We appreciate the letters of Dr Granger and of Mr Demers in response to the Committee on Infectious Diseases (COID) statement on ribavirin in the January 1996 issue of Pediatrics.1
Dr Granger expresses concern about the use of cohort studies to assess clinical effectiveness of ribavirin therapy. The potential for nonrandom distribution of severely ill patients to treat versus no treatment (or placebo) groups in cohort studies was acknowledged in the statement.1 However, all three recent cohort studies cited in the statement2-4 performed multivariate analyses controlling for disease severity on admission and other factors known to influence clinical outcome.