Full text of DOI:10.1111/j.1467-842X.2002.tb00264.x

Article
Mortality
Placing Aboriginal and Torres Strait Islander
mortality in an international context
Yin Paradies and Joan Cunningham
Abstract
Cooperative Research Centre for Aboriginal and Tropical Health and Menzies
School of Health Research, Northern Territory
Objective: To assess whether the patterns
of mortality observed among Indigenous
Australians were seen in other countries or
sub-populations. Previous reports have
indicated that the life expectancy of
ccording to the few published re- Australian data
Indigenous Australians compares
ports available, the mortality expe-
rience of Indigenous Australians tralians were obtained from the Australian
that is, Aboriginal people and Torres Strait Bureau of Statistics (ABS, special tabulation
Data on the mortality of Indigenous Aus-
unfavourably with that of Indigenous groups
in other developed countries, and is similar
to that in some developing countries.
However, in contrast to many developing
countries, low life expectancy of Indigenous
Australians is the result of relatively high
and early adult mortality, rather than high
infant mortality.
A
(
Islanders) compares unfavourably with that request). To minimise the impact of year-to-
of Indigenous groups in other countries such year variation in the number of deaths and
as New Zealand, the United States and to maximise the accuracy of the Indigenous
Canada.1-5 Estimated life expectancy for population estimates, we selected a three-
Indigenous Australians is similar to that of year period (1995-97) centred on the date of
developing countries, such as Pakistan, the most recent Australian Census (1996).
Methods: Using routinely available
6
Indonesia, Guatemala and Iran. However,
The Indigenous population is a relatively
administrative data on age-specific
in contrast to many developing countries, the small percentage of the total Australian
mortality and estimated life expectancy at
birth, we compared data for Indigenous
Australians (from the Northern Territory,
Western Australia and South Australia
combined) with corresponding data for 200
countries world-wide, as well as for several
population groups of interest, including
African Americans, Native Americans,
Canadian Natives and New Zealand Maori.
Results: Patterns of mortality among
Indigenous Australians were markedly
different to those of most other populations
with available data, with the exception of
the Russian Republic. The age-specific
mortality rates for Russian males in 1990-95
were almost identical to those of Indigenous
males in 1995-97. The similarities among
females were less pronounced, but stronger
than for any other country.
7
low life expectancy of Indigenous Austral- population (2.1% in 1996). As in other
ians is the result of relatively high and early countries with small minority populations,
adult mortality, rather than high infant not all deaths of Indigenous people are iden-
mortality.7
tified as such. The analysis reported here is
Although comparisons of population char- restricted to the three jurisdictions – North-
acteristics across and within countries are ern Territory (NT), WesternAustralia (WA),
limited by the quality and availability of data, and South Australia (SA) – determined by
such investigations can highlight important the ABS to have the most complete identifi-
similarities and differences. Patterns of mor- cation of Indigenous deaths in 1995-97.7
tality and life expectancy reflect a range of Even in these jurisdictions, however, some
economic, social, cultural, environmental under-ascertainment of deaths is likely.
and political factors that vary from place to
place and over time.
The Indigenous population estimates used
in the analysis were provided by the ABS
In order to identify countries or groups (special tabulation request) and are derived
with mortality patterns similar to those of from 1996 Census-based experimental esti-
Indigenous Australians, we compared data mates for the Aboriginal and Torres Strait Is-
for IndigenousAustralians with correspond- lander population in NT, WA and SA
ing figures for a number of countries and combined. Estimates for the Indigenous popu-
Conclusion and implications: The
dramatic fall in Russian life expectancy has
been studied extensively and several
important social and contextual factors
have been identified. These factors are also
relevant for the Indigenous population, and
this may help to explain the similar mortality
patterns of the two groups.
population groups from around the world.
