Saturday 14 July 2012

Dangerous Results: To Publish or Not To Publish?


Gunnar De Winter

So you thought the avian flu controversy was over? Far from it. A little while ago, two studies on the avian flu H5N1 sparked some controversy. Both studies detailed how the researchers were able to produce a flu strain that, in contrast to the known natural strains, would be highly transmissible between human beings. This research has potentially dangerous consequences. Debate ensued. Should these studies be fully or partially published, or not at all? Both the National Security Advisory Board for Biosecurity (NSABB) and the World Health Organization (WHO) issued press releases describing their (different) recommendations. Editors of Nature and Science postponed publishing the respective studies. But this is part of a wider issue: should studies with potentially dangerous results be published or not? Let’s look at some options.

Don’t Publish
Simply don’t publish the studies. Problem solved. Or not? This is the least popular ‘solution’ for the problem. An even more extreme version of this option is to simply not allow research that might yield dangerous results. But luckily almost nobody seriously considers this. After all, one can’t predict potential applications of research. And besides potential misuse, there might be great benefits as well (such as vaccine development in the case of the avian flu studies). Moreover, freedom of research is at stake here. If all research that might be abused by others is suspended, surely the scientific enterprise will wither until almost nothing remains. So, this option provides no solution. Moving on, then.

Publish Partially
Avian influenza A H5N1 viruses
Publish the studies, but leave out some key methodological details (which was the NSABB recommendation for the flu studies). Intuitively, this appears to make (some) sense. But there are problems here as well. First, who decides what to leave out? Furthermore, if other researchers want to build upon the results, they should be able to acquire the omitted details. But at the moment, there is no system to properly coordinate this. Finally, some people point out that just knowing that it’s possible is enough for others to figure it out. It might take a bit longer, but they’ll get there. Well, it seems option two also has its problems. On to the final option.

Publish Completely
Just publish the studies in full (the WHO recommendation). Most scientific research comes with a risk of potential misuse. This, however, should not stop enquiry. Besides, the best protection against abuse, so some argue, is to spread knowledge about whatever is being researched. In case of the flu studies, publishing the studies in their entirety is the best chance of finding a vaccine, thereby counteracting possible abuse. Scientific research in itself is not good or bad. It’s how it is used by people, or by society. The solution therefore is not to curb research, but to promote ethically and morally sound use of scientific knowledge. Achieving this, of course, is not an easy task, and there are many questions to be addressed, but it seems to be the best option we’ve got…

For more information about the avian flu debate:

Nature’s Mutant Flu News Special (http://www.nature.com/news/specials/mutantflu/index.html), where the latest news concerning the studies and several opinion pieces are aggregated.

A similar news and commentary collection can be found on Science’s Public Health, Biosecurity and H5N1 feature (http://www.sciencemag.org/site/feature/data/hottopics/biosecurity/index.xhtml).