Sunday, January 27, 2008

Scientific Integrity

I came across two articles just this last week which, yet again, run amok with the Peer-Review system: Sex Bias in Science and Ghost Management.

The latter was of particular interest to me.

"There are many reports of medical journal articles being researched and written by or on behalf of pharmaceutical companies, and then published under the name of academics who had played little role earlier in the research and writing process. In extreme cases, drug companies pay for trials by contract research organizations, analyze the data in-house, have professionals write manuscripts, ask academics to serve as authors of those manuscripts, and pay communication companies to shepherd them through publication in the best journals. The resulting articles affect the conclusions found in the medical literature, and are used in promoting drugs to doctors."


This is not a new given and hardly surprising. Pharmaceutical companies are first and foremost businesses. It is only natural then, that they would explore all venues possible, to promote their products. A well-run business also strives for cost-efficiency, which in this day and age, generally means you outsource certain tasks to specialised companies. Scientific copy-writing is a specialised skill. Any academic knows how labour-intensive the process of writing and submitting a journal paper can be. Whilst the article above talks about "extreme cases", I would dare suggest this approach is the norm rather than the exception.

The article goes on to state: "Ghost writing and honorary authorship are not in and of themselves scientific problems, though they become so when they shape science to meet particular interests."

A fair point. As a Scientific Management Consultant, I spend about a third of my professional time on copywriting. Whilst I do this within my function as a Management Consultant, I am first and foremost a Scientist, motivated to communicate science to a variety of target audiences. However, scientific copywriting is by nature, an arena where commerce aligns with science. After all, a client raises a purchase-order with you for a specific document and you create this document according to your client's specifications. The key here is to safe-guard your scientific integrity. Your briefing may include a focus on certain marketing messages, but in the end, your responsibility is to create a scientifically balanced document.

I can't help but feel that the focus of articles such as the above, point their arrows at the wrong target. The influence of pharmaceutical companies on the integrity of our research and publications, is but a symptom of the underlying disease. The pressures on the scientific world are rising. For centuries, scientists were protected by an Academic Bubble that was subject to few, if any, external pressures. Scientists were relatively free to research whatever topics they were interested in and publish however frequently they saw fit.

The last few decades, the business world has taken an active interest in science. And Academics increasingly seize opportunities to cash in on their reputation and research efforts. The Academic Bubble pretty much burst from both inside and out.

Inevitably, the pressures on scientists are rising. Modern academics are measured not only by the quality of their work, but equally by the rate of their publications and the amount of funding they can generate for their research institutions.

It should be no surprise then that these pressures gradually erode the scientific integrity. Most of us remember the downfall of MIT Professor Van Parijs, who regrettably falsified study data, most likely after caving in to internal pressures to perform.

I think the target question here is: how do we, scientists, maintain our scientific integrity when faced with external pressures whose interests are differently focused. Blaming pharmaceutical companies for the downfall of scientific integrity seems tantamount to blaming Maccy D's for the obesity epidemic.

Scientists too, are subject to evolutionary principles: when faced with changing external pressures, we need to evolve, select the best new approach and amplify the approach that works. We've been protected by the bubble for too long. It is time the scientific community as a whole scrutinizes its own approach to these new challenges. Much like the medical profession has its medical oath and ethics, scientists would do well to create a new integrity frame for themselves to abide by. It will be impossible to avoid rogues amongst the pack, but we need to re-design our institutions and processes such that the opportunities for these individuals are minimised.

We could start by giving the Peer-Review system an overhaul. By severing links between Publishing houses and copy-writing agencies (Such as Elsevier and Excerpta Medica), and by bringing back the independent status of our Journals. We should redefine the standards against which our researchers are judged, to ensure scientific integrity and freedom within the new science-business merger.

Science and Business interests CAN co-exist. But it requires an intensive effort from our end, to define the framework within which this can happen. In the end, this will serve the purpose of the business world best by raising the credibility of their public image, and it will serve our interest by safe-guarding our intellectual integrity. The opportunities are immense. The pitfalls, dangerous. Scapegoat-ing will do little in terms of constructive efforts to maximise this new environment. Perhaps its time we turn our scientific methods on ourselves at last.

