I have been asked to forward this to the list.
Please reply to the details given and not to me.
A PSI Training Course in
Meta-analysis for Drug Development
presented by
Stephen Senn
University of Glasgow
Meta-analysis has now achieved a century of use and controversy:
· from Karl Pearson's analysis in 1904 in the BMJ of inoculation
for enteric fever
· to Juni et al's analysis in 2004 in the Lancet of risks of
cardiovascular events in taking rofecoxib
In this two-day course Stephen Senn will prepare delegates for the second
century of meta-analysis, covering the practicalities of carrying out such
analyses, and using case studies where appropriate.
The course will be invaluable for all those involved in meta-analysis of
pharmaceuticals, whether as consumer or producer, or working for sponsor,
regulator, contract research organisation or academia.
The practicalities of carrying out this kind of analysis
· Different models for analysis
· When and when not to pool
· The respective roles of fixed and random effects analyses
· Frequentist approaches
· Bayesian approaches.
Parallels to the analysis of multi-centre trials
· The many modes of meta
· Ordinary least squares versus meta-analysis weighting
· Handling interactions
· Type II and Type III attitudes
Graphical methods
· Scatter and bubble plots
· Forest plots
· Galbraith plots
Differences in practice between 'public' organisations and commercial
organisations such as pharmaceutical sponsors.
· The Cochrane Collaboration vs pharmaceutical sponsors
· Publication bias vs appropriate analysis
Common pitfalls
· Double counting
· Different trial designs
· Baseline risk
Case studies
· Screening for breast cancer
· Rofecoxib and heart disease.
· Formoterol and asthma
Interactive and discussion sessions will be included
Registration begins at 9.30 on Day 1. The course runs from 10.00 to 17.30
on Day 1 and from 09.00 to 16.00 on Day 2.
Further Details: Ali Houghton, PSI Executive Office
Tel: +44 (0)1625 267882, Fax: +44 (0)1625 267879, E-mail:
[log in to unmask]
|