South Asian Cochrane Network Workshop

South Asian Cochrane Network-coordinating site (

Christian Medical College, Vellore, Tamil Nadu, India Workshop on Standard and Advanced Statistics in Meta-analysis;

August 21-26, 2006

Sponsored by the Cochrane Statistical Methods group

Hands on workshop on statistics used in meta-analysis. The workshop on standard statistics used in meta-analysis is aimed at authors of Cochrane Reviews and protocols, Cochrane Review group editors and statisticians and will be followed by the advanced statistics workshop that is aimed at those who have completed the standard workshop and for experienced review authors, editors, and statisticians.

Limited to 35 participants. Priority for those residing in Bangladesh, Bhutan, Maldives, Nepal, Sri Lanka, Pakistan and India.


To support review authors, editors and statisticians to understand the concepts and methods used in meta-analysis

Part I: Standard statistical methods used in meta-analysis

Meta-analysis: - Calculating and interpreting different measures of treatment effects for binary, continuous and survival data

- Choosing the right effect measure

- Indirect estimation of treatment effect when direct estimation is not possible from published data

- Carrying out fixed effects meta-analysis using different methods (Mantel-Haeszel, Peto, Inverse Variance) and interpreting pooled estimates

Heterogeneity and meta-regression:

- Understanding heterogeneity, and the random effects assumptions

- Exploring heterogeneity and interpreting random effect pooled estimates

- Understanding methods, pitfalls and interpretation of meta-regression

Combining data from different randomised controlled trial designs:

- Effectively incorporating data from cross-over and cluster randomised trials in a meta-analysis

Bias in Randomised Controlled Trials:

- Being aware of the different bias that can be introduced in randomised controlled trials and consequently in the meta-analysis

- Understanding and adequately producing the methods section of a review according to the Cochrane guidelines.

Part II: Advanced methods in meta-analysis

Multiple-treatments meta-analysis:

- Introduction to the concepts and methodology of MTM

- Understanding the benefits of combining direct and indirect information (greater precision and ranking of treatments)

- Exploring different implementation alternatives (Bayesian versus frequentist)

- Specific methodological issues (underlying assumptions of the MTM models and evidence consistency)

- Bias due to confounding in MTM

Meta-analysis in Stata 8:

- Carrying out meta-analysis, subgroup analysis, meta-regression, and investigating bias

- Analyzing datasets

Meta-analysis using S-plus/R software:

- Carrying out meta-analysis, subgroup analysis, meta-regression, and investigating bias

Bayesian meta-analysis:

- Understanding basic principals of Bayesian inference and the advantages of Bayesian meta-analysis

- Understanding convergence issues, prior specification and sensitivity to prior distributions

- Performing Bayesian Meta-analysis using WinBUGS

Resource person: Dr. Georgia Salanti, member, Cochrane Statistical Methods Group; Statistical Advisor, CCochrane Psychosocial, Developmental and Learning Difficulties Group; Lecturer in Biostatistics and Epidemiology, University of Ioannina, Greece.

Cost: Rs. 3000 for both courses: this covers course materials, refreshments and lunch on workshop days and a social evening.

Closing date: August 15, 2006

Remember we are NOT Doctors and have NO medical training.

This site is like an Encyclopedia - there are many pages, many links on many topics.

Support our work with any size DONATION - see left side of any page - for how to donate. You can help raise awareness of CAM.