“Something is rotten in the state of…
…science” if I may borrow from and rephrase Shakespeare.
The highest ethical standards – we expect no less from men and women of science. Yet, recent events brought to light just how short they can fall from these expectations.
“Neither a borrower nor a lender be: For loan oft loses both itself and friend.”
Case # 1: Duplication and plagiarism.
Borrowing from the works of others and our own without proper citation is, in the scientific community, unethical behaviour of the highest magnitude. Using the software eTBLAST, bioinformatics searched through 62,213 Medline entries and found 0.04% cases of potential plagiarism and 1.35% cases of potential duplication [1, 2]. These citations are recorded on Déjà vu, a database of potentially duplicated and plagiarized scientific articles.
One of its first victims comes from no less than that icon of higher learning – Harvard itself. Two review papers, published about 1 year apart by different authors in 2 different journals have 55% similarity [3].
The countdown is running for other papers listed in the Déjà vu database as scrutiny continues. The journals are suddenly waking up to the fact that unethical behaviour in science is more rampant than previously thought. But isn’t it ironic that anti-plagiarism software - the very tool that academia is using to prevent plagiarism by students – is the one that would cast serious doubt on scientific credibility?
“To tell, or not to tell, — that is the question.”
Case #2: Breach of confidentiality.
A referee for the New England Journal of Medicine committed a major breach of confidentiality by leaking an unfavourable manuscript to a drug company [4]. I’m referring to the Avandia (rosiglitazone) issue (see 25 Nov 2007post) and how a referee faxed a copy of the manuscript [5] to the drug’s manufacturer GSK before it was officially published. The confidentiality of the peer review process is sacrosanct and breaching this confidentiality is a major scientific misconduct.
This reflects current concerns about conflicts of interest among authors, editors, and referees (who are most of the time, all one and the same) due to financial support from pharmaceutical companies. Another reason to call for non-industry-sponsored clinical research.
“There is nothing either good or bad, but thinking makes it so.” And getting caught.
References:
[1] Errami M et al. (2008). Déjà vu—A study of duplicate citations in Medline. Bioinformatics 24(2):243–249.
[2] Errami M & Garner H (2008). A tale of two citations. Nature 451:397-399.
[3] http://www.boston.com/news/health/blog/2008/01/a_harvard_rheum.html
[4] Vastag B (2008). Reviewer leaked Avandia study to drug firm. Nature 451:509.
[5] Nissen SE & Wolski KN (2007). NEJM 356:2457–2471.
No Comments
Be the first to comment!