THE COST OF INACTION ON THE SOCIAL DETERMINANTS OF HEALTH

REPORT NO. 2/2012
STRICTLY EMBARGOED UNTIL  1AM (AEST), JUNE 4, 2012

CHA-NATSEM Second Report on Health
Inequalities

PREPARED BY Laurie Brown, Linc Thurecht and Binod Nepal

PREPARED FOR Catholic Health Australia

MAY 2012

ABOUT NATSEM
The National Centre for Social and Economic Modelling (NATSEM), a research centre at
the University of Canberra, is one of Australia’s leading economic and social policy research
institutes, and is regarded as one of the world’s foremost centres of excellence for
microsimulation, economic modelling and policy evaluation.
NATSEM undertakes independent and impartial research, and aims to be a key contributor to
social and economic policy debate and analysis in Australia and throughout the world through
high quality economic modelling, and supplying consultancy services to commercial,
government and not-for-profit clients. Our research is founded on rigorous empirical analysis
conducted by staff with specialist technical, policy and institutional knowledge.
Research findings are communicated to a wide audience, and receive extensive media and public
attention. Most publications are freely available and can be downloaded from the NATSEM
website.

Director: Alan Duncan
© NATSEM, University of Canberra

All rights reserved. Apart from fair dealing for the purposes of research or private study, or
criticism or review, as permitted under the Copyright Act 1968, no part of this publication may be
reproduced, stored or transmitted in any form or by any means without the prior permission in
writing of the publisher.

National Centre for Social and Economic Modelling

University of Canberra ACT 2601 Australia

Building 24, University Drive South, Canberra University, Bruce, ACT 2620

Phone + 61 2 6201 2780

Fax + 61 2 6201 2751

Email natsem@natsem.canberra.edu.au

Website www.natsem.canberra.edu.au

CONTENTS

About NATSEM i
Acknowledgements v
General caveat v
Abbreviations and Acronyms vi
Foreword vii
Executive Summary ix
1 Introduction 1
1.1 Objectives of this Report 2
1.2 Structure of this Report 3
2 Measuring Health and Socio-Economic Disadvantage 3
2.1 Key Health and Socio-Economic Indicators 3
2.2 Measuring Lost Benefits – the Costs of Inaction 4
2.3 Missing Data 6
2.4 Profile of the Study Population 7
3 How Many Disadvantaged Australians of Working Age Are Experiencing Health Inequity? 8
4 Costs To Well-Being - Potential Gains in Satisfaction With Life 11
5 Lost Economic Benefits – Potential Economic Gains From Closing Health Gaps 13
5.1 Potential Gains in Employment 13
5.2 Income and Gains in Annual Earnings 17
5.3 Government Pensions and Allowances and Savings in Government Expenditure 20
6 Savings To The Health System From Closing Health Gaps 24
6.1 Reduced Use of Australian Hospitals 24
6.2 Reduced Use of Doctor and Medical Related Services 26
6.3 Reduced Use of Prescribed Medicines 27
7 Summary and Conclusions 32
References 35
Appendix 1 - Technical Notes 37

Boxes, figures and tables
Table 1 Socio-economic and health domains and variables 4
Table 2 Socio-economic classification 4
Table 3 Outcome measures 5
Table 4 Per cent distribution of men and women aged 25-64 years by selected socio-
economic characteristics 7
Table 5 Inequality in self-assessed health status – potential increase in numbers of most
disadvantaged Australians reporting good health through closing the health
gap between most and least disadvantaged Australians of working age 9
Table 6 Inequality in long-term health conditions – potential increase in numbers of
most disadvantaged Australians reporting no long-term health conditions
through closing the health gap between most and least disadvantaged
Australians of working age 10
Table 7 Percentage disadvantaged persons satisfied with life by health status and
increase in those satisfied through closing the health gap between most and
least disadvantaged Australians of working age 11
Table 8 Percentage persons satisfied with life by presence of a long-term health
condition and increase in those satisfied through closing the health gap
between most and least disadvantaged Australians of working age 12
Table 9 Distribution of employment status among most disadvantaged groups by
health status 14
Table 10 Distribution of employment status among most disadvantaged groups by
prevalence of long-term health conditions 15
Table 11 Difference in employment between those with good and poor health status and
change in employment status from closing the health gap between most and
least disadvantaged Australians of working age 16
TABLE 12 Difference in employment between those without and with a long-term health
condition and change in employment status with reduction in prevalence of
chronic illness from closing the health gap between most and least
disadvantaged Australians of working age 17
Table 13 Weekly gross income from wages and salaries (2008) and increase in annual
earnings from improved health status from closing the health gap between
most and least disadvantaged Australians of working age 19
Table 14 Weekly gross incomes from wages and salaries (2008) and increase in annual
earnings from reduction in prevalence of long-term health conditions from
closing the health gap between most and least disadvantaged Australians of
working age 20
Table 15 Government pensions and allowances per annum (2008) for those in poor and
good health and savings in government welfare expenditure from improved
health from closing the health gap between most and least disadvantaged
Australians of working age 22
Table 16 Government benefits and transfers per annum (2008) for those with and
without a long-term health condition and savings in government welfare
expenditure from reduction in prevalence of long-term health conditions from
closing the health gap between most and least disadvantaged Australians of
working age 23
Table 17 Hospitalisation in 2008 for Australians of working age in the bottom income
quintile and reductions in persons hospitalised through closing the health gap
between most and least disadvantaged Australians of working age 25
CHA-NATSEM Second Reporton Health Inequalities, May 2012
Table 18Estimated number of hospital separations in 2008 for Australiansofworking
age inthe bottom income quintileandreductions in personshospitalisedthrough closing the health gapbetween most and least disadvantaged
Australians of working age25Table 19Average length of hospital stay in2008 for Australians of workingagein thebottomincomequintile andreductions in patientdaysstay through closing thehealthgapbetween most and least disadvantaged Australiansofworkingage26Table 20Estimated number ofdoctor and medically related services used in 2008 byAustralians of working age in the bottom income quintile andreductionsin
MBS services through closingthe health gapbetween most and leastdisadvantaged Australians of workingage27Table 21Estimated MBS benefitsin 2008 for Australians of working age in thebottom
income quintile and savings in MBS benefits through closingthe health gapbetweenmostand least disadvantagedAustralians ofworking age27Table 22Estimated number of PBSscripts used in 2008 by Australiansof working age inthe bottom income quintile andreductionsin PBS scriptvolume through
closing the healthgapbetween most and leastdisadvantaged Australians of
working age29Table 23Comparison of MediSim and Medicare Australia average costs of PBS scripts30Table 24Estimated Government expenditure onPBS medicines in 2008 for Australiansof workingage in the bottomincome quintile and savings inbenefits throughclosing the healthgapbetween most and leastdisadvantaged Australians ofworking age31Table 25Estimated patient co-paymentsto PBS medicines in 2008 by Australians ofworking age in the bottom income quintile and savings in PBS patient coststhrough closing the health gapbetween most and least disadvantaged
Australians of working age31Figure 1Additional number ofmost disadvantaged Australians who wouldbe free oflong-term health conditions if the health gapbetween most and least
disadvantaged Australians of workingage wasclosed.xFigure 2Percentage ofdisadvantaged persons of working age satisfied with life byhealth statusxFigure 3Expectedincrease in numbers employedthrough areduction in the prevalence
of chronicillness from closing the healthgapbetween most and leastdisadvantaged Australians of workingagexiFigure 4Expectedincrease in annual earnings from wages and salaries through eitheranimprovement in self-assessed health status (SAHS)or areduction in the
prevalence of long-term health conditions (LTC) from closing the health gapbetweenmostand least disadvantagedAustralians ofworking agexii

AUTHOR NOTE
Laurie Brown is a Professor and Research Director (Health), Dr Linc Thurecht is a Senior
Research Fellow and Dr. Binod Nepal is a Senior Research Fellow at the National Centre for
Social and Economic Modelling, University of Canberra.

ACKNOWLEDGEMENTS
The authors would like to acknowledge Martin Laverty, Chief Executive Officer and Liz
Callaghan, Director Strategic Policy, of Catholic Health Australia for their support of the project.

This paper uses unit record data from the Household, Income and Labour Dynamics in Australia
(HILDA) Survey. The HILDA Project was initiated and is funded by the Australian Government
Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) and is
managed by the Melbourne Institute of Applied Economic and Social Research (MIAESR). The
findings and views reported in this paper, however, are those of the authors and should not be
attributed to either FaHCSIA or the MIAESR.

ABBREVIATIONS AND ACRONYMS
ABS Australian Bureau of Statistics
AIHW Australian Institute of Health and Welfare
ALOS Average Length of Stay
CSDH Commission on Social Determinants of Health
Disadv. Disadvantaged
HILDA Household Income and Labour Dynamics in Australia survey
IRSD Index of Relative Socio-economic Disadvantage
LTC Long-term Health Condition
MBS Medicare Benefits Schedule
NATSEM National Centre for Social and Economic Modelling
NHMRC National Health and Medical Research Council
NILF Not in Labour Force
PBS Pharmaceutical Benefits Scheme
SAHS Self-assessed Health Status
SEIFA Socio-Economic Indexes for Areas
vs. versus
WHO World Health Organisation

FOREWORD
Half a million Australians could be freed from chronic illness, $2.3 billion in annual hospital costs could be
saved and the number of Pharmaceutical Benefits Scheme prescriptions could be cut by 5.3 million
annually.

These staggering opportunities are what new approaches to health policy could achieve, yet counter-
intuitively they do not require radical change to the way in which our health system operates. In fact, the
opportunity to reduce chronic illness and save on hospital and pharmaceutical expenditure requires
action outside of the formal health system.

Australia suffers the effects of a major differential in the prevalence of long-term health conditions. Those
who are most socio-economically disadvantaged are twice as likely to have a long-term health condition
than those who are the least disadvantaged. Put another way, the most poor are twice as likely to suffer
chronic illness and will die on average three years earlier than the most affluent.

International research points to the importance of factors that determine a person’s health. This research,
centred on the social determinants of health, culminated in the World Health Organisation making a
series of recommendations in its 2008 Closing the Gap Within a Generation report. The recommendations
of that report are yet to be fully implemented within Australia.

Drug-, alcohol-, tobacco- and crisis-free pregnancies are understood to be fundamental to a child’s
lifelong development. So, too, is early learning that occurs in a child’s first three years of life. School
completion, successful transition into work, secure housing and access to resources necessary for
effective social interaction are all determinants of a person’s lifelong health. These are factors mostly
dealt with outside of the health system, yet they are so important to the health of the nation.

Part of Catholic Health Australia’s purpose is improving the health of all Australians, with a particular
focus on the needs of the poor. It’s for this reason NATSEM was commissioned to produce The Cost of
Inaction on the Social Determinants of Health to consider economic dynamics of ignoring the World
Health Organisation’s recommendations for Australia on social determinants of health.

The findings of The Cost of Inaction on the Social Determinants of Health appear to suggest that if the
World Health Organisation’s recommendations were adopted within Australia:

• 500,000 Australians could avoid suffering a chronic illness;
• 170,000 extra Australians could enter the workforce, generating $8 billion in extra earnings;
• Annual savings of $4 billion in welfare support payments could be made;
• 60,000 fewer people would need to be admitted to hospital annually, resulting in savings of $2.3
billion in hospital expenditure;
• 5.5 million fewer Medicare services would be needed each year, resulting in annual savings of $273
million;
• 5.3 million fewer Pharmaceutical Benefit Scheme scripts would be filled each year, resulting in
annual savings of $184.5 million each year.

These remarkable economic gains are only part of the equation. The real opportunity for action on social
determinants is the improvements that can be made to people’s health and well-being.

Australia should seek the human and financial dividends suggested in The Cost of Inaction on the Social
Determinants of Health by moving to adopt the World Health Organisation’s proposals. It can do so by
having social inclusion agendas adopt a “health in all policies” approach to require decisions of
government to consider long-term health impacts.

This research further strengthens the case Catholic Health Australia has been making through the two
reports prepared by NATSEM on the social determinants of health – and the book Determining the
Future: A Fair Go & Health for All published last year – that a Senate Inquiry is needed to better
understand health inequalities in Australia.

No one suggests a “health in all policies” approach is simple, but inaction is clearly unaffordable.

Martin Laverty
Chief Executive Officer, Catholic Health Australia


EXECUTIVE SUMMARY
Key Findings
The findings of the Report confirm that the cost of Government inaction on the social determinants of
health leading to health inequalities for the most disadvantaged Australians of working age is substantial.
This was measured in terms not only of the number of people affected but also their overall well-being,
their ability to participate in the workforce, their earnings from paid work, their reliance on Government
income support and their use of health services.

Substantial differences were found in the proportion of disadvantaged individuals satisfied with their
lives, employment status, earnings from salary and wages, Government pensions and allowances, and use
of health services between those in poor versus good health and those having versus not having a long-
term health condition. Improving the health profile of Australians of working age in the most socio-
economically disadvantaged groups therefore would lead to major social and economic gains with savings
to both the Government and to individuals.

(a) Health inequity
If the health gaps between the most and least disadvantaged groups were closed, i.e. there was no
inequity in the proportions in good health or who were free from long-term health conditions, then an
estimated 370,000 to 400,000 additional disadvantaged Australians in the 25-64 year age group would see
their health as being good and some 405,000 to 500,000 additional individuals would be free from chronic
illness depending upon which socio-economic lens (household income, level of education, social
connectedness) is used to view disadvantage (Figure 1). Even if Government action focussed only on those
living in public housing, then some 140,000 to 157,000 additional Australian adults would have better
health.

(b) Satisfaction with life
People’s satisfaction with their lives is highly dependent on their health status. On average, nearly 30 per
cent more of disadvantaged individuals in good health said they were satisfied with their lives compared
with those in poor health (Figure 2). Over eight in every 10 younger males who had poor health and who
lived in public rental housing were dissatisfied with their lives. If socio-economic inequalities in health
were overcome, then as many as 120,000 additional socio-economically disadvantaged Australians would
be satisfied with their lives. For some of the disadvantaged groups studied, achieving health equality
would mean that personal well-being would improve for around one person in every 10 in these groups.

Figure 1 Additional numbers of most disadvantaged Australians in good health status (SAHS)
or free from long-term health conditions (LTC) from closing the health gap between
most and least disadvantaged Australians of working age


050100150200250300350400450500Male 25-44Male 45-64Female 25-44Female 45-64TotalAge Group (years)
Number
(
'
000)
In Bottom Income Quintile SAHSEarly School Leavers SAHSSocially Excluded SAHSPublic Housing Renters SAHSIn Bottom Income Quintile LTCEarly School Leavers LTCSocially Excluded LTCPublic Housing Renters LTC

Figure 2 Percentage of disadvantaged persons of working age satisfied with their lives by
health status

0102030405060708090100Male 25-44Male 45-64Female 25-44Female 45-64Age Group (years)
Percent
Satisfied
with
Life (
%
)
In Bottom Income Quintile Poor HealthEarly School Leavers Poor HealthSocially Excluded Poor HealthPublic Housing Renters Poor HealthIn Bottom Income Quintile Good HealthEarly School Leavers Good HealthSocially Excluded Good HealthPublic Housing Renters Good Health

(c) Gains in employment
Rates of unemployment and not being in the labour force are very high for both males and females in low
socio-economic groups and especially when they have problems with their health. For example, in 2008,
fewer than one in five persons in the bottom income quintile and who had at least one long-term health
condition was in paid work, irrespective of their gender or age. Changes in health reflect in higher
employment rates, especially for disadvantaged males aged 45 to 64. Achieving equity in self-assessed
health status (SAHS) could lead to over 110,000 new full- or part-time workers when health inequality is
viewed through a household income lens, or as many as 140,000 workers if disadvantage from an
educational perspective is taken (Figure 3). These figures rise to over 170,000 additional people in
employment when the prevalence of long-term health conditions (LTC) is considered.

