Ethics in medical publishing
Gonzalo Herranz, department of Bioethics, University of Navarra, Spain.
I Bioethics Course
department of Medical and Surgical Sciences. School of Medicine. University of Cantabria
Santander, February 9, 1991
Scientific communication and moral rectitude
The epidemic of scientific fraud
Something about etiology and pathogenesis
Taxonomy of offenses against scientific communication
research, punishment and prevention procedures
Scientific communication and moral rectitude
The Code of Medical Ethics and Deontology devotes its Chapter X ethics of professional publications. In the three parts of its article 38, it proposes the main rules that should guide the physician's conduct when publishing the findings of his research or the conclusions of his programs of study. article 38.1 speaks of the moral obligation to publish and to do so with priority in the professional press, submitting the findings obtained to the consideration of colleagues, before communicating them to the general public. This article enshrines the well-known Ingelfinger Rule. article 38.2 requires authors of clinical research articles to state in their publications that the work has been reviewed C an Ethics committee . article 38.3, which is the one that interests us most, enumerates a list of ethical misconduct in the subject of scientific publications: premature or sensational publication of procedures of undetermined efficacy, or exaggeration of this; the expression of capricious or unfounded opinions; falsification or invention of data; plagiarism what has been published by other authors; the inclusion as author of someone who has not contributed substantially to the design and execution of the work and, finally, the repeated publication of the same findings.
This is the first time that a code of medical ethics or professional conduct has included the topic of the deontology of scientific communication. In the text of the project prepared by the Central Commission of Deontology, it was stated why it was necessary to codify in this subject: "Scientific publications -it said-, due to the important role they play in the permanent Education of physicians and in the professional promotion of their authors, must have scientific quality and comply with certain ethical requirements ".
If it has been necessary to formulate a deontology of academic publication , it is because the corresponding delinquency is in our journals and books. Until about fifteen years ago, the environment in which scientific research was cultivated and its results disseminated appeared clean, innocent. Scientific research was a business that placed great emphasis on honesty and objectivity. The researcher worked, calmly or feverishly, in the search for truth; he checked again and again his results and the experimental conditions under which he obtained them; he did not feel pressured except for the correctness and precision of his findings and wished, before publishing them, to be criticized by his colleagues. His commitment was to study the phenomena, not necessarily to obtain brilliant results. He was interested, above all, in confronting his intelligence and his shrewdness with a problem, to take a pulse on the unknowns he was trying to clarify. He did not seek primarily to elevate his staff, nor to defeat his competitors in a difficult degree program of obstacles, nor to secure a priority over anyone else, or to present results so convincing and promising that they would make the continuity of financial support for his work certain. Those were the days of brilliant minds and simple tools.
The good behavior of the researcher was something connatural, it was in the environment. All his effort was concentrated on walking with ease the way back and forth between the hypothesis and the results, to verify or falsify it; or to refine the methods, to better choose the material on which he worked. But - it was said and is still said - the facts are sacred; the laboratory notebook , a holy book, where the results are scrupulously noted down at the very moment of their birth and confirmation. Its pages, numbered like those of a notarial deed, might be stained with reagents, but they could never be torn out or admit tampering.
The ethos of the research task has always been clear: scrupulous accuracy in obtaining, recording and publishing results is an essential component of the scientific process. Dishonesty in these operations not only morally demeans the scientist who consents to it and clashes head-on with the very nature of research, which is the search for truth, but also betrays the collective responsibility of the academic community. The latter is obliged to maintain the public's trust, since this trust is a condition both for the achievements of research to be accepted without reticence and for the vigorous support that scientific business receives from the public for its further development and expansion. Advanced societies know that investment in research is big business, producing a very high welfare dividend. To impurify research would be to kill the goose that lays the golden eggs.
