Norwegian researcher admits that his data were faked

The New England Journal of Medicine has joined the Lancet in issuing a statement of concern about papers the journals published by the Norwegian researcher Jon Sudbø, currently at the centre of allegations of research fraud.

It follows an announcement by officials at Radium Hospital in Oslo that Dr Sudbø, a senior consultant who wrote a paper on the link between non-steroidal anti-inflammatory drugs and oral cancer, has told his employers that data were faked.

The study published by the Lancet in October (2005;366: 1359-66) claimed to be based on information from a national database.

In its summary of the methods used, Dr Sudbø's paper said: “We undertook a nested case-control study to analyse data from a population-based database (Cohort of Norway; CONOR), which consisted of prospectively obtained health data from all regions of Norway.”

It has emerged that, although the database in question did exist, it was not open to anyone outside the Norwegian government.

The Lancet was told on 13 January about the suspected fraud. Initially officials from Radium Hospital said they had information that “strongly indicates that material published in the Lancet has not been based upon data from our national databases, but on manipulated data.”

The next day they told the Lancet that “it was not manipulation of real data—it was just complete fabrication.”

In a statement issued on its website on 20 January (www.nejm.org, doi: 10.1056/NEJMe068020) and due to appear in print on 9 February, the New England Journal of Medicine states that a study by Jon Sudbø and others that it published in 2001 (344:1270-80) has two figures (figure 3B and figure 3C) that the authors claim are of two different patients and stages of oral epithelial dysplasia. They are in fact different magnifications of the same photomicrograph, says the journal.

The results of another study in the journal by Dr Sudbø (2004;350:1405-13) were derived from the same database as that used in the Lancet paper, the journal says, adding that as a result “we have similar concerns.”

Radium Hospital has set up an investigating committee, led by the Swedish epidemiologist Anders Ekbom, to look into the allegations and to review Dr Sudbø's other published research, including the two papers in the New England Journal of Medicine.

In the statement published in last week's Lancet (2006; 367: 196), the journal's editor, Richard Horton, said that the Lancet has been told that Dr Sudbø had made a verbal admission of fabrication.

A spokesman at Radium hospital, Stein Vaaler, told the BMJ that Dr Sudbø had admitted the fraud, was on sick leave, and would not be responding to press inquiries.