6 Health Effects of Particles in New Zealand
An estimate of the impact of existing PM10 concentrations on total mortality in New Zealand has been made by Fisher et al (2002) based on the results of the Kunzli (1999) study. This includes estimates of effects not detected using the standard time series epidemiological studies. Thus the number of deaths estimated for Christchurch is considerably greater than the 40-70 estimated by Wilton (1999) for the acute impacts associated with the standard time series study design. Similarly estimates of the impact of PM10 concentrations on mortality in Nelson will be about four times greater than those estimated in Wilton (2001) if the chronic and other acute effects are included.
Although there are some uncertainties with the mortality estimates of Fisher et al (2002) associated with the exposure estimates and allocation to sources, these are based on the best available information. Further improvements in these estimates without considerable investment in PM10 monitoring and exposure assessment are unlikely.
Table 6.1 collates the existing health impact estimates for PM10 in New Zealand and includes additional estimates of morbidity impacts in the main urban areas. These estimates are based on the standard time series studies and therefore include only the acute health impacts that occur shortly after the pollution event. The mortality estimates in this risk assessment are much higher than the hospitalisation because they also include estimates of impacts that do not occur directly following the pollution event. Thus the hospitalisation estimates are likely to be extremely conservative. Estimates of health endpoints from this study have been rounded to the nearest five deaths and hospitalisations and the nearest 10,000 for RAD. These data are based on the following assumptions and should be treated as indicative only.
- Measured PM10 concentrations in these areas are indicative of average exposure across the urban areas.
- Increases in hospitalisations associated with PM10 concentrations as per the McGowan et al (2002) study for Christchurch.
- Increase in RAD of approximately 91,200 RAD each year per million of population for every 1 µgm-3 increase in annual average PM2.5 (AmericanLung Association, 1995).
- Estimates of the relationships between PM10 and PM2.5 in areas where these are not known.
- No threshold below which effects do not occur.
Table 6.1: Estimates of health impacts of particle concentrations in New Zealand
View estimates of health impacts of particle concentrations in New Zealand (large table)
The estimates of mortality for Christchurch and Nelson shown in Table 6.1 are greater than the values indicated in Wilton (1999) and (2002) of 40-70 and 14 respectively. This is because the former estimates were based on the time series methodology, which is unlikely to estimate the reduced life expectancy associated with long-term morbidity enhanced by air pollution. Estimates included in this study are based on the relationships described by Kunzli et al (2000).
Figure 6.1 provides an illustration of estimates of the impacts of existing PM10 concentrations on mortality, hospital admissions and restricted activity days in the main urban areas of New Zealand. These data, from Table 6.1, are based on assumptions relating to existing PM10 and PM2.5 exposure and health endpoints and should be treated as indicative only.