lation are referred to as ‘experimental’because
of the volatility in the Census counts and the
deficiencies in the quality of births, deaths
Methods
We compared age-specific mortality rates and migration data for this population.4
and estimated life expectancy at birth for
Age-specific mortality data for the total
Indigenous Australians with corresponding Australian population were obtained from
8
measures for the populations and sub- published reports. Estimates of Indigenous
(Aust N Z J Public Health 2002; 26: 11-6)
populations of a number of other countries. and total Australian life expectancy at birth
Correspondence to:
Submitted: May 2001
Yin Paradies, Cooperative Research Centre for Aboriginal and Tropical Health, PO Box 41096,
Casuarina, Northern Territory 0811. Fax: (08) 8927 5187; e-mail: yin.paradies@menzies.edu.au
Revision requested: October 2001
Accepted: December 2001
2002 VOL. 26 NO. 1
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
11

Paradies and Cunningham
Article
were calculated using PopulationAnalysis Spreadsheet software,
which constructs life tables from age-specific death rates.9
Americans, adjusted for miscoding of ethnicity on death certifi-
cates, was published by the Indian Health Service.12 These esti-
mates refer to the period 1994-96 and are based on the IHS service
area population, rather than the whole of the United States.
For the New Zealand Maori population, life expectancy at birth
for the period 1995-97 was available on-line from Statistics New
Other national-level and regional data
Age-specific death rates and estimated life expectancy at vari-
ous ages for males and females are published annually in the
United Nations’ Demographic Yearbook for a large number of
countries. The data are obtained from the relevant countries when-
ever possible, but many of the figures are based on the United
Nations’ own estimates. In the 1996 edition, life expectancy at
1
3
Zealand. Maori age-specific mortality rates were calculated us-
ing published age- and sex-specific death registrations for 199614
and the 1996 estimated resident population by age and sex.15
Estimated life expectancy at birth for 1996 was available for
the Canadian Registered Indian population (Indian and Northern
Affairs Canada, unpublished data), but no age-specific death rates
could be obtained.
6
birth was available for 200 countries. For the majority of these
200 countries, the latest available data were for a single year or
group of years 1990-95 but for some countries the latest data were
from the late 1980s. Age-specific death rates (in five-year age
groups) were also available for 93 of the 200 countries, although
some countries were missing data for some age groups.
Results
Life expectancy at birth
Regional estimates of age-specific death rates for 1990 were
Figures 1 and 2 show the distribution of estimated life expect-
ancy at birth for males and females, respectively, for all countries
included in the analysis as well as for the sub-populations
described above.
1
0
available from the Global Burden of Disease project. The eight
regions were described as: Established Market Economies (EME);
Formerly Socialist Economies of Europe (FSE); India (IND);
China (CHN); OtherAsia and Islands (OAI); Sub-SaharanAfrica
For males, life expectancy at birth ranged from 22.1 years in
Rwanda (1990-95) to 76.5 years in Iceland (1994) (see Figure 1).
Life expectancy at birth for IndigenousAustralian males in 1995-
(
SSA); LatinAmerica and the Caribbean (LAC); and Middle East-
ern Crescent (MEC).
9
7 was 60.0 years, approximately 16 years lower than the figure
Other sub-populations of interest
for all-Australian males in 1996 (76.2 years).
For the African American and Native American populations of
the United States, age-specific mortality rates by sex for 1996
were obtained from published reports of the US National Center
for Health Statistics.11 African American life expectancy at
birth was also available for 1996.11 Life expectancy for Native
Female life expectancy at birth ranged from 23.1 years in
Rwanda (1990-95) to 82.9 years in Japan (1995) (see Figure 2).
For Indigenous females, the estimated life expectancy at birth in
1995-97 of 67.6 years was about 15 years lower than that for all-
Australian females in 1996 (82.7 years).
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
Figure 1: Estimated life
Au
expectancy at birth, countries
C
M N
Af
with available data and selected
population groups, males.
I
Abbreviations: I = Indigenous Australian, A f = African Am erican,
M = NZ M aori, N = Native A m erican, C = Canadian Native, Au = all-Australian.
Data sources: UN Departm ent of Econom ic and Social Affairs 1998 (6); A BS 1999 (8); US B ureau of Census 2000 (9); Peters et al. 1998 (11); Indian
Health Service 1999 (12); Statistics New Zealand 2000 (13) ; A BS, unpublished data; Indian and Northe rn Affairs Canada, unpublished data.
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
A u
Figure 2: Estimated life
C
Af
N
expectancy at birth, countries
with available data and selected
population groups, females.
M
I
Abbreviations: I = Indigenous Australian, Af = African Am erican,
M
=
NZ M aori,
N
=
Native Am erican,
C
=
Canadian Native, Au = all-Australian.
