The question of CIs for LRs has come up several times before on the list. I
don't know if any of these articles or softwares calculate power and sample
sizes; however, with an equation set up in a spread sheet or with
off-the-shelf software you can simply plug in sample sizes until you find
what gives you an acceptable CI by trial and error.
Here are LR CI methods from the previous postings:
Simel DL, Samsa GP, Matchar DB. Likelihood ratios with confidence:
sample size estimation for diagnostic test studies. Journal of Clinical
Epidemiology 1991; 44:763-770.
You can find a Bayesian clinical calculator at:
Statsdirect calculates LR CIs.
or WWW.STATSDIRECT.COM I think there is a free sample there.
"you might want to check the statistics appendix section in Sackett's latest
book: Evidence-based Medicine: how to practise and teach: publ Churchill
Livingstone 2000. There's an equation that allows you to calculate LR
confidence intervals." However, there is apparently an error in that
method. But a correction has been published. "The correction of the
equation from the EBM book is posted on the
university of toronto website (see erratum 21 at
Apologies for quoting other people's e-mails without credit.
Something here should work for you.
David L. Doggett, Ph.D.
Senior Medical Research Analyst
Health Technology Assessment and Information Services
ECRI, a non-profit health services research organization
5200 Butler Pike
Plymouth Meeting, Pennsylvania 19462, U.S.A.
Phone: (610) 825-6000 x5509
FAX: (610) 834-1275
e-mail: [log in to unmask]
From: Dr. Giuseppe Giocoli [mailto:[log in to unmask]]
Sent: Thursday, August 23, 2001 6:46 AM
To: [log in to unmask]
Subject: CIs of LRs
Dear List Members,
Is there any practical way to calculate a suitable sample size for
confidence intervals of Likelihood ratios?
Dr. Giuseppe Giocoli
Gruppo di Lavoro EBM
Associazione Microbiologi Clinici Italiani
Via Sarca, 19
25015 DESENZANO d/G (BS) Italia