The validity of routine ultrasonography (US) in the evaluation of patients with blunt abdominal trauma (BAT) was investigated in a prospective study. From April 1989 to April 1990, 140 patients with suspected BAT were included in this study. Ultrasonography was performed by 17 surgeons using a standardized technique. The influence of the investigators' experience in US was manifested in the positive predictive value (PPV). Surgeons with a learning period of less than 1 year had a PPV of 60%. Investigators with experience of more than 1 year but less than 3 years had a PPV of 76%. For the most experienced investigators (> 3 years) a PPV of 92% was recorded. The sensitivity for intra-abdominal lesions was 100%, 100%, and 92%, and the specificity 94%, 89%, and 98% for the three levels of experience, respectively. We conclude that US is a suitable test for screening patients with BAT since it is highly sensitive, highly specific, complication free, and easy to learn. The positive findings of surgeons who are learning to use this method should be verified by reinvestigation by an experienced sonographer, by diagnostic peritoneal lavage (DPL), or by a CT scan if the patient is hemodynamically stable.