Research Misconduct: Dealing with Allegations of Academic Fraud and Other Transgressions “There are instances. I mean, there have been scandals in the past, but they’re almost always disputed. There’s almost always some explanation. Look, . . .t come out in refereed journals and lie about their work. It’d never get away with it. There are too many safeguards. They’by their principal investigator. Or the referees would find something fishy in the article.And then, even if the thing did get published, no one would be able to reproduce thework later on, and the scientific community would catch them. Think about it logically.Once your paper is published, everyone else is going to try to follow in yourexperimental footsteps, and they’ll get bad data of their own. It’months later or six years later, the truth will out. You suppress bad data – it’cost you your career.” Allegra Goodman, Intuition (Random House 2006).
These are the first reactions of one character in Allegra Goodman’s recent novel, Intuition,
on hearing a friend express disturbing suspicions that a scientific colleague has falsified data. Theyalso may sum up the views of many today about fabrication, falsification, and plagiarism,academe’
s cardinal sins: we may occasionally see sensational stories in the press, but really,
transgressions are rare, sinners few; and because, reliably, the sin will be exposed and the sinnerpunished, the ultimate impact of such fraud is very limited.
But this view may no longer be correct. When looked at nationwide (and even more so,
internationally), in the past five to ten years, public cases do not appear to be nearly as rare or aslimited in impact as Ms. Goodman’
s character surmises. All signs are that the baseline level of
academic fraud in academe is now higher than it was thirty years ago. The New York Times hasrecently described the trend as follows:
The South Korean scandal that shook the world of science last week is just one sign of aglobal explosion in research that is outstripping the mechanisms meant to guard against errorand fraud. . . . Experts say the problem is only getting worse, as research projects, and thejournals that publish the findings, soar.2
The rising tide of fraud has been sufficient to support the launch of a new online journal, Plagiary,devoted solely to reports and discussions of fraud, fabrication, and plagiarism.3 Statistics kept by theU.S. Department of Health and Human Services’
Office of Research Integrity (“ORI”) suggest an
1 Please note that the views presented in this paper are those of the author, and not of Yale University. 2 Altman, L., and Broad, W., “Global Trend: More Science, More Fraud,”New York Times (Dec. 20, 2005). 3 Plagiary: Cross-Disciplinary Studies in Plagiarism, Fabrication, and Falsification,. Lesko, ed.),(visited on May 27, 2006).
National Association Of College and University Attorneys
overall increase in research misconduct involving Public Health Service-supported work as well.4Plagiarism by public figures, some of them based in prestigious colleges and universities, has alsobeen the subject of repeated, high profile scandal in recent years.5 In short, we should expect ourinstitutions to handle more frequent, more complex, and more public academic fraud cases incoming years. If an institutional policy lumps its treatment of academic fraud with other research-related misconduct (such as negligent supervision of research assistants and inappropriateauthorship practices), it should anticipate all the more activity.
Why this increase? There surely has always been some amount of chronic academic fraud,
the crime being no more than a specialized expression of the general human susceptibility totemptation. The increasing competition for decreasing research dollars (with faculty salarysometimes directly tied to success in obtaining research awards) and the voracious institutionalappetites and incentives for publications create additional reasons for researchers to cut corners inillegitimate ways. The ready availability and common use of software tools that make imagemanipulation and copying easy certainly are factors as well. More importantly, expanding amountsof interdisciplinary, domestic inter-institutional, and international research make it more difficult forcollaborators to know each other’
s work thoroughly, and possibly for reviewers to review it: these
factors, too, appear to contribute to the problem and suggest, cumulatively, that growth can beexpected to continue.6 Research into research misconduct suggests that perceptions of being treated“unfairly” also impel scientists to commit fraud.7 In a well-known study published in Nature in2005, more than 30 percent of several thousand surveyed NIH-funded scientists admitted toengaging in some form of research misconduct in the three years before they answered the 2002survey– a startlingly high percentage, even if for present purposes it should be discounted becauseof the breadth of the survey’s approach.8 This study’
s findings suggest that standards overall have
been dropping, educational programs about research ethics have weakened, and detectionmechanisms may be inadequate.
4 ORI Newsletter (June5 See, e.g., accounts of the cases of Stephen Ambrose and Doris Kearns Goodwin, in Hoffer, P., Past Imperfect: Facts,Fictions, Fraud – American History from Bancroft and Parkman to Ambrose, Bellesiles, Ellis, and Goodwin (PublicAffairs 2004). See also, Special Report on Plagiarism, Chronicle of Higher Education, (December 17, 2004)(including anumber of related articles, including Bartlett, T. and Smallwood, S., “Four Academic Plagiarists You’ve Never HeardOf: How Many More Are Out There?”)6 Reports of common plagiarism and other inappropriate conduct by students in general and at some specific institutionsalso suggest that the situation is likely to become only more fraught in future years. See, e.g., Bartlett, T., “Ohio U. Investigates Plagiarism Charges,”Chronicle of Higher Education (March 10, 2006)(“Ohio University is investigating44 possible cases of plagiarism by current and former engineering graduate students, all of which were discovered by aformer graduate student who believes that professors there have fostered a culture of cheating. … “); Wasley, P.,“Review Blasts Professors for Plagiarism by Graduate Students,”Chronicle of Higher Education (June 16, 2006);Glater, J., “Colleges Chase As Cheats Shift to Higher Tech,” (New York Times, May 18, 2006). 7 Martinson, B., et al., “Scientists behaving badly,”Nature, 435: 737-38 (June 9, 2005). 8 Id. See also, Wadman, M., “One in three scientists confesses to having sinned,”Nature 435: 718-719 (June 9,2005)(“Overall, about a third admitted to at least one of the ten most serious offences on the list – a range ofmisbehaviors described by the authors as ‘striking in its breadth and prevalence.’
about the incidence of the following “top ten” behaviors: (1) falsifying or ‘cooking’ research data; (2) ignoring majoraspects of human-subject requirements; (3) not properly disclosing involvement in firms whose products are based onone’s own research; (4) relationships with students, research subjects or clients that may be interpreted as questionable;(5) using another’s ideas without obtaining permission or giving due credit; (6) unauthorized use of confidentialinformation in connection with one’s own research; (7) failing to present data that contradict one’s own previousresearch; (8) circumventing certain minor aspects of human subject requirements; (9) overlooking others’
data or questionable interpretation of data; (10) changing the design, methodology or results of a study in response topressure from a funding source.
National Association Of College and University Attorneys
The broadest issue for consideration in the higher education community is how American
universities and publications should address the problem, because any perception that eitheracademic fraud is common or not well handled when found seriously damages the entire academicenterprise. Modifications of the peer review system might be considered, as should imposition ofincreased obligations on collaborators, and renewed and intensified efforts to educate members ofthe academic community (including undergraduates) about the appropriate standards for behaviorand their application in real life situations. An international databank of researchers with anestablished history of fraud should be considered for development, as should a depository systemfor data supporting publications in leading journals. The university community, together withpublishers of academic journals, professional associations, and leading research sponsors(including, but not limited, to the federal government), needs to take a fresh look at the entire issue,with a horizon broader than federally funded research.
In the meantime, however, counsel need to identify ways in which institutional policies and
procedures can be refined or modified to handle with an appropriate level of formality the full rangeof matters that they can be expected to address. Counsel also should be prepared to offer thoughtfuljudgments on nuanced procedural and substantive questions that can be expected to arise inparticular types of cases under adjudication, always aware that these matters can take on a publiclife of their own. These questions combine amorphous principles of scholarly practice and academicethics, vaguely legal notions of intellectual property, and quasi-constitutional concepts of dueprocess. The result is an intriguing mixture. This paper offers some perspective and reflections onthese topics, with the goal of imparting useful guidance.
In Part I, it provides background on the basic concepts and procedures and a brief overview
of the history of academic fraud and its regulation in the United States. In Part II, it identifies anddiscusses a series of questions for counsel to consider about institutional policies regarding researchmisconduct. Part III offers thoughts on some practical issues that arise during a research misconductinquiry and investigation. Part IV presents closing notes. An attachment offers a list of “SelectedReadings on Academic Misconduct,” for those who wish to read further on the subject. (Some ofthe “selected readings” also appear as footnoted sources in the text.)
Part I General background
The Office of Research Integrity (“ORI”), in the U.S. Department of Health and Human
Services, has in the past taken the lead in establishing procedural standard for handling all academicfraud and other research misconduct matters. It continues to do so now in its new regulations fordealing with research misconduct in Public Health Service-funded research (42 CFR Part 93,replacing Part 50),9 which implements guidance from a government-wide directive issued by theOffice of Science and Technology Policy (“OSTP”) in December 200010 and further articulatesacceptable institutional processes for addressing allegations of research misconduct.11 But many
9 70 Fed. Reg. 28370-28400 (May 17, 2005; effective June 16, 2005). 10 OSTP, “Federal Policy on Research Misconduct,” (December 2000), available at11 The National Science Foundation has its own policies and procedures, 45 CFR Part 689, also based on the OSTPgovernment-wide guidance, but I think it is fair to say that ORI and misconduct in NIH-funded research have been morein the forefront of public discussion on the issue than has NSF, possibly simply because of NIH’
National Association Of College and University Attorneys
research misconduct and academic fraud allegations fall outside the ORI purview, either becausethey do not fit within the new ORI definition of misconduct, or because the research in questiondoes not involve PHS funding.12 They may not involve science at all. Especially with the increased(and in some cases, quite appropriate) transformation of adjudications under ORI jurisdiction intomore trial-like proceedings, complete with burdens of proof and of going forward and otherevidentiary rules, universities must revisit the question posed when federal research misconductregulation was first imposed: whether the ORI/OSTP mechanism is appropriate for dealing withallegations that lie outside ORI or other federal jurisdiction. Many of our institutions are nowengaged in this sort of broad reconsideration, as we focus on the narrower task of ensuring that ourprocedural statements comply with the new ORI mandates for PHS-funded research.
