Reprint
Doctoring the evidence: what the science establishment doesn’t want you to know
The Sunday Times, August 12 2012
Science has a dirty secret: research is plagued by plagiarism, falsification and fabrication. A new voluntary code is meant to prevent fraud, but it lacks teeth. So how can the sociopaths in lab coats be stopped, asks Brian Deer
In nearly seven years since being appointed editor-in-chief of the British medical Journal, Dr Fiona Godlee has won a name for upsetting consensus.
Publishing hard-hitting investigations alongside traditional research articles, she has challenged drug companies and professional societies like an old-school muckraking reporter rather than the editor of the establishment’s house journal.
Hammering out an editorial last week at the British Medical Association’s redbrick headquarters off Euston Road in London, she tackled one of the toughest problems of all: science fraud and research misconduct.
“It’s difficult to know how prevalent misconduct is,” Godlee wrote, “but there is evidence that it happens frequently.”
She was commenting on a voluntary “concordat”, just signed by a crowd of research funding agencies, that is supposed to outlaw scientific fraud but lacks the teeth to do so.
The science establishment’s consensus is that there is no need for outside scrutiny because, apart from the odd sociopath, given credence by an “irresponsible” media, science is above the kind of misconduct that has tainted the Roman Catholic Church, politics, the press and, of course, the banks.
This is a little like the church saying, as it did, that everything was fine but for a little bit of regrettable priestly paedophilia – or the press claiming that phone hacking was confined to one “rogue reporter”.
READ HOW DEER’S WAKEFIELD
INVESTIGATION WAS CHECKED
For too long, science grandees have refused to confront the ethical misconduct in their midst, which is driven by the need to generate research funding.
If the mandarins of science shirk a house cleaning, others will do it for them. In recent months, the quiet, polite voices of traditional science editors such as Godlee have been joined by noisy and knowledgeable – whistleblowers on the worldwide web.
Science, like other fallen pillars of modern British society, faces a reckoning.
SCIENTIFIC FRAUD is classified under three big sins. The first is plagiarism, best exemplified in Britain by the case of Dr Raj Persaud, the celebrity psychiatrist. He cut and pasted other’s work into his books and articles, and in July 2008 was briefly suspended from practising.
The second sin is falsification, such as in the case of Andrew Wakefield, the so-called MMR doctor. He was erased from the medical register in 2010 over what Godlee calls “an elaborate fraud” exposed by The Sunday Times.
And the third is fabrication, admitted only four months ago by Dr Peter Francis, a British ophthalmic geneticist working in the United States. He made up the results of work never done, leading to sanctions by America’s National Institutes of Health.
Science editors have been complaining about such behaviour for almost a quarter of a century. In December 1988, Dr Stephen Lock, one of Godlee’s predecessors at the British Medical Journal, wrote a similar editorial not far from her present desk.
Lock wrote that he knew of two cases of scientific fraud “both entailing non-existent laboratory work associated with reports of drug trials”. He demanded tough action. None followed.
Godlee writes this weekend: “Whilst scientific misconduct is troublesome in any field, within medical research it is uniquely problematic. It not only harms the immediate participants in clinical trials, it damages the wider population of patients when spurious or misrepresented findings are incorporated into clinical practice.”
It is not just in medicine that dishonesty is well recognised. The problem is much the same in every field. Godlee’s editorial includes a reference to a 2009 study by Edinburgh researchers. Trawling surveys, they reported “conservatively” that 14% of scientists had said they knew of fakery by colleagues and three quarters knew of other questionable practices.
Godlee is not alone in taking this on. In an editorial titled “Face up to fraud”, the editors of Nature wrote last January: “Some fraudulent researchers might be sociopaths who don’t care about the rules. But many others simply believe that they can anticipate the outcome of a research project, and see no downside to fabricating the required results to achieve a stronger signal.”
The new 24-page “concordat to support research integrity”, with a foreword by David Willett, minister for universities and science, is meant to be the blueprint for change.
If its suggestions are followed, universities and other public bodies involved in scientific research will in future have written procedures in place to encourage the reporting of wrongdoing, and systems in place to respond.
“This is good, but much that is mentioned in the concordat remains to be clarified,” Godlee notes. “Who will lead and deliver these changes? Who will be accountable for them? Where will the money come from?” She could have gone further. Although stuffed with phrases such as “highest standards”, the concordat is, at best, encouragement, not action.
Godlee knows there is a long way to go. In 1988, Lock called for a government enforcement agency: a go-in-and-get’em approach to fraud like America’s.
The US Office of Research Integrity requires institutions receivinh grants from the National Institutes of Health to take direct action against suspects. That can mean labs being sealed, computer data seized and researchers interviewed under caution.
A year ago, a report from the all-party House of Commons science and technology committee called for a statutory watchdog to root out misconduct, similar to America’s.
“In the same way that there is an external regulator overseeing health and safety,” the committee recommended to the government, “we consider that there should be an external regulator overseeing research integrity.”
