine reliably precedes a bad night for you, that observation is yours, it is specific to you, and acting on it costs nothing. The CBT self-help guide used in the MENOS trials takes exactly this approach, telling readers that "studies of women with menopausal symptoms suggest that about 50 per cent of flushes may be triggered by particular events" while warning that these triggers "vary widely between women; in fact, some women can't identify any triggers at all" (Myra Hunter and Melanie Smith, Managing Hot Flushes and Night Sweats, 2nd edition, Routledge, 2021). A personal trigger is a real thing. A printed trigger list handed to every woman is not the same thing, and has never been shown to work as one.

Alcohol: a firmer reason to cut down, and it is not hot flashes

Alcohol does carry a separate, better-measured health argument in midlife, and it belongs in its own paragraph rather than smuggled into a symptom claim. The National Cancer Institute lists breast cancer among the cancers linked to alcohol and gives the risk as 1.04 times as likely in light drinkers, 1.23 times in moderate drinkers and 1.6 times in heavy drinkers (National Cancer Institute, reviewed 2 May 2025).

Those are relative risks, and the NCI says so itself, adding that for a common cancer such as breast cancer "even a small relative risk can translate into a large absolute risk." The same fact sheet gives the absolute version, drawn from the Surgeon General's Advisory and covering alcohol-related cancers as a group rather than breast cancer alone: of 100 women who have less than one drink a week, about 17 will develop an alcohol-related cancer; of 100 women who have one drink a day, 19 will; of 100 who have two drinks a day, about 22 will. An increase of 2 per 100 at one drink a day, and 5 per 100 at two.

For the terms: the Dietary Guidelines for Americans, 2020–2025 define moderation for women as one drink or less a day, and the NIAAA defines heavy drinking for women as four or more drinks on any day or eight or more in a week.

That is a real reason some women drink less in midlife. It is not a hot flash argument, and presenting it as one does the reader a disservice in both directions.

Caffeine: a small association, in a different group of women

Caffeine has one study behind the standard advice, and reading it closely matters. In a cross-sectional survey of 1,806 women attending a menopause clinic at Mayo Clinic, caffeine users scored slightly higher for vasomotor symptom bother than non-users, 2.30 against 2.15 on the clinic's scale (P = 0.011), a difference that held after adjustment for menopause status and smoking (Faubion et al., 2015).

Two limits sit on that finding. The authors' own conclusion is specific: "Caffeine use is associated with greater vasomotor symptom bother in postmenopausal women." The abstract reports no equivalent vasomotor finding for women still in the transition, which is the group reading this page. And the study asked women once, at a clinic visit, which cannot separate caffeine causing symptoms from symptoms changing how much coffee someone drinks. Set against it, SWAN's longitudinal analysis found no significant association with caffeine over five visits (Gold et al., 2006).

If coffee reliably sets off a flush for you, cut it. There is no good evidence that cutting it will do anything for a woman who has not noticed that pattern.

What diet trials have actually tested

Very little, and that is the honest summary. The Menopause Society grades dietary modification Level III for vasomotor symptoms, its lowest grade, consensus and expert opinion, and places it on the not-recommended list, writing that "there is limited evidence from clinical trials to support the use of dietary modification for improving VMS."

How the panel reached that grade is more informative than the grade itself. Two studies carry most of the weight, and they point in different directions.

The larger one is an analysis inside the Women's Health Initiative Dietary Modification trial, covering 17,473 postmenopausal women aged 50 to 79 who were not taking hormone therapy. Among the women who had symptoms at the outset, assignment to the low-fat, high-fruit-and-vegetable intervention was associated with a modestly higher chance of those symptoms disappearing altogether (odds ratio 1.14, 95% CI 1.01–1.28). Separately, and measured against women who simply maintained their weight, those who lost 10% or more of their body weight in the first year were more likely again to be rid of them (odds ratio 1.56, 95% CI 1.21–2.02). But the authors' sharpest finding is the one worth carrying away: "Large weight loss (>22 lb), but not dietary changes, was related to the elimination of moderate/severe VMS" (Kroenke et al., 2012). The weight came out as the active ingredient; the food changes, on their own, did not.

