Single vessel coronary artery disease (SVD) has a generally benign prognosis and many patients with this pattern of disease are managed conservatively. Although coronary artery bypass grafting is sometimes recommended, balloon angioplasty and other interventional procedures provide the principal alternative management strategy. These are of proven value in the treatment of symptomatic disease but no study has yet found them to confer prognostic advantage. The aim of this study was to identify factors predicting the allocation of SVD patients to medical or interventional treatment and to test the hypothesis that the disposition of the attending cardiologist, as manifest by the nature of his or her practice, will strongly influence the process Some 1860 consecutive adult diagnostic catheter procedures were prospectively examined. Patients with previous intervention, acute coronary syndromes or coexisting valvar and myocardial disease were excluded. A final cohort of 76 patients manifesting non-occlusive SVD was identified. These patients were under the care of 11 different cardiologists who were grouped on the basis of whether or not they themselves performed coronary angioplasty procedures (Yes n = 4: No n = 7). The baseline characteristics of the patients of these groups (Yes n = 26: No n = 50) showed no significant differences with respect to age, anginal symptoms and treadmill exercise test duration or positivity by ST segment criteria. Patients were then regrouped on the basis of initial treatment allocation (Intervention n = 28: Medical n = 48). The means of normally distributed, continuous variables (age, exercise test duration) were compared by t test for independent samples and categorical variables were examined by a chi squared regression model. No significant differences were found in age, anginal symptoms or exercise test positivity. Being under the care of a cardiologist performing coronary angioplastywas however strongly associated with allocation to an interventional management strategy (p = 0.4). A multiple logistic regression analysis confirmed attending cardiologist type as a significant and independent predictor of treatment allocation. This study demonstrates that, in patients with SVD, physician prejudice may be an important factor in treatment recommendation and while this may have little influence on the group's uniformly favourable prognostic outlook it has clear implications for individual patients and for resource allocation.