Limitations of trial

doi:10.1136/bmj.b2531

Published 23 June 2009, doi:10.1136/bmj.b2531
Cite this as: BMJ 2009;338:b2531

Letters

Colorectal cancer

Limitations of trial

The first 150 words of the full text of this article appear below.

In the NORCCAP trial of colorectal cancer screening in Norway,¹the number of patients screened and the range of intestinalsites examined was too limited for even an apparent fourfoldreduction in mortality from rectosigmoid cancer to be reliablydetectable by a standard intention-to-treat analysis of overallcolorectal cancer mortality. In these circumstances, it seemsmore appropriate for the medical interpretation of the availableresults to be based on separate consideration of rectosigmoidcancer mortality and colon cancer mortality, and for the mainemphasis to be on the separate results for those who acceptedthe screening invitation; those who, although allocated screening,rejected the invitation; and those allocated control.

Among those screened, all had rectosigmoid examination, butonly 19% had, in addition, full colonoscopy; so, protectionagainst rectosigmoid cancer would be expected to be greaterthan protection against colon cancer mortality. The table ofthe numbers of colorectal cancer deaths . . . [Full text of this article]

Richard Peto, professor of medical statistics and epidemiology¹

¹ University of Oxford, Oxford OX3 7LF

richard.peto@ctsu.ox.ac.uk

CiteULike

Complore

Connotea

Del.icio.us Add to Digg

Digg

StumbleUpon Add to Technorati

Technorati What's this?

Relevant Article

Risk of colorectal cancer seven years after flexible sigmoidoscopy screening: randomised controlled trial: Geir Hoff, Tom Grotmol, Eva Skovlund, Michael Bretthauer, and for the Norwegian Colorectal Cancer Prevention Study Group
BMJ 2009 338: b1846. [Abstract] [Full Text] [PDF]