Sensitivity, Specificity, and Predictive Values in the 'Sensitivity and Specificity of Clinical Diagnostics'

Abstract

To the Editor.— article entitled The Sensitivity and Specificity of Clinical Diagnostics During Five Decades: Toward an Understanding of Necessary Fallibility1fueled a spirited discussion in our journal club. For this I heartily congratulate the authors. article splendidly illustrates some problems with the medical literature. authors introduced an interesting idea that could have been the subject of a brief essay or op-ed, but the idea was presented as a scientific article complete with data and Methods, Results, and Comment sections. result was turgid pseudo-science, posing as meta-analysis, which added nothing to the body of knowledge of science and contributed little to its understanding. studies, directly compared, ranged from autopsy rates of 96% to 23% to unknown; clinical diagnosis ranged from collaboration between clinician and pathologist to (presumably) routine completion of death certificates; the studies span the Atlantic; and no information is given about the