>Can any one help? I'm struggling with the technical point of how to deal
>with a proportion of zero,
>especially in a small study. e.g. "none of the eight study patients
>responded to treatment therefore the treatment
>should never be used clinically for similar patients".
>Is there a way of generating a confidence interval for a proportion of
>zero?
Beyond exact probability methods, there is an approximation. See:
Hanley JA, Lippman-Hand A
If Nothing Goes Wrong, Is Everything All Right-
Interpreting Zero Numerators
JAMA 1983;249(13):1743-1745
David Birnbaum, PhD, MPH
Clinical Assistant Professor
Dept. of Health Care & Epidemiology
University of British Columbia, Canada
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|