Colorectal carcinoma and Haemoccult

Publisher Website
Google Scholar
Abstract
Meta-analysis has been performed on the world literature relating to Haemoccult screening for colorectal cancer. The analysis indicated a sensitivity of about 50%, specificity of about 97%, positive predictive value of 4.3% and negative predictive value of 99.85%. These figures assume a prevalence of colorectal cancer in the population of 0.3%. A model calculation using these figures together with some additional realistic assumptions suggests no reduction in mortality compared to an unscreened population. It is concluded that many questions about faecal occult blood screening remain and that uncontrolled mass population screening is not justified at this stage.