Data sources: UN Depart ment of Eco n o mic and Social Affairs 1998 (6); ABS 1999 (8); US Bureau of Census 2000 (9); Peters et al. 1998 (11); Indian
Health Service 1999 (12); Statistics New Zealand 2000 (13); ABS, unpublished data; Indian and Northern Affairs Canada, unpublished data.
12
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
2002 VOL. 26 NO. 1

Mortality
Aboriginal andTorres Strait Islander mortality
For both males and females, life expectancy was lower for
Indigenous Australians than for New Zealand Maori, African
Americans, Native Americans and Canadian Natives.
African American populations. While there were some similari-
ties in the overall shapes of the mortality curves, the rates were
considerably higher in adulthood for IndigenousAustralians than
for other Indigenous and non-Indigenous sub-populations in the
US and New Zealand.
Age-specific mortality rates
Figures 3 and 4 show Indigenous age-specific mortality rates in
Of all the nations examined, the Newly Independent States of
the former USSR were the most similar to Indigenous Austral-
ians in their age-specific mortality patterns. Among males the
Russian age-specific mortality rates were the closest match to the
Australian Indigenous rates and were, in fact, virtually identical
(see Figure 7). The similarities were not as pronounced among
females (see Figure 8), but the mortality pattern for Russian
females matched that of Indigenous Australian females more
closely than any other national population examined.
comparison with figures for the eight world regions as defined by
10
Murray and Lopez in the Global Burden of Disease Project. These
graphs show the unusual mortality profile of the Indigenous Aus-
tralian population. For both males and females, Indigenous mortal-
ity rates at younger ages are low by world standards but rise by
middle age to be higher than any of the regional estimates.
To determine whether the IndigenousAustralian pattern of rela-
tively low infant and childhood mortality rates followed by high
mortality in the young and middle adult years was present in other
populations, we examined the age-specific mortality rates of all
national and sub-populations with available data and compared
them with the corresponding rates for Indigenous Australians.
Figures 5 and 6 present age-specific mortality rates for the
Australian Indigenous, NativeAmerican, New Zealand Maori and
Discussion
There is an enormous gap in life expectancy between the Indig-
enous population of NT, WA, and SA and the totalAustralian popu-
lation for both males and females. Life expectancy at birth for
Indigenous Australians is also considerably lower than the
Figure 3: Age-specific
mortality rates for males,
by Global Burden of
Disease region (1990)
and for Indigenous
Australians (1995-97).
1
0000
10
1
000
1
00
1
0
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75+
Sub-Saharan Africa
India
Latin America and the Caribbean
Former Socialist Economies of Europe
Indigenous Australian
Other Asia and Islands
Middle Eastern Crescent
China
Established Market Economies
Data sources: Murray and Lopez 1996 (10); ABS, unpublished data.
Figure 4: Age-specific
mortality rates for
females, by Global
1
0000
Burden of Disease region
(1990)10 and for
1
000
Indigenous Australians
(
1995-97).
1
00
1
0
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75+
Sub-Saharan Africa
India
Latin America and the Caribbean
Other Asia and Islands
Middle Eastern Crescent
China
Former Socialist Economies of Europe
Indigenous Australian
Established Market Economies
Data sources: Murray and Lopez 1996 (10); ABS, unpublished data.
2002 VOL. 26 NO. 1
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
13

Paradies and Cunningham
Article
corresponding figures for other Indigenous and non-Indigenous sub-
populations in New Zealand, the United States and Canada.
The Australian Indigenous population has a mortality profile
that is distinct from the profiles of most national populations
worldwide and differs from the profiles of the eight regions de-
Although the results described here should not be considered pre-
cise, it is unlikely that the broad trends observed in this analysis
are explained by limitations in the quality and availability of data.
