Scientific misconduct may be far more extensive than the occasional high-profile cases of fabricated research or plagiarism, according to survey results published in today's issue of the journal Nature.
That's because scientists appear to be engaging far more often in "mundane" kinds of scientific wrongdoing - transgressions that could be an even greater threat to science, authors from the University of Minnesota and a Minneapolis research foundation conclude.
The researchers found that 33 percent of the scientists surveyed admitted to engaging in at least one of 10 research misbehaviors in the previous three years. The results were even more striking for mid-career scientists, some 38 percent of whom acknowledged such misbehavior.
A tiny percentage of the 3,247 scientists said they had committed research sins that can result in federal sanctions and end a career - such as falsification or plagiarism. The study found they were far more likely to admit to other transgressions.
Some 15.5 percent admitted to "changing the design, methodology or results of a study in response to pressure from a funding source." About 12.5 percent said they had overlooked "others' use of flawed data."
The scientists had all received National Institutes of Health funding but were not NIH employees. Their names, which were withheld, were culled randomly from NIH databases.
The 2002 survey was funded by the NIH and supported by the federal Office of Research Integrity, which investigates scientists accused of research fraud and can bar them from receiving federal funds.
"Our evidence suggests that mundane 'regular' misbehaviors present greater threats to the scientific enterprise than those caused by high-profile cases such as fraud," the authors wrote. They speculated that intense competition among scientists to attract grants and publish in prestigious journals, among other things, might be the problems.
That pressure may be increasing, co-author Brian C. Martinson of the nonprofit HealthPartners Research Foundation said in an interview, as the NIH limits growth in research funding and universities turn to researchers to woo funding from drug manufacturers and other companies.
"A lot of this is sort of known in science," Martinson, an epidemiologist and HealthPartner researcher, said about the findings, adding that researchers may justify their behavior to themselves by saying they are simply doing what they have to do to get by. "The scientific community hasn't owned up to it: Scientists are under pressure. They're human beings and they (may) react to pressures in an inappropriate way."
The study comes amid a national debate about whether drug companies' influence over study results is marring the integrity of science.
In May, for example, the New England Journal of Medicine published a study that found that about half of 107 U.S. medical schools surveyed would let companies that sponsor academic research draft resulting articles that appear in medical journals.
Both the authors of today's Nature study and other scientists said that there is reason to be cautious about its precise results. If anything, they said, the results likely underestimate the amount of research misbehavior because scientists guilty of it might be less likely to respond.
However, others said the survey questions encompass gray areas. For example, there may be legitimate reasons for scientists to change study design when asked to do so by a study's sponsor, said Michelle Mello of the Harvard School of Public Health. One reason might be inappropriate risks to people enrolled in a clinical trial.
"Sometimes, you have a lot of adverse events," said Mello, an associate professor of health policy and law who co-authored the May study looking at companies' influence on medical school research. "My point is, you need to know more about who the (clinical trial) sponsor was and what the nature of the requested changes might have been."
Other questions left unanswered by the results include why mid-career scientists were more likely to admit misbehavior than younger researchers doing postdoctoral work. The authors speculated, however, that more established researchers may be more confident of their ability to weather ethical storms without consequence to their careers. They also may have more opportunities to misbehave.
Some 1,768 mid-career scientists responded to the survey, a response rate of 52 percent. There were 1,479 early-career respondents, a response rate of 43 percent.
Adil Shamoo, a University of Maryland bioethicist and editor of the journal Accountability in Research, said some scientists are in denial, believing all scientific flaws eventually get corrected.
"If somebody publishes something wrong, another guy will publish another experiment and correct it," the thinking goes, Shamoo said. However, he said, "That is not true in all cases." Clinical trials, for example, are rarely repeated by others. "Something can be done wrong and go on for 20 to 30 years."
The Association of American Medical Colleges opposed a 2002 effort by the Office of Research Integrity (ORI) to conduct a similar survey, thinking it might have been aimed at expanding the definition of research misconduct. However, while it opposed expanding the regulations, it doesn't object to discussions about what's ethical and unethical, or professional and unprofessional, said Susan Ehringhaus, the AAMC's associate general counsel. Those discussions are likely to be sparked anew by today's study, she said.
Percentage of scientists who say that they engaged in the behavior listed within the previous three years.
Falsifying or 'cooking' research data 0.3
Ignoring major aspects of human-subject requirements 0.3
Not properly disclosing involvement in firms whose products are based on one's own research 0.3
Relationships with students, research subjects or clients that may be interpreted as questionable 1.4
Using another's ideas without obtaining permission or giving due credit 1.4
Unauthorized use of confidential information in connection with one's own research 1.7
Failing to present data that contradict one's own previous research 6.0
Circumventing certain minor aspects of human-subject requirements 7.6
Overlooking others' use of flawed data or questionable interpretation of data 12.5
Changing the design, methodology or results of a study in response to pressure from a funding source 15.5