Laboratory ethics: What makes some scientists cheat?

Is it a matter of shoddy work in the lab? A problem of excessive deference by junior researchers to senior scientists? Or does the case of the suspect stem-cell experiments in South Korea - an episode that is shaking the biomedical field worldwide - point to a severe lapse of research ethics?

With a probe at Seoul National University just beginning, it is likely to be some time before investigators can explain what led to apparent flaws in research - once celebrated as groundbreaking - by scientist Hwang Woo-Suk. His work involved cloning human embryos to garner highly prized stem cells specific to individual patients - an ability seen as an advantage in any future stem-cell therapies.

In the meantime, the case is prompting a closer look at how scientific journals screen research reports before publication, as well as forcing a deeper recognition of the intense pressures scientists can experience while working on cutting-edge, high-stakes research.

"Scientists are not a special breed of human being," says Thomas Murray, president of the Hastings Center, a bioethics institute in Garrison, N.Y. "But they function in a special environment.... They are bright people working in a community where the best ideas rise to the top. If you're not in first place, you're no place."

South Korea's probe at the dawn of what some dub "the biotech century" caps a year of research-ethics challenges.

• Late last month, a US district court judge in Albany, N.Y., sentenced a former Veterans Administration cancer researcher to 71 months in jail for criminally negligent homicide. Paul Kornak admitted that he had forged medical records, opening the way for people to take part in drug trials who should have been excluded because of existing medical conditions. One participant whose records were altered died during the experiment.

• In October, the Massachusetts Institute of Technology fired a young biologist and promising immunology researcher. MIT officials say Luk Van Parijs was dismissed after he admitted to school investigators that he fabricated and altered evidence in research papers to support grant applications.

• In March, a University of Vermont obesity scientist admitted faking data in order to buttress grant applications. (He netted $3 million in government grants.) Under a deal with US prosecutors, Eric Poehlman agreed to plead guilty to criminal fraud and to retract or correct several research papers.

Meanwhile, the National Institutes of Health has tightened its rules for NIH researchers who also serve as consultants to drug companies. The NIH is trying to walk a tightrope between avoiding conflicts of interest and ensuring that scientists can engage in the open give and take today's complex research efforts require. In a survey of NIH-funded scientists, released in June, only 1.5 percent of 3,000-plus respondents acknowledged having falsified or plagiarized information. But 15.5 percent admitted to altering their research approach under pressure from funding sources, and 12.5 percent admitted to looking the other way when colleagues used flawed data.

Surveys show that the public consistently holds scientists in high esteem, perhaps leading many people to assume an unrealistic ethical purity among them.

If lapses happen in business, "the public says, 'Well, what did you expect?' " says Mark Frankel, director of Scientific Freedom, Responsibility, and Law at the American Association for the Advancement of Science, which publishes the journal Science. The public tends to be more surprised when the violators are scientists - although public esteem remains high despite the lapses, he adds.

Biomedical research in particular is a hotbed of economic and scientific competition. Several countries, including South Korea, are vying with the US for leadership. So the pressure to lead can be huge.

Dr. Hwang's request last week to withdraw his team's paper on stem-cell work, published in May in Science, has spurred editors at the journal to ask whether their peer-review process of evaluating papers for publication could have caught problems in the Hwang submission.

"A paper that apparently achieves a result that others have tried to get and failed is subject to especially careful scrutiny," says Donald Kennedy, Science's editor. "I expect a certain amount of skepticism" among reviewers as they give papers the once-over. "On the other hand, I think reviewers generally tend to trust explicit representations" of the information in the papers.

Science is generally self-correcting, says the Hasting Center's Dr. Murray. If a paper is published and other scientists fail to reproduce the results, it is likely to get relegated to the trash bin.

In the end, no system is infallible, ethicists note. "If you have someone determined to fabricate evidence, no screening system will catch that," says Alto Charo, a law professor at the University of Wisconsin who specializes in biomedical and research ethics. "You have to rely on the integrity of the individual."

In the past decade, federal funding agencies have put more emphasis on ethical research practices, requiring grant recipients to take ethics courses or giving grants to scientists at universities with ethics classes for graduate students, notes the AAAS's Dr. Frankel. These courses have undergone little evaluation for effectiveness, but several cases that made headlines this year came to light after fellow researchers or young protégés became suspicious of data being used and blew the whistle on their errant colleagues.