FUNDING IN THE HEALTH SCIENCES
(August 2005)
These days, no idea or innovation can get off the ground without money, and scientific research is no exception. But unlike many commercial pursuits, research often does not produce lucrative rewards in the short term. Many research projects do not make marketable discoveries for years, sometimes even decades. And to make things even more financially difficult, many research projects require substantial investment prior to the actual research process in order to obtain proper resources and equipment. Because of this, the issue of funding becomes tricky business for many researchers. And it can become especially difficult to attract investors when you know that the odds of you discovering anything financially profitable are almost dismally slim.
Ninety percent of scientists that have ever lived exist today, and as these scientists struggle to find niches of uncharted territory in an overpopulated field of work, their projects rapidly become more specialized, to ensure that research doesn’t overlap. Unfortunately, the evolution of highly specialized research projects results in the evolution of research topics that are also not very applicable to the rest of the world. This becomes another obstacle to be overcome in the quest for funding. In spite of these problems, researchers still require reliable, independent financial support and their current method of procuring it is to apply for a grant.
Researchers presently use the grant-peer review system to get funding and this system has been criticized for diluting the effort towards pure, theoretical research by encouraging researchers to investigate applied science ventures of questionable scientific merit. To determine whether or not these criticisms are valid and to deduce where to go from here in improving the grant distribution process the development of this system needs to be properly understood and discussed.
From Prizes to Grants
It was actually quite usual in the seventeenth and eighteenth century for scientists to spend part of their private income to support their research [1]. Although this sharply restricted the actual amount of capital researchers could spend, it did allow them total freedom to investigate whatever they wanted (within financial means). However the explosive growth of scientific progress in the 19th Century quickly consumed more money than scientists could supply privately, and so new sources of funding became necessary.
France was the first country to provide government support for research projects, through the Academie des Sciences in Paris. Before long, the Academie was distributing privately funded prizes for discoveries made in select areas of the scientific field. But prizes required that the work be done before any money was offered. This often meant that those without their own resources couldn’t do any work at all. Thus, there was a growing need for a system by which worthy applicants could access funds prior to starting their research. Before long the peer-review system for scientific research was created.
By various tortuous paths, the Academie des Sciences in Paris and other institutions such as the American Academy of Arts and Sciences were able to convert prize money into grant money. The reasoning behind this change was that it turned out modest sums of money given to young investigators to undertake scientific research often resulted in a more effective use of resources than giving large sums of money as rewards for completed work. Soon enough it became normal for a scientist to look to other organizations for money instead of supplying it themselves.
The Origins of Peer-Review
The change from prizes to grants was a matter of efficiency more than anything else. Grants were simply a better way to utilize available funds. However, the emergence of grants as the prevailing financial support system also required efficient and just methods of distribution. Indeed, our current peer-review committees were originally created to resolve the issue of properly evaluating a scientist’s work, thus determining whether it is worthy of funding. The practice of peer-review began in France in the eighteenth century due to increasing competition for membership into the Academie des Sciences [2]. The Academie had severely restricted numbers, and election to membership was difficult. A formal procedure for the election of a new member was put in place during the first half of the nineteenth century. Candidates compiled and submitted notices des travaux, reports about themselves and their work, in order to appear before the Academie in the most favourable light. Candidates were also evaluated by the section of the Academie to which they had applied. By 1860, it had become standard practice for the senior member of the section to write a report on all candidates, assessing their research, their published works and their notices des travaux. Nowadays, not only grants and scientific memberships but also publications, appointments, promotion, in other words, the overall success of one’s entire career, depends on peer evaluation.
Contemporary Funding and allocation of Resources in Health Sciences
Funding sources for health research may be roughly classified in two groups: private and public funding. Private funding -with the exception of non-profit organizations- are usually targeted to areas of research that are more likely to generate large profits, such as the development of cosmetics, pharmaceuticals or technology. For example, Pfizer, a major pharmaceutical company that funds its own research, had a net income of 9,126 million USD last year. Bayer, another pharmaceutical company, generated a net income of 9,852 million USD in spite of the fact that one of their drugs, Baycol, used to treat high cholesterol, had to be withdrawn from the market for safety reasons in August 2001, following reports linking its use with a potentially life-threatening muscle disorder [3].
Public funds, on the other hand, are used in a wide variety of health research fields and support research with objectives that differ from the profit-oriented goals pursued by privately funded research. The objectives stated by the Canadian Institutes of Health Research (CIHR), the main federal agency responsible for funding health research in Canada, illustrates this, stating its commitment “to excel, according to internationally accepted standards of scientific excellence, in the creation of new knowledge and its translation into improved health for Canadians, more effective health services and products and a strengthened Canadian health care system” [4].
Public funds are also the main source of funding for health research and development (R&D) in Canada. In 2002 the public sector contributed 2.8 billion CAN to the health research budget, which is equivalent to 63% of all funds used in health R&D that year [5]. In fact, CIHR’s financial budget came to around 533 million CAN that year “to ensure talented investigators are provided with the resources and training needed to address the health challenges faced by Canadians” [6] . Even so, the private sector and drug companies in particular, are beginning to play a much stronger and more noticeable role in funding health research in Canada. The most obvious indication of this is that universities, the most expansive performers of health R&D in Canada [5], are receiving more money from private resources. Private for-profit firms were funding 12% of university-based health research in 1996 compared with 2% in 1976. Both government and not-for-profit funding at Universities dropped from 73% and 25% in 1976 to 70% and 18% in 1996, respectively [7]
Although the peer-review system has been in place now for some time, in the last forty to fifty years some critics have begun to oppose it. Their main criticism is that the funding system has created a legion of professional researchers more interested in fostering careers than advancing the field of science [8]. Several studies have since investigated how the granting system influences who and what research gets financial support. A study conducted in the seventies by Cole et al.(1981)[9] on the National Science Foundation’s (NSF) peer review showed a high correlation between reviewer ratings and grants made, as expected. In other words, the NSF review ratings were the primary determinant of an applicant’s success. They also showed that the awarding of grant money to an applicant was independent of their previous scientific performance, affiliation and professional age. Surprisingly, they found that while the fate of a particular grant application depended partly on the characteristics of the proposal and the principal investigator, the rest was dependent on whether or not the program director’s choice of reviewers were in the applicant’s favour. And because the reviewers disagreed so dramatically among themselves, it was largely the random chance of having favourable reviewers that determined whether an applicant received funding.
