Acr 2 D0 F

October 2006
The economic costs of obesity
Report by Access Economics Pty Limited to
Diabetes Australia
TABLE OF CONTENTS
Acknowledgements...............................................................................................................i
Glossary of common abbreviations....................................................................................ii
Executive summary.............................................................................................................iii
1. Introduction .................................................................................................................1
1.1 Overview of this report ............................................................................................................1
1.2 Cross-cutting methodological issues ......................................................................................1
2. Obesity.........................................................................................................................7
2.1 Definitions and measures .......................................................................................................7
2.2 Risks .......................................................................................................................................9
2.3 Aetiology ...............................................................................................................................12
2.4 Prevalence in Australia .........................................................................................................16
2.5 Prevalence trends and projections .......................................................................................20
3. Health Impacts of Obesity ........................................................................................27
3.1 Diabetes ................................................................................................................................27
3.2 CVD.......................................................................................................................................31
3.3 Osteoarthritis.........................................................................................................................35
3.4 Cancers.................................................................................................................................38
4. Health Expenditures..................................................................................................42
4.1 Methodology .........................................................................................................................42
4.2 Diabetes ................................................................................................................................42
4.3 CVD.......................................................................................................................................45
4.4 Osteoarthritis.........................................................................................................................47
4.5 Cancer ..................................................................................................................................50
5. Other (Non-health) Financial Costs .........................................................................53
5.1 Methodology .........................................................................................................................53
5.2 Diabetes ................................................................................................................................58
5.3 CVD.......................................................................................................................................66
5.4 Osteoarthritis.........................................................................................................................71
5.5 Cancer ..................................................................................................................................74
6. Burden of Disease.....................................................................................................87
6.1 Methodology – Valuing life and health..................................................................................87
6.2 Diabetes ................................................................................................................................90
6.3 CVD.......................................................................................................................................92
6.4 Osteoarthritis.........................................................................................................................95
6.5 Cancer ..................................................................................................................................96
Disclaimer
While every effort has been made to ensure the accuracy of this document, the uncertain nature of economic data,
forecasting and analysis means that Access Economics Pty Limited is unable to make any warranties in relation to the
information contained herein. Access Economics Pty Limited, its employees and agents disclaim liability for any loss or
damage which may arise as a consequence of any person relying on the information contained in this document.
7. Cost Summary and Interventions.............................................................................98

7.1 Summary of the cost of obesity ............................................................................................ 98
7.2 Summary of the cost of diabetes........................................................................................ 100
7.3 The microeconomics of obesity.......................................................................................... 102
7.4 Taxation, regulation and subsidised programs .................................................................. 104
7.5 Pharmacological and surgical interventions....................................................................... 108
REFERENCES...................................................................................................................111
TABLE OF FIGURES
Figure 1-1: Incidence and prevalence approaches to measurement of annual costs 2

Figure 2-1: Prevalence of Obesity, 2005 19
Figure 2-2: Trends in Obesity prevalence (%) for adults, 1980 to 2000 20
Figure 2-3: Trends in self-reported Obesity prevalence (%), 1995 to 2004-05 21
Figure 2-4: Trends in Obesity prevalence rates for NSW school children: 1985-2004 23
Figure 2-5: Prevalence projections to 2025, Baseline 24
Figure 2-6: Prevalence projections to 2025, Elimination 25
Figure 2-7: Prevalence projections to 2025, Growth 26
Figure 3-1: Prevalence (%) of Type 1 and 2 Diabetes by gender, Australia, 1999 30
Figure 3-2: Age-adjusted Relative Mortality Risk, by BMI for US sample Populations 32
Figure 3-3: Prevalence of Obesity related CVDs, 2005, by Age and Gender 33
Figure 3-4: Prevalence of Obesity in Selected CVD Populations, 2005 34
Figure 3-5: Prevalence of Osteoarthritis, in 2005, by Age, Gender and Weight Range 37
Figure 3-6: Prevalence of Obesity related Cancers, 2005, by Age and Gender 39
Figure 3-7: Prevalence and AFs of Obesity in Selected Cancer Populations, 2005 40
Figure 4-1: Type 2 Diabetes, allocated health expenditure by cost type cost, 2005 (%) 44
Figure 4-2: Type 2 Diabetes, allocated health expenditure by Age & Gender, 2005 ($m) 44
Figure 4-3: CVD health costs, 2005, $’000, by Age, Gender and Weight 46
Figure 4-4: CVD health expenditure due to Obesity by cost type, 2005 (% total) 47
Figure 4-5: Osteoarthritis, Total health expenditure, 2005, by Age, Gender & Weight 49
Figure 4-6: Osteoarthritis, health expenditure due to Obesity by cost type and gender, 2005
(%) 50
Figure 4-7: Cancer Health System Costs, 2005, $’000, by Age, Gender and Weight 52
Figure 5-1: Productivity Losses 54
Figure 5-2: DWL of Taxation 57
Figure 5-3: Type 2 Diabetes, non-health financial cost summary, 2005 (% total) 66
Figure 5-4: Productivity and carer costs per Person by Age and Cancer Type ($) 81
Figure 5-5: Summary of Other Financial Costs by Cancer Type and Gender 86
Figure 6-1: Loss of wellbeing due to Type 2 Diabetes (DALYs), by age and gender, 2005 92
Figure 6-2: CVD Obesity Related BoD and Gross Cost of lost wellbeing, 2005 94
Figure 6-3: Osteoarthritis, BoD, 2005, by Age & Gender 95
Figure 6-4: Cancer, BoD, 2005, by Age and Gender 97
Figure 7-1: Financial costs of Obesity by type of cost, 2005 (% total) 98
Figure 7-2: Total costs of Obesity by bearer, 2005 (% total) 100
Figure 7-3: Financial costs of Obesity by bearer, 2005 (% total) 100
Figure 7-4: Financial cost of Diabetes by type, 2005 (% total) 101
Figure 7-5: Total cost of Diabetes by bearer, 2005 (% total) 102
Figure 7-6: Financial cost of Diabetes by bearer, 2005 (% total) 102
Figure 7-7: Externalities and the case for intervention 103
TABLE OF TABLES
Figure 1-1: Incidence and prevalence approaches to measurement of annual costs 2

Figure 2-1: Prevalence of Obesity, 2005 19
Figure 2-2: Trends in Obesity prevalence (%) for adults, 1980 to 2000 20
Figure 2-3: Trends in self-reported Obesity prevalence (%), 1995 to 2004-05 21
Figure 2-4: Trends in Obesity prevalence rates for NSW school children: 1985-2004 23
Figure 2-5: Prevalence projections to 2025, Baseline 24
Figure 2-6: Prevalence projections to 2025, Elimination 25
Figure 2-7: Prevalence projections to 2025, Growth 26
Figure 3-1: Prevalence (%) of Type 1 and 2 Diabetes by gender, Australia, 1999 30
Figure 3-2: Age-adjusted Relative Mortality Risk, by BMI for US sample Populations 32
Figure 3-3: Prevalence of Obesity related CVDs, 2005, by Age and Gender 33
Figure 3-4: Prevalence of Obesity in Selected CVD Populations, 2005 34
Figure 3-5: Prevalence of Osteoarthritis, in 2005, by Age, Gender and Weight Range 37
Figure 3-6: Prevalence of Obesity related Cancers, 2005, by Age and Gender 39
Figure 3-7: Prevalence and AFs of Obesity in Selected Cancer Populations, 2005 40
Figure 4-1: Type 2 Diabetes, allocated health expenditure by cost type cost, 2005 (%) 44
Figure 4-2: Type 2 Diabetes, allocated health expenditure by Age & Gender, 2005 ($m) 44
Figure 4-3: CVD health costs, 2005, $’000, by Age, Gender and Weight 46
Figure 4-4: CVD health expenditure due to Obesity by cost type, 2005 (% total) 47
Figure 4-5: Osteoarthritis, Total health expenditure, 2005, by Age, Gender & Weight 49
Figure 4-6: Osteoarthritis, health expenditure due to Obesity by cost type and gender, 2005
(%) 50
Figure 4-7: Cancer Health System Costs, 2005, $’000, by Age, Gender and Weight 52
Figure 5-1: Productivity Losses 54
Figure 5-2: DWL of Taxation 57

Figure 5-3: Type 2 Diabetes, non-health financial cost summary, 2005 (% total) 66
Figure 5-4: Productivity and carer costs per Person by Age and Cancer Type ($) 81
Figure 5-5: Summary of Other Financial Costs by Cancer Type and Gender 86
Figure 6-1: Loss of wellbeing due to Type 2 Diabetes (DALYs), by age and gender, 2005 92
Figure 6-2: CVD Obesity Related BoD and Gross Cost of lost wellbeing, 2005 94
Figure 6-3: Osteoarthritis, BoD, 2005, by Age & Gender 95
Figure 6-4: Cancer, BoD, 2005, by Age and Gender 97
Figure 7-1: Financial costs of Obesity by type of cost, 2005 (% total) 98
Figure 7-2: Total costs of Obesity by bearer, 2005 (% total) 100
Figure 7-3: Financial costs of Obesity by bearer, 2005 (% total) 100
Figure 7-4: Financial cost of Diabetes by type, 2005 (% total) 101
Figure 7-5: Total cost of Diabetes by bearer, 2005 (% total) 102
Figure 7-6: Financial cost of Diabetes by bearer, 2005 (% total) 102
Figure 7-7: Externalities and the case for intervention 103
ACKNOWLEDGEMENTS
This report was prepared by Access Economics to Diabetes Australia. It was funded by an
unrestricted grant from sanofi-aventis who had no part in the direction or findings contained in
this report. Access Economics would like to acknowledge with appreciation the comments,
prior research and expert input from:
Professor Stephen Colagiuri

Director, Department of Endocrinology and Diabetes,
Prince of Wales Hospital, Randwick, NSW
Mr Brian Conway
Executive Director, Diabetes Australia
i
GLOSSARY OF COMMON ABBREVIATIONS
ABS Australian Bureau of Statistics

AF Attributable Fraction
AIHW Australian Institute of Health and Welfare
ALSWH Australian Longitudinal Study of Women’s Health
AWE Average Weekly Earning
BMI Body Mass Index
BoD Burden of Disease
CAPANS Child and Adolescent Physical Activity and Nutrition Survey (WA)
CHD Coronary (ischaemic) Heart Disease
CVD Cardiovascular Disease
DALY Disability Adjusted Life Year
DSP Disability Support Pension
DWL Deadweight Loss
GP General Practitioner
HDL High-Density Lipoprotein
HHD Hypertensive Heart Disease
LDL Low-Density Lipoprotein
NHMRC National Health and Research Council
NHS National Health Survey
NNS National Nutrition Survey
NPV Net Present Value
NSW New South Wales
PAD Peripheral Arterial Disease
PWO People with Obesity
QALY Quality Adjusted Life Year
RAC Residential Aged Care
RR Relative Risk
SDAC Survey of Disability, Ageing and Carers (ABS)
SES Socioeconomic Status
SPANS Schools Physical Activity and Nutrition Survey (NSW)
VLY Value of a Life Year
VSL Value of a Statistical Life
WA Western Australia
YLD Years of Healthy Life Lost due to Disability
YLL Years of Life Lost due to Premature Mortality
ii
EXECUTIVE SUMMARY
In 2005, 3.24 million Australians were estimated to be obese – 1.52 million males (15.1%
of all males) and 1.72 million females (16.8% of all females).
Obesity is the accumulation of excessive fat in the body, defined here in terms of Body
Mass Index (BMI) over 30 for adults and, for children and adolescents aged 2 to 18
years, a set of age-gender specific BMI-thresholds are used. Obesity is linked to
genetic, perinatal, socioeconomic and other factors, but is primarily due to energy
imbalance.
BMI is defined as body weight (in kg) divided by height (in metres squared).
Whilst well accepted for people of Caucasian extraction, the definition of obesity is
not appropriate for application to other major ethnic groups in Australia, especially
people of Asian extraction, where it will otherwise underestimate 'true obesity'.
Moreover, this report focuses on obesity alone, excluding ‘overweight’ (defined
generally as BMI between 25 to 30) so the costs estimated are far less than the
costs of all excess body weight.
The 55-59 year age group contained the largest number of obese people for both men
(159,000) and women (203,000).
Over 280,000 young Australians (aged 5-19 years) are obese.
Prevalence rates are based on Australian measured anthropomorphic data from
AusDiab, the National Nutrition Study and the NSW Schools Physical Activity and
Nutrition Survey (SPANS) study for children.
PREVALENCE OF OBESITY, 2005 (AUSTRALIANS)

250,000
Males
Females
200,000
150,000
100,000
50,000
-
4
+
0-
5-
-1
-1
-2
-2
-3
-3
-4
-4
-5
-5
-6
-6
-7
-7
-8
-8
90
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
Despite serious weaknesses in data, obesity prevalence rates appear to be increasing for both
adults and children, although it is unclear at exactly what rate. A baseline prevalence
projection (with no further change in age-gender prevalence rates, such that all further
increases are due to demographic ageing alone) indicates that, by 2025, a total of 4.2 million
Australians (16.7% of the population) are forecast to be obese.
However, if rates continue to increase at historical rates, there could be as many as
7.2 million obese Australians by 2025 (28.9% of the population).
iii
People with Obesity (PWO) have increased overall risk of death, as well as higher Relative
Risk (RR) of:
Type II diabetes (RR up to 3.2);
Cardiovascular Disease (CVD), including Coronary Heart Disease (CHD) RR up to 1.8,
stroke (RR up to 1.8) and hypertension (RR up to 2.35), which in turn causes
Hypertensive Heart Disease (HHD) and Peripheral Arterial Disease (PAD);
osteoarthritis (RR up to 2.45);
various cancers – colorectal, breast, uterine and kidney (RR up to 1.75); and
other health conditions.
Where the RR for a disease is raised in obese people, a portion of the cases of that disease
are directly attributable to obesity; the Attributable Fractions (AFs) are derived from RRs and
used to estimate costs.
This report estimates that in 2005:

102,204 Australians had Type 2 diabetes as a result of being obese (10.8% of all people
with Type 2 diabetes);
over 379,00 Australians had CVD as a result of being obese (obesity causing 14% of
hypertension, 12% of CHD and 12% of stroke);
over 225,000 Australians had osteoarthritis as a result of being obese (14% of all people
with osteoarthritis); and
20,430 Australians had cancer as a result of being obese (obesity causing 13% of
colorectal and kidney cancers, and 16% of breast and uterine cancers).
These health impacts have a number of cost impacts on the Australian economy, namely:
direct financial costs to the Australian health system include the costs of running
hospitals and nursing homes, General Practitioner (GP) and specialist services, the cost
of pharmaceuticals, allied health services, research and other direct costs (such as
health administration);
other financial costs, which include:
productivity losses – short and long-term employment impacts and premature
mortality;
carer costs – the value of community care services provided primarily by informal
carers;
Deadweight Loss (DWL) from transfers – taxation revenue foregone, welfare
and other government payments;
other costs – aids, equipment and modifications, transport and accommodation
costs, respite and other government programs and the bring-forward component of
funerals; and
non-financial costs – the disability, loss of wellbeing and premature death that result
from obesity and its impacts, measured in Disability Adjusted Life Years (DALYs), known
as the Burden of Disease (BoD).
Different costs of diseases are borne by different individuals or sectors of society – the
individual, their friends and family, Federal and State governments, employers, and other
members of society. Costs were measured using a broad range of data sources.
iv
Costs of obesity
The total financial cost of obesity in 2005 was estimated as $3.767 billion.
Of this, productivity costs were estimated as $1.7 billion (45%), health system
costs were $873 million (23%) and carer costs were 804 million (21%).
DWL from transfers (taxation revenue foregone, welfare and other government
payments) were $358 million (10%) and other indirect costs were $40 million (1%).
The net cost of lost wellbeing (the dollar value of the burden of disease, netting out
financial costs borne by individuals) was valued at a further $17.2 billion, bringing the
total cost of obesity in 2005 to $21.0 billion.
FINANCIAL COSTS OF OBESITY BY TYPE OF COST, 2005 (% TOTAL)

1.1%
9.5%
23.2% Health system
Productivity
Carers
21.3%
DWL
Other indirect
44.9% Total = $3.767billion
Of the financial costs, 29.1% are borne by individuals, 16.4% by family and friends, 37.0% by
Federal Government ($1.4 billion per annum), 5.0% by State Governments, 0.1% by
employers and 12.4% by the rest of society. However, if the cost of lost wellbeing is included,
the individual’s share rises markedly to 87.3% of the total.
FINANCIAL COSTS OF OBESITY: BY WHO BEARS THEM, 2005 (% TOTAL)

12.4%
Individuals
0.1%
29.1% Family/Friends
5.0%
Federal Government
State Government
Employers
Society/Other
37.0%
16.4%
Total = $3.767billion
v
COST SUMMARY, OBESITY ($M) 2005
Individuals Family/ Federal State Employ- Society/ Total

Friends Gov’t Gov’t ers Other
Type 2 diabetes
BoD 1,269 0 0 0 0 0 1,269
Health System 23 0 54 25 0 15 116
Productivity 277 0 162 0 3 0 442
Carers 0 456 23 0 0 0 479
DWL 0 0 0 0 0 76 76
Other indirect 6 1 0 0 0 0 7
Transfers 0 -18 18 0 0 0 0
Total financial 305 439 257 25 3 90 1,119
Total inc. BoD 1,574 439 257 25 3 90 2,389
CVD
BoD 11,263 0 0 0 0 0 11,263
Health System 84 0 198 93 0 54 428
Productivity 334 0 138 0 0 0 472
Carers 0 217 90 0 0 0 306
DWL 0 0 0 0 0 184 184
Transfers -16 -39 55 0 0 0 0
Total financial 402 178 480 93 0 237 1,390
Total inc. BoD 11,665 178 480 93 0 237 12,653
Osteoarthritis
BoD 1,172 0 0 0 0 0 1,172
Health System 44 0 102 48 0 28 221
Productivity 164 0 397 0 0 0 561
Carers 15 0 0 0 0 0 15
DWL 0 0 0 0 0 47 47
Transfers 0 0 0 0 0 0 0
Total financial 233 0 499 48 0 75 855
Total inc. BoD 1,405 0 499 48 0 75 2,027
Cancer
BoD 3,542 0 0 0 0 0 3,542
Health System 21 0 50 23 0 13 107
Productivity 136 0 80 0 2 0 218
Carers 0 2 1 0 0 0 3
DWL 0 0 0 0 0 51 51
Transfers -21 -2 24 0 0 0 0
Total financial 154 2 157 23 2 66 403
Total inc. BoD 3,696 2 157 23 2 66 3,945
Total
BoD 17,246 0 0 0 0 0 17,246
Health System 172 0 403 189 0 109 873
Productivity 911 0 777 0 5 0 1,693
Carers 15 674 114 0 0 0 804
DWL 0 0 0 0 0 358 358
Transfers -37 -60 97 0 0 0 0
Total financial 1,095 618 1,393 189 5 468 3,767
Total inc. BoD 18,340 618 1,393 189 5 468 21,013
There has been much speculation regarding the causes of obesity in the population, and ways
to address it. There is no case for a ‘fat tax’, which would be inefficient and inequitable due to
inelastic demand, and little evidence that mandatory regulatory approaches would be either
vi
effective, worthwhile given their regulatory burden or superior to voluntary codes developed in
partnership with industry. Subsidised programs offer more hope of efficacy and cost
effectiveness, although a number of factors make it difficult to evaluate the best interventions.
The limited number of randomised controlled trials with long term follow up and the paucity of
cost effectiveness data constrain the evaluator’s ability to clearly identify and compare the
relative value and effectiveness of individual and combined weight loss programs over
substantial periods (five years or more). Combined lifestyle modification in severely obese
adults can achieve weight loss and a reduction in comorbidities in some patients but does not
always achieve sustained significant long-term weight loss, while many patients regain weight.
There is some evidence that workplace weight management programs are effective
substitutes for physician-directed programs.
Pharmacotherapy combined with lifestyle modifications have been shown effective to achieve
weight loss and improve comorbidities, but longer term cost effectiveness has not yet been
fully established, although it appears promising. Surgery for selected patients, with
appropriate follow-up care, has achieved weight loss and improvement of some comorbidities
and quality of life at five years, but risks are higher. Bariatric surgical technology has
improved, stimulating new consumer demand which is expected to increase. Surgery has
been shown to be cost effective, although riskier than pharmacotherapies.
Access Economics
13 October 2006
vii
1. INTRODUCTION
1.1 OVERVIEW OF THIS REPORT
Access Economics was commissioned by Diabetes Australia to estimate the economic cost of
obesity in 2005. Obesity is defined in terms of BMI over 30. The study adopts a prevalence
approach to cost measurement, as the data sources lend themselves to utilisation of such an
approach, and as this avoids the uncertainty surrounding estimates of future treatment costs
associated with an incidence approach.
PWO have a higher risk of Type II diabetes, CVD, various cancers, osteoarthritis and other
health conditions. The following chapters present evidence in relation to:
the prevalence and epidemiology of obesity by age and gender in Australia, including a
brief analysis of past trends and future projections (Chapter 2);
a description of the health impacts of obesity, by each of the main areas of impact –
diabetes, CVD, osteoarthritis and cancers (Chapter 3);
health system costs resulting from greater utilisation of health services including acute
hospitalisations, pharmaceuticals, diagnostics, medical and Residential Aged Care
(RAC) and allied health services (Chapter 4);
other financial impacts (Chapter 5) including:
loss of productivity in the paid workforce;
greater need for personal care and household services provided either by informal
family carers or by formal sector community care services;
greater need for aids, equipment or home modifications, largely to enhance
mobility;
loss of wellbeing (Chapter 6) as a result of the morbidity (disability from the downstream
diseases caused by obesity) and premature mortality (death) – known as the BoD; and
a summary of the various costs of obesity and a brief discussion of potential policy
responses to the obesity epidemic (Chapter 7).
Specific methodologies relevant to each section are presented in the chapters, although the
remainder of this chapter addresses various cross-cutting methodological issues.
1.2 CROSS-CUTTING METHODOLOGICAL ISSUES

1.2.1 INCIDENCE AND PREVALENCE APPROACHES
This report utilises the prevalence (annual costs) approach to estimating the costs of
obesity, as the data sources generally lend themselves to utilisation of such an approach, and
as this avoids the uncertainty surrounding estimates of future treatment costs associated with
the alternative incidence (lifetime costs) approach. The difference between incidence and
prevalence approaches is illustrated in the following diagram, which considers three different
cases:
a, who became obese in the past and has incurred the associated costs up to the year in
question, with associated lifetime costs of A + A*, shaded in green;
1
b, who became obese in the past and has incurred the associated costs in 2005 as well
as in the past and future, with associated lifetime costs of B + B* + B**, shaded in blue;
and
c, who became obese in 2005, with lifetime costs of C + C*, shaded in red.
All costs should be expressed as present values relative to 2005.
Using an incidence approach, only cases like ‘c’ would be included, with the total cost
estimate equivalent to the sum of all the costs in the base year (ΣC) plus the present value of
all the future costs (ΣC*).
Using a prevalence approach, costs in 2005 relating to a, b and c would all be included, with
total costs equal to Σ(A + B + C). Costs in all other years are excluded.
FIGURE 1-1: INCIDENCE AND PREVALENCE APPROACHES TO MEASUREMENT OF ANNUAL COSTS

Past Base Future
year
A* A
B* B B**
C C*
Annual prevalence costs in the base year = Σ(A + B + C);

Annual incidence costs in the base year = Σ(C + present value of C*)
Note that Figure 1-1 also defines the lifetime costs of obesity for each person, as follows:
Lifetime cost for person c (= Incidence cost) = C + present value of C*
Lifetime cost for person b = B + present values of B* and B**
Lifetime cost for person a = A + present value of A*
1.2.2 CLASSIFICATION OF COSTS

Conceptual issues relating to the classification of costs include the following.
Direct and indirect costs: Although literature often distinguishes between direct and
indirect costs, the usefulness of this distinction is dubious, as the specific costs included
in each category vary between different studies, making comparisons of results
somewhat difficult. This report thus distinguishes instead between the health system
expenditures, other financial expenditures and loss of wellbeing (BoD).
Real and transfer costs: ‘Real costs use up real resources, such as capital or labour,
or reduce the economy’s overall capacity to produce (or consume) goods and services.
Transfer payments involve payments from one economic agent to another that do not
use up real resources. For example, if a person loses their job, as well as the real
production lost there is also less income taxation, where the latter is a transfer from an
individual to the government. This important economic distinction is crucial in avoiding
double-counting. It has attracted some attention in the literature’ (Laing and Bobic,
2002:16).
2
Economic and non economic costs: Economic costs encompass loss of goods and
services that have a price in the market or that could be assigned an approximate price
by an informed observer. ‘Non-economic’ costs include the loss of wellbeing of the
individual as well as of their family members and carers. This classification is ill-defined,
since ‘non-economic’ costs are often ascribed values and the available methodologies
are becoming more sophisticated and widely accepted. This report acknowledges that
greater controversy and uncertainty still surround the valuation of ‘non-economic’ costs
and thus the dollar estimates for the loss of wellbeing are presented separately.
Prevention and case costs: It is important to distinguish between the costs following
from and associated with a condition and costs directed towards preventing that
condition. Prevention activities include public awareness and education about obesity.
There are three types of costs associated with obesity and its downstream impacts:
1 Direct financial costs to the Australian health system include the costs of running
hospitals and nursing homes (buildings, care, consumables), GP and specialist services
reimbursed through Medicare and private funds, the cost of pharmaceuticals
(pharmaceutical benefits scheme and private) and of over-the-counter medications,
allied health services, research and “other” direct costs (such as health administration).
2 Other financial costs, which include:
Productivity costs include productivity losses of PWO such as long-term
employment impacts, absenteeism and/or premature mortality.
Carer costs include the value of care services provided in the community primarily
by informal carers and not captured in health system costs.
Transfer costs comprise the DWL associated with government transfers such as
taxation revenue foregone, welfare and disability payments.
Other costs include government and non-government expenditure on aids,
equipment and modifications that are required to help cope with illness, transport
and accommodation costs associated with receiving treatment, programs such as
respite and community palliative care and the bring-forward component of funerals.
3 Non-financial costs are also very important—the disability, loss of wellbeing and
premature death that result from obesity and its impacts. Although more difficult to
measure, these can be analysed in terms of the years of healthy life lost, both
quantitatively and qualitatively, known as the BoD.
Different costs of diseases are borne by different individuals or sectors of society. Clearly the
PWO bears costs, but so do employers, government, friends and family, co-workers, charities,
community groups and other members of society.
It is important to understand how the costs are shared in order to make informed decisions
regarding interventions. While the PWO will usually be the most severely affected party, other
family members and society (more broadly) also face costs as a result of obesity. From the
employer’s perspective, depending on the impact of obesity, work loss or absenteeism will
lead to costs such as higher wages (i.e. accessing skilled replacement short-term labour) or
alternatively lost production, idle assets and other non-wage costs. Employers might also face
costs such as rehiring, retraining and workers’ compensation.
While it may be convenient to think of these costs as being purely borne by the employer, in
reality they may eventually be passed on to end consumers in the form of higher prices for
goods and services. Similarly, for the costs associated with the health system and community
services, although the Federal and State/Territory Governments meet a large component of
3
this cost, taxpayers (society) are the ultimate source of funds. However, for the purpose of
this analysis, a ‘who writes the cheque’ approach is adopted, falling short of delving into
second round or longer term dynamic impacts.
Society bears both the resource cost of providing services to PWO, and also the ‘deadweight’
losses (or reduced economic efficiency) associated with the need to raise additional taxation to
fund the provision of services and income support.
Typically the groups who bear costs and pay or receive transfer payments are:
PWO;
The Household
friends and family (including informal carers);
employers;
Federal Government;
State and Local Government; and
the rest of society (non-government, i.e. not-for-profit organisations, workers’
compensation groups etc).
Classifying costs by type and allocating them by who bears the costs enables a framework for
analysis (see Table 1-1).
TABLE 1-1: SCHEMA FOR COST CLASSIFICATION

Conceptual Subgroups Bearers of Cost Comments
group
1. Health Costs by type of service (and PWO*, governments and
System Costs prevalence in 2001) society
2. Other
Financial
Costs
Productivity Lost productivity from temporary PWO, employer and
#
Costs absenteeism government
Lost management productivity Employer and
#
government
#
Long-term lower employment PWO and government Includes premature
rates retirement
#
Premature death PWO and government Loss of productive
capacity
Additional search and hiring Employer Incurred when
replacement prematurely leave job
Carer Costs Lost carer productivity Friends and family, and Includes both paid and
employer# unpaid work
Transfer costs DWL Society Relate to transfers from
taxation, welfare etc
Other costs Various, as able to be Governments, PWO, Aids, modifications,
measured, but tend to be Friends and family and travel, accommodation,
relatively small society, respite/ palliative care,
funeral costs etc
3. Non-financial BoD (YLLs, YLDs, DALYs). PWO* The net value of BoD
(loss of should exclude other
wellbeing) costs borne by the
individual to avoid
double counting
* Friends/family may also bear loss of wellbeing, health costs and lower living standards as a result of obesity;
however, care is needed to assess the extent to which these are measurable, additional (to avoid double counting)
and not follow-on impacts. For example, a spouse may pay a medical bill and children may share in lower
household income when the PWO’s work hours are reduced – but as this is simply redistribution within family
4
income it is not measured here. Moreover, if a family carer develops depression or a musculoskeletal disorder, it
would be necessary to estimate the aetiological fraction attributable to obesity, allowing for other possible
contributing factors.
# Where earnings are lost, so is taxation revenue and frequently also there are other transfers, such as welfare
payments for disability/sickness/caring etc, so Governments share the burden.
1.2.3 CALCULATING PARAMETERS

There are essentially two ways of estimating each element of cost for each group:
Top-down: These data may provide the total costs of a program element (eg, health
system); or
Bottom-up: These data may provide estimates of the number of cases in the category
(‘n’) and the average cost for that category. The product is the total cost (eg, the wage
rate for lost earnings multiplied by the average number of days off, and the number of
people to whom this applies).
It is generally more desirable to use top-down national datasets in order to derive national cost
estimates, to ensure that the whole is not greater or less than the sum of the parts. On the
other hand, it is often difficult to obtain top-down estimates. In this report, the top-down
approach is applicable to health system and BoD costs and the bottom-up approach applies in
other cases.
Data on health system costs and BoD are derived from the Australian Institute of Health
and Welfare (AIHW), which are in turn based on other data sources, such as the
Australian Hospital Statistics and Bettering The Evaluation And Care Of Health data for
GP costs.
Data on other financial costs are drawn from a variety of sources – for example, the
literature (focussing on Australian literature but sometimes supplemented by
international material), data from the Australian Bureau of Statistics (ABS) Survey of
Disability, Ageing and Carers (SDAC) and Average Weekly Earnings (AWE), and so on.
The main limitations of the data are in relation to timeliness, comparability and objectivity. For
example:
Health cost data were most recently calculated by the AIHW for 2000-01 (AIHW, 2005b)
but only include 86% of recurrent costs, and so are factored up in this report and
extrapolated to 2005.
The National Health Survey (NHS) and SDAC use self-reported data on adults, where
there is no medical verification of obesity or its impacts, but the medically verified
alternative (AusDiab) was not as recent in its age-gender prevalence measures.
There were differences in data collections in relation to different diseases caused by
obesity. For example, ‘other financial’ costs of cancer cost were measured in great
detail but it was not possible to achieve the same detail for diabetes.
AFs are derived from odds ratios and RR parameters in source studies that may reflect
correlation rather than causation, although best attempts are made to correct for other
confounding factors such as comorbidities or socioeconomic issues (eg, low income may
be both a cause and consequence of obesity and its health impacts).
1.2.4 DISCOUNT RATES

A discount rate is used to convert future income or a cost stream into the equivalent value in
today’s dollars.
5
Choosing an appropriate discount rate for present valuations in cost analysis is a subject of
some debate, and can vary depending on what type of future income or cost stream is being
considered. There is a substantial body of literature, which often provides conflicting advice,
on the appropriate mechanism by which costs should be discounted over time, properly taking
into account risks, inflation, positive time preference and expected productivity gains.
The absolute minimum option that one can adopt in discounting future income and costs is to
set future values in current day dollar terms on the basis of a risk free assessment about the
future (that is, assume the future flows are similar to the certain flows attaching to a long-term
Government bond).
Wages should be assumed to grow in dollar terms according to best estimates for inflation and
productivity growth. In selecting discount rates for this project, we have thus settled upon the
following as the preferred approach.
Positive time preference: We use the long-term nominal bond rate of 5.8% pa (from
recent history) as the parameter for this aspect of the discount rate (If there were no
positive time preference, people would be indifferent between having something now or
a long way off in the future, so this applies to all flows of goods and services).
Inflation: The Reserve Bank has a clear mandate to pursue a monetary policy that
delivers 2% to 3% inflation over the course of the economic cycle. This is a realistic
longer run goal and we therefore use a value of 2.5% pa for this variable (It is important
to allow for inflation in order to derive a real (rather than nominal) rate).
Productivity growth: The Commonwealth Government's Intergenerational report
assumed productivity growth of 1.7% in the decade to 2010 and 1.75% thereafter. We
suggest 1.75% for the purposes of this analysis as many of the productivity costs extend
past 2010.
There are then three different real discount rates that should be applied:
To discount income streams of future earnings, the discount rate is:
5.8 - 2.5 - 1.75 = 1.55%.

To discount health costs, the discount rate is:
5.8 - (3.2 - 1.75) - 1.75 = 2.6%.

To discount other future streams (healthy life) the discount rate is:
5.8 – 2.5 = 3.3%
While there may be sensible debate about whether health services (or other costs with a high
labour component in their costs) should also deduct productivity growth from their discount
rate, we argue that these costs grow in real terms over time significantly as a result of other
factors such as new technologies and improved quality, and we could reasonably expect this
to continue in the future.
6
2. OBESITY
2.1 DEFINITIONS AND MEASURES
Obesity is defined as the accumulation of excessive fat in the body. Body fat is not measured
directly. Instead, commonly used measures of obesity are based on easily obtained
anthropomorphic measures such as weight, height, waist and hip circumference.
2.1.1 BODY MASS INDEX

BMI is the most commonly used measure of obesity and is calculated as the ratio of weight in
kilograms to the square of height in metres.
BMI = weight/(height)2
For adults, weight classifications based on BMI are as follows, used in the ABS NHS.
Underweight <18.5
Normal range 18.5 to < 20.0 and 20.0 to < 25.01
Overweight 25.0 to < 30.0
Obese ≥ 30.0
Source: ABS (2005a)
These weight classifications are not necessarily suitable for all ethnic groups. For example
Asian people would generally be considered obese at a lower BMI, while Polynesians would
not be considered obese until they reached a higher BMI cut-off (AIHW, 2006). Nonetheless a
BMI of 30kg/m2 is accepted as the standard international definition of adult obesity.
For children and adolescents (2-18 years), a set of age and gender specific BMI-thresholds
are used to define childhood obesity. These thresholds were developed by Cole et al (2000)
and are based on data for Brazil, Great Britain, Hong Kong, the Netherlands, Singapore and
the United States, aligning with the adult obesity threshold of 30 kg/m2 by age 18. The
thresholds have been accepted as the international standard for measuring obesity in children
and adolescents. Prior to these measures, a series of different approaches were taken to
measuring childhood obesity using BMI. For example, in the US a BMI in the 95th percentile
was used to define obesity. Other countries defined obesity as a measure (such as BMI,
waist-to-height ratio) which was greater than two standard deviations from the median
(Ebbeling et al, 2002). These measures were somewhat arbitrary, and because they were
defined in terms of an individual’s measure relative to others in the population, meaningful
comparison of prevalence rates over time, or across countries was not possible.
The age and gender specific thresholds for overweight and obesity in children and adolescents
are shown in Table 2-1. These thresholds, while less arbitrary than previous measures, still
provided a statistical definition of obesity. That is, they are not defined in terms of an increase
1
The split group enables categories to be reported against both the World Health Organization and National Health
and Medical Research Council guidelines.
7
in health risk (something which has been difficult to establish in children and adolescents), and
as such the implications for any one child determined to be obese are not clear.
TABLE 2-1: AGE AND GENDER-SPECIFIC THRESHOLDS

FOR OVERWEIGHT AND OBESITY IN CHILDREN AND ADOLESCENTS.
Source: Cole et al (2000)
Self-reported vs Measured BMI
In Australia, a number of major surveys of obesity, such as the ABS NHS, have used self-
reported data on height and weight, rather than conducting more costly and intrusive physical
examinations to collect this information. However, studies of obesity prevalence that use
physical examination to determine BMI are more accurate than those based on self-reported
data, as people misreport their height and weight, intentionally and unintentionally.
The ABS 1998 report How Australians Measure Up, noted differences between the 1995 NHS
and the 1995 National Nutrition Survey (NNS). The NNS took physical anthropomorphic
measures for a subset of NHS participants and showed that people tend to overstate their
height and understate their weight. Further, the tendency to overstate height is in inverse
relation to measured height, and the tendency to understate weight is in inverse relation to
measured weight.
8
The net result is that people under-report their BMI. In the ABS study 23% of males and 25%
of females were categorised in a higher weight category when physical measures were used
rather than self-reported ones.
2.1.2 WEIGHT DISTRIBUTION MEASURES

While BMI is by far the most commonly reported measure of obesity, a number of measures
which reflect the distribution of fat stored in the body are also popular. These measures
include the waist-to-hip ratio, and waist circumference. Definitions of overweight and obesity
based on waist circumference are provided in Table 2-2.
TABLE 2-2: DEFINITIONS OF OVERWEIGHT AND OBESITY

BY WAIST CIRCUMFERENCE
Males Females
Overweight 94.0-101.9 cm 80.0-87.9 cm
Obese ≥102.0 cm ≥88.0 cm
Measures of weight distribution are technically measures of a subset of obesity known as

abdominal obesity or central adiposity.
2.2 RISKS
For adults, a BMI of 30 kg/m2 or more has been shown to be associated with a raised risk of
pre-mature illness and mortality. However, there are problems with such thresholds given that
risk appears to be related continuously with adiposity. Studies have generally reported a
linear or curvilinear relationship between BMI and mortality risk in adults starting as low as
22kg/m2. However, the extent to which small degrees of overweight increase the risk of
morbidity is still debated in the literature.
The National Health and Research Council (NHMRC) report that epidemiological studies have
found that fat-distribution measures, such as waist circumference or waist-to-hip ratio, are
better predictors of mortality risk than the BMI measure of obesity.
2.2.1 EVIDENCE FOR ASSOCIATED HEALTH RISKS

Accumulation of excess fat in the body is associated with well established health risks. PWO
have been shown to have higher RR of developing diseases including Type 2 diabetes, CVD,
osteoarthritis, a number of cancers, as well as many other diseases, than non-obese people,
generating public health concern.
This report measures the cost of obesity in terms of the costs of four major types of disease
that have been shown to be linked to obesity, attributing only the proportion of costs that are
aetiologically attributable to obesity, rather than other risk factors or causes:
Type 2 Diabetes;
CVD, including
CHD;
stroke, and
hypertension, which causes
i HHD and
9
ii PAD
Osteoarthritis; and
Cancer
Breast
Colorectal
Uterine and
Kidney.
Evidence on the precise quantitative link between obesity and the specific diseases costed in
this report is derived from a summary of RRs from the literature, reported in Mathers et al
(1999), determined from analysis of numerous studies of the relationship between excess
weight and specific diseases. The RRs are presented in Table 3-1 in the preamble to Chapter
3. The RR of a person with a BMI less than 25 is unity, by definition.
Where the RR for a disease (such as CVD) is raised in obese people, a portion of the cases of
that disease are directly attributable to obesity. This is the portion of cases that would not
occur if obesity were eliminated, and is known as the Attributable Fraction (AF). The cost
estimates in this report (Chapters 3 to 6) are then based on AFs derived from the RRs.
Increases in obesity prevalence, such as those that have occurred in Australia in recent
decades (see Section 2.5), lead to increases in the prevalence of related diseases. This has
negative impacts for the individuals involved and high costs for society as a whole. According
to the AIHW (2006:183), overweight (including obesity) was the leading determinant of BoD in
Australia in 2003.
Other Sources
Since the publication of Mathers et al (1999), further evidence has been reported linking
obesity to the diseases included in this report as well as numerous other diseases and
conditions.
The AusDiab 2005 report found that in the five years between their original survey (1999-
2000) and the follow-up, people with a BMI of 30 or more had an incidence of diabetes which
was four times that of those in the normal BMI range at baseline (annual incidence for obese
males was 1.8% compared with 0.4% for those in normal BMI range, annual incidence of
diabetes in obese females was 1.4% compared to 0.4%).
AusDiab 2005 also reported the incidence of hypertension by baseline BMI status. The annual
incidence of hypertension for people who were obese at baseline (1999-2000) was three times
that of those in the normal BMI range. For women the risk was even higher, with 5.6% of
obese women developing hypertension each year compared to 1.5% of women who had BMI
in the normal range. For men the difference in incidence was 5.2% compared to 2.4%.
The NHMRC has also reviewed the evidence for the relationship between obesity and disease
in their report “Clinical Practice Guidelines for the Management of Overweight and Obesity in
Adults”. The NHMRC summarises diseases by the degree to which RR is raised in PWO
(slightly increased; moderately increased; greatly increased) compared to the non-obese. The
NHMRC findings are shown in Table 2-3.
10
TABLE 2-3: NHMRC - DISEASES ASSOCIATED WITH OBESITY BY CATGORY OF RR
RR Associated with metabolic Associated with excess weight

consequences
Greatly increased Diabetes (Type 2) Sleep apnoea
(RR>3) Hypertension Breathlessness
Gall bladder disease Asthma
Dyslipidaemia Social isolation and depression
Insulin resistance Daytime sleepiness and fatigue
Non-alcoholic fatty liver disease
Moderately increased CHD Osteoarthritis
(RR 2-3) Stroke Respiratory disease
Gout/hyperuricaemia Hernia
Psychological problems
Slightly increased Cancer (breast, endometrial, Varicose veins
(RR 1-2) colon and others) Musculoskeletal problems
Reproductive abnormalities/ Bad back
impaired fertility
Stress incontinence
Polycystic ovaries
Oedema/cellulitis
Skin complications
Cataract
Source: NHMRC (2003)
As Table 2-3 shows, a number of diseases that have not been included for costing in this
report are also associated with greatly increased risk in obese people relative to those in the
normal weight range. As with those diseases that have been costed (Chapters 3 to 6),
diseases such as gall bladder disease, sleep apnoea, and depression are likely to be
accompanied by significant personal and economic costs (for example, Hillman et al, 2006).
For this reason the overall estimate of the cost of obesity presented here is likely to be an
underestimate of the true cost.
2.2.2 RISKS FOR CHILDREN

As noted earlier, the link between BMI and ill-health is less clear in children and adolescents
than it is in adults. This is because many of the consequences of obesity take time to develop.
Exceptions to this are Type 2 diabetes (which is becoming increasingly common in children),
insulin resistance syndrome, and some orthopaedic complications. Insulin resistance
syndrome is defined by Ebbeling et al (2002) as the clustering of CVD risk factors including
hypertension, dyslipidaemia, chronic inflammation, increased blood clotting tendency,
endothelial dysfunction, and hyperinsulinaemia.
Psychosocial problems are also identified in the literature as an immediate consequence of

obesity, with substantial costs for the children involved. For example Waters and Baur (2003)
report a US study which found that obesity carries a social stigma in children greater than any
physical disability. This is echoed by Ebbeling et al (2002), who state that “obese children are
stereotyped as unhealthy, academically unsuccessful, socially inept, unhygienic, and lazy”
even to an extent by health care providers. Moreover, childhood obesity increases the RR of
obesity in later life, adding to alarm about increasing prevalence of childhood obesity.
11
2.2.3 THE CAUSAL LINKS

The NHMRC distinguish direct and indirect causal links between obesity and disease. Obesity
leads to morbidity directly as a result of excess body weight, for example by placing additional
pressure on certain parts of the body. It is in this way that obesity can lead to back problems
and pulmonary complications (from increased diaphragmatic pressure) such as sleep apnoea
and asthma.
The indirect route in which obesity causes disease is as a result of metabolic consequences
(usually insulin resistance). AusDiab 2001 defines metabolic syndrome as “the clustering of
the ‘deadly quartet’ of risk factors: Type 2 diabetes, abdominal obesity, dyslipidaemia, and
hypertension.”
Insulin resistance is also a component of ‘metabolic syndrome’, or ‘syndrome X’,

which includes hypertension, dyslipidaemia and hyperglycaemia, the combination
of which has been shown to be highly predictive of end-point diseases such as
coronary artery disease, stroke and diabetes. NHMRC (2003:5).
Dyslipidaemia is a disorder of lipoprotein metabolism, including lipoprotein overproduction or

deficiency. Dyslipidemias may be manifested by elevation of the total cholesterol, the "bad"
Low-Density Lipoprotein (LDL) cholesterol and the triglyceride concentrations, and a decrease
in the "good" High-Density Lipoprotein (HDL) cholesterol concentration in the blood. The
primary dyslipidemia related to obesity is characterized by increased triglycerides, decreased
HDL levels, and abnormal LDL composition. Much work has been done to elucidate the
pathogenesis of the dyslipidemia of obesity, which seems to be closely related to insulin
resistance in obese individuals; however, more studies in humans are needed to further
understand the metabolic mechanisms underlying the changes, and to distinguish between the
roles of insulin resistance and body fat in the lipoprotein changes. The dyslipidemia
associated with obesity no doubt plays a major role in the development of atherosclerosis and
CVD in obese individuals. All of the components of the dyslipidemia, including higher
triglycerides, decreased HDL levels, and increased small, dense LDL particles, have been
shown to be atherogenic. Weight loss and exercise, even if they do not result in normalisation
of body weight, can improve this dyslipidemia and thus reduce CVD risk. In addition, obese
individuals should be targeted for intense lipid-lowering therapy, when necessary (Howard et
al, 2003).
Mortality: One measure that emphasises the adverse impact of obesity on health is the RR of
mortality. This is because risk of mortality is increased by a number of the diseases
associated with obesity (such as CVD and cancer). The NHMRC report the results of the
Framingham study in the United States which found that obesity reduced life expectancy in 40-
year-old males and female non-smokers by 5.8 and 7.1 years respectively.
2.3 AETIOLOGY
This section outlines very briefly current knowledge regarding what causes obesity in
individuals and an overview of the debate on what is driving changes in obesity prevalence in
the population (these changes are outlined in section 2.5 on prevalence trends). However, it
should be noted that the main purpose of this report is to estimate the economic cost of
obesity in Australia, rather than to draw conclusions about the relative importance of causes,
so the overview is quite perfunctory.
12
2.3.1 OBESITY IN THE INDIVIDUAL

Accumulation of fat in an individual can be due to a number of factors. This section briefly
discusses a few – genetic, perinatal and energy imbalance (including metabolism).
Genetic factors
Genetic and medical factors can explain the build up of excess fat in the body in some
individuals. According to Ebbeling et al (2002), five genetic mutations have been identified
which cause human obesity. Additionally a number of candidate alleles have been linked to
risk of early-onset obesity. Overall, however, only a small fraction of obesity prevalence can
be attributed to single gene defects.
Perinatal factors
Ebbeling et al (2002) outline research into whether factors during pregnancy (such as obesity
in the mother) can increase a person’s likelihood of developing obesity. The hypothesis is that
an increase in nutrients across the placenta can induce permanent changes in appetite, or
energy metabolism. This hypothesis is supported by some experiments in animals but so far
evidence for humans is lacking. Bottle feeding has been linked to increase risk of obesity in
later childhood.
Energy imbalance
An energy imbalance (excess) of some kind is at least the partial cause of obesity in the
majority of obese individuals. Energy imbalance refers to when an individual’s energy intake
exceeds the energy that he or she expends2. At the simplest level, there are thus two
components that can lead to an energy imbalance: what one eats; and what energy-burning
activity one undertakes. When an individual is in energy imbalance over a sustained period,
they will experience weight changes. Even small imbalances in the energy intake-expenditure
equation can lead to large changes in weight over time.
Small changes, weighty consequences
According to Ebbeling et al (2002) an energy surplus equivalent to one serve of

soft drink per day (500kJ) could lead to a 50 kg weight gain over a 10-year period,
ceteris paribus.
While the concept of energy imbalance is a simple one, there is still a degree of uncertainty
and complexity about how many individual elements affect the accumulation and metabolism
of fat in the body. Some people seem to be ‘more efficient’ at storing fat, while others may
metabolise the same kilojoules (seemingly) faster at any given activity level. It is this
uncertainty and complexity that has led to debate over what have been the most important
drivers of increased obesity prevalence in Australia and worldwide. Apart from these
metabolic factors, the most well-established factors that affect energy balance are linked to
physical activity and diet.
Numerous studies have tried to establish whether sedentary activities such as television
watching or computer games are related to obesity. These studies generally show a strong
2
Conversely, weight loss occurs when energy expended exceeds that consumed (an energy deficit) over a period.
13
positive correlation between participation in sedentary activities and obesity – for example, a
study cited in Ebbeling (2002) of children in Mexico City found that with each additional hour of
television viewing per day, risk of obesity increased by 10%). However, it is difficult to
establish a direct causal link because there may also be at least some reverse causality (being
obese is a discouragement from participation in physical activity). Television (and computer or
other electronic games) are also thought to affect both sides of the energy balance equation
because of the increased tendency to consume energy dense foods while watching or playing.
Moreover, increased exposure to advertising of energy dense foods while watching TV is
speculated to increase consumption of these food groups at other times.
More study is required in all these areas, as there are still many grey areas.
Different types of diet (eg. high fat or high carbohydrate diets) may affect the
accumulation of fat. The nature and balance of proteins and other vitamins and minerals
may also be relevant, as may be the greater prevalence of (therapeutic) drugs for which
weight gain is a side-effect (eg, anti-depressants).
The type of fat (or carbohydrate) consumed in the diet may be more important than the
portion of diet that comes from fat (the relative balance of HDL and LDL cholesterol, of
saturated and unsaturated fats and of trans-fatty acids). A number of US studies have
shown that, concurrent with the rising prevalence of obesity, there has been a decrease
in the share of calories in the diet consumed as fat.
The intensity of physical activity and the overall duration of completely sedentary time
may also play a role (the dose-response relationship).
Socioeconomic, locational and environmental ethnicity (as opposed to genetic)
factors also require more study.
Some studies have found a relationship between Socioeconomic Status (SES) and
obesity (inversely correlated), although this relationship is not yet well explained
and may also suffer some two-way causality (obese people have health impacts
that reduce their capacity to earn). For example, a relationship between obesity
and SES was found in the (self-reported) 2001 NHS data (AIHW, 2006:186).
However, in an analysis including data from three studies of childhood obesity
conducted between 1995 and 1997, Booth et al (2001) found no statistically
significant relationship between obesity and SES. Energy-dense foods may be
less expensive dollar for calorie (but not dollar per nutrient) than fresh fruit and
vegetables. However, complex educational and environmental aspects
(relationship and parenting patterns, self-efficacy etc) are likely to play a role as
are the time aspects of food preparation. Moreover, people do not always behave
completely rationally in relation to diet or activity, as borne out in eating disorders,
the addictive properties of some foods, eating ‘free’ food even when satiated, or
even simple price substitutions (tap water being a ‘cheap’ drink but not preferred
by some favouring sweetness or satiety).
Booth et al (2001) found that obesity prevalence was significantly higher in urban
than rural boys, but there was no significant relationship between obesity and
rurality in girls.
Finally, Booth et al (2001) found a higher prevalence of obesity among students
from European and Middle-Eastern cultural backgrounds compared to children
from other backgrounds.
2.3.2 OBESITY IN SOCIETY

Rising obesity prevalence is unlikely to be explained substantially by genetic factors, given the
genetic stability of most populations, although there are some theories relating to selective
14
partnering. Similarly, better nutrition in pregnancy and the possibility of bottle feeding are
unlikely to explain much of the growth. Most theories attempt to explain why larger numbers of
people experience sustained periods of energy imbalance than in the past. In Australia, it is
generally though that energy intake has risen while levels of physical activity have fallen,
although even this is controversial depending on the length of the time series (energy intake
may have fallen compared to a century ago), distribution of the energy intake (median or
percentiles) or other source study aspects. Some arguments in the ‘lifestyle’ debate are
summarised in dot-point form below, for ‘now’ compared to (an unspecified time in) the past.
Energy intake = diet

more food eaten in and from restaurants
more take-away and highly processed food due to the convenience and ‘palatability’ of
energy-dense foods and advertising
more soft drinks, being additional rather than substituting for food (in part due to the
proximity of drink vending machines, failure to satiate and other reasons)
less time preparing meals (time scarcity and a preference for the ‘instant’)
optimal food eating temperatures (air-conditioning and central heating that counter the
body’s tendency towards lower appetite in extreme heat and greater energy expenditure
in extreme cold)
greater wealth, so there is greater consumption of almost all goods and services
different social mores, that work against self-discipline and cultivate a consumption-
oriented society
Energy expenditure = physical activity

more sedentary work environments: post-industrial economic development leads to a
contraction in primary production, mining and manufacturing and a rapid expansion in
the less physically active services sectors (as borne out by the proportion of ‘white collar’
workers in the labour force)
more sedentary leisure activities: computers, television, video games, less free play and
more ‘structured’ activities for children
less incidental exercise (ie using cars instead of walking or cycling, elevators rather than
stairs), particularly true of transport to and from school for children (however, the
advantage may be a reduction in child morbidity and mortality from motor vehicle
accidents and from ‘stranger danger’)
greater urbanisation (with urban planning that may not have prioritised physical activity)
and higher crime rates, reducing opportunistic exercise
Much debate remains over which factors have been most important in driving population
changes in obesity prevalence in part because many of the factors are difficult to measure at
the ‘whole of society’ level or it is difficult to determine how the factor has changed over time
because no historical data exist. For example, TV viewing may have displaced other
sedentary activities such as listening to the radio, reading or playing cards, and therefore not
substantially increased overall time spent in sedentary activities. The Western Australia (WA)
Child and Adolescent Physical Activity and Nutrition Study (CAPANS) found that of the
variables considered, aside from age (which for children was expected to be a strong predictor
of BMI), lower levels of physical activity and regularly skipping breakfast were the main
predictors of BMI. Mother’s education limited to year 10 or below and eating fast food more
15
than once a week were also important. There was no significant difference in BMI across
schools.
As well as the debate over which are the leading causes of population obesity, there is also
debate about its extent. This is addressed in the next sections.
2.4 PREVALENCE IN AUSTRALIA

In this report ‘prevalence’ of obesity refers to the number of PWO in a population at a given
point, or over a certain period of time, usually one year, while the ‘prevalence‘ rate refers to
those people expressed as a proportion of their respective source population.3
The best method of measuring community prevalence is through well-designed clinical studies
of populations, preferably longitudinal and prospective. However, studies of obesity
prevalence in Australia suffer from inconsistencies in methodology (self-report vs
measurement), the definition of obesity which is used (BMI vs abdominal), gaps in age groups
covered (under fives, young adults 17 to 24 years) and irregular repetition (most recent
measured adult prevalence figures are for 1999-00, most recent national survey of childhood
and adolescent obesity was 1997). The lack of hard data is somewhat surprising given the
topical importance of the issue.
Baseline prevalence estimates in this report reflect the most recently available data on
measured obesity as defined by BMI (>30 for adults and based on the Cole et al (2000)
thresholds for children and adolescents). The use of BMI measures is necessitated by the
availability of data. It is noted that where measures of waist circumference have been
collected alongside measures of BMI, the prevalence of abdominal obesity has regularly far
exceeded the prevalence of obesity as determined by BMI. For example in the 1999-2000
AusDiab study 30.5% of all people were defined as obese based on waist circumference
compared with 20.8% of people based on BMI.
2.4.1 PREVALENCE IN ADULTS

In adults the most recent study of measured obesity is the AusDiab study.4 This study was
conducted between May 1999 and December 2000, included 11,247 respondents and
comprised a household interview followed by a biomedical examination. Based on a BMI
measure of over 30, adult obesity prevalence was found to be 19.3% for men and 22.2% for
women (Cameron et al, 2003). Age and gender specific prevalence rates are provided below
(Table 2-4).
3
This differs from the original clinical definition, but has come to be accepted in common parlance.
4
It is noted that the AusDiab 1999-2000 study is not entirely nationally representative because rural areas and
areas with a large proportion of the population who were Aboriginals or Torres Strait Islanders, were excluded
during the sampling process. The AusDiab 2005 follow up survey only reported incident cases of obesity in the
original study group.
16
TABLE 2-4: AUSDIAB 1999-2000

PREVALENCE RATES (PER CENT)
Age Group Males Females

25-34 17.4% 12.4%
35-44 17.8% 19.5%
45-54 20.8% 26.9%
55-64 25.5% 32.8%
65-74 19.9% 29.4%
75+ 12.7% 15.6%
Total 19.3% 22.2%
Source: Cameron et al (2003)
Because the AusDiab study only included individuals aged 25 years and older, it was
necessary to estimate obesity prevalence for the 20-24 year age group using the prevalence
rate found in the NNS.
2.4.2 PREVALENCE IN CHILDREN AND ADOLESCENTS

The most recent national estimates of obesity prevalence in children and adolescents are from
the 1995 NNS conducted by the ABS. The survey population was a subset of the NHS
population, although it includes information on individuals aged 2 years or older. Because the
NNS preceded the AusDiab study by five years, the prevalence estimate for this age group is
likely to be less accurate than for other adults.
The prevalence of obesity for 20-24 year olds in the NNS was 9.9% of males and 8.6% of
females. In the age group 2-18 years, 4.5% of boys and 5.3% of girls were found to be obese
(Magarey et al, 2001).
A number of regional studies have been conducted more recently than the NNS. These
include the SPANS in New South Wales (NSW), and the CAPANS in WA.
CAPANS included children and adolescents in the school years 3, 5, 7, 8, 10 and 11,
conducted between August and December 2003. The prevalence of obesity in the CAPANS
study was 4.9% for both boys and girls.
SPANS included children and adolescents in school years Kindergarten, 2, 4, 6, 8 and 10, and
was conducted between February and May 2004. The prevalence of obesity of boys in the
SPANS study was 7.7%, and the prevalence of obesity of girls in the study was 6.1%.
Table 2-5 summarises the results of these three Australian childhood prevalence studies.
17
TABLE 2-5: PREVALENCE OF OBESITY IN CHILDREN AND ADOLESCENTS, STUDIES
Study Age Number of respondents Prevalence rate of

(region; year) Range obesity (%)
Boys Girls Boys Girls
NNS (Australia;1995) 2-18 1,515 1,447 4.5 5.3
CAPANS (WA; 2003) 7-16 959 876 4.9 4.9
SPANS (NSW; 2004) 5-165 3,128 2,893 7.7 6.1
In this report the NSW SPANS childhood prevalence rates are used. This survey was seen as
the best available estimate of current prevalence, given the date it was conducted and the size
of the survey population. It would be preferable to use nationally representative data however
given NNS was conducted over a decade ago it was considered to be unacceptably dated.6 A
review of self-reported data from the 2004-05 NHS, revealed that (for adults at least) obesity
prevalence in NSW was the same as that for Australia (AIHW, 2006).
One prevalence rate was used for all children of each gender for ages 5 to 19 years. This is
because of the high level in fluctuation in obesity prevalence in individual school years in
SPANS in 2004, reflected also in similar surveys of NSW school children from 1997 and 1995.
The prevalence rate for obesity found in SPANS was applied to children aged 17-19 despite
the likelihood that few children of this age were included in the study, because five year age
groups were used and because they appear to be reasonably consistent with prevalence rates
for the 20-24 year age group.
This report conservatively assumes that prevalence of obesity is zero in the 0-4 year age
group, although there is mounting evidence that this is not the case. For example, Vaska and
Volkmer (2002) reported prevalence of obesity in four year olds in South Australia for 2002 as
4.1% for boys and 5.8% for girls. The assumption of zero prevalence of obesity in children
aged less than 5 years is unlikely to significantly affect the costs of obesity estimated later in
this report due to the very low prevalence of obesity related diseases in this age group.
2.4.3 BASELINE PREVALENCE ESTIMATES

Table 2-6 provides the baseline prevalence estimates for obesity (by age and gender) which
are used in this report. Overall, in 2005 obesity prevalence was higher for women (16.8%)
than for men (15.1%); prevalence was higher for women in each age group over 35 years but
in younger age groups prevalence was higher in males (aged 5-34 years). For both males and
females, prevalence rates for obesity peak in the 55-64 year age group.
5
SPANS study does not report any results by age but rather by school year. The age range in this table
corresponds to normal age range for students in those school years.
6
Given the consequences of potentially increasing prevalence of childhood obesity, on both the individuals and
society, a national survey of childhood obesity is long overdue.
18
TABLE 2-6: BASELINE PREVALENCE RATES

(PER CENT) BY AGE AND GENDER
Age Group Males Females

0-4 0.0% 0.0%
5-19 7.7% 6.1%
20-24 9.9% 8.6%
25-34 17.4% 12.4%
35-44 17.8% 19.5%
45-54 20.8% 26.9%
55-64 25.5% 32.8%
65-74 19.9% 29.4%
75+ 12.7% 15.6%
Total* 15.1% 16.8%
* Note: Total is for 2005
Figure 2-1 shows the estimated prevalence of obesity for 2005, calculated using the
prevalence rates from Table 2-6 and figures for the Australian population for 2005 (from the
Access Economics Demographic Model based on ABS demographic data).
In 2005, 3.24 million Australians (1.52 million males and 1.72 million females) were
estimated to be obese.
The 55-59 age group contained the largest number of obese people for both men
(159,000) and women (203,000).
FIGURE 2-1: PREVALENCE OF OBESITY, 2005
250,000
Males
Females
200,000
150,000
100,000
50,000
-
4
+
4
9
-1
-1
-2
-2
-3
-3
-4
-4
-5
-5
-6
-6
-7
-7
-8
-8
0-
5-
90
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
19
2.5 PREVALENCE TRENDS AND PROJECTIONS

Concern over growing obesity rates has been largely fuelled by the apparent increase in
obesity between the mid-1980s and mid-1990s. The pattern of changes in obesity prevalence
since the mid-1990s is less clear. Due to the number of factors which can influence obesity
prevalence in the population (outlined in Section 2.3), a degree of volatility over time should
perhaps be expected.
2.5.1 TRENDS IN ADULT PREVALENCE

2.5.1.1 MEASURED OBESITY PREVALENCE
Prevalence trends in measured adult obesity were reported in the report of the AusDiab 2000
study findings, Diabesity and Associated Disorders in Australia – 2000: the Accelerating
Epidemic. Studies included in the comparison were the National Heart Foundation Risk Factor
Prevalence Survey (conducted in 1980 and 1989), the NNS (from 1995) and the AusDiab
study (from 1999-00). In order to compare results from the four surveys, prevalence of obesity
among capital city residents only was considered.7 Prevalence rates from the four studies
were age-standardised to the 1991 Australian population.
FIGURE 2-2: TRENDS IN OBESITY PREVALENCE (%) FOR ADULTS, 1980 TO 2000
20%
Males
18% Females
16%
14%
12%
10%
8%
6%
80
89
95
00
19
19
19
20
Source: International Diabetes Institute (2001)
Figure 2-2 shows obesity prevalence over the period 1980 to 2000, measured at 1980, 1989,
1995 and 2000. Over the period 1980 to 2000 the AusDiab data suggested that obesity
7
Urban or capital city residents were included for the NNS.
20
prevalence more than doubled its share of the population (from 7% to 19% in females and
from 7% to 17% in males), with the average yearly increase in the prevalence rate being 0.5%
for males and 0.6% for females. For both males and females the period of fastest growth in
measured obesity prevalence was between 1989 and 1995. Between 1995 and 2000 (the
most recent estimate available for the rate of change in obesity prevalence) the prevalence
rate of obesity among women grew by 0.56% each year, while the prevalence rate of obesity
among men actually dropped by 0.1%. Unfortunately the data preclude an analysis of any
age-specific or weight specific trends in obesity prevalence (for example, international trends
have shown the heaviest people getting heavier, and the distribution is relevant to RR for
health impacts).
2.5.1.2 SELF-REPORTED OBESITY PREVALENCE IN ADULTS
The NHS has been conducted by the ABS in the years 2004-05, 2001, and 1995. The survey
collected self-reported anthropomorphic data including height and weight for adults aged
18 years and above, and reported BMI category (underweight, normal range, overweight,
obese) by ten year age groups. In 2004-05 approximately 25,900 people (including children)
from all States and Territories were included in the survey (ABS, 2005a:50-52).
FIGURE 2-3: TRENDS IN SELF-REPORTED OBESITY

