In May, South Korean prosecutors indicted their nation's disgraced former scientific superstar, Woo Suk Hwang, and five lesser members of what was once the world's hottest research team. The charges focused international attention anew on the multiple roles that early-career scientists played in the stem cell scandal--both in perpetrating and in uncovering the fraud and other violations—and also on the intense pressures that young scientists can feel to produce career-boosting "good" results.
Just how prevalent is encouragement or even coercion to enhance or finagle one's data or to ignore or keep mum about infractions by others? More common than you'd suspect, according to an unscientific sample of postdocs who shared their views with Next Wave in response to queries posted on postdoc listservs . None mentioned abuses that rise to the level of outright fabrication practiced in Seoul, but the comments make clear that life in some labs can tax the conscience of the ethically scrupulous. Bad behavior is hardly universal, however; postdocs also commended exemplary supervisors who consistently model and demand adherence to the highest ethical standards.
Postdoc equals pressure?
"The concept of 'postdoc' implies pressure," says one we'll call Heidi Results. "There's a lot of temptation, and it becomes a reality when the boss supports fraud and manipulation of data." There's no mystery about why early-career scientists who ought to know better nonetheless succumb, explains postdoc Ben Derules (not his real name): "Getting a [faculty] job hinges entirely on how many papers you have." A postdoc whose name isn't A. Paul Dobserver, and who has worked for a half-dozen principal investigators (PIs), admires "the integrity, scientific and otherwise, of exactly two" of them. Adds Sue Spicious, another pseudonymous postdoc with experience in numerous labs, "No supervisor I ever [worked for] had explicit standards about results."
Some of the PIs who have supervised postdoc Chuck Sumdata (not his real name) have also had "fixed notions" about the results he should produce. "This is not what we wanted," he was told when his results turned out otherwise. The senior scientists would then "work it to prove they were right or delete certain data to fit their hypotheses," Chuck says. In a "difficult" lab where Sue spent time, "it was determined beforehand what the result should be, and there was no convincing [the PI] later" if data varied from that "preconceived result."
"In the high-pressure environment of that lab, if you did not have continuous results, you would be harassed and demeaned," Sue continued. The lab produced "publications containing inaccurate data." Once, for example, postdocs wrote up what they believed to be solid results. With the article under review at a top journal, however, they discovered their observation to be "not real. The paper was accepted, yet the wrong data ... was not corrected or retracted. [The paper] continued onto the press and was published," Sue says. This "malicious falsification" resulted from ''pressure to falsify ... from the supervisor, who 'made' them ignore" their misgivings. Another of the lab's papers was found, after publication, "to have a statistical analysis mistake that decreased the main significance of the work a lot." The postdoc authors "wanted to publish a correction, but the supervisor ... [ruled that] out of question."
Several of Ben's supervisors "tried to tell me to leave some data out. ... Every lab I've been in has told me to do that when they didn't think [the data] fit with the 'story.' " When he produced "data that disagreed with what my supervisor previously published, he refused to let me publish it. ... He wanted to have his previous paper stand." Heidi also has seen ethical infractions that include "misleading presentation of problems and existing literature [implying] that [the authors are] the first to discover something when it's not the case; not citing relevant work in research articles and even reviews; comparing results with results obtained with archaic methods" rather than the best methods; "citing only your friends; and more."
Nor is all pressure to exaggerate, falsify, or delete explicit. A postdoc we'll call Maura Lee Upright had a PI who "outwardly expressed support for 'true' results but would often arbitrarily dismiss some subject's data [and] never published negative results." When she alerted him that staff had enrolled subjects "who clearly were not qualified ... based on IRB-approved study criteria, ... he labeled me a 'troublemaker' and tried to force me out," she says. "He wanted data and he wanted it fast, and that was his priority. He definitely let [staff] know that and pressured them. His lab mantra was, 'It's better to ask forgiveness than permission,' 'If I don't know about it, it doesn't happen,' [and] 'Get it done at all costs, and don't tell me how,' " says Maura.
Fudging doesn't happen just in journal articles. When A. Paul applied for a prestigious fellowship several years ago, "my mentor told me to list a certain paper as 'in press' even though it was not yet submitted [because] it will be in press by the time the fellowship was reviewed." His graduate adviser urged him to list two other as-yet-unpublished papers in the same way. "None of the three manuscripts were even submitted" in time for the review, A. Paul continues. Although uncomfortable about the deception, "I went [along] with what my postdoc mentor and my Ph.D. adviser recommended." He won the award.
Rewards, however, rarely seem to go to postdocs who protest unethical practices. Sue and her colleagues tried to notify university officials of abuses but were "basically discouraged from proceeding, almost laughed at by the university higher personnel," she says. The situation, these officials said, was "your word against the supervisor's, ... a department head and key figure in the department." The complaints could "have repercussions in the postdocs' career and ... do them more harm than good," the officials warned, adding that ethical violations "happen more than you think and [are] not uncommon." "Unhappy" postdocs, they recommended, should "just leave." Maura got similar advice when she tried to raise ethical concerns with "senior faculty. They essentially said the [PI] brought in much-needed grant money and wouldn't get fired and advised me to 'cut my losses' and either leave or be quiet."
Setting an example
Postdocs also cited lab chiefs who are models of ethical excellence. A. Paul holds his "current mentor in awe because he does not ever cut corners ... Every potential flaw in the study is meticulously reported." Maura, too, now has a "highly ethical" mentor who "drives home in e-mails, memos, and staff meetings the importance of high standards and accuracy, and above all, that 'the data are the data.' ... All results should be published," this supervisor insists, "even those that are 'negative' or report no statistical significance, because that is important knowledge as well." He also "encourages confidential discussion, always, if you have any ethical concerns and addresses them immediately," Maura says. Chuck, too, mentions two ethically aware supervisors who "trained me well [and] let me repeat and repeat my experiments so I could prove my points."
Our small sample cannot reveal how common and representative these experiences might be. The postdocs agreed, however, that competition for grants and the publications that fuel them is intense, regulatory mechanisms are weak, and wrongdoing is rarely detected or punished. They suggested corrective measures including mandatory ethics training for postdocs and supervisors, changes in journals' acceptance policies, and greater willingness by university administrators to hear postdocs' complaints. What seem to matter most from the postdocs' point of view are the values that lab chiefs teach, practice, and enforce from day to day.
"It is very common for misconduct to happen at places where there is an atmosphere of terror and pressure," Sue concludes. "Where one is present, there is almost certainly the other. I have read ... about the Korean scandal lab's atmosphere, [and] it appeared to me ... exactly like the lab I was in! Pressure to work more hours, pressure to have results, pressure to have 'preconceived results,' or else you get berated," often publicly. "This atmosphere of stress and pressure," she says, "cannot be conducive to much more than trouble."