Statistical Methods for Meta-Analysis

Course summary
The course will cover general techniques used in meta-analysis. This includes summary statistics arising from a collection of independent studies and based upon effect measures such as risk or odds ratio, standardized difference or correlation coefficient.

Course Description
The course will provide an overview on the state-of-the-art methodology and aim to enable participants to undertake the statistical analysis of their own meta-analytic data independently. Fixed effects and random effects approaches will be discussed. The course will focus on techniques dealing with heterogeneity such as sub-group analysis and meta-regression. Cumulative meta-analysis and publication bias assessment will be covered. Special topics such as rare event meta-analysis using generalized linear mixed models will also be touched on. Practical sessions (about 50% of the course time) will be used to illustrate the methodology at hand of case-studies. The course will use the package STATA and special add-ons such as METAN, METAREG, METACUM or METAFUNNEL. The introductory lecture is attached as a sample of how the material would look.

Course timetable
On the first day, the course will start with registration and coffee at 9.30am with formal teaching starting at 10.00 am, on the last day formal teaching will end at about 4.00pm. Afterwards there will be an opportunity for participants to ask questions about the course and to discuss with the instructor how to analyse their own data (until about 5pm). Participants are welcome to bring their own data if they wish.

Effect measures and basic concepts of meta-analysis (day 1)
Introduction to STATA and METAN (day 1)
Heterogeneity in meta-analysis including subgroup analysis (day 1)
Cumulative meta-analysis (day 1)
Small study size effects (publication bias) (day 2)
Meta-regression (day 2)
Meta-regression (for rare events data) using generalized linear mixed models (day 2)

Course Code
S3RISMA2016

Course Dates
19th September 2017 – 20th September 2017

Course Presenter
Professor Dankmar Böhning