IJE Advance Access originally published online on January 19, 2008
International Journal of Epidemiology 2008 37(1):1-3; doi:10.1093/ije/dym289
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Editor's Choice |
Big business, big science?
E-mail: George.Davey-Smith{at}bris.ac.uk
What is the appropriate nature of the relationship between researchers and their sponsors or employers? This question receives an energetic outing in the current issue of the IJE. Neil Pearce—president elect of the International Epidemiological Association (IEA), the organization for which the IJE is produced—makes a compelling case that corporate influences can distort the knowledge base of epidemiology.1 He relates the story of how research on the adverse effects of an asthma drug, fenoterol, was attacked by epidemiologists working as the hired guns of industry, with clearly detrimental effects to the health of New Zealand, where this saga played itself out. His recent book, Adverse Reactions: The Fenoterol Story,2 details the whole sorry tale.
An energetic series of commentaries follow,3–6 which make the point that epidemiologists working full time for industry are in a different position to hired consultants (particularly when the consultants do not make their funding source clear). In the process of making such arguments it is even suggested that ‘balanced evaluation and discussion ... seldom sell books’,6 and perhaps the IJE has a conflict of interest in appearing to help promote its president's elegant monograph. Carl Phillips, who also administers a gentle slap to the IJE with respect to handling of one of his papers, suggests that ‘too many in the field feel that an ad hominem attack about someone's finances is legitimate scientific criticism, but actually criticising the substance of published research is strange and rude’.4
Most of our commentators appear to agree that full declaration of conflicts of interest is appropriate, although Phillips’ own ‘competing interests’ statement reflects frustration with this. Can full conflicts of interest really ever be documented? Some epidemiologists have taken to increasingly long generic online conflicts of interest statements.7 But what constitutes a conflict of interest? Is it just money—or, learning from the sociology of science, should we recognize cognitive conflicts of interest (for example, over-attachment to ones own ideas), interpersonal conflicts of interest, cultural conflicts of interest (would the eurocentric IJE ever publish a positive evaluation of female circumcision, whatever the quality of the science?), etc.? Viewed in this light, the recent entertaining autobiography by DNA pioneer (others use different terms) Craig Venter8 could be read as a long conflict of interest statement regarding any future work he does. For this brief ‘editor's choice’, so far my own conflicts of interest statement would be long—I have been critical of (and criticized by) the pharmaceutical industry for many years; Neil Pearce has bought me beer in the past (though less, I think, than I have bought him); Craig Venter has not responded to an e-mail I sent, etc. The IJE approach is to ask authors to declare what they perceive to be conflicts of interest, since alternative approaches are essentially impossible to police.
So much for big business. Increasingly, epidemiology is becoming ‘big science’, and this is reflected in the current issue of the IJE. Sample sizes of studies are growing dramatically. The first of our two cohort profiles, the Western Australian Family Connections Genealogical project relates to around a million people9; the second, The Norwegian Women and Cancer Study10 to only 165 000. Much of the impetus for the increasing expansion of sample sizes has come from genetic epidemiology. Tim Frayling11 and John Ioannidis and colleagues12 point out that the established effect sizes for the influence of common genetic variants are small, the number of hypotheses tested are large (requiring strong evidence of statistical deviation from the null to be credible), and therefore vast studies are required to advance knowledge. Adapting Moore's law (that digital capacity doubles every 2 years), a decade ago the diabetologist Edwin Gale advanced Gale's Law, ‘that the sample size considered adequate for genetic studies also doubles every two years’, with the implication that someone, somewhere, is cashing in. So far Gale's law seems to apply to genetic epidemiology studies, but how large can they get?
Our diversion documents the other end of the spectrum of the size of genetic studies, as Alun Evans reports on a n of 1 sib-pair study in Banbridge, County Down, where initially one of two brothers with methaemoglobinaemia was successfully treated with vitamin C, the other serving as the control.13 Remarkably, 60 years later the brothers were found and the mutation underlying their recessive congenital methaemoglobinaemia identified.
Meta-analysis is another example of big science in epidemiology, and as often is the case the current issue contains both an empirical meta-analysis14 and a methodological paper.15 The obvious hook for an ‘editor's choice’ regarding meta-analysis would be to refer to our regular ‘Cochrane Column’ feature, where we briefly report on a recent Cochrane systematic review. Unfortunately due to editorial slackness no such column appears in this issue; normal service will be resumed as soon as the responsible editor gets out of detox.
Finally, for those of you (who like me) thought we were attempting to remedy the clear gender bias in our ‘Reprints and reflections’ with Alison Glover's classic paper on surgical variations16–19, we provide a photograph of Dr Glover here (Figure 1).
|
References
1 Pearce N. Corporate influences on epidemiology. Int J Epidemiol (2008) 37:46–53.
2 Pearce N. Adverse Reactions: The Fenoterol Story. (2007) New Zealand: Auckland University Press.
3 Greenland S. Addressing corporate influence through ethical guidelines. Int J Epidemiol (2008) 37:57–59.
4 Phillips CV. Lack of scientific influences on epidemiology. Int J Epidemiol (2008) 37:59–64.
5 White A, Robinson N, Egger P, et al. Collaboration between industry-based and academic epidemiologists. Int J Epidemiol (2008) 37:56–57.
6 Haas J. Epidemiology and the pharmaceutical industry: an inside perspective. Int J Epidemiol (2008) 37:53–55.
7 Sackett DL. Commentary: Measuring the success of blinding in RCTs: don’t, must, can’t or needn’t? Int J Epidemiol (2007) 36:664–65.
8 Venter JC. A Life Decoded: My Genome: My Life. (2007) New York: Viking Books.
9 Glasson EJ, de Klerk NH, Bass AJ, Rosman DL, Palmer LJ, Holman CDJ. Cohort profile: the Western Australian family connections genealogical project. Int J Epidemiol (2008) 37:30–35.
10 Lund E, Dumeaux V, Braaten T, et al. Cohort profile: the norwegian women and cancer study – NOWAC – Kvinner og kreft. Int J Epidemiol (2008) 37:36–41.
11 Frayling TM. Commentary: Genetic association studies see light at the end of the tunnel. Int J Epidemiol (2008) 37:133–35.
12 Ioannidis JPA, Boffetta P, Little J, et al. Assessment of cumulative evidence on genetic associations: interim guidelines. Int J Epidemiol (2008) 37:120–32.
13 Evans A. Sibling trials in Banbridge, County Down. Int J Epidemiol (2008) 37:42–45.
14 Nnoaham KE, Clarke A. Low serum vitamin D levels and tuberculosis: a systematic review and meta-analysis. Int J Epidemiol (2008) 37:113–19.
15 Minelli C, Thompson JR, Abrams KR, Thakkinstian A, Attia J. How should we use information about HWE in the meta-analyses of genetic association studies? Int J Epidemiol (2008) 37:136–46.
16 Alison Glover J. The incidence of tonsillectomy in school children. Proc R Soc Med (1938) 31:1219–36. Reprinted in Int J Epidemiol 2008;37;09–19.
17 McPherson K. Commentary: James Alison Glover (1874–1963), OBE (1919) CBE (1941) MD (1905) DPH (1905) FRCP (1933): health care variations research then and now. Int J Epidemiol (2008) 37:19–23.
18 Burton MJ. Commentary: Tonsillectomy - then and now. Int J Epidemiol (2008) 37:23–25.
19 Wennberg J. Commentary: A debt of gratitude to J. Alison Glover. Int J Epidemiol (2008) 37:26–29.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||