Before proceeding further, we need to acknowledge the persistent confusion in this field that
is reflected by chronic use of inconsistent terminology, in particular, “academic fraud,” “academicmisconduct,” “scientific misconduct,” and “research misconduct.” Although these terms seem to beused interchangeably, it is worth considering what they convey to the ordinary listener
• “Academic fraud” seems most clearly to denote intentional fabrication, falsification,
or plagiarism, but probably excludes negligent supervision or improper authorshippractices.
• “Scientific misconduct” doesn’t suggest the same level of intentionality, but may
include reckless or sloppy handling of data and loss of research records, for example;by using the word “scientific,” however, it seems to exclude breaches of ethicalstandards in the fields of law or history.
• “Research misconduct” might include breaches in scholarship about law or history,
but not quite as well when the transgression does not directly involve research – forexample, the theft of an original idea in an essay.
• “Academic misconduct” seems broadest, encompassing all fields and all types of
activities in academe, but it seems too mild to use for crimes of intentionaldeception, especially to the lay person.
As discussed below, lumping treatment of the “crimes of moral turpitude” (fabrication,
falsification, and plagiarism) with what might be called the “civil offenses” concerning, forexample, certain authorship disputes among collaborators, creates a host of issues. For presentpurposes, I will refer to “academic fraud” when I mean the “crimes,” and “academic or researchmisconduct” when I mean the broader array of offenses.
When taking a fresh look at an institutional policy and procedure or seeking insight for
judgments on particular matters, it is very useful to review some of the history of researchmisconduct policy and proceedings, and to become familiar with some individual cases. Thesedemonstrate the many nuanced judgments that are involved in handling allegations effectively;
but also possibly because fraud in clinical and biological research seems a more personal betrayal of the public’s trust. Regarding NSF procedures generally, see “Dear Colleague” letter dated November 26, 2002, Peggy L. Fischer, Ph.D.,Associate Inspector General for Investigations, OIG, NSF, available at 12 See 42 CFR §93.102, “Applicability.”
National Association Of College and University Attorneys
illuminate the complex psychological realities of complainants, perpetrators, and adjudicators; andmake all too clear how easily such cases can spin out of control. They also provide a roadmap to thehistory of regulation of fraud in government-sponsored research, which can serve as a useful tool indetermining an institutional approach to procedure. A bibliography of some helpful reading in thisvein is attached below. A sketch of some of the high points follows. A little history
With apparent regularity since the 1970s in the United States (and probably before), there
have been outbreaks of serious and well-publicized incidents of research fraud, each of which hasprompted a strong reaction in the scientific, governmental, and public communities13, and eachtime, almost as if academic fraud were a shocking new phenomenon. Beginning, for example, in1974 with the notorious episode of William T. Summerlin’s “genetically altered” mice that were infact crudely painted black,14 through the Baltimore affair in the 1980s,15 the Schön case at Bell Labsin the 1990s,16 and the front page Korean stem cell scandal this year,17 high profile incidents orserious allegations of fabrication and falsification have periodically exploded in the news, oftenrevealing a lengthy and appalling history of fabrication by an individual with a previous andperhaps deserved reputation for brilliance. They have occurred at institutions of all stripes. Theseepisodes have predictably fueled outraged calls for regulatory changes (when federal funding isimplicated); improved review and reaction by scientific publications; and prompted periods of moreattentive university management of both the research process and the detection and adjudication ofallegations.
Recent cases have exposed some problems in the current methods of handling such matters
and screening for fraud in the scientific community and in the legal and regulatory systems. In2005, the Poehlman fabrication case in clinical research conducted largely at the University ofVermont become public.18 The investigation, prompted by a whistleblower, was by all accountswell handled by the University, but the scope of the fraud and its duration without detection wastroubling: the matter resulted in the retraction of numerous papers spanning a decade of publication,54 separate findings of scientific misconduct, related civil, criminal, and administrativeproceedings, a lifetime bar from participation in U.S. government programs, and a guilty plea in arelated criminal felony prosecution.19 Also in 2005, the British Medical Journal published a lengthy
13 See Mitcham, C., “Ethics and Scientific Research, A Selective Chronology with References,” (March 2003), availableat 14 Judson, H., The Great Betrayal: Fraud in Science (Harcourt 2004) 104-08 (describing Summerlin case). 15 This case of alleged data fabrication and falsification is described in detail in Kevles, D., The Baltimore Case: A Trialof Science, Politics, and Character (Norton 1998). 16 See “Report of the Investigation Committee on the Possibility of Scientific Misconduct in the Work of Hendrik Schönand Coauthors,” (September 2002), 17 Wade, N., et al., “Korean Scientist Said to Admit Fabrication in Cloning Study,”New York Times (Dec. 16, 2005). 18 ORI Press Release, Eric T. Poehlman (Mar. 17, 2005),19 See Dahlberg, J. and Mahler, C., “The Poehlman Case: Running Away From the Truth,” Science and EngineeringEthics (2006) 12: 157-173; ORI press release, (March 17, 2005); ORI report, On June 28, 2006, Dr. Poehlmanwas sentenced to a year and a day in federal prison. AP, “Former UVM Researcher Sentenced for Falsifying Work,”(June 28, 2006)(Dr. Poehlman “will serve the sentence at a federal prison work camp in Maryland.”)k/.
National Association Of College and University Attorneys
and very troubling account of its decade-long efforts to deal with years of apparently fraudulentresearch published in the ‘
90s by Dr. Ram J. Singh, who, absent a home institution and no obvious
professional body to “own” his work, seems to have escaped responsibility. In Nutrition, 2005 sawthe retraction of a paper that one pre-publication reviewer had said “had all the hallmarks of havingbeen entirely invented.”20 The “self-correcting nature of science,” the peer review, and publicationreferee systems that have provided comfort in the past clearly did not work in these cases. The costcan be years of researchers relying unwittingly on fraudulent results, and lasting damage to thereputation of science and publicly funded scientists.21
Since the first days of 2006, the continuing cascade of scandal reported in the general press
suggests that it is time for a broader review of the systems in place to prevent and deal with suchfraud. Even an incomplete list is striking: We have seen the downfall of the Korean stem cellresearcher Dr. Hwang, noted above, for fabricating stem cell cloning research results and a related“research misbehavior” finding by University of Pittsburgh with regard to Dr. Hwang’scollaborator, Gerald Schatten22; the announcement that Prof. Chen Jin, at Jiaotong University inChina, falsely presented as his own research results in microchip design that he had stolen23;findings of plagiarism, falsification, and fabrication against Prof. Ward Churchill at University ofColorado at Boulder, the timing of which (perhaps unjustifiably) raises the question of whether suchproceedings have been used in retaliation for other behavior by Prof. Churchill24; and the retractionby a Columbia University professor of five papers that had been based on research by a graduatestudent. 25 (The student reportedly has disputed the retractions and reportedly said, from her currentlocation at another university in Germany, that “she had not been told that the papers were beingwithdrawn.”26)
The recent exposés of fraud in China and South Korea highlight a special need for current
procedures to anticipate how to address allegations of research misconduct that arise in the contextof international collaborations. In a vein similar to that of the Nature findings mentioned above forthe United States and some of those from Ohio State University, the Christian Science Monitorreported recently based on one study that research fraud in China is “rampant,” with over 60 percentof Ph.D. candidates having admitted to plagiarism and bribery in connection with their academicwork.27 Standards in other cultures regarding topics such as authorship but also regardingplagiarism and even treatment of data may be different from those in the United States, and foreignresearchers based here as well as collaborators from abroad may need to be educated specially aboutthe applicable standards.
20 White, C., “Suspected research fraud: difficulties of getting at the truth,”BMJ 331:281-288 (July 30, 2005); R. Smith,“Investigating the previous studies of a fraudulent author,”BMJ 331:288-291 (July 30, 2005)(discussing Nutritionretraction). 21 See Sox., H. and Rennie, D., “Research Misconduct, Retraction, and Cleansing the Medical Literature: Lessons fromthe Poehlman Case,”Ann. Intern. Med. 2006: 144:E-7-E-11 (discussing Poehlman case and other instances). 22 Wade, N., “University Panel Faults Cloning Co-Author,”New York Times (Feb. 11, 2006). 23 Barboza, D., “In a Computer Scientist’s Fall, China Feels Robbed of Glory,”New York Times (May 15, 2006). 24 Report of the Investigation Committee of the Standing Committee of the University of Colorado at Boulderconcerning allegations of academic misconduct against Professor Ward Churchill (May 9, 2006)25 Chang, K., “Professor at Columbia Retracts Papers Over Research Questions,”New York Times (Mar. 16, 2006)26 Chang, K, “Columbia Chemistry Professor is Retracting 4 More Papers,”New York Times (June 15, 2006). 27 Marquand, R., “Research Fraud Rampant in China,”Christian Science Monitor (May 16, 2006).
National Association Of College and University Attorneys
Procedures for handling allegations need to be flexible but effective and predictable for a
number of reasons. First, despite the common use of the term “whistleblower,” it is important toremember, always, that allegations of fraud and other misconduct may be erroneous. Sometimes,close examination reveals misguided allegations and a factual record susceptible to significantlydifferent interpretations. For example, the fabrication and falsifications charges against TherezaImanishi-Kari, which resulted in the resignation of her faculty supervisor and senior author, DavidBaltimore, from Rockefeller University, were found not to be substantiated after years ofinvestigations, hearings in Congress, and prolonged agony for all concerned. (They are still debatednow.) 28 It is critically important, therefore, for the procedure to be scrupulously fair and timely, toafford due process to the respondent, and to resolve expeditiously, without becoming a full-blown“criminal” proceeding unless that is clearly warranted.
