Colorectal cancer screening in Italy: Feasibility and cost effectiveness in a model area

Objective. To evaluate the feasibility and cost-effectiveness of screening programmes for colorectal cancer in Italy. Design. We compared five types of programmes: annual faecal occult blood testing, sigmoidoscopy (every 5 years), faecal occult blood testing plus sigmoidoscopy (every 1 and 5 years), colonoscopy (every 10 years) (all in the age group 55-69 years, last examination at 70 years) and 'filter' colonoscopy. The latter had to be performed in persons at 50 years of age and repeated every 10 years until the age of 70. Costs for the tests and colon cancer care were paid by the Regional Health Office to the hospitals performing the procedures/treatments. Setting. Data were applied to a small model area in northern Italy (Gemona, 80,000 inhabitants) with well-known demographic (age distribution) and epidemiological (colon cancer incidence) features. Results. All-inclusive 10-year costs per screenee and per death prevented (in US dollars) were: 965 and 77,200 for faecal occult blood testing; 436 and 15,500 for sigmoidoscopy; 1521 and 35,000 for sigmoidoscopy plus faecal occult blood testing; 510 and 15,100 for colonoscopy; 510 and 14,000 for 'filter' colonoscopy. With 'filter' colonoscopy the programme required 870 colonoscopies per year, while with colonoscopy 13,700 colonoscopies were needed at time zero. Conclusions. In Italy, screening programmes based on sigmoidoscopy/colonoscopy are more cost effective than those based on faecal occult blood testing. 'Filter' colonoscopy at age 50 appears superior to the other types of endoscopy-based screening programmes because it utilizes, at any point in time, a much smaller fraction of available resources.