Background: Double contrast barium enema (DCBE) is the examination carried out most frequently for investigation of patients with large bowel symptoms. The aim of this study was to compare the sensitivity of DCBE and colonoscopy for the detection of colorectal cancer and neoplastic polyps Ն1 cm. Methods: All patients undergoing DCBE (1389) or colonoscopy (1081) as the primary investigation for large bowel symptoms or for cancer or polyp surveillance in the first 9 months of 1997 at a large teaching hospital were included in this study. At 1 and 2 years following investigation, a computerized search of appropriate diagnosis and procedure codes to detect any missed cancers or polyps was performed for all patients with a normal investigation. Results: Almost 19% of patients in both groups went on to have an additional large bowel investigation over the 2-year period. In the DCBE group, 47 patients (3.5%) had a cancer diagnosed; eight of them had been missed at the primary investigation (sensitivity 83%). In the colonoscopy group, 37 patients (3.4%) had a cancer; one of them had been missed at the primary investigation (sensitivity 97.5%). Neoplastic polyps Ն1 cm were diagnosed in 1.6% of the DCBE group and in 7.7% of the colonoscopy group, with sensitivities of 21.7% and 91.4%, respectively. Nine patients (0.6%) had a false positive diagnosis of cancer in the DCBE group; one had an iatrogenic bowel perforation following flexible sigmoidoscopy. Conclusions: Where adequate facilities and expertise exist, colonoscopy should be the investigation of choice for most patients with large bowel symptoms suggestive of neoplastic disease.Colorectal cancer is the second most common cause of cancer death in the United Kingdom [10]. Despite improvements in surgery and chemotherapy, survival rates over the last 50 years have remained static. One reason for this is that Ն35% of patients are at an advanced stage when first diagnosed [4]. Early detection of colorectal cancer and neoplastic polyps is therefore the key to improvement in this disease, since cure can be effected by surgery and polypectomy may prevent the development of colorectal cancer [16].Double contrast barium enema (DCBE) is the examination carried out most frequently for patients with large bowel symptoms [2]. Its sensitivity for Dukes' A and B colorectal cancer ranges from 55% to 85% [9]. Some authors have suggested that this rate can be improved to 96-98% if combined with flexible sigmoidoscopy [5,6]. National Health Service (NHS) Executive guidelines [9], however, acknowledge colonoscopy as the gold standard for the detection of cancer and polyps, with a reported sensitivity of 96.5% and 94%, respectively [11,15]. In addition, the wider use of colonoscopy has recently been shown to improve the detection rate for early colorectal cancer [7]. The 5-year survival for such lesions is 97.6% [7]; thus, colonoscopy offers real hope for cure of this disease.There have been no large randomized prospective trials comparing the sensitivity of DCBE and colonoscopy for detec...