This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
Up to a decade ago, descriptions of clinical investigations frequently suffered from a failure to apply statistical methods. Investigators have since become so sensitized to the value of biometrics that nowadays rarely does an article appear without its standard errors and chi squares duly recorded. Unfortunately, at the same time there has been a tendency to overlook the fact that mathematics is no substitute for common sense. If the premise from which an investigation starts is false and the observations inaccurate, even the most painstaking calculations will lead to erroneous conclusions. A possible example of this is provided by the recent epidemic of articles dealing with the distribution of the blood groups in various diseases.
More than 20 years ago studies on the distribution of the ABO blood groups in disease enjoyed considerable vogue. At that time, the blood groups were studied in such conditions as dementia precox, malaria, general
Wiener AS, Wexler IB. BLOOD GROUP PARADOXESGUEST EDITORIAL. JAMA. 1956;162(16):1474–1475. doi:10.1001/jama.1956.02970330046013
Customize your JAMA Network experience by selecting one or more topics from the list below.