Thiru S, et al. - Patients referred to a chest pain clinic were prospectively analyzed to compare the relative values of the Duke Clinical Score (DCS) and the Diamond–Forrester (DF) model, and of pain characteristics only to predict the presence of coronary artery disease (CAD). This study included 1,376 patients. Using DCS, researchers assigned participants to one of five CAD likelihood groups (< 10, 10–29, 30–60, 61–90, and > 90%), and using the DF model, to three CAD likelihood groups (< 15, 15–85 and > 85%). Either the presence of obstructive (> 70%) coronary stenoses or a positive functional test confirmed CAD diagnosis. A significantly lower CAD prevalence was noted in patients referred for suspected angina pectoris compared with that expected by using historical prediction models. No improvement in diagnostic accuracy was evident, even with the use of risk factors profile and demographics in addition to symptoms characteristics.