Commentary: Author Response
have suggested that both modeling principles and basic beliefs are important in discussing how we should prepare for possible smallpox terrorist attacks.1 Eichner and Schwehm (ES) amplify these aspects.2 ES focus on “surveillance” as a prevention tool. In the ES world, any contact under “surveillance” would be prevented from transmitting further infections at the moment the contact becomes infectious, irrespective of the total number of contacts under watch. Thus, “surveillance” becomes equivalent to quarantine (which was considered in my earlier modeling work3,4).
Presuming that 75% of transmission from infected contacts can be prevented in this manner strikes ES as reasonable, but strikes me as extremely optimistic in the event of a bioterror attack. The potential for societal chaos in the wake of a smallpox attack should not be underestimated. Nonetheless, combined with the additional assumptions in their Figure 2 (that 90% of the population is susceptible and that isolation of symptomatic cases reduces transmission by 75%), their conclusion that outbreaks with R0 <17.7 can be contained follows directly from equation 16 in my article. (Note that their strategy presumes a relative reduction in transmission of π = 1 - .9*[1/4]*[1/4]). Similarly, assuming that R0 = 6, my equation 15 predicts that the expected total number of untraced cases resulting from an initial attack of size 100 would be 150, consistent with their reported upper limit of 200 obtained from simulation (that presumably includes cases detected under “surveillance” as well).
ES note that under their assumptions, postexposure vaccination is irrelevant, but they fail to acknowledge that under their assumptions, all vaccination is irrelevant to preventing new infections! Indeed, ES's assumption that two thirds of contacts are vaccinated successfully did not figure in any of my calculations replicating their results. Taken literally, Figure 2 of ES is telling us that vaccination is unnecessary for purposes of controlling an outbreak.5 This also strikes me as unreasonable.
Although ES are correct that in a single, small outbreak, mass vaccination could lead to greater total casualties than traced vaccination as a result of vaccine complications (a conclusion also reported in earlier work3), they fail to consider the other extreme, namely, that delay in implementing large-scale vaccination in a large attack would cost many more lives. At the instant an attack is detected, one will not know whether it is large or small. However, once it has been established that an attack has occurred, the probability of further attacks can only increase (what Richard Danzig has called reload6). Combined with the goal of controlling outbreaks as quickly as possible, preparation for mass vaccination becomes even more important in the bioterror setting (in contrast with a natural outbreak).
I agree with ES that it is important to consider nonvaccine control alternatives such as quarantine, but for another reason: if terrorists are able to weaponize smallpox, they might be able to reengineer the virus to defeat the vaccine.5 Even if such reengineering is not feasible, terrorist interference with an orderly plan of vaccination is certainly plausible. In short, there is much more than smallpox epidemiology at play in preparing for a bioterror attack. As 9/11 sadly reminds us, one cannot overestimate the consequences of the terrorist imagination.
1.Kaplan EH. Preventing second generation infections in a smallpox bioterror attack. Epidemiology
2.Eichner M, Schwehm M. Smallpox: A vulnerable specter [commentary]. Epidemiology
3.Kaplan EH, Craft DL, Wein LM. Emergency response to a smallpox attack: the case for mass vaccination. Proc Natl Acad Sci U S A
4.Kaplan EH, Craft DL, Wein LM. Analyzing bioterror response logistics: the case of smallpox. Math Biosci
5.Eichner M. Case isolation and contact tracing can prevent the spread of smallpox. Am J Epidemiol
6.Danzig R. Catastrophic Bioterrorism—What Is to Be Done?
Washington, DC: Center for Technology and National Security Policy, National Defense University; August 2003.