The majority of clinical trials of antiepileptic drugs (AEDs) fall into two
main groups: (i) phase II and early phase III add-on studies in people wit
h drug-resistant, chronic epilepsy, and (ii) late phase III parallel-group
monotherapy studies in newly diagnosed patients. Both types of trial record
the occurrence of seizures over a set period from which a large number of
endpoints may be derived for analysis. Although the main objective of treat
ment is the suppression of seizures, other associated endpoints such as dru
g side-effects, quality of life, and costs are also of relevance. Within an
y particular trial, the required sample size will depend crucially on the c
haracteristics of the population sampled, the type (and perhaps number) of
endpoints. losses to follow-up and other deviations from protocol, as well
as important statistical considerations, especially the type of design, inc
luding Type I and II error rates, and methods of analysis. (C) 2001 Elsevie
r Science B.V. All rights reserved.