And yet, despite all these ideals, academic publication has an ethical Achilles tendon in its very structure. Always, to a greater or lesser extent, the researcher needs to prepare, refine and refine the results for presentation. This need obliges the author to decide which results to communicate and which not to communicate, and how to do so. He cannot report them all. Failures are not usually discussed, although someone has proposed to launch a journal dedicated to publishing failures: The Journal of Negative Results. The shavings of information that all research work produces are swept up and thrown into the waste bin. Only what is considered significant deserves to be reported. But then you have to decide whether to arrange the data in tables, preserving maximum accuracy, or to present them graphically, emphasizing observed trends, not accuracy; if you sift them through the statistical methods, you will have to decide which sieve, in your particular case, eliminates useless information, but cannot mask erratic or difficult-to-explain data . We all know that the presentation of scientific data includes a certain measure of artifice, of streamlining, of cosmetics.
But these "fixes" to the results, or rather, to the way of presenting the results, have a limit: that of respect for the truth. "Any scientist," said Medawar, "who is reasonably inventive and imaginative, will almost certainly make mistakes in matters of interpretation.... If you confine your mistakes simply to that field, you will never do much harm, nor need you lose much sleep. That is part of the business of science. It does not matter much, because where one imagines wrongly, another will imagine rightly. But if the mistake is about a fact - if the scientist says that the litmus paper has turned blue when, in fact, it had turned red - then he has good reason to lose sleep and be tormented, every early morning, by those cruel feelings that make one feel on the verge of losing credit . For an error of that subject makes it very difficult, even impossible, for others to interpret correctly the findings of the fallacious scientist: that is, they will not be able to devise a reasonable hypothesis to accommodate them." So much for Medawar's quotation .
If an error has crept into the published results -inadvertent, due to an impurity in the material, due to the omission of a step in the applied method- that the author discovers later, it is easy to fix: the author must communicate it as soon as possible. The academic community is grateful. Moreover, the public confession of the error not only forgives the sin, but also brings the wrongdoer back to a state of grace, gaining him prestige among his colleagues. Just a few weeks ago, an article publishing house of the New England Journal of Medicine, signed by Arnold Relman, praised the honesty of authors who confess their mistakes. It is worth transcribing a few sentences from this article, as they help us to focus this afternoon's topic :
" research is fraught with errors. In the ordinary course of things, the most important errors will be discovered in later work. In a sense, much of the progress in biomedical research consists of discovering and rectifying errors, as new theories and better methods allow researchers to reanalyze previous conclusions. There is nothing reprehensible in good faith error, if it is accompanied by a willingness to look for it and to report it faithfully. Errors discovered in the course of subsequent research are usually, and sometimes only implicitly, corrected by more recent articles dealing with the same problem. If, on the other hand, a major error is discovered in the data, it should be reported by means of a letter to the publisher or some other form of short communication. Typographical errors or minor mistakes made by authors (or editors) are another matter, which are usually corrected by means of a common erratum.
In this atmosphere of trust and honesty, a curious epidemic began 15 years ago.
The epidemic of scientific fraud
Indeed, between the mid-1970s and the mid-1980s, some disturbing cases of scientific falsification and plagiarism were discovered and received tremendous advertising . Some observers of the phenomenon, prone to sensationalism, went so far as to claim that an epidemic of deliberate deception, of alarming dimensions, was corrupting science, especially biomedical science, and killing public confidence in it.
These cases surprised us all. I still remember the astonishment we felt and the comments we made when we heard, in one of the weekly bibliographic sessions of the department of Anatomy Pathology, about the fraud of Summerlin, an immunologist working at Sloan-Kettering in New York, who painted patches of black skin on white mice as if they were allografts. The surprise was repeated each time that, in successive years, a case of fraud was unmasked and added another episode to this new epidemic.
But was it really a new phenomenon? Was there fraud, and how much, before, to give a date, 1975? I have already pointed out that, in general, there is the idea that science was once poor but honest, and that, in the face of the wide advertising received by so many recent scandals, this claim of innocence of the old science is opposed to the conviction that scientific fraud is something new. And the idea will probably endure.
