Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Meta-Analysis of Hazard Ratios and Confidence Intervals

From   "Roger B. Newson" <>
Subject   Re: st: Meta-Analysis of Hazard Ratios and Confidence Intervals
Date   Wed, 19 Sep 2012 10:08:11 +0100

Hazard ratios have weighted geometric means, which are calculated by -metan- in exactly the same way as weighted geometric means for odds ratios, with the same choice between the fixed-effect and randomly-variable effect methods.

An alternative possibility might be to do meta-analyses using the SSC packages -eclplot-, -parmhet-, and -parmest- (which has a -metaparm- module for this purpose). These offer the user the additional choice of a nonrandomly-variable-effect meta-analysis (as well as the fixed-effect and randomly-variable-effect options). However, -metan- can provide a more instant solution, if Belinda is in a hurry.

Best wishes


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

Opinions expressed are those of the author, not of the institution.

On 19/09/2012 05:19, Belinda Butcher wrote:
I had read this, but am unsure what to do with the confidence intervals, all the examples I have seen calculate the logodds and selogodds basically from the 2x2 table (which I don't have)?


Belinda Butcher BSc(Hons) MBiostat PhD
Director - Biostatistics & Medical Writing
WriteSource Medical Pty Ltd

PO Box 1521
Lane Cove NSW 1595

M: 0418 286 014

On 19/09/2012, at 2:12 PM, JVerkuilen (Gmail) wrote:

On Tue, Sep 18, 2012 at 8:03 PM, Belinda Butcher
<> wrote:

I am conducting a meta-analysis of some cancer data using the hazard ratio for overall survival and the upper and lower 95% confidence intervals.

Usually meta-analyses of a ratio estimator like a hazard ratio,
relative risk or odds ratio are better conducted on the log scale.

JVVerkuilen, PhD

"Out beyond ideas of wrong-doing and right-doing there is a field.
I'll meet you there. When the soul lies down in that grass the world
is too full to talk about." ---Rumi
*   For searches and help try:

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index