Encourage ethics in the laboratory

Hwang Woo-Suk has gone from bearing the lofty title "Supreme Scientist" of South Korea to facing possible criminal charges. His collaborator, American Gerald Schatten, has been charged with "research misbehavior" by a university review panel.

In what may be the biggest scientific fiasco since the supposed discovery of cold fusion in 1989, the two men headed a group of Korean scientists who published findings last year claiming they had produced 11 stemcell lines using cloned human embryos. In January, Korean investigators concluded the data in the study had been faked. The US-based journal Science retracted papers it had published describing the findings.

Unfortunately, the wayward scientists aren't alone. In December, editors at the New England Journal of Medicine (NEJM) said that a report it published in 2000 on the drug Vioxx contained inaccurate information about its potentially severe side effects. And a Norwegian cancer researcher, Dr. Jon Sudbo, admitted falsifying data and conclusions in studies published in the NEJM, the Journal of Clinical Oncology, and the British medical journal The Lancet.

Why lie? Fraudsters sometimes yield to the temptation to exaggerate their results to gain the eye of editors at top scientific journals. Acceptance for publication can boost careers and win fresh funding for future projects.

Editors of scientific journals are in the midst of soul-searching, asking what they can do to reduce such transgressions. "Even the very best journals have published rubbish they wish they'd never published at all," says Richard Smith, former editor of BMJ (British Medical Journal), in the current issue of The Scientist.

Editors of prominent journals receive thousands of submissions and have neither the time nor funds to duplicate experiments or investigate articles with the thoroughness of a detective or a courtroom prosecutor. The time-honored process of peer review, in which potential articles are sent to experts in the given field for secret vetting, is itself under fire.

So what can be done?

Fortunately, scientists aren't scam artists who disappear and reemerge with another identity. Their conclusions eventually will be tested by others. If the results can't be duplicated, their research loses credibility, and they ultimately face ostracism.

Computer programs are beginning to be used to check if data contains suspect patterns or an image has been altered. All the authors of a paper should reveal exactly what their role has been and accept responsibility for the findings. Peer reviewers should reveal if they have any conflicts of interest.

In addition, courses on ethics and integrity in research should be given higher prominence in college science programs and at research firms. (A recent study found that half of today's US college students admit to serious cheating on written assignments.)

The British government is introducing a code of scientific ethics next month that will call on scientists to act with "rigor, respect, and responsibility" and to present findings honestly and accurately.

It's important for citizens to feel that scientists have made their highest motive the quest for knowledge and the advancement of the public good - not fame or personal gain.