Estimation of the duration of vaccine-induced residual protection against severe and fatal smallpox based on secondary vaccination failure

Background: Understanding the loss of vaccine-induced immunity against smallpox is essential in determining the fraction of those who are still protected in the present population and in constructing effective countermeasures against bioterrorist attacks. Method: Three small Australian outbreaks from the 1880s to early 1900s were investigated. Each documented individual age at infection. The case records for Launceston, 1903, further documented the age at vaccination and disease severity, enabling estimates of the duration of protection against severe and fatal smallpox. Results: A significant association between vaccination and death was observed in the outbreak in Sydney, 1881 (odds ratio of death among vaccinated individuals = 0.3; 95% confidence interval (CI): 0.1, 0.8; p = 0.02), where the time since last vaccination was similar for all vaccinated cases. In Launceston, 1903, where the age at vaccination varied widely, the median duration of partial protection against severe and fatal smallpox was estimated to be 31.7 (95% CI: 13.2, 116.2) and 53.9 (95% CI: 25.6, 123.5) years after vaccination, respectively. Whereas those in their 20s are expected to have the highest frequency of vulnerability to smallpox death in the present population, infections among those in their 30s or 40s are expected to be much less fatal. Conclusion: Long lasting partial protection was suggested from the outbreak records, the estimated durations of which were roughly consistent with those reported previously. In the event of a bioterrorist attack, those involved in emergency tasks before emergency vaccination practices are re-established should ideally be previously vaccinated individuals in their 30s or 40s.