Content Header Mask
Science Blog

The Science Blog

19/07/2013

Study shows animal studies are biased

In a week where the numbers of animals used across Great Britain showed yet another rise, we received reports of a US study that concludes that animal tests are biased and give false hope to patients, researchers and doctors.

Statisticians from Stanford University in the USA looked at reports from ‘meta-analyses’ for 160 potential treatments for brain diseases including Parkinson’s disease, Alzheimer’s disease, multiple sclerosis, stroke and spinal cord injury. (1)

Out of a total 4,445 animal studies included in the meta-analyses nearly half reported a significant positive result (1,719). Based on their statistical analysis, the Stanford researchers argued that actually only about half of these should have been (919 out of 1,719). Overall they found there was ‘excess significance’ in 31% of the meta-analyses for the 160 treatments. 

The researchers went on to say that they considered that only eight of the 160 drugs should have really been progressed to human trials on the basis of the results. Out of those eight only two had actually showed some promise in human trials (NOS donors for stroke and bromocriptine for Parkinson’s disease). For example, they found that a meta-analysis of 33 animal studies all testing a treatment called MBP for multiple sclerosis suggested that it would work but a subsequent clinical trial failed to show any improvement above the placebo treatment. (2) They concluded that "there is great discrepancy between the intervention effects found in preclinical animal studies and those found in clinical trials of humans with most of these interventions rarely achieving successful translation".

According to the LA Times the Stanford study, "suggests animal research is riddled with bias that allows too many treatments to advance to human trials". (3) A researcher at University Hospital in Basel, Switzerland, Dr. Matthias Briel, admitted, "A lot of clinical researchers are not aware that animal studies are not as well planned as clinical trials". (4). The head researcher of the Stanford study said to Nature that there should be a database of all animal studies, similar to that for clinical trials. 

 

References

1. Evaluation of Excess Significance Bias in Animal Studies of Neurological Diseases. PLoS Biol 11(7) 2013: e1001609. doi:10.1371/journal.pbio.1001609

2. A phase III study evaluating the efficacy and safety of MBP8298 in secondary progressive multiple sclerosis. 2011. Neurology 77: 1551–1560.

3. Animal studies riddled with bias, report finds. LA Times 16 July 2013. http://www.latimes.com/news/science/sciencenow/la-sci-sn-animal-studies-bias-20130716,0,6790412.story 

4. Animal studies produce many false positives: Examination of neurological disease research shows pervasive ‘significance bias’. Nature news 16 July 2013. Nature doi:10.1038/nature.2013.13385 http://www.nature.com/news/animal-studies-produce-many-false-positives-1.13385

ARCHIVE