To the Editor.
—We were most interested in the recent article by Graboys et al1 that concluded "as many as 50% of all coronary angiographies in the United States may be unnecessary." Based on the methods used and the conceptual organization of the study, we believe alternative explanations may account for their findings.The 168 patients in the study are not representative of all angiography patients. Self-referred patients seeking a second opinion are often individuals with significant doubts about the recommendation they received for angiography, perhaps for reasons other than those presented. The patients have relatively mild disease, whereas a typical population of patients with coronary angiography includes a substantial fraction who have had life-threatening symptoms, such as an acute myocardial infarction or unstable angina. Angiography is more likely to be appropriate for patients with these latter conditions.In addition, the patients included in this sample appear to prefer