Six historical studies were investigated to clarify the obtainable information on postexposure vaccination against smallpox. Using the distribution of incubation period, the frequency of cases by time from exposure to vaccination was obtained. More than half of all failures happened within 7d after exposure in all six records investigated. Based on two studies (n=36 and 28), the probability of escaping severe smallpox was further analyzed using logistic regression, showing an inverse association between severe smallpox and time from vaccination to onset (p<0.01 and p=0.04, respectively). Whereas the relationship between the probability of developing severe disease and the time from vaccination to onset could be analyzed with the available information, our findings indicate that previous epidemiologic records showing cases alone, rather than also showing individuals probably protected, are not useful for clarifying the effectiveness of postexposure vaccination by time after exposure.