Journal : vol 2 : March/April : Conference report

Fraud in research

Peter Wilmshurst FRCP, Consultant Cardiologist, The Royal Shrewsbury Hospital, Shrewsbury

Clin Med JRCPL 2002;2:159-60

This debate was held at the Royal College of Physicians on 5 November 2001

Science is meant to be objective but when reporting it, it is subject to bias and competing interest. I agreed to report the meeting for Clinical Medicine before I was asked to second the motion 'This house believes that Britain should be ashamed of its response to research fraud and take action at once'.

A vote taken by the chairman, Dr Richard Smith (Editor, British Medical Journal), before the debate showed that the audience was fairly evenly divided, but there was a large swing in the final vote to support the motion by those originally opposed to it. Not a single voter moved the other way. As a result, the motion proposed by Professor Michael Farthing (Chairman, Committee on Publication Ethics) was carried overwhelmingly, despite opposition from Professor Sir George Alberti (President, Royal College of Physicians) and Sir Donald Irvine (President, General Medical Council).

However, both sides acknowledged that we have a problem with research fraud, that our attempts to deal with it have been inadequate and that more needs to be done. It was also agreed that a national body is required to educate, audit, detect and deal with research fraud. The points of contention were whether our performance was bad enough for us to be 'ashamed' and whether action must be taken 'at once'.

Both sides agreed that for the last 15-20 years there has been recognition that research fraud has been as great a problem in Britain as in other countries. The difference between Britain and other developed countries has been our failure to make any real attempt to deal with the problem. The USA held congressional hearings on fraud in biomedical research in 1981 and 1985, and soon afterwards founded the Office of Scientific Integrity (now the Office of Research Integrity). The Scandinavian countries set up mechanisms for investigating and dealing with allegations of research misconduct a decade ago. Germany and France now have mechanisms.

The British medical establishment has discussed the problem during the last decade, but failed to take adequate action. In 1990, the Royal College of Physicians (RCP) set up a commission that reported in 1991, but the report was not launched with the press conference usual when a college report is published. A few years later, a survey showed that the report had been so effectively concealed that postgraduate deans were generally unaware of its existence - and certainly unaware that they were charged with implementing it. In October 1999, a Joint Consensus Conference on Misconduct in Biomedical Research was held at the Royal College of Physicians of Edinburgh. The report did not appear until October 2001 in the Journal of the Royal College of Physicians of Edinburgh. The British medical establishment has been quick to talk, but slow to take action.

Because the mechanisms for dealing with research misconduct are ineffective, a group of editors of medical journals formed the Committee on Publication Ethics in 1997 to advise editors on how to deal with cases of research misconduct. Admittedly the General Medical Council (GMC) has considered a few cases that involve medical practitioners, but it is unable to deal with research fraud involving non-medical scientists. Even where it has had the power to act, the GMC has been slow to do so. Witnesses have died in the time taken to hold a hearing, and defendants have escaped sanctions because the delays have been considered unacceptable according to European human rights legislation.

A recent case in which the GMC found two doctors guilty of serious research misconduct illustrates the problems1. The research fraud had occurred a decade earlier and had been known for over seven years before it was brought to the attention of the GMC. King's College, where the fraud occurred, held an inquiry 10 years ago and found that fraud had occurred, but no robust action was taken. Whistle-blowers at King's College had been silenced and documentary evidence was destroyed. It took an external whistle?blower to bring the complaint to the GMC, and it was a further three years before the GMC held a hearing before the Professional Conduct Committee. The falsified research was retracted a decade after its publication. In the interim, the research has been cited - which may have led other researchers up blind alleys and wasted resources.

Patients may be directly harmed by research fraud, yet in the public perception it does not carry the same stigma as clinical misconduct. Misconduct by clinicians grabs the attention of the media because the general public can empathise with the personal tragedies of patients or bereaved relatives who provide testimony of misconduct. In these cases, the public galleries of the GMC are full. Clinicians who are found guilty may need the police to protect them from a mob. Such media images of betrayal have severely damaged public trust in the doctor-patient relationship.

In contrast, research misconduct attracts little interest from the public, the media or the medical profession. The GMC's public galleries are empty, there is no human interest and little media coverage. Time and distance isolate the guilty from their victims. Researchers who falsify results may, like the pilot of a bomber, be able to divorce their action from the suffering they cause far away. Like the bomber pilot, a research fraudster may do more harm to the general population than either a lone gunman or a bad doctor, but often we are unable to link research fraud to individual victims.

I was the only person in the public gallery during two recent GMC hearings of doctors found guilty of serious professional misconduct for research fraud. The one press agency reporter who was present for part of the hearing told me that the only matter of interest to the press and the public was that one doctor had used aliquots from a single specimen of his own urine instead of samples from 12 patients as claimed.

In one case, the barrister defending a doctor later found guilty spent much time emphasising the 'betrayal' of his client by a whistle-blower who reported the fraud. Perception is so clouded by the issue of human betrayal that a whistle-blower who reports research fraud by a colleague appears to some to be the villain not the hero, while the research fraudster appears to be the victim. We forget that a dishonest scientist has betrayed public trust just as a doctor who commits misconduct.

The surveys that have been conducted and the databases available in other countries suggest that research fraud is more common than generally believed. Surveys of medical students and junior doctors suggest that tolerance of misconduct becomes greater as careers advance. We are sending trainees the wrong message. Training in research ethics must start in medical schools.

There are many pressures on researchers. There is pressure to publish for advancement and financial incentives from pharmaceutical companies. Journals also contribute to research misconduct because they are often unwilling to consider negative studies. Sir Donald Irvine stated that the GMC has started to take a tougher line on research fraud in the last few years. Since 1995, 11 doctors have been found guilty of serious professional misconduct because they have committed research fraud. Eight were erased from the medical register, two were suspended and only one was permitted to remain in practice with conditions attached. Sir Donald's message was that a doctor found to have committed research fraud is likely to lose his or her licence to practise as a doctor. He believes that we will soon have a system of governance in place to deal with research that mirrors that for clinical governance. This sounds like jam tomorrow.

The most convincing evidence of the deficiencies in dealing with research fraud in Britain is that when cases eventually come to attention, there is often proof of cover up. We consistently find that those who expose research fraud suffer. Many whistle-blowers suffer a worse fate than those who commit the fraud.

Unless researchers address the issue speedily, they may find a solution imposed by others. Some evidence for this comes from organisations such as the Wellcome Trust that is insisting that grants for research are tied to agreements on ways of dealing with research misconduct. However, this issue is too important to be left to individual grant?awarding charities. National rules are required that govern both clinicians and non-clinicians. We need to make clear to students from early in their training that research integrity is crucial and as closely linked to the right to practise as clinical integrity.

Reference

1. Farthing MJG. Retractions in Gut 10 years after publication. Gut 2001; 48:285-6.

This page last updated on November 15, 2002