Friday, September 9, 2011

Medicine and the Media - Once Again

The accuracy of medical news stories included in almost every nightly network news broadcast is questionable.  The recent suggestion that the blood test for c-reactive protein, CRP, should be requested by everyone over forty as a screening test for blood vessel disease that could lead to a heart attack is the insignia of a 90-second news report glossing over the complexity of an issue and misleads by presenting, as fact, speculation and unproven hypotheses.  

It is true that high blood pressure and obesity are risk factors for heart disease, heart attack and stroke.   Although the association of two things is not proof of causality, there is a great deal of literature supporting the hypothesis that lowering blood pressure reduces the risk of a heart attack, and diet and exercise appear to have a similar effect.  It's not clear that the same is true for CRP.  Much work has to be done to validate any claim that lowering an elevated CRP will reduce a patient’s risk of heart disease.  There haven't been any studies that establish how you can reliably reduce an elevated level of CRP.  Some life style changes have been suggested but none has been rigorously tested and proven to be effective.

I hope Brian Williams doesn’t think I am picking on him.   But the story he and his colleague Dr. Nancy Snyderman recently presented on screening for disease ignored some very important points that were not even mentioned in the story. Testing for disease in healthy individuals who have no symptoms or history suggesting an increased risk for the disease under study is known as population screening.  If a disease is relatively rare and you screen everyone, the vast majority will be negative and the cost per case identified can be staggeringly high.  Since no test is accurate 100% of the time, a certain number of individuals who do not have the disease will have a positive test result and a certain number who have the disease will have a negative test result.  The false positive and false negative rates of any test are crucial to the decision to use it as a screening test.  

 Those with no disease who test positive - the false positives - will undergo further testing which may be expensive and could involve the risk of significant complications.  The false negative individual will be reassured and not pursue additional testing when, in fact, they have the disease or risk of the disease in question.

I realize this is complicated statistical stuff.  But for the men who are reading this, consider the example of the dilemma of screening for prostate cancer using the blood test for prostate specific antigen, PSA.      Simply summarized, how many men who have been found to have an elevated PSA have undergone a prostate biopsy because of an elevated PSA and been found not to have prostate cancer?  These biopsies are not without risk and are quite expensive - involving a urologist’s fee for doing the procedure, a pathologist’s fees for reviewing the microscopic slides of the tissue biopsied and the laboratory charges associated with the processing of the specimen.   Furthermore, it is not reassuring to learn that many men who have PSA results within normal limits actually have prostate cancer.   The rationale for screening for prostate cancer is the high fatality rate among men who are found to have the disease and the relatively high frequency of the disease.  But since watchful waiting is one of the accepted options for men found to have prostate cancer, it's important to evaluate the strategy for screening in each individual patient in light of exactly what would be done based on the test results.

With apologies to Mr. Williams, I present this discussion to demonstrate how misleading it is to attempt to inform the public about a complex issue sandwiched between advertisements for the treatment of  bladder control and erectile dysfunction. 

No comments:

Post a Comment