May 2, 2014

Mammography ping-pong

Hilda Bastian at Scientific American

It’s like a lot of evidence ping-pong matches. There are teams with strongly held opinions at the table, smashing away at opposing arguments based on different interpretations of the same data.

Meanwhile, women are being advised to go to their doctors if they have questions. And their doctors may be just as swayed by extremist views and no more on top of the science than anyone else.

She explains where the different views  and numbers come from, and why the headlines keep changing.


Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »


  • avatar
    Megan Pledger

    I find over-diagnosis very confusing.

    It says on wikipedia that “Overdiagnosis is the diagnosis of “disease” that will never cause symptoms or death during a patient’s lifetime.”

    (So if a person gets told they have cancer, has no symptoms but walks out in front of a bus that’s over-diagnosis????)

    Wikipedia says you can’t estimate it because people diagnosed generally get treated but if you have a randomised trial of screening “then a persistent excess of detected disease in the tested group years after the trial is completed constitutes the best evidence that overdiagnosis has occurred.”

    But couldn’t you also say that early detection and treatment has meant that those “excess” are really the beneficiaries of screening? Shouldn’t we be saying about those ones “Yay, screening worked, look how well they survived”?

    3 years ago

    • avatar
      Thomas Lumley

      The idea is that if you wait long enough all the disease in both groups will be diagnosed, so the “Yay, screening worked” people will eventually be balanced out by fatal disease cases in the control group.

      If there is a persistent excess in the screening group even after many years, then these must be people who would never have been diagnosed without screening. That means they would never have developed symptomatic disease.

      There are two kinds of over-diagnosis, though. There is disease that is real but that would not progress before death from other causes (this is the big issue for prostate cancer). Some of these would have been people who just walked under a bus, but if there are systematically more bus-walk-underers in the screening group, that’s evidence for screening being harmful.

      There are also tumours that spontaneously regress. This used to be regarded as very rare, but some recent research based on follow-up of screening trials suggests it might be a lot more common that was thought. The situation is still pretty unclear, though.

      3 years ago