Is This the End of Mammograms to Screen for Breast Cancer?


Internet discussions usually descend into abuse within a few postings. Whether this is because people these days are less polite and restrained than they used to be, or because the internet allows them to publish their first reactions without the time to cool off that older means of communication entailed, I do not know; but the fact is that those who take part in such discussions seem to confuse insult with argument and are seldom able to keep to the point for very long.

Doctors, if the internet discussions that follow articles that appear in the New England Journal of Medicine are anything to go by, are better than average. Often, indeed, though not always, they employ rational argument. Perhaps there is something to be said after all for a long and rigorous education.

Not long ago there appeared an article in the NEJM throwing doubt on the wisdom and even the ethics of screening by mammography for early cancer of the breast. A Swiss commission examined the evidence for the benefits and found none, or some so slight that it hardly counted. The best estimate they could come up with was that mammography saved one life from breast cancer per 1000 women aged 50 who underwent screening, but that the all-cause death rate in the screened and unscreened was practically the same. Not only did many screened women undergo unnecessary operations and much radiotherapy for false positive results, but those who offered the screening were little more honest than second-hand car salesmen, in effect preying on the women’s complete misunderstanding of the supposed benefits of screening. Women believe that even with screening they are twenty times more likely to die of the disease than in fact they are; and they overestimate the benefits of screening by eighty times.

In the circumstances, then, the Swiss panel suggested that screening by mammography should be stopped forthwith.

The discussion that followed by doctors around the world was, by the low standards prevailing on the internet, rational and even-tempered. No one suggested that the Swiss commission was corrupt or actuated by discreditable hidden motives, a strong favorite of participants in many internet discussions.

A psychiatrist, for example, pointed out that the authors of the article in the NEJM believed that screening saved one woman per thousand from death by cancer of the breast at a cost of over-diagnosis of tumors and hence of over-treatment of other women. What he missed in the article was any philosophical consideration of how the benefit of screening might be weighed against the harm. Perhaps the over-treated women would be happy to accept their unnecessary treatment if they knew it had helped to save the life of another woman. This would have been an entirely reasonable point, had the psychiatrist not overlooked the fact that all-cause death rates after screening are more important than death rates from the screened-for disease: unless, that is, it is particularly important for some reason not to die of the screened-for disease rather than from any other cause. This is not inconceivable – I would rather die aged 80 of a heart attack than of the consequences of Parkinson’s disease – but it is not self-evident in this case either.


Several doctors weighed in with statistics, both in favor and against mammography. Those against seemed to me more convincing than those in favor. The only completely irrational comment came from a doctor with prostate cancer. He wrote:

As a person who had his prostate cancer diagnosed purely from a rising PSA with a negative sononography and prostate exam, I take exception to these articles telling us to curtail screening exams

This is no better than the patient who refuses to give up smoking because he had an uncle who smoked 80 cigarettes a day and lived to be 99. But which of us has never resorted to such an argument?


images via shutterstock / : Blue Planet Earth / Tyler Olson