Despite eradication, smallpox still presents a risk to public health whilst laboratory stocks of virus remain. One factor crucial to any assessment of this risk is R0, the average number of secondary cases infected by each primary case. However, recently applied estimates have varied too widely (R0 from 1.5 to >20) to be of practical use, and often appear to disregard contingent factors such as socio-economic conditions and herd immunity. Here we use epidemic modelling to show a more consistent derivation of R0. In isolated pre-twentieth century populations with negligible herd immunity, the numbers of cases initially rose exponentially, with an R0 between 3.5 and 6. Before outbreak controls were applied, smallpox also demonstrated similar levels of transmission in 30 sporadic out-breaks in twentieth century Europe, taking into account pre-existing vaccination levels (about 50%) and the role of hospitals in doubling early transmission. Should smallpox recur, such estimates of transmission potential (R0 from 3.5 to 6) predict a reasonably rapid epidemic rise before the implementation of public health interventions, because little residual herd immunity exists now that vaccination has ceased.
Bibliographical noteFunding Information:
We thank T. Boller and A. Wiemken for allowing part of this work to be conducted in the Botanical Institute Basle, T. Mes and M. Wilkinson for advice on the analysis, A. Rodriguez and J. P. Clapp for providing 28S sequences, E. Stöckli and H. Reichert for use of the confocal microscope, P. Heslop-Harrison for advice on in situ controls and J. P. Clapp, M. G. A. van der Heijden, L. Keller, A. M. Koch and F. Mery for critically reading the manuscript. We thank the Swiss National Science Foundation for supporting this work with a standard research grant and a fellowship awarded to I.R.S. under the professorial fellowship programme.
This work was funded by the Department of Health, UK. The views expressed in the publication are those of the authors and not necessarily those of the Department of Health. We thank C. Penn and G. Lloyd for their help with this work and the preparation of the manuscript, and D. Jones, S. Duncan, N. Gay, and members of the DH Steering Group for their comments and help with model parameterization.
We thank D. Garrett and F. Delaglio for software support; C. A. Bewley, M. Caffrey, W. A. Eaton, J. Kuszewski, L. Murphy, C. Schwieters, A. Szabo and N. Tjandra for discussions; and C. A. Bewley for critically reading the manuscript. This work was supported in part by the AIDS Targeted Antiviral Program of the Office of the Director of the National Institutes of Health (to G.M.C.).