5 Comments
I just participated in the QUEST for Research Excellence program where I spoke on a number of issues that affect research integrity. Since it has been on my mind, I thought I’d blog on the topic. Research integrity is the reason why we have many of the regulations, guidelines, and policies in place that safeguard our research, but, ultimately, you, the researcher, are responsible for conducting your research in a responsible, ethical manner. This shared responsibility we have with you enables us to maintain the public trust and have confidence in the research record and results. Every contributor in NIH-funded research has a role in maintaining this integrity—from student to PI to grantee institution to federal staff. I thought it would be useful to give you a sense of NIH’s role when we receive an allegation of research misconduct, which is only a small piece of the world of research integrity, but an important one.
When one of our staff members receives an allegation of research misconduct, they report it to a central individual within their institute or center. This person then contacts my research integrity staff in the Office of Extramural Research. We do a preliminary review of the information. In some cases, the allegation involves issues such as human subject protections, animals, or misuse of funds, and we forward those to the appropriate offices, the Office of Human Research Protections, Office of Laboratory Animal Welfare or Office of Management Assessment, respectively.
For the remaining cases, where there is sufficient and appropriate information, the allegation is forwarded to the Office of Research Integrity (ORI) within the Department of Health and Human Services. ORI is responsible for investigating allegations of research misconduct for all of the HHS divisions, including NIH. ORI works together with the researcher’s institution to investigate the allegation, which can take several months or more. Ultimately, if ORI determines that misconduct was committed, ORI places an announcement on their website that explains what was found and if administrative actions are imposed, what sanctions are placed on the investigator. You will also see these notices published in the NIH Guide for Grants and Contracts when they involve NIH-supported research.
Obviously, we hope, and it proves true, that these cases are few in number. We believe that training scientists in the responsible conduct of research early in their careers is an important way to maximize research integrity. To that end, all researchers supported by NIH training grants are required to receive training in this area, and the training plan is evaluated when the application is peer reviewed.
I am the Agency Extramural Research Integrity Officer and have been involved in research integrity throughout my entire career. I am continually impressed at the level of respect scientists receive from the public for the phenomenal research that they do, particularly our biomedical research scientists. But because of our impact on the lives of these same people, which is felt every day by most everyone, this respect can be lost by a single case where integrity is questioned. So that is where preventative measures maintain the integrity of the valuable and precious research we support. I look to you to work at the highest levels of integrity possible, always keeping the public trust in mind.
“We believe that training scientists in the responsible conduct of research early in their careers is an important way to maximize research integrity.”
While this belief may indeed be valid, the NIH and US science enterprise has little control of early experiences in research integrity since many of the current scientists vying for funding received their early training in other countries, under mentors who did not receive NIH training fellowships etc., or under less rigorous situations. I suggest that the NIH/study sections consider putting more emphasis on a scientist’s past training and contributions when evaluating funding. How reliable has previous work been by the individual and their mentors. Has it held up to scrutiny for 10 years or more? Was the contribution ever really of sufficient impact to ever receive scrutiny?
A good sales job on a current proposal should not be allowed to mislead grant/manuscript reviewers and the US public on the ultimate value of the research. If the more reliable, higher impact, thoroughly scrutinized scientist were poor in the sales job area, the current review system would be greatly biased toward funding unreliable and short-lived research. Further, some study sections are clearly better at scrutinizing integrity than others- the NIH should start putting more pressure on the weak study sections to scrutinize the integrity of researchers and their proposals. There may be a flaw in the current study section arrangement of review because of the differences in scientific/medical cultures and training that govern what is considered scientific integrity and the grouping of like minds(virtues) in the review of the same proposals and manuscripts, etc.
Thanks for this good overview. There are a few tems troubling to me , regarding misconduct…
1) The sanctions against the individual are either not harsh enough, or uniform enough. Banning an individual from serving in an advisory capacity or from recieving federal funding is a good start, but the ban is often only for a couple of years. A more widespread and uniformly applied ban would seem to be more appropriate. The current system sends a message that “if you get caught, you might get banned, it depends”. The repayment/recovery of pilfered grant money also seems to be handled in an arbitrary manner. Sanctions against the administrators at instutitions also seem weak – if you are the Dean and misconduct occurs on your watch, surely that should influence things such as setting of indirect cost rates?
