TO SCREEN OR NOT TO SCREEN?

 

CANCER SCREENING: WHEN IS IT WORTHWHILE?

A Michigan Fall

She loved dry leaves

Essentially the entire issue of the November 27, 2012 NCI Cancer Bulletin is devoted to a discussion of cancer screening.  As both of the groups I try to help are engaged in just that, I read it with interest (and more than my usual comprehension.)  It is really good stuff, and (as always) I urge you to read it.  However, I know you won’t, so I will toss out a few observations and maybe some analysis.  I am tempted to weave a few egregious lies into some future blog, to see if anybody notices and checks.  I promise not to do it this time, though.

First, there is a long piece by Dr. Virginia Moyer, who is the head of the USPSTF: that is, the United States Preventative Services Task Force.  This is the group that has caught so much flack over their recommendation that healthy men not be screened for prostate cancer using PSA.  (All males over 60 will know very well what that is.)  To get current with the Prostate Cancer Wars, re-read the blog of the same name (3/27/12) and especially the appended Comments. 

It turns out that the USPSTF is an all-volunteer group of 16 and is substantially independent of control by government or anything else.  They were set up in 1984 to counter the perception that the medical community was neglecting prevention for cure.  They make recommendations for preventative measures, including screening.  They have had their bad days; one especially bad occurred in 2009, when they issued guidelines for mammography.  They recommended that women under the age of 50 – and over 75 – be not screened at all.  The AMA went ballistic.

Dr. Moyer does a good job of explaining their reasoning.  Any screening regimen involves false negatives.  False negatives engender anxiety, discomfort, and expense.  In the case of cancer, there are some which are easily cured – and others that are incurable.  It doesn’t make sense to screen for these.  Pancreatic cancer is a case in point; there is no viable treatment.  For the rest of cancers, screening might save lives, and whether to screen or not is a judgment call.  I reason that ovarian cancer is one of these.  If caught early, it’s completely curable; if later, rarely so.

The problem here is that one does not know how to value a few lives saved relative to a whole bunch of needless treatment and anxiety.  I tend to think that one live saved easily balances a mountain of anxiety, discomfort and expense.  But that’s just me. 

Another great article concerns the role of statistics.  You know Mark Twain’s famous analysis of statistics.  Well, it seems still to be at work.  Some statistics show that the life-span of cancer victims from diagnosis to death has lengthened dramatically in recent decades.   Good news.   But there is a problem: does this represent better treatment, or is it simply an artifact of screening?  For instance, assume treatment for cancer X is totally ineffective.  Persons not screened will be diagnosed later than persons in the screening arm of the trial.  Thus, the diagnosis-to-death interval will be shorter for the first group than the second, and it will appear the “progress is being made”, when in fact it hasn’t.  In fact, the screened group is less well off in that they had to live longer with anxiety.  So, I guess that screening is useful only for those cancers that are easier to cure in an early stage.  Ovarian is one of these.