Reduced risk of pertussis in whole-cell compared to acellular vaccine recipients is not supported when data are stratified by age.

TO THE EDITOR—Kiraly and coauthors recently commented [1] that confounding by age might explain the apparent reduced risk of pertussis described by Witt et al [2] among persons vaccinated with a whole-cell pertussis vaccine (wP) compared to recipients of acellular pertussis vaccine (aP). We note that Table 3 of the original article [2] provides sufficient data to estimate the pertussis attack rate per 100 000 population (AR) according to vaccine type received, stratified by year of age (Table 1), and to produce an age-adjusted relative risk (RR). Our calculations lend support to the concern expressed by Kiraly et al. Among individuals aged 8–14 years, the number within each year of age that received ≥1 wP dose ranged only from 42 to 155, providing very limited power to accurately assess ARs or to avoid the risk of erroneously observing 0 cases within a given year of age. In contrast, among those aged 15–20 years, the number exposed to ≥1 wP dose was higher and was better balanced with the number exposed only to aP (404 to 10 825 wP and 3038 to 8959 aP by year of age), thereby increasing power and limiting the risk of erroneously observing 0 cases in a given year of age. Examining the data by year of age showed 2 patterns: