Hormone Therapy: What the Evidence Actually Shows

For more than two decades, many women heard a simplified message: menopausal hormones are dangerous. That message grew out of early interpretations of the Women's Health Initiative (WHI) and was often applied broadly to women of different ages, different risk profiles, and hormone regimens the WHI did not actually study.

In 2025 and 2026, the FDA formally changed that framework. Boxed-warning language related to cardiovascular disease, breast cancer, and probable dementia was removed from several menopausal hormone therapy products. The endometrial-cancer boxed warning remains for systemic estrogen-alone therapy.

That does not mean hormone therapy has no risks. It means the risks are more individualized than the old blanket warning suggested.

Read the FDA February 2026 announcement

Modern menopausal hormone therapy may involve:

• Estradiol — often delivered by patch, gel, cream, oral medication, or other route
• Progesterone — used when endometrial protection is needed and sometimes for other clinical reasons
• Testosterone — used selectively in women when clinically appropriate
• Different routes matter: oral and transdermal estrogen do not have identical effects on clotting and metabolism
• Different progestogens matter: micronized progesterone is not pharmacologically identical to medroxyprogesterone acetate (MPA)

The WHI primarily studied oral conjugated equine estrogen (CEE), with or without MPA. It did not test every modern estradiol/progesterone regimen.

SSRIs and SNRIs can reduce hot flashes and are useful options for some women, especially when systemic estrogen is contraindicated or unwanted.

But they are not hormone replacement.

They do not replace estrogen. They do not restore estrogen-responsive vaginal and urinary tissue. They do not provide estrogen's bone-preserving effect. They treat selected symptoms through neurotransmitter pathways.

For an eligible woman whose symptoms are being driven by menopause, the decision should therefore not be framed as though an SSRI/SNRI and hormone therapy are interchangeable treatments.

SSRIs/SNRIs also have their own risks and tradeoffs

Depending on the medication and the patient, these can include:

• sexual dysfunction, reduced libido, delayed orgasm or anorgasmia
• weight gain with some agents
• discontinuation symptoms when treatment is stopped
• persistent sexual symptoms reported after discontinuation in a subset of patients; the true incidence is uncertain

Nonhormonal therapy may be the right choice for an individual patient. But patients deserve the same informed-consent discussion about these drugs that they are routinely given about hormones.

The WHI produced two very different breast-cancer findings that are often collapsed into one.

In women with a prior hysterectomy who were randomized to CEE estrogen alone, long-term follow-up found lower breast-cancer incidence and lower breast-cancer mortality compared with placebo.

The estrogen-plus-MPA arm had a different breast-cancer outcome.

That does not prove that every estrogen regimen prevents breast cancer, nor that all combined regimens carry the same risk. It demonstrates why "HRT causes breast cancer" is too broad to accurately describe the evidence.

Read the JAMA 2020 WHI long-term follow-up

Oral estrogen passes through the liver first and alters hepatic production of clotting factors.

Transdermal estradiol bypasses that first-pass effect. Observational studies and meta-analyses consistently find higher venous-thromboembolism risk with oral estrogen, while transdermal estradiol has shown little or no measurable increase in VTE risk in most studies.

This is one reason a woman's route of therapy should be individualized rather than discussing "estrogen risk" as though every preparation were equivalent.

The WHI combined estrogen with medroxyprogesterone acetate (MPA).

Micronized progesterone is a different molecule with different pharmacology. Observational studies suggest it may have a more favorable thrombotic and breast-risk profile than some synthetic progestins, although long-term randomized head-to-head outcome data remain limited.

This distinction deserves to be part of informed consent.

Current FDA/HHS messaging emphasizes that systemic menopausal hormone therapy generally has its most favorable benefit-risk profile when initiated before age 60 or within approximately 10 years of menopause onset, assuming the patient is otherwise an appropriate candidate.

A healthy 52-year-old several years into menopause should not automatically be counseled using risk estimates derived from women who began therapy much later in life.

These areas require nuance.

FDA/HHS now emphasizes evidence suggesting more favorable cardiovascular and overall outcomes when HRT is initiated near menopause. Observational studies have also raised the possibility of cognitive benefit when therapy is started near the menopausal transition.

However:

• hormone therapy should not be sold as a proven dementia-prevention treatment
• cardiovascular benefit is not identical for every woman or formulation
• the overall WHI randomized cohort did not show increased long-term all-cause mortality
• age, timing, route, baseline risk and hormone formulation all matter

The appropriate question is not simply, "Are hormones risky?" It is:

"What are the likely benefits and risks of this specific hormone regimen for this specific patient?"

1. Are you aware of the FDA's 2025-2026 menopausal hormone therapy labeling changes?
2. Which risks apply to me specifically based on my age, time since menopause, medical history and route of treatment?
3. Are you treating oral estrogen and transdermal estradiol as though their clotting risks are identical?
4. If I need progesterone, have we discussed micronized progesterone versus synthetic progestins?
5. If you recommend an SSRI or SNRI for menopausal symptoms, have we discussed sexual side effects, weight effects and discontinuation symptoms as part of informed consent?
6. If I am an appropriate candidate for HRT, what is the rationale for treating only selected symptoms rather than the underlying estrogen deficiency?