Quality and validity in animal research

Published on October 31, 2016   35 min

Other Talks in the Series: Animal Models in Biomedical Research

0:00
My name is Anna Olsson. I'm a researcher at the Institute For Research and Innovation in Health at University of Porto in Portugal, the i3S. And I will be talking to you today about quality and validity in animal research.
0:18
So what's the reason to think about translation, quality, and validity of animal research? Well, one important reason is that some of your critics will be thinking about this.
0:33
There is the traditional criticism which you may be familiar with from some of the anti-vivisection organizations, websites, and reports. That's the criticism which is based on general arguments of biological differences between species. This is often anecdotal and not based on systematically collected data or in-depth analysis. And it's published outside mainstream biomedical science journals, often in books or in reports, or pamphlets from the organizations producing the information. More recently, over the last 15 years, a different kind of criticism has appeared which is based on meta-analysis and systematic reviews of large number of data, large amounts of data, and which is published in top biomedical science journals.
1:28
And if you haven't read these papers already, I highly recommend that you'll look them up after this lecture. These are two papers published in two of the top Plos journals, in Plos medicine we've got, "Can animal models of disease reliably inform human studies?" A pretty provocative question. And assisted to that study in Plos biology, "Publication bias in reports of animal stroke studies leads to major overstatement of efficacy." These are two papers which I will come back to, they come out of the CAMARADES group and I will use some of their results later on in this talk. The third paper published in Plos One, "Survey of the quality of experimental design, statistical analysis and reporting of research using animals." If you're familiar with the Arrive guidelines and I hope you are, then this is the paper batch, led to the publication of the Arrive guidelines for how to report research with animals.
2:31
And all this discussion actually led a group of researchers to publish a rather provocative paper in British Medical Journal, where they ask, where is the evidence that animal research benefits humans? So quite a lot of discussions going on right now, quite a lot of issues to attend to, presenting us with challenges.
2:54
We can say that research with animals informs us about, how the specific factor, affects the specific species, under the specific circumstances. And we should be aware that it provides no exact answer on the situation with other factors, in other species, and under other circumstances.
3:15
But usually we're not satisfied with these cautious assumptions, we actually usually want to make much more far-reaching assumptions. We want to assume that the results from the sample represent the population the sample was drawn from, which if we bring it down to a concrete animal research example, means that what I found in my 12 male 6-week-old C57BLACK6/J mice is true for all male 6-week-old C57BLACK6/J mice. We want to assume, to make the assumption that the results translate into other populations, which means that what I found in my 6-week-old mice is true for young male of other species. And quite often, we also actually want our results to have clinical relevance which then would mean that, what I found in my male mice applies to a population of human patients with the disease that I'm trying to model in these mice.
4:20
So if we want to illustrate these examples in terms of the different types of validity that they represent, then the first example, that the results from the sample represent the population the sample was drawn from, that's what we usually call internal validity. That the results translate into other populations is what we usually call external validity. And that the results have clinical relevance is what we usually call predictive validity.
4:52
So when is the right time to think about validity and translation? We need to think about that when we design and carry out studies, so to make sure that the results can be relied on. We also need to think about this when we look at the results of other studies if we aim to base our own research on these studies. This is for us to know that the results are really what they look like. And we also need to think about validity and translation, when we evaluate proposed studies for funding and authorization. This is to ensure that money and animal lives and welfare are not spent on fruitless research, research that is simply not designed in a good enough way for it to produce useful results.
Hide

Quality and validity in animal research

Embed in course/own notes