BioEd Online

A paper that analyses a hypothetical poison attack on the United States has been published despite the government's objections.

The paper's authors modelled the health and economic losses that would result if a terrorist poisoned the US milk supply with the botulinum toxin. The Proceedings of the National Academy of Sciences (PNAS) originally said it would publish the paper on 25 May, but it delayed publication to address concerns from the US health department. After consideration, the journal has decided to publish the paper without any substantive changes.

Bruce Alberts, president of the National Academy of Sciences, which publishes PNAS, explained the journal's decision in an editorial, which was posted with the paper online¹.

"It is important to recognize that publishing terrorism-related analysis in the open scientific literature can make the nation safer," Alberts wrote.

But Stewart Simonson, assistant secretary for public health emergency preparedness at the health department, was critical. "The assistant secretary respects the decision, but he doesn't agree with it," said Marc Wolfson, a spokesman for Simonson.

Dual-use research

The episode raises thorny issues about the proper handling of 'dual use' scientific research. This is research that aims to bolster defence but that could be used maliciously. Scientists and security experts have been wrestling over what to do with dual-use research and information since the terrorist attacks on 11 September 2001.

The PNAS paper, by Lawrence Wein and graduate student Yifan Liu from Stanford University, California, considers what might happen if botulinum toxin were poured into a milk tanker on its way to a holding tank. The toxin is a potent nerve poison that is sometimes used in very low doses in plastic surgery to smooth the skin. Wein and Lu calculate that, when diluted, this would deliver potentially lethal doses to about 568,000 people.

They then estimate the number who would fall ill, with symptoms ranging from cramps to paralysis, depending on factors such as when the poison is detected and whether the milk is pasteurised. In some scenarios, the great majority of these people could fall ill or die.

Health officials contend that the paper amounts to a blueprint for terrorists. But Alberts and the paper's authors say that much of the information is readily available on the Internet, so it does not provide any new ideas. Instead, they said, it will inform the nation's defences. Wein notes that publications on vulnerabilities can help the government to strengthen the nation against attacks, and can help the public put pressure on policy-makers when they drag their feet. "There has been no perceptible move in the food industry or in the government to shift from a food safety to a food security mentality," says Wein.

"In a free society, the notion that we become more secure through transparency is important," said biophysicist Steven Block of Stanford University. "The government needs to be reminded of that from time to time."

Critical timing

The PNAS episode raises the profile of these issues at a crucial time, occurring just before the first meeting of a US government group that has been set up to tackle difficult issues in science and security. The National Science Advisory Board on Biosecurity, based in Bethesda, Maryland, was created in March 2004 and is scheduled to hold its first meeting on 30 June. In his editorial, Alberts suggested that the board use the botulism paper as a case study for further analysis.

In 2003, many journals, including PNAS, said they would specially review dual-use research papers for security concerns, and PNAS did follow this procedure for the botulism paper.

But Elisa Harris, an analyst at the Center for International and Security Studies at the University of Maryland in College Park, says the dispute over this paper shows that the scientific community needs more guidance.

"I think this whole exercise demonstrates that the publishers' statement from 2003 is not sufficient," Harris said. "We need real guidelines for PNAS, other scientific journals and scientists to use in assessing the biosecurity risks of a given manuscript."