The fraud squadBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d4017 (Published 28 June 2011) Cite this as: BMJ 2011;342:d4017
Eric Poehlman holds a unique place in the annals of medical research. Once a highly regarded US researcher on obesity, metabolism, and ageing, he published hundreds of papers and garnered millions of dollars in grants over his 20 year career. In 2006 he became the first biomedical scientist in the US to go to jail for falsifying research data.1 Last year another US serial research fraudster, anaesthesiologist Scott Reuben, was jailed for conducting a series of fraudulent clinical trials of multimodal analgesia over six years.2
Prosecutors in Germany launched a criminal investigation this year after the editors of 16 medical journals retracted 88 articles by professor of anaesthetics Joachim Boldt because he failed to obtain ethical approval. He was stripped of his professorship, and the hospital where he was chief anaesthetist set up an investigating commission to review his work for data fabrication, falsification, or misrepresentation.3 4
In the United Kingdom, the Andrew Wakefield saga has highlighted the inadequacies in the country’s system for tackling research misconduct. Dr Wakefield, a reader in experimental gastroenterology at London’s Royal Free Hospital School of Medicine, sparked a worldwide scare by suggesting a link between autism and the MMR vaccine. He was struck off the medical register in 2010, and only then did the Lancet, which published the research in 1998, retract the article. Although he was ousted from his job in 2001 for refusing to replicate his controversial study, the medical school did only a cursory investigation in 2004 when it was made aware of substantial concerns about his work, and it took a journalist, Brian Deer, to provide clear evidence that the study was not only unethical but fraudulent.5
The effects of the failure to tackle his wrongdoing more quickly and decisively are still being felt. Vaccination levels in the UK have still not fully recovered, and measles cases are on the rise around the world. 6 University College London, which merged with the Royal Free medical school a few months after the Lancet article was published, announced only three months ago that it would at last investigate the events behind the Lancet paper and Dr Wakefield’s other work.7
Research misconduct can often take years to come to light. It can go undetected for decades, poisoning the evidence base on which doctors rely in choosing treatments for their patients. The scale of the problem is uncertain, although in 2009 the first meta-analysis of surveys questioning scientists about the incidence of research misconduct produced worrying findings. Around 2% admitted fabricating, falsifying, or altering data and up to 34% admitted other questionable research practices, with medical and pharmacological researchers the worst offenders.8 Yet proposals by the Academy of Medical Sciences for a new Health Research Agency in the UK make no mention of integrity and skate over the need for measures to ensure that the research it hopes to facilitate is honest and reliable.9
Meanwhile the sole UK body dedicated to promoting research integrity is the UK Research Integrity Office (UKRIO). Set up in 2006 with the backing of the research community and the Department of Health, UKRIO is a low key independent body with no statutory basis and no investigatory powers. It was initially funded by the Department of Health, Universities UK, Research Councils UK, and others. But in a climate of economic stringency that funding has been withdrawn and it is now seeking new backers.
UKRIO’s role is limited. It provides advice and assistance in investigations free of charge but relies on the institutions where research misconduct is suspected to carry out the necessary inquiries. With no power to instigate an investigation, it will advise when whistleblowers or institutions alert it to questionable research practices but can take no action on inadequate institutional investigations and cannot impose sanctions or publicise wrongdoing.
But universities, as a recent editorial in CMAJ, the Canadian Medical Association journal, noted, have an inherent conflict of interest, concerned as they are with “academic reputation, high-profile faculty and the imperative to keep grant and sponsorship money flowing,”10 The temptation is to get rid of the researcher (as happened in the Wakefield case), and often the whistleblower too, and sweep the mess out of sight.
Peer review and other mechanisms for ensuring the integrity of research wholly failed in Dr Wakefield’s case. Some argue that only a statutory oversight body with investigatory powers conferred by parliament can properly safeguard the public reputation of science and the integrity of the research base. Peter Wilmshurst—a cardiologist and one man crusader against research misconduct who has reported around 25 doctors to the UK medical regulator, the General Medical Council (GMC), for research malpractice—would like to see “a statutory body with some powers to investigate to go into universities and compel people to produce their lab books and their results, so they can be investigated.” Researchers found to have misbehaved should have all their results checked, and the universities employing them should face fines and be forced to repay grants, he suggests.
