Stanford University researchers Lawrence M. Wein of the Graduate School of Business and Yifan Liu of the Institute for Computational and Mathematical Engineering have created a model of the likely effects should terrorists put botulinum toxin into a single milk delivery tanker before it delivers its load to a milk processing plant.
STANFORD GRADUATE SCHOOL OF BUSINESS — A mere 4 grams of botulinum toxin dropped into a milk production facility could cause serious illness and even death for 400,000 people in the United States. Investments that would cost the public only 1 cent more per half-gallon of milk could prevent this nightmare scenario, according to Lawrence M. Wein of the Stanford Graduate School of Business.
Wein, the Paul E. Holden Professor of Management Science, has been conducting a series of studies on the effects of various potential terrorist activities in United States. Not only milk, but soft drinks, fruit and vegetable juices, processed tomato products, and even grains—anything that goes through large-scale storage and production and rapid distribution—could be at risk for such an attack, with catastrophic consequences for the American public, Wein says in his most recent study, conducted with Yifan Liu, a PhD candidate at the Institute for Computational and Mathematical Engineering at Stanford University.
In the case of milk, says Wein, all it would take is for someone to obtain a suitable strain of botulinum toxin—the most poisonous substance known to humans—from an overseas black market lab, grow it in culture, and pour it into an unlocked milk tank or milk truck. From there, the contaminated milk would make its way into large processing silos, where it would poison at least 100,000 additional gallons. Only a fraction of the toxin would remain active after pasteurization, but according to Wein's mathematical model, that could be enough to infect the approximately 400,000 people who would drink the milk. "Only 1 millionth of a gram is enough to poison an adult," says Wein, "and there would be more than that per person remaining in the distributed milk to do the job."
Wein and Liu’s paper was published in the July 12 issue of the Proceedings of the National Academy of Sciences (PNAS) accompanied by an unusual editorial addressing some of the debate on scientific research vs. national security that has arisen since pre publication copies were circulated to a limited list of academics.
Based on their mathematical models of current distribution of milk, Wein and Liu estimate that within 48 hours of ingesting contaminated milk, consumers would begin to display symptoms. If authorities were able to notify the public within the subsequent 24 hours to stop drinking milk (an ambitious time period), the contamination could be reduced. However, of those eventually exposed, "about half would die," says Wein. The death toll would be high (as much as 50,000 he estimated), due to the current insufficient supply of ventilators and antitoxins in the U.S. medical system.
However, an ounce of prevention is worth a pound of poison, says Wein, who last September presented his findings on the case of bioterrorism and milk to the Assistant Secretary of Public Health Preparedness and several members of the Department of Homeland Security.
Wein calls for the FDA to make current volunteer safety guidelines mandatory, such as requiring that milk tanks and trucks be locked and that two people be present when milk is transferred from one stage of the supply chain to the next.
Before releasing milk into silos, milk-tank truck drivers should be required to employ a new 15-minute test that can detect the four types of toxins associated with human botulism. Drivers currently are required to wait for an antibiotic residue test and the toxin test could be conveniently accomplished at the same time. "A single set of tests can be performed on each 5,500 gallon truck at a cost to milk producers that would raise consumer prices only several cents a gallon," says Wein. "We the public need to ask ourselves whether the elimination of this catastrophic threat is worth a one cent increase in the cost of a half-gallon of milk."
Changes made in voluntary pasteurization processes since the 9/11 terror attacks may mean that Wein and Liu’s original calculations of the human toll should be lowered, said Wein. The PNAS editorial, signed by Bruce Alberts, president of the National Academy of Sciences (NAS), agreed but argued that the issues raised by the paper are still of value to both academics and security agencies. “There is everything to be gained by alerting the public and state governments to the dangers so they can help the federal government in its ongoing, highly laudatory attempts to reach 100 percent compliance” with pasteurization processes that will protect the milk supply.
In the original paper (which has free web access) the researchers say a few major inputs into their model have large levels of uncertainty associated with them.
There are also three aspects of the model that have not been discussed in the open literature, although presumably studies can and perhaps have been performed: the inactivation rate attained by pasteurization, the specificity of an ELISA test in milk, and the release size that a terrorist organization is capable of. Such studies would allow our results to be sharpened considerably. The dose–response curve, pasteurization inactivation rate, and terrorists' release-size capabilities each contain several orders of magnitude of uncertainty, and together they essentially determine the release threshold required to achieve a sufficiently high milk concentration. There is much less uncertainty about how many people would drink this contaminated milk.
One big question has to do with the effectiveness of pasteurization at deactivating the toxin.
Taken together, we have a reasonably accurate estimate of the number of people who could be poisoned but a very poor estimate of how much toxin is required to cause a large outbreak. The main uncertainties related to the number of people who could be poisoned are how quickly the attack would be detected via early symptomatics and how quickly and completely consumption would be halted: we optimistically assumed that consumption is halted instantaneously and completely within 24 h after the early symptomatics are detected, even though it took several weeks to identify the source of the two large but more subtle Salmonella outbreaks in the dairy industry (26, 27). Even if the reducible uncertainty resolves itself favorably (e.g., heat pasteurization inactivates 99% of toxin rather than 68.4%), a catastrophic event is not implausible, and the way forward seems clear: invest in prevention, investigate inactivation processes that do not affect nutrition or taste and, most importantly, develop and deploy a sub-45-min highly specific in-process test.
Check out a graph from the paper that shows rate of poisoning as a function of initial dose added to a milk tanker. Note that the highest dose modelled is 1 kilogram. Terrorists who managed to contaminate multiple tanker trucks going to multiple mlk processing facilities could poison a much larger number of people.
Note the key role of pasteurization in reducing the extent of poisoning. Imagine some other food has similar characteristics to milk in that it gets processed at large facilities and rapidly distributed but which does not get pasteurized. Contamination of such a food would deliver a much larger portion of the toxin to the public.
Automated centralized food distribution systems greatly reduce costs. But they also introduce central points of vulnerability. However, on the bright side the big food production facilities also introduce central points for testing and for safety measures to protect foods and neutralize toxins.
One other point: Technologies that increase the shelf life of foods will have the effect of reducing the scale of bioterrorism attacks on food. If food goes bad quickly people will buy it more often and eat it sooner. Fresh milk will get consumed pretty quickly for this reason. But if foods can last a long time then people can buy larger lots at a time for months in advance. Then if toxins are introduced into the food supply a smaller fraction of the public will eat the contaminated food before the contamination causes symptoms and the cause is identified. For example, someone who buys canned or dried milk would run a much lower risk of getting poisoned by botulinum toxin in milk. Most of the consumers of dried milk on a particular day are consuming milk produced at a variety of times in the past.