Most Animal Research Studies May Not Avoid Key Biases

A recent article in Science Magazine reveals that the majority of animal research studies are not protected from the kinds of biases that are routinely guarded against in human research, thereby often leading to misleading results. For example, a large number of seemingly successful animal studies reported on in scientific journals are, in reality, either impossible to replicate in larger animal studies and/or in human clinical trials. This means, in turn, that many drugs tested on animals may appear to be much more effective for humans than they actually are.

Neurologist Malcolm MacLeod of the Centre for Clinical Brain Sciences (University of Edinburgh) conducted an analysis of 2500 journal articles and discovered that that the bulk of animal studies on drugs for the treatment of eight human diseases failed to adopt and/or address their implementation (or lack thereof) of four essential measures required to prevent the presence of biases in research: randomization (to indicate that both the random and control groups are not manipulated to produce desired results); blinding (so that the researchers do not know what animal(s) underwent what procedure); calculation of sample size in advance (to prevent accumulation of data until desired outcome is achieved); and statement of conflict of interest(s).

These measures are standard practice in human research but are not as yet required in animal research, which suggests that a double standard is in place and that animal experiments are not subject to the same degree of scrutiny as human clinical trials.

Click here to read the article in its entirety.