A definite diagnosis of anterior myocardial infarction is often difficult to make in patients when a pattern of poor R wave progression in the precordial leads is present on the electrocardiogram. The purpose of this study was to determine whether a mathematical model could be devised to identify patients with anterior infarction among 102 consecutive patients with poor R wave progression. Each patient underwent exercise testing with thallium scanning. The diagnosis of anterior infarction was established in 20 (20%) of the 102 patients by the presence of fixed thallium-201 perfusion defects in the anterior wall or septum, or both. With the use of a multivariate stepwise discriminant analysis of clinical and electrocardiographic variables, five variables (sex, ST-T changes, S wave amplitude in leads V2 and V3 and the sum of the R wave amplitude in leads V3 and V4) that were statistically significant by univariate analysis were selected by the model to identify patients with anterior infarction (sensitivity 85%, specificity 71%). The discriminant model was subsequently applied prospectively to an additional 21 patients with poor R wave progression and provided a sensitivity of 85% and a specificity of 88%. Thus, anterior infarction (fixed thallium-201 defects in the anteroseptal segments) was present in 20% of patients with poor R wave progression in the precordial leads; and a mathematical model can be used to identify a subset of patients with anterior infarction in a group of patients with poor R wave progression.