In the Antiarrhythmics Vs Implantable Defibrillators (AVID) trial, all patients who meet the study entry criteria are followed in a registry. During the period between June 1993 and June 1995, of 1,117 patients who were enrolled in the registry and met the study entry criteria, 476 were randomized to receive either implantable cardioverter-defibrillators (ICDs) or drug therapy (amiodarone or sotalol), and 641 patients were not randomized for a variety of reasons including: patient refusal (42%); physician refusal (43%); concerns about ability to maintain follow-up over several years (10%), and others (6%). There were no significant differences between the 476 randomized and 641 nonrandomized patients with regard to clinical characteristics, left ventricular function, history of congestive heart failure, medical history, and previous cardiac procedures performed before the index event, except that randomized patients were slightly older (65 vs 62 years) and had a slightly higher prevalence of coronary artery disease and previous myocardial infarction. The index event and location of the index event were not significantly different between the 2 groups. Although 14% of registry patients received neither ICD nor antiarrhythmic drug therapy, ICDs were no more frequently used in the registry patient than antiarrhythmic drugs (45% for ICD vs 42% for drugs). Thus, randomized AVID patients have very similar clinical characteristics, cardiac history, and presenting arrhythmias as to nonrandomized eligible patients. Therefore, the results of the AVID trial may be generalized for all patients with AVID-eligible arrhythmias.