Doubts over three influential head injury studies mean that patients are receiving treatment that may be unsound, warn doctors in this week?s BMJ.
Professor Ian Roberts and colleagues describe the worrying story of Dr Julio Cruz, a neurosurgeon who published three trials on the use of the drug mannitol for head injuries between 2001 and 2004.
The results showed that high dose mannitol greatly reduced death and disability six months after the head injury, but they have recently been called into question.
Dr Cruz committed suicide two years ago and his co-authors and the editors of the journals in which the three papers were published have failed to respond adequately to the concerns raised about the integrity of the data.
Despite considerable efforts, Professor Roberts and his colleagues say they are ?left with serious doubt about important studies but with no way of determining with confidence whether the results are fabricated or real.?
The implications for patients are serious,? they add. ?They are being treated on the basis of potentially unreliable evidence.? They believe that a strong case exists for an international body to respond to the problem of research misconduct.
These events have several important consequences, warn Charles Young and Fiona Godlee in an accompanying editorial. For instance, if some or all of Cruz?s data are false, then doctors will be providing their critically ill patients with uncertain and possibly harmful treatment.
The failure to retract unsound data also has long term consequences as the data become integrated into subsequent reviews and guidelines, they add.
Young and Godlee believe that editors should react promptly to alert readers in situations where inaccuracies or misleading statements may have been published. In the case of Cruz?s data, they also call on the editors and co-authors involved in the disputed papers to act quickly either to confirm the veracity of what they have published or to withdraw it, and they urge clinicians treating patients with head injuries to approach these data with caution.