The UK is not the only country currently debating whether its research integrity mechanisms need beefing up. The issue has moved up the agenda worldwide. Canada, which has had its own scandals involving faked research, is currently considering introducing a new research integrity regime. The Council of Canadian Academies has proposed a federal agency without investigatory powers, focusing on education and prevention.11 Not good enough, say CMAJ editor in chief Paul Hébert and colleagues in their editorial.10
They call for a new agency or an existing authority to be given “the power and mandate to investigate all allegations of research misconduct, along with the authority to compel researchers to come before panels and institutions to provide access to all necessary information to exonerate or find fault.” For maximum deterrence, they insist, the authority must “publish the names of all individuals involved in serious misconduct, release the outcomes of all investigations and issue regular reports.”
Such powers require legislation, and few countries so far have research integrity regulators set up by statute. Those that have legislated have usually done so in response to high profile scandals that have been seen as a threat to public confidence in science. But several countries are looking at their research integrity systems and debating the extent to which problems can be dealt with by advisory bodies, codes of conduct, and a focus on prevention and training, or whether stronger regulatory powers are needed.
Norway, Denmark, and Croatia have independent bodies created by legislation, with some investigatory powers, and Poland’s parliament last year adopted measures for a national commission to oversee research ethics. In Sweden, universities are obliged by law to investigate any allegations of research misconduct and can refer more complex investigations to a central body. The US has the Office of Research Integrity (ORI), a legislated body within the federal government’s Department of Health and Human Services, which oversees the regulation of research funded by the federal public health service (box).
US Office of Research Integrity
The roots of the US Office of Research Integrity (ORI) date back to 1981, when congressional hearings followed a public furore over the discovery of research misconduct at four major US centres. It was created from the merger of two earlier offices in 1992.
The ORI lost its powers to initiate its own investigations more than a decade ago, and fact finding is carried out by the institutions where misconduct is suspected. But institutions receiving federal funds must have a policy on research misconduct that includes a proper investigation. Findings of investigations must be submitted to the ORI, which reviews them for “timeliness, objectivity, thoroughness and competence.” The ORI may reanalyse the research data, publications, or other source documents, and make recommendations to the assistant secretary for health on administrative actions to be taken against the researcher.
Possible actions include corrections or retractions of journal articles and a ban on receiving federal funds or acting as a consultant to a public health body. The office saw Dr Poehlman jailed for a year for using fraudulent data in securing millions of dollars in federal research funds and in academic articles over a decade.
Transparency is a major feature of the ORI’s remit. Findings and actions taken are publicised and can spell the end of a research career. The ORI’s openness contrasts with the practice of the UK office, which has not yet published any details of its cases. UKRIO has only an advisory role in investigations and its participation is absolutely confidential. It will provide experts if an institution asks but has no power to insist on a thorough investigation or require the findings to be published. It plans to publish a report on its first four years soon, with anonymised summaries and statistics of the types of cases it has handled, and suggesting lessons to be learnt.
The impetus for setting up UKRIO can be traced back to a scandal in 1994 when Malcolm Pearce, a senior lecturer at St George’s Hospital Medical School in London, reported that he had salvaged an ectopic pregnancy by transferring the embryo to the womb. The paper, which attracted worldwide attention, was published in the British Journal of Obstetrics and Gynaecology, of which Dr Pearce was an assistant editor, with his head of department, Geoffrey Chamberlain, listed as a coauthor.12 Professor Chamberlain happened to be the editor of the journal and president of the Royal College of Obstetricians and Gynaecologists. An investigation found that—unbeknown to Professor Chamberlain—the patient never existed, and Dr Pearce’s report in the same issue of the journal of a randomised trial of treating women prone to miscarriage with chorionic gonadotrophin was also shown to be a fiction.13 Dr Pearce was struck off the medical register, and a BMJ editorial called on the UK “to abandon its lax approach to scientific fraud.”14
Dr Pearce was at least unmasked quickly. A fraudulent paper and abstract by Anjan Banerjee, published in Gut in 1990, were not retracted until 2001, after he was suspended from the medical register and a year before he was struck off by the GMC.15 In the intervening years he received various promotions and honours, including a Hunterian professorship from the Royal College of Surgeons. His medical school, King’s College London, had carried out an investigation in 1991 that concluded that his research was unreliable, but it failed to inform the GMC or Gut of this. The GMC was later told that the laboratory books from the time he carried out his research had disappeared, which impeded its investigation.16 Dr Wilmshurst, who reported Dr Banerjee to the GMC, observes: “In my experience, when it comes to a choice between exposing misconduct in their institution or covering it up, the senior officials in the institution will usually choose cover-up.”
