“Fifty percent of physicians in the U.S. graduated in the bottom half of their medical school class...”

It's been a little crazy the past several days, so I have taken a brief hiatus from my blog.  I heard a statistic cited the other day by some television news reporter that seemed pretty obvious and maybe even a bit silly.  Apparently, fifty percent of all physicians in the United States graduated in the bottom half of their medical school class.  Really?!?!  I wonder if the television news reporter would be surprised to hear that fifty percent of all physicians in the United States graduated in the top half of their class.  Everywhere you look these days, someone is manipulating statistics to either make a statement or prove some point.  Unfortunately, you have to pay attention carefully so that you don't fall into their trap.

During the 2007 election campaign for New York City Mayor, Rudy Giuliani tried to make a point that his chances of surviving prostate cancer were much better because he lived in the United States rather than in England, which had "socialized medicine."  Giuliani stated, "My chance of surviving prostate cancer - and thank God, I was cured of it - in the United States?  Eighty-two percent.  My chance of surviving prostate cancer in England?  Only 44 percent under socialized medicine."  The New York City Mayor was trying to drive home the point that he was lucky to be living in New York, where his chance of surviving prostate cancer was almost twice as high had he been living in London, England.  The problem with his statement was that it was simply not true. 

According to Dr. Gerd Gigerenzer, former Director of the Center for Adaptive Behavior and Cognition at the Max Planck Institute for Human Development in Berlin, Germany, journalists and politicians aren't the only ones who misquote statistics.  Many physicians (and their patients) don't understand even the most basic concepts of statistics and probability (for a review, see a great article by Gigerenzer here).  So, where did Mayor Giuliani (and many physicians, journalists, and others too) go wrong with his prostate cancer statement?  Apparently Giuliani used data published in 2000 that showed that 49 British men per 100,000 were diagnosed with prostate cancer.  Of these, 28 men died within 5 years.  Similar published results from the United States showed that 136 American men per 100,000 were diagnosed with prostate cancer in 2000 (the number is higher because more screening is conducted in the U.S. compared to England).  Of these, 26 men died within 5 years.  If you use the statistics to calculate a percent survival at 5 years, you get about 82% survival for American men versus 42% survival for British men.  Unfortunately, this is absolutely incorrect.  You should not use incidence statistics to calculate percentage survival.  Giuliani, like so many journalists, politicians, and (unfortunately) physicians today, massaged the numbers to make a dramatic point.

My point is this.  We would all do well to go back and review some basic principles of statistics and probability.  Armed with this new knowledge of statistics and probability can be a powerful thing.  Now (well, at least learning more about statistics and probability), whenever you hear statistics reported on the television, radio, or Internet, you can take a moment to background check the facts.  Don't fall into the trap by accepting everything you hear.  There's a really good chance that what you hear isn't actually correct.