Evaluation of safety and efficacy of new drugs is based largely on data from clinical trials involving a limited number of patients. This approach does not necessarily detect the rare adverse events that may only be observed when very large numbers of patients are studied. Consequently, we designed a double-blind 12-week trial comparing the new angiotensin-converting enzyme (ACE) inhibitor, quinapril (n = 5,053), with a well-established beta-adrenergic receptor blocker, metoprolol (n = 506). Essentially hypertensive patients (diastolic blood pressure 95-114 mm Hg) received either 10 mg quinapril or 50 mg metoprolol once daily, and the doses were doubled at 4-week intervals to a maximum of 40 and 200 mg, respectively, in nonresponders. Responder rates were similar under both regimens. Adverse events were assessed by interview as well as by a standard questionnaire. The overall prevalence of adverse events reported by standard questionnaire was higher than that reported spontaneously during interviews. With respect to typical ACE inhibitor adverse reactions (e.g. cough and taste disturbances), there was no difference between quinapril and metoprolol independent of the mode of reporting. In summary, both drugs showed comparable overall tolerance and safety. The discrepancy between spontaneously reported and questionnaire-reported adverse events was noteworthy, and this finding prevailed in a volunteer group of 327 patients who were treated with quinapril for 52 weeks. Thus, a questionnaire is of great significance in addition to the patient history/interview in a large-scale, double-blind study designed to learn about details of drug safety.