“…14 However, using the latter method, a serosurvey of preschool children from the US in the early 1990s reported a similar proportion of seropositive subjects to that in our preschool-aged group, 15 whereas a more recent (1997) US study of preschoolers found seropositivity levels of ≥94%, 16 and a study of US army recruits aged 17 to 23 years found 97% seropositive to type 1, 99% to type 2 and 87% to type 3. 17 The correlation between cut-off antibody titre, test method and protective immunity is uncertain, but a method with less sensitivity and greater specif icity should provide a more conservative estimate of immunity than a highly sensitive, less specific method. Serosurveys cannot easily be compared between different laboratories, but are useful to identify differences in seroprevalence in different age groups and monitor changes, in the same population, if the method remains constant.…”