The Effect of ambient air quality on lung function, respiratory symptoms and bronchodilator use among symptomatic children

Numerous overseas studies have linked both short and long-term exposures to outdoor air pollution to a range of health effects. The differences in air pollution sources, climate and geography in Australia challenged the generalisability of these overseas findings to the Australian setting. In response, the Hunter Illawarra Study of Airways and Air Pollution (HISAAP) was undertaken. The aim of Phase II of HISAAP was to assess the short-term effects of particulates on respiratory health amongst symptomatic children. This thesis presents the results of an analysis of the 345 primary school children eligible for Phase II of the Hunter component of HISAAP. There were multiple daily diary measures on each child, different types of outcomes such as continuous, dichotomous and count variables, as well as several sources of exposure data on pollutants. Because of the complex and hierarchical nature of data, there are several possible methods of analyses that could be used. The thesis begins with a description of the sampling methods used in the study. Next, an overview of the literature on the relationship between air pollution and respiratory health, followed by a review of the methods of analyses appropriate for longitudinal diary studies of this nature. The methods and results are then presented for the analyses of the association between the three main outcomes of interest – evening peak flow, day cough and bronchodilator use – and air quality variables: particulates (PM10 and TSP), sulphur dioxide, pollens and fungi, using three modelling approaches. These include a representative of data reduction methods (Aggregate analysis), subject-specific or mixed-model methods (Korn-Whittemore analysis) and marginal methods (Generalised Estimating Equations). All estimates were adjusted for climate-related covariates and trend. The final chapter discusses the advantages and disadvantages of the various methods of analyses, and a recommendation for analytic techniques for further studies.