Figure 3 Expected increase in numbers employed through a reduction in the prevalence of
chronic illness from closing the health gap between most and least disadvantaged
Australians of working age


020406080100120140160180Male 25-44Male 45-64Female 25-44Female 45-64TotalAge Group (years)
Number
(
'
000)
In Bottom Income Quintile SAHSEarly School Leavers SAHSSocially Excluded SAHSPublic Housing Renters SAHSIn Bottom Income Quintile LTCEarly School Leavers LTCSocially Excluded LTCPublic Housing Renters LTC



(d) Increase in annual earnings
If there are more individuals in paid work, then it stands to reason that the total earnings from wages and
salaries for a particular socio-economic group will increase. The relative gap in weekly gross income from
wages and salaries between disadvantaged adult Australians of working age in good versus poor health
ranges between a 1.5-fold difference for younger males (aged 25 to 44) who live in public housing or who


experience low levels of social connectedness to over a staggering 6.5-fold difference experienced by
males aged 45 to 64 in the bottom income quintile or who are public housing renters.

Closing the gap in self-assessed health status could generate as much as $6-7 billion in extra earnings and,
in the prevalence of long-term health conditions, upwards of $8 billion (Figure 4). These findings reflect
two key factors – the large number of Australians of working age who currently are educationally
disadvantaged having left school before completing year 12 or who are socially isolated and the relatively
large wage gap between those in poor and good health in these two groups. In terms of increases in
annual income from wages and salaries, the greatest gains from taking action on the social determinants
of health can be made for males aged 45 to 64.

Figure 4 Expected increase in annual earnings from wages and salaries through either an
improvement in self-assessed health status (SAHS) or a reduction in the prevalence
of long-term health conditions (LTC) from closing the health gap between most and
least disadvantaged Australians of working age


01,0002,0003,0004,0005,0006,0007,0008,0009,000Male 25-44Male 45-64Female 25-44Female 45-64TotalAge Group (years)
Extra Annual
Earnings (
$m)
In Bottom Income Quintile SAHSEarly School Leavers SAHSSocially Excluded SAHSPublic Housing Renters SAHSIn Bottom Income Quintile LTCEarly School Leavers LTCSocially Excluded LTCPublic Housing Renters LTC

(e) Reduction in income and welfare support
A flow-on effect from increased employment and earnings and better health is the reduced need for
income and welfare support via Government pensions and allowances. Those in poor health or who have
a long-term health condition typically received between 1.5 and 2.5 times the level of financial assistance
from Government than those in good health or who were free from chronic illness. Irrespective of
whether an income, education or social exclusion lens is taken, closing the gap in health status potentially
could lead to $2-3 billion in savings per year in Government expenditure, and in the order of $3-4 billion
per year if the prevalence of chronic illness in most disadvantaged socio-economic groups could be
reduced to the level experienced by the least advantaged groups.


(f) Savings to the health system
Potential savings to the health system through Government taking action on the social determinants of
health were difficult to estimate because of the lack of socio-economic coded health services use and cost
data. As an example of the possible savings that might accrue, changes in the use and cost of health
services – hospitals, doctor and medically related (Medicare) services, and prescribed medicines
subsidised through the PBS – from changes in self-assessed health status for individuals in the lowest
household income quintile were modelled.

Nearly 400,000 additional disadvantaged individuals would regard their health as good if equity was
achieved with individuals in the top income quintile. Such a shift is significant in terms of health services
use and costs as there were very large differences in the use of health services by individuals in the
bottom income quintile between those in poor versus good health. More than 60,000 individuals need
not have been admitted to hospital. More than 500,000 hospital separations may not have occurred and
with an average length of stay of around 2.5 days, there would have been some 1.44 million fewer patient
days spent in hospital, saving around $2.3 billion in health expenditure.

A two-fold difference in the use of doctor and medical services was found between disadvantaged
persons in poor versus good health. Improving the health status of 400,000 individuals of working age in
the bottom income quintile would reduce the pressure on Medicare by over 5.5 million services. Such a
reduction in MBS service use equates to a savings to Government of around $273 million each year. With
respect to the use of prescription medicines, in 2008, disadvantaged individuals in the 45 to 64 age group
and who were in poor health and who were concession cardholders used 30 prescriptions on average
each. While those aged 25 to 44 averaged 19 scripts, both age groups used twice as many scripts as
concessional patients in good health. Over 5.3 million PBS scripts would not have been required by
concessional patients if health equity existed. However, a shift to good health through closing socio-
economic health gaps would shift around 15,000 persons in low-income households from ‘having’ to ‘not
having’ concessional status, resulting in a net increase of 41,500 scripts (a 6 per cent increase) for general
patients. Health equity for concessional patients was estimated to yield $184.7 million in savings to
Government and a $15.6m reduction in patient contributions. However, there would be an increase in the
out-of-pocket cost of medicines to general patients by some $3.1m.

Conclusions

This is the first study of its kind in Australia that has tried to gauge the impact of Government inaction on
the social determinants of health and health inequalities. Reducing health inequalities is a matter of social
inclusion, fairness and social justice (Marmot et al, 2010). The fact that so many disadvantaged
Australians are in poor health or have long-term health conditions relative to individuals in the least socio-
economically disadvantaged groups is simply unfair. So are the impacts on people’s satisfaction with their
lives, missed employment opportunities, levels of income and need for health services. This study shows
that major social and economic benefits are being neglected and savings to Government expenditure and
the health system overlooked. The findings of this Report are revealing and are of policy concern
especially within the context of Australia’s agenda on social inclusion. However, in this study the health
profile of individuals of working age in the most socio-economic disadvantaged groups only was
compared with that of individuals in the least disadvantaged groups. The first CHA-NATSEM Report
(Brown et al, 2010) on health inequalities showed that socio-economic gradients in health exist in
Australia. It is not only the most socio-economically disadvantaged groups that experience health
inequalities relative to the most advantaged individuals, but also other low and middle socio-economic
groups. Thus, this Report provides only part of the story of health inequalities in Australians of working
age.

Socio-economic inequalities in health persist because the Social Determinants Of Health are not being
addressed. Government action on the Social Determinants Of Health and health inequalities would require
a broad investment, a focus on health in all policies and action across the whole of society. In return,
significant revenue would be generated through increased employment, reduction in Government
pensions and allowances, and savings in Government spending on health services. The WHO Commission
on the Social Determinants of Health called for national governments to develop systems for the routine
monitoring of health inequities and the Social Determinants Of Health, and to develop more effective
policies and implement strategies suited to their particular national context to improve health equity
(http://www.who.int/social_determinants/en/ ). This Report continues the work of demonstrating how
improving health equity could have a major impact on the health and well-being of Australians, as well as
a significant financial impact for the country.

Key words

Socio-economic disadvantage, health inequalities, Social Determinants Of Health, Government action

1 INTRODUCTION
There are no regular reports that investigate and monitor trends in Australia in health inequality over time nor
whether gaps in health status between ‘rich’ and ‘poor’ Australians are closing. In September 2010, Catholic
Health Australia (CHA) and the National Centre for Social and Economic Modelling (NATSEM) released the first
CHA-NATSEM Report on Health Inequalities “Health lies in wealth: Health inequalities in Australians of working
age” (Brown and Nepal, 2010). That Report investigated socio-economic inequalities in health outcomes and
lifestyle risk factors of Australians of working age, i.e. individuals aged 25 to 64. The Report received widespread
media attention. Taking a social determinants of health perspective, the study showed health inequalities exist
for Australians of working age; social gradients in health were common, i.e. the lower a person’s social and
economic position, the worse his or her health is; and that the health gaps between the most disadvantaged and
least disadvantaged socio-economic groups were often very large. The Report further showed that household
income, a person’s level of education, household employment, housing tenure and social connectedness all
matter when it comes to health. Socio-economic differences were found in all the health indicators studied –
mortality, self-assessed health status, long-term health conditions and health risk factors (such as smoking,
physical inactivity, obesity and at-risk alcohol consumption) – and were evident for both men and women and
for the two age groups (those aged 25-44 and 45-64) studied.

As Professor Marmot and his review team remark in the Strategic Review of Health Inequalities in England post-
2010, serious health inequalities that are observed do not arise by chance (Marmot et al, 2010). Social
inequalities in health occur because of the inequalities in the conditions of daily life under which we are born,
develop as young children, grow into teenage years and adulthood, and live into old age. The material and social
circumstances under which we live are in turn shaped by the unequal distribution of money, power and
resources at both the local and national levels. We have different access to household goods and services, to
health care, schools and higher education, conditions of work and leisure, housing and community resources,
and different opportunities to lead flourishing and fulfilling lives. A collection of societal factors will play out over
an individual’s lifetime and will be expressed through their health and health behaviours. Evidence collected by
social determinants of health researchers shows that it is the social determinants of health that are mostly
responsible for health inequities – the unfair and avoidable differences in health status seen within countries
(http://www.who.int/social_determinants/en/).

Health inequalities persist because inequalities persist across key social and economic domains – early child
development and education, employment and working conditions, housing and neighbourhood conditions,
standards of living, and, more generally, the freedom to participate equally in the benefits of society (Marmot et
al, 2010). The Australian Government’s vision of a socially inclusive society is one in which all Australians feel
valued and have the opportunity to participate fully in the life of our society. Achieving this vision means that all
Australians will have the resources, opportunities and capability to: learn by participating in education and
training; work by participating in employment, in voluntary work and in family and caring; engage by connecting
with people and using their local community’s resources; and have a voice so that they can influence decisions
that affect them (www.socialinclusion.gov.au). Australian families and individuals may experience social
exclusion if they lack certain resources, opportunities or capabilities so that they are unable to participate in
learning, working or engaging activities and are unable to influence the decisions affecting them.

What would it mean for Australians of working age if the gaps in health between the least socio-economically
disadvantaged and most socio-economically disadvantaged were closed? How many more individuals would feel


satisfied with their life? How many more would be in full-time work or even employed part-time? How would
earnings from paid work increase and the reliance on Government welfare payments reduce? If the most
disadvantaged Australians of working age enjoyed the same health profile of the most advantaged, what savings
would occur through reduced use of hospitals, doctors, medical services or prescribed medicines for example?
These potential social and economic benefits are the costs of Government inaction on the social determinants of
health and on socio-economic health inequalities.

1.1 OBJECTIVES OF THIS REPORT
The aim of this research is to provide an indication of the extent of the cost of Government inaction in
developing policies and implementing strategies that would reduce socio-economic differences within the
Australian population of working age (25-64 years) that give rise to health inequities.

The cost of inaction is measured in terms of the loss of potential social and economic outcomes that might
otherwise have accrued to socio-economically disadvantaged individuals if they had had the same health profile
of more socio-economically advantaged Australians. For the purposes of this report, the contrast is made
between those who are most socio-economically disadvantaged and those who are least disadvantaged defined
in terms of household income, level of education, housing tenure and degree of social connectedness.

Four types of key outcomes are considered – the number of disadvantaged Australians of working age
experiencing health inequity, satisfaction with life, economic outcomes (including employment, income from
paid work, savings to Government expenditure on social security payments and transfers) and savings to the
health system.

Thus the Report aims to address five key questions:

• If the most socio-economically disadvantaged Australians of working age had the same self-
reported health status profile of the least disadvantaged groups,how many more individuals
would be in good health rather than poor health?
• If the most socio-economically disadvantaged Australians of working age had the same
prevalence of long-term health conditions as the least disadvantaged groups,how many more
individuals would be free from chronic long-term illness?
• If individuals in the most socio-economically disadvantaged groups had the same health profile
– in terms of self-assessed health status and long-term health conditions – of the least
disadvantaged groups, how many more individuals would be satisfied with their life?
• If individuals in the most socio-economically disadvantaged groups had the same health profile
of the least disadvantaged groups, what improvements in employment status, income from
paid work and reductions in government pensions, allowances and other public transfers are
likely to be gained?
• If individuals in the most socio-economically disadvantaged groups had the same health profile
of the least disadvantaged groups, what savings might occur to the health system in terms of
reduced number of hospital separations, number of doctor- and medical-related services and
prescribed medicines and associated costs to Government?

1.2 STRUCTURE OF THIS REPORT
The following section outlines the key health and socio-economic indicators that have been chosen to explore
the cost of inaction in addressing health inequalities. The data sources and variables used are identified and
explained. A profile of the study population and a brief overview of the statistical analyses are provided.

How many disadvantaged Australians of working age are experiencing health inequity is explored in Section 3.
Potential gains in satisfaction with life are then investigated in Section 4 and economic gains from closing socio-
economic health gaps in Section 5. Section 6 addresses possible savings to Australia’s health system and some
concluding remarks are provided in Section 7.

2 MEASURING HEALTH AND SOCIO-ECONOMIC DISADVANTAGE

2.1 KEY HEALTH AND SOCIO-ECONOMIC INDICATORS
The analyses in this Report draw on the same data sources and variables used in the first CHA-NATSEM Report
“Health lies in wealth: Health inequalities in Australians of working age” (Brown and Nepal, 2010). The choice of
these was based on the commonality and importance of different social determinants of health reported in the
national and international literature and measures that represent key dimensions of health. The health and
socio-economic variables chosen for the analyses are described briefly in Table 1 below.

All of the variables in Table 1 are derived from the person-level data contained in Wave 8 of the Household,
Income and Labour Dynamics in Australia (HILDA) Survey and all involve self-reported data. The interviews for
Wave 8 were conducted between August 2008 and February 2009, with over 90 per cent of the interviews being
conducted in September-October 2008 (Watson, 2010). HILDA is a broad household-based social and economic
longitudinal survey which started in 2001. As Watson (2010) describes:

The HILDA Survey began with a large national probability sample of Australian households occupying private dwellings.
All members of the households providing at least one interview in Wave 1 form the basis of the panel to be pursued in
each subsequent wave. The sample has been gradually extended to include any new household members resulting from
changes in the composition of the original households. (Watson, 2010, p2).



More information on the variables can be found in Appendix 1.

The groups compared in this research representing the most and least disadvantaged Australians of working age
for the four socio-economic indicators are given in Table 3.




Table 1 Socio-economic and health domains and variables

Domain

Variable description

Socio-economic status



Household income

Annual disposable (after-tax) household income including government
transfers (government benefits) in the past financial year. Income is
equivalised to household size and structure, and is reported by quintile.

Education


Highest educational qualification categorised into three groups: year 11 and
below, year 12 or vocational qualification, and tertiary education.

Housing

Tenure type of the household – owner, purchaser, private renter, public
renter or rent other/free.

Social connectedness

A summary measure constructed on the basis of rating of three questions on
frequency of gathering with friends/relatives, perceived availability of
someone to confide in at difficult times, and feeling of loneliness. Classified
as low connectedness, moderate connectedness or high connectedness.

Health outcomes



Self-assessed health status



The five standard levels of self-assessed health status have been collapsed
into two: “good health” and “poor health” where “good health” includes
excellent, very good and good health; and “poor health” refers to fair and
poor health.

Presence of a long-term health
condition

Has any long-term health condition, impairment or disability that restricts an
individual in their everyday activities, and has lasted or is likely to last for six
months or more.





Table 2 Socio-economic classification



Most Disadvantaged

Least Disadvantaged

Income

bottom quintile

top quintile

Education

= year 11 schooling

tertiary qualification

Housing

public renter

homeowner

Social connectedness

low

high





2.2 MEASURING LOST BENEFITS – THE COSTS OF INACTION

As previously stated, the cost of Government inaction on social determinants of health is viewed in terms of the
loss of potential social and economic benefits that otherwise would have accrued to individuals in the most
disadvantaged socio-economic groups if they had had the same health profile as those who are least
disadvantaged. In the first CHA-NATSEM Report it was shown, for example, that only 51 per cent of males aged
45 to 64 who were in the bottom household income quintile reported that they were in good health compared
with 87 per cent in the top income quintile. So, what would happen in terms of their overall satisfaction with
their life, employment or income or need for government assistance, or their use of health services if an
additional 36 per cent of disadvantaged 45- to 64-year-old males enjoyed good health rather than being in poor
health?




Table 3 Outcome measures

Domain

Measure

Definition

Health Inequity







Inequity in self-assessed health status

Increase in number of most disadvantaged
individuals in good health if self-assessed health
profile was the same between most and least
disadvantaged groups



Inequity in long-term health conditions

Increase in number of most disadvantaged
individuals with no long-term health condition if
self-assessed health profile was the same
between most and least disadvantaged groups

Satisfaction with Life

Satisfaction with life overall

Classified as ‘not satisfied’ or ‘satisfied’ to the
question in HILDA ‘all things considered, how
satisfied are you with your life?’

Economic







Employment status

Classified as: employed full time, employed part
time, unemployed looking for full-time work,
unemployed looking for part-time work, not in the
labour force marginally attached, and not in the
labour force not marginally attached



Wages and salaries

Individual weekly gross wages and salary from
all jobs as at 2008



Government pensions & allowances

Total Government pensions & allowances
including income support payments and
payments to families, all age and other pensions,
Newstart and other allowance payments as at
2008

Health System







Hospital use

Number of persons hospitalised in public or
private hospital, number of separations and
number of patient days in 2008.



Use of doctor- and medical-related service

Number of Medicare Benefits Schedule (MBS)
services in 2008



Government expenditure on doctor- and
medical-related service

Benefits paid for MBS services in 2008



Use of prescribed medicines

Number of prescriptions dispensed through the
Pharmaceutical Benefits Scheme (PBS) in 2008



Government expenditure on prescribed
medicines

Benefits paid under the PBS in 2008



Consumer expenditure on prescribed
medicines

Co-payments paid on PBS medicines in 2008





A number of outcome measures were chosen for the analysis. These are described in Table 3. Data used to
address the first three domains are from the 2008 HILDA survey. An important category in terms of employment
status is ‘not in the labour force’ (NILF). Individuals who are not participating in the labour force are often
described as ‘marginally attached’ or ‘not marginally attached’ to the labour market. If a person is marginally
attached to the labour force then in many ways they are similar to those who are unemployed. However, while
they satisfy some, they do not satisfy all of the criteria necessary to be classified as unemployed. The marginally
attached include those who want to work and are actively looking for work, but were not available to start work;
or were available to start work but whose main reason for not actively looking for work was that they believed
they would not be able to find a job, i.e. discouraged jobseekers. Persons not in the labour force are classified as
‘not marginally attached’ to the labour force if they do not want to work or want to work at some stage but are
not actively looking for work and are not currently available to start work.


The data to assess potential savings to the health system were derived from three of NATSEM’s health
microsimulation models:

• HospMod – a static microsimulation model of the use and costs of public and private hospitals in
Australia (Brown et al, 2011);
• MediSim – a static microsimulation model of the use and costs of the Australian Pharmaceutical Benefits
Scheme (Abello and Brown, 2007); and
• the health module in APPSIM – a module within the dynamic microsimulation model APPSIM that
simulates lifestyle risk factors, self-assessed health status, health service utilisation and costs in
Australia over 50 years (Lymer, 2011).


These data were supplemented by administrative data on the MBS and PBS from Medicare Australia.

The steps taken to estimate potential benefits if the health inequity between the most and least disadvantaged
individuals disappeared are described below (and as represented in Figure 1).

1. The proportion of individuals in the most disadvantaged group (for each of the socio-economic
characteristics above) who were in good health, or who had a long-term health condition, was
compared with the percentage of individuals in the least disadvantaged group.
2. The number of additional individuals in each most disadvantaged group who would be expected to have
good health (or be free from chronic illness) if the most disadvantaged group had the same percentage
as the least disadvantaged group was calculated.
3. It was then assumed that the number of individuals ‘shifting’ from poor to good health, or having to not
having a long-term health condition, would have the same level of satisfaction with life, employment
profile, income, government benefits and payments, and use of health services as those belonging to
individuals in the same most disadvantaged socio-economic group but who reported in the HILDA
survey that they were in good health. Thus, it is assumed that any ‘improvement’ in health does not
‘shift’ individuals out of their socio-economic group but rather they take on the socio-economic
characteristics of those in the group but who were ‘healthy’. The difference between the profiles of all
individuals having poor health and the mix of some individuals remaining in poor health and some
shifting to good health gives a measure of the potential gains that might occur if health equity was
achieved between the most and least disadvantaged socio-economic groups in Australia.


The HILDA survey population weights were applied to the person-level records to generate the estimates for the
Australian population of working age. As in the first CHA-NATSEM Report, the study population is broken down
by gender and into two age groups: those aged 25 to 44 and those aged 45 to 64. Youth under 25 years of age
were excluded as many of these individuals could be studying. In the first Report, simple cross-tabulations
between the various socio-economic and health indicators were generated and the percentages of the different
socio-economic groups having a particular health characteristic calculated (Brown and Nepal, 2010).

2.3 MISSING DATA

The HILDA Wave 8 data had a total of 8,217 unit records for people aged 25to 64. For some variables, however,
a slightly fewer number of records were available for analyses owing to non-response. To deal with this, we
compared the socio-demographic profiles of people with missing and non-missing responses. Differences were
not sufficiently large to bias the results for whom responses were known.




2.4 PROFILE OF THE STUDY POPULATION

The basic socio-economic profile of the Australian population of working age is given in Table 4. In 2008, nearly
14 per cent of persons of working age lived in Australia’s poorest 20 per cent of households1. One of every four
Australians aged between 25 and 64 had left high school before completing year 12, with nearly two of every
five females aged 45 to 64 being an early school leaver. Although the majority of individuals were home-owners
(either outright owners or purchasers), nearly 500,000 (4%) Australians of working age lived in public rental
accommodation. Over one in five individuals of working age experienced a low level of social connectedness –
gathering infrequently with friends or relatives, having no one or struggling to find someone to confide in at
difficult times, and often felt lonely.

1 Defined by annual disposable (after-tax) household income including government transfers (government benefits) in the past
financial year where income is equivalised to household size and structure, and is reported by quintile.



Table 4 Per cent distribution of men and women aged 25-64 years by selected socioeconomic
characteristics



Menc



Womenc



25-44

45-64



25-44

45-64

Equivalised disposable HHa income quintileb











Bottom

10

15



13

17

Second

20

17



20

18

Third

22

21



22

18

Fourth

23

22



22

22

Top

25

26



22

25

Education











Year 11 and below

18

25



20

38

Year 12 / vocational

55

52



45

40

Tertiary

27

22



35

22

Housing tenure











Owner

16

45



17

47

Purchaser

49

37



51

34

Renter private

28

13



26

12

Renter public

4

4



4

6

Rent other/free

3

1



3

1

Social connectedness











Low connectedness

20

28



19

24

Moderate connectedness

30

33



30

32

High connectedness

30

25



35

30

Population (million)

2.97

2.63



2.99

2.70

Number records in HILDA

2,007

1,879



2,230

2,101



Source: HILDA Wave 8 datafile.
Note: aHH = household. b Equivalised disposable household income quintile is based on all responding households in the full HILDA
sample, and weighted by population weights. c Percentage totals may not add to 100 owing to rounding or missing data..






3 HOW MANY DISADVANTAGED AUSTRALIANS OF WORKING AGE ARE
EXPERIENCING HEALTH INEQUITY?

As many as one in nine 25- to 44-year-old Australians and over one in five Australians aged 45 to 64 believe their
health to be poor or at best fair. However, the proportion of individuals who report their health as being poor
differs greatly by socio-economic status, with inequalities in self-assessed health status being significant for both
men and women, and for both the younger and older age group studied. For example, three-quarters of those
aged 25 to 44 and half of individuals aged 45 to 64 and who live in poorest income quintile households report
poor health compared with 85 to 95 per cent of those living in the top 20 per cent of households. Around 15 per
cent of Australians aged 25 to 44 and a third of those aged 45 to 64 have at least one long-term health condition,
impairment or disability that restricts them in their everyday activities and that has lasted, or is likely to last, for
six months or more. Health conditions included under the term ‘long-term health conditions’ are very broad,
ranging from, for example, a person having hearing problems, loss of sight or visual impairment, long-term
effects of a head injury or stroke, chronic or recurring pain, limited use of their arms or legs, a mental health
condition, arthritis, asthma, heart disease, dementia and so on. However, the key factor is that whatever health
problem or problems an individual has, this impacts on their daily life and is long-lasting. As with self-assessed
health status, there is a major socio-economic differential in the prevalence of long-term health conditions –
those who are most socio-economically disadvantaged are twice as likely as those who are least disadvantaged
to have a long-term health condition, and for disadvantaged younger men up to four to five times as likely
(Brown and Nepal 2010).

If the health gaps between the most and least disadvantaged groups were closed, i.e. there were no inequity in
the proportions in good health or who were free from long-term health conditions, then how many more most
disadvantaged Australians of working age would be in good health or have no chronic health problem?

Tables 5 and 6 show the number and health profile of individuals in the most disadvantaged income,
educational, housing and social exclusion groups and compares the proportion in ‘good’ health or ‘does not have
a long-term health condition’ with individuals in the least disadvantaged groups. The number of individuals who
are socio-economically disadvantaged differs substantially between the four indicators. Nonetheless, it is clear
that many socio-economically disadvantaged Australians experience poor health including chronic illness, and
that the rates of ill-health are significantly higher (p<0.05) than those for least disadvantaged individuals. Over
700,000 of the 2.8 million working-aged Australians who left school before completing high school report their
health as poor – this is a significant number of Australians. Of the 485,000 living in public rental accommodation,
44 per cent (211,000 people) report their health as poor. And, more individuals report having at least one long-
term health condition (Table 6) with typically between 750,000 and 1 million people reporting a chronic health
problem.

Combined with these large numbers is the significant difference in the health profile of the most and least
disadvantaged groups. While inequity occurs across all four socio-economic measures, the most striking
differences are by household income and housing tenure where the percentage point difference for both males
and females aged 45 to 64 is between 30 and 40 per cent. The final columns in Tables 5 and 6 give estimates of
the number of individuals who would be expected to be in good health or have no long-term illness if the
prevalence rates for the least disadvantaged group also applied to most disadvantaged individuals. In other
words, these estimates are a measure of the number of individuals experiencing health inequity.


Leaving housing tenure aside, a staggering number of around 370,000 to 400,000 additional disadvantaged
Australians would see their health as being good if socio-economic inequalities in health disappeared – this
number is equivalent to the entire population of the ACT (Table 5). Government action on the social
determinants of health would particularly benefit females in terms of self-assessed health status. With respect
to long-term health conditions, an estimated 405,000 to 500,000 additional individuals (approaching the
population of Tasmania) would be free from chronic illness if prevalence rates were equalised. Again in
numerical terms, the group that would benefit the most are females aged 45 to 64 (Table 6).

Table 5 Inequality in self-assessed health status – potential increase in numbers of most
disadvantaged Australians reporting good health through closing the health gap between
most and least disadvantaged Australians of working age



Most Disadvantaged Group

Least
Disadv.
Group

Difference
in % Good
Health

Increase
in No. of
Most
Disadv. in
Good
Health



Group
Pop (No.)

No. In
Poor
Health

No. In
Good
Health

%
Good
Health

% Good
Health



Income Quintile













Male 25-44

301,333

70,158

231,175

76.7

93.3

16.6

49,864

Male 45-64

384,626

188,624

196,003

51.0

86.5

35.6

136,889

Female 25-44

398,476

88,084

310,392

77.9

92.4

14.5

57,906

Female 45-64

468,563

218,833

249,730

53.3

85.8

32.5

152,327

Total

1,552,998

565,699

987,300

-

-

-

396,986

Educational Attainment













Male 25-44

541,677

97,419

444,258

82.0

92.5

10.5

44,911

Male 45-64

669,051

229,672

439,379

65.7

85.0

19.3

127,315

Female 25-44

605,230

86,467

518,763

85.7

93.2

7.5

60,548

Female 45-64

1,028,959

284,585

744,374

72.3

88.3

16.0

146,878

Total

2,844,917

698,143

2,146,774

-

-

-

379,652

Housing Tenure













Male 25-44

104,525

31,634

72,892

69.7

92.4

22.7

23,659

Male 45-64

93,698

51,035

42,663

45.5

78.2

32.7

30,624

Female 25-44

114,649

32,498

82,151

71.7

90.5

18.8

21,549

Female 45-64

172,503

94,699

77,804

45.1

83.4

38.3

66,033

Total

485,376

209,866

275,510

-

-

-

141,865

Social Connectedness













Male 25-44

604,147

110,338

493,809

81.7

94.0

12.3

74,191

Male 45-64

735,361

213,866

521,495

70.9

81.8

10.9

79,896

Female 25-44

568,955

110,978

457,978

80.5

94.2

13.7

77,913

Female 45-64

645,296

227,592

417,704

64.7

86.1

21.4

137,606

Total

2,553,759

662,774

1,890,986

-

-

-

369,606



Source: HILDA Wave 8 datafile.

Top four




Table 6 Inequality in long-term health conditions – potential increase in numbers of most
disadvantaged Australians reporting no long-term health conditions through closing the
health gap between most and least disadvantaged Australians of working age



Most Disadvantaged Group

Least
Disadv.
Group

Difference
in % Does
not have a
LTC

Increase in
No. of Most
Disadv. who
do not have
a LTC



Group
Pop (No.)

Has a
LTC

Does not
have a
LTC

% Does
not have
a LTC

% Does
not have
a LTC

Income Quintile













Male 25-44

301,333

114,859

186,474

61.9

90.9

29.0

87,464

Male 45-64

384,626

239,988

144,638

37.6

73.8

36.2

139,107

Female 25-44

398,476

118,288

280,188

70.3

87.2

16.9

67,387

Female 45-64

468,563

277,850

190,713

40.7

76.6

35.9

168,008

Total

1,552,998

750,985

802,013

-

-

-

461,966

Educational Attainment













Male 25-44

541,677

123,533

418,144

77.2

90.6

13.4

72,353

Male 45-64

669,051

308,982

360,069

53.8

75.1

21.3

142,402

Female 25-44

605,230

131,533

473,697

78.3

89.2

10.9

66,012

Female 45-64

1,028,959

420,330

608,629

59.1

80.2

21.1

216,934

Total

2,844,917

984,378

1,860,539

-

-

-

497,701

Housing
Tenure















Male 25-44

104,525

50,919

53,606

51.3

83.3

32.0

33,479

Male 45-64

93,698

62,933

30,765

32.8

66.4

33.6

31,406

Female 25-44

114,649

51,931

62,718

54.7

80.1

25.4

29,129

Female 45-64

172,503

114,308

58,195

33.7

70.2

36.5

62,871

Total

485,375

280,091

205,284

-

-

-

156,885

Social Connectedness













Male 25-44

604,147

144,800

459,347

76.0

88.0

12.0

72,599

Male 45-64

735,361

317,018

418,343

56.9

73.7

16.8

123,615

Female 25-44

568,955

138,865

430,090

75.6

88.3

12.7

72,219

Female 45-64

645,296

304,702

340,594

52.8

74.1

21.3

137,769

Total

2,553,759

905,385

1,648,374

-

-

-

406,202



Source Data: HILDA Wave 8 datafile.



Top four



If the health gap between the most and least disadvantaged groups were closed,how many more socio-
economically disadvantaged Australians of working age would be satisfied with their lives, how would
employment status change, what gains might be made in earnings from paid work and reductions in government
welfare payments, and what savings might accrue to the health system? These potential benefits are
investigated in the following sections.




4 COSTS TO WELL-BEING - POTENTIAL GAINS IN SATISFACTION WITH LIFE

In the HILDA survey, respondents were asked about how satisfied or dissatisfied they are with some of the things
happening in their lives. This includes a wide range of experiences – the home in which they live, their
employment opportunities, their financial situation, how safe they feel, feeling part of their local community,
their health, the neighbourhood in which they live and the amount of free time they have. After considering
these aspects of their lives, they are asked ‘all things considered, how satisfied are you with your life?’ Tables 7
and 8 present differences in the proportion of those in the most disadvantaged groups who are satisfied with
their lives according to their health status and presence or absence of long-term illness. The last columns in
Tables 7 and 8 give the expected increase in number of disadvantaged individuals satisfied with their lives, based
on the estimated increase in numbers of individuals expected to be in good health or free from chronic illness
from closing the health gap between most and least disadvantaged Australians of working age (last columns in
Tables 5 and 6) and the differences in proportion of disadvantaged persons satisfied with life by level of health
(Tables 7 and 8).

Table 7 Percentage disadvantaged persons satisfied with life by health status and increase in those
satisfied through closing the health gap between most and least disadvantaged Australians
of working age



Poor Health
(%)

Good Health
(%)

Difference
(%)

Increase in
Number
Satisfied

Lowest Income Quintile









Male 25-44

53.4

84.1

30.7

15,308

Male 45-64

55.7

86.5

30.8

42,162

Female 25-44

47.9

86.7

38.8

22,468

Female 45-64

61.3

88.9

27.6

42,042

Total







121,980

Year 11 or below









Male 25-44

52.7

83.6

30.9

13,877

Male 45-64

62.9

86.9

24.0

30,556

Female 25-44

63.4

84.3

20.9

12,655

Female 45-64

71.4

93.6

22.2

32,607

Total







89,695

Public Renters









Male 25-44

18.9

71.3

52.4

12,397

Male 45-64

61.9

86.8

24.9

7,625

Female 25-44

58.6

63.8

5.2

1,121

Female 45-64

76.7

85.3

8.6

5,679

Total







26,822

Low Social Connectedness







Male 25-44

51.1

79.6

28.5

21,144

Male 45-64

50.8

87.1

36.3

29,002

Female 25-44

46.0

76.3

30.3

23,608

Female 45-64

64.9

86.0

21.1

29,035

Total







102,789



Source Data: HILDA Wave 8 datafile.

Top four




Table 8 Percentage persons satisfied with life by presence of a long-term health condition and
increase in those satisfied through closing the health gap between most and least
disadvantaged Australians of working age



Has LTC
(%)

Does not
have a LTC
(%)

Difference
(%)

Increase in
Number
Satisfied

Lowest Income Quintile









Male 25-44

68.7

81.7

13.0

11,370

Male 45-64

62.9

82.8

19.9

27,682

Female 25-44

60.8

81.1

20.3

13,680

Female 45-64

63.3

93.0

29.7

49,898

Total







102,631

Year 11 or below









Male 25-44

72.3

81.0

8.7

6,295

Male 45-64

70.2

84.8

14.6

20,791

Female 25-44

69.3

82.1

12.8

8,450

Female 45-64

73.3

91.2

17.9

38,831

Total







74,366

Public Renters









Male 25-44

45.9

73.0

27.1

9,073

Male 45-64

62.7

84.4

21.7

6,815

Female 25-44

53.9

67.5

13.6

3,962

Female 45-64

69.8

85.1

15.3

9,619

Total







29,469

Low Social Connectedness







Male 25-44

61.0

78.7

17.7

12,850

Male 45-64

68.0

83.4

15.4

19,037

Female 25-44

56.1

75.5

19.4

14,010

Female 45-64

73.9

82.8

8.9

12,261

Total







58,159



Source Data: HILDA Wave 8 datafile.



Top four

With respect to self-assessed health status, there are substantial differences in the proportion of disadvantaged
individuals satisfied with their lives between those in poor versus good health – with the exception of female
public housing renters. Typically only between 45 and 65 per cent of individuals in poor health are satisfied with
their life whereas, for those in good health, the proportion increases to around 80 to 90 per cent. On average,
nearly 30 per cent more of disadvantaged individuals in good health said they were satisfied with their lives
compared with those in poor health. More than eight in every 10 younger males who had poor health and who
lived in public rental housing were dissatisfied with their lives.

If the health status of those in the most socio-economically disadvantaged groups could be improved to be on
par with the least disadvantaged groups, then as many as 120,000 individuals could shift from being dissatisfied
to satisfied with their lives. For some groups, the gain in numbers equates to around 10 per cent of the group’s
total populations, in particular, men and women aged 45 to 64 living in the poorest 20 per cent of households
and male public housing renters. Thus these numbers are not inconsequential.


The patterns for long-term health conditions (Table 8) reflect those in Table 7 for self-assessed health status,
with slightly fewer individuals in each group shifting to greater satisfaction with their life. Gains occur for all four
socio-economic indicators, but targeting health inequities by household income quintile would lead to the
greatest number of disadvantaged individuals benefitting from Government action.



5 LOST ECONOMIC BENEFITS – POTENTIAL ECONOMIC GAINS FROM CLOSING
HEALTH GAPS

5.1 POTENTIAL GAINS IN EMPLOYMENT

It is well known that health influences the participation of individuals in the labour force. Tables 9 and 10 show
the distribution of employment status of the four study groups broken down by self-assessed health status and
the presence of long-term health conditions. A key point to note is that while these groups are of working age,
they are also socio-economically disadvantaged, which is reflected in relatively high rates of unemployment or
not being in the labour force. Both distributions adhere to general patterns of employment in that it is the
younger males who have the highest rates of full-time employment, females the highest rates of part-time
employment and the older females the highest rates of having no attachment to the labour force. These broad
patterns are consistent across health status and long-term illness and the four socio-economic groupings.

The differences in employment between those in good and poor health and those not having or having a long-
term health problem are given in Tables 11 and 12. These tables also show what might happen to employment if
the health inequities between the most and least disadvantaged groups of individuals are overcome. The figures
show ‘shifts’ in employment states where increases in the number of individuals employed are matched by
numbers moving out of unemployment or into the labour force from not being in the labour force.

In terms of full-time employment, it is the older males, i.e. those aged 45 to 64, followed by younger males, who
experience the greatest health differentials, while in terms of part-time employment it is females in both age
groups who are most disadvantaged through health. The potential gains in the number of individuals in paid
work if the health gaps between the most and least disadvantaged groups could be closed are substantial.
Targeting inequality in health status would, for example, suggest an additional 141,000 early school leavers
would be employed full time or part time (Table 11). Even more individuals would be in the paid workforce if the
prevalence of long-term health conditions was reduced – the findings indicate that targeting long-term health
issues in either those living in the lowest income households or those who did not complete high school would
see more than 172,000 additional persons participating in paid work.

What do the numbers in the final column of Tables 11 and 12 represent? Improvement in the health status of
males aged 45 to 64 who either live in the poorest 20 per cent of households or who live in private rental
accommodation would lead to an additional 55,000 or 14,000 men respectively being in full- or part-time
employment. These figures equate to an additional one man in every seven males aged 45 to 64 in the bottom
income quintile or public renter disadvantaged groups being in paid work. With the exception of public renters,
the figures for younger males and for females represent about one additional person in 20 of the group
population being employed. For those in public rental accommodation this rises to about one in 10 individuals,
which is socially important given that those living in public rental accommodation are most often those
individuals who are suffering multiple and cumulative disadvantage.


When improvements in long-term health conditions are considered, then the magnitude of the impact rises, and
it is not only the older males who seem to benefit the most, but also the younger males. The figures in Table 12
suggest an additional one man in every five males aged 45 to 64 in the bottom income quintile or public renter
disadvantaged groups would be employed (either full or part time) and for the younger males in these two
groups an additional one male in every six and eight respectively. For the older females, the figures start to
approach an additional one female in 10 being employed.


Table 9 Distribution of employment status among most disadvantaged groups by health status

Employment Status

Poor Healtha

Good Healtha



M25-44

(%)

M45-64

(%)

F25-44

(%)

F45-64

(%)

M25-44

(%)

M45-64

(%)

F25-44

(%)

F45-64
(%)

Lowest Income Quintile

















Employed FT

21.6

10.3

2.6

2.2

49.1

38.5

11.3

9.2

Employed PT

5.8

3.9

8.1

13.4

16.9

15.7

30.8

20.7

UnEmpl looking FT work

12.1

8.1

0.0

0.6

9.6

4.6

8.7

2.5

UnEmpl looking PT work

0.0

0.0

4.0

3.1

0.5

0.4

2.5

2.3

NILF marginally attached

14.1

18.2

28.4

10.8

17.3

8.7

14.2

7.6

NILF not marginally attached

46.4

59.6

57.0

69.9

6.7

32.0

32.5

57.7

Total population (n)

70,158

188,624

88,084

218,833

231,175

196,003

310,392

249,730

Year 11 or Below

















Employed FT

42.8

32.8

17.2

8.4

73.4

67.2

31.2

28.3

Employed PT

6.1

4.7

19.6

18.5

10.8

10.2

31.8

33.7

UnEmpl looking FT work

8.2

2.8

2.2

0.7

2.4

2.0

2.9

0.5

UnEmpl looking PT work

0.0

0.0

2.5

0.9

0.9

0.2

3.0

0.8

NILF marginally attached

27.0

9.8

20.9

6.3

10.2

1.4

7.8

4.0

NILF not marginally attached

15.8

49.9

37.6

65.3

2.3

19.0

23.1

32.8

Total population (n)

97,419

229,672

86,467

284,585

444,258

439,379

518,763

744,374

Public Renters

















Employed FT

25.9

9.6

19.6

13.2

45.6

47.5

21.5

25.6

Employed PT

0.0

2.4

0.0

4.9

23.6

11.3

21.1

20.5

UnEmpl looking FT work

4.8

0.0

0.0

0.7

0.8

0.0

9.6

3.9

UnEmpl looking PT work

0.0

0.0

3.9

1.1

0.0

0.0

0.0

4.4

NILF marginally attached

57.7

39.6

29.5

35.6

22.9

1.3

20.3

8.6

NILF not marginally attached

11.6

48.3

47.1

44.5

7.1

39.8

27.5

36.9

Total population (n)

31,634

51,035

32,498

94,699

72,892

42,663

82,151

77,804

Low Social Connectedness















Employed FT

56.0

26.6

23.3

14.9

83.5

71.6

41.8

36.5

Employed PT

6.3

5.9

22.5

18.8

5.5

10.4

26.5

31.1

UnEmpl looking FT work

6.7

7.1

1.5

0.5

5.1

2.0

5.0

1.5

UnEmpl looking PT work

0.0

0.5

2.3

3.5

0.2

0.3

3.5

1.8

NILF marginally attached

16.8

8.5

19.8

14.6

3.0

3.0

7.1

4.1

NILF not marginally attached

14.2

51.3

30.6

47.7

2.7

12.7

16.0

25.0

Total population (n)

110,338

213,866

110,978

227,592

493,809

521,495

457,978

417,704



Source Data: HILDA Wave 8 datafile.


Note a Percentage totals may not add to 100 owing to rounding or missing data.




Table 10 Distribution of employment status among most disadvantaged groups by prevalence of
long-term health conditions

Employment Status

Has a LTCa

Does not have a LTCa



M25-44

(%)

M45-64

(%)

F25-44

(%)

F45-64

(%)

M25-44

(%)

M45-64

(%)

F25-44

(%)

F45-64

(%)

Lowest Income Quintile

















Employed FT

10.2

7.6

8.1

2.2

64.5

49.2

12.3

12.1

Employed PT

9.6

6.2

8.1

12.6

15.5

16.0

32.1

21.8

UnEmpl looking FT work

11.3

5.1

6.8

2.1

8.6

6.7

7.5

0.3

UnEmpl looking PT work

5.7

0.0

5.1

3.0

0.0

0.5

1.4

1.1

NILF marginally attached

22.1

15.8

23.0

8.5

8.7

8.7

13.7

10.2

NILF not marginally attached

41.1

65.4

48.8

71.5

2.7

18.8

33.0

54.5

Total population (n)

114,859

239,988

118,288

277,850

186,474

144,638

280,188

190,713

Year 11 or Below

















Employed FT

30.7

29.0

15.0

15.6

81.3

74.6

32.9

29.6

Employed PT

19.1

7.5

20.9

20.4

7.1

10.1

31.9

32.4

UnEmpl looking FT work

4.4

0.8

5.1

1.1

2.8

3.2

2.8

0.1

UnEmpl looking PT work

5.7

0.0

5.3

1.1

0.5

0.2

1.9

0.7

NILF marginally attached

22.2

7.3

13.6

4.9

7.4

1.1

8.5

4.3

NILF not marginally attached

17.8

55.4

40.1

57.0

0.9

10.7

22.1

32.8

Total population (n)

123,533

308,982

131,533

420,330

418,144

360,069

473,697

608,629

Public Renters

















Employed FT

25.3

6.2

8.7

11.4

56.2

58.3

26.6

27.4

Employed PT

5.7

6.9

11.4

9.1

20.9

10.1

21.4

18.8

UnEmpl looking FT work

3.2

0.0

7.5

0.5

5.3

0.0

8.9

4.2

UnEmpl looking PT work

2.4

0.0

1.8

3.1

0.0

0.0

2.4

0.0

NILF marginally attached

39.0

29.9

29.9

32.3

16.9

1.4

10.8

15.5

NILF not marginally attached

24.4

57.0

40.7

43.5

0.6

30.2

29.8

34.1

Total population (n)

50,919

62,933

51,931

114,308

53,606

30,765

62,718

58,195

Low Social Connectedness















Employed FT

49.7

34.8

25.6

13.4

87.6

76.5

42.2

42.4

Employed PT

7.5

8.5

17.6

21.9

5.0

9.4

28.5

31.7

UnEmpl looking FT work

11.6

4.3

2.4

1.6

3.4

3.2

4.9

0.8

UnEmpl looking PT work

1.0

0.4

4.2

3.5

0.0

0.4

2.9

1.3

NILF marginally attached

17.8

7.5

22.3

11.8

1.6

2.3

5.8

4.1

NILF not marginally attached

12.4

44.6

27.9

47.8

2.3

8.2

15.7

19.8

Total population (n)

144,800

317,018

138,865

304,702

459,347

418,343

430,090

340,594



Source Data: HILDA Wave 8 datafile.

Note a Percentage totals may not add to 100 owing to rounding or missing data.






Table 11 Difference in employment between those with good and poor health status and change in
employment status from closing the health gap between most and least disadvantaged
Australians of working age



Difference in Employment (%)

Change in Number of People



M25-44

M45-64

F25-44

F45-64

M25-44

M45-64

F25-44

F45-64

Total

Lowest Income Quintile



















Employed FT

27.5

28.2

8.7

7.0

13,663

38,876

5,096

10,663

68,298

Employed PT

11.1

11.8

22.7

7.3

5,535

16,153

13,145

11,120

45,953

UnEmpl looking FT work

-2.5

-3.5

8.7

1.9

-1,247

-4,791

5,038

2,894

1,894

UnEmpl looking PT work

0.5

0.4

-1.5

-0.8

249

548

-869

-1,219

-1,291

NILF marginally attached

3.2

-9.5

-14.2

-3.2

1,596

-13,004

-8,223

-4,874

-24,505

NILF not marginally attached

-39.7

-27.6

-24.5

-12.2

-19,796

-37,781

-14,187

-18,584

-90,348

Year 11 or Below



















Employed FT

30.6

34.4

14.0

19.9

17,349

44,479

6,397

32,579

100,804

Employed PT

4.7

5.5

12.2

15.2

2,673

7,111

5,496

24,884

40,164

UnEmpl looking FT work

-5.8

-0.8

0.7

-0.2

-3,299

-1,034

315

-327

-4,345

UnEmpl looking PT work

0.9

0.2

0.5

-0.1

512

259

225

-164

832

NILF marginally attached

-16.8

-8.4

-13.1

-2.3

-9,556

-10,861

-5,901

-3,765

-30,083

NILF not marginally attached

-13.5

-30.9

-14.5

-32.5

-7,679

-39,953

-6,532

-53,206

-107,370

Public Renters



















Employed FT

19.7

37.9

1.9

12.4

4,661

11,606

409

8,254

24,930

Employed PT

23.6

8.9

21.1

15.6

5,584

2,726

4,547

10,301

23,158

UnEmpl looking FT work

-4.0

0.0

9.60

3.2

-946

0

2,069

2,113

3,236

UnEmpl looking PT work

0.0

0.0

-3.90

3.3

0

0

-840

2,179

1,339

NILF marginally attached

-34.8

-38.3

-9.2

-27.0

-8,233

-11,729

-1,982

-17,829

-39,773

NILF not marginally attached

-4.5

-8.5

-19.6

-7.6

-1,065

-2,603

-4,224

-5,019

-12,911

Low Social Connectedness

















Employed FT

27.5

45.0

18.5

21.6

20,319

20,403

35,873

14,492

91,087

Employed PT

-0.8

4.5

4.0

12.3

-591

-594

3,595

3,117

5,527

UnEmpl looking FT work

-1.6

-5.1

3.5

1.0

-1,182

-1,187

-4,075

2,727

-3,717

UnEmpl looking PT work

-

-0.2

1.2

-1.7

148

148

-160

935

1,071

NILF marginally attached

-13.8

-5.5

-12.7

-10.5

-10,197

-10,238

-4,394

-9,895

-34,724

NILF not marginally attached

-11.5

-38.6

-14.6

-22.7

-8,497

-8,532

-30,840

-11,375

-59,244



Source Data: HILDA Wave 8 datafile.



Top four






TABLE 12 Difference in employment between those without and with a long-term health condition and
change in employment status with reduction in prevalence of chronic illness from closing
the health gap between most and least disadvantaged Australians of working age



Difference in Employment (%)

Change in Number of People



M25-44

M45-64

F25-44

F45-64

M25-44

M45-64

F25-44

F45-64

Total

Lowest Income Quintile



















Employed FT

54.3

41.6

4.2

9.9

47,493

58,147

2,763

16,465

124,868

Employed PT

5.9

9.8

24.0

9.2

5,160

13,632

16,173

15,457

50,422

UnEmpl looking FT work

-2.7

1.6

0.7

-1.8

-2,362

2,226

472

-3,024

-2,688

UnEmpl looking PT work

-5.7

0.5

-3.7

-1.9

-4,985

696

-2,493

-3,192

-9,974

NILF marginally attached

-13.4

-7.1

-9.3

1.7

-11,720

-9,877

-6,267

2,856

-25,008

NILF not marginally attached

-38.4

-46.6

-15.8

-17.0

-33,586

-64,824

-10,647

-28,561

-137,618

Year 11 or Below



















Employed FT

50.6

45.6

17.9

14.0

36,538

65,078

11,750

30,805

144,171

Employed PT

-12.0

2.6

11.0

12.0

-8,682

3,702

7,261

26,032

28,313

UnEmpl looking FT work

-1.6

2.4

-2.3

-1.0

-1,158

3,418

-1,518

-2,169

-1,427

UnEmpl looking PT work

-5.2

0.2

-3.4

-0.4

-3,762

285

-2,244

-868

-6,589

NILF marginally attached

-14.8

-6.2

-5.1

-0.6

-10,708

-8,829

-3,367

-1,302

-24,206

NILF not marginally attached

-16.9

-44.7

-18.0

-24.2

-12,228

-63,654

-11,882

-52,498

-140,262

Public Renters



















Employed FT

30.9

52.1

17.9

16.0

8,772

16,363

5,243

9,997

40,375

Employed PT

15.2

3.2

10.0

9.7

5,089

1,005

2,913

6,098

15,105

UnEmpl looking FT work

2.1

0.0

1.4

3.7

703

0

408

2,326

3,437

UnEmpl looking PT work

2.4

0.0

0.6

-3.10

804

0

175

-1,949

-970

NILF marginally attached

-22.1

-28.5

-19.1

-16.8

-7,399

-8,951

-5,564

-10,562

-32,476

NILF not marginally attached

-23.8

-26.8

-10.9

-9.4

-7,968

-8,417

-3,175

-5,910

-25,470

Low Social Connectedness

















Employed FT

37.9

41.7

16.6

29.0

27,588

51,671

11,988

39,815

131,062

Employed PT

-2.5

0.9

10.9

9.8

-1,815

1,113

7,872

13,501

20,671

UnEmpl looking FT work

-8.2

-1.1

2.5

-0.8

-5,953

-1,360

1,805

-1,102

-6,610

UnEmpl looking PT work

-1.0

0.0

-1.3

-2.2

-726

0

-939

-3,031

-4,696

NILF marginally attached

-16.2

-5.2

-16.5

-7.7

-11,761

-6,428

-11,916

-10,608

-40,713

NILF not marginally attached

-10.1

-36.4

-12.2

-28.0

-7,332

-44,996

-8,811

-38,575

-99,714



Source Data: HILDA Wave 8 datafile.



Top four

5.2 INCOME AND GAINS IN ANNUAL EARNINGS



If there are more individuals in paid work then it stands to reason that total earnings from wages and salaries by
individuals within a particular socio-economic group will increase. Potential gains in annual earnings from wages
and salaries were estimated based on the difference in average weekly personal income between those in poor
versus good health. A conservative approach to measuring income was taken in that weekly gross (i.e. before tax
or anything else is taken out) income from wages and salaries was averaged across almost all individuals in a


group. Only those records in HILDA where data on income were missing or where income was stated as being
negative2 were excluded. Records for individuals stating they had zero earnings were included in the analysis.
This allows for different employment patterns and change in employment status across a full year. For example,
in the HILDA survey, employment status is based primarily on whether or not an individual undertook any paid
work at all during the last seven days prior to the survey. Individuals may have been in and out of the workforce
over the course of the year with their weekly earnings reflecting this fluctuating attachment to the labour
market. Hence, the average weekly incomes given in Table 13 are lower than if only either those in paid work at
the time of the survey or those in full- or part-time employment for all of the past year were considered.

2 Income may be negative when a loss accrues to a person as an owner or partner in unincorporated businesses or rental properties.
Losses occur when operating expenses and depreciation are greater than total receipts.

Conceptually the annual gains in earnings given in the last columns of Tables 13 and 14 represent the extra
earnings from those additional workers joining the workforce through improved health plus any increase in
weekly wages and salaries from those already in the workforce but whose health shifts from poor to good (or
from having to not having a long-term health condition).

The greatest absolute differentials in average weekly wages and salaries between those in good versus poor
health occur for males 45 to 64 years of age who are either socially isolated or early school leavers or live in
public housing, followed by younger males of working age who left school before completing year 12. The
relative gap in weekly gross income from wages and salaries ranges between a 1.5-fold difference for younger
males (aged 25 to 44) who live in public housing or who experience low levels of social connectedness to over a
staggering 6.5-fold difference experienced by males aged 45 to 64 in the bottom income quintile or who are
public housing renters.

Depending upon which socio-economic lens is used, closing the gap in self-assessed health status could lead to
anywhere between $1.4 billion and $7 billion in extra earnings. The largest benefits accrue for those who are
most educationally disadvantaged or who are socially excluded – this occurs for both men and women and for
younger and older individuals. These findings reflect two key features – the large number of Australians of
working age in these two disadvantaged socio-economic groups who would enjoy better health if socio-
economic inequalities in health did not exist and the relatively large wage gap between those in poor and good
health. Increase in earnings is most significant for males aged 45 to 64.

Potential benefits from closing the health gap in the prevalence of long-term health conditions replicate those
for self-assessed health status, although the health differential in wages and salaries are larger as well as the
resulting gains in annual earnings exceeding those from closing the socio-economic gap in health status.




Table 13 Weekly gross income from wages and salaries (2008) and increase in annual earnings from
improved health status from closing the health gap between most and least disadvantaged
Australians of working age



Poor Health

($)

Good Health

($)

Difference
($)

Ratio Good to
Poor Health

Gain in
earnings
($Millions pa)

Lowest Income Quintile









Male 25-44

174

372

198

2.1

513

Male 45-64

41

279

238

6.8

1,694

Female 25-44

42

130

88

3.1

265

Female 45-64

41

84

43

2.0

341

Total

-

-

-



2,813

Year 11 or Below











Male 25-44

331

733

402

2.2

939

Male 45-64

222

652

430

2.9

2,847

Female 25-44

161

359

198

2.2

623

Female 45-64

144

351

207

2.4

1,581

Total

-

-

-



5,990

Public Renters











Male 25-44

320

477

157

1.5

193

Male 45-64

71

470

399

6.6

635

Female 25-44

114

247

133

2.2

149

Female 45-64

199

333

134

1.7

460

Total

-

-

-



1,438

Low Social Connectedness









Male 25-44

668

1,034

366

1.5

1,412

Male 45-64

313

873

560

2.8

2,327

Female 25-44

250

477

227

1.9

920

Female 45-64

171

499

328

2.9

2,347

Total

-

-

-



7,005



Source Data: HILDA Wave 8 datafile.



Top four










Table 14 Weekly gross incomes from wages and salaries (2008) and increase in annual earnings
from reduction in prevalence of long-term health conditions from closing the health gap
between most and least disadvantaged Australians of working age



Has a LTC
($)

Does not
have a LTC
($)

Difference
($)

Ratio Good to
Poor Health

Income Gain
($ Millions pa)

Lowest Income Quintile









Male 25-44

150

429

279

2.9

1,269

Male 45-64

36

312

276

8.7

1,996

Female 25-44

82

147

65

1.8

228

Female 45-64

39

95

56

2.4

489

Total

-

-

-



3,982

Year 11 or Below











Male 25-44

334

800

466

2.4

1,753

Male 45-64

208

715

507

3.4

3,754

Female 25-44

165

377

212

2.3

728

Female 45-64

193

352

159

1.8

1,794

Total

-

-

-



8,029

Public Renters











Male 25-44

262

627

365

2.4

635

Male 45-64

46

598

552

13.0

902

Female 25-44

68

287

219

4.2

332

Female 45-64

142

395

253

2.8

827

Total

-

-

-



2,696

Low Social Connectedness









Male 25-44

633

1,074

441

1.7

1,665

Male 45-64

373

961

588

2.6

3,780

Female 25-44

303

480

177

1.6

665

Female 45-64

207

537

330

2.6

2,364

Total

-

-

-



8,473



Source Data: HILDA Wave 8 datafile.



Top four



5.3 GOVERNMENT PENSIONS AND ALLOWANCES AND SAVINGS IN GOVERNMENT EXPENDITURE

Many individuals of working age in disadvantaged socio-economic groups receive welfare support through the
Australian Government benefit and transfer system. This includes a variety of payments including, for example,
Newstart Allowance, Austudy Payment, the Disability Support Pension, Sickness Allowance, Widow Allowance,
Partner Allowance or the Parenting or Carers Payments. Family tax benefits have also been included in the
analysis. Eligibility for these pensions and allowances typically depends on individuals and families meeting
specified income and assets tests. With increased employment and earnings, an increased number of individuals
would no longer qualify for these payments, hence, there is potential for significant savings in Government
expenditure on welfare support with health equity. The results of this aspect of the modelling are provided in
Tables 15 and 16.


Leaving tenants of public housing aside for the moment, the difference in Government assistance in 2008
between those in poor versus good health was numerically greatest for those aged 45 to 64, typically ranging
between approximately $6,000 and $9,500 each year, with older males receiving slightly more financial
assistance than older females. The difference in Government benefits and allowances by health status varied
considerably by the socio-economic indicator used for those aged 25 to 44. For those living in the lowest income
quintile households, those in poor health received only around $1,000 more than those in good health. In
contrast, if younger working age adults are socially isolated and in poor health, then they received upwards of
$7,500 more in Government assistance than those in better health. Those in poor health typically received
between 1.5 and 2 times the level of financial assistance than those in good health. Irrespective of which of the
three socio-economic lenses is taken, closing the gap in health status could potentially lead to $2-3 billion in
savings per year in Government expenditure.

Similar patterns are shown in Table 16 when long-term health conditions are investigated. However, reducing
the prevalence of chronic illness to levels on par with those in the most socio-economically advantaged groups
could produce savings in Government spending in the order of $3-4 billion per year.

The findings for renters of public housing draw attention to some issues different to those found for the other
three socio-economic indicators. Individuals living in public housing are most often single persons living alone or
a single adult living with one or more children. They frequently will be unemployed or not looking for work
because of disability or ill health or through parenting or caring responsibilities (AIHW, 2011). Males aged 45 to
64 who were in good health or free from chronic illness but who through a range of social and economic
circumstances needed public housing for their accommodation received 80 per cent more in income from
Government benefits and allowances in 2008 than those in poorer health. The net result of this finding is that
closing the gaps in health inequity would increase public expenditure by around $450-475 million each year.

When considering self-assessed health status, both males and females aged 25 to 44 living in public housing who
were in poor health received considerably more Government assistance than those in good health when
compared with the differences in Government expenditure for these two groups by household income, level of
education or social connectedness. In contrast, the difference in welfare support by either health status or long-
term health conditions for women aged 45 to 64 living in public housing is considerably lower than those found
for the other three socio-economic lenses, primarily due to relatively higher payments to women in good health.
These findings for public renters reflect the complexity of the needs of those in public housing and the Australian
public benefits and transfer system in supporting those with disability and health needs and carers, support for
the long-term unemployed, and support for Australian families, especially in helping with the cost of raising
children.








Table 15 Government pensions and allowances per annum (2008) for those in poor and good health
and savings in government welfare expenditure from improved health from closing the
health gap between most and least disadvantaged Australians of working age



Poor Health

($)

Good Health

($)

Difference
($)

Ratio Poor to
Good Health

Govt Spending
($Millions pa)

Lowest Income Quintile









Male 25-44

19,559

18,623

-936

1.1

-47

Male 45-64

19,092

12,713

-6,379

1.5

-873

Female 25-44

23,038

21,989

-1,049

1.0

-61

Female 45-64

19,114

12,857

-6,257

1.5

-953

Total









-1,934

Year 11 or below











Male 25-44

16,794

10,221

-6,573

1.6

-295

Male 45-64

17,195

7,587

-9,608

2.3

-1,223

Female 25-44

20,654

13,742

-6,912

1.5

-419

Female 45-64

14,120

7,615

-6,505

1.9

-955

Total









-2,892

Public Renters











Male 25-44

27,038

18,187

-8,851

1.5

-209

Male 45-64

18,326

32,959

14,633

0.6

448

Female 25-44

33,076

22,433

-10,643

1.5

-229

Female 45-64

17,698

14,833

-2,865

1.2

-189

Total









-180

Low Social Connectedness









Male 25-44

13,427

6,249

-7,178

2.1

-533

Male 45-64

15,543

6,150

-9,393

2.5

-750

Female 25-44

13,189

10,676

-2,513

1.2

-196

Female 45-64

14,958

7,278

-7,680

2.1

-1,057

Total









-2,536



Source: Source Data: HILDA Wave 8 datafile.



Top four














Table 16 Government benefits and transfers per annum (2008) for those with and without a long-term
health condition and savings in government welfare expenditure from reduction in
prevalence of long-term health conditions from closing the health gap between most and
least disadvantaged Australians of working age



Has a LTC
($)

Does not have
a LTC ($)

Difference
($)

Ratio Poor to
Good Health

Govt Spending
($Millions pa)

Lowest Income Quintile









Male 25-44

22,605

14,990

-7,615

1.5

-666.0

Male 45-64

18,592

10,300

-8,292

1.8

-1,153.5

Female 25-44

24,182

21,008

-3,174

1.2

-213.9

Female 45-64

19,045

12,116

-6,929

1.6

-1,164.1

Total









-3197.5

Year 11 or below











Male 25-44

16,174

9,282

-6,892

1.7

-498.7

Male 45-64

15,907

6,628

-9,279

2.4

-1,321.4

Female 25-44

18,770

14,035

-4,735

1.3

-312.6

Female 45-64

14,986

6,807

-8,179

2.2

-1,774.3

Total









-3907

Public Renters











Male 25-44

24,188

17,522

-6,666

1.4

-223.2

Male 45-64

17,624

32,774

15,150

0.5

475.8

Female 25-44

23,575

26,143

2,568

0.9

74.8

Female 45-64

18,989

15,967

-3,022

1.2

-190.0

Total









Low Social Connectedness









Male 25-44

13,509

5,686

-7,823

2.4

-567.9

Male 45-64

12,820

5,971

-6,849

2.1

-846.6

Female 25-44

13,485

10,353

-3,132

1.3

-226.2

Female 45-64

14,052

6,317

-7,735

2.2

-1,065.6

Total









-2706.3



Source Data: HILDA Wave 8 datafile.

Top four


6 SAVINGS TO THE HEALTH SYSTEM FROM CLOSING HEALTH GAPS

Differences in the use of health services and potential savings to the health system are investigated in this
section of the Report. A key problem, however, in trying to estimate the impact of social determinants of
health and socio-economic inequalities in health is the lack of suitable socio-economic coded health data.
Socio-economic differentials in health services use and costs are typically limited in Australia to reporting
by composite socio-economic area-based measures such as the Index of Relative Socio-economic
Disadvantage (IRSD) – an index that reflects the aggregate socioeconomic status of individuals and
families living in a geographic unit (ABS, 2008). Measures of socio-economic status, such as income, at the
person or household (family) level that are linked to a person’s health status and use of health services
are not generally available.

For this reason, the analysis below takes changes in self-assessed health status for individuals living in
households in the lowest income quintile as an example to illustrate the possible savings that might
accrue to Australia’s health system from improvements in the health profiles of socio-economically
disadvantaged individuals of working age. Based on the findings in earlier sections of the Report, looking
at potential reductions in health services use and costs through a ‘household income lens’ will provide a
reasonable view as to likely benefits from conquering health inequalities. As shown in Section 3, an
additional 400,000 Australians of working age would assess their health as ‘good’ if health equity was
achieved between individuals living in the lowest versus the highest income quintile households. How
might this change in health status impact on the use and cost of Australia’s health system?

The necessary data for the analyses presented below were accessed from the 2008-09 output of three of
NATSEM’s health microsimulation models: HospMod, MediSim and the health module in APPSIM.

6.1 REDUCED USE OF AUSTRALIAN HOSPITALS

In 2008-09, there were a total of 8.148 million hospital separations from public and private hospitals in
Australia, 4.891m (60%) occurring in public hospitals. One-fifth of these were by Australians aged 25 to 44
(males 0.584m or 7.2% separations; females 1.108m or 13.6% separations) and nearly 30 per cent by
individuals aged 45 to 64 (males 1.186m or 14.6% separations; females 1.159m or 14.2% separations)
(AIHW, 2010). An estimated $41.8 billion was spent on Australia’s hospitals in 2008–09 (AIHW, 2011).

As would be expected, there is a significant difference in the likelihood that a person living in the bottom
income quintile households would be hospitalised by their health status (Table 17). In 2008, over one in
three disadvantaged persons in poor health needed a hospital either as a day-only patient or for at least
one overnight stay. Although this rate is considerably higher than for those in good health, still between
one and two in every 10 significantly socio-economically disadvantaged persons who thought their health
to be good was hospitalised. Using the findings in Table 5 on the potential increase in numbers of those
living in the bottom income quintile households likely to regard their health as good through closing the
health gap between the most and least disadvantaged income quintiles and the health status differences
in rates of hospitalisation for those in the bottom quintile, the potential reduction in the number of
disadvantaged persons hospitalised can be estimated. The results are shown in Table 17.These data
suggest that over 60,000 fewer people would use Australian hospitals each year if health equity could be
achieved.




Table 17 Hospitalisation in 2008 for Australians of working age in the bottom income quintile
and reductions in persons hospitalised through closing the health gap between most
and least disadvantaged Australians of working age





No of Disadv. Persons

% Disadv.
Persons
Hospitalised

No Disadv. Persons
Hospitalised

Reduction in
No of Disadv.
Persons
Hospitalised



In Poor
Health

Remain
in poor
health

Shift to
good
health

Poor
Health

Good
Health

In Poor
Health

Remain
in poor
health

Shift to
Good
Health

Male 25-44

70,158

20,294

49,864

30.4

12.8

21,328

6,169

6,383

8,776

Male 45-64

188,624

51,735

136,889

34.1

15.8

64,321

17,642

21,628

25,051

Female 25-44

88,084

30,178

57,906

37.9

22.9

33,384

11,437

13,260

8,687

Female 45-64

218,833

66,506

152,327

32.4

20.1

70,902

21,548

30,618

18,736

All persons

565,699

168,713

396,986

-

-

189,935

56,796

71,889

61,250



Source Data: NATSEM’s Microsimulation model ‘HospMod’



The average number of separations per year experienced by persons who were hospitalised also varies by
health status with those in poor health having a much higher rate of re-admission – especially males aged
45 to 64 (Table 18). The modelling from HospMod suggests that in 2008, individuals aged 25 to 64 who
were in the bottom income quintile households and who were in poor health contributed to nearly 1
million hospital separations, i.e. nearly 12 per cent of all hospital separations in Australia. However, over
500,000 hospital episodes could be prevented if the health gap between these individuals and those living
in the top income quintile households could be closed (Table 18).



Table 18 Estimated number of hospital separations in 2008 for Australians of working age in
the bottom income quintile and reductions in persons hospitalised through closing
the health gap between most and least disadvantaged Australians of working age



Ave No Separations
per disadv. person
hospitalised

No. of Separations

Reduction in
No of
Separations



In Poor
Health

In Good
Health

Disadv
Persons in
Poor Health

Disadv
Persons
Remain in
Poor hHealth

Disadv
Persons
Shift to
Good Health

Male 25-44

4.4

3.3

93,843

27,145

21,063

45,635

Male 45-64

6.6

2.2

424,517

116,435

47,583

260,499

Female 25-44

3.0

1.7

100,152

34,312

22,543

43,296

Female 45-64

4.8

2.4

340,329

103,430

73,483

163,417

All persons

-

-

958,841

281,323

164,671

512,847



Source Data: NATSEM’s Microsimulation model ‘HospMod’



Average length of stay (ALOS) in hospital varies between two and four days depending on age, gender and
health status (Table 19). With reductions in the number of persons hospitalised and number of
separations, and difference in ALOS, removing health inequality could ultimately result in 1.44 million
fewer patient days spent in hospital by socio-economically disadvantaged persons of working age.




Table 19 Average length of hospital stay in 2008 for Australians of working age in the bottom
income quintile and reductions in patient days stay through closing the health gap
between most and least disadvantaged Australians of working age



ALOS (Days)

No. of Patient Days

Reduction in
No of
Patient Days



Disadv
Persons
In Poor
Health

Disadv
Persons
In Good
Health

Disadv
Persons in
Poor
Health

Disadv
Persons
Remain in
Poor Health

Disadv
Persons
Shift to
Good Health

Male 25-44

3.7

1.8

347,220

100,437

37,913

208,870

Male 45-64

2.7

2.4

1,146,196

314,374

114,198

717,624

Female 25-44

2.2

2.5

220,333

75,487

56,357

88,489

Female 45-64

2.4

1.9

816,790

248,232

139,617

428,941

All persons

-

-

2,530,540

738,531

348,085

1,443,924



Source Data: NATSEM’s Microsimulation model ‘HospMod’

Estimating potential savings in dollar terms is problematic because of the variation in the causes of
admission (i.e. the casemix) to hospital, whether public or private hospitals were used and the variation in
costs by size and type of hospital. However, in 2008-09, the average cost per separation adjusted for
differences in casemix (and excluding depreciation) for a range of selected public hospitals was $4,471.
Given the focus is on socio-economically disadvantaged individuals and those living in the poorest
households, it is highly likely that the majority of hospital visits would have occurred in public hospitals.
Thus, a reduction of nearly 513,000 separations at an average cost of $4,471 would give a total savings of
nearly $2.3 billion each year. This is equivalent to 5 per cent of Australia’s total expenditure on hospitals.

6.2 REDUCED USE OF DOCTOR AND MEDICAL RELATED SERVICES

In 2008-09, there were over 294 million doctor- and medical-related services subsidised through
Australia’s Medicare Benefits Schedule (MBS) at a cost to Government of $14.3 billion. Nearly 25 per cent
of these were by Australians aged 25 to 44 (males 21.8m or 7.4% MBS services; females 45.3m or 15.4%
services) and over 30 per cent by individuals aged 45 to 64 (males 38.7m or 13.2% services; females
50.4m or 17.1% services). Visits to GPs are a major component of this service use. For both younger and
older females (of working age), visits to GPs account for around 30 per cent of all MBS doctor and medical
related services. For males aged 25 to 44, GP attendances contribute to 38 per cent of all MBS services
and for the older males 29 per cent.

Results from NATSEM’s health module in the APPSIM dynamic microsimulation model show there is a
two-fold difference in the number of MBS services used on average in 2008 between disadvantaged
persons in poor versus good health (Table 20). Use by females outstrips males especially for younger
women in child-bearing age (25-44). If 396,986 individuals in the bottom income quintile households
changed their health status from poor to good then the number of MBS services used in 2008 would have
been reduced by over 5.5 million services. The reduction in service use is most noticeable for both males
and females aged 45 to 64 and with almost 1 million services potentially not needed for females aged 25
to 44.

Focussing on Government expenditure on the benefits paid from the public purse for doctor and
medically related services (for which cost data are available by age and gender although not by health or
socio-economic status), then this reduction in MBS service use would equate to a savings to Government
of around $273 million annually (Table 21).