PREVALENCE (%), 1995 TO 2004-05
19%
Males
18%
Females
17%
16%
15%
14%
13%
12%
11%
10%
95
01
5
/0
19
20
04
20
Source: NHS data
The trend in the prevalence rate for self-reported obesity over the period 1995 to 2001 was for
an annual increase in prevalence rates of 0.6% for males and 0.7% for females. For men,
these figures contradict those found in the data for measured obesity over a similar period8
8
1995-2000 vs 1995-2001.
21
(+0.6% per annum compared to -0.1% per annum measured). For women the figures are
consistent (0.7% per annum compared to 0.6% per annum measured). Over the 9.5 years
between the 1995 and 2004-05 NHS, self-reported obesity prevalence rates increased by
6.7% for males (from 11% to nearly 18%) and 4.1% for females (from 11% to 15%) and the
annual average change in obesity prevalence rates was 0.7% for men and 0.4% for
women.
The NHS data provides the most recent indicator of trend in adult obesity prevalence. While
self-reported data on BMI have been consistently shown to under-estimate BMI (because
individuals tend to over-estimate their height and under-estimate their weight), this is unlikely
to substantially affect the trend in the data as the surveys use consistent methodology.9 This
is an advantage of using the trend from the NHS data compared with the trend in measured
obesity which is compiled from studies which used different methodologies. Another
advantage of using the NHS trend is that it contains more recent data than is available for
measured obesity.
2.5.2 TRENDS IN PREVALENCE IN CHILDREN AND ADOLESCENTS

Booth et al (2003) analysed changes in the prevalence of obesity among young Australians
aged 7 to 15 years between 1969, 1985 and 1997. Data for this analysis was drawn from five
studies: the Australian Youth Fitness Survey (1969); the Australian Health and Fitness Survey
(1985); the South Australian Schools Fitness and Physical Activity Survey (1997); the NSW
Schools Fitness and Physical Activity Survey (1997); and the Health of Young Victorians Study
(1997). Booth et al found a 2-4 fold increase in obesity prevalence rates between 1985 and
1997 (consistent across states and between sexes). Over the period 1969 to 1985 obesity
prevalence rates in boys trebled while there was no significant change in prevalence rates for
girls. Due to the different age groups used in the surveys it is not possible to calculate the
average change in prevalence rates for the whole period of 1969 to 1997. Between 1969 and
1985, the annual change in obesity prevalence rates (for children aged 13-15 years) was an
increase of 0.09% for boys and a decrease of 0.06% for girls. Over the period from 1985 to
1997 prevalence rates (for children aged 10-12 years) increased annually by 0.18% for boys
and 0.16% for girls.10
Magarey et al (2001) compared prevalence of obesity in children aged 7-15 years between
1985 (Australian Health and Fitness Survey data) and 1995 (NNS data). Over the ten year
period between the surveys, the prevalence rate of obesity rose by 3.3% for boys (rising from
1.4% to 4.7%), and 4.3% for girls (rising from 1.2% to 5.5%). This increase in obesity
prevalence rates equates to an annual increase of 0.33% for boys and 0.43% for girls.
Looking at data for NSW alone (SPANS), prevalence data are available for 1985, 1997 and
2004 (Booth et al, 2006). Over the period from 1985 to 2004, the prevalence rate of obesity in
NSW school children increased by 6.3% in boys and 4.6% in girls, or on an annual basis by
0.33% for boys and 0.24% for girls. Over the most recent period, from 1997 to 2004, obesity
prevalence rates increased annually by 0.39% for boys and 0.24% for girls. Trends in
obesity prevalence rates for NSW school children are shown in Figure 2-4.
9
It is noted that there are a number of characteristics of self-reported BMI data that could influence the trend. For
example in the ABS 1998 How Australians Measure Up, (a document comparing differences in findings of the NNS
and NHS in relation to BMI measures), it was found that heavier people generally under-report their weight by more
than lighter people – suggesting that as the population gets heavier, self-report data may get even less accurate.
10
Booth et al (2006) found that obesity in NSW increased substantially faster over the same period.
22
FIGURE 2-4: TRENDS IN OBESITY PREVALENCE RATES FOR NSW SCHOOL CHILDREN: 1985-2004
8%
Boys
7% Girls
6%
5%
4%
3%
2%
1%
0%
85
97
04
19
19
20
Increasing childhood obesity prevalence rates are not unique to Australia. Ebbeling et al
(2002) report on trends in childhood obesity in the United States, England and Egypt. In the
US, prevalence of obesity increased by 2.3-fold to 3.3-fold over approximately 25 years to
1999. In England, obesity prevalence rates increased by 1.1% for boys and 1.3% for girls in
the decade to 1994. In the 18 years to 1996, Egypt experienced a 3.9-fold increase in obesity.
Ebbeling et al (2002) also noted the skewed fashion in which the distribution of BMI has
shifted: children with higher BMI also have the greatest increase in BMI.
2.5.3 PROJECTIONS
This section presents projections of obesity prevalence for the years 2005, 2010, 2015, 2020
and 2025. The projections are based on three different scenarios:
1 Baseline: prevalence rates remain unchanged to 2025;
2 Elimination: prevalence declines to zero by 2025; and
3 Growth: prevalence increases in-line with recent trends.
Scenarios for prevalence projections are not intended to indicate what is considered likely to
happen but rather what could happen. The future prevalence of obesity is not predetermined
by current levels of obesity. Obesity is a condition that is highly open to intervention. What
obesity prevalence in 2025 looks like will be determined largely by what society and individuals
choose to do about it in the meantime.
2.5.3.1 BASELINE PROJECTION: PREVALENCE RATES UNCHANGED TO 2025
Under this first scenario the prevalence rates remain constant at the levels in Table 2-6. The
number of obese people in Australia grows due to population growth and demographic aging
23
(a larger proportion of the population in the age groups with highest prevalence rates for
obesity). Figure 2-5 shows projected obesity prevalence under Baseline assumptions. By
2025 a total of 4.2 million Australians are forecast to be obese. This is over 900,000 new
cases of obesity over a 20 year period.
FIGURE 2-5: PREVALENCE PROJECTIONS TO 2025, BASELINE
2,500,000
Males - total
Females - total
2,000,000
1,500,000
1,000,000
500,000
-
2005 2010 2015 2020 2025
2.5.3.2 ELIMINATION PROJECTION: PREVALENCE TO ZERO BY 2025
This second scenario assumes that obesity is completely eliminated by 2025, and that
prevalence decreases at a linear rate between 2005 and 2025. Prevalence rates for 2005 are
as for the baseline scenario (Table 2-6). For example, under the Baseline scenario, the rate of
obesity in men aged 55-59 for 2005 was 25.5%. Assuming a linear decline in prevalence, by
2015, the prevalence of obesity in 55-59 year old men is modelled to halve to 12.75%.
Figure 2-6 shows projected obesity prevalence under elimination assumptions. Under this
scenario, the number of PWO decreases by 615,000 in the five years to 2010. In the five
years from 2020 to 2025 the number of PWO decreases by 994,000.
24
FIGURE 2-6: PREVALENCE PROJECTIONS TO 2025, ELIMINATION
2,000,000
Males - total
1,800,000
Females - total
1,600,000
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
-
2005 2010 2015 2020 2025
2.5.3.3 GROWTH PROJECTION: RATES CONTINUE TO INCREASE TO 2025
The third scenario assumes a constant growth in obesity prevalence until 2025. The annual
growth rates assumed in this scenario are taken from the most-recent available 10-year trend
in adult prevalence (NHS change from 1995 to 2004-05) and the most recent available 7-year
trend for children and adolescents (SPANS change from 1997 to 2004). The annual
percentage increase in prevalence rates (by age and gender) are provided in Table 2-7.
TABLE 2-7: ANNUAL INCEASE IN OBESITY PREVALENCE RATES
Age Males Females
5-19 0.39% 0.24%
20+ 0.7% 0.4%
It is assumed that prevalence rates have grown by these amounts since the date when the
prevalence rates for the baseline assumptions (Table 2-6) were measured. This means that
the prevalence estimate for 2005 is also different under this scenario.
Figure 2-7 shows projected obesity prevalence under Growth assumptions. Under this
scenario, there are estimated to be 3.7 million Australians who are obese in 2005, increasing
to 7.2 million obese Australians by 2025. Over this period an additional 2.0 million men and
1.4 million women would become obese.
25
FIGURE 2-7: PREVALENCE PROJECTIONS TO 2025, GROWTH
4,500,000
Males - total
4,000,000
Females - total
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
2005 2010 2015 2020 2025
26
3. HEALTH IMPACTS OF OBESITY

AFs are the proportion of a health condition (eg, its prevalence, mortality, disease burden or
dollar costs), that is caused by – ie, aetiologically attributable to – a particular risk factor, after
controlling for other potentially confounding factors. AFs are useful in understanding the
extent to which the prevalence – and hence costs – of various conditions could be reduced
through changes in their modifiable risk factors, such as obesity. The application of AFs aims
to identify the scope for health gains, in terms of both dollar expenditures and healthy life
years lost as a result of mortality or morbidity, that may be possible through improvements in
obesity prevalence rates.
Even the best estimates used for the AF of obesity contain an amount of uncertainty. Other
confounding (risk) factors may have been inadequately controlled in source studies or may be
associated with obesity, such as physical inactivity, alcohol abuse or inadequate diet.
Statistical problems in regression analysis (such as multicollinearity) may be encountered,
precipitated by the complex biochemical and physiological inter-relationships between what,
ideally, should be ‘independent’ variables.
That said, a number of epidemiological studies have provided a body of evidence that there is
an overall increase in risk among people who are obese in the conditions represented in this
study. Mathers et al (1999) used these studies and some other supporting works to estimate
the RR for overweight or obese persons of incurring each causally related disease. These
risks are presented in Table 3-1.
TABLE 3-1: RR ASSOCIATED WITH OVERWEIGHT AND OBESITY

Overweight (BMI 25-59) Obese (BMI 30 and over)
Males Females Males Females
<65 65+ <65 65+ <65 65+ <65 65+
CHD 1.35 1.00 1.40 1.00 1.80 1.20 1.20 1.25
Ischaemic Stroke 1.35 1.00 1.35 1.00 1.50 1.15 1.60 1.20
Hypertension 1.40 1.40 1.40 1.40 2.35 2.35 2.35 2.35
Osteoarthritis 1.35 1.35 1.35 1.35 2.40 2.40 2.40 2.40
Colorectal Cancer 1.20 1.20 1.20 1.20 1.40 1.40 1.40 1.40
Kidney Cancer 1.00 1.00 1.00 1.00 1.00 1.00 1.50 1.50
Breast Cancer - - 1.00 1.00 - - 1.30 1.30
Uterine Cancer - - 1.00 1.00 - - 1.75 1.75
Source: Mathers et al (1999: Table 7-8)
3.1 DIABETES
Diabetes mellitus is a condition characterised by persistently high blood glucose level,
resulting either from the body’s inadequate production of the hormone insulin or an inadequate
response of target cells to insulin or a combination of these factors.
There are many causes and forms of diabetes. The three most common forms are Type 1
(juvenile-onset or insulin-dependent) diabetes, Type 2 (adult-onset or non-insulin dependent)
diabetes and gestational diabetes. Type 1 diabetes accounts for 10% to 15% of Australian
people diagnosed with diabetes, Type 2 diabetes for 85% to 90% of Australians. Gestational
27
diabetes accounts for the remaining (small number of) cases, as it is developed by only 3% to
8% of pregnant women not previously diagnosed with diabetes. However, about half of these
women will develop Type 2 diabetes later in life.
Diabetes is the sixth leading cause of death in Australia, claiming over 3,300 lives each year.
The death rate for diabetes in 2004 was higher for males (21.0 per 100,000) than for females
(13.8 per 100,000) after adjusting for age differences (ABS, 2006a). In 2004, diabetes (Type 1
and 2) was the underlying cause of death for 3,599 people: 1,869 males and 1,730 females
(ABS, 2006b).
While Type 2 diabetes has strong age and genetic associations, there are a number of other
risk factors that increase the possibility of developing this form of diabetes (Table 3-2). These
risk factors may act alone, but often act together in complex interplay. Obesity and lack of
exercise are major risk factors and it has been shown that the incidence of Type 2 diabetes
(which used to be known as adult-onset diabetes and was rarely experienced in people under
45) in children and adolescents is rising in parallel with higher rates of obesity. However, on
the same account, diabetes may be prevented or delayed by lifestyle interventions.
TABLE 3-2: RISK FACTORS FOR TYPE 2 DIABETES
Source: AIHW (2002)
Diabetes can result in many secondary long term health conditions, especially if it is
undetected or poorly controlled. Hence, there are not only direct health costs associated with
diabetes but also indirect health costs due to an increased risk of other diseases and
complications arising from diabetes: one third of people with diabetes experience
complications such as eye problems, kidney damage, foot ulcers, heart attack, stroke and
amputation (Colagiuri et al, 2003).11 Of 8,536 people with Type 2 diabetes aged 40 years and
over, eye problems were the most common complication (experienced by 27% of these
people), followed by kidney damage (10%), foot or leg ulcers (9%) and heart attack (9%).
Amputation (5%) was a less common but important complication. The presence of
complications has a significant impact on use of health services, receipt of government
benefits and the likelihood of having a carer. The number of GP, outpatient and emergency
department visits increases with the severity of complications.
Diabetes has become one of the most common non-communicable diseases in the world. For
this study for Australia, a number of sources were investigated for age-gender prevalence
rates. The AusDiab study (International Diabetes Institute, 2001) was found to be the best
source methodologically as it uses clinical diagnosis and thus uncovers undiagnosed as well
as diagnosed population prevalence.12 However, the data are now rather dated, being from
1999 and so the results are possibly conservative if age-gender rates have increased
11
It is important to note that diabetes is an independent risk factor for heart disease and stroke, so the
cardiovascular impacts are not double counted in this analysis.
12
AusDiab showed that, between 1981 and 1999, the number of adults with diabetes trebled.
28
significantly since. The AusDiab follow-up in 2004-05 was biased in it prevalence estimates
(although incidence estimates showed an increase), and it is important that another
epidemiological study is undertaken using similar methods to AusDiab in order to make more
accurate current estimates and analyse rates of change to inform policy.
The AusDiab study showed that, in 1999, around 5% of the Australian population were
estimated to have Type 1 or 2 diabetes. Interestingly, for every known case of diabetes, there
was one undiagnosed case (diagnosed through physical examination as part of the study).
The prevalence rate for diabetes increases with age and overall prevalence is higher for males
than females (5.1% and 4.8% respectively). Males had higher rates of diabetes than females
for all age groups between 35 and 74 years, while females had higher rates under 35 years
and over 75 years (Table 3-3 and Figure 3-1).
TABLE 3-3: PREVALENCE (%), DIABETES (TYPE 1 AND 2), AUSTRALIA, 1999
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+ Total
Males 0.0* 0.0* 0.1 2.6 6.8 16.1 21.6 22.4 5.1
Females 0.0* 0.0* 0.4 2.3 5.5 9.9 16.1 24.5 4.8
Persons 0.0* 0.0* 0.3 2.4 6.1 13.1 18.6 23.7 5.0
Source: International Diabetes Institute (2001)
* Data are not available, but youth prevalence rates for diabetes are low and are therefore rounded to 0
Two weaknesses in the AusDiab data from the perspective of this analysis were the lack of
robust reported prevalence in Australians aged under 25 years and the need to split Type 1
from Type 2, since Type 2 diabetes is the type strongly aetiologically associated with obesity.
Since Type 2 diabetes occurs overwhelmingly in people aged over 25 years, this age-group is
thus (a little conservatively) the basis of prevalence estimates in this report. The ABS NHS
data from 2001 were used to establish the relative rates of Type 1 and Type 2 diabetes in
these adults, noting that although more recent data are available from the NHS than AusDiab
(the most recent NHS was in 2004-05) for both prevalence rates, the data are self-reported –
ie, excluding undiagnosed cases – hence AusDiab was overwhelmingly favoured. The year
2001 was used as it most closely matches the AusDiab year. The NHS data are also useful in
investigating trends in diabetes prevalence over time (the first NHS study was in 1995), while
noting that if rates of awareness and diagnosis have improved over the past decade then the
NHS data would overstate the increase in prevalence of diabetes. Table 3-4 shows that, in
2001, Type 2 diabetes accounted for 83% of total diabetes in males and 87% of total diabetes
in females.
TABLE 3-4: TYPE 2 DIABETES AS A SHARE OF TYPE 1 AND 2 DIABETES (%), AUSTRALIA, 2001
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+ Total
Males --- --- 3% 61% 81% 90% 84% 92% 83%
Females --- --- 34% 59% 77% 96% 92% 89% 87%
Source: ABS (2002)
29
FIGURE 3-1: PREVALENCE (%) OF TYPE 1 AND 2 DIABETES BY GENDER, AUSTRALIA, 1999
30
25
20
% 15
10
0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Males, Type 1&2 Males, Type 2 Females, Type 1&2 Females Type 2
Source: Access Economics based on International Diabetes Institute (2001) and ABS (2002)
The age-gender prevalence rates established from the data sources were applied to Australian
demographic data from the ABS for the year 2005. In 2005, there were an estimated
1.1 million Australians with diabetes. The vast majority, 86%, had Type 2 diabetes. Of the
943,334 people with Type 2 diabetes, 52% were male, while 95% were aged 45 years or over
and more than half (53%) were aged 65 years or over. Most men with diabetes were aged 55
to 64 years while most women with diabetes were aged 75 years or older (Table 3-5).
TABLE 3-5: PREVALENCE, DIABETES (TYPE 1 AND 2), ADULTS 25 YEARS OR OLDER, 2005
25-34 35-44 45-54 55-64 65-74 75+ Total
Males, Type 1 & 2 1,438 38,890 94,415 177,427 147,231 115,272 574,672
Females, Type 1 & 2 5,729 34,746 77,351 107,702 115,127 184,157 524,813
Persons, Type 1 & 2 7,167 73,636 171,766 285,129 262,358 299,429 1,099,485
Males, Type 2 43 23,810 76,051 159,356 123,460 105,986 488,706
Females, Type 2 1,946 20,646 59,248 103,381 106,086 163,321 454,628
Persons, Type 2 1,989 44,457 135,298 262,736 229,546 269,308 943,334
Source: International Diabetes Institute (2001), ABS (2002)
These estimates are conservative as they assume a constant prevalence rate between 1999
and 2005. However, it is likely that the prevalence rate may have increased during that time
period. AusDiab 2005 (International Diabetes Institute, 2006), the follow-up survey of the
earlier AusDiab Report, claims that every year 0.8% of Australian adults aged 25 years or
older develop Type 1 or Type 2 diabetes.13 Every day in Australia approximately 275 of these
adults develop diabetes – ie, around 100,000 adults each year (including around 50,000 newly
diagnosed cases). Diabetes Australia estimates that there may be closer to 75,000 newly
diagnosed cases each year. If only ageing is taken into account, the number of people with
13
This estimate is based on the AusDiab finding (from examination results of 11,247 adults aged 25 years and
older) that eight out of every 1000 people in Australia developed diabetes.
30
diabetes would have increased by less than 200,000 between 1999 and 2005 – from 938,700
to around 1.1 million), implying a mortality rate of 6.5% per annum for people with diabetes
(the average mortality rate of an 80 year old Australian male or an 83-year old female).
According to Mathers et al (1999:Table 7-8) – based on studies by Carey et al (1997), Colditz

et al (1990 and 1995) and Njolstad et al (1998) – overweight people are 1.8 times more likely
to develop adult-onset (Type 2) diabetes than those of normal weight, while obese people are
3.2 times more likely. NHMRC (2005) gives a broader overview of RR measures, showing the
results of 22 studies that examined the relationship between weight and Type 2 diabetes
(including the four studies used in Mathers et al, 1999). It was found that a BMI ≥ 30 results in
a 2-fold increase in the risk of Type 2 diabetes. Increased waist height ratio and weight
circumference are also associated with increased risk of Type 2 diabetes but available data do
not allow quantification of risk relative to a clinically applicable measurement level. According
to the AIHW the impact of obesity on diabetes is even stronger - overweight adults are three
times more likely to develop Type 2 diabetes than those of ideal weight, while obese people
are ten times more likely than persons of ideal weight (AIHW, 2002).
Based on a ratio of 52.9% of Australians being of normal weight, 31.1% of people being
overweight and 15.9% of people being obese14 and the RRs associated with obesity from
Mathers et al (1999) – for consistency with the other conditions due to obesity discussed in
this report, it was estimated that 10.8% of people with Type 2 diabetes are diabetic as a
consequence of being obese.
Based on these estimates, in 2005 102,204 Australians had Type 2 diabetes as a

result of being obese.
3.2 CVD
CVD is also known as ‘circulatory disease’ or as ‘heart, stroke and vascular disease’ and
refers to all diseases and conditions of the heart and blood vessels. CVD is one of the main
sources of mortality (and morbidity) due to obesity, particularly in the longer term. Obesity is a
substantial risk factor for CVD. Figure 3-2 displays the positive correlation between increased
BMI and higher CVD mortality risk. The RR of death from CHD exceeds 2 for both males and
females when BMI levels exceed 30kg/m2, while the risk of death from stroke is elevated 25%
for males and 50% for females.
14
Based on prevalence estimates from Chapter 2.
31
FIGURE 3-2: AGE-ADJUSTED RELATIVE MORTALITY RISK, BY BMI FOR US SAMPLE POPULATIONS
Females Males
Source: NHMRC Obesity Guidelines
AFs of obesity in CVD are calculated based on the estimated prevalence of CHD, stroke and
hypertension in the Australian population and the RR (Mathers et al 1999) of contracting these
conditions given the individual is obese.
The 2001 NHS (ABS, 2002) provides detailed information on the self-reported prevalence of a
variety of CVDs in Australia in 2001 with demographic breakdowns. The NHS data are limited
due not just to self-reporting problems but also to non-clinical categorisation of CVD, thus the
prevalence of ischaemic stroke is derived from AIHW (2004), which is also, however, based on
NHS original data. Two other barriers to CVD estimation in Australia are the lack of
universally agreed definitions (eg, hypertension can be considered a risk factor and a disease
in its own right) and difficulties in diagnosis, particularly when the condition is mild.
In this report, hypertension (high blood pressure) is treated as a disease, with
downstream risk for HHD and PAD. While hypertension is also a risk factor for CHD and
stroke, this link is not included, to avoid potential double counting of people and costs.
There is a continuous relationship between blood pressure levels and the risk of
CHD, stroke, heart failure, peripheral vascular disease and kidney failure. World
Health Organization (WHO) and the National Heart Foundation of Australia
guidelines define ‘high’ blood pressure as systolic pressure at or above 140 mmHg
or diastolic pressure at or above 90mmHg, or receiving medication for high blood
pressure. Major contributors to high blood pressure include poor diet (especially
high salt intake), overweight, excessive alcohol consumption and insufficient
physical activity.
CHD or ischaemic heart disease is the most common cause of sudden death in
Australia. Its main manifestations consist of acute myocardial infarction (AMI, or heart
attack) and angina. The common underlying problem is atherosclerosis, which is
plaque build-up on the inside of arteries.
Stroke (or cerebrovascular disease) is Australia’s second greatest killer after CHD and
the leading cause of long term disability in adults. Stroke occurs when a blood vessel
that carries oxygen and nutrients to the brain is either blocked by a clot (ischaemic
stroke) or, les frequently, bleeds (haemorrhagic stroke). This can cause death, or
damage part of the brain, which in turn can impair a range of functions such as
movement of body parts, vision and communication. About one-third of people
32
sustaining stroke die within 12 months and half of the survivors are disabled in the
longer term.
Using these definitions, accounting for the demographic population change between 2001 and
2005 and assuming that prevalence rates remain constant over time, Access Economics
estimates that 2.8 million Australians were affected by obesity related CVD in 2005,
representing approximately 13.9% of the population. Figure 3-3 provides breakdowns by age
and gender.
FIGURE 3-3: PREVALENCE OF OBESITY RELATED CVDS, 2005, BY AGE AND GENDER
600,000
Column 1 -
Hypertension
500,000
400,000
Prevalence (No. People)
300,000
Column 2 - CHD
200,000
Column 3 - Stroke
100,000
0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male Female
Source: Access Economics derived from ABS (2002:Table 5) and AIHW (2004)
A relatively small number of children and young adults (aged 0 to 34 years) have stroke
and CHD. However hypertension is evident in 50,700 young Australians – 10% of the
people aged under 35 years). Overall, age as a risk factor for CVD is clearly illustrated.
Among the working population, hypertension affects 1.1 million Australians aged 15 to
64 years (with an approximately equal split between males and females). A further
96,000 and 146,400 working age Australians report stroke and CHD as long term
conditions.
Within the population of PWO related CVD, females comprise 52% of the total (or
1.5 million people). Older females (75 years and older) are more at risk of CVD (62% of
the oldest age group) compared to males (38%).
33
FIGURE 3-4: PREVALENCE OF OBESITY IN SELECTED CVD POPULATIONS, 2005
300,000
250,000
Column 1 -
Obesity Prevalence in CVD Population
Hypertension
200,000
150,000
Column 2 - CHD
100,000
Column 3 - Stroke
50,000
0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male Female
Figure 3-4 shows the prevalence of obesity in each CVD group, by age and gender. From a
total of 980,500 Australians with obesity related CVD in 2005, 561,800 (57%) were female,
while 790,600 (81%) had hypertension, 120,700 (12%) had CHD and 69,200 (7%) had stroke.
As noted previously, people who are obese are at a higher risk of contracting hypertension
which, in turn, causes HHD with an AF of 100% and PAD with an AF of 9% imputed by Access
Economics from Mathers et al (1999) data. Therefore, the obesity AF for PAD can be derived
as 9% of the HHD/hypertension obesity AF. Table 3-6 shows the age and gender distribution
of obesity AFs in relation to CVD.
Based on these estimates, in 2005 over 379,000 Australians had CVD (CHD,
ischaemic stroke and hypertension) as a result of being obese.
34
TABLE 3-6: FRACTION OF CVD PREVALENCE ATTRIBUTABLE TO OBESITY
Age Group Male Female Total Male Female Total

HHD CHD
0-14 4% 6% 4% 4% 6% 4%
15-24 7% 6% 7% 4% 6% 6%
25-34 11% 9% 10% 11% 13% 12%
35-44 11% 14% 12% 11% 13% 12%
45-54 12% 17% 15% 12% 17% 13%
55-64 14% 19% 17% 14% 19% 16%
65-74 11% 17% 15% 11% 17% 13%
75+ 8% 11% 10% 8% 10% 9%
Total 12% 16% 14% 11% 14% 12%
Ischaemic Stroke PAD
0-14 0% 0% 0% 0% 0% 0%
15-24 7% 6% 6% 1% 1% 1%
25-34 11% 9% 10% 1% 1% 1%
35-44 11% 13% 12% 1% 1% 1%
45-54 12% 17% 13% 1% 2% 1%
55-64 15% 19% 16% 1% 2% 2%
65-74 11% 17% 14% 1% 2% 1%
75+ 8% 10% 9% 1% 1% 1%
Total 12% 14% 12% 1% 1% 1%
3.3 OSTEOARTHRITIS
Obesity is one of the most preventable risk factors for osteoarthritis due to extra weight placing
pressure on joints, particularly knee and hip joints. The US National Centre for Chronic
Disease Prevention and Health Promotion (Centre for Disease Control, 2004) cites studies
showing the effectiveness of weight loss and physical activity in reducing arthritis symptoms.
A randomised controlled study among women showed that the amount of weight lost
was strongly correlated with improvements in signs and symptoms of knee osteoarthritis
(Felson et al, 1992).
Regular exercise reduced pain and improved physical performance among older people
with disabling osteoarthritis of the knee (Ettinger et al, 1999; Minor and Allegrante,
1997).
The prevalence and costs of osteoarthritis in Australia have previously been calculated for the
year 2004 by Access Economics (2005b) and are used as the basis of the prevalence and
costing in this analysis. Since there were no superior Australian epidemiological studies of
osteoarthritis prevalence, and since self-reported data are relatively reliable for osteoarthritis,
osteoarthritis prevalence data by age and gender were obtained in Access Economics (2005b)
from the 2001 NHS (ABS, 2002). Population projections by age and gender from ABS (2003a)
were applied to the NHS data to generate age and sex specific prevalence estimates for 2005.
The 2001 NHS reported 1.39 million people with osteoarthritis in Australia, 46% of all people
with arthritis and 7.3% of the population, significantly higher than the 6.4% who reported the
condition in the previous 1995 NHS (largely due to demographic ageing). Accounting for
demographic change from 2001 to 2005 and with prevalence rates constant for each age-
35
gender group, Access Economics estimates that 1.6 million Australians were affected by
osteoarthritis in 2005 (Table 3-7), approximately 7.9% of the population.
TABLE 3-7: PREVALENCE OF OSTEOARTHRITIS, 2005, BY AGE AND GENDER
Age Male Female Total

No. (000) % No. (000) % No. (000) %
0-24 5.3 0.15 5.0 0.15 10.3 0.15
25-34 15.8 1.10 22.5 1.56 38.4 1.33
35-44 38.2 2.55 61.8 4.07 100.3 3.32
45-54 102.7 7.40 164.0 11.64 266.8 9.53
55-64 165.4 14.97 262.7 24.06 428.5 19.51
65-74 126.7 18.57 237.4 33.16 365.1 26.11
75+ 121.2 23.60 281.9 37.43 405.2 31.98
Total 575.4 5.70 1,035.3 10.11 1,614.4 7.94
Given current demographic trends, overall prevalence is expected to increase further over
time. By 2020, Access Economics estimates that one in every ten Australians will have
osteoarthritis, a projected increase of 24% from 2005. These projections, however, do not
allow for other factors – such as a new intervention that might delay or prevent osteoarthritis.
A forecast increase of this magnitude nonetheless suggests substantial scope for potential
policy impacts. A reduction in obesity rates, as the primary risk factor impacting on
osteoarthritis prevalence, could thus potentially greatly reduce the health impacts and costs of
osteoarthritis in Australia.
Figure 3-5 illustrates osteoarthritis prevalence by age groups, gender and weight range.
Osteoarthritis is more prevalent in women than in men, and in older people, with the
gender bias towards women being more pronounced in the older age cohorts.
Of Australians in the age group 75 years or older, 32% had osteoarthritis in 2005 and, of
those with osteoporosis, 27% were obese (24% of the oldest males with osteoarthritis
and 29% of the oldest females with osteoarthritis were obese).
Despite the higher prevalence of arthritis in older age groups, over 52% of all people with
osteoarthritis were of working age (defined, due to data restrictions, as 25-64 years of age).
Of the working age population, 43% were obese (38% of working age males with osteoarthritis
and 46% of working age females with osteoarthritis were obese).
36
FIGURE 3-5: PREVALENCE OF OSTEOARTHRITIS, IN 2005, BY AGE, GENDER AND WEIGHT RANGE
40 300
35
250
30
Population with Osteoarthritis (000)

Population with Osteoarthritis (%)
200
25
Column 2 - Females
20 150
Column 1 - Males
15
100
10
50
5
0 0
0-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Normal Overweight Obese Male (%) Female (%)
Table 3-9 presents obesity AFs for osteoarthritis by age group and gender calculated for the
Australian population from the data for the prevalence of obesity, prevalence of osteoarthritis
and from the RR ratios in Table 3-1 at the beginning of this chapter. Notably, Table 3-1
showed that obese people are 2.4 times more likely to have osteoarthritis (the highest RR in
that table). This is reflected in the results in Table 3-9, which displays the substantial
contribution to osteoarthritis prevalence as a result of obesity, particularly among working age
Australians – 14% of people with osteoarthritis have the condition as a consequence of being
obese.
TABLE 3-8: FRACTION OF OSTEOARTHRITIS PREVALENCE ATTRIBUTABLE TO OBESITY
Age Group Male Female Total

