Interesting project to to develop a novel method for summarizing and comparing class-related safety in order to inform future schizophrenia drug development. While the methods used to create an integrated summary of safety events among patients enrolled in the risperidone trials are straightforward, as are the methods used to create a table with all AE Preferred terms and their EBGM value for antipsychotics, no information is provided on how the authors plan to compare these two tables. How will they conclude whether the FAERs data represent a reasonable proxy for safety given the AEs experienced in the trial? Or is the only purpose to estimate the proportion of subjects in each clinical study having an AE Preferred Term as a cumulative function of each Preferred Term’s class-related risk in real-world reporting database FAERS? The proposal would be strengthened by making this more clear.