The large number of empirical studies trying to replicate Caspi's GxE findings for depression have been summarized in several reviews focusing on GxE with 5-HTTLPR (see for example 72,81,82-88). These reviews ultimately fueled a heated debate regarding the plausibility of the Caspi findings. Including somewhat identical individual studies, review papers have drawn opposing conclusions about the support for GxE effects, with some studies finding consistent GxE effects and others failing to detect them.84,85 Meta-analyses have provided a quantitative summary of these studies, but have also reached opposing conclusions. Specifically, the results of two meta-analyses,82,86 which found evidence against a consistent GxE effect, differed from a third meta-analysis,83 which concluded there was strong evidence to support the 5-HTTLPR GxE. These conflicting results may be explained by differences in the selection of studies for inclusion in the meta-analyses.89,90 For example, the meta-analyses that used the most stringent inclusion criteria82,86 failed to support the GxE association.91 Some have also noted there is an inverse relationship between the power of the replication studies and support for the 5-HTTLPR association, precisely the opposite of what
