Clustering of childhood asthma hospital admissions in New Zealand, 1999-2004

Abstract

The context for this study is public concern about aerial spraying of biological insecticides over Auckland, New Zealand between January 2002 and May 2004. We analysed childhood asthma hospital admissions for the whole of New Zealand, July 1999 – December 2004 using a spatial scan statistic. We found spatial clustering of asthma admissions in many New Zealand cities, and spatiotemporal clustering in a few cities. We hypothesize that many of the purely spatial clusters might be explained by characteristics of the local population or health services.

This explanation is less plausible in the case of the observed space-time clusters of asthma admissions, which we consider more likely to be related to local exposures. In spatiotemporal models, there were significant clusters in Auckland, Palmerston North, Lower Hutt, Christchurch and Invercargill. Two of the four Auckland clusters overlap biological insecticide spray zones, and two do not; the majority of the observed spatiotemporal clusters are unrelated to aerial spraying of biological insecticides in space and time. While the present results do not allow us to identify which local exposures are most relevant in explaining the observed spatiotemporal clusters, we hypothesize that air pollution, including fine particles of biological and non-biological origin, might play a role.