Does hormone replacement therapy actually work? What the evidence says in 2026

Few areas of medicine have swung as hard, or as publicly, as hormone replacement therapy. Women were told it was the key to healthy aging, then told it caused cancer, and are now being told, more or less, that it works after all. The truth is more interesting than any of those headlines.

A quick primer on the biology

Estrogen and progesterone are steroid hormones. That matters because steroid hormones don't just knock on the outside of cells the way most drugs do. They pass right through the cell membrane, enter the nucleus, and change what DNA gets transcribed. That's why their effects are so broad, and why we still don't fully understand everything they do.

During perimenopause and menopause, estrogen falls off a cliff. Progesterone drops too. The idea behind HRT is straightforward: put those levels back where they were, not higher. That's different from oral contraceptives, which use much higher doses to suppress a woman's own hormone production.

Estrogen is the hormone doing most of the symptomatic work. Progesterone is there for a specific reason: to protect the lining of the uterus. Estrogen alone stimulates the uterine lining to grow, and unchecked, that raises the risk of uterine cancer. If you have a uterus, you need progesterone with your estrogen. If you don't, estrogen alone is an option, and some of the outcomes on estrogen-only regimens actually look slightly better.

The Women's Health Initiative, and what it actually showed

In the early 1990s, the NIH launched the Women's Health Initiative. About 27,000 postmenopausal women, average age 63, were randomized to HRT or placebo. In 2002, one arm was halted. Invasive breast cancer was up about 25% in relative terms. Cardiovascular events were up about 30%. Prescriptions for HRT in the US dropped by roughly 80% almost overnight.

Here's where the framing went sideways. A 25% relative increase sounds catastrophic. The absolute numbers were 30 cases of breast cancer per 10,000 women per year in the placebo group versus 38 per 10,000 in the HRT group. Eight extra cases per 10,000. Not nothing, but not what most women heard.

The bigger problem was external validity. The WHI wasn't designed to test whether HRT relieves menopausal symptoms. It was designed to test whether HRT protects the heart in postmenopausal women. Seventy percent of the women enrolled were more than 10 years past menopause. They weren't the women who would actually be prescribed HRT for symptoms today.

When researchers later stratified by age, the picture changed. Women aged 50 to 59, closer to menopause, actually showed reduced cardiovascular disease and reduced all-cause mortality on HRT. The harm signal was concentrated in the older cohort. We took a result from women in their 60s and 70s and applied it to women in their 40s and 50s, and a lot of women lost access to a treatment that probably would have helped them.

What HRT actually does for symptoms

Hot flashes are where HRT shines. Randomized trial data show a 70% to 80% reduction in vasomotor symptoms. For women who are waking up drenched every night, that is life-changing.

Sleep improves too. The KEEPS trial, in recently menopausal women, showed a significant improvement in global sleep quality on HRT versus placebo, with most of the benefit concentrated in women who had severe hot flashes. A lot of the vaguer complaints, brain fog, fatigue, mood, may simply be downstream of sleeping through the night again.

The data on cognition itself is less impressive. Randomized trials of neurocognitive testing don't show dramatic effects, with the possible exception of verbal memory. The "scales fell from my eyes" stories you see online are real for some women, but I'm skeptical of any single intervention doing that much.

Genitourinary symptoms, including vaginal dryness, respond well to topical vaginal preparations. These can even be an option for women who aren't candidates for systemic estrogen, because very little gets into the bloodstream. There's also evidence that HRT reduces fracture risk in women at risk for osteoporosis.

Pills, patches, and why the route matters

Oral estrogen has to pass through the liver before reaching the bloodstream. The liver breaks down about 95% of 17-beta estradiol on that first pass, which means oral doses have to be high to produce normal blood levels. Patches and vaginal preparations bypass the liver. Some of the clot risk seen with HRT appears to be driven by the oral route, and while we don't fully understand why, it's plausible that variable liver metabolism makes oral dosing trickier to get right.

Right now there's actually a shortage of patches, which complicates things further. The point is that "HRT" isn't one thing. The hormone, the dose, the route, and whether you have a uterus all change the calculus.

Testosterone in women

Women make testosterone too, at roughly a tenth of male levels. It peaks around age 20 and declines slowly, without the menopausal cliff of estrogen. The best-studied indication is hypoactive sexual desire disorder. The largest meta-analysis, published in The Lancet and pooling data from about 8,500 women, showed that testosterone therapy produced approximately one additional satisfying sexual event per month compared to placebo.

Beyond sexual function, the evidence is thin. Testosterone can raise cholesterol and may promote atherosclerosis, so it's not the unambiguous win that estrogen is. And side effects like acne and hair growth are real.

A word on hormone tests and telehealth

There is a large industry now selling hormone panels direct to consumers, along with HRT prescriptions. The tests are mostly a cash grab. In perimenopause, estrogen levels swing wildly from day to day. The indication for HRT is clinical: are you at an age consistent with perimenopause or menopause, and do you have symptoms? That's it. Repeated hormone testing doesn't change the treatment.

Bottom line

HRT works, and it works especially well for the specific symptoms it's used to treat: hot flashes, night sweats, sleep disruption, vaginal dryness, and fracture risk. The Women's Health Initiative scared a generation of women and their doctors away from a treatment that, for symptomatic women in their 40s and 50s, probably does more good than harm. The risks are real but smaller in absolute terms than the headlines suggested, and they depend heavily on age, route of administration, and whether you're taking estrogen alone or with progesterone. This is genuinely a conversation to have with your doctor, not a decision to make from a podcast or a direct-to-consumer website.

Here's the "What's the deal with hormone replacement therapy?" segment from the episode:

I covered this in depth on Wellness, Actually — listen below.

Wellness, Actually Podcast

"What's the deal with hormone replacement therapy?" — Listen to the full episode, including the week's health news and listener Q&A.