Having been out of town a lot the last couple of weeks, I haven’t read the New York Times Magazine the last couple of Sundays. It’s a magazine that I either read cover to cover or toss aside having read only The Ethicist, something that usually happens when it delves into some annoying hipster topic or lets the fashion and food stuff expand to fill too much of the magazine. This week, I came across an article that made me think: What should be the penalty for academic misconduct in science? Consider the case of Dr. Eric Poehlman:
On a rainy afternoon in June, Eric Poehlman stood before a federal judge in the United States District Court in downtown Burlington, Vt. His sentencing hearing had dragged on for more than four hours, and Poehlman, dressed in a black suit, remained silent while the lawyers argued over the appropriate sentence for his transgressions. Now was his chance to speak. A year earlier, in the same courthouse, Poehlman pleaded guilty to lying on a federal grant application and admitted to fabricating more than a decade’s worth of scientific data on obesity, menopause and aging, much of it while conducting clinical research as a tenured faculty member at the University of Vermont. He presented fraudulent data in lectures and in published papers, and he used this data to obtain millions of dollars in federal grants from the National Institutes of Health — a crime subject to as many as five years in federal prison. Poehlman’s admission of guilt came after more than five years during which he denied the charges against him, lied under oath and tried to discredit his accusers. By the time Poehlman came clean, his case had grown into one of the most expansive cases of scientific fraud in U.S. history.
“I need to start out by apologizing,” Poehlman said now, standing at the lectern before the judge. Speaking quickly and stammering occasionally, he apologized to friends and former colleagues, some of whom were listening in the back of the courtroom. He apologized to his mother, who sat in the front row, crying. And he apologized to Walter DeNino, the former protégé who turned him in, who was also sitting in the courtroom, several rows back on the prosecution’s side.
“I have wanted to say I’m sorry for five years,” Poehlman said, without turning around to face DeNino. “I want to make it very clear I am remorseful. I accept the responsibility. There’s no way that I can turn back the clock. And I’m not that individual that I was years ago.”
Before his fall from grace, Poehlman oversaw a lab where nearly a dozen students and postdoctoral researchers carried out his projects. His research earned him recognition among his peers and invitations to speak at conferences around the world. And he made nearly $140,000, one of the top salaries at the University of Vermont. All of that began to change six years ago, when DeNino took his concerns about anomalies in Poehlman’s data to university officials. The subsequent investigation — a collaboration among the University of Vermont, the Office of Research Integrity (which is within the Department of Health and Human Services) and the United States Department of Justice — uncovered fraudulent research that stretched back through almost half of Poehlman’s career. The revelations led to the retraction or correction of 10 scientific papers, and Poehlman was banned forever from receiving public research money. He was only the second scientist in the United States to face criminal prosecution for falsifying research data.
At 50, with his career in ruins and his reputation destroyed, Poehlman could only hope to avoid one final humiliation: becoming the first researcher sentenced to prison for scientific misconduct.
There is no doubt that this was far more than simply a difference in interpretation. It was far more than a matter of “spin,” which all scientists do to some extent or another whenever they present or publish their data. It was far more than a matter of overplaying certain evidence or ignoring evidence that didn’t fit in with the hypothesis, which some scientists fall prey to the temptation to do at times and for which arguments of interpretation can be made. In other words, there was not a gray area here. As the article describes, Poehlman had been found to have altered or even fabricated out of whole cloth primary data. His research involved studying the changes in lipid profiles in the blood that occur with aging in general and, in women, during menopause. Moreover, the scope of his deception was staggering, going on for ten years and attracting millions of dollars in grant support.
The story began six years ago, when Walter DeNino was hired by Poehlman as a laboratory technician. Here’s an account of the first time DeNino suspected that things weren’t entirely kosher in Dr. Poehlman’s lab:
The fall that DeNino returned to the lab, Poehlman was looking into how fat levels in the blood change with age. DeNino’s task was to compare the levels of lipids, or fats, in two sets of blood samples taken several years apart from a large group of patients. As the patients aged, Poehlman expected, the data would show an increase in low-density lipoprotein (LDL), which deposits cholesterol in arteries, and a decrease in high-density lipoprotein (HDL), which carries it to the liver, where it can be broken down. Poehlman’s hypothesis was not controversial; the idea that lipid levels worsen with age was supported by decades of circumstantial evidence. Poehlman expected to contribute to this body of work by demonstrating the change unequivocally in a clinical study of actual patients over time. But when DeNino ran his first analysis, the data did not support the premise.
When Poehlman saw the unexpected results, he took the electronic file home with him. The following week, Poehlman returned the database to DeNino, explained that he had corrected some mistaken entries and asked DeNino to re-run the statistical analysis. Now the trend was clear: HDL appeared to decrease markedly over time, while LDL increased, exactly as they had hypothesized.
Although DeNino trusted his boss implicitly, the change was too great to be explained by a handful of improperly entered numbers, which was all Poehlman claimed to have fixed. DeNino pulled up the original figures and compared them with the ones Poehlman had just given him. In the initial spreadsheet, many patients showed an increase in HDL from the first visit to the second. In the revised sheet, all patients showed a decrease. Astonished, DeNino read through the data again. Sure enough, the only numbers that hadn’t been changed were the ones that supported his hypothesis.
Confused by the discrepancy between the data sets, DeNino went back to Poehlman and asked to see the patient files. When Poehlman brushed him off, a disquieting feeling came over DeNino. Seeking advice, he e-mailed Andre Tchernof, a former postdoctoral fellow of Poehlman’s who had recently left to head his own lab in Quebec City. Tchernof confided to DeNino that something similar had happened before to another lab member.
