This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
To the Editor:—
The article titled "Conditional Probability Program for Diagnosis of Thyroid Function" by Overall and Williams (JAMA183:5 [Feb 2] 1963) was timely and well written. I believe, however, it should be indicated that, in clinical application, such procedures may function as useful "aids" to the practicing physician in focusing his attention on the more probable diagnoses.The emphasis on the word "aid" is necessary for two reasons. First, the practicing physician has a natural and proper reluctance to accept any system in which his diagnostic acumen is apparently replaced by a computer. This misapprehension can delay acceptance of these methods in proper context. The second aspect is a serious functional one which I have encountered in application of a somewhat similar method to the control of industrial work groups. The problem in this case is the result of a phenomenon which we call "depersonalization." That is
Nertney RJ. Computer Procedure as Aid in Diagnosis. JAMA. 1963;185(4):329. doi:10.1001/jama.1963.03060040113045