This is difficult to accomplish. Resolution of the factual claims may be extraordinarily
difficult, requiring significant scientific expertise and judgment, and highly sophisticated scientifictools. Also, there is significant pressure for more process to be considered “due,” because the stakesin the worst of these cases are very high for the accused. They can also be significant for a long listof others: the complainant; the home institution (or institutions); the institution where thechallenged work was conducted; the respondents’ and complainants’ collaborators, mentors, andstudents; research funders; the journals that published the falsified work; and those who publishedand who have relied on previously published work, which is often implicated as well. Theseinterests may tend to pressure the proceedings to become increasingly legalistic at the adjudicationphase – something that many institutions find difficult to handle, not having been established toserve as a trial court. What have standard processes been to date? We are all familiar with the multi-layered
protocol for the standard proceedings: since the early 1980s, those of our institutions that acceptfederal research funding have been required to establish and use a fairly uniform tripartite protocolfor addressing complaints of this nature: (1) assess the allegation briefly to ensure that it isnontrivial and apparently made in good faith; (2) conduct an inquiry to determine whether a fullblown investigation is warranted; (3) if warranted, conduct a formal investigation, probably on therecord. We also make provision for an appeal process, to review (but not re-do) the fact findingprocess. If PHS funding is involved, the entire matter may then be handed off to ORI for further,sometimes de novo, proceedings. In addition, complainants are to be protected from retaliation; andconfidentiality is to be maintained, at least for some period of time.
In 2000, the policies of the National Institutes of Health and National Science Foundation,
which until then had differed in significant procedural and definitional respects,29 were brought intoharmony (as were those of other agencies) by the action of the Office for Science and TechnologyPolicy. OSTP required all federal agencies to align their approaches with its directive. Asmentioned above, ORI amended its policies accordingly, issuing its final rule on May 17, 2005.30Most significantly, the ORI definition of “research misconduct” was altered to exclude thepreviously used catch-all “other practices that seriously deviate from those that are commonly
28 See Goodman, A., Intuition, supra; Judson, H., The Great Betrayal: Fraud in Science, supra. 29 For example, NSF treated breaches of human subjects regulations as falling within its definition of researchmisconduct but ORI did not; at NSF, the Office of Inspector General handled allegations brought to the agency; at NIH,the Office of Research Integrity and its various predecessors handled the allegations. Regarding NSF researchmisconduct proceedings generally, see Regulation of Research”). 30 70 Fed. Reg. 28370-28400 (May 17, 2005).
National Association Of College and University Attorneys
accepted within the scientific community for proposing, conducting, or reporting research.”31 Thenew definition of “research misconduct” under ORI procedures is limited to “fabrications,falsification, and plagiarism” (“FFP”) in “proposing, performing, or reviewing research, or inreporting research results.”32 Only those FFP that are of a type that constitute a “significantdeparture from accepted practices of the relevant research community” – a kind of materialitystandard - are actionable. In addition, the new regulation establishes a federal intentionality element– to constitute research misconduct for federal purposes, the act must be demonstrated to have beencommitted “intentionally, knowingly or recklessly”: negligence and gross negligence areinsufficient.33 The burden of proof remains “by a preponderance.”34 Further, as my co-presenter,Christian Mahler of the Office of Research Integrity, describes, as of 2005, institutions receivingfederal funds from PHS are required by the ORI to apply a number of new procedural standardswhen adjudicating charges of research misconduct related to PHS-funded research, such as burdensof proof and of going forward on various elements of a charge.35
Because NACUA’s recent archives include some very good primers and commentary on the
basics of the federal regulation related to research misconduct,36 which have changed and beensupplemented in the particular ways commented on by Christian Mahler, this paper will addressother issues: a selection of considerations to take into account when re-evaluating the institution’spolicy for handling allegations of research misconduct, and then some concerns that arise whenmanaging an inquiry or working closely with an investigation committee and managing aninvestigation. Part II Questions to consider about institutional policies and procedures The scope of the institutional policy. A key question in any review of an
institutional policy and procedure on academic misconduct is, “What behaviors should its processesaddress, adjudicate, and sanction?”
Since the 1980s, many institutions have relied on their research misconduct policies and
procedures to address fabrication, falsification, and plagiarism, and “other serious departures fromaccepted scientific practice” -- the catchall phrase for a range of misconduct important in theacademic community but not captured by ORI’
s FFP. That kind of misconduct might include, for
• authorship disputes (e.g., who is entitled to first authorship or senior
• failures to maintain data with appropriate rigor
• breaches of customary practices in publication (e.g., the corresponding
s failure to obtain permission from co-authors before submitting an
31 54 Fed. Reg. 32446, 1989 WL 282580 (Aug. 8, 1989). 32 42 CFR § 93.103. 33 42 CFR § 93.104. 34 42 CFR § 93.104(c). 35 42 CFR 93.106 (“Evidentiary standards”). 36 See Appendix I, Selected References.
National Association Of College and University Attorneys
• a first author’s refusal to supply collaborators with primary data
• a first author’s failure to obtain consent of co-authors to publish
• a principal investigator’s failure to supervise adequately researchers in the lab
• Unethical behavior, as judged by applicable professional standards
• Breach of human subjects, lab animal welfare, biosafety, and other research
Other institutions have excluded certain misconduct from the purview of their policy. Some do notinclude noncompliance with human subjects, lab animals, conflicts of interest, and other regulatoryresearch schemas under this adjudicatory umbrella, preferring to use other available grievancemechanisms. Also, an institution might choose to follow ORI’s lead in declining to adjudicateplagiarism allegations between collaborators (or former collaborators).
Now, OSTP (and hence, ORI) requires that for all federally-funded research (from NIH or
any other agency), recipient institutions use the prescribed procedures to address “fabrication,falsification, and plagiarism,”37 and excludes from its definition of research misconduct “otherserious departures.” The “serious departure” language now appears to establish a standard ofmateriality, only, rather than a different species of misconduct: that is, now, to constitute researchmisconduct, the alleged fabrication, falsification or plagiarism must have “seriously departed fromaccepted practices in the relevant research community.” But institutions may still choose to use thetripartite procedure (assessment, inquiry, investigation) for handling allegations related either tonon-federally funded research or to non-FFP charges (or both). This is because the process is now afamiliar one; it commits early phase fact-finding to experts in the field; and it may seem like a betterfit for fact-finding purposes than ordinary grievance procedures.
But there are several reasons why an institution might choose not to use the three-layer
approach for some allegations. First, there may be no need for the redundancy of the assessment andthen two different committee reviews. Second, the very confidential and increasingly legalisticinvestigation process may be neither optimal nor sensible for resolving those non-FFP misconductcharges that are more in the nature of personal disputes, such as quarrels among authors and formercollaborators over credit, or that are possibly negligent failures to supervise research associates. Forexample, it may not make sense for an investigation committee to apply the new materialitystandard, “significant departure from accepted practices,” in the setting of non-FFP wrongs. ORIprovisions regarding burden of proof and burden of going forward, and other standards, may notmake sense in the context of non-scientific, non-FFP proceedings. If other regulatory schemes (suchas those involving human subjects research) are implicated, ORI processes may not be appropriateto apply.
In reviewing scope, consider also title. Will it be confusing to the public audience (and even
defamatory) for a finding to be called research misconduct, when the transgression is a non-deceptive, minor “civil” breach whose punishment is likely to be a reprimand or re-education?Should a university make public a finding of research misconduct (as it might if there is fabrication)when a faculty member’s “misconduct” was (for example) negligent supervision of a lab or abehavior that a professional body has decided may be unethical? If the public equates “research
37 These are captured now whether they occur in the proposing, performing, reviewing the research or reporting theresults.
National Association Of College and University Attorneys
misconduct” with intentional misrepresentation of data or theft of another’
definition now encourages), use of the phrase in circumstances of a wrong not involving deceptionmay inappropriately harm the professional reputation of the researcher. Even triggering the processmay have disproportionately damaging effects.
Finally, an institution may wish to assess the scope and adequacy of its other grievance and
disciplinary procedures in determining how broad a reach to assign its research misconduct process. Other processes may need to be broadened to avoid overutilizing the research misconduct process.
Many of these determinations will be institution-specific. In my view, however, fairness and
accuracy suggest that “academic fraud” and (because of the ORI nomenclature) “researchmisconduct” be reserved for FFP (including FFP outside the research context), and another title beused for improper behaviors that are neither FFP nor regulatory violations. (To fill the gap, theUniversity of Pittsburgh has resorted to the phase “research misbehavior”38 and Duke’s publishedpolicy nicely uses the phrase “research practices that are inappropriate but which do not generallyrepresent misconduct.”39) Breaches of the regulatory types might best be treated independently,within their own framework, and like other violations of university policy, with investigations,grievance proceedings, or other disciplinary process, because the regulatory scheme is tooindependent to integrate well into the FFP process, and sanctions, standard of proof, and optimalprocess, are likely to be different. Jurisdictional issues: cross-institutional research and allegations. In a time when
some star researchers seem to change their home institutions as often as free agents leave theirteams in baseball, it is not unusual for an institution to find itself addressing a research misconductallegation that involves another university. This can also occur when research collaboration crossesuniversity boundaries.
