Meta-analysis of genetic association studies: magic tool or dangerous black box?

Excerpt

In the field of genetic epidemiology, where primary studies are often underpowered to detect small genetic effects and where conflicting results and non-replication of initial findings are often encountered [1], the importance of meta-analysis has been widely recognised, as demonstrated by the steep increase in the number of published genetic meta-analyses over time [2]. What has not received sufficient attention is the fact that the results of a meta-analysis, as with any other study design, are only valid if the meta-analysis has been conducted following appropriate methodology. Therefore, the paper by Boccia and colleagues published in this issue [3], which reports the findings of a systematic review on the quality of meta-analyses of genetic association studies in the field of cancer research, represents an important contribution to a rather neglected area. …