A Norwegian researcher dreamed up the lives and lifestyles of some 900 people — and used them in a study on cancer. Then, last October, Jon Sudbø had his results published in The Lancet.

The revelation comes hard on the heels of the Woo Suk Hwang scandal, in which several important advances in human cloning reported by the South Korean researcher turned out to be faked.

The blatant nature of Sudbø's fiction emphasizes questions already being asked about the effectiveness of peer review, even in top journals, and about who should be responsible for catching fraud (see page 243).

The latest deception was discovered by Camilla Stoltenberg, a director of epidemiology at the Norwegian Institute of Public Health in Oslo, who was catching up on her literature reading over Christmas. Sudbø's paper claims to analyse a public-health database and show that taking anti-inflammatory drugs can reduce the incidence of mouth cancer (J. Sudbø et al. Lancet 366, 1359–1366; 2005). “I was surprised because it refers to the Cohort of Norway, for which I am responsible,” Stoltenberg says. She knew that this could not have been the source of the lifestyle data as the paper claimed.

Novel approach: Jon Sudbø simply invented his test subjects. Credit: NORWEGIAN RADIUM HOSPITAL

Last week, Sudbø, who is based at the Norwegian Radium Hospital in Oslo, admitted that the data had not come from that database or any other, but from thin air.

Many details of this case still need to be worked out. There is some indication that Sudbø may have mental health problems. It is also not clear what his 13 co-authors knew about the fraud — the paper identifies three others as contributing equally to the research, and among the other co-authors are Sudbø's wife and his identical twin. None of the authors could be reached for comment.

The hospital has asked that Sudbø's other work be examined in an independent investigation, to be set up this week by Anders Ekbom, an epidemiologist at the Karolinska Institute in Stockholm, Sweden. But the case is already set to change research policy in Norway. The country's health minister, Sylvia Brustad, announced on 16 January that previously stalled reforms on medical research will probably be law by the autumn. The new rules would put more responsibility for catching fraud on the shoulders of the institutions where the research was done.

Richard Horton, editor of The Lancet, insists his journal is not at fault. “This is all so similar to the Hwang thing that we have just been through,” he says wearily. “Peer review is a great system for detecting badly done research, but if you have an investigator determined to fabricate an entire study, it is not possible to pick it up.” The mechanism of peer review at his journal is currently being examined as part of the largest study ever conducted into the process (see page 252).