But no one knows for sure if the past was a better time. There is now a sensitivity to the problem that there was not before. Moreover, we know how to diagnose the problem better. But no one has investigated the real incidence of fraud before the 1970s, and perhaps no one is interested in doing so. When Broad and Wade, whose tendency to exaggerate is manifest, included Galileo, Newton and Mendel in the list of science cheats, the academic community was not too happy. It does not seem to be of much interest to start removing the foundations of science. It happens, moreover, that the immense mass of scientific publications of the past are, rather than old and venerable, simply obsolete and no longer attract anyone's interest: they are not yet old enough to arouse the attention of the historian, but they are already sufficiently "outdated" to have to be put aside, since neither their methods nor their results are valid anymore. In today's research , the bibliography to be critically discussed is the recent one, that of a few years ago. That of the day before yesterday has no relevance, especially that which describes research on therapeutic agents or diagnostic methods that have already been abandoned. It can, in any case, be cited in passing as a historical antecedent, but its data are not analyzed in depth in terms of their authenticity or intrinsic value. It is, therefore, to some extent indifferent whether its data are correct or apocryphal or have been manipulated. There is in biomedical bibliography , it seems to me, a kind of scientific prescription of data, just as there is in the legal system a prescription of faults. The passage of time forgives everything. The need to look ahead, to progress, makes us lose interest in the insignificant past.
But it was from 1975 onwards that the issue of scandals grew. According to Lock, there were more than 20 cases of very serious fraud published up to 1985, although it seemed that many others followed a more private procedural path and had not made it to the pages of scientific journals. Perhaps these were not sensational offenses. Today we have detailed reviews of the most notorious cases of scientific crime. By way of example, I will offer the most essential features of four of them, to illustrate the wide spectrum of deceptions used and effects produced.
Elias Alsabti stood out as a plagiarist. He was a bright young Jordanian physician, from a wealthy family with ties to the Amman court, when he came to the United States in 1977 to train as a clinical researcher . His story has been told by Broad. It was his somewhat bizarre ideas about how to treat cancer, his incompetence and his tendency to falsify data that got him quietly fired from some of the centers where he started working. Finally, he managed to be admitted to the well-known M. D. Anderson Hospital in Houston, Texas, from which he was expelled when one of his articles was found to be plagiarized. By then, between the manuscripts he had misappropriated in Philadelphia and Baltimore and the articles he was plagiarizing (retyping the entire text of an already published article , redrawing the graphs and putting his name in place of the original authors), he managed, in two years, to publish more than thirty papers in second-rate journals in Japan and Europe. He was admitted to work in other hospitals, obtained a doctorate in absentia at the American University of the Caribbean and, when in 1980 it was found that his entire list of published articles was the result of plagiarism, as published in the Lancet, he replied that it was the other authors who had plagiarized him. Finally, having cleared things up, he had to leave the United States. The Alsabti case, with its comic overtones, sample how passive or indifferent academic institutions can be when it comes to providing information on the moral conduct of those who work or have worked in them. Institutions opt for silence, because crudely realistic references can make it difficult to get rid of an unscrupulous scientist: saying nothing makes it easier to get rid of undesirables.
The case of Vijay Soman, also reported by Broad, is illustrative, on the one hand, of how the publishing house process can be exploited to appropriate other people's work and, on the other hand, of how excessive confidence in an unscrupulous subordinate can ruin the degree program of his superior. The latter, P. Felig, a very prestigious endocrinologist, asked Soman to evaluate some colleagues' work for publication. Soman recommended that the work be rejected, but he kept a copy: the subject matter was of great interest to him, as it had been his field of work for some years. Mixing his own observations with those of the stolen article , he prepared a manuscript which he sent for publication. Curiously, as if it were the plot of a novel, the manuscript was sent for evaluation to one of the authors of the stolen article . When this entanglement was discovered, there followed lengthy negotiations about priorities - the editors of both groups were childhood friends - which failed, auditors' evaluations, academic records: the final result was the expulsion of Soman and his retraction of 12 articles containing falsified data (eight of them also signed by Felig). These decisions came at a time when Felig had just accepted a Chair at Columbia University. A Columbia University committee recommended that he resign because of the "ethical insensitivity" he had demonstrated.
The cases of John Darsee and Robert Slutsky have certain features in common and some significant differences. Both were highly gifted and ambitious young clinical researchers who applied their intelligence to the production of a massive curriculum of publications. Their mastery of the technique of writing, of the ability to correlate basic research with its clinical applications through the mastery of some state-of-the-art technologies, put them in a position to be judged by no one, as no one person mastered at the same time all the methods they employed. Darsee worked in the laboratory of Harvard cardiologist Eugene Braunwald, whose protection he won, to the extent that the latter did not accept the first accusations of fraud that were made against his partner or applied minor remedies. He associated him with his team in multicenter programs of study and in the course of one of them unexplained discrepancies were discovered in the data provided by Darsee. It was only then that Darsee's work was subjected to a serious research that succeeded in demonstrating that a large part of the articles bearing his signature were faked: not only those carried out at Harvard under Braunwald, but also those produced in the centers where he had worked before, and even during his student days.
The conditions under which Slutsky worked were different. He found, at the University of California, San Diego, that he had to work on his own, as the chair of his department was almost always vacant. In addition, his work at the boundaries of various disciplines was not easy to supervise or evaluate. He worked alone, at a furious pace: during the first seven years of graduate alone, he published 137 papers. In the biennium 1983-1984 he published more than half of them (72), i.e., at the rate of one article every ten days. Slutsky was very generous in offering author positions for his papers, which endeared him to many of his younger colleagues. When his papers were submitted to a commission for promotion within the University, some discrepancies in the results concerning the same group of animals were discovered. When asked for an explanation, he said that he had lost the laboratory notebooks. And seeing that things were taking an unfavorable turn, he left the University of California at San Diego. A review of his papers by a committee showed that 12 of them were the result of imagination, 46 contained forgeries and 79 were considered valid.
For some, such high-profile cases are just the small tip of a huge iceberg. Broad and Wade have suggested that for every serious case that is published, some 100,000 hidden frauds of varying magnitude take place. Stewart and Feder, who examined the work of John Darsee's collaborators, come to the pessimistic conclusion that one in four researchers cheats or indulges in scientific misconduct. When one asks those who have bothered to measure more objectively the magnitude of the phenomenon, one gets much more reassuring answers. According to Miers, the National Institutes of Health Office that monitors extramural research detects about ten cases of research misconduct annually out of a total of 20,000 subsidy projects, which indicates that there is one case of misconduct for every 2,000 projects. While this is regrettable, it does not seem, however, that fraud, with such a leave prevalence rate, will be able to mortally wound today's science. Garfield, the B of the ISI, compares the general crime fees in the United States with those specific to scientists and concludes that scientists show unusual honesty compared to taxpayers, car drivers, unemployment benefits or insurance policy holders.
In Spain, as far as I know, the matter has not been studied. We do not know the incidence of fraud and plagiarism. The cheating in the Chair examinations was a very cruel and inadequate scenario to bring to public light some dirty laundry. I know of a striking case of blatant plagiarism , in which 85 percent of the text of a review article was reproduced verbatim by a group of six co-authors who, I presume, concerted to perpetrate the crime. A complaint of plagiarism was recently filed against an article published in Medicina Clinica, but the authors were able to get out of trouble. We lack serious programs of study on the problem. It is an enormously difficult topic to study, admittedly, so we have to be content with the occasional anecdote. However, I have the impression that minor plagiarism of foreign publications is widespread: it is enough to appreciate the frequency of lexical or syntactic anglicisms to realize that the authors are textually "inspired" by the works of others. I believe that the crime of fictitious authorship is extremely widespread and that a serious concept of authorship has not taken root among us. Given, moreover, our status as a third-rate scientific power, this parasitism on foreign medical bibliography is practically inevitable.
Something about etiology and pathogenesis
What are the causes of authors succumbing to the temptation to falsify data, to invent results, to plagiarize the articles of others, to unduly multiply the issue of authors?
