The Elusive Truths of the Baltimore Case


By John D. Dingell
Thursday, July 18, 1996 ; Page A27

Vindication. Exoneration. Political grandstanding. Bullying. McCarthyism.

These words are being bandied about following the recent decision of a government appeals panel in the matter of allegations of scientific misconduct against Dr. Thereza Imanishi-Kari. The panel's decision has provided some of my longtime critics with an opportunity to settle old scores and rack up rhetorical points in attacks suggesting that the Subcommittee on Oversight and Investigations, under my chairmanship, wrongly targeted Imanishi-Kari and her colleague, Dr. David Baltimore.

Fair enough. These people are entitled to their opinions. But they are not entitled to their own versions of the facts.

First, a brief history. In 1986 Dr. Baltimore, with the assistance of Dr. Imanishi-Kari, published a paper in the journal Cell. Margot O'Toole, a young postdoctoral student working in Imanishi-Kari's lab (not, as the Post's June 30 editorial stated, in Baltimore's lab), attempted to replicate research conducted by Imanishi-Kari. She could not. Her subsequent questions about the research were summarily dismissed by the individuals and institutions involved. At this point the matter came to the attention of the Oversight Subcommittee, which held bipartisan public hearings (as it had on several other cases of alleged scientific misconduct). The subcommittee held its last hearing on this case in 1990. From there, it went to the Office of Scientific (now Research) Integrity within the Department of Health and Human Services, and then to the most recent appeals panel.

Bullying? If there was any bullying, it was of Margot O'Toole, who was driven from science and found herself answering phones at a relative's moving company.

Vindication? Hardly. The HHS panel's opinion confirms that many of Dr. O'Toole's concerns were justified. The research she questioned was "rife with errors of all sorts," said the panel's decision. Moreover, the panel cited findings of "significant inaccuracies or misstatements" by previous scientific reviews of the Cell paper. Had Imanishi-Kari dealt openly and appropriately with O'Toole's criticisms in the privacy of their laboratory, rather than letting the problem fester and retaliating against O'Toole, the controversy would never have erupted.

Is David Baltimore a victim and martyr in this cause? Dr. Baltimore, not the subcommittee, forced the scope of the inquiry to be expanded. Imanishi-Kari did not conduct her work in Dr. Baltimore's laboratory. Had she done so, his direct supervision might have prevented the publication of the Cell paper with the flawed data. Nonetheless, Baltimore sought to defend his friend and collaborator by turning the subcommittee's investigation into a referendum on his own honor and reputation -- despite the fact that research he conducted was never at any time under question.

Baltimore's understandable impulse was misguided. The subcommittee set out only to examine the process by which scientists receiving vast sums of taxpayer money resolve allegations of misconduct leveled by other scientists. Baltimore insisted instead on pitting the competence and veracity of O'Toole against that of Imanishi-Kari. In effect, he argued that the public's representatives had no right and no cause to examine how the public's money is spent on scientific endeavors.

The HHS appeal panel did not find a preponderance of the evidence to support a charge of outright fraud or intentional misconduct by Imanishi-Kari. But neither did its opinion rebut much of the Secret Service's sworn testimony that portions of her laboratory records had apparently been altered. "We recognize," said the panel, "that some of the apparent anomalies and peculiarities in the records may be impossible to explain fully at this point." The panel acknowledged that data supporting a key table in the Cell paper were simply "inexplicable," and thus "evidence that it is unlikely that Dr. Imanishi-Kari would fabricate such results." (This is most curious logic: if the data supporting the table were "explicable," would that have been evidence of fabrication?) The panel simply accepted Imanishi-Kari's claim of haphazard record-keeping.

Some scientists familiar with the Secret Service's analysis remain convinced that the forensic, statistical, and scientific evidence of falsification is overwhelming. But apart from one's view of the panel's decision, The Post correctly notes the fundamental point at stake in the debate on scientific misconduct: "the government pays for scientific research and is justified in investigating allegations of fraud when they are leveled."

Until the subcommittee addressed the scientific misconduct issue, the scientific community had no effective and systematic method for dealing with such allegations where federal funds were involved. Other subcommittee inquiries in that period and afterward documented clearly and without dispute that a cardiologist at Harvard had fabricated data and that the university had failed to uncover it; that a Pittsburgh professor had published "research" on therapies for severely retarded children based on nonexistent data and on subjects who were never tested (he later pleaded guilty to two felonies); and that several breast cancer patients were treated by a renowned cancer researcher with a highly toxic drug without being informed or given a chance to consent.

Political grandstanding? If political gain had been the objective, as some have suggested, it would have been foolhardy. The political utility of sorting through the facts of complicated cases in the face of vitriolic criticism from the scientific establishment is not exactly on a par with opposing crime or supporting welfare reform. If a political cost-benefit analysis had been the only criterion for getting involved, I would have stayed miles away from the issue of scientific misconduct.

The charge of McCarthyism is most ironic, and moronic. In spite of the cries of unfairness, the subcommittee in the Imanishi-Kari matter received sworn public testimony from every relevant party, including Drs. Baltimore and Imanishi-Kari, as it tried to determine whether O'Toole's allegations were addressed adequately by the scientists and institutions involved. All witnesses received a full opportunity to state their cases.

What lessons can be learned from this investigation? Congressional hearing rooms are most certainly not the best forum in which to adjudicate complex and conflicting scientific claims, just as they are not the best forums in which to try lawsuits or write regulations. I hope HHS and the scientific community have learned some lessons too. The government, as well as academic institutions and scientists themselves, must do better by the Margot O'Tooles of the world than in this case.

Lost in what has been portrayed as a clash of personalities is a serious public policy issue. Scientific research forms the basis of government policy on the environment, workplace safety, the safety and efficacy of drugs, and vital matters of public health. The American taxpayer supplies some $9 billion annually to the National Institutes of Health alone for scientific research. The inability of science to police and correct itself poses a far greater danger to its future than appropriate inquiry and oversight.

At least and at last, the decision of the appeals panel closes the door on the Imanishi-Kari case. Sadly, the real truth in this matter will likely always elude us.

The writer, a Democratic representative from Michigan, was chairman of the House Energy Subcommittee on Oversight and Investigations.

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