Table 20 Estimated number of doctor and medically related services used in 2008 by
Australians of working age in the bottom income quintile and reductions in MBS
services through closing the health gap between most and least disadvantaged
Australians of working age



Number of Disadv. Persons

Ave No MBS
Services per
Disadv. Person

No. MBS Services (‘000)

Reduction
in MBS
Services
(‘000)



In Poor
Health

Remain
in poor
health

Shift to
good
health

Poor
Health

Good
Health

Disadv
Persons
in Poor
Health

Disadv
Persons
Remain
in poor
health

Disadv
Persons
Shift to
Good
Health

Male 25-44

70,158

20,294

49,864

16.5

6.1

1,157.6

334.9

304.2

518.5

Male 45-64

188,624

51,735

136,889

27.4

12.3

5,168.3

1,417.5

1,683.7

2,067.1

Female 25-44

88,084

30,178

57,906

30.4

13.7

2,677.8

917.4

793.3

967.1

Female 45-64

218,833

66,506

152,327

30.2

17.1

6,608.8

2,008.5

2,604.8

1,995.5

Total

565,699

168,713

396,986

-

-

15,612.5

4,678.3

5,386.0

5,548.2



Source Data: NATSEM’s Microsimulation model ‘APPSIM’



Table 21 Estimated MBS benefits in 2008 for Australians of working age in the bottom income
quintile and savings in MBS benefits through closing the health gap between most
and least disadvantaged Australians of working age



Ave Benefit
per MBS
service ($)

MBS Benefits ($m)

Savings in
MBS Benefits
($m)



Disadv
Persons In
Poor Health

Disadv
Persons
Remain in
Poor Health

Disadv
Persons
Shift to
Good Health

Male 25-44

46.32

53.6

15.5

14.1

24.0

Male 45-64

48.61

251.2

68.9

81.8

100.5

Female 25-44

52.65

141.0

48.3

41.8

50.9

Female 45-64

49.08

324.4

98.6

127.8

98.0

Total

-

770.2

231.3

265.5

273.4



Source Data: http://www.medicareaustralia.gov.au/provider/medicare/mbs.jsp



6.3 REDUCED USE OF PRESCRIBED MEDICINES

In 2008-09, more than 181 million prescriptions were subsidised under the Pharmaceuticals Benefits
Scheme (PBS) at a total cost to Government of over $6.6 billion. Government expenditure accounted for
83.4 per cent of the total cost of PBS prescriptions, the remaining cost being met by consumer out-of-
pocket co-payments. Around 10 per cent of PBS medicines are used by individuals aged 25 to 44 and 30
per cent by those aged 45 to 64. Persons on low income as measured by having an Australian Government
Pensioner Concession Card, Commonwealth Seniors Health Card, DVA White, Gold or Orange Card, or
Health Care Card are eligible for receiving PBS medicines at a concessional rate. Out-of-pocket co-
payments are reduced and the safety net threshold is lower, beyond which Government meets the full
cost of the medicines. Data from NATSEM’s MediSim microsimulation model shows that over 85 per cent
of individuals aged 25 to 64 and who are in the bottom income quintile access PBS medicines at
concessional rates, irrespective of whether they are in good or poor health (Table 22). Concessional and


general patients have very different patterns of prescription medicine use and costs, and therefore it was
important to split the number of individuals modelled into these two patient groups.

Males and females aged 45 to 64 who were in poor health and were concessional patients had an average
of 30 and 33 prescriptions filled respectively in 2008. In contrast, males and females of the same age but
who were in good health and were general patients (e.g. among the ‘working poor’ but not meeting
income and asset tests to be eligible for concessional rates) used only 12 and 11 scripts on average.
General patients aged 25 to 44 with good health filled as few as four or seven scripts depending on
whether they were female or male (Table 22). A shift to good health through closing socio-economic
health gaps will shift some persons in low-income households from ‘having’ to ‘not having’ concessional
status (e.g. through changes in their employment status and household earnings). More than 4,300
additional males aged 25 to 44 and some 11,500 males aged 45 to 64 would lose their concessional status
and become general patients. In contrast, females aged 25 to 44 who are in good health are fractionally
more likely to be concessional patients than those in poor health, hence with improvements in health
status more younger adult females (around 2,400 individuals) become PBS concession cardholders. These
changes in concessional status impact on potential reductions in script volumes and costs.

For example, over 5.3 million scripts would not have been dispensed for concessional patients if health
equity had been achieved, but there would have been a net increase of 41,500 scripts for general patients
(Table 22). This reflects a 2.6-fold increase in scripts for males aged 25 to 44 and 1.6-fold increase for
males aged 45 to 64 in the general patient group (higher proportions of males in good health are general
patients than females).

If these changes in script volume were achieved,what changes might occur in Government and consumer
out-of-pocket expenditure on the PBS? The findings are given in Tables 24 and 25. The results are based
on cost estimates from MediSim. As an indicator of the reliability of the MediSim data, the MediSim costs
were aggregated by age, gender and health status to provide overall costs for concessional and general
patients and compared with available administrative PBS data compiled by Medicare Australia (DoHA,
2010). As shown in Table 23, there is a good concordance between the two data sources noting that the
administrative data is for the total population (age-sex specific data was not available) and MediSim
output is for the 25- to 64-year-old age group.

There is little difference in the average Government benefit paid per script to concessional patients by
age, gender or health status (Table 24). The cost of a PBS script on average to Government is slightly
higher for both males and females aged 25 to 44 who are in good health overall compared with those in
poor health. The opposite occurs for those aged 45 to 64 with the average cost of a script to Government
being higher for those in poor health. Improvement in health status for concessional patients would yield
substantial savings to Government – an estimated $184.7 million.


Table 22 Estimated number of PBS scripts used in 2008 by Australians of working age in the bottom income quintile and reductions in PBS script
volume through closing the health gap between most and least disadvantaged Australians of working age





% Concessional
or General
Patient

Number of Disadv. Persons

Average No of
PBS Scripts per
Disadv. person

No. PBS Scripts (‘000)

Reduction in
PBS Scripts
(‘000)



In Poor
Health

In Good
Health

In Poor
Health

Remain
in poor
health

Shift to
good
health

In Poor
Health

In Good
Health

Disadv.
Persons In
Poor Health

Disadv.
Persons
Remain in
Poor Health

Disadv.
Persons
Shift to
Good Health

Concessional























Male 25-44

96.8

88.1

67913

19645

43930

19

7

1,290.3

373.2

307.5

609.6

Male 45-64

93.9

85.5

177118

48579

117040

30

12

5,313.5

1,457.4

1,404.5

2,451.6

Female 25-44

87.7

91.8

77250

26466

53158

19

8

1,467.7

502.9

425.3

539.5

Female 45-64

88.2

88.2

193011

58658

134352

33

20

6,369.4

1,935.7

2,687.0

1,746.7

Total





515292

153348

348480





14440.9

4269.2

4824.3

5347.4

General























Male 25-44

3.2

11.9

2245

649

5934

8

7

18.0

5.2

41.5

-28.7

Male 45-64

6.1

14.5

11506

3156

19849

16

12

184.1

50.5

238.2

-104.6

Female 25-44

12.3

8.2

10834

3712

4748

8

4

86.7

29.7

19.0

38.0

Female 45-64

11.8

11.8

25822

7848

17975

14

11

361.5

109.9

197.7

53.9

Total





50407

15365

48506





650.3

195.3

496.4

-41.4

All persons





565,699

168,713

396,986





15,091.2

4,464.5

5,320.7

5,306.0



Source Data: NATSEM’s microsimulation model MediSim


Table 23 Comparison of MediSim and Medicare Australia average costs of PBS scripts



Government
Cost ($)

Patient
Copayment ($)

Total Cost
($)

Concessional







PBS MA Data*

33.04

3.98

37.02

MediSim+

32.93

3.92

36.85

General







PBS MA Data

54.61

26.26

80.87

MediSim

57.44

24.38

81.82



* for the total population

+ for the population aged 25-64 years



For general patients, the picture is more complicated. There are very different average Government script
costs combined with increasing numbers of male general patients, but a reduced number of younger
female general patients. The net effect is that for three of the four age-sex general patient groups, total
Government expenditure would likely rise. These increases in Government costs are offset in the
modelling by savings from female general patients aged 25 to 44. However, this could be artificially
inflated as female general patients aged 25 to 44 who are in poor health appear to receive a very high
average benefit per script (Table 24). Nevertheless, any rise in Government expenditure on general
patients would not outweigh the savings from reduced script use by concessional patients.

Likely changes in out-of-pocket payments by consumers are shown in Table 25. At January 1, 2009, PBS
co-payments were set at $5.30 per script for concessional patients and $32.90 for general patients. The
average co-payments in Table 25 are lower because they take into account scripts dispensed ‘above’ the
safety net thresholds. Concessional patients reaching the safety net have any additional scripts, i.e. above
the safety net, dispensed at no out-of-pocket cost and for general patients the co-payment reduces to the
concessional rate (i.e. $5.30). If health equity was achieved for concessional patients then there would be
a $15.6 million reduction in out-of-pocket costs. However, there would be an increase in the cost to
general patients by some $3.1m.






















Table 24 Estimated Government expenditure on PBS medicines in 2008 for Australians of
working age in the bottom income quintile and savings in benefits through closing
the health gap between most and least disadvantaged Australians of working age



Ave Benefit per PBS script
($)

PBS Benefit ($m)

Savings
in PBS
Benefits
($m)

Disadv
Persons In
Poor Health

Disadv
Persons In
Good Health

Disadv
Persons In
Poor Health

Disadv
Persons
Remain in
Poor Health

Disadv
Persons
Shift to
Good Health

Concessional













Male 25-44

31.82

34.74

41.1

11.9

10.7

18.5

Male 45-64

35.00

30.57

186.0

51.0

42.9

92.1

Female 25-44

29.28

32.95

43.0

14.7

14.0

14.3

Female 45-64

33.03

32.28

210.4

63.9

86.7

59.8

Total

33.35

32.34

480.5

141.5

154.3

184.7

General













Male 25-44

24.49

28.05

0.4

0.1

1.2

-0.9

Male 45-64

54.20

58.79

10.0

2.7

14.0

-6.7

Female 25-44

191.52

26.80

16.6

5.7

0.5

10.4

Female 45-64

43.19

63.77

15.6

4.7

12.6

-1.7

Total

63.39

53.31

42.6

13.3

28.3

1.1

All persons





523.1

154.7

182.6

185.8



Source Data: NATSEM’s microsimulation model MediSim

Table 25 Estimated patient co-payments to PBS medicines in 2008 by Australians of working
age in the bottom income quintile and savings in PBS patient costs through closing
the health gap between most and least disadvantaged Australians of working age



Ave copayment per PBS
script ($)

Copayment ($m)

Savings in
PBS Co-
payments
($m)

Disadv
Persons In
Poor Health

Disadv
Persons In
Good Health

Disadv
Persons In
Poor Health

Disadv
Persons
Remain in
Poor health

Disadv
Persons
Shift to
Good Health

Concessional













Male 25-44

4.06

4.66

5.2

1.5

1.4

2.3

Male 45-64

3.47

4.37

18.4

5.1

6.1

7.2

Female 25-44

4.16

4.61

6.1

2.1

2.0

2.0

Female 45-64

3.50

4.26

22.3

6.8

11.4

4.1

Total

3.60

4.38

52.0

15.4

21.0

15.6

General













Male 25-44

29.45

27.67

0.5

0.2

1.1

-0.8

Male 45-64

24.36

24.36

4.5

1.2

5.8

-2.5

Female 25-44

25.42

26.75

2.2

0.8

0.5

0.9

Female 45-64

18.71

27.17

6.8

2.1

5.4

-0.7

Total

21.57

26.33

14.0

4.3

12.8

-3.1

All persons





66.0

19.7

33.8

12.5



Source Data: NATSEM’s microsimulation model MediSim




7 SUMMARY AND CONCLUSIONS

Social gradients in health are common in Australia – the lower a person’s social and economic position,
the worse his or her health – and the health gaps between the most disadvantaged and least
disadvantaged groups are typically very large. This Report confirms that the cost of Government inaction
on the social determinants of health leading to health inequalities for the most disadvantaged Australians
of working age is substantial. This was measured in terms not only of the number of people affected but
also their overall well-being, their ability to participate in the workforce, their earnings from paid work,
their reliance on Government income support and their use of health services.

Health inequality was viewed through a number of different socio-economic lenses – household income,
education, housing tenure and social connectedness – with attention being focussed on the health gaps
between the most and least disadvantaged groups. The cost of Government inaction was measured in
terms of the loss of potential social and economic gains that might otherwise have accrued to socio-
economically disadvantaged individuals if they had had the same health profile of more socio-
economically advantaged Australians. The modelling ‘shifted’ disadvantaged individuals from poor to
good health, or having to not having a long-term health condition, to replicate the health profile of the
least disadvantaged group. It was assumed that any ‘improvement’ in health did not move individuals out
of their socio-economic group but rather that they took on the socio-economic characteristics of those in
the group who were ‘healthy’.

If the health gaps between the most and least disadvantaged groups were closed, i.e. there was no
inequity in the proportions in good health or who were free from long-term health conditions, then an
estimated 370,000 to 400,000 additional disadvantaged Australians in the 25-64 year age group would see
their health as being good and some 405,000 to 500,000 additional individuals would be free from chronic
illness, depending upon which socio-economic lens (household income, level of education, social
connectedness) is used to view disadvantage. Even if Government action focussed only on those living in
public housing, then some 140,000 to 157,000 additional Australian adults would have better health.

Substantial differences were found in the proportion of disadvantaged individuals satisfied with their
lives, employment status, earnings from salary and wages, Government pensions and allowances, and use
of health services between those in poor versus good health and those having versus not having a long-
term health condition. As shown in the Report findings, improving the health profile of Australians of
working age in the most socio-economically disadvantaged groups therefore leads to major social and
economic gains with savings to both the Government and to individuals.

For example, as many as 120,000 additional socio-economically disadvantaged Australians would be
satisfied with their lives. For some of the disadvantaged groups studied, achieving health equality would
mean that personal well-being would improve for around one person in every 10 in these groups. Rates of
unemployment and not being in the labour force are very high for both males and females in low socio-
economic groups and especially when they have problems with their health. For example, in 2008, fewer
than one in five persons in the bottom income quintile and who had at least one long-term health
condition was in paid work, irrespective of their gender or age. Changes in health reflect in higher
employment rates, especially for disadvantaged males aged 45 to 64. Achieving equity in self-assessed
health status could lead to more than 110,000 new full- or part-time workers when health inequality is
viewed through a household income lens, or as many as 140,000 workers if disadvantage from an


educational perspective is taken. These figures rise to more than 170,000 additional people in
employment when the prevalence of long-term health conditions is considered.

If there are more individuals in paid work then it stands to reason that the total earnings from wages and
salaries for a particular socio-economic group will increase. The relative gap in weekly gross income from
wages and salaries between disadvantaged adult Australians of working age in good versus poor health
ranges between a 1.5-fold difference for younger males (aged 25-44) who live in public housing or who
experience low levels of social connectedness to over a staggering 6.5-fold difference experienced by
males aged 45 to 64 in the bottom income quintile or who are public housing renters.

Closing the gap in self-assessed health status could generate as much as $6-7 billion in extra earnings, and
in the prevalence of long-term health conditions upwards of $8 billion. These findings reflect two key
factors – the large number of Australians of working age who currently are educationally disadvantaged
having left school before completing year 12 or who are socially isolated and the relatively large wage gap
between those in poor and good health in these two groups. In terms of increases in annual income from
wages and salaries, the greatest gains from taking action on the social determinants of health can be
made from males aged 45 to 64.

A flow-on effect from increased employment and earnings and better health is the reduced need for
income and welfare support via Government pensions and allowances. Those in poor health or who have
a long-term health condition typically received between 1.5 and 2.5 times the level of financial assistance
from Government than those in good health or who were free from chronic illness. Irrespective of
whether an income, education or social exclusion lens is taken, closing the gap in health status potentially
could lead to $2-3 billion in savings per year in Government expenditure, and in the order of $3-4 billion
per year if the prevalence of chronic illness in most disadvantaged socio-economic groups could be
reduced to the level experienced by the least disadvantaged groups.

