The scientific community is founded on honesty, collaboration, and open access to information and resources. All of its members are expected to meet high standards of integrity, accuracy, and objectivity. I suspect many members of ABLE, like me, were never taught these professional expectations formally. It was assumed we would learn these by observing our mentors and colleagues. Unfortunately this does not appear to work. Research misbehavior and misconduct are being reported ever more frequently, and each case that comes to light further erodes public confidence in science and support. While some high-profile cases involve senior scientists, it is students, trainees, and early-career scientists who are most likely to be reported for violations of federal regulations.
Studies from the federal Office of Research Integrity (ORI), Council of Graduate Schools (CGS), and other groups indicate that stricter enforcement and oversight is not the solution to this problem. Instead, well–designed training programs are the most effective way to reduce misbehavior and misconduct. Federal funding agencies have responded with new regulations stating all undergraduates, graduate students, and post–doctoral fellows supported by their grant dollars must receive formal, ongoing training in Responsible Conduct of Research (RCR). The general goal of such training is to teach students the uniformly accepted standards for conducting and reporting research in a discipline. More specifically, RCR training is meant to:
- Raise situational awareness, so individuals are not caught in or negatively affected by misbehavior or misconduct of other scientists (mentors, peers, etc.)
- Reduce the likelihood that individuals will engage in misbehavior or misconduct themselves.
Over the past four years, my colleagues and I have developed and tested a PBL-based RCR training program for graduate students. In this article I want to share some of what we learned along the way about underlying principles of RCR, and how new federal regulations are going to affect faculty and staff who train undergraduate and graduate students on a day-to-day basis. This article is not meant to be an in-depth review, but rather to provide readers with a working knowledge of the issue, and a greater understanding of the added training responsibilities we are likely to face in the immediate future.
Will RCR training requirements even apply to students at your institution?
If your department or program uses any NSF or NIH grant money to support students or curriculum reforms, then the answer is yes.
In 2007, NSF’s current appropriation was signed into law as the America COMPETES Act. Section 7008 states that all undergraduates, graduate students, and post–doctoral fellows supported with NSF money in any way must receive RCR training. Formal policies are being developed currently, but are expected to be similar to those of the NIH (which are described next). Formal NSF policies could be in place as early as mid–2009.
NIH already requires a formal RCR training plan as part of all new R25 (Research and Education), and T34 (MARC) undergraduate program proposals. NIH’s requirements for graduate students depend on how the student is funded. All pre–doctoral trainees who apply for a grant (NRSA, F31, F32, etc.) are required to have training. The same rule applies to all post–doctoral trainees applying for any type of NIH grant. If a student or fellow does not have documented training, the grant they apply for will not be awarded. The same requirement applies to pre- and post-doctoral trainees supported by a T32 or other training grant. Individual investigators still can support graduate students who have not had RCR training on an R01, R29, or similar grant, but rumor is that NIH will set a uniform requirement that all students supported by grant or contract funds in any way must receive RCR training.
Currently, RCR training is not required for graduate students who are supported by internal teaching assistantships from their institutions. However, the Council of Graduate Schools (CGS) is strongly encouraging member institutions to develop programs that provide significant, ongoing RCR training to ALL graduate students and fellows, regardless of funding source.
Why is ORI so important in current RCR training efforts?
The Office of Research Integrity (ORI) is an agency within the US Dept. of Health and Human Services. ORI’s mission is to oversee and ensure the integrity and accuracy of research conducted by Public Health Service (PHS) agencies, including the National Institutes of Health (NIH). It is responsible for developing policies, procedures and regulations related to the detection, investigation, and prevention of research misconduct. ORI conducts investigations, and provides technical assistance to institutions as they respond to charges of misconduct. More recently, ORI has begun to implement activities and programs to teach the responsible conduct of research, promote research integrity, and prevent research misconduct.
Given its mission and history, ORI has become the primary resource and opinion leader on issues relating to RCR, both in and outside of the federal government. ORI’s definitions of misconduct and misbehavior, statements of policy, and training guidelines have been adopted by other federal and private agencies, including NSF.
How does ORI define inappropriate behavior for scientists?
ORI recognizes two types, the more serious being research misconduct. Three behaviors (often abbreviated FFP) fall under this heading: fabrication, which is making up data or results and recording or reporting them; falsification, which is manipulating research materials, equipment, or processes, or changing or omitting data or results so the research is not accurately represented in the research record; and plagiarism, that is, appropriation of another person's ideas, processes, results, or words without giving appropriate credit.
Misconduct is more likely to grab headlines, but is reported less often than research misbehavior. Sometimes called questionable research practices (abbreviated QRP), they are just as corrosive to the research enterprise. Here are some examples (the list is not exhaustive).
- Failing to properly record data.
- Failing to retain records to support published data.
- Improper or insufficient protection of sensitive data or private information.
- Giving or denying authorships improperly.
- Misrepresenting speculation as fact.
- Refusing to grant reasonable requests for access to reagents created using public funds
- Using inappropriate statistical methods
- Providing inadequate supervision to subordinates, or exploiting them for personal or professional gain.
What should RCR training include?
Given the circumstances in which misbehavior and misconduct have occurred in the past, ORI recommends all RCR training programs cover nine core topics.
- Rules of data acquisition, management, sharing and ownership
- Guarding against conflicts of interest and commitment
- Protection of human subjects
- Ensuring the welfare of animals used in research
- Research misconduct: fabrication, falsification, plagiarism
- Acceptable publication practices and responsibilities of authorship
- Responsibilities of mentors and their trainees
- Responsibilities when conducting peer reviews
- Responsibilities towards scientific collaborators
What are some of the problems associated with implementing RCR training?
Most everyone agrees in principle with the intended spirit and tenets of RCR training. That said, implementation has proven difficult for several reasons. ORI and other agencies have left the details of developing a training program that covers these topics up to individual institutions. Currently there is no uniformly accepted standards for course structure, length of training, or format. For a time this threatened to permanently derail the initiative. Fortunately, ORI and NIH are starting to provide additional guidance, and institutions are starting to make training materials and program models available.
It should not be a surprise that there is significant resistance from institutions who do not want to shoulder the cost of course development, and from investigators who think the time their trainees will be away from the bench is being wasted. The latter argue that the vast majority of scientists still are hard working, honest, and diligent; why add more regulations if the informal apprentice system works fairly well already? CGS has published an excellent short report intended to assist deans and administrators in overcoming local resistance to RCR training. It can be purchased online at www.cgsnet.org/; under “Publications,” search for Item Code “GRADRCR.”
At this early stage, the cost–effectiveness of systematic RCR training is unknown. Will money spent on curricular development actually produce any positive change in behavior? There are two reasons for continuing to provide RCR training, even when there is no clear evidence (as yet) that it is effective. First, operating RCR programs are needed in order to evaluate their effectiveness. Second, most RCR training touches on areas for which there is a well–documented need, specifically mentor/trainee relationships, and protection of human subjects. Ultimately the best argument for continuing current RCR efforts (at least for now) may be the aphorism “imperfect action is better than perfect inaction.”
Where can instructors and trainers go for assistance in developing programs?
The best place to start online is ORI itself, at www.ori.dhhs.gov/education. They have an extensive list of suggestions for how to incorporate RCR training as elements within existing undergraduate and graduate training programs, along with links to active programs at other institutions. ORI also provides many publications as free PDFs. Nicholas Steneck has written a particularly good introduction to RCR, which is available at http://www.ori.dhhs.gov/documents/ rcrintro.pdf.