Letters Composite outcomes

“Cherry picking” did not occur in studied example

BMJ 2010; 341 doi: https://doi.org/10.1136/bmj.c5009 (Published 15 September 2010) Cite this as: BMJ 2010;341:c5009
  1. Mark J Bolland, senior research fellow1,
  2. Andrew Grey, associate professor of medicine1,
  3. Ian R Reid, professor of medicine1
  1. 1Department of Medicine, University of Auckland, Private Bag 92 019, Auckland 1142, New Zealand
  1. m.bolland{at}auckland.ac.nz

    In their discussion of composite end points,1 Cordoba and colleagues highlight the reporting of vascular events in our trial of calcium supplements2 as an example of “cherry picking”: the post-hoc selection of a composite end point to create a significant P value. An allegation of cherry picking is an indictment of both authors and editorial staff, and should be supported by …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription