Globally urban air pollution is recognised as an important environmental risk and a major focus of public health concern in developed and developing countries. A comprehensive study sought to establish whether ambient air pollution levels should be of public health concern in Auckland, New Zealand. Urban airshed modelling was applied over a relatively large area at a fine-scale resolution and GIS-based techniques were adopted to quantify long-term exposure to ambient air pollution levels and measure associated health effects. This involved the translation of 9km2 grid cell resolution annual average nitrogen dioxide (NO2) concentrations to census area unit (CAU) concentrations. Multivariate modelling was used to establish an exposure-response relationship for the effect of air pollution on mortality in the Auckland region for the period 1996 to 1999. After adjusting for age, sex, ethnicity, socio-economic status, and urban/rural domicile there was a 1.3% (95%CI: 1–1.5%) increase in non-external cause mortality, and 1.8% (95%CI: 1.5–2.1%) increase in circulatory and respiratory causes per 1 μg/m3 increase in annual average NO2. The New Zealand Deprivation Index (NZDep96) is a unique opportunity to study any confounding caused by socio-economic status with a small area GIS approach. There was an association between NZDep96 and NO2, as well as NZDep96 and mortality, and our analysis could take such confounding into account. Based on these exposure response relationships and applying an annual average threshold of 13 μg/m3, the average annual (for 1996–1999) number of people estimated to die from non-external causes and circulatory and respiratory causes attributable to air pollution in Auckland is 268 (95% range: 227–310) (3.9% of total all cause deaths) and 203 (95% range: 169–237) (5.9% of total circulatory and respiratory deaths) per year, respectively. The number of attributable deaths found in this study are consistent with a previous New Zealand risk assessment using a different methodology, and is approximately twice the number of people dying from motor vehicle accidents in the region, which is on average (1996–1999) 103 per year. Spatial differences in air pollution levels and exposure, and mortality and population characteristics between CAUs were illustrated using GIS-based exposure maps. The GIS-based exposure maps identify high exposure areas for policy developers and planners in a simple and realistic manner. Taken together with overseas studies the study provides additional evidence that long-term exposure to poor air quality, even at levels below current standards, is a hazard to the public health.
(1) National Institute Of Water & Atmospheric Research Ltd
(2) National Centre for Epidemiology and Population Health, The Australian National University
(3) National Institute of Water & Atmospheric Research Ltd
(4) Department of Community Health, University of Auckland