The early diagnosis of small-bowel obstruction is critical in preventing complications, particularly strangulation. Traditionally, the clinical diagnosis of small-bowel obstruction has depended on plain film confirmation. Unfortunately, findings on the plain film may not be confirmatory in 20-52% of cases. The purpose of this study was to determine whether CT is superior to the traditional clinical-radiographic evaluation in prospectively establishing the diagnosis, severity, and cause in cases of suspected obstruction of the small bowel and to see what impact this information might have on treatment, costs, and the need for additional gastrointestinal contrast studies. Physicians from three surgical services referred all patients with suspected small-bowel obstruction for plain film and CT evaluation. Eight-five patients were evaluated on 90 occasions during an 11-month period. Obstruction was classified on the basis of clinical and plain film findings as absent, indeterminate, or present (partial or complete). CT scans were obtained in all patients and were interpreted and graded without knowledge of the clinical-radiographic classification. The results of gastrointestinal contrast studies (barium enema, small-bowel series, and enteroclysis) performed in 21 cases were also compared. The gold standard for the diagnosis was surgical findings in 61 cases and clinical course in 29 cases. On the basis of the combined clinical-radiographic findings, the diagnosis was complete obstruction in 21 of 46 cases (sensitivity, 46%; confidence interval (CI), 32-60%). When CT was used, the diagnosis was established in all 46 cases (sensitivity, 100%; CI, 86-100%). In the 25 cases in which the traditional evaluation failed, the early CT diagnosis of complete obstruction prevented a 12-72 hr delay in surgery with its attendant increased morbidity, mortality, and costs. On the basis of the combined clinical-radiographic findings, partial obstruction of the small bowel was diagnosed in six of 20 cases (sensitivity, 30%), whereas all cases were detected with CT. False-positive CT findings for complete obstruction of the small bowel occurred in three cases of paralytic ileus (one each due to small-bowel infarction, lower lobe pneumonia, and peritonitis due to rupture of the urinary bladder). One case of colonic obstruction due to carcinoma in the hepatic flexure was mistakenly diagnosed as partial obstruction of the small bowel. The clinical and plain film evaluation was never precise enough to provide the exact location or cause of small-bowel obstruction. Gastrointestinal contrast studies provided additional useful information regarding colonic abnormalities (four cases), functional grading of partial obstruction of the small bowel (six cases), and exclusion of a false-positive CT diagnosis of complete obstruction in a case of reflex ileus. CT is sensitive for diagnosing complete obstruction of the small bowel and for determining the location and cause of obstruction. In comparison, the traditional clinical and plain film evaluation is relatively insensitive. CT should be used when the results of clinical and plain film evaluation are inconclusive. Gastrointestinal contrast studies play an important diagnostic role in partial obstruction of the small bowel and in colonic obstruction with predominant small-bowel dilatation.