HRT: What the 2002 Study Actually Found

In July 2002, one arm of a large American study called the Women's Health Initiative was stopped early. The headline that went round the world was that HRT increased breast cancer risk by 26%.

Prescribing collapsed — by roughly two thirds within a couple of years. A generation of women went through the menopause untreated, and many are still frightened of HRT because of a study most of them have never read.

It is worth going back and looking at what it actually said.

What 26% meant

It was a relative increase, and relative figures without absolute ones are close to meaningless.

In absolute terms, the combined oestrogen-and-progestogen arm showed roughly eight extra cases of breast cancer per 10,000 women per year. Put another way: a change of well under one tenth of one percent per year.

That is not nothing, and it should not be waved away. But "26% increase" and "eight women in ten thousand" produce completely different reactions, and only one of them was on the front pages.

Who the study actually studied

This is the part that matters most, and it is where the study's findings were stretched furthest beyond what they could support.

• The average participant was 63 years old, and many were more than a decade past their menopause. That is not who typically starts HRT — most women begin in their late forties or early fifties, for symptoms
• Many already had cardiovascular risk factors
• The preparations were oral conjugated equine oestrogens with medroxyprogesterone acetate — not the transdermal oestrogen and micronised progesterone typically prescribed in the UK today

Applying the results to a 50-year-old with hot flushes taking a modern patch was never justified by the data. It happened anyway.

The finding that barely got reported

The study had a second arm: women who had had a hysterectomy and therefore took oestrogen alone, without a progestogen.

That arm showed no increase in breast cancer — and in the longer follow-up, fewer breast cancers than placebo.

It received a fraction of the coverage. But it points at something important: much of the breast cancer signal appears to relate to the progestogen component rather than the oestrogen, which is why the type of progestogen used now matters.

What is understood now

Timing changes the picture

Starting HRT within about ten years of the menopause, or before the age of 60, appears to carry a different — and more favourable — cardiovascular profile than starting it fifteen or twenty years later. The WHI largely tested the latter.

How the oestrogen is taken matters

Oestrogen through the skin — patch, gel or spray — does not carry the increased clot risk that oral oestrogen does, because it bypasses the liver. This is a genuinely significant difference and it means transdermal HRT is suitable for many women who would otherwise be told they could not have it: those with migraine, raised BMI, or a history of clotting risk factors.

The type of progestogen matters

Micronised progesterone appears to carry a lower breast cancer signal than the older synthetic progestogens used in the WHI. It is body-identical and is now widely used in the UK.

Putting the risk in proportion

The breast cancer risk associated with combined HRT is real, and it is broadly comparable in magnitude to several everyday factors — drinking a couple of units of alcohol daily, carrying excess weight after the menopause, or being physically inactive. None of those are treated as reasons for alarm, and none receive a fraction of the attention.

The point is not that the risk is trivial. It is that it belongs in a list with other risks, weighed against benefits, rather than standing alone as a verdict.

What HRT is genuinely good at

• Vasomotor symptoms — hot flushes and night sweats. It is by a wide margin the most effective treatment available
• Sleep, mood and cognitive symptoms during perimenopause, which are often what women find hardest and least often connect to hormones
• Bone protection. HRT reduces fracture risk, and this is well established. It is particularly relevant for anyone with early menopause or osteoporosis risk
• Genitourinary symptoms — dryness, discomfort, recurrent urinary infections
• Early menopause under 45, and premature ovarian insufficiency under 40, are a different category altogether. Here HRT is replacing hormones a woman should still have, and the risk conversation runs the other way — not taking it carries the risk

Vaginal oestrogen is a separate conversation

This deserves emphasising because it is so often lumped in with systemic HRT and refused on that basis.

Vaginal oestrogen — a small dose applied locally — is very poorly absorbed into the bloodstream. It does not carry the risks discussed above, it can be used long term, and it can be used alongside systemic HRT or on its own.

It treats dryness, discomfort, painful sex and recurrent urinary infections, and it is markedly under-prescribed. Many women with a history of breast cancer can also use it after discussion with their oncology team — that is a conversation, not an automatic refusal.

Who should not take systemic HRT

• Current or past breast cancer, or another oestrogen-dependent cancer — which needs a specialist discussion rather than a flat no
• Undiagnosed vaginal bleeding — which must be investigated first
• Active liver disease, or a current blood clot
• Untreated high blood pressure, until it is controlled

Two practical points: HRT is not a contraceptive, and contraception is still needed for two years after the last period under 50, or one year over 50. And over 45, the menopause is diagnosed on symptoms, not blood tests — hormone levels fluctuate too much to be informative.

What to expect if you start

Give it three months before judging. Early breast tenderness, bloating and some irregular bleeding are common and usually settle. Dose and route often need adjusting, and needing a change is normal rather than a sign it is not working.

Testosterone is sometimes added, off-label but with specialist guidance, where low libido persists despite HRT.

The point of all this

Not that HRT is risk-free — nothing effective is. The point is that a decision this consequential deserves the actual numbers, the study's actual population, and the twenty years of evidence since — rather than a headline from 2002 that was misapplied at the time and has been quietly corrected ever since without anyone announcing it.