The smaller one is where the striking number on most diet pages comes from. In the Women's Study for the Alleviation of Vasomotor Symptoms, 84 postmenopausal women with at least two moderate-to-severe hot flashes a day were randomised to a low-fat vegan diet plus half a cup of cooked soybeans daily, or to no dietary change. Moderate-to-severe hot flashes fell by 88% in the diet group against 34% in the control group, and half the women who completed the intervention reported none at all by week 12 (Barnard et al., 2023).

That result deserves both its hearing and its caveats, and it rarely gets the second half:

The disclosure is not an accusation; it is printed in the paper. "The study was funded by the Physicians Committee for Responsible Medicine," and the first author "serves without compensation as president" of that organisation. In a trial where the outcome is a symptom the participant reports and nobody is blinded, a 34% drop in a group told to change nothing is a useful reminder of how much movement is available without any intervention at all. The panel that graded this literature worked, in its own words, "with the knowledge that trials of nonhormone treatments of VMS have a placebo improvement rate of 20% to 66%" (The Menopause Society, 2023).

Where that leaves things: one large trial says the weight, not the food. One small unblinded trial says the food, dramatically. A panel reading both graded the evidence Level III. If a page tells you some way of eating is proven to fix hot flashes, that 88% is almost always where it came from, minus everything in the table above.

For the options that do have Level I evidence for symptoms, see perimenopause treatment options, ranked by evidence. And under STRAW+10, perimenopause does not end at the final period, it ends 12 months later (Harlow et al., 2012), as explained in what perimenopause actually is.

What actually changes in your body, and why food matters here

Three things change that nutrition can plausibly influence, and each has been measured directly rather than inferred, all three in SWAN, the long-running Study of Women's Health Across the Nation.

The lipid finding is worth reading twice. SWAN tested whether 1,054 women's cardiovascular risk factors changed steadily with age or jumped around menopause. Most followed age; total cholesterol, LDL cholesterol and apolipoprotein B did not, and the authors concluded that monitoring lipids in perimenopausal women should improve prevention of coronary heart disease. If you have never had a lipid panel, this is the life stage where the number starts to matter, and our guide to normal cholesterol levels covers what to do with the result.

One correction to the version that circulates: the line that HDL falls at menopause is not what SWAN measured. In its HDL ancillary study of 471 women, HDL cholesterol increased across the transition while HDL subclasses, lipid content and cholesterol efflux capacity per particle changed for the worse (El Khoudary et al., 2021). Quality down, reported number up. We could not verify the plain "HDL falls" claim, so we are not making it.

Bone is the most time-sensitive of the three. Across ten years of measurements in 862 women, spine bone density fell 10.6%, but 7.38% of that went in the roughly three years from one year before the final period to two years after it, which the researchers named the transmenopause. Femoral neck loss had the same shape: 9.1% overall, 5.8% inside that window. And the body-composition finding reframes the usual advice, because total weight gain does not accelerate, only its composition changes. That is a case for defending muscle rather than eating less.

Protein, and why it matters more now

Protein becomes more relevant during this window for one specific reason: lean mass switches from slowly accumulating to slowly declining, and protein is the raw material for the tissue being lost.

Here is where this article stops short of what you may have read elsewhere. The general adult Recommended Dietary Allowance for protein is 46 g a day for women, derived from 0.8 g per kilogram of body weight (Food and Nutrition Board DRI summary tables). That is an adequacy floor for healthy adults, not an optimum for holding onto muscle while it is actively being lost.

Higher targets exist, but for a different population. The PROT-AGE study group, convened by the European Union Geriatric Medicine Society, recommends 1.0 to 1.2 g per kilogram per day for people over 65 to maintain and regain lean body mass and function, and at least 1.2 g/kg for those also exercising (Bauer et al., 2013). Women in their forties and early fifties are not that population, and we could not verify any protein target set specifically for the menopause transition, so this page does not give you one. The floor is checkable. The case for sitting above it while lean mass is being lost is reasonable. The number that would be right for you has not been established by anyone we can cite. The same applies to resistance training, which follows logically from the lean-mass data but was not tested for that outcome in it.

