A bioterror attack on the US should be tackled though containment and targeted treatment rather than mass vaccination, according to new computer simulations.
Emory University scientist Ira Longini simulated an outbreak of smallpox resulting from an attack initially targeting about 500 people in a population of 48,000. He used statistical methods to simulate social interaction and model the way the virus would spread into the wider population.
Under current US policy, any smallpox attack would be dealt with by a policy of containment and selective vaccination. Once the outbreak was identified, the authorities would quarantine everyone exposed during the initial attack and then try to trace everyone who had subsequent contact with them. All these people would be vaccinated to try to prevent them falling ill.
The US is stockpiling millions of doses of smallpox vaccine, in case terrorists unleash the virus, but there has been a debate about how to deploy it if an outbreak occurs.
Advertisement
Mass vaccination would in some ways be the simplest approach. But the existing live smallpox vaccine can cause complications that would kill one in a million recipients and cause serious illness in 10 times as many cases.
Quick action
Previous modelling work, carried out at the Massachusetts Institute of Technology in July 2002, had suggested that mass vaccination would be the most effective way of controlling an outbreak.
But Longini, who says he studied an outbreak on a more detailed scale, found the spread of the virus could be controlled adequately if the containment process began quickly.
“If you isolate the first few cases you should only get about 100 more cases,” he says, adding that mass vaccination could easily cost more lives than it would save.
Longini’s study also showed containment to be more important than vaccination in the early stages of an outbreak. “Any delay in individuals withdrawing to their homes or to the hospital after they become ill would have much more of an adverse effect that would failing to vaccinate,” he says.
Early warning systems
“When you take into account the number of people who would die from vaccination, it would have to be a pretty big attack before you started saving more lives [than you lost],” says Michael Soto, a researcher at the RAND corporation, a non-profit think-tank focused on public policy issues.
But Soto cautions that the early warning systems developed to spot an outbreak remain unproven. The systems are intended to sound the alarm as soon as a sudden increase in relevant symptoms is reported, allowing containment measures to be introduced rapidly.
The results of Longini’s study were announced at the American Association for the Advancement of Science in Seattle, Washington and paper has been submitted to the American Journal of Epidemiology for review.
