Cumulative incidence of pertussis in an unvaccinated preschool cohort based on notifications, interview and serology.

The cumulative incidence of pertussis at six years of age in the first unvaccinated cohort after cessation of immunisation in Sweden was investigated by serological methods and by a validation of the national epidemiological surveillance system by reports from the Child Health Centres (CHCs). In 312/385 (81%) eligible children, both an interview with the parents and a blood sample from the child could be obtained. The CHC reports yielded a cumulative incidence of 31%, whereas the serological assays found 54%. Of the cases reported by the parents to the CHCs, 89% were seropositives. In the additional cases of pertussis reported at the study interview, only 61% could be confirmed (p less than 0.001). Among children with reported severe cough not suspected to be pertussis, 46% were seropositive, distributed as 33% seropositives in cases with cough duration of less than 4 weeks and 69% for longer coughs (p less than 0.01). In the CHC reports, the parental diagnosis was found to have been confirmed in three-quarters of cases by medical personnel. The CHC reporting system was thus found to be reliable with an observed specificity of 93%, but sensitivity of only 52%. Thus, even this surveillance system, which yields the highest incidence rates, underestimates the incidence of the disease.