Hormone therapy’s risks depend on which hormones, when and how, and most herbs fell short

Key takeaways

• After 20 years in a large federal trial, estrogen alone was linked to fewer breast cancers, while estrogen plus a progestin added about 9 cases per 10,000 women a year, by our math⁵.
• The Menopause Society says that for most healthy women under 60 and within 10 years of menopause who have bothersome symptoms, the benefits of hormone therapy outweigh the risks³⁴.
• In a British study of more than 80,000 women with blood clots, estrogen pills were linked to higher clot risk, while patches and gels were not⁷.
• The National Academies said in 2020 that compounded hormones should be limited to narrow medical needs, and FDA investigators found records of 4,202 unreported problems tied to one pellet seller⁸⁹.
• In an NIH trial, black cohosh and a 10-herb blend did no better than placebo for hot flashes, while hormone therapy cut about 4 symptoms a day more than placebo¹⁰.

In 2020, researchers reported a twist in the long-running hormone story. After about 20 years of follow-up, women who had taken estrogen alone in a major federal trial had fewer breast cancers than women given a placebo⁵. They also had fewer breast cancer deaths⁵.

Women in a sister trial who took estrogen plus a progestin, a man-made form of progesterone, had more breast cancers⁵.

Same trial program, opposite results. That’s why the question “Is hormone therapy safe?” has no single answer. It depends on which hormones you take, when you start and how they enter your body.

The question returned to the news in July 2025, when the FDA convened an expert panel¹. Here are the answers, with absolute numbers where the sources give them.

A 2025 FDA panel reopened the debate, and critics pushed back

On July 17, 2025, the FDA held a meeting with 12 experts, whom Axios described as mostly like-minded¹². They urged removing the boxed warning, the FDA’s most prominent label caution, from menopause hormone products¹. The panel took no vote and had no power to change policy¹.

FDA Commissioner Marty Makary had already criticized those warnings in his book “Blind Spots” as an example of medical groupthink¹.

Critics said the panel was stacked toward a set conclusion and could sway policy without public comment¹. The New York Times reported that several panelists consult for drug companies or earn money prescribing hormones¹. Adriane Fugh-Berman, a Georgetown University pharmacologist, said the panelists leaned on observational studies, which can make benefits look bigger¹. An HHS spokesperson called claims of a one-sided process insulting¹.

The scientists who run the Women’s Health Initiative (WHI), the large federal hormone trials, filed a formal comment defending their work². They called some panelists’ statements inaccurate, misleading or out of context².

They disputed a claim that 50,000 women died from avoiding estrogen². They also called the “timing hypothesis,” the idea that starting early brings lasting heart benefits, untested in trials². Yet they took no position on removing the boxed warning and supported short-term use for hot flashes in younger women, whose absolute risks are low².

Our menopause sleep article covers the label changes that followed.

For many women under 60, benefits outweigh the risks

The Menopause Society’s 2022 position statement calls hormone therapy the most effective treatment for hot flashes³. For most healthy women with symptoms who are under 60 and within 10 years of menopause, it says benefits outweigh risks³⁴.

Starting later looks less favorable. Beginning more than 10 years after menopause, or at 60 or older, carries greater absolute risks of heart disease, stroke, blood clots and dementia⁴.

The WHI data back up the age pattern. In a 2021 talk, WHI investigator JoAnn Manson showed these trial figures for each 1,000 women over five years⁶:

Outcome per 1,000 women over 5 years	Ages 50–59	Ages 70–79
Deaths, estrogen plus progestin	5 fewer	2 more
Deaths, estrogen alone	6 fewer	13 more
Serious events overall, estrogen plus progestin	6 more	19 more
Serious events overall, estrogen alone	10 fewer	25 more

“Serious events overall” is the trial’s combined count of heart disease, stroke, lung clots, breast, colon and uterine cancer, hip fracture and death⁶. The slides didn’t say which age-group differences were statistically reliable, and smaller subgroups are less certain than whole-trial results.

Her conclusions: hormones shouldn’t be used to prevent heart disease or other chronic diseases, and the best candidates are women with symptoms who are recently menopausal and in good health⁶.

Estrogen alone and estrogen plus progestin act differently on the breast

The 20-year WHI analysis followed 27,347 women with an average age of 63 at the start⁵. The 16,608 women with a uterus took estrogen plus progestin or a placebo for a median of 5.6 years. The 10,739 women who’d had a hysterectomy took estrogen alone or a placebo for a median of 7.2 years⁵.

