A total of 190 patients, referred by general practitioners for a double-contrast barium enema, were subsequently examined with colonoscopy. With colonoscopy and histology as the reference standard, sensitivity, specificity, positive and negative predictive values, and accuracy for the radiological detection of cancer and polyps were calculated. No cancer was overlooked by the radiological examination, but there were four false positives. The overall sensitivity for polyps was 70%, increasing to 81% for polyps 10 mm. The predictive value was 93–97% for the exclusion of polyps. The caecum was reached in 187 patients by double-contrast barium enema (98%) and in 164 patients (86%) by colonoscopy. Lesions in four of 12 patients who had radiological changes were undetected at the first colonoscopy, but a repeat examination showed polyps 10 mm in size. Although colonoscopy is a more sensitive technique for the detection of small mucosal lesions, the general practitioners may, in the vast majority of patients, rely on a negative result for polyps and cancer obtained by the double-contrast barium enema. The latter is linked with a number of false-positive cases, while colonoscopy is associated with technical difficulties; both techniques may lead to repeated examinations, regardless of which was the first choice.