It is common clinical practice to test stools for occult blood and to rely heavily upon the result in deciding whether to investigate further. Because the accuracy of such testing has been questioned, a study was conducted to estimate the sensitivity of the Hemoccult II slide test in detecting colonic carcinoma and polyps. Patients with lesions demonstrated on barium enema roentgenography had their stools examined for occult blood. One to 2 weeks before colonoscopy, stool specimens were collected for 3 consecutive days according to the manufacturer's instructions. The sensitivity of occult blood testing was 38% for patients with benign polyps and 64% for patients with polypoid carcinomas. Thus, the authors conclude that the Hemoccult II test is not reliable as a diagnostic tool and that a negative test result does not exclude the presence of a colonic neoplasm in any patient.