Disagreements are prevalent not only within grant committees, but also among reviewers of manuscripts submitted for publication. A study by Rothwell and Martyn (2000) [10] on reproducibility of assessments made by independent reviewers in the clinical neuroscience field showed that agreement between the reviewers as to whether manuscripts submitted for publication should be accepted, revised or rejected was either poor or not significantly greater than what was expected by chance. Regarding abstracts submitted for presentation at conferences, only 10-20% of the total variance of the abstracts’ scores could be accounted for by the differences between individual abstracts. Thus, agreement between reviewers in clinical neuroscience was shown to be a little greater than would be expected by chance alone.
Several reasons have been given to account for these differences, but the basic idea behind all of them is that a proposal that is familiar, supports the reviewer’s own ideas, validates hypotheses that serve powerful interests or comes from recognized candidates or institutions is more likely to be accepted and funded. This idea has been supported by different studies with different degrees of emphasis [11, 8]. Particularly, a study by Glantz and Bero (1994) [12] assessed the level of interest and enthusiasm peer reviewers might have for a given proposal, depending on their own personal involvement in the field. This personal involvement was measured by the number of publications each referee had in the field they were judging. The authors concluded that reviewers that were not actively involved in the field effectively precluded those that had.
Because there is little doubt that the peer-review system has its problems, many scientists -and policy makers- have proposed mechanisms to improve it. One of these proposals for example, is to systematically conceal the author’s identity to the reviewers in order to reduce favouritism and discrimination by forcing reviewers to judge the proposal on its intrinsic merits. Another possibility is to reveal the referee’s identity to the authors, thus reducing unkind and unfair statements that anonymous judges are more likely to enunciate than known ones. Regarding these points, two studies by van Rooyen et al. (199813,1999 [14]) showed that the first of this strategies -anonymity of authors-would imply almost no extra cost on the editor or the reviewers. On the other hand, asking reviewers to consent to being identified to the author had no important effect on the quality of the review, the recommendation regarding publication, or the time taken to review, but it significantly increased the likelihood of reviewers declining to review.
Conclusions
Research funding is the lifeblood of science, the most important single determinant of its current organization. Securing and maintaining research grants to buy expensive operating equipment, pay research staff and maintain the university premises has become one of the principal investigator’s main concerns. The ultimate party responsible for the assignment and distribution of financial support are the scientists themselves, who are appointed to committees to evaluate and rank colleagues’ work. This system is not perfect, because scientists are only human, and can’t help but pursue their own objectives quite passionately on occasion. This inevitably colours their assessment of their peers’ work. This in turn, can lead to differences in judgment among reviewers, resulting in the odds of being awarded grant money amounting to little more than the odds of being chosen randomly.
It is certainly hard to evaluate the entire peer-review process after such a cursory glance at the research. However, some important points can be made. First, the actual system has generated an unprecedented number of scientists, that working under the above outlined conditions, have substantially improved the health of the Canadian population. Secondly, our research funding system does indeed have flaws, but recognizing them is the first step towards their resolution.
References
1. Crosland M., ‘From Prizes To Grants In The Support Of Scientific Research In France In The 19th Century: The Montyon Legacy’, Minerva 1979, 17 (3), 355-380.
2. Crosland M., ‘Assessment by peers in the nineteenth-century France: the manuscript reports on candidates for election to the Academie des Sciences’, Minerva 1986, 24 (4), 413-432.
3. Hoover’s Company Database http://www.hoovers.com/
4. CIHR’s mandate http://www.cihr-irsc.gc.ca
5. Statistics Canada, Science Statistics 2003, 27(6).
6. Health Canada.http://www.hc-sc.gc.ca
7. Silversides A., ‘Private Sector Becoming the Key to Research Funding in Canada’, Can Med Assoc J 1998, 159 (4), 397-398.
8. Savan B., ‘Science Under Siege’, CBC Enterprises 1988, 130-131.
9. Cole S., Cole J.R. and Simon G.A., ‘Chance and Consensus in Peer Review’, Science 1981, 214, 881-886.
10. Rothwell PM and Martyn CN, ‘Reproducibility of Peer Review in Clinical Neuroscience. Is agreement between reviewers any greater than would be expected by chance alone?’ Brain 2000, 123 (Pt 9), 1964-9.
11. Kuhn, T. The structure of scientific revolutions. University of Chicago Press, 1962.
12. Glantz S.A. and Bero L.A., ‘Inappropriate and appropriate selection of ‘peers’ in grant review’, JAMA 1994, 272 (2), 114-117.
13. van Rooyen S., Godlee F., Evans S., Smith R. and Black N., ‘Effect of blinding and unmasking on the quality of peer review: a randomized trial’, JAMA 1998, 280(3), 234-7.
14. van Rooyen S., Godlee F., Evans S., Black N. and Smith R., ‘Effect of open peer review on quality of reviews and on reviewers’ recommendations: a randomized trial.’ , BMJ 1999, 318(7175), 23-7.