2) Researcher ID. Communicating a ban to all affected agencies and individuals/institutions requires an assurance that we’re all talking about the same person. Without a uniform identification system it is often difficult to track down individuals who have performed misconduct. They can often get jobs in obscure (or not so obscure) institutions and fly under the radar for a few years before resurfacing to perform more misconduct. PubMed, for example, could then include a simple identifier tag – “author has had N other articles retracted” – to alert readers to potential problems.
3) The time currently taken for investigations is far too long. Investigations can drag on for months or years at a time, and during this period the accused inviduals continue to “poison” the field. I can recount at least 3 cases where individuals have been accused, and published 20-30 papers before being officially reprimanded in public. A mechanism to inform journal editors of an ongoing/preliminary invesigation would be useful, so at least their papers can be afforded higher scrutiny.
4) There is currently very little incentive to report misconduct, and in the case of junior investigators, significant disincentives exist – from political backlash to concerns over confidentiality/privacy. Many institutions do not publish the names/contact info’ for their research integrity officers (this should be mandated for receipt of federal funding). Even this article details multiple places where misconduct can be reported… OHRP, OLAW, ORM, ORI, OER. In many cases, one also wishes to report directly to a journal rather than involve funding agencies. When is this an appropriate course of action? A single landing page with a flow-chart to direct people to the correct place to report misconduct, would be a great help.
5) Time and time again, misconduct comes down to a very simple list of things that people do wrong. At the top of the list is faked western blot loading controls. This method is so open to abuse, some would like to see it banned altogether. Some journals have standards for image integrity, but again this is far from uniform. NIH is in a position to do something about this… mandate quality control standards for submitted images. Any journal that does not comply, articles in that journal cannot be included in progress reports for NIH grants. “Publication charges” in grant budgets could only be spent at compliant journals. The drop off in submissions to non-compliant journals would be rapid. There is a definite feedback loop wherein publications –> funding –> publications. That loop needs to be broken when people commit misconduct, and better communication between publishers and funding agencies is one way to ensure this.
Academic dishonesty is not limited to the content of publications. A critical issue is hubris and self-promotion as contrasted to collaboration with an open mind to new information. Within lab groups, failure to help results in work overload for responsible investigators, and hazing of new lab members is a cultural norm in some labs. Lack of oversight results in defamation, unfair competition, and turnover of top scientists. Ethics, truth, and cooperation are essential for academic excellence in publications.
I agree completely with Paul. We’re heading in the right direction, but we have a long way to go. Every instance of misconduct weighs the scales of public opinion against science, just when we need public support the most. Federal monies are going to become even more limited if the perception exists that we are misusing funds. It’s tough enough convincing Joe Public that scientists can disagree without there being fraud committed. We need, as a community, to realize how this looks to non-scientists.
Not only the number of publications have increased with fraudulent data but also grant applications and graduate theses. “Preliminary data” gives investigators to fabricate data as well as to alter data. They could always claim that the data didn’t bear out. PhD theses are dime a dozen and if one compares thesis from a given mentor, they would notice blatant plagiarism, self-citations, duplication of results, and scientific fraud. Internal thesis advisers are often chosen because of their connection to the mentors. Thesis approval has mostly become internal departmental approval. There is little incentive for the students to be original or creative as their own mentors just want to get you out and publish something.
My suggestion:
1. Applications from senior, credible investigators should be one page summary of whet will be done and the justification. Most of them are tenured. Salary component should be adjusted for tenure status and should not exceed 20%
2. Mid-tire investigators: Two pages: Creativity, originality, institutional support and budget justification. Some of them are tenured. Salary component should be adjusted for tenure status and should not exceed 30%.
3. For beginners level applicants: Five pages: Short and long term goals, specific objectives, establishing independence, creativity, originality and institutional sup[port, with a fixed budget. Salary sup[port to be 75%
They all should be taught in the beginning at NIH a couple of weeks long workshop on ethics, consequences, and how they could climb up to item 2 and 1.
There are cultural variations about ethics. An Indian, European, Chinese, Russian, and an American don’t view duplication, statistical analysis and plagiarism as ethical violations.
I am aware of someone who wanted to patent the outcome off a single experiment, with the blessing of the institution. Institutions are happy as long as they meet research dollar criteria; ethics, they don’t care as long as “it” doesn’t hit the fan.