Integrity in Scientific Research
Last year, the Institute of Medicine published a major report1 that does not seem to have inspired much response. A fairly thorough search showed that only one journal, JAMA, mentions the IOM's Integrity in Scientific Research: Creating an Environment that Promotes Responsible Conduct (in a book review).2 Furthermore, the IOM's Web page, in announcing that a town meeting would be held to discuss the report, promises, but does not deliver, the workshop's agenda and audio presentations.3
The Committee on Assessing Integrity in Research Environments commissioned me to write three background papers4; I also attended the town meeting, held in Washington last year. I hope to stimulate more public discussion by commenting briefly on the report and making my own suggestions.
The IOM report makes six recommendations. The first, acknowledging the dearth of empirical knowledge on the matter, urges funding agencies to "establish research grants to identify, measure, and assess those factors that influence integrity in research." The second and third recommendations ask research institutions to "promote integrity in research" and to offer "effective educational programs that enhance the responsible conduct of research."
I concur, as did most meeting attendees.
The last three recommendations concern "institutional self-assessment of integrity in research," which research institutions should undertake as a part of accreditation processes; they would be tracked by the Office of Research Integrity. The meeting attendees found these recommendations more troublesome. Accreditation, some said, is itself a problematic process that does not always accomplish its stated goals, and several others opined that no one knows how to do a self-assessment in research integrity.
Although Committee members made a strong case for self-assessment, I doubt that it will be broadly implemented. As I commented to them, it's a sad day for science when integrity in research cannot be taken for granted. While serious cases of research misconduct have occurred, they've been few; less serious cases of irresponsible research, while certainly more numerous, also probably are not serious enough to warrant this level of self-scrutiny. While it would be good to know the state of integrity at every research institution, I am not convinced that asking researchers to spend their limited time on yet another self-assessment is worth it.
I offer these suggestions for improving the research climate.
1. Reduce the pressure to publish: If you want to encourage someone to cut corners, pay on a piecework basis. In science, quality should be more important than quantity. All funding agencies and research universities should base decisions on grants and promotions on three (five? 10?) of a researcher's best publications. The researcher would select those for consideration; criteria for judging a publication's quality should be publicly stated and at least partly objective.
2. Create disincentives for multiple publication. Publishing two or more substantially similar papers without adequate citation should be treated as plagiarism--research misconduct.
3. Audit the practice of research. The systems of research oversight, including institutional review boards (IRBs) and institutional animal care and use committees, are dependent on trust. If researchers do not follow an approved protocol, there is little chance the IRB will ever know about it. Those bodies responsible to oversee research are ill-prepared and poorly designed to take on an auditing function. A separate office, coordinated closely with other relevant committees, should be established in all research centers to audit conduct regularly. This office also could investigate misconduct allegations and trace protocol violations back to their sources. A separate office would have responsibility for judging the outcome of investigations and designating and enforcing recommended sanctions.
4. Provide positive incentives for training in the responsible conduct of research (RCR). The only reason some cooperate with oversight bodies is because of the clear threat: "Do your paperwork or you can't do your research." This creates a counterproductive atmosphere of hostility. Instead, the federal government should encourage universities to try positive incentives, rewarding researchers who undertake RCR training and who display positive attitudes and behaviors.
No one would seriously deny that research institutions and funding agencies have a right and a responsibility to demand high standards of ethical conduct. But somewhere along the line, "upholding high standards" shades into "setting them up for failure." Not every scientist can produce at the levels expected without being tempted to cut corners. And some cannot resist.
Kenneth D. Pimple is the director of Teaching Research Ethics Programs at the Poynter Center for the Study of Ethics and American Institutions, Indiana University-Bloomington.
1. "Integrity in scientific research: Creating an environment that promotes responsible conduct," National Academies Press, available online at www.iom.edu/report.asp?id=4438
2. M. Kalichman, "Integrity in scientific research: Creating an environment that promotes responsible conduct," JAMA, 289:2433-4, 2003.