Limitations of trialBMJ 2009; 338 doi: http://dx.doi.org/10.1136/bmj.b2531 (Published 23 June 2009) Cite this as: BMJ 2009;338:b2531
- Richard Peto, professor of medical statistics and epidemiology1
- 1University of Oxford, Oxford OX3 7LF
In the NORCCAP trial of colorectal cancer screening in Norway,1 the number of patients screened and the range of intestinal sites examined was too limited for even an apparent fourfold reduction in mortality from rectosigmoid cancer to be reliably detectable by a standard intention-to-treat analysis of overall colorectal cancer mortality. In these circumstances, it seems …
Log in using your username and password
Log in through your institution
Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial