Global Views
   Middle East & Africa
 Embassy News
 Arts & Living
 Travel & Hotel
 Medical Tourism New
 Letters to Editor
 Photo Gallery
 News Media Link
 TV Schedule Link
 News English
 Hospitals & Clinics
 Flea Market
 Moving & Packaging
 Religious Service
 Korean Classes
 Korean Weather
 Real Estate
 Home Stay
 Room Mate
 English Teaching
 Job Offered/Wanted
 Hotel Lounge
 Foreign Exchanges
 Korean Stock
 Business Center
 PR & Ads
 Arts & Performances
 Restaurants & Bars
 Tour & Travel
 Shopping Guide
 Foreign Missions
 Community Groups
 Foreign Workers
 Useful Services
 ST Banner Exchange
Science Takes Hard Look Inward
By Rita Rubin
Support for Dr. Hwang — Thousands of South Koreans poured out on street of downtown Seoul on Jan. 11, 2006 in their support for Dr. Hwang Woo-Suk after an investigation panel concluded that Dr. Hwang's two landmark papers were fake earlier in the day. They staged candle-light vigil protest, demanding that Dr. Hwang given a chance to produce embryonic stem cells.

Journal editors, considered the gatekeepers of scientific information, are grappling with how they can better uncover liars. Last May and in March 2004, South Korean stem cell researcher Hwang Woo Suk fooled the editors of Science, which is regarded as one of the world's premier journals.

A new report from an investigating committee at Seoul National University, where Hwang worked, shows just how badly the editors were fooled. Hwang never cloned human embryos, from which he said he had derived a total of 12 stem cell lines, the panel said Monday. His data and illustrations were all fabricated.

One recent Hwang paper apparently was true, though. Nature announced Tuesday that independent blood tests showed Hwang most likely did clone a dog named Snuppy, as reported in that journal in August.

Editors from a host of respected journals, such as the Lancet and the Annals of Internal Medicine, have had to rescind fraudulent articles in the past year. Two articles dated back to 1992, which allowed plenty of time to throw legitimate researchers — not to mention doctors who turn to journals for guidance in treating patients — off track.

Given that there are tens of thousands of scientific journals, "this is probably happening on quite a large scale, and we just have inadequate mechanisms for sorting this out," says Richard Smith, former editor of the British Medical Journal, who wrote about fabricated research in July.

University of Michigan science historian Nicholas Steneck, a consultant to the federal Office of Scientific Integrity, says he is seeking money for an international conference to address fraudulent research. Steneck says he hopes to capitalize on the publicity surrounding Hwang.

"This field is one of lots of activity when major cases break," Steneck says. "Then, when the attention dies down, people forget about it."

Though major cases are few, "we don't know whether that is because others aren't being caught."

Unlike some other human endeavors, Smith says, "in science, the whole thing is based on trust."

Sometimes peer review, in which journals seek scientists' opinions of submitted papers, detects falsehoods, Smith says. But journals are not in the habit of asking authors to prove, say, that they studied 300 patients as they said they did.

"Even unusually rigorous peer review of the kind we undertook in this case may fail to detect cases of well-constructed fraud," Science editor in chief Donald Kennedy said in a statement Tuesday.

So how are journal editors, grant reviewers and medical school deans supposed to sniff out fraud?

"We don't give the authors lie-detector tests," says Catherine DeAngelis, editor of the Journal of the American Medical Association.

"The whole idea is that there's so much trust involved, and there has to be. If we completely lose trust in our authors, then we might as well hang it up."

Science is calling on outside experts to help come up with additional safeguards, Kennedy said. One option might be to require all authors to explain their specific contributions and to sign a statement saying they concur with the conclusions.

If a study looks too good to be true, then maybe it isn't, says Michael Meguid, editor of the journal Nutrition.

In February, Meguid retracted a 2001 paper by former University of Newfoundland researcher Ranjit Chandra, who concluded that vitamin supplements improved cognition in the elderly.

"Science is 99% failure," says Meguid, a surgery professor at the State University of New York Upstate Medical University. "When people have spectacular successes one after another, like this Korean guy did, you're jubilant, but then you start saying, 'Wait a minute.' "

Meguid says the fact that Chandra was the only listed author — as was the case on nearly all of his publications — should have raised suspicions. For example, Meguid says, Chandra wrote that he drew blood from 100 patients twice a week, quite an undertaking for someone who also was running the study.

Logistics aside, a lack of co-authors helps prevent fabrications from being discovered, Meguid adds.

Sometimes editors just have to go with their gut. In one instance, all of the scientists asked to review a paper submitted to JAMA thought it was fraudulent, DeAngelis says. She wrote the heads of the authors' institutions and asked them to investigate.

"They all came back and said, 'No, this is legitimate,' " DeAngelis says. "But I had what I call a niggle in the back of my head, and we didn't publish it. We always err on the side of being careful. Sometimes it means that we'll turn down a paper that actually is straightforward and honest. But I'd rather make that error than publish something that just doesn't feel right."

However, just because she's suspicious doesn't give her the right to alert other editors that a study might be fraudulent, DeAngelis says. She says she hasn't checked whether the paper she rejected has ended up being published elsewhere and adds, "I'd be surprised that it hasn't been."

The above article is from USA Today.






The Seoul Times Shinheungro 25-gil 2-6 Yongsan-gu, Seoul, Korea 04337 (ZC)
Office: 82-10-6606-6188
Copyrights 2000 The Seoul Times Company  ST Banner Exchange