NHS Breast Cancer Screening Has Major Issues

[Zaune]

The Guardian

Breast cancer screening causes more harm than has previously been recognised, even though it saves lives, according to an independent review set up following years of scientific controversy surrounding the NHS programme.

Around 1,300 lives are saved every year by mammography, which women are invited to undergo between the ages of 50 to 70, said the review, which recommends that screenings should continue.

But 4,000 women will undergo unnecessary treatment, including surgery, radiotherapy and chemotherapy, for a cancer they would not otherwise have known about and which would have done them no harm in their lifetime. Some breast cancers are so tiny and slow growing that they would never be a threat to a woman's health, the review says.

The government is embarking on an immediate revision of the leaflet which invites women for screening, said national cancer director Sir Mike Richards, so that women can weigh up the benefits against potential harm and make their own decision as to whether to be screened.

While he welcomed the review panel's support for screening, he added: "The key thing is that we communicate this new information to women so they can make an informed choice for themselves.

"NHS Cancer Screening Programmes have already asked independent academics to develop new materials to give the facts in a clear, unbiased way. I hope to see them in use in the next few months. If any woman has concerns about breast screening she should talk to her GP or health professional."

The independent review panel was set up by the government and the charity Cancer Research UK under the chairmanship of Sir Michael Marmot and comprised scientists who had never published on breast screening before, in order to establish their impartiality on an issue that has provoked impassioned argument between epidemiologists.

The leading critic of screening is Professor Peter Gøtzsche, director of the Nordic Cochrane Centre, whose team has spent years analysing the data from the trials carried out in several countries prior to the setup of national screening programmes – as well as more recent studies. Gøtzsche says the harm of screening outweighs the benefits.

Marmot's team also reviewed data from original breast screening trials, many of which took place in Scandinavia. They acknowledged that they had limitations because they mostly date from the 1980s or earlier – the NHS programme was set up in 1988. But they dismissed other criticisms, such as that some of the women had died of other diseases and not breast cancer.

The review judged that screening reduces the risk of dying from breast cancer by 20%. It prevented 43 deaths for every 10,000 women invited to be screened, or one death per 235 women invited. Among those 10,000 women, 681 cancers would be diagnosed and 129 of those would be over-diagnosed – the mammogram would have picked up an otherwise undetectable tumour which the woman would never have known about. As with many prostate cancers in men, it would grow so slowly that without treatment she would die with it, rather than of it.

"Our best estimate is around 4,000 cancers a year which is about 19%," said Professor David Cameron from Edinburgh University, who presented the findings which are published today in the Lancet medical journal.

"Breast cancer screening should continue, but we felt that the information given to women needs to be reviewed with our findings taken into account, so women are given an accurate picture of the benefits and harms."

The current NHS leaflet mentions misdiagnosis almost in passing. "Screening can find cancers which are treated but which may not otherwise have been found during your lifetime," it tells women.

Marmot said the information must now be available for women to make a choice. "Clear communication of these harms and benefits to women is essential, and the core of how a modern health system should function," he said.

The government and the NHS screening programme will be hoping that the controversy will die away, but Gøtzsche is unlikely to change his opinion. The review panel was wrong to assume that the effect of screening is the same today when we have much better treatments than in the original screening trials, he said, and the panel has also underestimated overdiagnosis substantially.

Although he is pleased that the panel recommended giving women better information, he does not accept that screening reduces the risk of dying from breast cancer by 20%. From more recent observational studies of ongoing screening programmes, he says, "we cannot see any effect of screening".

Much has changed since the early days of screening. Most important, the numbers of advanced cancers have not dropped as a result of screening, he says - which was the main reason for setting up the programmes. "Screening has not reduced the number of advanced cancers per 100,000 women, and when that's the case, screening cannot work," he added.But most worrying is that the numbers of advanced cancers, which carry the greatest threat of death, have not dropped as a result of screening, he says - which was the main reason for setting up the programmes. "Screening has not reduced the number of advanced cancers per 100,000 women."

Three breast cancer charities in a joint statement said they urged women to go for screening. Breakthrough Breast Cancer, Breast Cancer Campaign and Breast Cancer Care said the review had provided much needed clarity. "This is good news for women as they can now be assured that breast screening can be beneficial," they said. "However, some women who attend screening may be diagnosed and treated for a cancer that may not have caused them harm in their lifetime. To ensure women understand what this may mean for them it is important they have access to clear and balanced information."

Now that is not a happy number...

[Terralthra]

This has actually been known for a while. The number needed to treat (NNT) on mammograms is very high - meaning that you need to give many women mammograms before finding a woman with breast cancer for whom the mammogram made the difference between her dying of cancer and her not. On the other hand, mammograms themselves are radiation, which can cause cancer, and with the amount of treatment a false or neutral positive test gets, the NNT on negative interactions and side effects is substantially lower.

It's similar to the argument against prescribing antibiotics for strep throat to prevent rheumatic fever.

If you're interested in this sort of issue, I highly recommend Hippocrates' Shadow.

[mr friendly guy]

Why don't they adjust guidelines to better identify which cancers aren't likely to be problematic.

[Broomstick]

Last I heard (back in the 1990's, when I was working with medical researchers) the problem was that they couldn't be sure which cancers would be aggressive and which were safe to leave alone. At that time, more research was needed to obtain the knowledge to make such distinctions.

Has there been sufficient such research in the last 10 years or so? I don't know.

[lance]

From what I heard about a 1/4 of breast cancer is caused by mammograms.

[Broomstick]

Cite, please?

Granted, mammograms involve x-rays, but that seems an awful high percentage of induced cancers.

[lance]

It was from my professor for radiology, and its possible I miss heard him, or that's a high end don't suck at your job number.

[Broomstick]

You also have to account for the fact that with each decade the radiation dose from a given type of x-ray usually goes down. This is counter-acted by things like CAT scans, which are many multiple x-rays taken at once. Frequent CAT scans are not only associated with cancer a couple decades out, but there have been instances of people suffering radiation sickness from too-frequent CAT scans.

Nonetheless, the average dental x-ray, bone-fracture diagnostic x-ray, and yes, the average mammogram delivers significantly less radiation to the patient than a quarter century ago. Any study would have to allow for those differences.

[lance]

I think the switch to digital increased the amount of radiation exposure for a given x-ray with the trade off of not needing repeat exposures

[Broomstick]

Really? Because I've heard the exact opposite.

Not that it matters for people in my area - ALL the testing centers have gone digital around here.

[lance]

I'll see if I can dig up the chart up, but its the big trade off of higher exposure for a much higher level of contrast. Like the use of grids to absorb scatter may require using 5 times the amount of exposure.

[Broomstick]

On the other hand, the media capturing the image (film, then electronics) are getting more sensitive with passing time, therefore needing less radiation than before to produce the same resolution of image.

But yes, if you can, provide some hard data. I'd be interested.