Let's deal with the bogeyman first.
If you are going through menopause and feeling awful, hormone therapy is probably safe for you and it probably works. The fear around it comes from one study reported badly in 2002, using drugs we no longer prescribe. If you have symptoms and you are otherwise well, it is worth sitting down and talking about.
Almost everything people believe about HRT comes from one study, published in 2002, reported badly. The study was real and it was worth doing. The reporting was not. Twenty-four years on, the drugs are different, the question that study asked was the wrong one, and the numbers were never as frightening as the headlines made them.
Thirty years ago HRT was everywhere. Women felt bad — weak, lethargic, brain fog, losing bone, losing muscle, ageing. GPs, gynaecologists, everybody was prescribing it. Single pill, multiple pill, oestrogen, progesterone, every combination. And there was no hard evidence either way. Nobody could say whether it worked or whether it harmed.
So it was decided, driven mainly by the NIH, to run a large study. That became the Women's Health Initiative. It is still one of the largest and most expensive studies ever done — close to a billion dollars, published in 2002. The question was simple: is HRT good for women?
Three things. None of them makes it a bad study. All of them make it the wrong study to frighten a 52-year-old with hot flushes.
These were not healthy everyday women starting HRT for menopausal symptoms. Many were older, heavier and already at high cardiovascular risk. That was never meant to be the target population.
The trial was designed primarily to test whether hormones could prevent chronic disease. That was the wrong question. The question that matters in clinic is whether hormone therapy treats menopausal symptoms, and whether it does so safely.
It tested conjugated equine oestrogen with synthetic medroxyprogesterone acetate. Not transdermal estradiol and micronised progesterone, which is what we use now and which is safer. Oral oestrogen has a much greater first-pass effect on clotting factors and triglycerides, and a higher clot risk than transdermal. We have largely stopped using it.
This was the first medical study to hit the front pages in a big way, and the public had no idea of the difference between relative risk and absolute risk. The headlines were too good to pass up. "30% increased risk of cancer." True, and close to meaningless on its own.
Here is what those percentages actually meant, per 10,000 women per year.
| Outcome | No HRT | On HRT | Difference | Headline |
|---|---|---|---|---|
| Breast cancer | 30 | 38 | +8 | +26% |
| Coronary heart disease | 30 | 37 | +7 | +29% |
| Stroke | 21 | 29 | +8 | +41% |
| Pulmonary embolism | 8 | 16 | +8 | +113% |
| Colorectal cancer | 16 | 10 | −6 | −37% |
| Hip fracture | 15 | 10 | −5 | −34% |
Women's Health Initiative, combined oestrogen and medroxyprogesterone arm, 2002. Rates per 10,000 women per year.
Eight extra breast cancers per 10,000 women per year. That is the 27% headline. Seven extra heart events. Eight extra strokes. The embolism figure doubled, and that was with oral conjugated equine oestrogen, which is precisely the preparation we have abandoned.
And the good half never made the front page. Colorectal cancer down. Hip fracture down by a third — in a disease where mortality after an osteoporotic hip fracture runs near 40% at two years.
There were two arms. Combined oestrogen plus medroxyprogesterone, and oestrogen alone in women who had had a hysterectomy. They gave different answers, and that difference is the useful part of the whole study.
That narrows the problem considerably. Breast cancer risk is not automatically attributable to oestrogen. Synthetic medroxyprogesterone is probably the bigger contributor. Later follow-up went further: oestrogen-only therapy after hysterectomy showed lower breast cancer incidence and mortality.
The later analysis by age mattered too. Women starting hormone therapy between 50 and 60 had around 30% lower all-cause mortality. Between 60 and 70, no real difference. Between 70 and 80, slightly worse. Timing is most of the argument.
All of this came afterwards. By then the damage was done.
Better drugs. We have bioidentical hormones now. Bioidentical does not mean organic or natural — they are mostly synthesised from yam. It means the molecule closely mimics the biochemical structure of the hormone the human body makes. Most older HRT was synthetic chemistry that carried out the action of a hormone without matching its structure. We have dropped most of the preparations that caused the trouble.
Longer lives. Women are living longer. Thirty or forty years after menopause is a long time to spend feeling unwell, losing bone and losing muscle. Quality of life over that span weighs more heavily than it did in 2002.
Whether any of this makes a woman live longer, the evidence does not say. Hormone therapy is a quality-of-life treatment and a prevention treatment — bone, and urinary infection. I would not claim more than that.
Where it earns its place is the symptomatic woman, the frail woman, the woman losing muscle and losing bone. There, hormone therapy has something real to offer.
This is what I do, not a published guideline and not a recommendation for anyone else. I believe in screening, because the cancers we cure are the ones we find early, and we have not changed cancer deaths as much as people assume.