This site is intended for health professionals only

At the heart of general practice since 1960

Kendrick breast cancer screening pointless?

Our clinical columnist, Dr Malcolm Kendrick, challenges another sacred cow ­ this week he argues mammography screening may be a waste of time and effort

Our clinical columnist, Dr Malcolm Kendrick, challenges another sacred cow ­ this week he argues mammography screening may be a waste of time and effort

There is something irresistibly comforting about screening for cancer. Pick up an early stage tumour, remove it, and you have probably saved a life.

This is clearly the sort of thing that a health service should be doing. It also seems like pure common sense. A stitch in time saves nine ­ and a few lives.

What makes it even more appealing is that it reinforces a caring, 'parental' role for the medical profession. We can watch over the people, look after them and protect them from harm.

And for the public the idea of being protected from dying of cancer appeals at a very deep, visceral level. They are being looked after and protected by wise people with the very latest technology. This is an activity that makes people feel safe.

So there is a very powerful emotive appeal underlying breast screening. Doctors like it, patients love it, the Government can hold up screening as one of the many wonderful things it is doing with taxpayers' money.

It's a win-win for everybody involved.And just consider the alternative ­ a return to a frightening world where breast cancer can just appear, at some late stage, and all we can do is the best we can with whatever treatments are available.

But does the evidence support widespread screening? For example, in 2001 two Danish researchers, Olsen and Gotzsche, reviewed all breast screening trials done up until that time and concluded there was no reliable evidence that screening reduced mortality, in a research letter on their Cochrane review in The Lancet1.

An internet report on their research highlighted the finding that screening led to greater use of more aggressive treatment, rather than 'increased treatment options'2.

Their study was roundly condemned for various reasons. But however you interpret the data, the benefits seem vanishingly small.

For instance, one year after Olsen and Gotzsche published their results, a group of researchers led by Nystrom in Sweden claimed to have proved that breast screening was, in fact, worthwhile3.

This group analysed data from four randomised mammography trials in Sweden up to and including 1996. These trials included information for 247,000 women based on records from the Swedish Cancer and Cause of Death Registers.

The median follow-up time for the trials was 15.8 years. No benefits were found in younger women ­ significant effects were only seen in women over the age of 55.

Overall, there was a 21 per cent reduction in death from breast cancer, which does sound pretty good. But this was a relative risk reduction.The absolute figures were somewhat less impressive.

There were 511 deaths from breast cancer in 1,864,770 women-years in the groups of women who had mammograms versus 584 breast cancer deaths in 1,688,400 women-years in the control groups (those who were not invited for screening).

In short, 73 fewer deaths from breast cancer as a result of nearly two million years of screening (1,864,770 women-years). Which is one death delayed for every 26,000 women-years of screening.

I find it hard to get overly excited about that result. Twenty-six thousand years takes us well into the early Paleolithic age.


Perhaps this would be worth it if there were no downsides to screening. But there are many ­ and the most significant is the rate of false positives. The published figures seem to vary from a 0 per cent rate to 20 per cent or even higher. Who knows what the true figure is?

What we do know is that very many more women have breast 'cancers' than will ever die of breast cancer. And it has become increasingly apparent that a very high proportion of 'cancers' do not grow or become invasive. They just sit there.

Professor Gilbert Welch from Dartmouth Medical School, New Hampshire in the US, talked about the discovery of small breast cancers in a radio programme4: 'Now what's happening to these small cancers? Well they either never grow at all, or maybe even get smaller with time. Or they grow so slowly that people die of something else before the cancer ever causes symptoms.'

That's the real conundrum with cancer screening. If you try to find cancer early, all of a sudden you find a lot more than anyone ever thought existed.

And that's one of the things we're beginning to learn as we look harder for cancer, that there's an awful lot of cancer out there that we never knew existed.

'About a third of all adults will have some pathologic evidence of thyroid cancer; about 40 per cent of women in their 40s will have microscopic evidence of breast cancer, the so-called ductal carcinoma in situ ­ that comes from an autopsy study of women who have died from something else.'

One in 15 women dies of breast cancer. Some 40 per cent of women have microscopic breast cancers, and if all of these could be found on screening, what would we do?

This highlights the main problem with screening, which is that we do not actually understand the natural history of cancer. Many men will develop prostate 'cancer', but most will never die of it.

Many women develop breast 'cancers' and yet most of these will never grow, and might well regress. Screening assumes a model of cancer growth that starts with a small 'error' in cell transcription that will then steadily grow and invade local tissues before metastasising.

But this clearly does not always happen.Breast screening, or any screening, is only going to be truly effective once we know more about what we are actually screening for.

Governments rushed into breast screening using some pretty flimsy data, and it may now be completely impossible to extricate ourselves from the model that we are using.

But I believe that ­ although it is tough and painful, and most people want to sweep all debate under the carpet ­ we need to question the value of breast screening. Because a strong possibility exists that it may actually be pointless ­ at best.

Malcolm Kendrick is a salaried GP in south Manchester.

Rate this article  (4 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say