Bosman, Julie. "Reporters Find Science Journals Harder to Trust, but Not Easy to Verify," New

 

Bosman, Julie. "Reporters Find Science Journals Harder to Trust, but Not Easy to Verify," New
York Times, 13 February 2006, pp. C11, C3.

When the journal Science recently retracted two papers by the South Korean researcher Dr.
Hwang Woo Suk, it officially confirmed what he had denied for months: Dr. Hwang had
fabricated evidence that he had cloned human cells.

But the editors of Science were not alone in telling the world of Dr. Hwang's research.
Newspapers, wire services and television networks had initially trumpeted the news, as they
often do with information served up by the leading scientific journals.

Now news organizations say they are starting to look at the science journals a bit more
skeptically.

"My antennae are definitely up since this whole thing unfolded," said Rob Stein, a science
reporter for The Washington Post. "I'm reading papers a lot more closely than I had in the past,
just to sort of satisfy myself that any individual piece of research is valid. But we're still in sort
of the same situation that the journal editors are, which is that if someone wants to completely
fabricate data, it's hard to figure that out."

But other than heightened skepticism, not a lot has changed in how newspapers treat scientific
journals. Indeed, newspaper editors openly acknowledge their dependence on them. At The Los
Angeles Times, at least half of the science stories that run on the front page come directly from
journals, said Ashley Dunn, the paper's science editor. Gideon Gil, the health and science editor
for The Boston Globe, said that two of the three science stories that run on a typical day were
from research that appeared in journals.

Beyond newspapers, papers from journals are routinely picked up by newsweeklies, network
news, talk radio and Web sites.

"They are the way science is conducted, they're the way people share information, they're the
best approximation of acceptance by knowledgeable people," said Laura Chang, science editor
for The New York Times. "We do rely on them for the starting point of many of our stories, and
that will not change."

There are limits to the vetting that science reporters, who are generally not scientists themselves,
can do. Most journal articles have embargoes attached, giving reporters several days to call
specialists in the field, check footnotes on an article and scrutinize the results.

"Scientific discoveries are more difficult because they often require in the generalist reporter a
good deal of study, follow-up interviews and some guidance on how to make sense of technical
matters," said Roy Peter Clark, a senior scholar at the Poynter Institute, which studies journalism.
"But I think the scandals do require both a new level of skepticism on the part of the reporter and
also maybe some new protocols between scientists and journalists."

The Hwang case was not the first time journals had been duped: recently, editors at The New
England Journal of Medicine said they suspected two cancer papers they published contained
fabricated data. In December, the same journal said that the authors of a 2000 study on the
painkiller Vioxx had omitted the fact that several patients had had heart attacks while taking the
drug in a trial. A study on the painkiller Celebrex that appeared in The Journal of the American
Medical Association was discredited when it was discovered that the authors had submitted only
six months of data, instead of the 12 months of data they had collected.

While the journals have a peer review process that is in part meant to filter out fallacious papers
by checking research techniques and conclusions, perhaps the greatest difficulty for science
reporters is trying to catch what journal editors have missed.

After hearing the news of Dr. Hwang's fabrications, Mr. Gil of The Globe said he immediately
remembered his newspaper's coverage of the stem cell papers.

"We were blown away, in part because we had written those stories on Page 1," Mr. Gil said.
"And when we wrote them, we called the leading experts in the world on all this embryonic stem
cell stuff, who are here in Boston. And they were as hoodwinked as anybody else."

Despite the fraud cases, most of what the journals publish is basically credible, said David
Perlman, the science editor of The San Francisco Chronicle. Among the most prestigious science
journals that reporters consult regularly are Nature, Science, The New England Journal of
Medicine and The Journal of the American Medical Association.

"I think they and we have been burned enough that they're making efforts," Mr. Perlman said.
"They're being more careful now, and I think reporters are too. I definitely have more of a ‘Hey,
let's look more carefully' attitude now that I did 5 or 10 years ago."

Donald Kennedy, the editor of Science, said in a statement in December that the journal itself
was not an investigative body. But when reporting on journal findings, most news outlets fail to
caution that studies must be replicated to be truly authenticated.

"Beyond Hwang, the more fundamental issue is that journals do not and cannot guarantee the
truth of what they publish," said Nicholas Wade, a science reporter for The New York Times.
"Publication of a paper only means that, in the view of the referees who green-light it, it is
interesting and not obviously false. In other words, all of the results in these journals are
tentative."

The journals' own peer review processes, which are intended to be the first barrier against fraud,
have come under criticism lately. A cover story in the February issue of The Scientist said that
the top-tier journals were receiving more submissions every year, overtaxing peer reviewers and
weakening the screening process.

After the Hwang scandal, Science announced it was considering a set of changes to better prevent
fraud: Dr. Kennedy said in January that new rules could include "requiring all authors to detail
their specific contributions to the research submitted, and to sign statements of concurrence with
the conclusions of the work," as well as "implementing improved methods of detecting image
alteration, although it appears improbable that they would have detected problems in this
particular case." (Through a spokeswoman, Dr. Kennedy declined to be interviewed and said the
editors were currently conducting a review of the episode.)

Some newspapers have adopted guidelines of their own to check for conflicts of interest
involving authors of journal articles. The Globe instituted guidelines last July requiring reporters
to ask researchers about their financial ties to studies, and to include that information in resulting
articles. In its weekly health and science section, The Globe outlines any shortcomings of a study
under the heading "Cautions."

Kit Frieden, the health and science editor for The Associated Press, said: "We've always had our
own peer review process, where on the major studies we seek outside expert comment. We've
always regarded scientific research cautiously because mistakes can be made, and I don't think
that's changed."

The growing competition for the most important research among the journals may contribute to
mistakes and fabrications, even in the most prestigious of the bunch. But in the end, the severe
consequences of presenting fraudulent research generally act as a deterrent, said Mr. Dunn of The
Los Angeles Times.

"Unlike financial fraud, where you can bamboozle somebody of their money and disappear and
then start over again, in science the researchers are in one place," he said. "If they get caught in
this type of thing, their careers are over."
The New York Times
Printer Friendly Format Sponsored By