But this plan was neudged into the laboratory long grass. The science establishment has conjured up the concordat in its place.
One of the concordat’s eight headline signatories is Sir Mark Walport, director of the Wellcome Trust, the leading charitable funder of medical research. “Our vested interest is in truth, and nothing but the truth,” he told me last week.
“But who would have a clear vested interest in stopping research misconduct?” I asked him.
“The whole research community has a vested interest,” he insisted, “because actually research is about uncovering the secrets of nature.”
Seen from Walport’s heights, his trade is protected by what scientists showcase as “reproducibility” and “peer review”. To be really accepted, novel scientific findings must be repeated by others, and relevant experts are consulted on plausibility.
These arrangements, however, also benefit big businesses in high profit but low-profile corporations. Research studies that may cost millions — paid from taxes via funding agencies — are submitted free to journals, allowing them to cash in on untruth in some cases.
Peer review is also free, or at least free to publishers. The science, technical and medical division of The Lancet’s publisher, Reed Elsevier, last year ran a 47% operating profit margin.
Again, the taxpayer picks up the tab. With about 27,000 new scientific reports being issued each week, recent estimates suggest that the cost to British universities of staff time spent on unpaid reviewing may total £165m a year.
But if the science establishment and business see little threat from research fraud, new voices are now launching a challenge.
At the sharp end of the change is a website called Retraction Watch, launched in 2010 by two senior American medical journalists. Nominally a blog logging formal withdrawals of scientific claims, it has evolved into a catalyst for an outpouring of concern, creating a feedback loop of disclosure with no national boundaries.
“Science is as human an enterprise as any enterprise,” says Dr Ivan Oransky, one of the website’s operators and executive editor of Reuters Health. “I don’t know that there are fewer sociopaths in science than in politics or corporate America. And there are the same pressures to cut corners, which is really how it all starts.”
Few science frauds are thought to go completely undetected. Two classes of witness usually know of it. The first are graduate students and post- doctoral researchers studying the same problem as the cheat. The second are very senior, sometimes rivals or retired, who can spot claims that are too good to be true.
For them, Retraction Watch and other web-based networks such as Science Fraud have created a new focus to challenge misconduct worldwide. While potential whistleblowers wanting careers once kept their concerns to themselves, and those of distinction spoke obliquely on old boy networks, today any anxieties can be furiously debated in post- publication open review.
One of the liveliest online debates is on allegations of misconduct by a medical research team that is being investigated by the authorities at Cardiff University.
Oransky says scientists are now venting their spleen in overwhelming quantities and languages — which leaves a dilemma for editors such as Godlee, trapped between what they know and what they can do.
The British Medical Journal more than most has campaigned for heightened integrity, but it is also the establishment’s voice.
“We have got progress, but there is still a lot of complacency,” she told me, revealing her concerns more sharply than in her editorial.
“The usual way of dealing with this is that it’s all between colleagues. But research misconduct is a serious matter. Time and money and people’s lives are at risk. I think scientific fraud should be a criminal offence.”
Brian Deer was named specialist reporter of the year in last year’s British Press Awards for his exposure of the Wakefield scandal
briandeer.com
The ones who didn’t get away
Dipak Das, 65 Heart surgeon; US; exposed 2012
Found guilty on 145 counts of fabrication and falsification, including in work on the possible health benefits of red wine.
Peter Francis, Ophthalmic geneticist; British researcher in US; exposed 2012
Admitted making up results from experiments on rats that had never been carried out.
Marc Hauser, 52 Evolutionary biologist; US; exposed 2010
Harvard University said an internal investigation had found him guilty of eight counts of research misconduct involving published and unpublished studies.
Li Liansheng, 45 Mechanical engineer; China; exposed 2009
Fired by Xian Jiaotong University after fabrication and plagiarism was found in published research on compressors.
Raj Persaud, 49 Psychiatrist; Britain; exposed 2006
Plagiarised other researchers’ findings, generally about bizarre disorders, and published them in books and magazine articles.
Scott Reuben, 54 Anaesthetist; US; exposed 2008
Found to have fabricated at least 21 published clinical trials, giving favourable results for new generation painkillers. In many cases, the work simply was not done.
Jan Hendrink Schön, 42 Physicist; German researcher in US; exposed 2002
Found to have made up and falsified some 28 papers relating to semiconductors, published in multiple scientific journals.
Jon Sudbo, 51 Dentist; Norway; exposed 2006
Found to have fabricated numerous studies, including one on painkillers and cancer. Of 908 purported patients in his landmark work, 250 were discovered to have the same birthday.
Andrew Wakefield, 55 Gut surgeon; Britain; exposed 2004-11
Found to have rigged his research suggesting that the MMR vaccine caused regressive autism and enterocolitis. Children in his study were discovered to have neither.
Hwang Woo-Suk, 59 Veterinarian; South Korea; exposed 2006
Found to have faked research results on embryonic stem cells. Nine “cell lines”, supposedly from different sources, had identical DNA.
