Investigators have proposed a project to apply “statistical theory to demonstrate that both number of seizures and time on treatment matter”. I do not mean to be obtuse, but I found the purpose of this project very difficult to discern. Perhaps the authors could clarify? It seems they are modeling past daily/weekly seizure burden to determine how long someone needs to be on treatment to discern non-effect, but that is not obvious. Instead, the main outcome is described as “The main outcome measure will be the number of days each patient was continued on therapy after lack of response was determined”. But the investigators are analyzing clinical trial data, where all patients randomized to inclusion are expected to adhere to whatever treatment was assigned for the duration of the trial. Additional clarification would help.