4 Comments:

At 2:01 pm, Blogger fi said...

This comment has been removed by the author.

 
At 2:21 pm, Blogger fi said...

Surely integrity can only be maintained with absolute transparency. In such an information driven society, any attempts at subterfuge will only be regarded with suspicion.

By employing "ghost-writers" and using the names of indirect associates with which to publish, it would appear to the sceptic that the pharamceuticals have some hidden agenda they are attempting to pass off as unbiased findings. If there were complete transparency, and all stake-holders were commonly known, at least people would be able to make informed decisions as to the level of credibility that should be assigned to the findings.

It is inherently paramount for good science to question the integrity of findings. In the mix of business and science, eyebrows will be raised all the more, as they should be. I would certainly be skeptical of an in-house consultant ecologist employed by a development agency! People just want to know if there are ulterior motives to the messages conveyed.

I can fully understand why companies employ experts such as yourself to write on their behalf, but to keep the sceptics happy there needs to be clarity about the intentions of interested parties, particularly when their are influential (and highly profitable) results at stake.

For example, as a consultant, how would you know yourself if you are being directed subversively? Do you get to verify the results and analyses from a study? This external scrutinity would obviously lend a lot more credibility than if you are handed results and just instructed what to write about them.

 
At 2:52 pm, Blogger Dr.Pew said...

I think the essential issue here is that business interests and science interests often don't line up.

First off, like you say, science demand transparency, openness and free exchange of information. From a business perspective, that is simply not possible. Businesses need to protect their intellectual property and strategies if they want to survive in the market place. They cannot afford bad publicity or results.

Whilst this is not exactly applaudable, we have to work with a few givens here. If we could start re-building society from scratch, then we could perhaps design a business world that functioned on entirely differently principles. But that's just not a viable option. So if we have to work with our current set of parameters, we have to accept that science and business operate on very different principles and perspectives. The key question is how to bring those together.

Second, there is no real hidden agenda. Everyone knows that businesses will promote their products in the best possible light. This is no secret and can therefore hardly be considered a hidden agenda. Moreover, the publications will generally mention either a funding or collaborator link to the company in question. So we generally are in the know.

Third, the pharmaceuticals and biotechs need to do their own research to back up value and health-economic dossiers for purposes of FDA approval and reimbursement. Because of the aforementioned premise however, people automatically assume that because it is research done by businesses, it will automatically be biased in their own favour. If pharmaceutical companies commission academic researchers to do their research for them, people generally still assume it will be biased, because of where the funding is coming from. So, it's pretty much a catch 22 situation here. Admittedly, some companies have shot themselves in the foot by validating people's suspicions of un-kosher reporting. But then, scientists have been known to falsify or leave-out essential data too. Rogues exist everywhere.

The point is that this merger of business and science has started to connect the "profit purpose" and its associated "bias" into scientific research. I merely want to point that our mechanisms for control are as good as non-existent.

Copy-writers should indeed see the study data. They often do. Our consultancy has a strong ethic in that we will not go against an individual employee's integrity. If no scientist is willing to blindly endorse someone else's claims, then perhaps we're all a bit closer to regaining that integrity.

You say you want all information to gauge the credibility and validity of study results yourself? Totally fair. The question is, is anyone really unbiased? If you get a research paper in your hands that states it's been carried out by Pharmaceutical Company X because they have product Y in the pipeline, won't you automatically discard it as biased? What controls do YOU need to assess that validity? How can we remove bias from either side? That's the point I was trying to make. We need new measures of control, and I don't think we quite know what those should be.

 
At 12:51 pm, Blogger Disco said...

Coincidentally, various aspects of publishing have been the hot topic here just recently, with it looking increasing likely that the next generation of RAE is going to be largely bibliometric driven and the whole question of open access publishing.

Peer review – the unsung hero and convenient villain of science

How to get good science

Challenging the tyranny of impact factors

Questionable judgement?

 

Post a Comment

<< Home