Common reasons for elective screening and surveillance colonoscopy, at predetermined intervals, are family or personal history of colorectal cancer (CRC) or advanced adenoma (AAP). Quantified, human haemoglobin (Hb)-specific, immunochemical faecal occult blood tests (I-FOBT) detect bleeding. To determine I-FOBT sensitivity for CRC or AAP before elective colonoscopy in patients at high-risk of cancer or advanced adenoma. Prospective double-blind study of 1000 ambulatory asymptomatic high-risk patients (555 family history of CRC, 445 surveillance for past neoplasm), who prepared three I-FOBTs before elective colonoscopy. I-FOBTs quantified as ngHb/mL of buffer by OC-MICRO instrument and results >or=50 ngHb/mL considered positive. At colonoscopy, eight patients had CRC, 64 others had AAP. Sensitivity for CRC and/or AAP was the highest, 65.3% (95% CI 54.3, 76.3), when any of the three I-FOBTs was >or=50 ngHb (15.4%), with specificity of 87.5% (95% CI 86.4, 90.5) identifying all CRCs and 62% of AAPs. All cancers or an AAP were detected every third I-FOBT-positive colonoscopy (47/154), so colonoscopy was potentially not needed at this time in 84.6% (846 patients). I-FOBT screening might provide effective supervision of high-risk patients, delaying unnecessary elective colonoscopies. This favourable evaluation needs confirmation and cost-benefit study by risk-group.