The Annals of Internal Medicine Suggests "Exorcising Ghosts"

The Annals of Internal Medicine just published an editorial on ghost-writing and guest writing[1]. The editorial includes a good summary of some recent evidence that ghost-writing affects the reporting and discussion of clinical research. (Note that we have posted extensively on this topic, for example, here, here and here.)
To its credit, the editorial admits that the Annals may have published a ghost-written article about a clinical trial of Vioxx, and noted evidence that up to 16% of past Annals articles may have been ghost-written (although ghost-writing was even more prevalent in the New England Journal). Also to its credit, the editorial summarized the Mello study, showing how officials at medical schools and academic medical centers are willing to sign contracts that allow research sponsors, rather than purported academic investigators, to control the design, analysis, and reporting of clinical research[2]. (See previous post here.)
This editorial is a significant development, the first open, honest discussion of ghost-writing to appear in a large-circulation medical journal (since a 1994 editorial in JAMA)[3]. Hopefully, it will make quite a few physicians, and even some health care policy makers aware of the wide-spread external threats to evidence-based practice, and more broadly, to physicians' core values.
My only criticism of the article is that its proposed remedies focus on journal editors and academic authors. For example, it notes the new Annals editorial policy that would notify an academic researcher's home institution and research sponsor if the researcher had been shown to have allowed ghost-writing of an article he or she ostensibly authored. But this is like notifying the foxes about the recent theft from the chicken-coop.
It avoids confronting how academic institutions and research sponsors may promote ghost-authorship and other deceptive research practices in the first place (as in the Mello paper noted above). The academic's home institution may have acquiesced to this ghost-authorship by signing away its faculty member's rights to control the research and its reporting to the ostensible research sponsor. Furthermore, the home institution may have been pressuring the faculty member to get external support for his or her research in lieu of any institutional support for such scholarship. And, of course, the impetus for ghost-writing often comes from research sponsors eager to promote their own (usually commercial) interests.
So if the goal really is to exorcise ghosts and other kinds of research manipulation, the onus will have to be on the large organizations that benefit from these practices, as well as the sometimes hapless researchers and physicians who get caught up in them.
References
1. Laine C, Mulrow CD. Exorcising ghosts and unwelcome guests. Ann Int Med 2005; 143: 611-612.
2. Mello MM, Clarridge BR, Studdert DM. Academic medical centers' standards for clinical-trial agreements with industry. N Engl J Med 2005; 352: 21.
3. Rennie D. Authorship! authorship! - guests, ghosts, grafters, and the two-sided coin. JAMA 1994; 271: 469-471.