EXPLORE PAST ISSUES
Evidence
Vol. 14 No 2 | Winter 2012
Feature
Painting the mice: research fraud
Dr Gerald Lawson
FRANZCOG


This article is 7 years old and may no longer reflect current clinical practice.

A handful of scientists commit research fraud. Such cases, when exposed, generate widespread publicity, with career-ending consequences. Despite this, and the fact that many countries and institutions have introduced audit and disciplinary measures to prevent such fraud, further examples keep emerging.

In the 1970s, William Summerlin was a dermatologist at the Memorial Sloan-Kettering Cancer Center in New York, conducting research into transplantation immunology. He received much acclaim after announcing that he could transplant tissue from unrelated animals by keeping the tissue in culture for four to six weeks. He demonstrated this by breeding white mice with patches of fur from a black mouse. In 1974, Summerlin was exposed when lab assistants discovered that the patches on the mice could be removed with alcohol. The ‘black patches’ turned out to be the result of a black marker pen. Summerlin later attributed his deceptive behaviour to a combination of mental and physical exhaustion, and pressure to publicise positive results. As a result of the Summerlin incident, the term ‘painting the mice’ became a synonym for research fraud.

Prof Asim Kurjak was a highly regarded obstetrician in the former Yugoslavia. In 1974, he published a paper in a Yugoslav medical journal entitled: ‘The effect of continuous lumbar epidural analgesia on the fetus, newborn child, and the acid-base status of maternal blood’. Fifteen years later, Dr Iain Chalmers became aware that it was a plagiarised article, with much of the text taken verbatim from a London study written in 1971. Chalmers’ colleague, Dr Jim Neilson, discovered an additional layer of deception – Kurjak’s paper was actually an amalgam of the 1971 article, combined with a similar 1973 article, both of which had been published in the Journal of Obstetrics and Gynaecology of the British Commonwealth.

Chalmers notified the University of Zagreb about this plagiarism. It responded by asking that ‘as Professor Kurjak is a very distinguished expert, with significant contributions to ultrasound medicine, we would appreciate your tactful handling of the case.’

Chalmers went along with this request and did not publicise it. However, in 2002, over a quarter of a century after the original incident, he became aware that Kurjak had done it again, plagiarising a Norwegian student’s PhD thesis for inclusion in a book he (Kurjak) had written. Chalmers regretted his earlier acquiescence in the request for ‘tactful handling’, and publicised the saga in the British Medical Journal.

Unfortunately, many of Kurjak’s colleagues rose to his defence, saying that it all happened a long time ago, there are more important problems, no one ever died from plagiarism and so forth.

The 1980s

Stephen Breuning was a rising academic star at the University of Pittsburgh. By the age of 30, he had published widely on medications for hyperactivity and individuals with learning difficulties. He argued that certain tranquilizers did more harm than good. At one stage, he had published almost one-third of the material in this field. However, his supervisor, Dr Robert Sprague, became suspicious at the remarkable speed of his publishing output and asked for an investigation by the National Institute for Mental Health (NIMH).

The NIMH found that he had fabricated results and that ‘none of the described studies of psycho-pharmacological treatment had been carried out’. Breuning blamed his behaviour on personal problems. He was convicted, in 1988, of scientific fraud and for making false statements on Federal grant applications, which is a criminal offence in the USA. Breuning was imprisoned for 60 days in a half-way house, and appears to have been the first person to serve time in prison for research fraud. This setback did not seem to derail his career. He resigned from the University of Pittsburgh and went on to establish a psychological practice in Rochester Hills, Michigan. His current website states that he is ‘accepting new patients’, but there is no mention of his misadventures in the 1980s.

Prof Michael Briggs was a graduate of Liverpool University, in the UK. He held a series of academic posts, as well as working for the pharmaceutical companies Wyeth and Schering. Briggs performed extensive research on the effects of oral contraceptives, especially the newly emerging triphasic contraceptives.

In 1973, he moved to Australia and, in 1976, was appointed professor of human biology at Deakin University, in Geelong. In the early 1980s, the chair of the Deakin University Ethics Committee, Dr Jim Rossiter, wrote to Briggs, querying his rapid recruitment of women on contraception. They were all perfect candidates: under 30 years old, within ten per cent of the ideal body weight, non-smokers, had never previously used the pill and were not on any medication. Not being satisfied with Briggs’ reply, Rossiter made a formal complaint to the university. The case dragged on and Rossiter was subject to threatening phone calls in the middle of the night. In 1985, a new inquiry was begun. However, in September of that year, Briggs resigned from the university and the investigation was terminated. He moved to Spain and died 15 months later, at the relatively young age of 51, of liver failure. Three months before he died, perhaps realising he had not long to live, he gave an interview to journalist Brian Deer of the Times, in which he admitted to many of the charges laid against him.