Certain environmental factors have obviously been blamed. These are the most studied, but, before considering them, it is worth drawing attention to certain psychological factors, certain irrational elements, which inevitably interfere in the scientist's action: self-love, ambition, the same exalted excitement before the emerging positive result or the new hypothesis confirmed, the unconscious rejection of unfavorable results, the need to defeat the rival group , superstitions about the time, the way, the person who has to perform a certain operation to succeed, or the need to work alone, or the complex of the unfaithful but shrewd administrator, who puts fifty where he should say a hundred: these are all circumstances that can numb the critical monitoring of circumstances and results and encourage the commission of errors or frauds.
The pathogenesis of fraud has been pointed out by different authors. Altman and Melcher include among the causal agents the pressure to publish or perish, the relaxation of the laboratory's daily discipline , the bosses' predilection for positive results and the reluctance to establish a system, always unpleasant, of surveillance. Koshland has drawn attention to other factors: the issue of individuals participating in research. Although one might think that the lone perpetrator is freer to falsify data, in fact, most of the crimes observed come from large groups of co-authors. This is explained by the fact that, in scientific bibliography , the single author is today an endangered species. It is precisely the works of several authors, not many, of interdisciplinary subject matter, sometimes carried out in different centers, that lend themselves more to falsification, because none of the several co-authors is an expert in all aspects of the complex research and one of them can introduce falsification. Also, too much work, and the excess of success that sometimes goes with it, causes senior researchers to lack the time to attend to the many research projects they have underway. The lack of regular, human, sympathetic contact , on the one hand, and, on the other, the demand for results that conform to what is expected or desired, can create a very dangerous environment that is conducive to falsification. There is a moral obligation among research managers, Koshland points out, to create a mode of communication that favors sincerity and objectivity, where good news and bad news are discussed with equal simplicity, where no one is isolated and without moral support in the low hours, when results refuse to come out.
Petersdorf, in his turn, thinks that the tendency to falsify stems from the savage competitive mentality that was instilled in the United States in candidates to study medicine, the excessive importance given to high grades, the need to win above all else: a culture that favors cheating one's classmates, cheating on exams, using lies to get ahead. He thinks that another factor is the gigantic size of today's scientific business , which creates a lower stratum of young researchers without much experience but who are determined to rise at all costs, which enshrines the need for research projects with too many co-authors, where scrupulous review is impossible.
Certainly, in the way research is done today, there are too many factors that favor, or at least do not curb, the tendency of a few to falsify.
Taxonomy of offenses against scientific communication
When reading the articles published on frauds in scientific communication, it can be observed that there is no defined taxonomy to name, classify and qualify the misconduct committed. Nor is there, in spite of the efforts, a consistent legal grade or a way to apply the criminal law to scientific misconduct. The very interesting joint symposium of the American cafeteria Association and the American Association for the Advancement of Science, held in Irvine, California, in February 1989, showed that law and science are still two heterogeneous cultures.
It is clear that there is a wide field for innocent error, in good faith. There are errors of observation, there are misinterpretations, mistakes in scientific judgment. These faults are unintentional and, to some extent, unavoidable, although it is sometimes difficult to distinguish between relaxed work born of familiarity with the methods, from negligent or positively careless work ; or to distinguish between a poor and erroneous judgment conditioned by a biased or inexperienced view of the problem, from the serious lapse in scientific judgment conditioned by a self-interested interpretation of the results.
I referred earlier to the Achilles tendon of preparing and refining data for publication. There is a fuzzy borderline between acceptable conduct and reprehensible practices. The cosmetics that are then applied to the data cannot substantially modify them. There is talk of dressing up, of making up, of streamlining the data, of rounding them off, as if they were innocent activities. Innocent, to a certain extent. They should not be abused, because they can come too dangerously close to the unethical territory of pruning annoying data , of silence about inconvenient facts, of selective quotation of the bibliography, of rounding numbers up to the point of exaggeration and of exaggerating conclusions to make them fit the predictions, even at the cost of abusive and fallacious distortion.
Participating in the features of involuntary error and culpable manipulation of the results is seriously biased research , that which is done by viewing the phenomenon, not from an extreme, but legitimate, observation position, but through an aberrant intellectual optic. Observation that is biased by prejudice leads to self-deception and, with it, to imposture.
The main form of crime against scientific communication is fraud, which includes such different forms as plagiarism and theft, piracy of other people's work , and falsification or invention of data and results.
There are specific faults of the publishing house process: unduly delaying the deadline for assessment of a work, rating it in a capricious manner, imposing the reviewer's own opinions on the authors, thus creating a more or less systematic bias in the published material, and, in particular, not making full use of the publishing house process to prevent delinquency in biomedical publication.
And there are specific faults in the publication process. In the first place, fictitious authorship, which some call honorary authorship, with its disturbing effect of creating borrowed prestige out of nothing, although with the risk of suffering the damage together with the other members of the "gang". A recent American judgment condemned equally all the co-authors of a work that had been falsified by only one of them, on the grounds that being listed as the author of an article makes each of the signatories a guarantor of the accuracy and reliability of the data it contains. Secondly, repeated publication, either in the variant of reiterating results already published (same control groups, groups of experimental animals or of partially coincident patients, same arguments with the same words of previous works), or in that of dividing a work into fragments and publishing them in a numbered series of articles: this is the stratified or "sausage" publication.
The classification and typification of publication offenses is being formalized. It ends up happening in scientific research what happens in civil society: no law, no crime. The publication of publication standards, the interest not only in style, but also in ethics, is typifying crimes, declaring certain conducts immoral. The process is moving forward. The International committee of Biomedical Journal Editors has issued rules on the conduct to be followed in case of conflicts of interest, on who is and who is not an author, on repeated publication. It is to be expected that the jurisprudence will develop in the future a legal doctrine on intellectual property, now that we already have in Spain a law on the subject, Organic Law 6/1987 (of November 11) which, among other things, creates Art. 534, bis, a of the Penal Code.
Not all misconduct has the same ethical-ethical or legal nature. Curiously, they may receive divergent qualifications: what is harmful in the field of scientific ethics may be considered not punishable in the legal field, and vice versa, what is punishable by law may be almost irrelevant in its effects on science. To the assertion by the U.S. National Institutes of Health committee researcher that plagiarism "is at least as serious a misconduct as the blatant falsification of experimental data ," Gordon responded that, while both are dishonest and abusive, falsification of data harms science in a far more serious way than plagiarism. False data constitute an active deception, tend to go unnoticed and can only be detected and corrected with great difficulty, whereas plagiarism, being inexcusable, merely creates a disturbing echo, an excess of the same data. the law, however, punishes plagiarism, but will probably never dare to do so with fabricated data .
research, punishment and prevention procedures
The academic community, becoming aware that within it there were false prophets and traitors of the truth, began to design and put in internship procedures to repress the delinquency of the academic publication and to establish guidelines for its prevention.
Several institutions have proposed procedures to investigate allegations of misconduct in publication: some universities, such as the University of California in San Diego; state entities that promote biomedical research , such as the National Institutes of Health in the United States; or journal editors, such as the International committee Editors of Biomedical Journals or the Vancouver group . As can be seen, such procedures have developed almost exclusively in the United States, although recently the British General Medical Council has established jurisprudence in a very well-founded judgment, in which it sentenced a researcher who deliberately falsified data in a clinical essay on Idaxozan to lifelong suspension. It is curious, but Zylke has noted that this policy has developed in the United States, not only because that is where the vast majority of publication offenses have been detected, but also because certain political vicissitudes (Watergate, Irangate, among others) made American society a fertile ground for the repression of any form of public misconduct. After all, public misconduct is also fraud in academic publication.
