Wednesday, April 16, 2008

Problems with the private sector...

A topic that seems to be increasingly relevant in the last few days of posting (here, and here) is the invasiveness of corporate practice on other areas of life. So with that in mind, I thought I'd throw some news from Nature out there, in the form of this article.

So this article gives us a quite disturbing set of events whereby

"a company conducting a clinical trial, having its employees design the trial, analyse the data, write the paper, and then towards the very end recruit academic authors to put their name on the paper to give it that seal of supposed authenticity "

Two words for this: not cool. Peer-review for a number of years has come under attack for encouraging, among other things, nepotism and sexism. The literature is vast, and a study (albeit, a potentially rewarding and important study) in itself, which I don't feel confident trying to portray to you in a single blog post. What is important to note, is that in the Journal of the American medical Association (JAMA), where the study mentioned in Nature is soon to appear, this phenomena of 'ghost authorship' has been mentioned for at least 10 years.

What is the problem here? Well, apart from the fact that, as Nature mentions, this particular cache of studies are actually so doctored as to be plain ol' fraudulent in their content, it's also a subversion of the current system of scholarship practiced over the world. If the peer-review system wasn't on shaky footing before, it's certainly been shown to be in even more strife. The tobacco industry has been employing researchers for decades to attempt to show that smoking isn't as bad as everyone thinks (nice, guys), but this is a new level of subversion, the actual hiring of external researchers as front-men for internal research.

Priority is one of the big prizes in scholarly research. It always has been, as we aren't generally the type to earn the big bucks. But lo and behold, the private sector has found a way to tap into this prize, by effectively giving away research to motivate their own ends. It seems like a win-win situation: the company gets their research, and a researcher gets to be a first name. This is not so much an issue in philosophy, but in the natural and life sciences there is a huge emphasis on being a first name. I mean, computational science/high-end physics papers can have tens, occasionally hundreds of names on a paper. No one wants to be obscured behind the impenetrable wall of `et. al.' And hence the problem - the prize of research, priority, has been subverted.

How does one counter this? It's not easy, and will most likely require a restructuring of peer-review (a shout out to my friends who are big on `designing in ethics'). The first, if somewhat cumbersome method I can think of to alleviate this problem is to require a defense of a paper funded by industry, but I don't think we have enough peer-reviewers for this. The second would be a change in emphasis on the importance of priority in research in the current academic climate, but that's even harder to accomplish.

the method of discovery mentioned in Nature may show some promise, but requires a diversification of talents in the academic community. I'm talking about encouraging and promoting the meta-study: a study of studies. Hell, this might even allow us to get our highly critical social theorist brethren into the fray. To examine a wide range of studies on a topic allows us to see trends in data that may not appear otherwise.

I'll quote a personal example to illustrate: my honours thesis in nuclear physics revolved around the debunking of a number of popular ideas in a particular field of plasma physics. The problem was that people had initially misunderstood a set of phenomena, and had published based on this misunderstanding, which was never picked up. It, over the span of 30-odd years, became canon within this field, until one day my then-supervisor was alerted by a PhD student that the description of the system was false. As in, the data that people continued to collect, to describe other phenomena, was all based on an interpretation of a system which was false. The data had been replicated time and time again as a side product of other projects, and still no one noticed. It followed that an analysis of the literature pointed to how and where the misunderstanding occurred.

That was a benign example, certainly far removed in intention from the current situation, but I hope I get my point across- by looking at what has come before, we can see patterns that each individual study can't display. In my example, it was a misunderstanding of theory, in Ross' example written up in nature, it was the subversion of a practice within the scholarly methods of reporting. Both of these are important, but it takes a particular type of research to see where things are going wrong. Academic maintenance, so to speak.

I can't see that corporations are going to stop trying this. It's up to the academy to make it that much harder for private sector interests to subvert our practices in an effort to push products.

6 comments:

Adam Man Tium said...

I repeat yesterday's question. Why are you against freedom?
What about freedom of speech? Don't these slacademics have a right to publish other people's data in their own name promoting poor information and increasing the coffers of hard working CEO's?

Catherine said...

Adam, lay off the conservapedia!

I'm personally wondering what academic is going to lend their name to such an enterprise in the first place. Getting bumped up the list of names on a ghost-written article must trigger the "ummmmm hang on" button for any self-respecting academic, surely? Especially when everyone knows that big scary drug companies are evil, right?

Adam Man Tium said...

Whilst communing with different spirits on this issue they have alerted me to perhaps other ways of addressing this issue, rather than paranoid rantings.
Am I right in thinking that the problem here is the 'drift' of economic concerns into areas that should be largely independent of economic influence? Academic work, scientific research, politics etc are areas, dare I say spheres who should be relatively free of economic influence. Following Michael Walzer and his Spheres of Justice, the insertion of economic concerns into spheres like scientific research is indicative of injustice.
Clearly many areas interact, and sticking one's head in the sand and pretending that economic factors don't exist, or shouldn't have any say is no solution. Is openness and accountability enough? In a situation like the one mentioned in Nature, who should actually be held accountable and/or responsible? Punishing the scientists, the company directors, the publishers even may go some way to minimising the chance of these situations occurring again, but if the system remains intact, it is hard to imagine what real change would actually result.
The question for you ethicists, public policy folks and even the odd academic is how to redesign the system in a way that prevents the influence of the economic on non-economic spheres.

Catherine said...

I don't have a medical background at all, what do they have to do currently? I assume they have to make some sort of "This research was funded by Company X" ... any other sorts of disclosures like that?

Nick said...

Catherine:

you know, I'm not really sure how they demonstrate their corporate ties. it's an interesting question, to be sure. In this casem I think the objection occurs when you hire a scientist to act as an 'external authority,' one who in a way is removed from the corporation. A bit of the ``but don't take my word for it...'' phenomenon.

Adam Man Tium said...

I have just had a look at a single article in Nature Medicine, and the Institutional arrangements for each author are listed, as well as the role that each author played in the research, but I could see no direct mention of conlficts of interest or any other such things.
Based on three or four seconds of research, I would say that disclosure of interests etc could be an issue.
Aside from disclosure issues, it seems to me that some of the problem is caused by public-private partnerships. The reduced funding for research forcing researchers into private enterprise etc etc.