Atmosphere Theme Report

Australia State of the Environment Report 2001 (Theme Report)
Lead Author: Dr Peter Manins, Environmental Consulting and Research Unit, CSIRO Atmospheric Research, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06746 9

Urban Air Quality (continued)

Indicators of policy response (continued)

Mortality and morbidity from respiratory disease [A Indicator 3.19]

  • Implications
  • -->

    Studies have been conducted on the relationship between air quality and mortality in Sydney (Morgan 2000), Brisbane (Simpson et al. 2000) and Melbourne (EPAV 2000b). The results indicate that in Brisbane and Sydney, short-term mortality (i.e. respiratory deaths not due to cancer) is related to increases in concentrations of particulate matter and ozone. The situation in Melbourne is different: in Melbourne, the statistics indicate that short-term mortality is related to nitrogen dioxide and ozone.

    The Air NEPM provides estimates of the annual short-term health effects of the criteria pollutants:

    • CO - loss of one day of earning for 50 000 people at a cost of $6 million (NEPC 1998, p. 52)
    • NO2 - 10 to 15% of the population display respiratory symptoms at a cost of $5 million (NEPC 1998, p. 61)
    • O3 - up to 10 deaths per year in Australia, with total costs up to $810 million (NEPC 1998, p. 75-76)
    • PM10 - up to 2400 deaths per year in Australia, with an associated health cost of $17.2 billion (NEPC 1998, pp. 122 & 127).

    In addition, hydrocarbons have long-term health effects that have been examined by Hearn (1995) for Melbourne. If his figures are extrapolated to all of Australia then there are about 1250 to 1785 deaths per year as a result of hydrocarbons (excluding deaths ascribed to the particulate matter in the hydrocarbons).

    The main health risk for Australians, in terms of mortality, arises from particulate matter and from hydrocarbons (Beer 2000).

    Basic statistics for Melbourne mortality (Table 26) show that there is substantial age variability in deaths as a result of respiratory diseases.

    Table 26: Mean daily deaths by cause in Melbourne, January 1991 to August 1996 (n=2070)
    Deaths(age group) Whole study period April-Oct. Nov.-March
    Mean s.d. Min. Max. Mean s.d. Min. Max. Mean s.d. Min. Max.
    Cardiovascular
    0-65 2.7 1.7 0 11 2.8 1.7 0 11 2.5 1.5 0 9
    65+ 21.6 5.1 7 41 22.9 4.9 8 39 19.7 4.7 7 41
    Total 24.3 5.4 8 43 25.7 5.3 9 43 22.3 4.8 8 43
    Respiratory
    0-65 0.5 0.7 0 4 0.6 0.7 0 4 0.4 0.6 0 3
    65+ 4.0 2.2 0 15 4.4 2.3 0 15 3.4 1.9 0 10
    Total 4.5 2.3 0 16 4.9 2.4 0 16 3.8 2.0 0 11
    All deaths
    0-65 10.9 3.4 1 25 11.1 3.3 1 25 10.6 3.4 1 23
    65+ 44.5 7.9 20 73 46.7 7.6 22 73 41.2 7.0 20 71
    Total 55.3 8.6 31 90 57.8 8.4 32 83 51.8 7.6 31 90

    Source: EPAV (2000b).

    Mortality as a result of respiratory diseases and respiratory cancers has been falling for males, but rising for females, since 1979 (Figure 117) which, presumably, mirrors the decline in cigarette smoking for males and the uptake in cigarette smoking for females. The peaks apparently represent influenza outbreaks. Studies to relate air quality and health variables need, first, to allow for major effects on mortality before examining the more subtle effects induced by air pollution. If Australian mortality rates due to cancer and diseases of the respiratory system are compared with those of other countries (Figure 118), it appears that respiratory deaths in most countries are dominated by short-term mortality.

    Figure 117: Standardised death rate due to respiratory diseases and respiratory cancers.

     Standardised death rate due to respiratory diseases and respiratory cancers

    Source: ABS (1996b)

    Figure 118: Standardised death rates for diseases of the respiratory system and cancers of the respiratory system for Australia and other OECD countries.

     Standardised death rates for diseases of the respiratory system and cancers of the respiratory system for Australia and other OECD countries

    Source: Based on data in ABS (1996b)

    Implications

    Graphs of the mortality data confirm that attempts to establish relationships between mortality data and air quality variables need to be based on large sample sizes because the subtle relationships are swamped by confounders such as smoking patterns and seasonal temperature variations. This illustrates the importance of good information on personal exposure as a data input for epidemiological studies. Such information needs to include data on personal habits (smoker or non-smoker) and the home environment (fuel used for heating and cooking) as well as information on the pollutant exposure at work, at home, during travel and during outdoor activity.