The yearly averages⁵:

• Estrogen alone: breast cancer in 0.30% of women, compared with 0.37% on placebo. By our math, that’s about 7 fewer cases per 10,000 women a year. Breast cancer deaths were 40% lower.
• Estrogen plus progestin: breast cancer in 0.45% of women, compared with 0.36% on placebo. By our math, that’s about 9 more cases per 10,000 women a year. The difference in deaths wasn’t statistically clear.

Lead author Rowan Chlebowski said the extra risk from the combined pills lingered long after women stopped taking them⁵. The WHI team says the combined pills’ risk was about as high in women under 60 as in older women².

Why not give everyone estrogen alone? The FDA lists cancer of the uterine lining as a risk for women with a uterus who take estrogen without a progestin¹⁵. Our article on observational studies tells how the WHI’s 2002 results first upended hormone advice.

One more limit matters. The WHI tested oral conjugated equine estrogens, alone or with medroxyprogesterone acetate². Its investigators say no comparable trials show whether estradiol or progesterone behave the same way².

Patches and gels may avoid the clot risk tied to pills

Estrogen comes as pills or as patches, gels and sprays that work through the skin¹⁵. A 2019 British study in The BMJ compared the two routes⁷.

Using medical records from 1998 to 2017, researchers matched 80,396 women aged 40 to 79 who’d had a blood clot in a vein with 391,494 similar women who hadn’t⁷.

Oral hormone therapy was tied to a 58% higher clot risk: 40% higher for estrogen-only pills and 73% higher for combined pills⁷. Hormones through the skin showed no added risk⁷.

The progestogen mattered, too. The pairing used in the WHI carried the highest risk, about double. Estradiol with a progestogen called dydrogesterone carried the lowest, an 18% rise that wasn’t statistically clear⁷.

In absolute terms, oral hormones added about 9 clots a year for every 10,000 women taking them⁷. Put another way, the researchers estimated one extra clot for every 1,076 women treated⁷. The study had no outside funding⁷.

This was an observational study, which shows links rather than proof. The Menopause Society agrees that skin-based routes and lower doses may lower the risk of clots and stroke³.

Vaginal estrogen, sold as creams, tablets, inserts and rings, is a separate case¹⁵. The Menopause Society recommends low doses for bothersome dryness and says they can be used at any age and for extended periods⁴. The WHI team says its results may not apply to vaginal products². Our companion article on midlife sexual health covers these treatments.

Compounded hormones and pellets skip the checks approved products get

Compounded hormones, often sold as “bioidentical,” are mixed by a pharmacy for one patient.

In 2020, the National Academies of Sciences, Engineering, and Medicine told prescribers to reserve these products for narrow medical needs, which our hormone-claims article describes⁸. A patient’s preference alone shouldn’t decide it, the committee said, and patients should be told the products aren’t FDA-approved⁸. Committee chair Donald Mattison said the evidence wasn’t strong enough to guide patients about benefits and harms⁹.

The report singled out pellets, small hormone implants placed under the skin. It said all compounded pellet products should be considered for the FDA’s list of drugs too difficult to compound⁸. It also wanted compounded estrogen and testosterone to carry boxed warnings like those on approved versions, and prescribers to disclose any financial stake in the products they prescribe⁸.

A viewpoint by FDA scientists in JAMA Internal Medicine illustrated the stakes. FDA investigators on a routine visit, for a purpose that had nothing to do with compounding, turned up records of 4,202 adverse events linked to compounded pellets sold by BioTE Medical⁹. Not one had been reported to the agency⁹.

The records included cancers, strokes, heart attacks, deep-vein clots, skin infections, pellets working their way out and hormone levels above the intended range⁹. Such reports don’t prove the pellets caused each problem. They do show what goes unseen when no one is tracking.

If natural to you means hormones identical to your body’s own, ask whether an FDA-approved version would work. Our article on hormone claims explains why you may not have to choose between your values and quality control.

Black cohosh went head to head with hormones and lost

The Herbal Alternatives for Menopause (HALT) trial is the kind of test many natural-health advocates ask for. It was funded by the National Institutes of Health, not a drug or supplement company, and it pitted herbs against hormones and a placebo for a full year¹⁰.

