© Tudor Banus
|
The
scientist must be free to ask any questions, to doubt any assertion, to seek for
any evidence, to correct any errors.
J.
Robert Oppenheimer,
American physician
(1904-1967)
|
|
Medical
journals play a pivotal role in establishing a drug’s safety. To curb the influence
of private firms on reporting results, the most prestigious journals are tightening
publication guidelines
Clinical trials are
a building block of current medical practice, carried out to show the advantages
of a new approach to treatment. But just doing the trial is not enough: scientific
rigour requires that it undergo peer review and publication in a medical journal
before changes in clinical practice can occur.
Physicians who had no part in the trial closely examine it for potential flaws and
provide advice to journal editors. The latter assume that the authors helped design
the study, had full access to the data and all relevant information before interpreting
the data. Once published, physicians can make informed decisions, confident that
the medical journal is providing a full and unbiased appraisal of the treatment.
But recent developments have put this procedure to the test. Two decades ago, trials
were done by an alliance between academic investigators and commercial sponsors.
Academics were largely responsible for designing trials. The sponsor provided the
money and material required to handle the huge volume of data. Manuscripts reporting
trial results were jointly written by academic and commercial sponsors; when they
were submitted for publication, the major authors were often the academic scientists
who had performed key roles in the trial, giving it enhanced credibility.
Much of this has changed. Over the past decade, many academics have moved to industrial
posts. As a result, sponsors now have substantial expertise in trial design, performance
and data interpretation. Scientists working for industry are more likely to design
trials which cast their product in a favourable light. Rather than including academic
investigators from the outset, these individuals may be presented with a trial that
has already been designed. The sponsor, who often prepares the manuscript for submission
to a medical journal, holds, and may limit access to, all the data. The problem is
that the sponsor is not dispassionate about trial results.
There have been a number of well-publicized cases (see p. 20), where sponsors have
suppressed results of clinical trials. What bothered editors the most, however, were
related practices that occur much more commonly, but in less egregious fashion. A
sponsor may prepare a manuscript in which all the data of importance are not included
or in which the unwanted side effects of a new approach to therapy are minimized.
Putting an interpretational spin on reported data is not fair to the patients who
volunteered for the trials or to the broader community of patients and physicians.
We deserve to get a complete and unbiased report of research results; drug marketing
should not be a consideration in preparing a report.
In September 2001, the International Council of Medical Journal Editors (ICMJE),
an alliance of the editors of 12 general medical journals, adopted new rules concerning
the ethics of clinical trial performance and reporting. Academic physicians now need
to give assurances that they have had a meaningful role in trial design, complete
and unlimited access to the data base, and an uncensored role in data interpretation
and manuscript preparation. The ICMJE believes these rules will help everyone. They
will provide academic physicians with needed leverage as they negotiate research
contracts with sponsors and will empower them to analyze all the available data fairly.
They will aid patients and physicians by providing additional assurances that all
the information about a new treatment is provided. They will help commercial sponsors
by levelling the playing field.
Medical treatment has been substantially advanced by new medical devices and pharmaceuticals.
We want this progress to continue, but we want to be sure that the process of reporting
trials is as free as possible from commercial bias; we believe that these new rules
will help achieve this end. |
