Letters Effect of study outcome on publication

Tackling submission and publication bias

BMJ 2017; 358 doi: https://doi.org/10.1136/bmj.j3436 (Published 17 July 2017) Cite this as: BMJ 2017;358:j3436
  1. Rainer Spiegel, senior physician, Division of Internal Medicine,
  2. Petra Opic, resident physician, Medical Intensive Care Units,
  3. Saskia Semmlack, resident physician, Medical Intensive Care Units,
  4. Sarah Tschudin-Sutter, senior physician, Division of Infectious Diseases and Hospital Epidemiology,
  5. Raoul Sutter, senior physician, Medical Intensive Care Units
  1. University Hospital Basel, University of Basel, Am Petersgraben 4, 4031 Basel, Switzerland
  1. rainer.spiegel{at}usb.ch

Evoniuk and colleagues analysed whether submission and publication bias are based on study outcome.1 They retrospectively reviewed 1064 drug research studies sponsored by GlaxoSmithKline over six years and did not find evidence for either bias.1 This finding is different from previous studies and discussions.234 Has the acceptance of studies reporting “negative” results truly improved in the scientific community, so that submission or publication bias is no longer relevant? No, for at least three reasons.

Firstly, Evoniuk and colleagues analysed only …

View Full Text

Log in

Log in through your institution


* For online subscription