Research misconduct policies may call for an institution to address allegations received
regarding individuals in its current employ, which is the most common case, and which makes theimposition of sanctions most manageable. But perhaps they should allow for a broader range ofcoverage. For example, if the respondent is now at Blue U., and the allegation is received at PurpleU., where all the research was done, the data reside, and all the other collaborators are located, doesit really make sense for the inquiry and investigation to be done remotely, at Blue U.? Shouldn’
Purple U. consider proceeding, itself? ORI has intimated in a passing statement to medical journaleditors that the proper home for adjudication of such issues is “where the research wasconducted.”40 There is no apparent reason in law that a university cannot conduct an inquiry andinvestigation into an allegation regarding research conducted on its campus, even though the allegedperpetrator is no longer employed by it, even acknowledging that the sanctions phase may be tricky.
38 Holden, C., “Schatten: Pitt Panel finds ‘Misbehavior,’ not Misconduct,”Science 311.5763.928 (Feb. 17, 2006),compare University of Pittsburgh Policy 11-01-01, ResearchIntegrity, and University of Pittsburgh Faculty Handbook,“Research Integrity” (“Other unethical or sloppy research practices are often referred to as "research misbehavior."),available at 39 Duke University Policy and Procedures Governing Misconduct in Research (Nov. 1995),40 ORI, “Managing Allegations of Scientific Misconduct: A Guidance Document for Editors,” (January 2000)(“[E]ditors are not responsible for conducting a full investigation or deciding whether scientific misconduct occurred. Those responsibilities rest with the institution where the work was conducted or with the funding agency.”)
National Association Of College and University Attorneys
This may be advisable in some cases for a number of reasons. First, the institution’
researchers may have knowledge of the matter, and the relevant data may be on its campus – easierto sequester, evaluate, and control. Second, although it may be adept at managing the problem, theemploying institution generally appears to have a structural conflict of interest in the managementof such disputes, since an adverse finding decreases its reputation and the value of its faculty: asRichard Smith, the Chief Executive of the British Medical Journal, observed about handlingallegations,
Employers face a clear conflict of interest with accusations. To expose one of youremployees as fraudulent is unpleasant and does harm to the brand of the institution. Oftenthe miscreant will be a friend, a respected colleague. The temptation is to bury the wholething, and Stephen Lock [editor of a history of fraud in research] has produced evidence thatthis did happen commonly. Many, even most, senior academics know of cases ofmisconduct, and yet they report that rarely were those cases fully investigated, the miscreantpunished, and the scientific record corrected. The miscreant may simply have beenencouraged to resign or take early retirement.41 (Footnote omitted.)
This issue arises upon receipt of an allegation, and continues through every decision to report or notreport, through the wording of any retraction, and at the findings and punishment stage. Thatconflict may incline an institution to hunker down, defend vigorously against an allegation or sweepit under the carpet, and blame the remote complainant or excuse misconduct. On the other hand, thecharging institution may have too strong an interest in protecting the reputation of the co-authors orof the complainant, if it is that individual’s home institution. The diplomatic challenge is there –each institution may be predisposed to credit the testimony and views of their own (one, thecomplainant; the other, the respondent). Each institution has to examine and re-examine itsjudgments to ensure that it is doing its utmost to treat allegations and proceedings impartially, andwith the highest allegiance to the preservation of the integrity of the academic mission.
Even when wishing to work together from an early stage, one might ask whether it makes
sense to transmit an allegation of fraud to the respondent’
sense whether it is at all meritorious. Doesn’
t that act become shadowed by potential defamation
liability, especially if there is no legal duty (or policy duty) to notify? If there is no related federalfunding, and if the other institution’s policies do not envision such cooperative effort, we mayquickly run into legal roadblocks when making such a communication.
Note that there does not appear currently to be any routine state or federal law privilege for
institutions reporting the fact of their proceedings or their findings to another educationalinstitution, and there is no immunity for individuals who either make one of these types of reportsor make the findings. In particular, while whistleblowers are protected under federal regulations,this provides no cover for disclosure of alleged wrongdoing in non-federally funded research. Toooften, and even when federal funds are involved, the approach of the respondent’
be, “Aw, it was just a sloppy mistake – why are you persecuting the guy?” – and a decision that aninvestigation isn’
t really warranted. Institutional policies could reasonably broaden the scope of the
reporting that the institution is required to make upon a finding of academic fraud. The policiescould require that an institution put in place other measures to ensure that researchers with whom
41 Smith, R., “Investigating the previous studies of a fraudulent author,”BMJ 331: 288-291 (Dec. 20, 2005).
National Association Of College and University Attorneys
their own are collaborating have agreed to permit notification to their own employers of the resultsof proceedings.
It is surprising to learn that ORI does not currently require notice to it of research
misconduct findings in non-federally funded research, nor do PHS funding applications (as of lastexamination) require an applicant to divulge that he or she has been found guilty of FFP in non-federally funded research. This is a deficiency that ORI might cure – and an absence of cure couldbe interpreted as signaling that it assigns a lesser level of gravity to FFP in industrially-sponsoredresearch.
In light of the amounts of cross-institutional research being done, and the mobility of
university researchers, institution policies should be shaped to require notification of other involveduniversities, and, possibly, to envision cooperative inquiries and investigations, that might helpensure appropriate neutrality and rigor in pursuing evaluations of such allegations. ORI proceduresempower universities to use a “consortium” approach in inquiries and investigations.42 Also, withinternational research increasing, institutions should consider establishing in their policy a duty tonotify international professional societies of their findings.
Institutional policy should be updated, then, to anticipate cross-institutional research and to
retain the discretion to proceed (if appropriate, depending on the quality of the other institution’sprocedures, for example) based on where the research was conducted, rather than solely where theindividual is employed when an allegation is received. Make sure your policy includes the option ofgiving notice of proceedings and findings to international professional societies. ORI and otherfederal agencies should reinforce the importance of institutional proceedings adjudicating FFP (butnot research misconduct other than FFP) by requiring disclosure in connection with applications forfederal research funding. Getting started: The earliest stages, from a policy perspective. If a study
“involves” federal funding (that is, if the alleged wrong is allegedly perpetrated by an individualwho received federal salary support or who utilized federally funded resources), it is clear that ORIor the funding agency must receive notice of any investigation and of the outcome, and that federalprocedures must be followed. The agency will exercise oversight authority and conduct furtherprocesses if a federal sanction is to be imposed.43
Once an allegation not involving federal funding has been received, however, it may not be
wise to begin an inquiry without any conversation at all with the expected respondent (depending,at the same time, on what judgment is made about the need for immediate sequestration of theresearch record). This is because simply being the subject of a research misconduct inquiry can beseen as damning, if it becomes public or even rumored. Fairness may suggest notice and a briefopportunity to respond, if possible, before launching a formal inquiry. Moreover, once a formalinquiry is launched, procedural constraints limit the institution’s flexibility in reaching a resolutionor pursuing mediation. On the other hand, early notice to a (guilty) respondent may provide him orher an opportunity to destroy evidence (computer files, samples, etc.), so the first steps need to becarefully considered.
4242 CFR § 93.306. 43 42 CFR § 93.309.
National Association Of College and University Attorneys
That availability of this kind of assessment and mediation, however, might also help
decrease the possibility that research misconduct allegations and proceedings are themselves beingused as a weapon, rather than a good faith protection of science. Especially when the allegation isbetween former collaborators and the allegation is one of plagiarism, formal research misconductprocedures are subject to abuse, and care must be exercised.
Co-authors must be advised, particularly if an allegation comes from outside and it is not
clear from the challenged publication who was responsible for what work. One might start with thesenior author, but in this it can protect all concerned if all co-authors are advised simultaneously andthe data files are sequestered. Sequestration can be overly dramatic and can needlessly hamperongoing work, but it may be necessary to protect both respondent and complainant. Then (asdiscussed below) the question may be whether the data is still available and accessible at all. Confidentiality concerns and public communications. Research misconduct
policies typically and correctly provide that proceedings should be confidential, from allegationthrough investigation. It is important to protect the reputation of the respondent, and the mere factof a charge against an individual can be very damaging. Confidentiality of proceedings can beimportant to the reputation of related co-authors as well, and to the complainant, to avoid ostracismor retaliation. In addition, confidentiality can help ensure that the proceedings and the inquiry andinvestigation committees are not subject to community pressure in favor of or against therespondent.
Confidentiality can easily be compromised, however. An outside complainant is under no
obligation not to repeat the allegation before proceedings are complete, though the threat of alawsuit might dissuade him. It is difficult to enforce a promise of witnesses who are interviewed tokeep the fact of or particulars about the proceedings in confidence. A retraction or expression ofconcern may need to be published before the (often lengthy) proceedings, including an appeal, arefinished. Any retraction prompts questions and rumors, and the communications with the journal –which do not appear to be privileged under any federal or state law – will generate rumors andpressure to divulge the fact and outcome of related proceedings. Co-authors may feel a need todefend their own reputations by implying that if there was a problem, it was not their doing. Othercircumstances may also dictate release of related information – for example, the possibility of arelated public health hazard. ORI has observed that:
[S]ome institutions have found it necessary to release information about an investigation toanother institution if, for example, the accused researcher requests a letter ofrecommendation. In ORI’
s view, the confidentiality provision relates only to the regulatory
investigation and reporting requirements mandated by the Public
Health Service Act and Federal regulations. It does not preclude an institution fromdisclosing information regarding actions that it may have taken pursuant to its internalprocedures as long as the Federal issues are not disclosed.44
Investigation and inquiry committees alike should be advised at the outset, in connection
with their appointments, that they are bound not to discuss their proceedings with any personoutside their own number. Investigation committees should not discuss the matter with inquiry
44 Gibbons, G., “Research Misconduct,”NACUA Presentation (Nov. 2, 2000 )(citing ORI Newsletter, Vol. 7, No. 2 at 2,March 1999).
National Association Of College and University Attorneys
committees. They should all be advised to begin their interviews by cautioning witnesses that theproceedings are confidential. Witnesses could be provided a notice in writing that they are expectedto maintain confidentiality. It is difficult to do more than try hard to keep confidentiality, however. It would be a challenge to impose serious discipline on a faculty member co-author who spokeabout his or her interview before an inquiry committee. Policies should stop well short of promisingconfidentiality for these reasons.