Calcium and vitamin D: the bone window

Because bone loss concentrates into roughly three years around the final period, calcium and vitamin D adequacy stop being generic good advice and become time-sensitive. The calcium requirement for women also steps up inside exactly this age band.

Dietary Reference Intakes from the Food and Nutrition Board, via MedlinePlus: calcium and MedlinePlus: vitamin D.

Three practical details that generic "get enough calcium" advice leaves out, all from the same MedlinePlus references:

• Absorption is best in amounts of no more than 500 mg at a time. A single 1,000 mg dose is not the same as 1,000 mg spread across a day.
• Calcium carbonate is absorbed better with food; calcium citrate is taken up well on a full or empty stomach. That matters most if you take acid-reducing medication.
• Food first is realistic here. A glass of milk or fortified soy milk, 45 g of cheese or 170 g of yoghurt runs roughly 300 mg each, and 85 g of sardines with bones 325 mg, which puts 1,200 mg in reach without a supplement.

MedlinePlus also notes that the Bone Health and Osteoporosis Foundation recommends more vitamin D for people aged 50 and over than the DRI does, 800 to 1,000 IU daily. That is a real disagreement between two credible bodies, and one to raise with your clinician rather than settle from a website. Supplement forms and quality get their own page on perimenopause vitamins and supplements.

Soy and phytoestrogens: the honest answer

Soy foods, soy extracts and the soy metabolite equol are all on the Menopause Society's not-recommended list for vasomotor symptoms at Level II, limited or inconsistent evidence. The reason is worth knowing, because it explains why soy trials keep disagreeing with each other.

The isoflavones are not the active ingredient people assume. Equol, a compound your gut bacteria may or may not make from the soy isoflavone daidzein, binds estrogen receptors with a strong preference for ER-β, and whether you produce it depends on your gut flora. In the cross-sectional study the position statement cites, 129 of 365 women (35%) were equol producers; among them, the highest quartile of dietary daidzein intake was associated with 76% lower odds of above-average hot flash frequency (odds ratio 0.24, 95% CI 0.07–0.83), while among non-producers there was no association at all (Newton et al., 2015). That is observational rather than a trial, and the trend across intake levels did not reach significance (P = 0.06).

Then the part that settles it practically: "tests to ascertain whether women are equol producers are not commercially available to the public or healthcare professionals." Even if the mechanism is real, neither you nor your doctor can find out whether it applies to you, and averaging a benefit only a third of women can experience across a whole trial population reliably produces mixed results.

The equol story also has a complication that the mechanism's supporters rarely mention. In the soybean trial described above, the one with the 88% figure, "neither seasonality nor equol production status was associated with the degree of improvement" (Barnard et al., 2023). If equol were the mechanism, it should have been the variable that mattered most.

None of which makes soy a food to avoid. It is an ordinary food with an ordinary place in a diet, and it has a separate, better-measured role in the lipid picture above: a meta-analysis of the 46 controlled trials the FDA used to review its soy heart-health claim found that soy protein lowers LDL and total cholesterol in adults (Blanco Mejia et al., 2019), an effect we cover with the gram amounts further down. Eat tofu because you like it, and because soy protein does something small and real to your LDL, not because it will stop the night sweats.

What do Japanese women do for menopause?

Less than the question implies, and the explanation usually attached to it does not survive the dataset that tested it directly.

The underlying observation is real and has been repeated. In SWAN's baseline community survey of 16,065 women aged 40 to 55, Japanese and Chinese women reported most symptoms least frequently, with odds ratios between 0.47 and 0.67 against white women (Gold et al., 2000). Following the same cohort for up to 17 years, Japanese women had the shortest median duration of frequent hot flashes and night sweats, 4.8 years against 7.4 years for the cohort overall (Avis et al., 2015). Our duration guide sets out how long hot flashes last in different groups of women. Two qualifications travel with that 4.8 and are usually dropped: in the adjusted models only the Chinese difference remained statistically significant, and SWAN recruited its Japanese participants in Los Angeles, so this is a comparison between ethnic groups living in the United States rather than between countries.

