Frightened Mouse Points to Lack of Scientific Rigour
The mouse is the uncontested, go to mammalian model organism, for several reasons. Firstly, their physiology and anatomy are fairly close to ours, as is their genetics, with over 95 % similarity between human and mouse genomes. Secondly, as they are small, and have a short generation time and accelerated lifespan, costs are kept low, allowing the sample size for an experiment to be quite large. Finally, mice are by far the most commonly genetically altered organism, with a large catalogue of animals available for study which have either had certain genes removed, or human genes added. This should enable researchers to draw conclusions which are not only meaningful, but reliable. However, should is the operative word here.
A study last year found that mice fear men, but not women. Consequently, significant results published by researchers may simply be the result of male-induced stress, rather than the variable under investigation. This does not only impact behavioural studies, as stress levels impact rodents at the cellular level too, potentially jeopardising medical research. It is remarkable that this effect had not been identified previously, as we have been using mice in our experiments for the past century. Such an oversight in the field reflects the pressure that researchers are under to publish in high impact journals, as confirming another researcher’s observations seldom leads to Nature.
Ben Goldacre has spoken on the subject of publication bias before, and how journals are often “not interested in publishing replication”, nor “negative data”. This is supported by statistical analyses which have demonstrated that there are a tremendous amount of negative results missing from the literature. It is important to note how catastrophic this can be. One infamous example concerns the anti-arrhythmic drug lorcainide. Early research suggested that such drugs could lead to an increase in mortality; however, this data went unpublished for 13 years. This meant that during the 1980’s, when several pharmaceutical companies invested in the commercial production of these drugs, there was no data available in the literature to indicate the harm they may incur. As it happens, these drugs dealt a great deal of harm, leading to a “death toll larger than the United States’ combat losses in wars such as Korea and Vietnam”.
It is crucial that we strive towards a more rigorous and less biased scientific process, but to do so, we must first address how we currently incentivise researchers.
Latest posts by Christian Matheou (see all)
- GM Crops: Why the Stigma? - July 27, 2015
- The Gap between the Brain and a Petri Dish - July 24, 2015
- A Less Invasive Route for Neuroregeneration - June 17, 2015
- “Lab-on-a-Chip” Delivery of Large particles into Cells - June 1, 2015
- Ultrasound as a Potential Treatment In Alzheimer’s - April 6, 2015