On the flip side, at the close of an investigation (and not an inquiry), we expect results to be
announced if there is a finding of guilt. If there is no such finding – if the respondent is exonerated,a decision must be made – perhaps with the input of the respondent – about whether anyannouncement would be more damaging or helpful to the respondent. ORI regulations do notexplicitly contemplate the arrival of that moment, however: Section 93.108, “Confidentiality,”seems to contemplate entirely confidential proceedings up to the point of an administrative hearing. Section 93.300(e) provides that institutions must “Provide confidentiality to the extent required by93.108 to all respondents, complainants, and research subjects … ” Section 93.318(g) provides that auniversity must notify ORI immediately at any time if “it has reason to believe that … [t]he researchcommunity or public should be informed.” – which in the context of the rest of the section implies asense of urgency that would not ordinarily be expected to exist at the close of a proceeding. Oncethe university has made a finding of research misconduct, there should be an understanding that thefinding may be made public.
In addition, while ORI procedures require notification to the federal government when an
investigation of misconduct “involving” federal funding is complete, obligations of notificationregarding privately funded research are often more ambiguous. Should sponsors that fundcollaborators of the respondent be notified of a retraction because of research misconduct? Shouldthey receive the whole report of the investigation committee, or simply notification of retraction?What should any related journal, or professional association, or future employer, be told?
In cases that have resulted in criminal prosecution or other involvement of the Department
of Justice, a press release may be issued. Universities do not typically announce their findingspublicly (and the policies I have examined do not provide for such announcements) except inconnection with well-publicized cases. But failing to announce a resolution can be as harmful ascontinuing confidentiality. The reputations of co-authors and collaborators, who may be entirelyinnocent or at most guilty of failing to review with adequate care the work of their collaborator,may be seriously harmed by association with the perpetrator.
Policies and procedures should anticipate the possibility that the university may make public
announcement of a resolution, advising a journal and professional societies, and funding agencies,when an investigation has made a finding of FFP, unless those proceedings are then reviewed orassessed anew by ORI (or another sponsoring agency). They should establish an obligation on theproceeding institution to notify other institutions that may employ the individual. The universitymight consider notification of journals that have previously published work of the individual. It is atruism in the literature of research misconduct that all of the work of an individual who is found tohave committed research fraud must be considered to be suspect until re-examined. Devastating asthis presumption is to the reputation of the individual, it appears that in most cases, instances of FFPare not isolated instances, but are typically pathological. Therefore, draconian as it may seem, aninstitution devoted to correcting the scientific record should seriously consider notifying otherjournals and previous employers.
National Association Of College and University Attorneys
Part III Practical considerations during an inquiry or investigation Who is the client? We all know that we “represent the institution, not any
s a lesson we learn in law school and in early days on the job in a university
s office. That concept gets played out in interesting ways in the context of research
misconduct proceedings, however, where the “institution” is embodied in the policy, the researchintegrity officer, the inquiry and investigation committees, possibly the complainant, possibly therespondent, any collaborators, the integrity of the research record, and the officers of the university.
When a possible complainant comes to you for legal advice, you do not act as his or her
personal counsel: you do your job best by asking questions, informing of policies and proceduraloptions, and encouraging negotiated settlement of differences where possible, while advising thecomplainant of an ethical obligation to pursue blatant fabrication, falsification, or plagiarism,especially where federal funding is involved. The National Academy of Sciences has advised,“Someone who has witnessed misconduct has an unmistakable obligation to act.”45 As onethoughtful commentator has observed, in response to the question “Should one report misconduct?”,“Professional ethics would require a resounding ‘yes’
to this seemingly rhetorical question. The
reality is more complex.”46 But counsel’s responsibility is to the institution, and therefore toenabling a sound process of addressing allegations, helping ensure that federal and other dollars arenot misused through academic fraud, and that the scientific or other academic process of researchand publication is protected.
A respondent in an investigation also should have some counsel or at least access to an
adviser. Some universities provide counsel to faculty as a matter of course, but many do not. Theinstitution’s indemnity policy may allow advancing of defense costs. This is an area in which thereis a lot of room for different approaches. 47
My practice is to counsel faculty complainants with concerns about work done (for example,
by collaborators) outside the university about how to interpret the practices and procedureselsewhere, and to review their complaints for presentation issues, and clarity, in order to facilitatethe consideration and adjudication of their complaint. It is useful, however, to avoid prejudging themerits of their complaint, and to keep some distance from the emotions such proceedings can evoke.
When working with the inquiry and investigation committees, it is important not to
substitute your judgment for theirs, but to assist them in identifying questions that they need to askand answer and to help them understand the procedures. In the investigation phase in particular,faculty may need assistance in preparing for on-the-record interviews, formulating questionsappropriately, using some legal terms, and thinking about burden of proof and standards of proof. Reviewing draft reports for internal consistency, gaps, and cite checking and ensuring that their
45 On Being a Scientist: Responsible Conduct in Research, (2d ed.)(NAS 1995) at 18. 46 Decoo, Wilfried, Crisis on Campus: Confronting Academic Misconduct (MIT Press 2002) at 151. 47 See Mello, M. and Brennan, T., “Due Process in Investigations of Research Misconduct Cases,”New Eng. J. Med. 349:1280-86 (Sept. 25, 2003). Where the complainant and the respondent are both faculty at one institution, special careis plainly required to ensure procedural evenhandedness.
National Association Of College and University Attorneys
characterizations are meticulously supported by the record is important. Particularly when dealingwith scientists, for whom this kind of engagement is foreign, effective assistance from counsel iscrucial. The job is to help the committee form and articulate its own views, based on its expertise,and ensure the adequacy of its supporting evidence. Jurisdictional issues: federally-funded research. An initial question upon receipt
of any allegation of fabrication, falsification, or plagiarism is, “Which external procedures (if any)bind the institution and to whom will the university need to report any investigation or findings ofmisconduct?” The answer will depend in part on how the research was funded. This determinationmay be less obvious, and may require a more detailed analysis, than one might expect.
Although ORI has become the federal government’s public face on research misconduct,
s jurisdiction extends only to research that involves funding by the Public Health Services.48
The National Institutes of Health are the most obvious of other funding sources within theDepartment of Health and Human Services, but note that not all of DHHS-sponsored research are inthe Public Health Services.49 The National Science Foundation’s separate policies and proceduresnow are more in line with ORI’s, thanks to the OSTP directive, but they still differ in some respects:for example, at NIH, ORI is the go-to office; at NSF, it is NSF’s Office of Inspector General.
In addition, it appears that even if PHS money funded part of the work on an article that is
published and that is retracted because of FFP, ORI does not assume jurisdiction for reporting andother purposes unless the precise work that was fabricated (and thus the author of that work) wasfunded by PHS funds. The federal support that creates ORI jurisdiction may be direct (e.g., salarysupport) or indirect (e.g., use of federally-funded core facility), but it must support the particulardisputed work. In multi-author studies, this can be a tricky determination.50 One might expect thatthe government would anticipate questions on retractions for work on which it is cited as a support,but I am advised that the current position is limited to the circumstances set forth above.
Further, authors may treat statements regarding research funding more casually in the
context of publication than they do in the context of a research misconduct allegation. For example,they may on occasion be tempted to over-acknowledge an NIH grant to help invigorate an otherwiseuninspiring progress report. Therefore, a careful review of the funding sources for each author on adisputed work, with direct input from the office responsible for financial services at your institution,should be undertaken early and a determination supported by appropriate documentation. If thework is sponsored research under an agreement with a private funding source, that contract must beexamined (as well as your institution’
s policy) to ascertain any legal reporting obligations. The
nature of some relationships between the institution and funding organizations may encourage earlynotification about the problem being investigated, especially if a retraction is anticipated. Butconfidentiality concerns (and defamation exposure) require caution and discretion in determiningwhom to notify of what, and when. Working with the Investigation Committee, on the record. Institutions may have
more experience dealing with the receipt of an allegation and an inquiry than conducting an
48 See, e.g., “Investigating allegations of scientific misconduct,”BMJ 2005:331:245-46 (July 30, 2005). 49 Although 42 CFR § 93.220 lists agencies to be treated at PHS agencies, there have reportedly been differences ofopinion in the past about the propriety of treating SAMSHA and certain other non-NIH agencies as such. See 42 CFR §93.220, 93.221. 50 Private conversation with A. Price, Director, Division of Investigations, ORI, (Spring 2005).
National Association Of College and University Attorneys
investigation itself. The investigation is obviously a critical step and can be tricky for many reasons. Here are some considerations and thoughts:
They are not lawyers. This is of course both good and bad. The investigation
committee is typically a group of substance matter experts (about science, history, or whatever isappropriate to the allegation) who are asked to perform an essentially adjudicatory function. Afterthe inquiry stage, they are often acting as prosecutor as well as fact finder: it is they who ask thequestions of the respondent and the witnesses, and then decide what the facts are. They need to doboth the direct and the cross examination, functionally, to make sure the strengths and weaknessesof both positions are exposed and tested.
