Letters Comparing risk prediction models

Systematic review did not consider problem of treatment effects

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e4355 (Published 03 July 2012) Cite this as: BMJ 2012;345:e4355
  1. Su May Liew, senior lecturer1,
  2. Jenny Doust, professor2,
  3. Paul Glasziou, professor2
  1. 1Department of Primary Care Medicine, University of Malaya, 50603 Malaysia
  2. 2Faculty of Health Sciences and Medicine, Bond University, QLD 4229, Australia
  1. su.liew{at}phc.ox.ac.uk

Siontis and colleagues’ systematic review comparing established risk prediction models for cardiovascular disease is a useful validation of previous studies,1 but they seem to have ignored the problem of treatment effects.

We found in our review that none of the available risk scores adjusted for the treatment of risk factors in their populations.2 Ignoring the effect …

View Full Text

Sign in

Log in through your institution

Free trial

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