In recent years a number of articles testing the reproducibility of preclinical biomedical research have been published in top-rated journals. Surprisingly most of this could not be reproduced – not even by the researchers themselves! There are two informative and commendable editorials shedding some more light on this issue – for more information see: Nature 23 May 2013 – p. 433 (six flags), and The Economist Oct 19, 2013 (Trouble at the lab). Approximately 75% of the contents published in international scientific journals with peer-review and high impact factor, were difficult to reproduce. What can the reason for this be? Issues such as the increased pressure to publish, competition, and high ambitions of publishing way too early are discussed as potential contributions to this. However, the researchers also point out that mistakes correct themselves, because other researchers continue to work on precisely this task. This has to be regarded as a controversial attitude. However, good advices were also provided as to how author, editor, referee, and the readers themselves could question the quality of the publication:
1. Was the experiment blinded?
2. Was the experiment repeated?
3. Were all results presented?
4. Were both positive and negative controls included?
5. Were used reagents subjected to quality control?
6. Were the statistical methods/tests optimal?
Happy reading and (personal?) reflection!
Roland