They are likely to be inexperienced in the niceties of legal proceedings, and the proceeding
they are conducting should not be over-legalized. And yet, because the proceedings may be career-ending for a respondent, they must be very carefully run and scrupulously fair. Committee memberswill have a sophisticated sense about analysis and fact-gathering, but they may not be practiced informulating questions that are appropriate, or in articulating the basis for their conclusions. Theymay need assistance in ensuring that they create a clear record of their questions and answers,particularly where data and figures are being referred to. They need to decide on appropriateevidentiary rules. Counsel need to assist them in these tasks, and coach them, without participatingas a decision maker and always trying to support, not direct. To enable them to do their jobappropriately, faculty committee members often want a short course in lawyering. Their jobtypically entails conducting deposition-like interviews on the record, which requires carefulpreparation and planning, and orderly assembly of documents (with copies) that they will use intheir questioning. They may need to create demonstrative displays, laying out figures and data, inorder to keep the record clear. Answering “Are we the Inquiry or the Investigation Committee?” It may be
difficult to distinguish what the inquiry committee has done from what the investigation committeeis being asked to do. Faculty conducting an inquiry are often loath to stop short of a full review ofdocuments, materials, and interviews with individuals. This may be so both because of their nature(as scientists, for instance, needing a conclusion), an unwillingness to accept their goal as a“probable cause” determination, and a legitimate recognition that their determination of whether aninvestigation is warranted triggers disclosure to ORI and funding sources, and serious, perhapsirrevocable reputational injury. No wonder they are unwilling to make a determination if theybelieve there is a significant chance that their judgment might be incorrect. In plagiarism cases inparticular, it often is the case that there is little if anything that the investigation committee wouldsee that the inquiry committee has not. In fact, the inquiry stage may take about 15 minutes: asimple side-by-side comparison and a brief interview with the accused may be all that are necessaryto know that an investigation is required. Dealing with outside counsel; providing outside counsel to a respondent .
Dealing with respondent’s counsel or advisors can be challenging. Policies should make it clearwhat level of participation is permissible. Institutions may find that they wish to provide counsel toa respondent faculty member to help ensure a full and fair adjudication in a complicated or highly-charged matter. Whether to allow cross examination or not, in person confrontation of witnesses,and other due process informed procedures, will vary from institution to the other, but mustcarefully be considered in conducting an investigation. In ORI-governed proceedings, there is nomandate. Some institutions may also wish their Research Integrity Officer to deal directly with the
National Association Of College and University Attorneys
respondent, and not transform the proceedings into a lawyer-to-lawyer event. Some institutions willprovide counsel to the respondent whom they employ. This may be wise in some circumstances. Ifthere is no ORI involvement and a subsequent review is unlikely, the matter is sufficientlycomplicated, and serious disciplinary consequences may result, it may be most efficient for theinstitution to provide counsel at the investigation phase to ensure fair proceedings and to preserveany possibility of imposing discipline without redundant process. Consider how many layers of fact-finding will there be. It is important to keep in
mind that when the allegations that are being adjudicated concern non-PHS funded research, theinvestigation proceedings amount to the final trial on the merits: any appeal that is providedinternally at an institution is unlikely to be at the level of the oversight and review provided by ORIand the Departmental Appeals Board, which seems in effect to revisit all factual and legaldeterminations. In PHS-funded proceedings, on the other hand, where those appeals are provided,the institution’
s investigation may be more like a grand jury proceeding, and less determinative. Yet
the same procedural guidance typically applies.51 In either case, the Investigation Committee needsto proceed as the ultimate fact-finder. What, precisely, are the allegations under review? The Investigation Committee
needs to be clear on its charge and on the allegation or allegations that its members will address.52This can be more challenging than it may seem at first glance. Allegations may have been receivedfrom several sources. When fabrication and falsification are involved, it may be difficult tocrystallize the charge. And, frequently, during the course of the review of a meritorious claim,additional instances of misconduct will be brought to the committee’
need advice on which to address and which to set aside for review at a later time. In cases in whichintentionality is an issue, they may wish to search for additional instances of what a respondentmaintains is “negligent but honest error” to determine whether there is a pattern of misconduct thatrefutes the defense – or additional instances may be brought to the institution’s attention. Typically,these matters are so time consuming and challenging to manage that an institution may decide tostop with an initial investigation, and refer its finding (after all appeals have been exhausted) to thejournals that have published other possibly tainted material, or the institutions where the relevantresearch was conducted. The duty to look for more: the pathology of research misconduct. In every FFP
case, it will be necessary to consider whether the institution conducting the investigation has anobligation to pursue all additional instances of possible research misconduct that have beenidentified during the course of the proceeding, and even to look for additional instances, before (orafter) making a finding of misconduct. In some circumstances, this may be impracticable toaccomplish before a finding is issued, and it may be preferable for an investigation on one or twopublications to be completed, the results made public, and then a determination made by theinstitution that housed the investigation, or the institution that employs the individual, how best topursue or review further publications. In the Poehlman case, for example, since Dr. Poehlman hadbeen at University of Vermont during the bulk of the publication years in question, that university
51 At the same time, one ORI representative has written, “As this is an investigation, not a trial, other due processconsiderations such as provision of counsel and cross-examination of witnesses are left to the individual institution’sprocedures.” Gibbons, G., “Research Misconduct,”supra at 7. But in cases where there is no ORI based proceeding, theinvestigation may have the lasting significance of fact finding done at trial. 52 Bombardieri, M., “MIT review clears school in delaying fraud probe,” Boston Globe, (May 20, 2006)(“The absenceof a clear, concise and consistent definition of the allegation complicated the conduct of the inquiry, the panel said.”)
National Association Of College and University Attorneys
(in concert with ORI) reviewed many but not all publications before the findings were issued. It hasnow developed an approach that, working with another university where Dr. Poehlman was brieflyemployed, and with Dr. Poehlman’
s collaborators, should eventually result in the review of all of
his prior publications.53 Two commentators have noted about this task, “Although meeting thisstandard will involve thousands of hours of investigation by faculty working with those mostimmediately responsible and affected, the coauthors, we think that institutions will rise to theoccasion because their reputations and the integrity of the scientific record are at stake.”54 In .short,it is an enormous task.
In cases where the individual has frequently changed institutions, or has resigned and is not
affiliated with an institution, the onus may be less easy to place. Journals that published the authormay not be willing or able to devote the time and expense;55 professional societies may also findthemselves without adequate resources to conduct an investigation as well;56 private fundinginstitutions are unlikely to have the resources or expertise, although they are often held up asresponsible; and absent federal funding, there may simply be no willing and able party. One hopesthat an employer or former employer will rise to the task, and institutions’
permit that undertaking. The obligation cannot be overlooked, but the burden should not beunderestimated. Is the standard really just “more likely than not”? The committee should be
briefed on the standard of proof it is to apply, which (under ORI provisions) is a preponderance ofthe evidence, but may be a higher standard for non–federally funded work. Many committees maytry to apply a higher standard, again recognizing the gravity of their assignment, but in the case offederally funded research, the committee can be assured that their work is subject to appeal and thatthey are required to determine only whether it is “more likely than not” that the offense wascommitted. What does the Investigation Committee report need to do? Its draft report should
be carefully reviewed by counsel to ensure that the report reflects the committee’s compliance withUniversity procedure and standards, and that the report is a stand-alone document that is self-explanatory. Factual assertions should be carefully referenced and cite checked against the record,to make sure that the committee has avoided any overbroad characterization or casualgeneralization. When the record is complex, the investigation and report may need to be carefullydivided into segments. A detailed record of all exhibits or documents referred to should be kept andcopied. Although procedures may call for interviews to be summarized, the transcripts seem betterthan any characterization, and particular should be addressed in the committee’s analysis.
The task of reaching a factual conclusion about the acts done and the accompanying intent is
hard to accomplish carefully and well. Although the particulars of proceedings may seem welldefined and the crime clear, closer examination usually reveals many layers of nuance and
53 Sox, H. and Rennie, D., “Research Misconduct, Retraction, and Cleansing the Medical Literature: Lessons from thePoehlman Case, Ann. Intern. Med. (2006) 144:E-7-E-11. 54 Sox, H. and Rennie, D., “Research Misconduct, Retraction, and Cleansing the Medical Literature: lessons from thePoehlman Case, supra, at E-8. 55 “Investigating allegations of scientific misconduct: Journals can do only so much; institutions need to be willing toinvestigate,”BMJ 2005:331:245-6. 56 The American Historical Association, for example, has decided that it will no longer investigate allegations ofscholarly misconduct. Glenn, D., “Judge or Judge Not?”, Chronicle of Higher Ed., (Dec. 17, 2004).
National Association Of College and University Attorneys
ambiguity in developing an accurate factual narrative. Repeated experience in this field reveals thatindividuals who to all appearances have devoted their lives to a search for truth, may sometimeshave an astonishing capacity for self-delusion. These are the respondents. But we all reachconclusions by choosing and connecting dots of factual observation. Some people seem to feel freerto add dots to the picture, ignore some dots that are unquestionably there, or rely totally on dotsdeveloped by others, in what one early commentator has called the “pathological science.”57 Inreaching its conclusions, your investigation committee is engaged in the same kind of enterprise asthe respondent was when he did the work being scrutinized. Work hard to make sure that thecommittee connects the dots in the clearest and most indisputable fashion. How can university counsel get help in advising the Committee? Investigations
are often unique in many respects, and although their frequency may be increasing, it is difficult toaccrue enough experience to address many issues that may arise. Counsel would do well to readbroadly in the field and carefully examine some published investigation reports – a review willhighlight needs for the committee and enable counsel to anticipate issues. (Some useful samples arecited in the references at the end of this paper.) Counsel may wish even to share published reportswith committee members, early on, so that they are aware of the level of formality they should aimfor in their fact-finding and have a model to follow. Consultations with litigation counsel may alsohelp guide university counsel in their role as adviser to the committee in decisions about the level ofprocess to provide and other formal aspects of the investigation, and on interpreting the universitypolicies and procedures in a highly adversarial context. On the administrative front, if possible, it isvery helpful to have a paralegal and secretarial assistance available to support the investigationcommittee and to help schedule and manage the interviews and the record, and to ensure a polishedproduct, confidentially produced.
In addition, the Office of Research Integrity is open to assist by sharing the fruits of its
experience on managing investigations, but its availability may be restricted to advice on handlinginvestigations concerning federally-funded research. Note, finally, that ORI procedures establishtimelines for investigation, but the office has acknowledged in other contexts that requests forextensions are to be expected in complicated data cases. Forensics. Managing the evidence with trial-like rigor in a university research
misconduct proceeding is a challenge. Note that although we may not expect faculty committees tomeet criminal law standards, counsel to a respondent in a serious matter will, and, if a prosecutionresults, the federal government may also wish to hold the university to a higher standard than theordinary grievance proceeding would provide. So, use care in copying computer files, sequesteringdocuments and electronic material; and document the transfer of files.
Interpretation of data and images in sophisticated cases can also be challenging, no matter
how expert the committee. Be aware that ORI makes available tools for examining data, and forfederally funded research will make available expert advice on statistical analysis and for otherspecialized needs in assessing an allegation of misconduct. Journals are also implementing somesafeguards and making tools available,58 as are other components of the scientific community. Seekhelp early.
57 Langmuir, I., “Pathological Science,” (Dec. 18, 1953), available at58 Wade, N., “It May Look Authentic; Here’s How to Tell It Isn’t,”N.Y. Times (Jan. 24, 2006); “ORI Forensic Tools forQuick Examination of Scientific Images, (last updated on April 19, 2006).
National Association Of College and University Attorneys
Dealing with the complainant. It is sometimes easy to forget the complainant in the
course of the intricacies of an investigation, particularly when the complainant does not havedetailed knowledge of anything but the result and was not, for example, the subject of plagiarism. Much like victims in criminal cases, however, the complainant needs to have some closure and insome cases it will make sense for there to be periodic communications and assurances that theprocess is working as intended. As detailed in Crisis on Campus: Confronting AcademicMisconduct by Wilfried Decoo, who herself was twice a complainant (of the temperate sort), it canbe very difficult to be a complainant even in the most apparently open and shut cases.59
In addition, as observed above, despite the use of the term “whistleblower,” complainants do
not have a monopoly on the truth. Sometimes they may have inadequate data on which to reach ajudgment, or have ulterior motives in bringing a complaint – motives that may not be apparent onfirst review.
Whatever the motive, it is important to communicate to the complainant that a process for
reviewing the complaint exists, is being followed diligently, and will be brought to a close asexpeditiously as possible, and within allotted time frames as much as possible. Complainants aresometimes the wronged party and if they do not feel satisfied by the proceedings – especially if theproceedings exonerate the respondent – complainants may also bring suit for the underlyingoffense, be it plagiarism recast as copyright infringement or theft of ideas recast as palming off.60
The case of the missing data. The absence of supporting original data for
questioned figures presented in a text seems to be all too common. Sloppy recordkeeping practices,changes in technology that make old data inaccessible, confusion connected with moves from oneinstitution to another, and unclear policies and practices for departing researchers and collaborators,all may cause supporting to be unavailable for examination. This creates a serious conundrum foran investigation committee. For example, the Schön committee at Bell Labs commented in its finalreport:
Proper laboratory records were not systematically maintained by Hendrik Schön in thecourse of the work in question. In addition, virtually all primary (raw) electronic data fileswere deleted by Hendrik Schön reportedly because the old computer available to him lackedsufficient memory. No working devices with which one might confirm claimed results arepresently available, having been damaged in measurement, damaged in transit or simplydiscarded. Finally, key processing equipment no longer produces the unparalleled resultsthat enabled many of the key experiments. Hence it is not possible to confirm or refutedirectly the validity of the claims in the work in question.61
ORI procedures now provide that “destruction, absence of, or respondent’
research records adequately documenting the questioned research is evidence of researchmisconduct,” but only in certain circumstances that are onerous to establish. ORI requires that, totreat the absence of records as evidence of misconduct, the institution must:
59 Decoo, Wilfried, Crisis on Campus: Confronting Academic Misconduct (MIT Press 2002). 60 My colleague Robert Bienstock has astutely observed that particular confidentiality problems in updating thecomplainant will also arise and need various treatment, depending upon whether the complainant is a fact witness, aninjured party, a disinterested bystander, or a supervisor who discovered the wrongdoing, for example. 61 Schön Report at 3, available at
National Association Of College and University Attorneys
[E]stablish by a preponderance of the evidence that the respondent [1] intentionally,knowingly, or recklessly had research records and destroyed them, [2] had the opportunity tomaintain the records but did not do so, or maintained the records and failed to produce themin a timely manner and [3] that the respondent’
s conduct constitutes a significant departure
from accepted practices of the relevant research community. [Numbers added.]62
The draft regulations, on the other hand, had created a rebuttable presumption of misconduct in theabsence of records, but this approach was ultimately rejected by the agency.63 In my view, the finalversion makes it too easy for a fabricator to destroy or lose access to evidence and escape theconsequences, especially when the scholarly expectation in many fields is that the author of apublication is generally considered to have a duty to retain the supporting data and make it availableto qualified inquiring scholars in the field, but ORI’
s approach prevails for now, at least for PHS-
University policies for responsible conduct of research should stress the importance of
retaining the complete research record and keeping it accessible (from a technology point of view aswell as from a physical existence point of view) for a set period of time after publication. Researchers should be reminded regularly about the importance of reviewing their records, andallocating responsibility among themselves multi-author studies who is responsible for preservingwhat data. Institutions should clearly articulate a protocol for handling the original data of departingfaculty (and departing post docs, grad students, and undergrads). The accurate assessment ofsubsequent research misconduct allegations (as well as intellectual property rights) and any relatedlitigation or further administrative proceedings may depend on the rigor and care with which theresearch record has been maintained. The journal and the comment, erratum, or retraction. It is often presumed that a
full retraction will be the consequence when published research is connected with a finding ofresearch misconduct, especially one involving fabricated or falsified data results or plagiarizedmaterials. On closer scrutiny, however, the correction of a flawed research record or publicationmay have numerous possible outcomes, of which wholesale retraction is only one.64 A fabricatedfigure could be withdrawn and replaced by a corrected figure, and a “corrigendum” or “correction”published. A small copied passage could be struck from the text and a paper re-issued. Anexperiment could be repeated and the results verified, and the article allowed to stand. The journalcould publish a comment, an “expression of concern,” an erratum or corrigendum, or a retraction inwhole or in part, and all of this may be subject to multilateral negotiations.
Even when complete retraction is the outcome, it can be challenging (especially in this
global, electronic age) to accomplish a thorough cleansing of the record: cached or downloadedelectronic versions, or hard copies of the original, may remain available to mislead the unwary.65
62 42 CFR § 93.106(b)(1). 63 70 Fed. Reg. 28370, 28371 (May 17, 2005). 64 See generally, Budd, J. et al., “Phenomena of Retraction: Reasons for Retraction and Citations to the Publications,”JAMA (1998) 280:296-97; Atlas, M., “Retraction Policies of high-impact biomedical journals,”J. Med. Libr. Assoc. 92(2) (Apr. 2004); National Library of Medicine, “Errata, retraction, duplicate publication, comment, update, andpatient summary policy for MEDLINE,ated Jan. 21, 2005). 65 See Sox, H., and Rennie, D., “Research Misconduct, Retraction, and Cleansing the Medical Literature: Lessons fromthe Poehlman Case,”Ann. Int. Med. 144 (Apr. 18, 2006).
National Association Of College and University Attorneys
Note that even in a “clean” case, in which the entire tainted publication is stricken from the
record, there is still the question of exactly what should be said about the reasons for the retraction,and who is entitled to influence that decision and wording. Again, there is a surprisingly wide arrayof formulations. Co-authors typically are required to agree on the phrasing of the retraction notice;but an errant co-author may be difficult to sign up. An investigation committee may have a differentview from that of the authors and yet it cannot dictate that all authors – especially if the authors arenot all at the committee’
s institution – take any particular action.
Within a journal, the decision to comment, correct, or retract may be committed to the editor
or editorial board. The journal may have its own policy, which it will have advised publishingauthors of at the time of submission. Short of that, medical journals at least may be guided by thestandard enunciated by the International Committee of Medical Journal Editors,66 or perhaps byother professional association standards,67 which call for publication of the reason for the retraction. But editors have little leverage for demanding a particular phrasing. A journal also may be put in adifficult position if it has any reason to question the objectivity of the proceedings of aninstitution.68
In a further interesting twist, an institution looking for complete retraction may be in an
awkward position if the objectivity of the editors of the journal in addressing a request for retractioncan be questioned. In small fields, and with journals that are run from an institutional base ratherthan by an independent professional structure, conflicts of a personal and reputational sort may alsoinfluence in surprising ways how a journal responds to questions presented about a manuscript thatit has published – how willing it is to take steps to make retractions widely known, at what stageand on what basis it will be willing to print a retraction or other comment, and so on. For these andother reasons, journals may have their own agendas.
Whether the retraction refers to the name of the individual found to have committed fraud
will need to be addressed, and may have different outcomes depending upon the stage of theproceedings at the time of the retraction. The institution may also need to decide whether to makethe investigation report available to the journal. The university policy should address these issues,creating options for the university where possible but allowing flexibility to shape the decision tothe circumstances.
66 ICMJE, “Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing forBiomedical Publication” (updated February 2006), available (visited 4/8/2006)67 American Historical Association. See generally Hoffer, P. C., Past Imperfect (Public Affairs 2004). 68 ORI has provided guidance to scientific publishers in addressing these and related questions, in “ManagingAllegations of Scientific Misconduct: A Guidance Document for Editors” (Jan. 2000), available through the ORIwebsite,
National Association Of College and University Attorneys
Part IV Closing notes Other possible legal fallout of an investigation: civil actions for defamation and breach of employment contracts; criminal prosecutions of the respondent; False Claims Act complaints against the institution.
Counsel should be prepared for other legal fallout from serious misconduct cases. Civil
defamation suits against complainants, the institution, and even committee members may beanticipated. Torts ranging from defamation to interference with prospective business advantage;Lanham Act “passing-off” claims (presenting faculty member A’s research as though it were facultymember B’s, for example); patent-based inventorship/authorship disputes, and employment-basedcontract or civil rights suits may all result from research misconduct proceedings. Although ORI hasin some contexts advised that “there may be a qualified privilege under State and common law toreport alleged misconduct to responsible institutional or governmental officials,”69 counsel shouldnot proceed on the assumption that such a privilege will be upheld.
Expect claims that due process has been violated: investigations conducted by universities
seldom attain a standard that the criminal justice system aspires to, and yet, because of the sentencethat may be imposed – debarment or the end of a career – counsel for the respondent can beexpected to attack the proceedings in many ways as if it had been intended to be a criminalproceedings. State institutions will be particularly vulnerable to those charges. An instructiveexample of a complaint filed against the university involved in the Poehlman case is attached.70
Be aware, too, that institutions do not appear to be immune from suit by qui tam relators or
pursuit by the Department of Justice by virtue of their compliance with federal standards. If a grantapplication including fabricated data was forwarded to the PHS by an institution, the Governmentmay assert a False Claims Act violation against the institution as well as the individual. Theinstitution may also have exposure for failure to comply with its assurance that it will establishpolicies and procedures and comply with PHS regulations on research misconduct.71
Student misconduct and faculty misconduct. In all of these, consider carefully
whether the same standards are being articulated to students and applied to them as are to faculty: itwould be unfortunate to discover, for example, that students are being expelled for breaches thathave been excused in faculty. Although it is appropriate to apply different disciplinary processes,available instruction on the university’s expectations and discussion of these aspects of theresponsible conduct of research is important to make available to students of all levels –undergraduates, graduates, and postdoctoral trainees. Globalization and research misconduct. American faculty and students are hardly
immune from the temptations that cause academic misconduct, but international faculty working onfederally funded research may derive a particular benefit from good university programs discussing
69 ORI, “Managing Allegations of Scientific Misconduct: A Guidance Document for Editors,”supra n.12, citing ORIPosition Paper #1, “The Whistleblower’s Conditional Privilege to Report Allegations of Scientific Misconduct,”December 1993 (available on ORI’s web). 70 See Dahlberg, J. and Mahler, C., “The Poehlman Case: Running Away from the Truth,”Science & EngineeringEthics (2006) 12: 157-203. 71 42 CFR § 93.413 (“HHS compliance actions”); see 42 CFR § 93.300 “General responsibilities for compliance.”)
National Association Of College and University Attorneys
U.S. standards and expectations on authorship and plagiarism, since cultural differences may createvulnerabilities and pitfalls for them that could be avoided. What is plagiarism? It’s obvious, and yet it isn’
directly copying extensive sections of text without attribution and head toward theft of an idea (forexample, a new application for a concept, first presented in a grant proposal), or theft of research(i.e., presenting original archival research as one’
s own), the rules of the game – of attribution,
copyright, “ownership,” become hazier.72 ORI Guidance provides that plagiarism can be “theft ofintellectual property,” but appears to be using the phrase “intellectual property” in the broadestsense: the academic notion of plagiarism is not confined to copyright violations.73 We could allbenefit from more thoughtful discussions about the topic. Review your institution’s resources on thetopic and determine whether you already have such tools, but if not, it could be helpful to all in yourcommunity to develop robust online resources about plagiarism.
72 For an excellent discussion of the contours of plagiarism, see Parrish, D., “Scientific Misconduct and the PlagiarismCases,”J. Coll. & U. Law 21:517-54 (Winter 1995); Hoffer, P., Past Imperfect, supra. See also Chronicle of HigherEducation, supra, Special Issue on Plagiarism, “What is Plagiarism?” (Dec. 17, 2004)73 “As a general working definition, ORI considers plagiarism to include both the theft or misappropriation ofintellectual property and the substantial unattributed textual copying of another's work. . . . The theft ormisappropriation of intellectual property includes the unauthorized use of ideas or unique methods obtained by aprivileged communication, such as a grant or manuscript review.” ORI Policy on Plagiarism, available at:ted June 21, 2006).
National Association Of College and University Attorneys
Appendix I Selected Readings On Academic Misconduct
Counsel who are suddenly faced with an academic misconduct case that looks quite substantial, orwho are in the process of reevaluating their institution’
s policy and procedures, would benefit from
reviewing secondary sources about the history of academic fraud generally, the particulars of pastmisconduct cases, and the development of relevant regulations, because many judgment calls willbe required, and internal precedent and policy guidance are likely to be sparse. The literature isextensive, but the following list provides a good starting point, in addition to the materials cited inthe footnotes to the text of this paper.
Decoo, Wilfried, Crisis on Campus: Confronting Academic Misconduct (MIT Press 2002)(thoughtful academic discourse with slight emphases on scientific misconduct)
Galison, P. and Biagioli, M (eds.), Scientific Authorship, Credit and Intellectual Property in Science(Taylor & Francis Books 2002)(history of science perspective and analysis of authorship)
Goodman, Allegra, Intuition (Dial Press, Random House 2006) (a very instructive and compellingnovel based on the Baltimore case. Enlightening about psychological realities and factualambiguities)
Hoffer, Peter Charles, Past Imperfect: Facts, Fictions, Frauds – American History from Bancroftand Parkman to Ambrose, Bellesiles, Ellis, and Goodwin (Public Affairs 2004) (analysis ofplagiarism, fabrication, and falsification in history by University of Georgia historian, formerdirector of American Historical Association program on misconduct)
Judson, Horace Freeland, The Great Betrayal: Fraud in Science (Harcourt 2004) (general overviewof major scientific fraud cases in U.S.)
Kevles, Daniel, The Baltimore Case : A Trial of Politics, Science, and Character (Norton 1998) (avery detailed history of the saga of the 1980s Baltimore case, describing handling of matter at earlyphases, through NIH involvement and Congressional hearings)
LaFollette, Marcel C., Stealing into Print: Fraud, Plagiarism, and Misconduct in ScientificPublishing, (Univ. of California Press 1992) (scholarly review and thoughtful analysis)
Robin, Ron, Scandals & Scoundrels: Seven Cases that Shook the Academy (Univ. Cal. Press 2004)(description and analysis of notorious cases in anthropology, history)
National Association Of College and University Attorneys
Selected prior NACUA presentations on Research Misconduct (available through the NACUA website)
• Gibbons, Gail L., “Research Misconduct,” (Nov. 2, 2000 NACUA Institute for Law
• Parrish, D., “Scientific Misconduct and the Plagiarism Cases,” J. Coll. & U. Law
• Price, A., “Difficult Issues for Institutions and ORI in Handling Scientific
• Patti, C., “Managing the Difficult Research Misconduct Case” (2004)
Selected Investigation Committee Reports that are instructive and available online
“Report of the Investigation Committee on the Possibility of Scientific Misconduct in the Work ofHendrik Schön and Coauthors” (“Bell Labs case”) (September 2002), available at
Inspection Report, “Inspection on the Review of Scientific Integrity Issues at Lawrence BerkeleyNational Laboratory” (July 2002)(U.S. Department of Energy Office of Inspector General,DOE/IG-0478)
Report of the Investigative Committee in the Matter of Professor Michael Bellesiles (July 10, 2002),
In the Matter of Eric T. Poehlman, Investigation Report (redacted)(2005),
Report of the Investigation Committee of the Standing Committee of the University of Colorado atBoulder concerning allegations of academic misconduct against Professor Ward Churchill (May 9,2006)
Other materials
AAAS-ABA National Conference of Lawyers and Scientists, Project on Scientific Fraud andMisconduct, Report on Workshops 1, 2 and 3 (AAAS Washington, D.C. 1989)
Barnes, B., et al., “Creating an Infrastructure for Training in the Responsible Conduct of Research:The University of Pittsburgh’s Experience,” Academic Medicine (2006) 81:119-127.
Chronicle of Higher Education, “Special Report: Plagiarism” (December 17, 2004)(collection ofarticles on the subject),
Gunsalas, C. K., “Research Misconduct: Selected Bibliography,” (February 2001)(Trustees ofIndiana University)(http://poynter.indiana.edu/sas/res/misconduct.pdf)(well-annotated guide)
National Association Of College and University Attorneys
Mello, M. and Brennan, T., “Due Process in Investigations of Research Misconduct Cases,”NewEng. J. Med. 349:1280-86 (Sept. 25, 2003)
Reynolds, G. H., “Thank God for the Lawyers: Some Thoughts on the (Mis)Regulation of ScientificMisconduct,” 66 Tenn. L. Rev 801 (Spring 1999)
PHOSPHODIESTERASES AS THERAPEUTIC TARGETSCHING-SHWUN LIN, ZHONG-CHENG XIN, GUITING LIN, AND TOM F. LUE T he advent of sildenafil (Viagra) for the treat- morethanoneisoformthroughalternativesplicing ment of erectile dysfunction has attractedor from multiple gene promoters. The recentlywidespread interests in phosphodiesterase type 5completed human genome sequence3 suggests that(PDE5), the ta
46th ICHA – 8 OCTOBER 2012 CLOSING REMARKS BY THE CHAIRMAN Today, 52 countries were represented (49 members and 3 observers). We have been listening to various presentations during the day and there were many other meetings organized during the previous week. So what are the principal conclusions of these meetings and of today’s Conference ? A- GENERAL ASSEMBLY AND REGULATORY ISSUE