0-24 5.5% 4.7% 5.1%
25-34 10.8% 9.3% 9.9%
35-44 10.8% 13.3% 12.3%
45-54 12.0% 16.6% 14.8%
55-64 14.1% 18.9% 17.0%
65-74 11.5% 17.4% 15.3%
75+ 8.1% 10.7% 9.9%
Total 11.5% 15.4% 14.0%
Based on these estimates, in 2005 over 225,000 Australians had osteoarthritis as

a result of being obese.
37
3.4 CANCERS
Obesity has been associated with higher risk of four types of cancer – colorectal, kidney,
breast and uterine cancers. In NSW in 2005 it has been estimated from notifications data
(Access Economics, 2006c) that there were around 33,700 new cases of cancer, affecting
18,400 males and 15,200 females (around 0.5% and 0.4% of the male and female population,
respectively). On an incidence basis, colorectal cancer (14%) and breast cancer (13%) were
two of the most common cancers, while kidney cancer and uterine cancer made up 3% and
4% respectively of new cancer cases. Together these account for one third (33%) of all
incident cancers.
Cancer prevalence in Australia is estimated using data calculated from NSW notifications,
extrapolated to the Australian population by age-gender group for each cancer type. The
NSW prevalence rates are complex to estimate as they depend on defining a timeframe of
remission after which cancer survivors are considered ‘cured’. Some patients may die from
the cancer or from other causes so estimating the annual prevalence of cancer is more
problematic than estimating annual incidence or mortality. Brameld et al (2002) used
incidence and survival rates to estimate the number of people actively being treated for cancer
in WA. A similar methodology was used by Access Economics – using relative survival
probabilities and treatment case timeframes – to estimate the prevalence of cancer in NSW for
the year 2005.
Point prevalence is the cumulative number of cancer patients diagnosed prior to 2005
but who have not yet died, at the current point in time. This is estimated by multiplying
the incidence of cancer in NSW in each of the past five years (by age, sex and cancer
type) by the corresponding survival rate (by age, sex and cancer type). For example,
2004 incidence is multiplied by the survival rate after one year, 2003 incidence is
multiplied by the survival rate after two years, and so on. If the patient did not die (from
cancer or other causes) in the past five years then the patient is assumed to be cured
and no longer have cancer.
Active prevalence is defined as the cumulative number of cancer patients diagnosed
prior to 2005 and who have not yet died, but on the basis of probability will die from their
cancer (thus will require health care for active cancer now or in the future). Assuming
that all cancer patients will receive some form of treatment (regardless of whether they
are likely to die), incidence in 2005 will also be added to this estimate.
Both methodologies may underestimate the small number of patients who continue to
experience the impacts of cancer past the five year cut off, but may overestimate those that
might be considered ‘cured’ (under a different definition) prior to five years. Furthermore these
methodologies will underestimate the impacts on patients who, while free from their primary
disease, require further treatment to cope with ongoing physical or psychosocial disabilities
(eg, replacement of prosthetic devices, long term impacts of chemotherapy).
In NSW in 2005 there were around 125,900 prevalent cases (point prevalence) of all types of
cancer (diagnosed since 2000 and still alive), of which 66,000 were male and 59,900 were
female – around 1.9% of the NSW population. The most prevalent cancer was breast cancer
(17%), while colorectal cancer, kidney cancer and uterine cancer prevalence were 13%, 3%
and 4% respectively – in total the obesity related cancers account for 37% of all prevalent
cases.
38
FIGURE 3-6: PREVALENCE OF OBESITY RELATED CANCERS, 2005, BY AGE AND GENDER
20,000
18,000 Column 1 - Breast
16,000
Prevalence (No. People)
14,000
12,000 Column 2 - Colorectal
10,000
Column 3 - Kidney
8,000
6,000
Column 4 - Uterine
4,000
2,000
0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male Female
Accounting for the demographic differences between NSW and Australia and assuming that
prevalence rates remain constant for each State/Territory population, Access Economics
estimates that 137,225 Australians had actively prevalent obesity related cancers in 2005, with
the age-gender distribution illustrated in Figure 3-6. This represents approximately 0.7% of
the population. Demographic trends presented in the chart include:
A relatively small prevalence of obesity related cancer in children and young adults,
aged 0 to 34 years. Generally, the most common cancers found in children are
leukaemia (34% in 2005), brain cancer (11% in 2005) and Non-Hodgkin’s Lymphoma
(6% in 2005).
Among the working population, the most prevalent cancers are breast cancer among
females (41% of all female obesity related cancer cases) and colorectal cancer among
males (33% of all male obesity related cancer cases).
In the older female population (65 years or older), colorectal cancer also becomes a
major concern (13% of all female obesity related cancer cases), second only to breast
cancer (20%).
Three quarters (76%) of prevalent obesity related cancers are in females (103,798
women in 2005). A reduction in the incidence of breast cancer would significantly
reduce cancer prevalence in the female population.
The prevalence of obesity among people with prevalent obesity related cancer is even more
strongly gender-related, with 81% (16,598) of the 20,430 obese cancer patients being female.
Figure 3-7 depicts the prevalence and AFs of obesity among the individual cancer
populations by age and gender.
39
FIGURE 3-7: PREVALENCE AND AFS OF OBESITY IN SELECTED CANCER POPULATIONS, 2005
20% 9,000
18% 8,000
16%
Column 2 - Breast 7,000
Obesity Prevalence in Cancer Population

14% Column 1 - Colorectal
Obesity Attributable Fraction
6,000
12%
5,000
10%
4,000
8%
Column 3 - Kidney
3,000
6%
2,000
4%
Column 4 - Uterine 1,000

2%
0% 0
+
14
75
-2
-3
-4
-5
-6
-7
0-
15
25
35
45
55
65
Age Group
Male Prevalence Female Prevalence Male AF Female AF
Using the RR ratios from Table 3-1 with the prevalence of the four cancers and of obesity and
overweight in Australian demographic groups in 2005, the AFs for obesity in each of the
cancer populations have been derived (Table 3-9). As observed previously, female obesity
has a relatively larger impact on cancer prevalence than male obesity, particularly within the
45 to 64 year old age group.
Based on the AFs by age-gender group for each cancer, in 2005, 20,430
Australians had cancer as a result of being obese.
40
TABLE 3-9: FRACTION OF CANCER PREVALENCE ATTRIBUTABLE TO OBESITY
Age Group Male Female Total Male Female Total

Colorectal Breast
0-14 4% 4% 4% 11% 6% 6%
15-24 7% 6% 6% 11% 6% 6%
25-34 11% 9% 10% 11% 9% 9%
35-44 11% 13% 12% 11% 13% 13%
45-54 12% 17% 14% 12% 17% 17%
55-64 15% 19% 17% 15% 19% 19%
65-74 11% 17% 14% 11% 17% 17%
75+ 8% 10% 9% 8% 11% 10%
Total 11% 15% 13% 11% 16% 16%
Kidney Uterine
0-14 4% 4% 4% - 4% 4%
15-24 7% 6% 6% - 6% 6%
25-34 11% 9% 10% - 9% 9%
35-44 11% 13% 12% - 14% 14%
45-54 12% 17% 14% - 17% 17%
55-64 15% 19% 16% - 19% 19%
65-74 11% 17% 14% - 17% 17%
75+ 8% 10% 9% - 11% 11%
Total 12% 15% 13% - 16% 16%
41
4. HEALTH EXPENDITURES
4.1 METHODOLOGY
Estimates for direct health system costs are derived in Australia by the AIHW from an
extensive process developed in collaboration with the National Centre for Health Program
Evaluation for the Disease Costs and Impact Study. The approach measures health services
utilisation and expenditure (private and public) for specific diseases and disease groups in
Australia. The Disease Costs and Impact Study methodology has been gradually refined over
the 1990s to now estimate a range of direct health costs from hospital morbidity data, case mix
data, Bettering the Evaluation and Care of Health data, the NHS and other sources. AIHW
(2005b) provides a summary of the main results of estimates of health expenditure by disease
and injury for the year 2000-01. The advantage of top-down methodology is that cost
estimates for various diseases will be consistent, enhancing comparisons and ensuring that
the sum of the parts does not exceed the whole (total health expenditure in Australia).
The AIHW recurrent data for 2000-01 were used as the base for Access Economics’ estimates
for health expenditure in 2005, disaggregated by age, gender and type of cost. These data
use categories based on the Tenth Revision of the International Classification of Disease
published by the World Health Organisation and the International Classification of Primary
Care Version 2. The AIHW includes only 86% of total recurrent health expenditure in its
estimate of expenditure by disease and injury, referred to as ‘allocated’ health expenditure.
The ‘unallocated’ remainder includes capital expenditures, expenditure on community health
(excluding mental health), public health programs (except cancer screening), health
administration and health aids and appliances.
The main factors contributing to the extrapolation were demographic growth by age-gender
groups (increasing prevalence) based on ABS data and health cost inflation, based on AIHW
(2005a). The latter measured 3.7% over 2000-01 to 2001-02, 4.1% over 2001-02 to 2002-03
and 3.8% over 2002-03 to 2003-04; for the 18 months from 2003-04 to the end of calendar
year 2005, health cost inflation is assumed to have averaged 3.2% per annum, which was the
average rate over the period 1997-98 to 2002-03. Thus overall inflation resulted in a 17.5%
increase over the whole period from 2000-01 to end-2005.
4.2 DIABETES
4.2.1 TYPES OF HEALTH SYSTEM COSTS ASSOCIATED WITH DIABETES

Based on health expenditure data provided by the AIHW for Type 1 and 2 diabetes and the
share of Type 2 diabetes provided by the ABS (2001), the allocated health costs arising from
Type 2 diabetes are estimated to be $921.6 million in 2005 (Table 4-1).
42
TABLE 4-1: TYPE 2 DIABETES, ALLOCATED HEALTH SYSTEM EXPENDITURE, 2005 ($M)
In- Out- Total Aged Out-of Pharma- Pharma- Other Re- Total
patients patients hospital care hospital ceuticals ceuticals health search
homes medical (pre- (over the profes-
services scription) counter) sionals
Males
0-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
15–24 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
25–34 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
35–44 4.7 1.2 5.8 0.8 3.7 4.5 0.3 0.7 0.7 16.4
45–54 19.6 4.9 24.5 3.2 15.5 18.7 1.1 2.8 3.0 68.8
55–64 45.8 11.5 57.2 7.5 36.2 43.8 2.6 6.5 6.9 160.8
65–74 33.1 8.3 41.4 5.4 26.2 31.7 1.9 4.7 5.0 116.3
75+ 31.2 7.8 39.0 5.1 24.7 29.8 1.8 4.5 4.7 109.5
Total 134.3 33.7 168.0 22.1 106.4 128.4 7.6 19.2 20.3 471.9
Females
0-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
15–24 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
25–34 0.2 0.1 0.3 0.0 0.2 0.2 0.0 0.0 0.0 0.7
35–44 3.9 1.0 4.9 0.6 3.1 3.8 0.2 0.6 0.6 13.8
45–54 14.5 3.6 18.2 2.4 11.5 13.9 0.8 2.1 2.2 51.0
55–64 31.7 8.0 39.7 5.2 25.1 30.4 1.8 4.5 4.8 111.5
65–74 31.3 7.8 39.1 5.1 24.8 29.9 1.8 4.5 4.7 109.9
75+ 46.3 11.6 57.9 7.6 36.7 44.3 2.6 6.6 7.0 162.8
Total 127.9 32.1 160.0 21.0 101.3 122.4 7.2 18.3 19.4 449.7
Persons
0-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
15–24 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
25–34 0.2 0.1 0.3 0.0 0.2 0.2 0.0 0.0 0.0 0.7
35–44 8.6 2.2 10.7 1.4 6.8 8.2 0.5 1.2 1.3 30.2
45–54 34.1 8.6 42.7 5.6 27.0 32.6 1.9 4.9 5.2 119.8
55–64 77.5 19.5 96.9 12.7 61.4 74.1 4.4 11.1 11.7 272.4
65–74 64.3 16.2 80.5 10.6 51.0 61.6 3.6 9.2 9.7 226.2
75+ 77.5 19.4 96.9 12.7 61.4 74.1 4.4 11.1 11.7 272.3
Total 262.2 65.8 328.0 43.1 207.7 250.8 14.8 37.5 39.7 921.6
Source: Access Economics based on AIHW (2005b)
Figure 4-1 shows health expenditure by type of health cost for people with Type 2 diabetes.
The majority (28.8%) of health expenditure is directed to pharmaceuticals –
$265.6 million in 2005. This item includes over the counter and prescription medication
such as insulin, oral hypoglycaemic agents, lipid lowering and blood pressure lowering
agents and non-steroidal anti-inflammatory drugs.
Inpatient expenditures were the second largest, comprising a further 28.4% or
$262.2 million.
Out-of-hospital medical services were the third most substantial cost element at
$207.7 million (22.5% of the total).
Outpatient expenditures accounted for 7.1% of total costs - 65.8 million.
Expenditure on aged care ($43.1 million), other health professionals ($37.5 million) and
research ($39.7 million) each accounted for 4% to 5% of total health costs.
43
FIGURE 4-1: TYPE 2 DIABETES, ALLOCATED HEALTH EXPENDITURE BY COST TYPE COST, 2005 (%)
4.3% Inpatients
4.1%
28.4% Outpatients
Aged Care
28.8%
Out-of hospital
medical
Pharmaceuticals
7.1%
Other health
professionals
4.7%
Research
22.5%
Figure 4-2 shows health expenditure by age and gender.

51% of total health spending ($471.9 million) is on males and 49% ($449.7 million) is on
females, reflecting prevalence proportions.
The greater prevalence of Type 2 diabetes in people aged 45 and over explains the
differences in expenditure patterns for different age groups. Most notably, 29.5% of
health expenditure is associated with adults aged 75 or older.
FIGURE 4-2: TYPE 2 DIABETES, ALLOCATED HEALTH EXPENDITURE BY AGE & GENDER, 2005 ($M)
300
Males Females
250
200
150
$m
100
50
0
0-14 15–24 25–34 35–44 45–54 55–64 65–74 75+
Age Group
44
Adjusting the health expenditure data for expenditures not allocated by AIHW (recall Section
3.1) brings the total cost of health expenditure on Type 2 diabetes to $1,071.7 million for
2005, or an estimated 0.12% of GDP.15 This is multiplied by the AF of 10.8% to estimate the
proportion of the health costs of diabetes due to obesity.
The total health costs arising from Type 2 diabetes caused by obesity were $116.1
million in 2005.
4.3 CVD
Direct health system costs for the four obesity related CVDs – CHD, stroke, HHD and PAD –
are based on AIHW 2000-01 data (AIHW, 2005b).
In 2000-01 the official health cost of the four CVDs was $2.6 billion, dominated by CHD
($1.5 billion), with stroke $895 million, PAD $200 million and HHD $72 million.
Hospital costs dominated the profile ($1.6 billion), followed by aged care ($473 million)
and pharmaceuticals ($254 million).
As noted earlier, these costs only include 86% of total recurrent health expenditure and need
to be factored up to include the cost of capital expenditures, expenditure on community health,
public health programs, health administration and health aids and appliances. In addition, the
2000-01 data were inflated to 2005 prices using health cost inflation data from AIHW (2005a)
and also increased to allow for prevalence growth based on increases for each age-gender
group in line with demographic changes.
The total health system costs are shown in Table 4-2 and Figure 4-3, together with the
proportion of costs attributable to obesity. In 2005, the health system cost of Australians
with the four obesity related CVDs was around $3.9 billion. The AFs by demographic
group and CVD type were used to derive the proportion of costs attributable to obesity.
The total health costs arising from CVD due to obesity were $428.3 million in 2005.
Overall, the health system costs caused by obesity for men ($222.7 million, 52%) were
slightly greater than for women ($205.6 million, 48%), despite the higher overall
prevalence of obesity related CVDs in women.
Over 85% of CVD health expenditure caused by obesity ($376.9 million, 88.2%) was
spent on people aged 55 years or older. Around two thirds of CVD health expenditure
caused by obesity ($281.8 million, 65.8%) was spent on people aged 65 years or older.
Health costs due to obesity for CHD ($264 million, 62%) and stroke ($149 million, 35%)
make up 97% of the total, while HHD and PAD cost only $12 million and $3 million
respectively.
15
Based on a GDP estimate of $892,486 million for 2005. The unallocated component is conservative as the AIHW
include in this component the National Diabetes Services Scheme Services Scheme which cost around
$103 million by FY2004-05 (ie, the 11% annual growth is much greater than the extrapolation based on the
average). The Scheme includes items such as Blood Glucose Testing Strips, Sharps (Needles and Syringes),
Minimal Urine testing strips & older style Pump Consumables and New Insulin Pump Consumables.
45
TABLE 4-2: TOTAL HEALTH SYSTEM COSTS FOR THE FOUR OBESITY RELATED CVDS, 2005
($m) Health Costs, four CVDs Health Costs attributable to Obesity

Age Group Male Female Male Female Total
0-14 $1.7 $1.4 $0.0 $0.0 $0.0
15-24 $4.1 $3.0 $0.1 $0.1 $0.2
25-34 $12.9 $7.3 $0.6 $0.4 $0.9
35-44 $69.4 $24.1 $6.9 $2.9 $9.8
45-54 $218.6 $91.6 $25.5 $14.8 $40.4
55-64 $466.7 $173.7 $63.5 $31.7 $95.2
65-74 $648.6 $346.9 $68.5 $55.4 $123.9
75+ $805.5 $1,027.2 $57.6 $100.2 $157.9
Total $2,227.4 $1,675.1 $222.7 $205.6 $428.3
Note: The four CVDs refer to CHD, stroke, HHD and PAD
FIGURE 4-3: CVD HEALTH COSTS, 2005, $’000, BY AGE, GENDER AND WEIGHT
Column 1 - CHD
180,000 $900
160,000 $800
Column 2 - Stroke
140,000 $700
120,000 $600
Total Health Costs ($m)