Researchers at Group Health in Seattle and the University of Washington, led by Katherine Newton, enrolled 351 women aged 45 to 55 with at least two hot flashes or night sweats a day, and reported results in 2006¹⁰. Each woman got one of five treatments¹⁰¹¹:

• black cohosh, 160 mg a day of a root extract standardized to 2.5% triterpene glycosides;
• a 10-ingredient blend with black cohosh, dong quai, chaste tree, licorice and others;
• the same blend plus counseling to eat 12 to 20 grams of soy protein a day;
• standard hormone therapy;
• a placebo.

At one year, none of the herbal options differed from placebo by even 0.6 symptoms a day¹⁰. Hormone therapy cut 4.06 symptoms a day compared with placebo¹⁰. Black cohosh showed one small early gain, 0.38 fewer night sweats a day at three months¹¹. The blend plus soy left symptoms more intense than placebo at one year¹¹.

After the trial began, lab tests couldn’t find dong quai, false unicorn or pomegranate in the blend, according to a commentary by Donald Brown, a naturopathic doctor who consults for the supplement maker Nature’s Way¹¹. The trial’s authors noted it was too small to detect changes of less than 1.5 hot flashes a day¹¹.

The wider record is mixed. A 2016 review found no high-quality, consistent evidence that black cohosh beats placebo, while a 2017 analysis of 47 trials with 8,326 women found it beat placebo, though not hormone patches with a progestogen¹². Rare, sometimes serious liver damage has been reported, and some products contained the wrong herb¹². Our menopause sleep article covers the resulting label warning.

Red clover, soy and other herbs show small or no effects

Red clover. A 2016 pooled analysis suggested it may reduce hot flashes, mainly in women with five or more a day, and may ease vaginal dryness¹². Studies of up to three years found no clear harm, but animal research raised concerns about hormone-sensitive tissue¹²¹³. The Menopause Society calls the evidence inconsistent¹³.

Soy isoflavones. A 2016 analysis of 62 studies with 6,653 women found modest drops in hot flashes, though many studies were low quality¹². Long-term use of soy extracts has been tied to a thickened uterine lining¹². Our estrogen article covers soy in depth.

Dong quai. Its one randomized trial found no help for hot flashes¹³. The Menopause Society says women with fibroids or clotting problems, and those taking the blood thinner warfarin, should never use it¹³.

Evening primrose oil. Its one placebo-controlled trial, in 56 women, found no benefit¹³. It shouldn’t be combined with blood thinners or a class of antipsychotic drugs called phenothiazines¹³.

St. John’s wort. This herb speeds the breakdown of many medicines, lowering their levels, and with some antidepressants it can raise the risk of serious serotonin-related side effects¹⁴.

Quality varies by brand and batch, so tell your clinician about any herb you use¹³. NCCIH suggests bringing a written list of every product and dose¹⁴.

The question card, and how strong the evidence is

This card is our own design, built from the sources above.

Question	What the evidence says	Key number	Source
Is it safe for me?	Most favorable under 60 or within 10 years of menopause	Ages 50–59: 5 to 6 fewer deaths per 1,000 over 5 years	46
Does the type matter?	Estrogen alone lowered breast cancer; adding a progestin raised it	About 7 fewer vs. 9 more per 10,000 a year	5
Does the route matter?	Pills tied to more clots; skin routes weren’t	About 9 extra clots per 10,000 women a year on pills	7
Is compounded better?	Not shown; limit to narrow medical needs	4,202 unreported pellet problems at one company	89
Will black cohosh work?	No better than placebo in a year-long NIH trial	Under 0.6 symptoms a day apart	10

Option	Rating	Basis
Hormone therapy for hot flashes	Strong evidence	Cut about 4 symptoms a day vs. placebo310
Lower clot risk with patches or gels	Promising but early	Large observational study, not a trial37
Compounded hormones as safer or better	Not enough evidence	Evidence too limited to guide patients89
Black cohosh for hot flashes	Evidence says no, or unsafe	Failed in HALT; pooled studies disagree; liver reports1012
Multi-herb blends	Evidence says no, or unsafe	No benefit, worse with soy, missing herbs1011
Red clover	Promising but early	Small, inconsistent benefits; an animal-study concern1213
Dong quai or evening primrose oil	Not enough evidence	One small trial each, no benefit; interaction risks13

Who shouldn’t use hormone therapy, and what to ask

An FDA consumer guide to menopause medicines, which warns that its risk information may change, says not to take hormone therapy if you¹⁵:

• have vaginal bleeding problems;
• have, or once had, some cancers, including breast or uterine cancer;
• have ever had a stroke, heart attack or blood clot;
• have liver disease or a bleeding disorder;
• have reacted allergically to a hormone medicine.

