[et2o]

The problem with this study is that the analysis is based upon intention to treat. Only 42% of those who were randomized to get a colonoscopy actually got one. The cancer incidence was 18% lower among all people who were invited to get a colonoscopy. If we assume rate among people who were invited to get a colonoscopy but did not go through with it is the same as the rate of people who did not get a colonoscopy then the cancer rate would be more like -2/3 among those who got a colonoscopy.