Commercial funding carries a hefty price tag

Industry provides essential funding but can present researchers with a conflict of interest, argues David Weatherall.

For vice-chancellors of impoverished universities, the growing links between academia and industry and the resulting sources of income must be welcome. The situation is equally attractive to academics who see it as a way of augmenting their miserable pay.

But the increasing pressures on universities to move in this direction from government funding bodies are presenting problems that could have serious implications. Nowhere are these stresses and strains more evident than in biomedical research.

Recently, the New England Journal of Medicine, which justly prides itself on its particularly tough conflict-of-interest guidelines, had to apologise publicly for falling standards over the past few years. An internal inquiry disclosed that nearly half the 40 articles on drug therapy published since 1997 were written by authors with serious conflicts of interest based on their relationship with the manufacturers of the drugs in question. Commenting on this revelation, the editor of another medical journal said that it was impossible to find a well-informed commentator on any medical topic who had not had money from the pharmaceuticals industry.

At the same time as this article appeared, Harvard Medical School announced that it was planning to ease its conflict-of-interest guidelines to increase opportunities for researchers to profit from their dealings with companies. The school said it was losing staff to universities with less stringent rules. One staff member said that, even with the revised policy, Harvard will be only as permissive as many other leading US universities, including Stanford and Johns Hopkins.

These events are only the tip of the iceberg of the disquieting press that the biomedical research scene has received recently. The tragic death of a young patient following an experimental procedure directed at gene therapy for a metabolic disease has raised grave concerns about the quality of clinical research at the interface between academia and industry.

Reports have appeared in the US press of authors agreeing to become ghost writers for reviews of drug efficacy written by companies and of academics allowing their names to be used on papers concerning drug trials carried out by a company, in which the author played little or no part.

With increasing numbers of university research workers developing close relationships with industrial sponsors, or founding start-up companies of their own, there are equally serious possibilities for conflict of interest within university laboratories. In a busy research laboratory in which a number of projects are being carried out simultaneously, it is often difficult to follow flows of funding. Thus it is quite possible for work to be carried out for the scientists' companies that may be funded for other purposes by granting bodies. And there is the question of confidentiality and an inability to discuss work openly within university research groups because of the constraints of industrial secrecy.

These problems will become greater over the next few months following the completion of the human genome project. Universities and industry will increasingly require each other's expertise; industry will need access to well-defined patient populations or those with individual diseases, while academia will require expensive technology to reap the fruits of the post-genome period.

Some years ago two US scientists, Julius Comroe and Robert Dripps, obtained a consensus on the most important discoveries that had been made for medical practice and analysed the original papers on which this work has been based. They found that nearly half the most important advances stemmed from original work based on curiosity-driven science rather than from research directed towards a particular goal. If similar opportunities are not to be lost in the future, and if universities are to maintain their integrity and ability to pursue knowledge for its own sake, governments and granting bodies will have to temper their pressures on research workers for early and exploitable results.

The spate of problems at the industry-university interface suggests that biomedical research is being pushed much too hard towards rapid exploitability. In the case of the clinical sciences, on both sides of the Atlantic this trend is being exacerbated by lack of funding, driving research workers towards industrial support.

These might be teething problems, but there are few indications that, in their rush towards the pot of gold offered by completion of the human genome project, either government or universities are fully aware of where this might lead.

David Weatherall is regius professor of medicine, University of Oxford. A longer version of this article appeared in The Lancet.