This Viewpoint argues for caution when interpreting the findings of meta-analyses of heterogenous studies with small samples.