Soy is the usual explanation, and it has genuine support. A prospective cohort in Takayama, Gifu followed 1,106 women aged 35 to 54 who were all premenopausal when they enrolled. Over six years, 101 developed new moderate or severe hot flashes. Risk was lower in the highest third of soy intake than in the lowest (hazard ratio 0.47, 95% CI 0.28–0.79), and lower again by isoflavone intake (hazard ratio 0.42, 95% CI 0.25–0.72) (Nagata et al., 2001).

Then SWAN tested the same idea inside the cohort where the ethnic difference was observed, and it did not hold: "Baseline genistein intake was not related longitudinally to vasomotor symptoms and did not account for reduced symptom reporting among Asian women after adjustment for covariates" (Gold et al., 2006). The same analysis found no significant association with physical activity, alcohol or caffeine either. So the gap is real, and diet does not explain it.

Part of the gap may not be physiological at all. Japanese has several words for what English flattens into "hot flash", and which one a survey uses changes the answer. Asking 45-to-55-year-old women in Kyoto and Fukushima prefectures, prevalence ran from 3.0% for hotto furasshu to 17.1% for hoteri, reaching 22.1% only when every hot-flash term was counted together. Overall vasomotor prevalence was 25.7%, and it peaked at 52.6% among late perimenopausal women. Self-assigned kōnenki status, the Japanese term the author glosses as climacteric, predicted hot flashes better than menstrual staging did (Melby, 2005). The earliest careful work in this area reached a similar conclusion from a survey of 1,141 non-hysterectomised Japanese women aged 45 to 55: symptom experience was "always significantly related to self-defined menopausal status" (Lock et al., 1988).

As for what Japanese women actually use: hormone therapy is uncommon. Among 7,915 participants in the Japan Nurses' Health Study, 13.8% had ever used it, with a median duration of 2 years (Yasui et al., 2022). Kampo, the traditional Japanese herbal medicine its authors describe as "widely used in Japan, especially in the field of menopause medicine", has been tested properly once. In what its authors call the first placebo-controlled double-blind randomised trial of the formula kamishoyosan, 101 women received it and 104 a matching placebo; after 8 weeks hot flashes had fallen in both groups, with no significant difference between them (Takamatsu et al., 2020). Looking wider at herbal medicine, a Cochrane review pooled 22 randomised trials of Chinese herbal medicine in 2,902 women and found insufficient evidence that it was any more or less effective than placebo or hormone therapy (Zhu et al., 2016).

There is no transferable remedy in any of this, and no national secret. There is a measurement problem, a real but unexplained difference between groups, and a traditional treatment that failed its first proper test.

What a genuinely useful eating pattern looks like

There is no meal plan on this page, and the omission is deliberate: a day-by-day plan implies a precision the evidence does not have. What the evidence supports is a short list of priorities following from the three measured changes above.

• Protein at a level you actually hit, because lean mass is declining rather than accumulating. Whether you sit above the 46 g floor is a conversation with a clinician or dietitian, not a number this page can hand you.
• Calcium to the age-appropriate target, food first, with vitamin D alongside it, because that 7.38% of spine density goes in about three years and the interval does not come back.
• An overall pattern aimed at LDL cholesterol, the number SWAN found jumping at the final period rather than drifting with age. The specifics are the same for anyone lowering LDL, and are covered in our cholesterol diet guide.

Notice what is not on that list: no named diet claimed as proven for perimenopause, no superfood, no supplement stack, no elimination list. We are not aware of any eating pattern with verified evidence for treating this transition's symptoms, and saying so beats filling the gap with something plausible. Symptoms are a separate question, covered in what's actually linked to perimenopause.

