• English
  • Español

B6O2 | Bayesian approach to meta-analysis. What can you gain?

Abstract text
The Bayesian approach gives new opportunities as aggregating different types of data, combining direct and indirect comparison or assessing clinical significance. On the other hand, the variety of Bayesian models can be a bit confusing, and difficulties of its implementation can cause unwillingness to apply it.
The aim is to systemize the knowledge of application of Bayesian statistics in area of meta-analyses, and to compare it with traditional statistical methods. We want to identify the situations in which Bayesian approach is really worth to use it.
Initially the systematical reviews on existing statistical methods for meta-analyses was conducted. The special attention was put on different Bayesian models which afterwards were implemented in WinBUGS environment and examined on different data sets.
In the case of regular meta-analysis of dichotomous data applying basic Bayesian models leads us, in fact, to similar results of estimation as Mantel-Haenszel or DerSimonian-Laird method. The real advantage of Bayesian approach is noticed if we expect something more than typical meta-analysis, especially if we have to deal with the following problems:
1. Assessing the clinical significance - for instance, assessing the chance that relative risk is greater than 1.25 (or any other level of significance).
2. Combining data from different type of studies -including extra information (e.g. results of non-randomized studies) to meta-analysis, keeping moderate “level of conviction” to this extra data.
3. Combining direct and indirect evidence (Mixed Treatment Comparison).
Bayesian statistics give us technical opportunities to improve meta-analysis, especially in area of aggregating multi-type data. On the other hand, there is no significant difference between results obtained by Bayesian and traditional approach in the case of simple meta-analysis of regular data. Moreover, if WinBUGS codes are once prepared, then conducting the calculations is not as difficult as one may think.

Walczak J1, Borowiack E2, Nikodem M1, Siedmiogrodzki K1, Zapalska A2, Khan K3, Meads C3, Mol B4, Oude Rengerink K4, Thangaratinam S3, Zamora J5
1 CASPolska Association, Poland
2 Arcana Institute, Poland
3 Queen Mary, University of London, Great Britain
4 Academic Medical Center Amsterdam, Netherlands
5 Hospital Universitario Ramon y Cajal, Madrid, Spain
Presenting author and contact person
Presenting author: 
Mateusz Nikodem
Contact person: 
Ewa Borowiack (Contact this person)
Mateusz Nikodem (Contact this person)
Date and Location
Oral session B6O2
Sábado 22 Octubre 2011 - 11:35 - 12:05