Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Disclaimer: The Parliamentary Library does not warrant or accept liability for the accuracy or usefulness of the transcripts. These are copied directly from the broadcaster's website.
Are big ideas too entrenched? -

View in ParlViewView other Segments

Norman Swan: Hello, and welcome to this week's Health Report with me, Norman Swan. Today, how court actions for deliberately infecting someone with HIV may be affected by changing scientific knowledge. How Australian research findings have been fast-tracked into clinical practice, potentially helping thousands of pregnant women.

And a leading critic of the current state of medical research argues that big ideas in medical science have become entrenched, creating fads and perhaps taking resources from more productive areas. One of the people arguing this is John Ioannidis who is professor in disease prevention and health research and policy at Stanford University in California. His targets include personalised medicine, stem cell research and genetics, particularly genomics.

John Ioannidis: There are a number of ideas that have dominated biomedical research and biomedical research funding over the last several decades, and these include lots of concepts around biology and genetics, genomics in particular, but also any sort of big data and large-scale information where you can combine in-depth knowledge of biology along with in-depth recording of technology and data at high dimensions to try to identify best treatments, best ways to preventing disease, best ways to inform people to do the right things, and best ways to help people live longer lives.

So this is where most of the funding has gone, at least in research savvy communities that spend tens of billions of dollars on biomedical research. Apparently, even though these ideas seemed to be fascinating, maybe they haven't really delivered what we have expected them to deliver.

Norman Swan: What's your evidence for saying that?

John Ioannidis: So I think that we need to disentangle the scientific curiosity component, which is very exciting, from what have we learned that can really make a difference in people's lives and in saving patients or making them live longer and better lives. So genomics, and under genomics one could include not only genetics but many other biological platforms that we can measure lots of things as well as pharmacokinetics where you try to identify genetic variation that would tell you which drug is the best to take if you need to take some drug or something else to treat some disease. Plus prevention based genomics where you use information to make sound choices about how to prevent disease. And other things like gene therapy where you try to correct the genetic effects that underlie disease. So all of these are fascinating ideas. I think we have learned a lot about biology. But if you look at how many applications have really moved the stage of affecting people's lives and really being documented to be effective and making a difference, there is very, very few.

The same applies to other technologies in that sphere, like stem cell research. Again, the concept of understanding how stem cells work is a fascinating question from the biology side. But pretty much the randomised trials that we have done have not really shown any major benefits from stem cell interventions, much like the clinical trials for gene therapy have not shown much benefit. So even though the science sounds fascinating and very complex and exciting, we don't have applications that really make a big impact on human health.

Norman Swan: So in your analysis, you talk about curiosity driven research, you've got all these researchers out there looking at what genes, how they work at the molecular level, talking about precision medicine or personalised medicine, all really hopeful things. And you argue in your paper, well, those ideas have become entrenched and therefore you've got billions of dollars of almost fixed investment…even though it might change between research groups, it's a fixed investment almost in those areas. What is the problem here? Is the problem the curiosity driven research is driven in the wrong direction of curiosity? Or it's a failure of implementation based on existing knowledge? Or we just haven't accepted they are a dead end?

John Ioannidis: I'm very much in favour of curiosity driven research and blue-sky science. In fact one of the main problems for these ideas having been entrenched in a research funding universe is they do not allow other curiosity driven research that is even more bold and high risk, and not within these big domains that are easily recognisable to thrive.

So there are two problems. One is that we cannot allow to shift the funding to truly high risk ideas that would explore entirely new concepts because there's tens and hundreds of thousands of scientists who are entrenched in these areas and they pay off in terms of funding, even though they don't really get very far.

And second, we are not really getting what we thought that these technologies and that type of data and information would deliver. But the precision medicine is probably one of the latest terms to describe the same concept that has been out there for many, many years. And even though there have been few successes with some information being identifiable that could inform a few people on what is best to do, these are really the exceptions, these are really rare exceptions. In the large scale of things, most people have not really benefited from precision medicine and I don't think that they will benefit. By definition, precision medicine is looking at information that each time will benefit very, very few people. At the extreme it will be just a single person who has a given biological profile and then we try to find out what to do with that single person. But we are talking about a global population of 7 billion, not easy to catch up just by adding one person at a time.

Norman Swan: So you're saying that you are sceptical that precision medicine will ever scale up.

John Ioannidis: I think it will scale up to some extent but it will be nowhere close to catering to the needs of population health. It will probably help hundreds, maybe thousands of people, but not hundreds of millions and billions that we are interested in.