Spurred on by growing concerns about the state of research integrity in the UK, representatives of the medical establishment, including the GMC and royal colleges, convened in Edinburgh in 1999 and produced a consensus statement calling for action to tackle research misconduct.17 UKRIO was one eventual result. Richard Smith, former editor of the BMJ and one of those involved, remembers: “There were high expectations because everybody came together, all the different groups. After years of people saying ‘this isn’t really a problem because it doesn’t happen very much and no patients are ever harmed,’ people sort of accepted that it was a problem and that something needed to be done. And then pretty well nothing happened.”
Eventually, he says, Michael Farthing, a former editor of Gut and now vice-chancellor of Sussex University and vice-chair of UKRIO, “picked up the ball and cobbled together UKRIO. It wasn’t really a consequence of that meeting. He just got so frustrated he began to do something himself.”
In the five years of its existence UKRIO has had some success. Its code of practice for research and its manual for the investigation of research misconduct have been adopted by more than 50 universities, as well as NHS bodies and others. Originally contacted mainly by whistleblowers, it is called in increasingly by institutions, says its chair, Ian Kennedy. But it is hopelessly unequal to the scale of the task. Its caseload has risen each year, reaching over 60 last year, and its remit now covers all fields of research, not just biomedicine. With the end of its original funding in October 2010, UKRIO reconstituted itself as a company limited by guarantee and plans to go on trying to fill the gaps, with a subscription model of support as one possibility.
Why wasn’t its funding continued? The UK Research Integrity Futures Working Group was convened by Universities UK and Research Councils UK in Spring 2009 to consider what research integrity arrangements the UK should have for the future. One option was to further fund and expand UKRIO. The independent science policy consultancy People Science and Policy was commissioned to produce a report on the various options. The consultancy interviewed 42 individuals from 36 organisations involved in research, mainly universities and research councils, and concluded that there were “some disadvantages” in continuing and extending support for UKRIO. The consultancy was told that UKRIO had “started very slowly,” leaving concerns about its capacity to deliver. The greatest risk was that “UKRIO would continue to be seen as an offspring of the biomedical world and fail to win wider support.”18
The Futures Working Group produced proposals for a national body with a £400 000 a year budget “to succeed and build on the work of UKRIO.” 19 While acknowledging that the UK system “does not offer sufficient assurance to the public or research funders that existing research integrity systems function effectively or are robust,” the report shies away from the idea of regulation and argues that investigation of individual cases is best left to organisations employing researchers. “Consultations on research integrity have shown that there is no perceived requirement or appetite in the UK for a new body which would have regulatory or investigatory functions,” concluded the group, chaired by Janet Finch, vice-chancellor of Keele University.
Modest though its proposals were, they coincided with financial cutbacks, and there was no consensus on taking them forward. The report argues that misbehaving researchers who are on the medical register can be taken to the GMC and “any new regulatory function would be likely to create overlap, uncertainty and inefficiency.”
The GMC has dealt with dozens of cases of research misconduct over the years, but typically comes on the scene only when investigations have been carried out by others. MedicoLegal Investigations, a small company which acts mainly for the drug industry, has referred more than a dozen cases to it. Most of its referrals and those from Dr Wilmshurst have resulted in findings of misconduct, but exposing research fraud is not the GMC’s forte. In Dr Wakefield’s case the allegations focused on whether his research was ethical, not on whether it was genuine. “Often there is delay, sometimes the GMC has misinvestigated and I’ve had to go back and say ‘you’re missing that point,’” says Dr Wilmshurst. In many cases, he considers the sanctions imposed to have been “pretty lenient.”
The Commons science and technology committee, in its current inquiry into peer review, has been asking questions about the need for regulation of research integrity and recommendations on the subject could feature in its forthcoming report. Research bodies have long argued that the process for getting approval to do health research is too bureaucratic in the UK, leaving little appetite for more regulation. But the picture is about to be redrawn.
The UK government, which sees the life sciences as an important engine for economic growth, announced in March that a special health authority—the Health Research Authority—would be set up later this year, with the National Health Research Ethics Service at its core. It will work closely with the Medicines and Healthcare Products Regulatory Agency, which regulates clinical trials, to “promote proportionate standards for compliance and inspection,” although as a body set up by statutory instrument it will have no investigatory powers.20
The next step in the plan is to establish this authority in primary legislation so that it can combine and streamline the approvals for health research, which are scattered across various bodies. The bill could also give the authority investigatory powers, but whether it will do so is still undecided, says Marc Taylor, deputy director of research and development and head of research systems and governance in the Department of Health.
