“…Eslava-Cobos (75) proposed that a more general assessment of outcome based on time to nth seizure can be obtained by using an individualized value of n corrected on the basis of baseline seizure frequency. For protocols that require exit of a patient from the study either when seizures are poorly controlled or when unacceptable side effects occur, retention in the trial is also an excellent endpoint that reflects dual aspects of efficacy of treatment (70,72,73).…”