The same page lists serious side effects, including stroke, blood clots, a clot in the eye that can cause vision loss, liver problems and high blood pressure¹⁵. Check your own product’s current label, too. Our estrogen article lists clot warning signs that need care right away. Problems with hormone products can be reported to the FDA’s MedWatch program⁸.

The FDA suggests asking whether hormones are right for you and why, what the benefits, serious risks and common side effects are, how long to use them, and whether non-hormone medicines would work¹⁵. We’d add three questions:

• Which estrogen, which progestogen and which route are you suggesting?
• Is this product FDA-approved?
• Does anything else I take, including herbs, interact with it?

Questions readers ask

The bottom line

Hormone therapy isn’t one product with one risk. In the WHI, estrogen alone and estrogen plus a progestin affected the breast differently. Pills and skin products differ on clots, and age at the start shifts the balance. For most healthy women under 60 with bothersome hot flashes, The Menopause Society says benefits outweigh risks.

The herbs sold as alternatives haven’t matched that record in fair tests, and compounded pellets add dose and quality questions. Bring the question card to your appointment and decide with a clinician who knows your history.

Related reading

• Hot flashes can last a decade, and scientists traced them to a thermostat in the brain
• Real hormone problems have names and tests, which “balance your hormones” claims lack
• Why observational studies got smoking right and hormone therapy wrong
• One man’s bones proved estrogen matters for men as well as women
• Hot flashes aren’t the only thing stealing sleep in menopause

Sources

1. Axios. “FDA Panel Backs Menopausal Hormone Therapy.” July 18, 2025. axios.com
2. Women’s Health Initiative Steering Committee. “WHI Response to FDA Expert Panel on MHT Use” (public comment to Docket FDA-2025-N-2589). September 2025. whi.org
3. Contemporary OB/GYN. “New Hormone Therapy Guidelines From The North American Menopause Society” (on the 2022 Hormone Therapy Position Statement). July 22, 2022. contemporaryobgyn.net
4. The North American Menopause Society (now The Menopause Society). “The 2022 Hormone Therapy Position Statement of The North American Menopause Society” (abstract and highlights reprinted by the Australasian Menopause Society). Menopause, 2022. menopause.org.au
5. Cimino S. “Postmenopausal Use of Estrogen Alone Lowers Breast Cancer Cases, Deaths” (report on Chlebowski RT, et al., JAMA, July 28, 2020, doi:10.1001/jama.2020.9482). MDedge, July 29, 2020. mdedge.com
6. Manson JE. “The Women’s Health Initiative: Hormone Therapy Trials.” WHI webinar slides, March 17, 2021. whi.org
7. MDedge. “Most Oral HRT Linked to Increased VTE Risk” (report on Vinogradova Y, Coupland C, Hippisley-Cox J, BMJ, January 9, 2019). mdedge.com
8. National Academies of Sciences, Engineering, and Medicine. “The Clinical Utility of Compounded Bioidentical Hormone Therapy: Recommendations.” National Academies Press, July 2020. Recommendations Insert.pdf
9. Fulton D. “Report Calls for Limits on Compounded Bioidentical Hormones.” Regulatory Focus (RAPS), July 2, 2020. raps.org
10. National Institutes of Health. News release on Newton KM, et al., the Herbal Alternatives for Menopause (HALT) trial, Annals of Internal Medicine, December 19, 2006. December 18, 2006.
11. Brown D. “NIH-Funded HALT Study Finds Black Cohosh Ineffective for Treating Menopausal Symptoms.” Alternative Medicine Alert (Relias Media). reliasmedia.com
12. National Center for Complementary and Integrative Health. “Menopausal Symptoms and Complementary Health Practices: What the Science Says.” Clinical Digest, NIH, February 2021. nccih.nih.gov
13. The Menopause Society. “Do Mother Nature’s Treatments Help Hot Flashes?” Accessed October 2026.
14. National Center for Complementary and Integrative Health. “Tips: How Herbs Can Interact With Medicines.” NIH, accessed October 2026. nccih.nih.gov
15. U.S. Food and Drug Administration, Office of Women’s Health. “Menopause: Medicines to Help You.” Content current as of August 22, 2019. fda.gov