Bd. Jovanovic et Rj. Zalenski, SAFETY EVALUATION AND CONFIDENCE-INTERVALS WHEN THE NUMBER OF OBSERVED EVENTS IS SMALL OR ZERO, Annals of emergency medicine, 30(3), 1997, pp. 301-306
A common objective in many clinical studies is to determine the safety
of a diagnostic test or therapeutic intervention. In these evaluation
s, serious adverse effects are either rare or not encountered. In this
setting, the estimation of the confidence interval (CI) far the unkno
wn proportion of adverse events has special importance. When no advers
e events are encountered, commonly used approximate methods for calcul
ating Cls cannot be applied, and such information is not commonly repo
rted. Furthermore, when only a few adverse events are encountered, the
approximate methods for calculation of CIs can be applied, but are ne
ither appropriate nor accurate. In both situations, CIs should be comp
uted with the use of the exact binomial distribution. We discuss the n
eed for such estimation and provide correct methods and rules of thumb
for quick computations of accurate approximations of the 95% and 99.9
% Cls when the observed number of adverse events Is zero.