BACKGROUND: The value of a risk factor analysis in the presence of a clinically suspected deep venous thrombosis (DVT) has been assessed mainly in inpatient populations. The aim of this prospective study was to evaluate the potential association between DVT and acquired circumstances suspected as risk factors, in a cohort of outpatients with a clinically suspected DVT. METHODS: Consecutive outpatients referred for a clinically suspected DVT, with recent clinical signs, not exceeding 1 week, were included. Before venography, all patients were interviewed by a trained physician to detect the presence of risk factors. RESULTS: From March 1992 to February 1994, 277 patients were included; venography was positive in 162 (58.4%). Five independent variables were significantly associated with the occurrence of DVT; in a multivariate analysis, 64.7% of patients were correctly classified; odds ratios for having DVT in the presence of these underlying conditions were respectively: 1.75 for age over 65 years, 1.68 for prior history of venous thromboembolism, 1.69 for high risk circumstances (any type of surgery or leg trauma within the past 3 months), 5.59 for malignancy, and 2.56 for varicose veins. CONCLUSIONS: In outpatients referred for a clinically suspected DVT, recognition of associated conditions might increase the certainty of the diagnosis.