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:—
A paper by Charles E. Bender, MD (199:529, 1967) evaluates the effect of treatment with corticosteroids on the outcome of infectious mononucleosis. This was done by comparing patients preselected for such treatment with those receiving no steroid, on a retrospective basis. An attempt was made to remove extraneous differences by selecting an untreated "control" patient for each steroidtreated patient on the basis of similarity in age, sex, disease status, and time of "treatment."It is most difficult to establish comparable groups in this way. Inevitable uncontrolled differences are not taken into account. Moreover, the author says:The usual approach to an assessment of drug effect by assignment of patients to treatment and control groups was inappropriate because of the well recognized dangers of steroid therapy and because many students treated by the Student Health Service are minors.This seems almost to guarantee the existence of differences
Chalmers TC, Muench H. Corticosteroids in Infectious Mononucleosis. JAMA. 1967;201(1):64-65. doi:10.1001/jama.1967.03130010090027