[Muanet] Research funding system versus science

[Dion Giles]

This article requires a password, but seems relevant and important enough 
to pass on to this small group of readers.  My comment on the last two pars 
-- if researchers find ways to divert R&D funding into genuine research to 
inform, best of luck to them!

Dion Giles

  http://www.nytimes.com/2005/06/14/health/  (subscription address)

NY Times June 14, 2005
	By CORNELIA DEAN

People who worry about research misconduct in science should look beyond
the fabrication, falsification and plagiarism that have traditionally
defined the term and consider less obvious misdeeds, like interpreting
data in a questionable way, changing study methods to suit a sponsor and
other problems, researchers from Minnesota say.

They say this kind of "mundane 'regular' misbehavior" occurs often and
threatens the integrity of science more than the occasional high-profile
case of scientific misconduct.

The researchers surveyed almost 1,500 scientists whose work had been
financed by the National Institutes of Health. Though fewer than 1 percent
admitted to misdeeds like falsifying results or disregarding rules for the
treatment of people participating in medical research, more than 15
percent said they had changed the design, methods or results of a research
study to suit a sponsor. Almost as many confessed to using "inaccurate or
inappropriate research designs."

"Our findings reveal a range of questionable practices that are striking
in their breadth and prevalence," said the researchers, who described
their findings in a commentary in the current issue of the journal Nature.

In the past, several eminent scientific organizations have objected to
suggestions that the definition of scientific conduct be widened beyond
the misdeeds known collectively as F.F.P., for fraud, fabrication and
plagiarism, on the grounds that it would be impossible to write guidelines
that would be applicable to everyone, and that detailed oversight would be
an unnecessary and disruptive intrusion into the research process.

In a 2003 report, "Integrity in Scientific Research," an expert panel
convened by the National Academy of Sciences emphasized the role that
individual institutions play in ensuring research is done properly.

The authors of the new commentary, Dr. Brian C. Martinson of Health
Partners Research Foundation in Minneapolis and Dr. Melissa S. Anderson of
the University of Minnesota and Dr. Raymond de Vries of the University of
Minnesota and St. Olaf College, say their study is the first to
quantitatively analyze behavior by researchers.

As a result, Dr. Martinson said, it was impossible to say whether behavior
patterns had changed in recent years, when commercial sponsors have taken
a larger role in biomedical research, the only scientific area covered by
the survey.

Also, "not all the questions worked the way we wanted them to," said Dr.
Martinson, a sociologist who researches health issues at Health Partners
Research Foundation. For example, changing a research design because a
government-appointed panel has safety concerns is quite different from
changing a design to give a commercial sponsor the results it wants - and
the survey question did not differentiate.

Another issue addressed in the survey, inappropriately assigning
authorship credit, may mean that someone who contributed minimally to a
report was listed as a co-author, a common practice among scientists. But
it may also mean that scientists allowed themselves to be listed as the
authors of a paper ghost-written by a commercial sponsor.

Dr. Michelle Mello, an associate professor of health policy and law at the
Harvard School of Public Health, said it was hard to design surveys to
address these issues. "In a survey about misconduct, you don't want to
frame the question in a way that makes no one want to admit that they did
it," she said. "But that ambiguity makes it difficult to form a conclusion
about how worried we should be."

Dr. Mello was an author of a report, published last month in The New
England Journal of Medicine, that medical schools varied widely in their
standards for testing new medicines for drug companies, with some
accepting far more control by the companies than others.

Still, she said in an interview, she doubted that questionable behavior
among researchers was widespread.

But Dr. Martinson said he believed that the 15.5 percent of scientists who
said in the survey that they had changed the design, methods or results of
a study to satisfy a sponsor "had to have some sense that they were
responding yes to behavior that was inappropriate in their own judgment."

He noted that the survey respondents admitted in much higher numbers to
other categories of conduct that are accepted among scientists, though
technically improper.

For example, he said, 70 percent of midcareer scientists surveyed said
they had applied funds from one research grant to another project. "The
federal government frowns on this," he said, but "scientists don't think
it's wrong. They see it as a way of getting the best and the most research
out of the dollars they have available to them."

Dr. Martinson, a sociologist, said his work on the issue grew out of his
interest in how scientists respond to the pressures of high-level research
and the stress of obtaining money to pay for it. "A lot of the behaviors
we are looking at fall into that category - the ways you have to behave if
you are going to continue to get funding," he said. "Does that make them
right? I don't know."