For two decades, the evaluation of antiarrhythmic drugs has essentially been based on the count of premature beats during Holter monitoring and the inducibility of tachyarrhythmias during electrophysiologic studies. These approaches are convenient to provide firmly established short-term evaluations of the effects of type I agents in relatively benign arrhythmias occurring in patients with good cardiac function. However, the situation is much more uncertain for long-term outcome in serious tachyarrhythmias occurring in severely diseased patients with an altered myocardium. In addition, indications for type I agents are not really adapted to beta-blockers and type III drugs, the effects of which are much more difficult to assess and depend at least in part on the target arrhythmia. There is an interest in evaluating drug efficacy in terms of responders and nonresponders and in understanding the reasons for a response rather than reliance on pure quantitative data, the meaning of which may be less relevant in the final analysis. We must also recognize the need to reconcile the multiple obstacles encountered when severely diseased patients must be managed efficaciously and safely in the long term with new drugs, the effects of which need objective assessment.