Norman Swan: So my question comes back to what I asked you earlier which is is this a problem that people call the Valley of Death, that we've got all these great ideas coming out of curiosity driven research and nobody is taking them and applying them, or the results of curiosity driven research are so complicated and dense and difficult than in fact it's really an almost impossible task. We haven't come up with the findings that actually are translatable.

John Ioannidis: I think it's a combination of both. There is an issue of complexity and not having easy answers, and also a difficulty in translation and implementation for things that we know, and if we could get them to work then they should work.

Norman Swan: And one example of that that you use in your paper is type II diabetes where it was thought that you would find the genes for type II diabetes but in fact it turns out to be a jungle of genes.

John Ioannidis: Exactly. So I think this is fascinating knowledge. Personally I have worked in that space, and some of my most brilliant colleagues have been leaders in that space, and it's very interesting to find out that there are all these hundreds of genes that may modify a little bit the risk of getting diabetes. Hopefully we should learn something from these genes, but there are so many and so complex and it's very difficult to make real progress and say that we are using this information to eradicate diabetes or even just to diminish diabetes around the world.

Norman Swan: So how do you disentangle…if you say it's both, it's the failure of translation of existing knowledge and that the curiosity driven research has come up with stuff that you will never be able to translate, almost a holistic view of that, how do you get beyond that?

John Ioannidis: My suggestion is that we need to invest more on very high risk, innovative, disruptive, out-of-the-box, blue-sky science. So instead of just going after that low-hanging fruit that seems to be so easy to churn more data and more and more of the same, I would argue we need to shift a little bit more to allowing ideas that are not within the mainstream to flourish. The failure rate, even for genetics before the genetic revolution would have been very high, but nevertheless we decided to go ahead in that direction and we did get some very interesting insights into the genetic architecture. At least we know that it's very, very complex.

So I think that there is lots of scientists who would be willing to explore high risk paths if they were supported and they didn't have to just do the type of research that hundreds of thousands of other scientists are doing. The second component is that we do have a number of things that we have discovered that do seem to work, and we have identified some major problems that hopefully should be easy to tackle and they could translate to saving many millions and potentially billions of lives.

To give you one example, smoking; we know that smoking is going to kill about a billion people in the next century unless we do something about it. And we also know that we have effective interventions. But these interventions do not really depend on complex biology, they do not depend on complex informational systems, they are very simple; just try to get rid of smoking by increasing taxes, by imposing penalties, by having laws that are pretty strict about advertisement and…

Norman Swan: But that happens in a different box from…

John Ioannidis: It's a completely different box. It has no strong biology and no fascinating complexity of biology behind it, but they are simple tasks and they are simple moves that if we adopt them more globally and more consistently we can save hundreds of millions of lives.

Norman Swan: How do you fix up what you are alleging which is the groupthink in organisations such as the National Institutes of Health or the National Health and Medical Research Council in Australia or the MRC in the United Kingdom, which is well recognised is that you've got an area of research which has forged a path, they know how to look at it and make assessments of it, and the people sitting on the committees are doing that research. So you get this sort of funnel. And organisations like the Gates Foundation and others have recognised this problem which is why they create the challenges so that you've got the disruptive technology on the side which loosens up on that. But the traditional funders who have the billions of dollars in addition to the foundations have not been very good at that.

John Ioannidis: I think that one needs some higher-level auditing of what's going on. As you say, some of the most successful scientists are leading these funding institutions. They are very smart, very successful in what they have done. But I think that they have a viewpoint that is pretty much stuck to their own universe, and their universe can only extend that far. So you need some oversight, you need someone who would be looking out-of-the-box and be able to tell these people that this is great, that's wonderful that we have reached so far and we have learned so much, but we really need to invest also in some avenues that probably will be more productive and would lead eventually to saving more lives of people in our country and globally.

So I think it's not easy. I think that the public could play a role, I think the public needs to be informed about what we know and what we don't know, and where the biggest gains lie and how could we get there. Research is wonderful, I think that investing more in research and in science as a society is one of the best investments, probably the best investment that a society could aim for. But at the same time there is a need to invest in things that could really make a difference, and it would be a pity if we don't do that.

Norman Swan: John Ioannidis, thank you very much for joining us on the Health Report.

John Ioannidis: Thank you.

Norman Swan: John Ioannidis, who is professor in disease prevention and health research and policy at Stanford University in California.

You're tuned to the Health Report here on RN, ABC News Radio and CBC Radio across Canada. I'm Norman Swan.