We found significant statistical heterogeneity in the pairwise meta-analyses, and the study population in our review included participants with different previous exposures and responses to ADHD medications. These characteristics were quite evenly distributed across the included studies and across the different nodes in the network, therefore, even if they contributed to statistical heterogeneity, it is unlikely that they have implications in terms of clinical heterogeneity and affected the validity of our results. On the contrary, heterogeneity can be seen as increasing the external validity of our findings, because the patients seen in real-world clinical practice tend to have similar variations. Although we included studies that used different rating scales to assess the core symptoms of ADHD, we selected carefully only validated scales that measure exclusively the same triad of symptoms—ie, inattention, hyperactivity, and impulsivity.