When to raise it with a clinician rather than adjust your diet. Heavy bleeding that soaks through protection hourly, bleeding between periods or after sex, periods closer together than 21 days, or any bleeding after 12 months without one needs assessment rather than a trigger diary. The same goes for unexplained weight loss, a lump in the breast, or symptoms that look like perimenopause but could be thyroid disease or anaemia. Diet is not the first move for any of those.

What weight loss does and doesn't mean here

Weight loss is one of the few lifestyle measures the 2023 statement affirmatively recommends for vasomotor symptoms, at Levels II-III, and its reasoning is specific to this stage: adiposity acts as a risk factor for symptoms earlier in the transition, in perimenopause and early postmenopause, but not later, so weight loss may do more when women are earlier in it. If you are reading this while still having periods, that is the window the evidence describes. The panel is also candid about the quality of what it is drawing on, noting that these studies "are either small pilot studies, nonrandomized trials, or post hoc analyses of studies designed for a different purpose."

The complication is that the scale is a poor instrument here. SWAN's data shows fat rising and lean mass falling on an unchanged total-weight trajectory, so an approach that sheds muscle alongside fat works against the bone and metabolic picture this article is about. That argument gets its own page: why weight changes in perimenopause.

Frequently asked questions about eating in perimenopause (FAQ)

What foods should I avoid during perimenopause? No health authority publishes a list, and the two bodies that address the question disagree. The NHS includes avoiding or reducing potential triggers, "such as spicy food, caffeine, hot drinks, smoking and alcohol", among the things to try for hot flushes and night sweats. The Menopause Society reviewed the same advice and found "no clinical trials assessing the effects of avoiding triggers for the alleviation of VMS", grading it Level II and not recommended. Noticing your own patterns costs nothing; treating a printed trigger list as reliable treatment is not supported.

Can changing my diet stop hot flashes? Not reliably, on current evidence. Dietary modification sits at Level III, consensus and expert opinion, on The Menopause Society's not-recommended list for vasomotor symptoms. In the Women's Health Initiative, large weight loss rather than the dietary change itself was what related to eliminating moderate or severe symptoms. That verdict covers hot flashes and night sweats only, and says nothing about bone, cholesterol and muscle.

Does soy help with hot flashes? Not dependably: soy foods, soy extracts and the soy metabolite equol are not recommended for vasomotor symptoms at Level II. Only about 35% of North American women carry the gut bacteria that convert the soy isoflavone daidzein into equol, and tests for equol-producer status are not commercially available to the public or to healthcare professionals. In the one trial that measured it, equol production status was not associated with how much women improved.

What do Japanese women do for menopause? Lower symptom reporting among Japanese women is a real and repeated finding, but the soy explanation fails inside the dataset that tested it: in SWAN, baseline genistein intake "was not related longitudinally to vasomotor symptoms and did not account for reduced symptom reporting among Asian women after adjustment for covariates". Measured rates in Japan also depend on which Japanese word a survey uses, ranging from 3.0% to 17.1% by individual term. Hormone therapy is uncommon, at 13.8% lifetime use among Japanese nurses, and the one placebo-controlled trial of the Kampo formula kamishoyosan found no significant difference from placebo.

What is the best diet for perimenopause? No eating pattern has been shown to reliably treat hot flashes and night sweats; The Menopause Society grades dietary modification Level III and does not recommend it for those symptoms. Nutrition still matters here for three measured changes: total and LDL cholesterol rise around the final period, spine bone density falls about 7.4% in the three years surrounding it, and lean mass shifts from gaining to losing.

Should I eat more protein in perimenopause? Protein adequacy matters more once lean mass starts declining, which SWAN measured happening during the transition. The general adult RDA is 46 g a day for women, based on 0.8 g per kilogram of body weight. We could not verify a protein target set specifically for perimenopausal women, so this article publishes none; the 1.0 to 1.2 g/kg figure that circulates was written for adults over 65.

Medical disclaimer: This article is for informational purposes only. Always consult a qualified healthcare provider before making changes to your health routine.

