Typical errors in medical studies
A recent article by Dr. Robert H. Shmerling, M.D., Beth Israel Deaconess Medical Center, on the Intelihealth® web site, deals with the issue of errors in medical studies. Dr. Shmerling lists three types of errors. The first type, results from lack of knowledge of all the "confounding factors" (factors other than the object of the study, that impact patients' health). The second, is a result of the finite (relatively short) duration of studies. The third is due to the way questions are posed, that often tend to favor the company (pharmaceutical?) that is funding the study.
Add these errors to the fact that one out of every twenty medical studies is wrong because of the statistical errors in such studies, and we have good reason to be careful about the initial hype we here when new results are released.
For more information, please see our detailed writeup!
Last Modification - November 20, 2005