Background. Patients after stroke and major orthopedic surgery have increased factors for developing deep vein thrombosis. We sought to determine the implications of screening high-risk patients to detect proximal deep vein thrombosis. Methods. We used decision analysis to determine the implications of screening vs not screening asymptomatic high risk patients with duplex ultrasonography to detect proximal deep venous thrombosis. The outcomes were bleeding, pulmonary embolism, death, and number of patients with true- positive, false-positive, and false-negative tests. Results. Screening with ultrasonography, all asymptomatic patients who receive appropriate prophylaxis, prevalence 5%, would result in the treatment of 3.1% patients with proximal deep vein thrombosis (true positives); 2.9% without proximal deep vein thrombosis (false positives) and in the lack of diagnosis in 1.9% patients (false negatives). At a prevalence of 20%, no prophylaxis, screening would result in the treatment of 12.4% patients with proximal deep vein thrombosis (true positives), 2.4% without proximal deep vein thrombosis (false positives), and in the lack of diagnosis in 7.6% of patients (false negatives). Conclusions. Screening high-risk patients who receive prophylaxis is not warranted. Patients who receive no prophylaxis should be screened with ultrasonography.