Several reasons have been suggested for the higher mortality
rates among Indigenous Australians compared with other
Indigenous groups world-wide. These include social, economic,
political and environmental factors such as the lack of a treaty in
Australia, greater barriers to accessing health services, the lack
of community-controlled health services in many areas, inadequate
resources and infrastructure, poor environmental health and the
relative lack of a skilled Indigenous workforce.2,3,16 It should be
noted that the analysis of Australian Indigenous mortality was
limited to NT, WA, and SA because of concerns about data qual-
ity in other jurisdictions during the time period of interest. These
three jurisdictions accounted for about 34% of the Indigenous
poulation in 1996 but 62% of all deaths identified as Indigenous
1
0
fined for the Global Burden of Disease Project. Indigenous age-
specific mortality was also higher than that of other Indigenous
and non-Indigenous sub-populations of interest, especially in
adulthood. However, there was a striking similarity between In-
digenous and Russian mortality patterns, especially among males.
Any comparison of mortality patterns between and within coun-
tries is limited by the quality and availability of data. The com-
pleteness of coverage of death registrations is variable and there
can be important differences in, for example, the definition and
identification of population subgroups, the accuracy of popula-
tion estimates, and the years for which data are available. Group-
ing of data across large regions may obscure variations in mortality
patterns for individual countries and for smaller regions, and in-
formation is not available for some countries. Interpretations of
comparative measures must therefore be made with caution.
4
in 1995-97. The mortality experience among those identified as
Indigenous in death records appears to be similar across all juris-
4
dictions but the extent to which deaths identified as Indigenous
reflect the overall mortality experience of Indigenous people is
Figure 5: Age-specific
10000
mortality rates for
selected population
groups, males.
1000
100
10
0
1
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75+
Age group
Indigenous Australian (1995-97)
NZ Maori (1996)
African American (1996)
Native American (1996)
Data sources: ABS, unpublished data; Peters et al. 1998 (11); NZ Health Information Service 1999 (14); Statistics New Zealand 2000 (13).
Figure 6: Age-specific
mortality rates for
selected population
groups, females.
1
0000
1
000
1
00
1
0
0
1
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75+
Age group
Indigenous Australian (1995-97)
NZ Maori (1996)
African American (1996)
Native American (1996)
Data sources: ABS, unpublished data; Peters et al. 1998 (11); NZ Health Information Service 1999 (14); Statistics New Zealand 2000 (13).
14
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
2002 VOL. 26 NO. 1

Mortality
Aboriginal andTorres Strait Islander mortality
unknown. Efforts to improve the quality of Indigenous mortality
data throughout Australia are under way and progress is being
The reduction in life expectancy and increase in mortality rates
were not as dramatic for Russian women compared with men,
and it has been suggested that traditional gender roles provided
some protection for women.20,21 Leon et al. have concluded that
data quality problems, such as deterioration in completeness of
death reporting or under-estimation of the population at risk, were
not significant contributing factors to the apparent mortality
increase.22
7
made. This should allow the inclusion of other jurisdictions in
future analyses.
The similarity in mortality rates for Indigenous males in 1995-
9
7 and Russian males in 1995 warrants further consideration.
Former Soviet Union countries including Russia have undergone a
great deal of turmoil and change since the early 1990s. These
changes in the social, economic and political arenas have been
accompanied by marked changes in health status. In the four-year
period 1990-94, for example, Russian life expectancy at birth fell
Indigenous Australians are exposed to many of the same factors
that appear to have contributed to the decline in health and increased
mortality in Russia, and this may explain in part the similarity of
their mortality patterns. Indigenous Australians are disadvantaged
relative to otherAustralians with respect to education, employment,
and income and are more likely to be exposed to poor living condi-
tions, poor nutrition, tobacco, excessive alcohol consumption, other
17
by a massive five years. Mortality rates rose for every age group,
but the largest increases were in the middle-aged groups. More
than 75% of the total decline in life expectancy resulted from in-
creased mortality rates among persons ages 25 to 64 years.17
Many factors appeared to be operating simultaneously to
produce this sudden change in mortality in Russia, including eco-
nomic and social instability, low employment and reduced in-
come, high rates of tobacco and alcohol consumption, poor
nutrition, stress, and deterioration of the health care system.17-19
7
drugs and harmful substances, and violence. They also experience
high levels of stress due to factors such as separation (especially of
children), trauma, grief, loss of culture, and ongoing effects of
poverty, discrimination and racism.7,23
According to Cockerham, health lifestyles in the Soviet Union
10000
Figure 7: Age-specific
mortality rates for
Russian, Indigenous and
all-Australian males.
1000
100
10
0
1
5
10
15
20
25
30
Age group
Russian (1995)
35
40
45
50
55
60
65
70
Indigenous Australian (1995-97)
all-Australian (1996)
Data sources: ABS, unpublished data; ABS 1999 (8); UN Department of Economic and Social Affairs 1998 (6).
10000
Figure 8: Age-specific
mortality rates for
Russian, Indigenous and
all-Australian females.
1000
100
10
0
1
5
10
15
20
25
30
Age group
Russian (1995)
Data sources: ABS, unpublished data; ABS 1999 (8); UN Department of Economic and Social Affairs 1998 (6).
35
40
45
50
55
60
65
70
Indigenous Australian (1995-97)
all-Australian (1996)
2002 VOL. 26 NO. 1
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
15

Paradies and Cunningham
Article
were influenced more by ‘life chances’ (e.g. income, education
References
1
8
and other structural factors) than by ‘life choices’. The socialist
system created an environment in which primary responsibility
for the health of the people was vested in the central government.
Individuals were not encouraged to take responsibility for their
own health and were passive recipients of largely curative health
care services, rather than active participants in preventing ill-
health. People were effectively disempowered, and when the sys-
tem collapsed, a void was created.
1. Hogg R. Indigenous mortality: placingAustralianAboriginal mortality within
a broader context. Soc Sci Med 1992;35(3):335-46.
2
3
.
.
Kunitz SJ. Disease and Social Diversity: The European Impact on the Health
of Non-Europeans. New York: Oxford University Press, 1994.
Ring I, Firman D. Reducing indigenous mortality in Australia: lessons from
other countries. Med J Aust 1998;169:528-33.
4. Cunningham J, ParadiesY. Mortality of Aboriginal and Torres Strait Islander
Australians. Canberra:Australian Bureau of Statistics, 2000. Cat. No.: 3315.0.
5
.
Ross K, Taylor J. Improving life expectancy and health: A comparion of Aus-
tralia’s Aboriginal and Torres Strait Islander People and New Zealand Maori.
J Popul Res. In press 2001.
6
7
.
.
United Nations Department of Economic and Social Affairs. Demographic
Yearbook 1996. New York: United Nations, 1998.
Australian Bureau of Statistics. The Health and Welfare of Australia’s Abo-
riginal and Torres Strait Islander Peoples 1999. Canberra: ABS, 1999. Cat.
No.: 4704.0.
InAustralia, so-called welfare dependency has been associated
with continuing loss of roles, loss of mastery and hopelessness
among Indigenous people,24 and disempowerment has been ex-
plicitly recognised by some health practitioners as the fundamen-
tal cause of the poor health of Indigenous Australians.25 These
factors, and those identified in Russia, are consistent with the
growing literature on the social determinants of health.26
8
9
1
.
.
Australian Bureau of Statistics. Deaths, Australia, 1998. Canberra:ABS, 1999.
Cat. No.: 3302.0.
Population Analysis spreadsheet software LTMXQXAD [computer program].
Washington: United States Bureau of Census, 2000.
0. Murray CJL, LopezAD. Global health statistics: A compendium of incidence,
prevalence, and mortality estimates for over 200 conditions. In: Global Bur-
den of Disease and Injury Series, Vol II. Boston: Harvard School of Public
Health, 1996.
There is recent evidence that the decreases in life expectancy
in Russia in the late 1980s and early 1990s were largely erased in
the middle part of the 1990s. Shkolnikov and others have reported
that life expectancy in 1998 was back to what it had been in 1984.27
This suggests that large positive as well as large negative changes
in life expectancy are possible over relatively short periods of
time. Recent data suggest that mortality is on the rise again, how-
ever, with female and male life expectancy decreasing by one and
two years respectively between 1998 and 2000.27,28
1
1. Peters K, Kochanek K, Murphy S. Deaths: Final Data for 1996. Maryland:
National Center for Health Statistics, 1998. National Vital Statistics Reports
No.: 47(9).
1
1
2. Indian Health Service. Regional Differences in Indian Health. Maryland: US
Department of Health and Human Services, 1999.