Sources

1. The North American Menopause Society. The 2023 nonhormone therapy position statement of The North American Menopause Society. Menopause 2023;30(6):573-590 — Clinical position statement (PDF). https://menopause.org/wp-content/uploads/professional/2023-nonhormone-therapy-position-statement.pdf
2. National Health Service. Things you can do to help menopause and perimenopause symptoms — National health service guidance, page last reviewed 19 May 2026. https://www.nhs.uk/conditions/menopause-and-perimenopause/things-you-can-do/
3. National Cancer Institute. Alcohol and Cancer Risk — Federal cancer institute fact sheet, reviewed 2 May 2025. https://www.cancer.gov/about-cancer/causes-prevention/risk/alcohol/alcohol-fact-sheet
4. Faubion SS, Sood R, Thielen JM, Shuster LT. Caffeine and menopausal symptoms: what is the association? Menopause 2015;22(2):155-8 — Peer-reviewed cross-sectional survey. https://pubmed.ncbi.nlm.nih.gov/25051286/
5. Kroenke CH, Caan BJ, Stefanick ML, et al. Effects of a dietary intervention and weight change on vasomotor symptoms in the Women's Health Initiative. Menopause 2012;19(9):980-8 — Peer-reviewed analysis within a randomised trial. https://pubmed.ncbi.nlm.nih.gov/22781782/
6. Barnard ND, Kahleova H, Holtz DN, et al. A dietary intervention for vasomotor symptoms of menopause: a randomized, controlled trial. Menopause 2023;30(1):80-87 — Peer-reviewed randomised controlled trial. https://pubmed.ncbi.nlm.nih.gov/36253903/
7. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multi-racial/ethnic population of women 40-55 years of age. American Journal of Epidemiology 2000;152(5):463-73 — Peer-reviewed community-based survey (SWAN). https://pubmed.ncbi.nlm.nih.gov/10981461/
8. Gold EB, Colvin A, Avis N, et al. Longitudinal analysis of the association between vasomotor symptoms and race/ethnicity across the menopausal transition: study of women's health across the nation. American Journal of Public Health 2006;96(7):1226-35 — Peer-reviewed longitudinal cohort analysis (SWAN). https://pubmed.ncbi.nlm.nih.gov/16735636/
9. Avis NE, Crawford SL, Greendale G, et al. Duration of menopausal vasomotor symptoms over the menopause transition. JAMA Internal Medicine 2015;175(4):531-539 — Peer-reviewed longitudinal cohort analysis (SWAN), via PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC4433164/
10. Nagata C, Takatsuka N, Kawakami N, Shimizu H. Soy product intake and hot flashes in Japanese women: results from a community-based prospective study. American Journal of Epidemiology 2001;153(8):790-3 — Peer-reviewed prospective cohort study. https://pubmed.ncbi.nlm.nih.gov/11296152/
11. Melby MK. Vasomotor symptom prevalence and language of menopause in Japan. Menopause 2005;12(3):250-7 — Peer-reviewed cross-sectional study. https://pubmed.ncbi.nlm.nih.gov/15879913/
12. Lock M, Kaufert P, Gilbert P. Cultural construction of the menopausal syndrome: the Japanese case. Maturitas 1988;10(4):317-32 — Peer-reviewed interview and survey study. https://pubmed.ncbi.nlm.nih.gov/3265758/
13. Yasui T, Ideno Y, Shinozaki H, et al. Prevalence of the Use of Oral Contraceptives and Hormone Replacement Therapy in Japan: The Japan Nurses' Health Study. Journal of Epidemiology 2022;32(3):117-124 — Peer-reviewed prospective cohort study. https://pubmed.ncbi.nlm.nih.gov/33132283/
14. Takamatsu K, Ogawa M, Higuchi T, et al. Effects of Kamishoyosan, a Traditional Japanese Medicine, on Menopausal Symptoms: A Randomized, Placebo-Controlled, Double-Blind Clinical Trial. Evidence-Based Complementary and Alternative Medicine 2020;2020:9285317 — Peer-reviewed randomised controlled trial. https://pubmed.ncbi.nlm.nih.gov/32733592/
