I used to have a reference for this, but the rule of thumb is that the upper
limit of the 95% confidence interval is 3/n, i.e. 3/8 for 3 patients.
Mark
Mark Ebell, MD, MS
Department of Family Practice
Michigan State University [log in to unmask]
Helping Physicians Become Medical Information Masters
http://www.infopoems.com
> -----Original Message-----
> From: Malcolm Daniel [SMTP:[log in to unmask]]
> Sent: Thursday, July 22, 1999 3:42 AM
> To: 'Paul Sullivan'; ebm group
> Subject: RE: Proportion of zero
>
> Instead of using the calculations ( which in this case give you a standard
> error of zero), access the Geigy tables. You should be able to find them
> in the reference section of your library. Turn to page 89 and you will
> find the "Exact Confidence Limits". In the of none of 8, the 95%CI
> range from 0% to 33.6%.
>
> Cheers
>
> Malcolm
>
> -----Original Message-----
> From: Paul Sullivan [SMTP:[log in to unmask]]
> Sent: 22 July 1999 08:12
> To: ebm group
> Subject: Proportion of zero
>
> 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?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|