Potential savings to the health system through Government taking action on the social determinants of
health were difficult to estimate because of the lack of socio-economic coded health services use and cost
data. As an example of the possible savings that might accrue, changes in the use and cost of health
services – hospitals, doctor and medically related (Medicare) services, and prescribed medicines
subsidised through the PBS – from changes in self-assessed health status for individuals in the lowest
household income quintile were modelled.

Nearly 400,000 additional disadvantaged individuals would regard their health as good if equity was
achieved with individuals in the top income quintile. Such a shift was shown to be significant in terms of
health services use and costs as there were very large differences in the use of health services by
individuals in the bottom income quintile between those in poor versus good health. More than 60,000
individuals need not have been admitted to hospital. More than 500,000 hospital separations may not
have occurred and, with an average length of stay of around 2.5 days, there would have been some 1.44
million fewer patient days spent in hospital, saving around $2.3 billion in health expenditure.

A two-fold difference in the use of doctor and medical services was found between disadvantaged
persons in poor versus good health. Improving the health status of 400,000 individuals of working age in
the bottom income quintile would reduce the pressure on Medicare by over 5.5 million services. Such a
reduction in MBS service use equates to a savings to Government of around $273 million annually. With
respect to the use of prescription medicines, in 2008, disadvantaged individuals in the 45- to 64-year-old
age group and who were in poor health and who were concession cardholders used 30 prescriptions on
average each. While those aged 25 to 44 averaged 19 scripts, both age groups used twice as many scripts


as concessional patients in good health. Over 5.3 million PBS scripts would not have been required by
concessional patients if health equity existed. However, a shift to good health through closing socio-
economic health gaps would shift around 15,000 persons in low-income households from ‘having’ to ‘not
having’ concessional status, resulting in a net increase of 41,500 scripts (a 6 per cent increase) for general
patients. Health equity for concessional patients was estimated to yield $184.7 million in savings to
Government and a $15.6m reduction in patient contributions. However, there would be an increase in the
out-of-pocket cost of medicines to general patients by some $3.1m.

This is the first study of its kind in Australia that has tried to gauge the impact of Government inaction on
the social determinants of health and health inequalities. Reducing health inequalities is a matter of social
inclusion, fairness and social justice (Marmot et al, 2010). The fact that so many disadvantaged
Australians are in poor health or have long-term health conditions relative to individuals in the least socio-
economically disadvantaged groups is simply unfair. So are the impacts on people’s satisfaction with their
lives, missed employment opportunities, levels of income and need for health services. This study shows
that major social and economic benefits are being neglected and savings to Government expenditure and
the health system overlooked. The findings of this Report are revealing and are of policy concern
especially within the context of Australia’s agenda on social inclusion. However, in this study the health
profile of individuals of working age in the most socio-economic disadvantaged groups only was
compared with that of individuals in the least disadvantaged groups. The first CHA-NATSEM Report
(Brown et al, 2010) on health inequalities showed that socio-economic gradients in health exist in
Australia. It is not only the most socio-economically disadvantaged groups that experience health
inequalities relative to the most advantaged individuals, but also other low and middle socio-economic
groups. Thus, this Report provides only part of the story of health inequalities in Australians of working
age.

Socio-economic inequalities in health persist because the social determinants of health are not being
addressed. Government action on the social determinants of health and health inequalities would require
a broad investment, a focus on health in all policies and action across the whole of society. In return,
significant revenue would be generated through increased employment, reduction in Government
pensions and allowances, and savings in Government spending on health services. The WHO Commission
on the Social Determinants of Health called for national governments to develop systems for the routine
monitoring of health inequities and the social determinants of health, and develop more effective policies
and implement strategies suited to their particular national context to improve health equity
(http://www.who.int/social_determinants/en/ ). This Report continues the work of demonstrating how
improving health equity could have a major impact on the health and well-being of Australians, as well as
a significant financial impact for the country.


References

Abello, A., and Brown, L. (2007). Model 18: MediSim (Static Microsimulation Model of the Australian Pharmaceutical
Benefits Scheme) In Modelling our Future: Population Ageing, Health and Aged Care, Gupta, A. and Harding,
A. (Eds), Elsevier, Amsterdam, International Symposia in Economic Theory and Econometrics, Volume 16, pp.
533-542.

Australian Bureau of Statistics (2010) The Health and Welfare of Australia's Aboriginal and Torres Strait Islander
Peoples, 2010, cat. no. 4704.0, ABS, Canberra.

Australian Bureau of Statistics (2009) Experimental life tables for Aboriginal and Torres Strait Islander Australians,
2005-2007, cat. no. 3302.0.55.003, ABS, Canberra.

Australian Bureau of Statistics (2008) An Introduction to Socio-Economic Indexes for Areas (SEIFA) 2006. Information
Paper, cat. no. 2039.0, ABS, Canberra.

Australian Bureau of Statistics (2007) 2007 Year Book Australia, cat. no. 1301.0, ABS Canberra.

Australian Institute of Health and Welfare (2011) Australia’s hospitals 2009–10: at a glance. Health services series
no.39. Cat. no. HSE 106. Canberra: AIHW.

Australian Institute of Health and Welfare (2011). National Social Housing Survey: a summary of national results.
AIHW bulletin no. 92. Cat. no. AUS 141. Canberra: AIHW. Viewed 24 April 2012
<http://www.aihw.gov.au/publication-detail/?id=10737420122>.

Australian Institute of Health and Welfare (2010a) Australia’s health 2010, Australia’s health series no. 12, cat. no.
AUS 122, AIHW, Canberra.

Australian Institute of Health and Welfare (2010b) Australia’s health 2010 in brief, AIHW, Canberra.

Australian Institute of Health and Welfare (2010). Australian hospital statistics 2008–09. Health services series no.
17. Cat. no. HSE 84. Canberra: AIHW.

Baum, F. (2008). The New Public Health, Oxford University Press, 3rd edition, Melbourne

Begg S., Vos T., Barker B., Stevenson C., Stanley L. and Lopez A.D. (2007) The burden of disease and injury in
Australia 2003, cat. No. PHE 82, AIHW, Canberra.

Brown, L.J., Abello, A., Thurecht, L. (2011) Length of Hospital Stay by Older Australians: Bed-blocking or Not?
NATSEM Working Paper WP8, NATSEM, University of Canberra.

Cai, L. (2009) Be wealthy to stay healthy: An analysis of older Australians using the HILDA survey, Journal of
Sociology, vol.45, no. 1, pp. 55-70.

Cai, L. and Kalb, G. (2006) Health status and labour force participation: evidence from Australia, Health Economics,
vol.15, no. 3, pp. 241-261.

Commission on Social Determinants of Health (2008) Closing the gap in a generation: health equity through action on
the social determinants of health, Final Report of the Commission on Social Determinants of Health, World
Health Organization, Geneva.

Department of Health and Ageing (2010). Expenditure and prescriptions twelve months to 30 June 2009 Data and
Modelling Section, Pharmaceutical Policy and Analysis Branch, DoHA, Canberra.

Department of Health and Ageing (1999) National Physical Activity Guidelines for Australians, Australian
Government, Canberra.

Hayes, A., Gray, M. and Edwards, B. (2008), Social Inclusion: Origins, Concepts and Key Themes, Social Inclusion
Unit, Department of the Prime Minister and Cabinet, Canberra.

Gaston , N. and Rajaguru , G. (2009) The long run determinants of Australian income inequality. Economic Record,
vol. 85, no. 3, pp. 260-275, September 2009.

Last, J.M. (1988) A Dictionary of Epidemiology. 2nd Ed. Oxford University Press, New York.

Laverty, M. (2009) The central place of Health in Australia’s Social Inclusion Agenda: Addressing the social determinants
of health to achieve social inclusion, Policy Paper, Catholic Health, Canberra.

Lymer, S. (2011) More Than Just Population Ageing: An Evaluation Using Dynamic Microsimulation Of The Escalation
Of Australian Government Expenditure On Health Over The Next 40 Years. PhD Thesis, NATSEM, University of
Canberra, Canberra.

Marmot, M. and Wilkinson, R. (eds) (2003) Social Determinants of Health: The Solid Facts, World Health
Organisation, Geneva.


Marmot, M., Allen, J., Goldblatt, P., Boyce, T., McNeish, D., Grady, M. and Geddes. I. (2010) Fair Society, Healthy
Lives – the Marmot Review, Strategic Review of Health Inequalities in England post-2010, London.

Meagher, G. and Wilson, S. (2008) Richer, but More Unequal: Perceptions of Inequality in Australia 1987-2005,
Journal of Australian Political Economy, vol. 61, pp. 220-243.

Miranti, R., McNamara, J., Tanton, R. and Harding, A. (2010) Poverty at the Local Level: National and Small Area
Poverty Estimates by Family Type for Australia in 2006, Applied Spatial Analysis, published online March 2010.

National Health and Medical Research Council (2009) Australian Guidelines to Reduce Health Related Risk from
Drinking Alcohol, Commonwealth of Australia, Canberra.

National Public Health Partnership (2001) Preventing Chronic Disease: A Strategic Framework, Melbourne.

Nepal, B., Payne, A., and Brown, L. (2009) Healthy, wealthy and wise? The relationship between health,
employment and earnings in Australia, AMP.NATSEM Income and Wealth Report, no. 23, July, AMP, Sydney.

Raphael, D. (2004). Social Determinants of Health: Canadian Perspectives, Canadian Scholar’s Press, Toronto, Canada.

Saunders, P., Hill, T. and Bradbury, B. (2008) Poverty in Australia: Sensitivity Analysis and Recent Trends, Social
Policy Research Centre Report, no. 4.

Tanton, R., McNamara, J., Harding, A. and Morrison, T. (2009) Rich suburbs, Poor suburbs? Small area poverty
estimates for Australia's Eastern Seaboard in 2006, in A. Zaidi, A. Harding, & P. Williamson (eds), New
Frontiers in Microsimulation Modelling, Ashgate, London

Tanton, R., Nepal, B. and Harding, A. (2008) Wherever I lay my debt, that’s my home. Trends in housing affordability
and housing stress 1995-96 to 2005-06, AMP.NATSEM Income and Wealth Report, no. 19, March, Sydney, AMP.

Turrell G., Oldenburg B., McGuffog I. and Dent R. (1999) Socioeconomic determinants of health: towards a national
research program and a policy and intervention agenda. Queensland University of Technology, School of Public
Health, Ausinfo, Canberra.

Victoria Department of Health (2010) Life expectancy at birth: Victoria 2003 - 2007,
http://www.health.vic.gov.au/healthstatus/le-03-07.htm, Accessed 1 July 2010.

Vinson, T. (2007). Dropping off the edge: The distribution of disadvantage in Australia. Richmond, Victoria: Jesuit Social
Services and Catholic Social Services Australia.

Vos, T., Begg, S., Chen, Y. and Magnus, A. (2001). Socioeconomic differentials in life expectancy and years of life
lost in Victoria,1992-1996, New South Wales Public Health Bulletin, 12(5), 126 – 130.

Vu, Q.N., Harding, A., Tanton, R., Nepal, B. and Vidyattama, Y. (2008) Advance Australia Fair? Trends in small
area economic inequality 2001 to 2006. APM.NATSEM Income and Wealth Report, no. 20, July 2008, AMP,
Sydney.

Watson, N. (ed) (2010) HILDA User Manual – Release 8, Melbourne Institute of Applied Economic and Social
Research, University of Melbourne.






















APPENDIX 1 - TECHNICAL NOTES

(a) Self-assessed health status

Self-assessed health is a key health variable analysed in this study. This variable represents the standard
self-assessed health status collected through the self-completed questionnaire. The question asked was:
“In general, would you say that your health is: excellent, very good, good, fair or poor?” Respondents
reported their health to be in any of the five levels. For the ease of analysis and interpretation, we have
grouped these five levels into two: “good health” and “poor health”. “Good health” includes excellent,
very good and good health; and “poor health” refers to fair and poor health. Non-response cases were
excluded from the analysis.

Use of self-assessed health status has some merits and some demerits that need to be taken into account
while interpreting the results presented in this report. This is an easily available indicator of health status
from socioeconomic surveys and this provides an opportunity to relate this indictor to various socio-
economic measures. The self-assessed health indicator has been widely used in the empirical research of
health status because it has been found to reflect the true health status of individuals reasonably well. A
number of previous Australian studies of relationships between health and socio-economic issues have
satisfactorily used this indicator (Cai and Kalb, 2006; Cai, 2009; Nepal, 2009). Yet the data for this indicator
come from individual’s perception rather than clinical assessment of their health. Therefore this measure
cannot be expected to be identical to an objective measure of health status.

(b) Long-term health condition

In the HILDA survey, data on long-term health conditions was collected through individual interview. The
question was: Looking at SHOWCARD K1, do you have any long-term health condition, impairment or
disability (such as these) that restricts you in your everyday activities, and has lasted, or is likely to last, for
six months or more?

(c) Income quintile

The income quintile used is the equivalised disposable household income quintile. HILDA data files
provided disposable income in the previous financial year that was calculated by applying a tax module to
the reported incomes:

In order to produce the disposable income variable, an income tax model is applied to each sample
member that calculates the financial-year tax typically payable for a permanent resident taxpayer in
the circumstances akin to those of the respondent. The information collected in the HILDA Survey does
not permit accounting for every individual variation in tax available under the Australian taxation
system, but most major sources of variation are accounted for. When aggregated, income tax
estimates from HILDA compare favourably with national aggregates produced by the Australian
Taxation Office (ATO). (Watson, 2010, p46).

Before calculating the equivalised disposable household income quintiles, negative income was set to
zero. Using the full sample of responding households, equivalent scale was calculated as 1 + (number of
remaining adults × 50%) + (number of children under 15 years × 30%). Total disposable household income
was divided by the equivalence scale to derive equivalised household income. Income is equivalised to
take account of the fact that two-person households do not need twice the amount of resources of a
single-person household, for example.


(d) Social connectedness

The indicator called social connectedness reflects the degree to which an individual is connected to the
family, friends and society. The indicator was derived on the basis of responses to the following three
questions or statements posed in a self-completed questionnaire:



i) How often get together socially with friends/relatives not living with you
ii) I don’t have anyone that I can confide in
iii) I often feel very lonely


Responses were sought in an ordinal scale of 1 to 7 (better to worse). The first three scales were
considered as reflecting a high score and the remaining a low score for the purpose of this study. Having a
high score in all these three dimensions was classified as high connectedness, a high score in any two
dimensions as moderate connectedness and just one or no high score as reflecting low connectedness.

(e) Public Housing



Public housing encompasses publicly owned or leased dwellings administered by State and Territory
Governments. It includes all rental housing owned and managed by Government. Public housing provides
affordable and accessible housing for largely low-income households who are in housing need. Public
housing and community housing are collectively referred to as ‘social housing’ (AIHW, 2011).





Table A.1 Sample size and population by analysis variables, persons aged 25-64 years

Variables

N

Population (thousands)

Self-assessed health status

7,178

9,520

Long-term health condition

8,217

11,293

Housing

7,086

9,844

Connectedness

7,164

9,496

Other SES

8,217

11,293



Source: HILDA Wave 8 datefile