15. Zhu X, Liew Y, Liu ZL. Chinese herbal medicine for menopausal symptoms. Cochrane Database of Systematic Reviews 2016;3(3):CD009023 — Cochrane systematic review. https://pubmed.ncbi.nlm.nih.gov/26976671/
16. Matthews KA, Crawford SL, Chae CU, et al. Are changes in cardiovascular disease risk factors in midlife women due to chronological aging or to the menopausal transition? Journal of the American College of Cardiology 2009;54(25):2366-2373 — Peer-reviewed longitudinal cohort analysis (SWAN). https://pmc.ncbi.nlm.nih.gov/articles/PMC2856606/
17. El Khoudary SR, Chen X, Nasr AN, et al. HDL (high-density lipoprotein) subclasses, lipid content, and function trajectories across the menopause transition: SWAN-HDL study. Arteriosclerosis, Thrombosis, and Vascular Biology 2021;41:951-961 — Peer-reviewed longitudinal cohort analysis (SWAN). https://pmc.ncbi.nlm.nih.gov/articles/PMC8105263/
18. Greendale GA, Sowers M, Han W, et al. Bone mineral density loss in relation to the final menstrual period in a multiethnic cohort: results from the Study of Women's Health Across the Nation (SWAN). Journal of Bone and Mineral Research 2012;27(1):111-118 — Peer-reviewed longitudinal cohort analysis. https://pubmed.ncbi.nlm.nih.gov/21976317/
19. Greendale GA, Sternfeld B, Huang M, et al. Changes in body composition and weight during the menopause transition. JCI Insight 2019;4(5):e124865 — Peer-reviewed longitudinal cohort analysis (SWAN). https://insight.jci.org/articles/view/124865
20. Newton KM, Reed SD, Uchiyama S, et al. A cross-sectional study of equol producer status and self-reported vasomotor symptoms. Menopause 2015;22(5):489-495 — Peer-reviewed observational study. https://pubmed.ncbi.nlm.nih.gov/25380274/
21. Blanco Mejia S, Messina M, Li SS, et al. A meta-analysis of 46 studies identified by the FDA demonstrates that soy protein decreases circulating LDL and total cholesterol concentrations in adults. The Journal of Nutrition 2019;149(6):968-981 — Peer-reviewed meta-analysis. https://pubmed.ncbi.nlm.nih.gov/31006811/
22. Bauer J, Biolo G, Cederholm T, et al. Evidence-based recommendations for optimal dietary protein intake in older people: a position paper from the PROT-AGE Study Group. Journal of the American Medical Directors Association 2013;14(8):542-559 — Expert group position paper. https://pubmed.ncbi.nlm.nih.gov/23867520/
23. Food and Nutrition Board, Institute of Medicine, National Academies. Dietary Reference Intakes: Recommended Dietary Allowances and Adequate Intakes, Total Water and Macronutrients — Summary table via NCBI Bookshelf. https://www.ncbi.nlm.nih.gov/books/NBK56068/table/summarytables.t4/
24. MedlinePlus, US National Library of Medicine. Calcium in diet — Medical encyclopedia. https://medlineplus.gov/ency/article/002412.htm
25. MedlinePlus, US National Library of Medicine. Vitamin D — Medical encyclopedia. https://medlineplus.gov/ency/article/002405.htm
26. Harlow SD, Gass M, Hall JE, et al. Executive summary of the Stages of Reproductive Aging Workshop +10: addressing the unfinished agenda of staging reproductive aging. 2012 — Consensus staging criteria, via PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC3340903/

Sources 1 and 16-26 accessed 5 August 2026; sources 2-15 accessed 20 October 2026.

Olivia Smith, RDN

Registered Dietitian Nutritionist (RDN)

A registered dietitian who would rather give you a principle with a source than a number without one.

View profileFollow on X