A randomised study of screening for colorectal cancer using faecal occult blood testing: results after 13 years and seven biennial screening rounds

Abstract

Background: Three randomised trials have demonstrated reduction in mortality from colorectal cancer (CRC) by repeated screening with faecal occult blood tests, including the trial presented here, which is the only one still in progress. Aims: To evaluate reduction in mortality after seven screening rounds and the possible influence of compliance on mortality from CRC. Methods: At Funen in Denmark, random allocation to biennial screening with Hemoccult-II in 30 967 subjects aged 45–75 years and 30 966 controls was performed in 1985 from a population of 137 485 of the same age. Only participants who completed the first screening round were invited for further screening. Colonoscopy was offered if the test was positive. The primary end point was death from CRC, and the 10 year results were published in 1996. Results: From the beginning of the first screening to the seventh round, mean age increased from 59.8 to 70.0 years in the screening and control groups, and the male/female ratio decreased from 0.92 to 0.81. Those who accepted screening were younger than non-responders. Positivity rates varied from 0.8% to 3.8%, the cumulative ratio of a positive test was 5.1% after seven rounds, and 4.8% of patients had at least one colonoscopy. Mortality from CRC was significantly less in the screening group (relative risk (RR) 0.82 (0.69–0.97)), and the reduction in mortality was most pronounced above the sigmoid colon. After seven rounds, RR was reduced to less than 0.70 compared with controls. Mortality rates from causes other than CRC did not differ. Non-responders had a significantly increased risk of death from CRC compared with those who accepted the full programme. Subjects who accepted the first screening, but not subsequent ones, demonstrated a tendency towards increased risk. Conclusions: The persistent reduction in mortality from CRC in a biennial screening program with Hemoccult-II, and a reduction in RR to less than 0.70 in those adhering to the programme, support attempts to introduce larger scale population screening programmes. The smaller effect on mortality from CRC in the rectum and sigmoid colon suggests evaluation by additional flexible sigmoidoscopy with longer intervals.