The May 22, 2014 issue of the New England Journal of Medicine has a perspective piece that is even more instructive than its authors and the journal’s editors realized: ‘Abolishing Mammography Screening Programs? A View from the Swiss Medical Board’. It describes the process that the Swiss used to evaluate the scientific foundation for the recommendation that women undergo routine mammography screening. The authors then relate the reaction that followed their recommendation that the practice be discontinued.
The figure below summarizes Biller-Andorno and Jüni’s findings; they are the authors of the piece. Panel A show the views of 50 year old women in the US. These women greatly overestimated the likelihood of dying from breast cancer over a 10 year period and they similarly overestimated the beneficial effect of screening mammography. Panel B show the real effect of mammography. The difference between perception and reality is striking. Biller-Andorno and Jüni’s conclusion that screening mammography was of little or no benefit was based on an exhaustive analysis of all the data regarding its effectiveness.
It was clear right from its inception that screening mammography was based on wishful thinking rather than good data proving its effectiveness. It was one of medicine’s continuous stream of unanalyzed good intentions. Every physician wants to prevent death from breast cancer. Early detection of the disease by screening mammography should result in the cure of cancers that if detected later might prove lethal. What could go wrong? So let’s not wait for data supporting our supposition of benefit. Let’s do it. And so we did.
Forty years ago those who said that the wide spread use of screening mammography was a case of ready fire aim were derided and ridiculed. Oncologists told the public that epidemiologists didn’t understand clinical medicine and not to pay attention to their nay saying. Of course, wishful thinking may come true. But in this instance there has been an avalanche of data over the past decade that this screening test is ineffective and that any benefit it might provide is at least balanced by the harm that follows overdiagnosis of breast cancer. I’ve written on this subject several times. I won’t comment further on this issue. Rather, I wish to examine what happened after the Swiss board made its report.
They “recommended that no new systematic mammography programs be introduced and that a time limit be placed on existing programs.” The Swiss oncologists exploded in outrage. You must be aware that oncologists as a group are almost immune to the results of epidemiological analysis. This is an opinion based on 50 years of close observation of their reactions to data that go against their preconceived opinions.
Pay very close attention to the following quotation from the NEJM paper. “One of the main arguments used against it [the recommendation to stop screening mammography] was that it contradicted the global consensus of leading experts in the field.” Does that sound familiar? If there’s a consensus among the experts the science must be settled. Science is never settled! It’s always subject to revision or even to the total destruction of previous theories and interpretations. Science is dynamic and not settled by a vote. At the conclusion of the 19th century physics was held by the world’s experts to be complete except for a couple of minor loose ends. These loose ends turned out to be relativity and quantum mechanics.
Any time you hear someone say the science is settled you can be sure that it’s not. If it were no one would need to make such a declaration. For example, it’s settled science – the world is round. Or, it’s settled science – the earth revolves around the sun. All the experts so agree. So does your taxi driver and every high school student which is why no one bothers to say that these facts are settled. Important issues deserve serious examination and vigorous debate. Anytime a politician or journalist tries to stifle this debate you’re hearing from the dark side. John Stuart Mill said that the truth could only be approached from the result of serious argument about whatever was at issue. And the truth, at least as conceived by human beings, is always provisional.