A recent study has highlighted serious flaws in the way that animal experiments are conducted, adding to growing evidence that animal research is riddled with bias.
The aim of the study, which was carried out by researchers from the University of California, was to find out if experiments funded by drug companies are more likely to report favourable outcomes than those funded by government and other non-industry sources.
The researchers carried out an analysis of 63 animal studies of popular cholesterol-lowering drugs called statins and found severe flaws and substantial risks of bias in all of the experiments, regardless of who funded them. For example, in about half of the studies, researchers were fully aware of which animals were given the drug and which ones were given a placebo.
It concluded that studies funded by drug companies are actually less likely to present overly optimistic findings. The explanation for this could be cost; human clinical trials are expensive and drug companies are reluctant to push forward a drug that could fail in later stages. Despite this however, the study found that the majority of drug experiments were still likely to be positive towards the drug and any negative side effects seen in animals were overlooked.
According to the authors, these findings highlight how easy it is for researchers to “spin” results to make them come across as more favourable.
1. Science Daily: http://www.sciencedaily.com/releases/2014/01/140127093214.htm
2. Nonindustry-sponsored preclinical studies on statins yield greater efficacy estimates than industry-sponsored studies: a meta-analysis. (2014). PLOS Biology, 12(1): e1001770: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001770