Richard Smith: Should scientific fraud be a criminal offence?

At Britain’s first and only summit meeting on research misconduct in 2000, Alexander McCall Smith, a professor of medical law and ethics, argued that research misconduct (the gentlemanly phrase for scientific fraud) should be a criminal offence. The idea seemed outrageous. Nobody took it seriously, but 13 years later Nature has published an editorial not promoting but contemplating the idea. To me it begins to look like an inevitability that research misconduct will eventually become a criminal offence.

Despite searching, I can’t find McCall Smith’s paper online, but my memory is that he had two main arguments. Firstly, he argued that scientific fraud is really no different from financial fraud, which is a criminal offence, in that resources are misused. We might use the word stolen. Secondly, scientific fraud needs careful investigation and collection of evidence, procedures that are very familiar to the police and unfamiliar to university authorities. He might have added (and perhaps he did) that scientific fraud might do much more harm than financial fraud in that it could lead to global misunderstanding, including perhaps widespread use of ineffective and dangerous treatments.

By 2000 I had been interested in research misconduct for nearly 20 years. My interest was prompted by my predecessor as editor of the BMJ, Stephen Lock, who was widely perceived as “somewhat nutty” when it came to scientific fraud. Now he looks like a man far ahead of his time. Although I’d been watching the scientific establishment fail completely to address the problem of research misconduct, I couldn’t take McCall Smith’s argument seriously. Now I do.

The scientific establishment has standard arguments to brush away research misconduct: it’s rare, science is self-correcting, and nobody is harmed. All the arguments now look threadbare.

A systematic review has shown that research misconduct is common, terrifyingly common. (Anybody who is sceptical about research fraud should sign up to the brilliant Retraction Watch that will bring you several cases every week of scientific misconduct.) There is such a crisis around much of science being plain wrong and never reproduced that an initiative, supported by Science and Nature, has been launched to encourage studies to be reproduced. And examples of people being harmed by fraudulent research mount: the many children who have contracted measles because of misleading information about the MMR vaccine; the women with breast cancer treated unnecessarily with bone marrow transplants because of fraudulent research; and millions of patients who have had their postoperative pain mismanaged because of fraudulent studies.

What is also highly unsatisfactory is how hundreds of studies (and probably many more) that are fraudulent remain in the scientific literature without any signal that they are inventions. I’ve been involved closely with two fraudulent researchers who between them have generated more than a hundred studies that are not retracted. Science is failing in its duty to the public.

I’ve also come to realise first hand the difficulty that research institutions have in investigating cases and gathering evidence. Fraudsters escape because of the incompetence of the institutions, whereas investigation and collection of admissible evidence is the daily job of the police.

It’s time, sadly, to criminalise research fraud.

Richard Smith was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.

Competing interest: RS has been writing about research misconduct and bemoaning lack of progress for 30 years. He’s a founder member of the Committee on Publication Ethics, a trustee of the UK Research Integrity Office, and has been involved in many cases of research misconduct (not yet as an offender). These views are his own.

See also: