Lessons from the Vioxx Debacle: What the Privatization of Science Can Teach Us About Social Epistemology

Abstract

Since the early 1980s, private, for‐profit corporations have become increasingly involved in all aspects of scientific research, especially of biomedical research. In this essay, I argue that there are dangerous epistemic consequences of this trend, which should be more thoroughly examined by social epistemologists. In support of this claim, I discuss a recent episode of pharmaceutical research involving the painkiller Vioxx. I argue that the research on Vioxx was epistemically problematic and that the primary cause of these inadequacies was faulty institutional arrangements. More specifically, the research was organized in such a way as to allow short‐term commercial interests to compromise epistemic integrity. Thus, the Vioxx case study, in conjunction with numerous case studies developed elsewhere, provides strong reasons for believing that the privatization of the biomedical sciences is epistemically worrisome, and it suggests that the primary response to this situation should be a social, or organizational, one. What kind of organizational response would be most beneficial? I briefly discuss two prominent social epistemological proposals for how scientific research should be organized—namely those of Philip Kitcher and Helen Longino—and I suggest that they are incapable of dealing with the phenomenon of privatization. I then draw upon the Vioxx episode in order to outline an alternative suggestion for reorganizing certain aspects of pharmaceutical research.

Keywords:

• Vioxx
• Social Epistemology
• Privatization of Science
• Commercialization of Science
• Longino
• Kitcher

Acknowledgements

An early version of this paper was presented to the “Neuere Themen Working Group” in the Department of Philosophy at the University of Bielefeld, Germany, December 14, 2004. Thanks to Martin Carrier, Torsten Wilholt, and the other members of this group for their valuable comments. Another version of the paper was presented at the “Science, Democracy, and Economics Conference” in Madrid, Spain, April 11–13, 2005. Thanks to the Fundación Urrutia Elejalde and the Universidad Nacional de Educación a Distancia for organizing the conference. I would also like to thank Don A. Howard, the director of the Graduate Program in History and Philosophy of Science at the University of Notre Dame, who directed the doctoral dissertation out of which this work has arisen. This essay has benefited greatly from his comments upon earlier drafts. Finally, much of my work on this paper was supported by a National Science Foundation Graduate Research Fellowship.

Notes

[1] This is the theory behind the Bayh‐Dole Act of 1980. For discussion of this piece of legislation, as well as others that facilitated the present trends towards privatization, see Krimsky (2003) and Slaughter and Rhoades (1996).

[2] See, for example, Angell (2004), Avorn (2004), Bekelman, Li, and Gross (2003), Brown (2000, forthcoming), Cho and Bero (1996), Friedberg et al. (1999), Krimsky (2003), and Stelfox et al. (1998).

[3] Exceptions to this unfortunate trend include Brown (2000, forthcoming), Carrier (forthcoming), and Wilholt (forthcoming).

[4] Kitcher has also addressed the issue of organization in his book (1993).

[5] See, for example, Biddle (2006), Brown (2004), Kourany (forthcoming), Solomon (2001), and Solomon and Richardson (2005). For an exchange between Kitcher and Longino, see Kitcher (2002a, 2002b) and Longino (2002b, 2002c).

[6] This press release can be accessed on a number of websites. See, for example, http://www.pslgroup.com/dg/1FC1E2.htm (accessed June 1, 2006).

[7] Although the jury awarded the widow $253.5 million, Texas law has caps on the amount of punitive damages that can be awarded to a plaintiff; as a result, Ernst received only $26.1 million.

[8] This phenomenon of companies choosing to remain ignorant about certain questions that, if answered in a certain way, would be detrimental to their economic interests is also pointed out by Brown (forthcoming).

[9] Merck’s lack of candor with regard to the safety of Vioxx is the subject of a revealing warning letter from FDA to Raymond V. Gilmartin, the then‐President and CEO of Merck, on September 17, 2001. See Abrams (2001).

[10] In March 2006, a response by the non‐Merck VIGOR authors to the expression of concern was published in the New England Journal of Medicine (Bombardier et al. 2006). No corrections were provided. In response to this, the journal reaffirmed its expression of concern. For further details, see Curfman, Morrissey, and Drazen (2006).

[11] Marketing studies, which provide no novel scientific information and, in some cases, do not meet standard norms of scientific procedure, constitute a growing segment of many pharmaceutical companies’ research agendas. In many cases, a company’s primary purpose in performing such studies is to expose doctors to its drugs, to get them into the habit of prescribing its drugs, and to create brand loyalty among patients. The Advantage trial, for example, was created by Merck’s marketing department “as a promotional tool, to introduce about 600 doctors to Vioxx” (Berenson 2005a). For additional details on the increasing prevalence of marketing studies, see Angell (2004).

[12] Ghost‐writing is another tactic that is becoming increasingly common in pharmaceutical research. For further details, see Krimsky (2003) and Angell (2004).

[13] There is now a literature in philosophy on moral epistemology. See the essays collected in DePaul and Zagzebski (2003).

[14] See, for example, Chapter 9 of Angell (2004), in which she discusses the systematic influence of marketers on pharmaceutical research.

[15] See the literature cited in footnote 2.

[16] I criticize the ideals of Kitcher and Longino in much more detail in Biddle (2006).

[17] Kitcher’s ideal faces similar difficulties. For example, Kitcher’s ideal begins with a set of ideal deliberators who are somehow representative of the public. Again, these deliberators discuss, among other things, their preferences for which lines of research should be undertaken and then, in consultation with ideal, disinterested experts, settle upon a list of specific research projects. But this ideal, among other problems, simply glosses over the problem of how we are to involve the public in scientific decision‐making. The ideal begins with a set of ideal deliberators who are somehow representative of the public, and thus simply assumes, from the start, that the public is involved in an appropriate way. The real‐world problem of how the public should participate thus does not even arise. For further discussion of Kitcher’s ideal, see Biddle (2006), Brown (2004), and Longino (2002b, 2002c).

[18] Both Don Howard, in a 2002 reading group on Kitcher’s Science, Truth, and Democracy at the University of Notre Dame, and Miriam Solomon (2001, 144) have made this point as well.

[19] Organized skepticism, of course, is one of Robert K. Merton’s (1942) famous norms of science.

[20] James B. Conant was a prominent proponent of a version of an adversarial system of science in the 1950s (see Conant 1951, 1952). Arthur Kantrowitz proposed a different version in the 1960s and 1970s (see Kantrowitz 1967, 1976). I discuss both proposals, and I articulate how such proposals should be modified in the case of pharmaceutical research, in Biddle (2006).

Additional information

Notes on contributors

Justin Biddle

Justin Biddle received his PhD in the History and Philosophy of Science from the University of Notre Dame, USA. He is currently a Postdoctoral Fellow (Wissenschaftlicher Mitarbeiter) in the Department of Philosophy at the University of Bielefeld, Germany. His primary research interests are in the philosophy of science, social epistemology, and the ethics of science.