Traditional randomized clinical trials for the monotherapy assessment of antiepileptic drugs (AED) involve allocation of newly diagnosed patients to long-term treatment with different AEDs in order to determine remission rates and side effect profile. Apart from being time-consuming, however, these trials are unlikely to show significant differences in seizure control between the various drugs, which may lead some regulatory agencies to argue that remission rates could be related to the natural history of the disease rather than to efficacy of the administered drugs. To circumvent this problem, a number of innovative designs for the monotherapy assessment of new AEDs have been developed in recent years. They all share the common feature of being aimed at demonstrating a difference in response rate over a short treatment period between a high dosage of a new AED and some form of suboptimal treatment (placebo or low-dose active control). Patients allocated to suboptimal treatment show unacceptable seizure control more rapidly than patients on high-dose active treatment and therefore they exit the trial at a faster rate: evidence of antiepileptic activity is therefore based on demonstration of differences in rate of deterioration rather than improvement. These trials are conducted with titration schedules, dosages and durations of treatment which are totally unrelated to optimal use of the same AEDs in routine clinical practice. No comparative data with an established reference agent are provided, and allocation of patients to suboptimal treatment raises serious ethical concerns. For these reasons, justification for the continued implementation of these trials is questionable. Randomized long-term comparative trials should be considered the gold-standard for the monotherapy assessment of new AEDs. A review of the literature, however, reveals that long-term trials with new AEDs completed to date had significant shortcomings in their design, including excessively rigid or inappropriate dosing schedules, enrollment of patients with heterogeneous seizure disorders, low statistical power and insufficient duration of follow-up. Because these studies are usually aimed at addressing regulatory requirements, the information obtained cannot be meaningfully applied to routine clinical practice. Large longer-term randomized comparative trials using more pragmatic approaches are highly needed to determine the real value of first-line therapy with new AEDs in patients with well defined seizure disorders.