John Darsee was a young researcher with a prolific output of publications. He worked at the lab of esteemed cardiologist Eugene Braunwald at Harvard University in the 1980s. In his first 15 months there, he produced five major papers. The discovery of his first fraud came to light when lab workers realised that he had changed laboratory data obtained over three hours to make it look as though the observations had been recorded over three weeks. Darsee said it was a single, isolated, foolish act and Dr Braunwild, after a preliminary investigation, unfortunately chose to believe him. That he was guilty of long-standing fraud only came to light when it was realised there were significant differences between the results that Darsee had reported and those achieved by other laboratories that were working in the same field. This prompted an in-depth review by the National Institutes of Health, which discovered that Darsee had fabricated extensive amounts of data to create experiments that had never taken place. The review also discovered that he had been falsifying data since his medical student days. He subsequently worked at Ellis Hospital, New York, in critical care, on the condition that he did not carry out any research.

Dr William McBride, an Australian obstetrician, achieved fame when he wrote a letter to the Lancet in 1961, raising the alarm on thalidomide as a teratogen. This earned McBride nationwide praise and he was named ‘Australian of the Year’ in 1962. France’s Institut de la Vie awarded him prize money for his discovery and, in 1972, he used the money to set up Foundation 41, a medical research organisation, investigating the causes of mental and physical handicap in babies.

In 1981, McBride published a paper suggesting that Debendox caused birth defects. Multiple lawsuits were launched by parents who felt their offspring had been handicapped by the mother taking Debendox during pregnancy. McBride was a willing expert witness in court for the claimants. However, two junior scientists working at Foundation 41, Dr Phil Vardy and Dr Jill French, discovered that they were the unwitting co-authors of the paper. They also noted discrepancies between the laboratory log data on Debendox and the data that McBride had published. A series of drafts of the paper revealed numerous changes in McBride’s handwriting. They notified the Foundation’s advisory committee of this finding. Nothing much happened until the ABC aired their complaints on the Science Show, broadcast by Norman Swan in 1987. This led to a formal investigation, which concluded that the scientific fraud charges were well-substantiated. He was de-registered from medical practice in 1993 on clinical and scientific misconduct grounds. In 1998, at the age of 71, he was reinstated.

Roger Poisson, was head of oncology, St Luc’s Hospital, Montreal. In 1985, the New England Journal of Medicine published a five-year study on treatment of breast cancer: a randomised controlled trial on mastectomy versus lumpectomy, co-ordinated by the University of Pittsburgh, in which he was one of the authors.

Five years later, a project co-ordinator noted that, of the 89 hospitals involved in the trial, Dr Poisson’s hospital had recruited a vastly disproportionate number of the patients: 354 out of 2163 (16.3 per cent); and that there were discrepancies in the patient data from this hospital. This was reported to the Office of Research Integrity (ORI), the US body that monitors and addresses research fraud. An audit subsequently revealed that patient records had been falsified.

In 1993, the ORI concluded that Dr Poisson was guilty of scientific misconduct. Poisson argued that what he had done was only ‘white lies’, that he had altered data because of his ‘devotion to patients’, so that they could be enrolled in the study; and that he was merely a scapegoat for the US medical establishment, which supported mastectomy, while he advocated lumpectomy. Poisson resigned from the hospital in April 1994.

The 1990s

Malcolm Pearce was an assistant editor of the British Journal of Obstetrics and Gynaecology (BJOG), and a senior lecturer at St George’s Hospital, London. Pearce had written a well-regarded textbook, Obstetric Ultrasound – How, Why and When. The professor at the hospital was Geoffrey Chamberlain, who was also the President of the Royal College of Obstetricians and Gynaecologists and editor of BJOG.

Pearce published a case report in 1994, claiming that he had removed a tubal ectopic pregnancy and then successfully re-implanted it within the uterus. The article reported how the pregnancy continued uneventfully to a normal delivery, the first recorded case of such a successful outcome. This was startling news to other members of the hospital staff. One of the anaesthetists reviewed the theatre records and could find no evidence that such a procedure had ever taken place. It transpired that Pearce had tampered with computer records to create a fictitious patient. The saga was complicated by the fact that Prof Chamberlain had agreed to attach his name to the article as co-author, even though he had nothing whatever to do with the case. He later said he did this out of ‘politeness’.

Pearce was dismissed from the hospital and reported to the British General Medical Council (GMC), where he was found guilty of serious professional misconduct and struck off the medical register. Chamberlain was seriously discredited and felt compelled to resign from both the editorship and the presidency. These events led to a review of other papers by Pearce and four more were retracted. Andrew Wakefield was the lead author in an article in the Lancet in 1998, stating that 12 children attending his clinic at the Royal Free Hospital in London had features of autism and colitis, which, in many cases, had followed earlier vaccination for measles, mumps and rubella (MMR). Wakefield held a television press conference to publicise his findings, an unusual action for a doctor. This generated widespread alarm in England and Wales and vaccination rates for MMR dropped from 92 per cent to 79 per cent over the next decade. However, epidemiological studies did not confirm any link between MMR vaccines and autism. Wakefield chose not to reveal to his co-workers or the editor of the Lancet a number of contentious issues. First, he had a lucrative financial arrangement with a law firm seeking to establish a link between MMR vaccine and autism. Second, many of the children studied were already on the books of the law firm and had been invited to come to his clinic, many from hundreds of miles away. One child came from California. Finally, Wakefield had previously lodged a commercial patent for a measles vaccine of his own, which would obviously have brighter commercial prospects if the MMR vaccine was discredited.