Does this sound familiar? It’s a recurring theme in such cases. It’s rarely just one person who notices the problem; however it is usually only one person who is brave enough or stubborn enough to risk destroying his or her career to pursue it. The easiest course of action for someone like DeNino is either acquiescence or, if he’s unable to stomach what is going on, to find another lab to work in. A faculty member, Dr. Dwight Matthews, to whom DeNino went to seeking advice put it this way:
“First, understand that no matter how you proceed, everyone loses,” Matthews told DeNino when they met to discuss Poehlman. “Your career will be ruined because no one is going to protect you.” Matthews was brutally frank. “The university will come out bad,” he continued, “and Eric’s reputation will be destroyed.” He told DeNino that he would have to decide for himself what to do. As an afterthought, Matthews told me in a recent interview, he offered one suggestion: “If you’re going to do something, make sure you really have the evidence.”
Good advice. Actually, it’s good advice for any whistleblower, regardless of whether it’s in science or not. Even whistleblowers who do really have the evidence can expect to suffer. Undeterred, DeNino acted on his suspicions, and uncovered enough to show the scope of Poehlman’s fraud. Fortunately, cases of this sort, with deception and data fabrication on such a massive scale are rare. When faced with such cases, nonscientists will often ask: Why wasn’t this detected through peer review? The reason is that the entire process of peer review depends upon a certain level of trust and an underlying assumption that the scientist presenting his work to be published is honest, or at least not dishonest enough to make up data out of whole cloth. It’s a simple practical matter that it would be far too onerous to have to go through all the primary data for each paper. No peer reviewer would volunteer to do that. What science relies on to catch fraud like this is the self-correcting nature of science itself, where other scientists take the work of others and try to replicate it as a prelude to building on it. The hotter the scientific topic, the more likely it will be caught quickly. In scientific topics of less interest, fraud may never be caught because no one else cares enough about the topic to work on it. (Of course, if the topic isn’t hot there isn’t much incentive to falsify data about it, either.) However, in such cases, the damage is usually minimal because no other scientist is relying on the work tainted by fraud. It is in the middle ground, where Poehlman operated, that fraud may be most pernicious:
The length of time that Poehlman perpetrated his fraud — 10 years — and its scope make his case unique, even among the most egregious examples of scientific misconduct. Some scientists believe that his ability to beat the system for so long had as much to do with the research topics he chose as with his aggressive tactics. His work was prominent, but none of his studies broke new scientific ground. (This may also be why no other scientists working in the field have retracted papers as a result of Poehlman’s fraud.) By testing undisputed assumptions on popular topics, Poehlman attracted enough attention to maintain his status but not enough to invite suspicion. Moreover, replicating his longitudinal data would be expensive and difficult to do.
“Eric excelled at telling us what we wanted to hear,” Matthews, Poehlman’s former colleague, told me. “He published results that confirmed our predisposed hypotheses.” Steven Heymsfield, an obesity researcher at Merck Pharmaceuticals in New Jersey, echoed Matthews’s sentiments and added that Poehlman’s success owed more to his business sense and charisma than to his aptitude as a scientist.
“In effect, he was a successful entrepreneur and not a brilliant thinker with revolutionary ideas,” Heymsfield wrote me via e-mail. “But deans love people who bring in money and recognition to universities, so there is Eric.”
So, should Poehlman have been sentenced to one year and one day in federal prison, followed by two years of probation? I have to say reluctantly yes. He used fraudulent data to obtain millions of dollars worth of federal money to support his research, money that would have been better spent to fund the labs of honest researchers. When suspicion started coming his way, he lied. He tried to destroy DeNino’s credibility as a whistleblower and suggested that DeNino was homophobic (Poehlman is gay). His actions not only defrauded the government, but, even worse, cases like his erode public trust in science:
Rockey, who delivered a statement to the court on behalf of the N.I.H., said that lost grant money was not the only, or even the most significant, cost incurred. “Science is incremental,” she said, explaining that most scientific advances build on what came before. “When there’s a break in the chain, all the links that follow that break can be compromised.” Moreover, she said, fraud as extensive as Poehlman’s would inevitably lead to further erosion of the public’s trust in science. Poehlman’s sentence, she said, should send a clear message to the scientific community and the public at large that fraud would not be tolerated.
Indeed. Even so, I couldn’t help but feel a little uncomfortable when I read Dr. Poehlman’s explanation for his behavior:
“I had placed myself, in all honesty, in a situation, in an academic position which the amount of grants that you held basically determined one’s self-worth,” he told the court in June. “Everything flowed from that.” With a lab full of people dependent on him for salaries, Poehlman said he convinced himself that altering some data was acceptable, even laudable. “With that grant I could pay people’s salaries, which I was always very, very concerned about.”
He continued: “I take full responsibility for the type of position that I had that was so grant-dependent. But it created a maladaptive behavior pattern. I was on a treadmill, and I couldn’t get off.”
Whether or not Poehlman’s concern for his laboratory staff was genuine or a rationalization (probably a bit of both, I suspect, given his later admission that he had been seduced by the glory of being invited to give so many talks at scientific meetings and in being such a prominent member of the community), his words touch a little too close to home for me and probably for any scientist with people working for him whose salaries are dependent upon grants. Even though I only have two people working for me right now, and my present NIH grant doesn’t expire until 2010, I feel acutely responsible for supporting them and hopefully launching some of them into successful careers of their own. I don’t know how I could face them to tell them they didn’t have a job anymore were I ever to lose my funding. I wouldn’t fabricate data to keep them, but well do I understand the pressure of having people depending upon me and my science for their professional livelihood, now that I’m in charge rather than working in someone else’s lab.
NOTE: For more, Chad and Janet have both discussed this case.