3. Statistics New Zealand. New Zealand Life Tables (1995-97). Wellington: Sta-
tistics New Zealand, 2000. Cat. no. 03.301.IN97. Available from: URL:
http://www.stats.govt.nz/domino/external/pasfull/pasfull.nsf/
d81599c2f96f306a4c2567ed00094120/26885cd2e5daf3604c2566d
7
00096988. Accessed 30 August 1999.
4. New Zealand Health Information Service. Mortality and Demographic Data
996. Wellington: Ministry of Health, 1999.
5. Statistics New Zealand. Maori Population Estimates Alltabls5.xls.
Wellington(NZ): Statistics New Zealand, 2000. Available from: URL: http://
www.stats.govt.nz/domino/external/web/prod_serv.nsf/874ea91c142289384
c2567a80081308eaf37c00481e963b9cc256b120074d2bb. Accessed 22 July
1
1
1
The implications of the Russian experience for Indigenous
Australians remain somewhat uncertain. It seems likely that fun-
damental social change will ultimately be required to achieve sig-
nificant long-term and sustainable improvements in the health
and longevity of Indigenous Australians. This may seem on the
surface to be an insurmountable task, but it need not be so. As
one author has noted, ‘Social problems seldom get solved,
because people define these problems in ways that overwhelm
their ability to do anything about them’.29 The alternative is to
redefine a large problem into a series of smaller, more manage-
able problems and to aim for ‘small wins’that eventually add up.
According toWeick, this approach can help to increase optimism,
stimulate innovation, attract allies and resources, and produce a
climate conducive to generating additional small wins,29 all of
which are sorely needed if we are to achieve lasting improvement
in the health and well-being of Indigenous Australians.
1
999.
16. Ross K. Recent trends in the demography of the indigenous populations of
Australia, New Zealand, Canada and the United States of America. Aborigi-
nal Hist. In press 2001.
17. Notzon F, KomarovY, Ermakov S, Sempos C, Marks J, Sempos, E. Causes of
declining life expectancy in Russia. J Am Med Assoc 1998;279(10):793-800.
1
8. Cockerham WC. Health lifestyles in Russia. Soc Sci Med 2000;51:1313-24.
19. Leon D, Shkolnikov V. Social stress and the Russian mortality crisis. J Am
Med Assoc 1998;279(10):790-1.
2
0. Watson P. Explaining rising mortality among men in Eastern Europe. Soc Sci
Med 1995;41:923-34.
2
2
1. Weidner G. Gender gap in health decline in East Europe. Nature 1998;395:835.
2. Leon DA, Chenet L, Shkolnikov VM, Zakharov S, Shapiro J, Rakhmanova
G. Huge variation in Russian mortality rates 1984-94: Artefact, alcohol or
what? Lancet 1997;350:383-8.
2
2
2
3. Human Rights and Equal Opportunity Commission. ‘Bringing Them Home’–
[
report]. Sydney: Human Rights and Equal Opportunity Commission, 1997.
4. Trudgen RI. Why Warriors Lie Down and Die. Darwin: Aboriginal Resource
and Development Services, 2000.
5. Royal Australasian College of Physicians Cottrell Conference 1997
[
communiqué]. Statement on the delivery of specialist services to remote and
rural Aboriginal and Torres Strait Islander communities. Royal Australasian
College of Physicians conference, Darwin, NT, November 1997. Available
from: URL: http://www.racp.edu.au/indig/cottrell.htm. Accessed 20 January
Acknowledgements
Joan Cunningham is supported by a fellowship from the
Menzies Foundation. Some of the analysis was performed while
the authors were at the National Centre forAboriginal and Torres
Strait Islander Statistics (Australian Bureau of Statistics). The
views expressed are those of the authors and do not necessarily
represent those of the Australian Bureau of Statistics.
2
001.
2
2
6. Marmot M, Wilkinson RG. Social Determinants of Health. New York: Ox-
ford University Press, 1999.
7. Shkolnikov V, McKee M, Leon DA. Changes in life expectancy in Russia in
the mid-1990s. Lancet 2001;357:917-21.
28. World Health Organization. The World Health Report 2001: Mental Health:
New Understanding, New Hope. Geneva:WHO, 2001.
2
9. Weick KE. Small wins: Redefining the scale of social problems. Am Psychol
984;39(1):40-9.
1
16
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH
2002 VOL. 26 NO. 1