| [ Summary ] |
Randomized controlled trials have demonstrated that the use of screening programs with guaiac-based fecal occult blood tests (FOBT) reduce mortality from colorectal cancer by 15-33%. On the other hand test sensitivity for guaiac-based FOBT was reported to be 23-33% on the basis of colonoscopy performed on all subjects in asymptomatic populations. However, these values are too low to account for the 15-33% reduction in colorectal cancer mortality. For immunochemical FOBT, apparent higher sensitivities than the guaiac-based test have been reported, eg. 67-89% vs 33%. However the sensitivity to detect small early cancers less than 1cm in diameter was as low as 30% even for the immunochemical FOBT.
There are several issues to be considered in interpreting test sensitivity in relation to screening efficacy. One important issue is that sensitivity based on colonoscopy in all subjects necessarily leads to an overestimate of the prevalence, which is a denominator in calculating sensitivity. Accordingly, calculated sensitivity is an underestimate of sensitivity. The program sensitivity of a series of annual screenings might be a better alternative to test sensitivity in understanding screening efficacy. Such program sensitivity was reported to be as high as 89% with the guaiac-based FOBT from a Minnesota study. To improve detection rates for early cancer, a program incorporating colonoscopy in to the present FOBT screening would be a promising choice. |