What changed about hormone therapy since 2002

In July 2002, the National Institutes of Health halted one arm of the Women's Health Initiative early, and by dinnertime the story was everywhere: hormones cause breast cancer. Prescriptions fell off a cliff. Doctors who'd been writing estrogen for twenty years stopped cold, and a whole generation of women got handed a fan and told to ride it out.

The detail that didn't make the evening news: the average woman in that trial was 63 when she enrolled. A lot of them were more than a decade past their last period. Some were in their seventies, and some already had plaque in their arteries. They weren't the 51-year-old waking up soaked at 3 a.m. four nights a week. The trial was designed to ask whether hormones prevent heart disease in older women. It answered that question. It got read as an answer to a different one.

Twenty-plus years of re-analysis later, the picture is more specific, and specificity is the whole ballgame.

What it's genuinely good at

Systemic estrogen is still the most effective treatment there's for hot flashes and night sweats. Nothing else in the cabinet matches it. If your sleep is being shredded four or five nights a week, that's the intervention with the biggest effect size, and it isn't close.

It also protects bone. Estrogen is FDA-approved for the prevention of postmenopausal osteoporosis, and that's not a side benefit you shrug at if you're small-framed, if you smoked, or if your mother broke a hip at 68.

What it's not: a heart attack preventive, a dementia preventive, or an anti-aging protocol. No major medical body recommends starting it for those reasons, and any clinic selling it that way is selling something. Treat symptoms. That's the indication.

The timing thing

The single most useful idea to come out of the last two decades is that when you start matters more than whether you start.

The general framing used by The Menopause Society and by ACOG is the under-60, within-10-years window. Start systemic hormone therapy before age 60 or within ten years of your final period, with no contraindications, and the benefit-risk balance leans favorable for most healthy women. Start at 68, fifteen years out, and the same drug is a different risk proposition, because the cardiovascular system it's landing in is different.

This is why "I read that hormones are dangerous" and "my sister's doctor said they're fine now" can both be true. Different women, different clocks.

The pill versus the patch is a real decision

Oral estrogen goes through your liver first, and that first pass bumps clotting factors. Transdermal estradiol, the patch or the gel or the spray, goes straight into the bloodstream and skips that step. Observational data has pretty consistently shown lower clot and stroke risk with transdermal than with oral.

That matters enormously if you have a migraine-with-aura history, a high BMI, or any personal or family clot history. It matters less if you're low risk. But if your clinician reaches for a pill by reflex and you carry any of that baggage, it's fair to ask why not the patch.

Generic estradiol patches are also usually the cheaper end of the shelf, which isn't nothing when you're looking at years of it.

If you still have a uterus, you need the second drug

Estrogen alone thickens the uterine lining and raises endometrial cancer risk. Unopposed estrogen in a woman with a uterus isn't a nuance, it's a mistake. You need a progestogen alongside it, most commonly micronized progesterone, sometimes a levonorgestrel IUD.

Women who've had a hysterectomy take estrogen alone. And here's the part that never got a headline: in the WHI's estrogen-alone arm, breast cancer incidence didn't go up. Some follow-up analysis pointed the other direction. The breast signal in 2002 came from the combined arm.

The breast cancer conversation, straight

Combination therapy does carry an increase in breast cancer risk, and the risk grows with duration of use. Don't let anybody wave that away.

The 2022 position statement from what was then the North American Menopause Society characterized that attributable risk as rare, on the order of fewer than one additional case per 1,000 women per year of use. Small isn't zero. For a woman with a dense-breast history and a mother and an aunt who both had it, that math reads differently than it does for a woman with no family history at all, and the honest answer is that it should.

What you can do is hold up your end. Stay current on mammograms. Know your family history well enough to say it out loud in the exam room.

Vaginal estrogen is a separate conversation entirely

Low-dose vaginal estrogen treats the tissue changes of menopause: dryness, irritation, burning with urination, recurrent UTIs. Systemic absorption is minimal. It's a local therapy with a local risk profile, and it gets lumped in with systemic hormones constantly by people who should know better, including some pharmacists.

In November 2025 the FDA moved to remove the boxed warning from menopausal hormone products, including low-dose vaginal estrogen. If you were handed a box with a black-bordered cancer warning and quietly stopped filling it, that's worth raising again. A label change isn't a personal green light, though. It's a labeling decision, not your chart.

Skip the compounded pellets

Compounded "bioidentical" hormone preparations aren't FDA-approved, aren't tested for batch-to-batch dose consistency, and are frequently sold alongside saliva hormone testing, which has no established value for dosing. Pellets are the worst of the category: once they're in, you can't adjust the dose or take them out, and supraphysiologic levels are common.

The irony is that FDA-approved bioidentical options already exist. Estradiol is estradiol whether it comes from a patch at the pharmacy or a pellet in a strip-mall clinic. One of them you can stop on a Tuesday.

Who needs the harder conversation

A history of breast cancer, endometrial cancer, blood clots, stroke, heart attack, active liver disease, or unexplained vaginal bleeding. Those aren't automatic nos in every single case, but they're all specialist territory, and the bleeding one needs worked up before anybody writes anything.

If you can't take estrogen, there are real options now. Fezolinetant, approved in 2023, targets the brain pathway behind hot flashes and isn't a hormone. It requires liver monitoring and carries a warning for liver injury, so it's a bloodwork commitment. A second drug in that same class has since been approved. Certain SSRIs and SNRIs help. Gabapentin at night helps some women sleep through the sweats.

Running the appointment

Go in with a symptom log. Two weeks is plenty. Count the flashes, count the wake-ups, note what you gave up because of them.

Ask three things directly: transdermal or oral and why, what my personal contraindications are, and when we reassess. Then give it about three months before you judge whether it's working, because doses get adjusted and the first one is a starting guess, not a verdict.

If your doctor's entire position is "hormones cause cancer, try layering," that's 2002 talking. The Menopause Society keeps a searchable directory of certified practitioners at menopause.org. Twelve minutes with someone who's read the last decade of data beats twenty years of layering.

You're not asking for a favor. You're asking for the standard of care, and it exists.