Case-By-Case Meta-Analysis from A Bayesian Perspective

Due to the advancement of science and the tendency to use the smallest of humans in needless clinical trials in many situations, the role of meta-analysis to summarize published study with the same objective, using statistics is becoming more and more fundamental. The integration of the knowledge available makes it easier to understand and allows strong conclusions. For example, increasing the number of clinical trials encourages a rising need to conduct meta-analyses, making complexity possible. The purpose of the research was therefore to suggest a Bayesian meta-analytical methodology. Bayesian method was proposed to carry out a case-by-case meta-analysis. The procedure was based on the mixture of the posterior distributions of the parameter in question from each study that is a percentage of the meta-analysis, thus producing a combined posterior distribution as a meta-analytical measure. In addition, a brief comparison is made between the proposed method and some methods used in meta-analysis. Some examples with simulated and real data are presented and discussed.