Tuesday, November 17, 2009

The Mammography Controversy – Interpreting Medical “Facts”

The practice of medicine is a game of probabilities.  While often (but not always – but that is the subject of a separate discussion) based on scientific fact, those facts themselves are always dynamic and a good physician is always trying to predict the future which while based on facts is ultimately unknowable.  Will the illness get better or worse?  Will this medication or treatment make the patient improve or will the side effects and risks of the therapy actually create more problems?  Will this screening test discover an illness that can be cured or will it actually lead to further tests and therapies that are, in and of themselves, dangerous?  All of these questions are asked about future events and the unique, individual nature of each of us means that medical “facts” are sometimes not facts at all, but educated guesses based on the interpretation of studies.  Even when dealing with population wide public health issues, the “facts” from studies and from expert panels cannot be divorced from simple laws of statistics and probability and are rarely incontrovertible.  These same facts when read by different people can also be influenced by the most human of emotions and even by political desires. 

Screening Mammography

We see that in the political firestorm that has erupted over the US Preventive Services Task Force (USPSTF) recommendations that were announced this week that state that screening mammography should only be performed on women over 50 and only once every two years.  This is different from the standard recommendations which are still supported by the American Cancer Society which recommend mammography every year for all women over 40.  What made the USPSTF take this step?  Let’s look at what makes a good screening test.  Any screening test is limited by its sensitivity and specificity.  They are:

  • sensitivity = probability of a positive test among patients with disease
  • specificity = probability of a negative test among patients without disease
  • If a test is sensitive but not specific, you will have many false positives.  In medicine a false positive often means great anxiety in the patient and may also mean invasive procedures (such as breast biopsies and even breast surgery) which puts people at risk.  In this case the USPSTF looked at all the studies that have been done concerning screening mammography, including recent studies that specifically looked at women aged 40-50 and determined that there were enough false positives in this group that the risk of unnecessary procedures outweighed the benefit of finding a breast cancer at an early enough stage to be curable.  The American Cancer Society (ACS) and some of the other organizations that are criticizing this move are looking at the same data and coming to a different conclusion.  They acknowledge the risk but believe that the benefit outweighs the risk.  My own approach is to give all these facts to women and let them decide.  Most women are smart enough and know their own ability to handle risk well enough to take in these various recommendations and make their own decisions. 

    The Insurance Effect

    The difficult decision is now with the insurance carriers who must decide whether to set their benefits to encourage a yearly mammography for women over 40 or an every other year mammography for women over 50.  Most companies follow the recommendations of both the American Cancer Society and the USPSTF.  They rarely disagree and when they do, there is a definite dilemma.  My own recommendation is to pay for the test based on the ACS but only to encourage it in materials and mailings based on the USPSTF.  I believe the studies done that are cited in the new recommendations are strong and sound but I also believe that breast cancer is a disease that causes fear in families and that fear is very real.  Many women may want the reassurance of earlier and more frequent mammograms even if that means they will be increasing their risk from unnecessary tests and even potential unnecessary surgeries.  I would not make it more difficult to have those tests by creating new financial obstacles.