Infants hospitalised with pertussis: Estimating the true disease burden

Aim:  New Zealand (NZ) has a large pertussis disease burden compared with other developed countries. Accurate ascertainment of disease burden is fundamental to controlling pertussis and informing immunisation policy. Disease burden estimates are primarily from passive surveillance, which underestimates disease incidence. The aim of this study is to use active surveillance to determine pertussis disease burden in infants hospitalised in NZ. Methods:  Using the NZ Paediatric Surveillance Unit, active surveillance from 08/2004 to 07/2005 for infants <12 months old, hospitalised with pertussis. Results:  110 infants identified (196 per 100 000), including six with complications, eight intensive care admissions and one death. The hospitalisation rate (per 100 000) varied with ethnicity, being higher for Maori (296) and Pacific (358) compared with European/other (117). Twenty-four per cent were too young to be immunised. Of infants 6 weeks and older 46% had received no immunisations. Despite being more likely to be immunised Pacific infants had a higher hospitalisation rate owing to a larger proportion acquiring pertussis prior to age 6 weeks. Cyanosis and apnoea were frequent symptoms in young infants. Under-identification, estimated using capture–recapture analysis, was modest for both active surveillance (16%) and passive notification (19%). Conclusions:  Infant pertussis hospitalisation rates are three to six times greater than rates in the USA, England and Australia. Underestimation of disease burden by passive notification in hospitalised infants is modest, suggesting a high degree of clinical awareness by paediatricians in NZ. New immunisation strategies are needed to protect infants from a younger age.