Association between industry affiliation and position on cardiovascular risk with rosiglitazone: cross sectional systematic review

The Mayo investigators have added an important paper (18 March) to
the growing body of evidence linking authors whose conclusions in support
of a particular drug under review may be influenced by their conflicts of
interest. This paper found financial disclosure rates "were unexpectedly
low (53%)," and its authors concluded, "These findings, while not
necessarily causal, underscore the need for further progress in reporting
in order for the scientific record to be trusted." How much proof is
required to prove causality, and, in light of author bias and conflicts of
interest, what specific reporting policies might build trust in the
scientific record? This statement implies better reporting is the
solution when, in fact, disclosure seems to have no effect upon author
behavior or their standing in academia but does lead to a lack of public
confidence in medical judgment. These conflicts would soon disappear if
editors could find the courage to refuse publication to those with close
ties to industry and a clear bias.

There are several intractable issues confronting journal editors,
beginning with the fact that many leaders in the medical community have
already internalized the corporate ethic and accept the reality that
conflicts of interest are inevitable- or too difficult to regulate. In a
previous BMJ Rapid Response (10 August 2005) this writer observed, “When
these persons [medical leaders and university presidents] pass through the
revolving door into the corporate sanctum sanctorum, they adopt an ethic
in which board room politics, interlocking directorates, personal
friendships and substantial equity holdings make it impossible to
dispassionately implement policy in the public interest. In fact, board
members have a primary fiduciary responsibility to their stockholders, in
direct conflict with student and patient welfare and the overall mission
of the university.”

When dealing with ghostwriting, plagiarism, falsification of data,
pressure to increase revenue by increasing advertising, and drug company
detailing, it is understandable that editors face a daunting task. One
possible corrective measure may be to establish an independent panel,
including persons outside the medical profession, to fact check authors
and submissions. An ongoing ethics review process is required since
patient protection must be Priority #1.

Lynn Howard Ehrle is a retired consumer law teacher now serving pro
bono as Senior Biomedical Policy Analyst for the Organic Consumers
Association and Chair- International Science Oversight Board, composed of
44 physicians, scientists, and policy analysts from 11 countries.

E-mail: ehrlebird@organicconsumers.org

Competing interests:
None declared