Researcher Absolved of Data Fraud

Long Saga Battered Science Community

By Rick Weiss
Washington Post Staff Writer
Saturday, June 22, 1996 ; Page A01

One of the longest and most tortuous investigations into scientific misconduct came to a close yesterday when an appeals board at the Department of Health and Human Services cleared a Tufts University scientist of all charges that she had fabricated data.

The decision exonerates immunologist Thereza Imanishi-Kari, whose publication of a research report with Nobel laureate David Baltimore in 1986 led to a decade of accusations and counter-accusations that shook the foundations of scientific integrity.

It reverses a 1994 decision by a federal investigatory board that found Imanishi-Kari guilty of 19 allegations of fraud, including charges that she lied to cover up her misconduct. That decision led to the loss of her faculty position at Tufts.

Although Baltimore was never formally charged with misconduct, his support for Imanishi-Kari dragged the renowned molecular biologist through a highly publicized scientific and political gantlet, climaxing in a series of high-tension hearings led by Rep. John D. Dingell (D-Mich.), then chairman of a powerful House oversight subcommittee, and Baltimore's resignation as president of Rockefeller University.

"I'm delighted. I'm elated. It's fantastic," said Imanishi-Kari, reached at the Tufts laboratory where she has been permitted to continue her work as a research associate. "This has been a terrible 10 years. Now I just want to pick up the pieces and do my research."

Paul Berg, a professor of biochemistry at Stanford University and a longtime friend and supporter of Baltimore's, said he saw the decision as an indictment of the way the National Institutes of Health, Congress and Rockefeller University handled the charges. "It's a sad fact that his exoneration comes only after he took such a beating," Berg said of Baltimore. "The abuse he took at Rockefeller, the way he was manhandled and basically forced to resign, I hope they hang their heads in shame."

The long saga that became known as "the Baltimore affair" started in 1986 with the publication of an arcane research report focusing on genes that regulate the immune responses in mice. The results challenged the standard view of how the immune system works and suggested some novel approaches to developing treatments for various diseases, including AIDS.

But in subsequent studies, one of the researchers involved in the work, Margot O'Toole, was unable to get the same results. When she came upon what appeared to be inconsistencies in the data, she blew the whistle on Imanishi-Kari, passing details to a team of self-styled "fraud busters" at the NIH.

The research eventually became the subject of formal investigations by two university review panels, the Office of Research Integrity (ORI) and, finally, Congress. At its peak, the case became symbolic of the larger question of whether scientists can be trusted to police themselves.

Although only minor errors in the original research were ever proven, forensic analysis performed by the Secret Service for the Dingell committee suggested that Imanishi-Kari had tampered with data in certain follow-up studies, results of which she had submitted to Dingell as part of her defense. In November 1994, ORI concluded, in part on the basis of those findings, that Imanishi-Kari was guilty of scientific misconduct. The office sentenced her to 10 years' ineligibility for federal research grants.

Imanishi-Kari appealed to the court of last resort, the research integrity adjudications panel of Health and Human Services. It was that three-person panel that handed down its decision yesterday afternoon.

"The Office of Research Integrity did not prove its charges by a preponderance of the evidence," the panel concluded. "The Panel recommends that no debarment be imposed and determines that no other administrative actions should be taken."

According to the report, the panel "examined piece-by-piece all the record evidence, including the original laboratory notebooks." It weighed testimony from numerous experts, and concluded that the forensic evidence presented by ORI was "at best inconclusive."

In part as a result of the Baltimore affair, the federal government has been revising the procedures by which it may accuse a scientist of misconduct. Already it has raised the standard of proof to the current level of "preponderance of the evidence."

Articles appear as they were originally printed in The Washington Post and may not include subsequent corrections.