Nonetheless, in the Lancet and to the British Medical Research Council, Wakefield had declared that all the children in the study had been referred to his clinic through ‘normal channels’ by ‘general practitioners, community paediatricians and child psychiatrists’; and that ‘no conflict of interest exists’. In 2004, a journalist at the Times, Brian Deer, broke the story that, before publication, Wakefield had received £55 000 through the legal connection. Further payments to Wakefield eventually amounted to £435 000. This revelation caused ten of the co-authors of the paper to dissociate themselves from its conclusion. Other charges against him were that he had ordered extensive invasive investigations on the children in the study, without any clinical reason or ethical approval. These investigations included lumbar punctures, MRIs, colonoscopes, colonic biopsies, EEGs and Schillings tests. In the face of mounting criticism, Wakefield remained defiant. However, in 2010, an enquiry set up by the GMC ruled decisively against him. Wakefield was de-registered as a doctor in the UK and the Lancet finally retracted his paper, 12 years after publication.

The 2000s

Dr Edward Erin, a consultant respiratory physician in his mid-forties, carried out research at the Imperial College Hospital in London. His 2006 paper ‘The Effect of a Monoclonal Antibody Directed against Tumor Necrosis Factor A in Asthma’ was cited by other writers in the field over 70 times. His downfall was not precipitated by any slip-up in his writings or by suspicious co-workers. Rather, it was his private life that was his undoing. A married man with two children, he had an affair with his medical secretary, which resulted in her becoming pregnant. She was keen to continue the pregnancy. In an attempt to cause her to abort, he spiked her tea and orange juice with methotrexate and misoprostol. She became aware that these drinks contained strange powder and took samples to the police. This led to criminal charges against Erin and, in October 2009, he was found guilty and sentenced to six years in prison.

Perhaps unsurprisingly, these events led Erin’s peers to question the integrity of his research papers. In 2011, the American Journal of Respiratory and Critical Care Medicine retracted four of his papers owing to ‘concern regarding the veracity of the data and the validity of his conclusions’.

Elizabeth Goodwin was a geneticist at the University of Wisconsin. In 2005, she asked her postgraduate student, Chantal Ly, to review some particulars in a grant application that Dr Goodwin was preparing. At this time Ly was seven years into her PhD program.

Ly realised that material in Goodwin’s submission, which was described as ‘unpublished data’, had already been published in 2004. She showed this to a fellow PhD student, Garett Padilla. They noted other discrepancies, such as experiments that had not been performed. When approached about the errors, Goodwin dismissed them as a ‘mix-up’.

The six postgraduate students in Goodwin’s department were deeply troubled and discussed among themselves what they should do. It was decided that Ly and Padilla should meet with the head of the genetics department and show him the suspect data. This led to a formal enquiry that, in April 2006, reported ‘evidence of deliberate falsification’ in studies that had been funded by Federal grants, totalling US$1.8 million. Goodwin resigned shortly before the report was released.

The six postgraduate students suffered collateral damage. Their supervisors informed them that their years of thesis work would now be regarded as contaminated and advised that they should start from scratch on new theses. Of the six, three abandoned their PhDs altogether. Disillusioned, Ly took on a job as a lab technician, while Padilla abandoned science to enrol in law school.

Jon Sudbo, a Norwegian oncologist, published a study in the Lancet in October 2005 that claimed that the risk of oral cancer could be reduced with non-steroidal anti-inflammatory agents. The article was read by Dr Camilla Stoltenburg, the head of the Division of Epidemiology at the Norwegian Institute for Public Health, and also the sister of the Norwegian Prime Minister. She immediately realised that a statement in the article about utilising a particular Norwegian database could not be true, because the database had not been operational when the study was conducted. Stoltenburg passed this information on to the Radium Hospital in Oslo, where Sudbo worked. An inquiry was begun, which concluded that the ‘bulk of Jon Sudbo’s scientific publications are invalid due to the fabrication and manipulation of the underlying data material.’ The article was retracted by the Lancet in February 2006. Dr Sudbo was stripped of his medical degree in November 2006, but this was re-instated in 2009, with the condition that he could not work in medical research.

It is hard to understand why intelligent people engage in such risky, fraudulent activity. The usual motivation appears to be a mixture of intense career and peer pressure to produce significant results and publications, financial incentives to obtain funding grants, and personality disorders or weaknesses, especially vanity and arrogance – the messiah complex.

There appear to be two main groups of fraudsters. The first group is the overly ambitious young researcher, determined to climb rapidly up the career ladder. The second group is more perplexing – senior doctors, often at the height of their careers, and often occupying prestigious positions. It is remarkable that they should jeopardise everything they have accomplished in their careers, for such little apparent gain. Perhaps they have been doing it for years, become addicted to it, and no longer know how to stop?


Leave a Reply

Your email address will not be published. Required fields are marked *