“If parliament decides that the authority should exercise powers of inspection and enforcement, then that’s how it will be. And if it decides that it should enforce through other bodies, then that’s how it will be. But I’m sure there will be powers of inspection and enforcement in some of the areas it deals with. It is an open question at the moment exactly what role this new research agency should have in promoting or enforcing research integrity and exactly what relationship that would have to the kind of thing UKRIO has been doing to support good practice in individual organisations.”
High profile cases of research misconduct around the world and major retractions by journals stand as a warning against complacency in the UK. “Anybody who sits in government in the UK and doesn’t think there is going to be something happening here is clearly in cloud cuckoo land,” declares Professor Farthing. When he became involved in the issue of research integrity back in the 1990s he favoured a statutory body. But as a pragmatist, when government is “cutting back on practically everything,” he questions whether now is the time to push for such a body. “I’ve spoken to a number of ministers of different colours over the years and I have found it difficult to get any real interest or commitment.”
Nicholas Steneck, director of the research ethics and integrity programme at the Michigan Institute for Health and Clinical Research, says the Research on Research Integrity Program set up by the ORI and the National Institutes of Health made a difference in the US for a time because it started to pinpoint where the problems lay. “That is what is missing around the world. Everybody is afraid of these offices because they’re afraid of what’s going to happen. What needs to happen is a systematic look at where the problems are and how you go about sensibly fixing them.”
Professor Steneck, who had a long association with ORI, is on the board of UKRIO and is a member of Canada’s Tri-Agency Research Integrity Advisory Group. He thinks a central office with overarching authority would strengthen research integrity in the UK. “I think a central office would make a difference. I think the knee jerk that we don’t want more regulation is not necessarily a good idea. That said, I don’t think that’s going to happen in the UK—none of the players want it.”
He says the failure to take the UK Research Integrity Futures Working Group report forward showed that the research institutions are happy with the authority they have. “You don’t want anyone else to have the authority and therefore you don’t move forward with it. That’s the attitude that I think will prevail.”
Research integrity in the UK
1992: US Office of Research Integrity created
1995: Malcolm Pearce struck off by General Medical Council (GMC) over fictitious reports in British Journal of Obstetrics and Gynaecology in 1994
1997: Committee on Publication Ethics (COPE) founded by journal editors in response to growing concerns about the integrity of authors submitting papers to journals
1999: Edinburgh consensus statement on misconduct in biomedical research calls for a national panel to provide assistance for investigation of research misconduct and “collect, collate and publish information on incidents of research misconduct”
2001: Anjan Banerjee struck off by GMC over fraudulent paper in Gut in 1990
2001: COPE publishes proposals for a national research integrity body
2006: Revelations that research by Norwegian oncologist Jon Sudbo and South Korean cloning expert Hwang Woo-suk, published in leading journals, was based on made-up data rocks the research establishment around the world
2006: UK Research Integrity Office (UKRIO) set up on pilot basis as an independent advisory body without investigative or regulatory powers
2009: UKRIO produces code of practice for research which is adopted by many institutions but is not mandatory
2009: First meta-analysis of surveys questioning scientists about the incidence of research misconduct finds around 2% admit fabricating, falsifying, or altering data and up to 34% admit other questionable research practices8
2010: Andrew Wakefield struck off by GMC for “irresponsible and dishonest” conduct over case studies published in the Lancet in 1998; Lancet retracts paper
2010: UKRIO loses funding
2010: UK Research Integrity Futures Working Group recommends a national body to succeed UKRIO without investigatory or regulatory powers, with a budget of £400 000 a year, but no consensus emerges on taking it forward
2010: Publication of European Code of Conduct for Research Integrity by the European Science Foundation
2011: Academy of Medical Sciences report A New Pathway For The Regulation And Governance Of Health Research does not mention research integrity
2011: Brian Deer’s articles on Wakefield in BMJ produce evidence his work was fraudulent. University College London announces it is investigating Wakefield’s work
2012?: New health research authority to be set up by statute to streamline approvals for research. No decision yet taken on whether it will have investigatory and enforcement powers
Cite this as: BMJ 2011;342:d4017
Competing interests: The author has completed the unified competing interest form at www.icmje.org/coi_disclosure.pdf (available on request from her) and declares no support from any organisation for the submitted work; no financial relationships with any organisation that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; externally peer reviewed.