Every so often I hear stories of people prominent in medical research who are suspected of research misconduct—or research fraud, as it was once more bluntly called. Could the stories possibly be true?
Some prominent people have been found guilty of misconduct, and we know that research misconduct is not rare. A systematic review found that about 2% of researchers reported that they themselves had been guilty of serious misconduct and that 14% of their colleagues had been. (The King James Bible tells us “Thou hypocrite, first cast out the beam out of thine own eye; and then shalt thou see clearly to cast out the mote out of thy brother’s eye,” but we are better at seeing motes in others’ eyes than beams in our own.) And people who do multicentre trials expect that about 1% of the centres may falsify data. Retraction Watch now does a marvellous job of highlighting dozens of studies being retracted every year because of misconduct, but all the data suggest that most misconduct goes unidentified, unreported, and uncorrected.
We might thus suspect that some prominent people have been guilty of research misconduct, but have not been exposed. We might like to hypothesise that prominent people are less likely to be guilty of fraud—they are after all widely respected and published, our leaders. But we could hypothesise with equal validity that they might be more likely to be guilty of research misconduct in that it was cutting corners, inflating the significance of their findings, and bending data to fit favoured theories that made them prominent.
This is where I began to think of sexual abuse of children. Britain is now experiencing an epidemic of people prominent in public life being accused of sexual abuse of children. The epidemic was precipitated by the discovery, after his death, that Jimmy Savile, a television personality known to almost everybody in Britain, had been abusing hundreds of children and young people for some 50 years. His prominence had helped him avoid detection and conviction.
Sadly it seemed that adults could abuse children with little fear that the child would make an accusation, and if they did they wouldn’t be believed. The abusers could also be confident that the authorities—whether it was the church, the BBC, hospitals, schools, and even the police—would cover up anything that was reported. And ironically the better known you were the more confident you could be that your abuse would be misbelieved, ignored, or covered up.
Could it be the same with research misconduct? Because research works on trust rather than detailed audit it can be easy to fabricate or falsify data. If you submit a study to a medical journals saying that there were 200 people in the study nobody asks to see pictures, signatures, or records; and nobody doubts that data you report. Then even when misconduct is suspected the bias seems to be that the problems arose from error rather than intention. And the culture of most organisations has been to cover up misconduct, especially if the accused is a prominent researcher. A university or research institute could be fairly confident that it could bury cases of misconduct without fear of exposure.
So it may well be that prominent people in medical research have committed research misconduct, but we have never heard of it.
But the culture is, I hope, changing. Liam Donaldson, former chief medical officer of England, has made this powerful statement: “To err is human; to cover up is unforgivable; and to fail to learn is inexcusable.” Covering up becomes more risky every day, and a research institution covering up is, as Donaldson says, a more serious crime than a researcher, no matter how eminent, committing 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 a long interest in research misconduct, was a founder member of the Committee of Publication Ethics (COPE), and is on the board of trustees the UK Research Integrity Office. These views are his own (did you ever doubt it?)