What is the content of the standards issued by these entities? Apart from some very specific ones (e.g., the Vancouver group on Retraction of research Results), they usually offer two different parts: one to guide the initiation and development of dossiers, and the other to propose preventive measures. These are their basic elements:
- The institution should inform its researchers of its criteria and requirements for honesty in research , as well as the procedures to be followed in the event of a complaint of ethical irregularities.
- The files shall be processed as promptly as possible and shall always be recorded in writing.
- If, after a preliminary inquiry, it appears that the complaint appears to be justified, a full-fledged transcript will be instituted and substantiated by a trainer body composed of competent persons who are not closely related to the scientist in question.
- The accused scientist must be given full opportunity to respond to the allegations made against him/her.
- The transcript as a whole should be protected by secrecy, so that the reputation of the accused is not damaged until it has been concluded that he or she has engaged in some subject investigative misconduct.
- If a guilty verdict is finally established, the entities that have financed the research and the journals in which the guilty parties have published must be informed. The same information will be transmitted to academic and research institutions that request references on the guilty party.
- In case of publication of falsified or plagiarized papers, journals should publish a retraction in a clearly visible place, which will be disseminated as widely as possible. (Since 1984, Index Medicus has included an entrance of retracted articles and the Medline service of the National Library of Medicine of the United States includes the grade of retracted articles, when it has written notice of the fact).
The document "Maintaining High Ethical Standards in the Conduct of research" of the association of American Medical Colleges, published eight years ago, recommended particularly harsh measures in the case of proven guilt. In addition to the above, it requested that all pending articles signed by the fraudulent author be placed under embargo and that notice of the verdict be given to journals in which the author had previously published; that academic institutions take serious action against those whose guilt had been proven; that if the falsified research had been financed with public money, the fact be brought to the knowledge of ordinary judges; that a press grade be given to inform the general public of the culprit and of the crime. They also recommended that appropriate measures be taken to ensure that the author could not publish or act as a reviewer of scientific articles for a certain period of time.
The association of American Universities (AAU), which groups more than 50 of the most important research universities, published in January 1989 general guidelines to help institutions develop their own procedures for responding to allegations of fraud, plagiarism and other forms of misconduct in publishing. They contain virtually similar guidelines to those already described, but add some that are based on more recent experience. For example, to avoid stagnation or paralysis of the progress of the transcript , it establishes that the allegations should be immediately submitted for examination, preferably by a commission, which should inform the accused within 30 days that he has been the subject of an accusation that appears to be serious. The formal procedure should never last more than 120 days. He adds that, if no faults are found, no disciplinary action will be taken against the complainant and efforts will be made to prevent any retaliation or discrimination against him/her. Recalls that the University has a responsibility to investigate allegations of misconduct, even if the researcher decides to leave the institution. Universities cannot resort to the easy transcript of letting the researcher leave quietly.
But more interesting than repression is prevention. Everyone agreement that in a field where one's reputation can suffer very serious damage, where the professional degree program can end in a fatal accident, prevention is very important.
It has been proposed that young researchers should be followed closely, not only to prevent them from cheating, but also to teach them good internship research and to help them in their daily problems. research should be carried out in an atmosphere of simplicity, sincerity and mutual financial aid , rather than overly ambitious goals, strenuous work regimes, selfishness or savage competitiveness. Sessions should be held to discuss the progress of the research, to create an atmosphere in which the obligation to validate the results, to confess difficulties and failures, to communicate concerns or dissatisfactions about the progress of the experiments, to discuss the results with the laboratory or field notebooks in sight, are accepted as a matter of course. Ultimately, collegiality can prevent a colleague from falling into the temptation to engage in falsification or fraud. But all this is insufficient, as the cases of Darsee and Slutsky demonstrated to varying degrees.
It is time to finish. The ethics of publication of scientific results has a thousand facets, it is a very important topic , in which we must make much progress, and about which we must speak with much more simplicity and frankness. A few years ago it was a non-existent eventuality or, perhaps, a taboo topic . Today it is in the pages of the journals that create public opinion on biomedical science. Scientific fraud, its epidemiology, treatment and prevention has already been dealt with in several books and extensive reviews. Moreover, I have seen a news item announcing the appearance, in 1989, of a journal devoted exclusively to collecting papers on the various aspects of academic publication misconduct. Its degree scroll is 'Research: Policies, Audit, and Quality Assurance', edited by Adil E. Shamoo, Professor of Biological Chemistry at the University of Maryland School of Medicine, Baltimore. Further discussion will follow.
I would like to make my own the last, paradoxical words of an article by Altman and Melcher. They are these: "It is an irony that, by awakening the conscience of the public and by stirring up the profession's concern about fraud, William Summerlin, John Darsee, Vijay Soman and their scientific fraud cronies have made a greater contribution to science than they would have made if they had devoted their whole lives to honest research ".
Every cloud has a silver lining. I would like this intervention of mine to inoculate us all against the temptation of fraud. Thank you very much.
AAAS-ABA National Conference of Lawyers and Scientists. Project on Scientific Fraud and Misconduct. Report on Workshop number Three. Washington D.C. American Association for the Advancement of Science, 1989.
Altman L, Melcher L. Fraud in science. Br Med J 1983;286:2003-6.
Anonymous. GMC professional conduct committee. Br Med J 1988;296:306
Association of American Medical Colleges. The maintenance of high ethical standards in the conduct of research. J Med Educ 1982;57:895-902.
Broad W. Woulb-be academician pirates papers. Science 1980:208:1438.
Broad W. Imbroglio at Yale (I). Emergence of a fraud. Science 1980:210:38-41.
Broad W, Wade N. Betrayers of the truth: fraud and deceit in the halls of science. New York: Simon and Schuster, 1982.
Engler RL, Covell JW, Friedman PJ, Kitcher PS, Peters RM. Misrepresentation and responsibility in medical research. New Engl J Med 1987;317:1383-9.
Friedman PJ. Fraud in radiologic research. AJR 1988;150:27-30.
Garfield E. What do we know about fraud and other forms of intellectual dishonesty in science? Part 1. The spectrum of deviant behavior in science. Part 2. Why does fraud happen and what are its effects? Curr Cont 1987(14);3-7. (15);3-9.
Garfield E. Some deviant behavior in science has nothing at all to do with fraud. Curr Cont 1987(49):3-5
Herranz G. On the concept of the author. Med Clin (Barc) 1985;84:275-6.
Kohn A. An outbreak of piracy in the literature. Nature 1980;285:429.
Kohn A. False Prophets: fraud and error in science and medicine. Oxford: Blackwell, 1986.
Koshland DE. Fraud in science. Science 1987:235:141.
Lindberg DAB. Retraction of research findings. Science 1987;235:1308.
Lock S. Fraud in Medicine. Br Med J 1988;296:376-7.
Miers ML. Current NIH perspectives on misconduct in science. Amer Psycol 1985:40:831-5.
Mazur A. The experience of Universities in handling allegations of fraud or misconduct in research. In: Project on Scientific Fraud and Misconduct, Report of Workshop Number two. Washington D.C.: Aspen Institute/American Association for the Advancement of Science, 1989:67-94.
Medawar PB. Advice to a young scientist. New York: Harper & Row, 1979:37-39.
National Institutes of Health Investigating Committee. Nature 1989;340:173.
Norman C. How to handle misconduct allegations. Nature 1989;340:315
Collegiate Medical Organization. Code of Ethics and Medical Deontology. Madrid: committee General de Colegios Médicos de España, 1990.
Sava P. Vous avez dit "fraude scientifique"? Prat Sud-Ouest 1989;7(6):5-6.
Stewart W, Feder N. The integrity of scientific literature. Nature 1987;325:207-14
Woolf PK. Ensuring integrity in biomedical publication. JAMA 1987;258:3424-7.
Zurer P. Research integrity is topic of new journal. Chem Engin News 1987; 6. Zylke JW. Investigation results in disciplinary action against researchers. Retraction of articles. JAMA 1989;262:1910-1,1913.