Obese CVD Population
100,000 $500
Column 3 - HHD
80,000 $400
60,000 $300
40,000 Column 4 - PAD $200
20,000 $100
0 $0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male Other Obese Male Female Other Obese Female Male Female
Note: The columns refer to the costs (right axis) and the lines to the population (left axis). “Obese male” and
“obese female” refer to the CVD health expenditure due to obesity
Figure 4-4 displays the contribution to obesity related CVD health costs by type of health cost.
Inpatient costs ($212 million, 49.5%), RAC costs ($67 million, 15.6%) and
pharmaceuticals ($35 million, 8.3%) comprise nearly three quarters of the total CVD
costs due to obesity.
Of the obesity related out-of-hospital medical costs, GP expenditure is $6.2 million
(1.5%), imaging and pathology $7.0 million (1.6%) and specialists $11.5 million (2.7%).
46
$4.0 million (0.9%) is spent on other (allied) health practitioners (OHPs) and research
into obesity related CVD is estimated as $10.3 million (2.4%) for 2005.
FIGURE 4-4: CVD HEALTH EXPENDITURE DUE TO OBESITY BY COST TYPE, 2005 (% TOTAL)
14%
Inpatients
2%
Outpatients
1% Aged Care
GPs
8%
Specialists
50%
2% Imaging & Pathology
3%
Pharmaceuticals
1%
OHPs
Research
Total Health Costs:

16% Unallocated
$3.90 billion
Obesity Related Costs: 3%

$428 million
4.4 OSTEOARTHRITIS
Based on the raw data provided by the AIHW, Table 4-3 presents the estimated 2005
allocated health expenditure on osteoarthritis of $1.48 billion.
47
TABLE 4-3: OSTEOARTHRITIS, ALLOCATED HEALTH SYSTEM EXPENDITURE, 2005 ($M)

In- Out- Aged GPs Specialists Imaging & Pharma- OHPs Research Total
patients patients care Pathology ceuticals Allocated
Males
0-14 0.0 - - 0.0 0.1 0.4 0.2 - 0.0 0.8
15–24 0.6 - - 0.1 0.0 0.0 0.2 0.5 0.0 1.5
25–34 2.6 0.1 - 0.4 0.2 0.4 1.0 - 0.1 4.9
35–44 8.4 1.0 - 1.2 0.5 1.8 2.5 4.7 0.2 20.5
45–54 24.9 1.2 - 3.1 0.8 4.7 7.5 3.7 0.5 46.6
55–64 67.7 10.1 - 5.5 1.8 7.7 16.8 6.8 1.4 117.9
65–74 97.7 7.7 5.7 5.5 2.6 6.5 20.7 12.2 1.9 160.5
75+ 78.4 14.0 45.0 5.9 3.4 5.7 18.1 3.2 2.1 175.6
Total 280.4 34.2 50.7 21.9 9.5 27.3 67.0 31.1 6.2 528.2
Females
0-14 0.0 - - 0.0 0.0 - 0.3 - 0.0 0.3
15–24 0.3 - - 0.1 0.0 0.1 0.2 - 0.0 0.7
25–34 1.4 1.5 - 0.3 0.1 0.7 0.6 - 0.1 4.7
35–44 5.7 3.9 - 1.1 0.5 1.5 3.1 3.2 0.2 19.2
45–54 24.1 3.4 - 4.6 1.9 9.7 13.0 7.1 0.8 64.4
55–64 67.8 19.7 - 8.2 3.5 11.5 28.8 14.4 1.8 155.6
65–74 112.9 14.4 11.3 8.8 3.5 12.7 31.9 11.0 2.5 208.9
75+ 117.6 14.4 273.0 12.1 4.7 9.7 38.0 17.8 5.8 493.0
Total 329.7 57.3 284.3 35.1 14.2 45.8 115.9 53.5 11.2 946.9
Persons
0-14 0.1 - - 0.1 0.1 0.4 0.5 - 0.0 1.1
15–24 1.0 - - 0.2 0.0 0.1 0.4 0.5 0.0 2.2
25–34 4.0 1.6 - 0.8 0.3 1.1 1.6 - 0.1 9.6
35–44 14.1 4.9 - 2.3 1.1 3.3 5.7 8.0 0.5 39.7
45–54 49.0 4.6 - 7.7 2.7 14.4 20.5 10.8 1.3 111.1
55–64 135.5 29.9 - 13.7 5.4 19.2 45.5 21.2 3.2 273.5
65–74 210.5 22.2 17.1 14.3 6.1 19.2 52.6 23.1 4.4 369.5
75+ 195.9 28.3 317.9 17.9 8.1 15.4 56.1 21.0 7.9 668.5
Total 610.1 91.5 335.0 57.0 23.7 73.1 182.9 84.6 17.4 1,475.2
Factoring up by 100/86 for the unallocated costs, total health system expenditures for
osteoarthritis in 2005 were $1.72 billion, or an estimated 0.19% of GDP. This is multiplied
by the AFs by age and gender to estimate the proportion of the health costs of osteoarthritis
due to obesity.
The total health costs arising from osteoarthritis caused by obesity were
$221.3 million in 2005.
Given there are an estimated 225,230 Australians with osteoarthritis due to obesity, this
equates to $1,582 per person per annum.
Reflecting greater prevalence in women, total osteoarthritis costs for females
($1.1 billion or 64% of the total) are higher than for males ($614 million).
Reflecting age as a risk factor for osteoarthritis, the health costs of osteoarthritis for
Australians aged 75 years or older were $573 million for females (over half of all
osteoarthritis costs for women) and $204 million for males (about one third of all
osteoarthritis costs for men).
48
Over all age groups, obesity caused osteoarthritis health expenditure of $154 million for
females and $67 million for males in 2005.
Figure 4-5 outlines the total health system expenditure for osteoarthritis and that attributable to
obesity, by age and gender.
FIGURE 4-5: OSTEOARTHRITIS, TOTAL HEALTH EXPENDITURE, 2005, BY AGE, GENDER & WEIGHT
20% 600
18%
500
16%
Attributable Fraction of Obesity (%)
14%
Total Health Expenditure ($m)

400
12%
10% 300
Column 2 - Females
8%
Column 1 - Males 200

6%
4%
100
2%
0% 0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Health cost due to obesity Remaining Osteoarthritis health cost Male AF Female AF
Figure 4-6 compares the health costs that are attributable to obesity between gender and type
of cost. The significant differences arise when comparing inpatient (47% for males versus
32% for females) and aged care components (6% male costs and 20% female costs). This
corresponds to the higher AF for obesity in younger men (shown previously in Table 3-8).
49
FIGURE 4-6: OSTEOARTHRITIS, HEALTH EXPENDITURE DUE TO OBESITY BY COST TYPE AND
GENDER, 2005 (%)
Males;
Total: $614 million
Obesity AF: $67 million
14% Inpatients
1% 14% Outpatients
5% 1%
32% Aged Care
5%
Total: $1,715 million
Out-of-Hospital
47%
12% Obesity AF: $221 million

11% Pharmaceuticals
Other Health Professionals

6%
10%
Research
10%
20%
Unallocated
6%
6%
Females;
Total: $1,101 million
Obesity AF: $154 million
4.5 CANCER
Total health system costs for cancer are based on work by AIHW (2005c), which estimated the
hospital costs (including admitted and non-admitted patients), out-of-hospital expenditure
(including GP services, imaging, pathology and out-of-hospital specialist services), high-level
residential care, pharmaceutical costs and other costs (including allied health professionals,
dental and research) associated with cancers in 2000-0116.
These costs were divided by the active prevalence of cancer in Australia in 2000-01 (since
this is the more appropriate denominator given the AIHW’s focus on active treatments in its
cost calculations for cancer) to estimate the health system cost per person, then indexed to
health inflation from 2000-01 to 2005, to estimate the expected health system cost per
person in 2005.
The AIHW estimates do not include the costs of community palliative care services (but
palliative care costs provided by hospitals and hospices are included) or respite services in
aged care homes or in hospitals. Nor do the estimates include the costs of aids and
modifications that are funded privately or supplied by community health centres (but aids and
modifications provided by hospitals are included). Estimates of these costs are included in
Section 5.5.5.6.
16
Excluding carcinoma in situ cervix uteri and other benign, in situ and unspecified neoplasms.
50
The estimates of health system costs per person in 2005 are calculated on an active
prevalence basis so that:
everyone incurs the health system costs per person once in 2005; and
a smaller proportion of patients (based on the likelihood of dying in each following year)
also incur these health costs in following years.
The total health system costs are shown in The majority of cancer costs caused by
obesity relate to females ($80.9 million for females compared to $26.4 million for males),
due to the number of obese females with the four cancers being 4.3 times higher than
the number of obese males.
Colorectal cancer ($49.9 million) and breast cancer ($34.5 million) health costs due to
obesity make up 79% of the total, while kidney cancer and uterine cancer cost
$8.5 million and $14.4 million respectively).
Table 4-4 and Figure 4-7, together with the proportion of costs attributable to obesity. In 2005,
the health system cost of Australians with the four obesity related cancers was around $791.6
million. The AFs by demographic group and cancer type were used to derive the proportion of
costs attributable to obesity.
The total health costs arising from cancer due to obesity were $107.3 million in
2005.
The majority of cancer costs caused by obesity relate to females ($80.9 million for
females compared to $26.4 million for males), due to the number of obese females with
the four cancers being 4.3 times higher than the number of obese males.
Colorectal cancer ($49.9 million) and breast cancer ($34.5 million) health costs due to
obesity make up 79% of the total, while kidney cancer and uterine cancer cost
$8.5 million and $14.4 million respectively).
TABLE 4-4: HEALTH SYSTEM COSTS FOR THE FOUR OBESITY RELATED CANCERS, 2005
Age Group Health Australians Health Obese

Costs, four with four costs Australians
cancers cancers attributable with
($m) to obesity cancer
($m)
0-14 $2.2 220 $0.1 9
15-24 $2.5 385 $0.2 24
25-34 $10.9 2,488 $1.0 237
35-44 $42.7 11,767 $5.6 1,555
45-54 $99.5 26,169 $15.8 4,232
55-64 $168.5 36,776 $29.7 6,686
65-74 $181.3 33,824 $27.2 5,219
75+ $284.0 25,595 $27.6 2,469
Total $791.6 137,225 $107.3 20,430
Note: The four cancers refer to colorectal, breast, kidney and uterine cancers
51
FIGURE 4-7: CANCER HEALTH SYSTEM COSTS, 2005, $’000, BY AGE, GENDER AND WEIGHT
14,000 $180,000
$160,000
12,000
$140,000
10,000 Column 1 - Colorectal

$120,000
Obese Cancer Population
Total Health Cost ($'000)

8,000 $100,000
Column 2 - Breast
6,000 $80,000
Column 3 - Kidney
$60,000
4,000
Column 4 - Uterine $40,000
2,000
$20,000
0 $0
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male Other Obese Male Female Other Obese Female Female Male
Note: The columns refer to the costs (right axis) and the lines to the population (left axis). “Obese male” and
“obese female” refer to the cancer health expenditure due to obesity
52
5. OTHER (NON-HEALTH) FINANCIAL COSTS

5.1 METHODOLOGY
Other financial costs are all those that are not direct health system costs or intangible costs –
the loss of health and wellbeing. They include employment impacts, absenteeism and taxation
revenue impacts, as well as the economic cost of care and of aids and home modifications.
5.1.1 PRODUCTIVITY COSTS

Obesity can have an impact on a person’s capacity to work. If employment rates are lower for
PWO and its impacts – Type 2 diabetes, CVD, osteoarthritis and cancers – this loss in
productivity represents a real cost to the economy.
Productivity costs due to obesity related conditions can be susbtantial, reducing disposable
household income and affecting overall national production possibilities. People can require
time off work or may retire early to receive treatment recover from their symptoms, and their
family carers may also take time off work to accompany them to treatments or care for them at
home, sometimes for extensive periods. Hours are diverted from unpaid – such as providing
childcare, doing the housework, yard work, shopping and so on, and these activities must be
undertaken by someone else or their value lost. If the condition is terminal then the labour
pool is reduced, decreasing the capacity of the economy to produce at any given level of
unemployment. If return to work is possible, for the person or their carer, it may be at a lower
level of productivity through reduced hours of work, decreased efficiency, changed
responsibilities and/or changed future career opportunities.
Access Economics measures the lost earnings and production due to health conditions using
a ‘human capital’ approach. The lower end of such estimates includes only the ‘friction’ period
until the worker can be replaced, which would be highly dependent on labour market
conditions and un(der)employment levels. In an economy operating at near full capacity, as
Australia is at present, a better estimate includes costs of temporary work absences plus the
discounted stream of lifetime earnings lost due to early retirement from the workforce, reduced
working hours (part-time rather than full-time) and premature mortality, if any. In this case, it is
likely that, in the absence of the health effects, PWO and its impacts would participate in the
labour force and obtain employment at the same rate as other Australians, and earn the same
AWE. The implicit and probable economic assumption is that the numbers of such people
would not be of sufficient magnitude to substantially influence the overall clearing of the labour
market.
ABS data from the ABS Average Weekly Earnings (Cat No 6302.0) series are used for AWE
calculations, based on full and part time earnings by age and gender from the August 2005
series. Overall the average for males and females was $805.40 but, where possible,
productivity losses are based on the specific age-gender AWE rates, which for early retirement
or premature mortality, translate into probable annual income streams lost (given expected
retirement age) for each particular age-gender group (Table 5-1). Other ABS data series are
also utilised in this chapter and, rather than referencing publications for numerous years, these
are referred to by series in the discussion, partly since some data are unpublished. Who
bears the costs of productivity losses is rather complex, and Figure 5-1 illustrates some
potential variations, and whether they are ‘funded’ through sick leave, Sickness Allowance, the
Disability Support Pension (DSP) or lost/reduced earnings.
53
FIGURE 5-1: PRODUCTIVITY LOSSES

Minimal Productivity Loss - Covered Entirely by Sick Leave Minimal Productivity Loss - Covered Partially by Sick Leave
Income Income
($) ($)
Lost current earnings

Sick leave Sick leave
Time Time
Diagnosis Return to work Diagnosis Return to work
Sickness allowance
Moderate Productivity Loss - Reduced Rate of Future Earnings Significant Productivity Loss - Permanent Disability
Income Income
($) ($)
Lost current earnings
Sick leave Sick leave
Lost future
earnings Lost current and
future earnings
Reduced future
earnings
Disability Support Pension
Time Time
Diagnosis Return to work Diagnosis
Sickness allowance
Significant Productivity Loss - Death
Income
($)
Sick leave
Lost current and

future earnings
Time
Diagnosis Disability Death

Support
Pension
54
TABLE 5-1: EXPECTED RETIREMENT AGE AND REMAINING LIFETIME EARNINGS (2005 DOLLARS)
Expected Retirement Expected Remaining
Age if Employed Lifetime Earnings ($m)
Age Males Females Males Females
0-4 63 60 1.03 0.57
5-9 63 60 1.12 0.62
10-14 63 60 1.20 0.67
15-19 63 60 1.30 0.72
20-24 63 60 1.36 0.75
25-29 63 60 1.34 0.70
30-34 63 60 1.22 0.62
35-39 63 60 1.10 0.54
40-44 63 60 0.92 0.46
45-49 63 60 0.74 0.36
50-54 63 61 0.53 0.24
55-59 64 62 0.32 0.13
60-64 65 64 0.15 0.05
65-69 68 68 0.04 0.01
70-74 72 72 0.01 0.00
75-79 77 77 0.00 0.00
80-84 82 82 0.00 0.00
85-89 87 87 0.00 0.00
90+ 92 92 0.00 0.00
Sources: ABS 6105.0, ABS 6310.0 (Indexed to $2005)
5.1.2 CARER COSTS

As with the productivity losses for the person with obesity, if they have a carer then the carer
costs are also measured using the implicit principle that the economy is operating at full
capacity (and therefore there is a net resource diversion from household tasks and leisure
activities), rather than a general equilibrium approach. In this context, there are several
possible methods for valuing the time foregone by caregivers including:
Opportunity cost: the value of lost wages foregone by the carer;
Replacement valuation: the cost of buying a similar amount of services from the formal
care sector; and
Self-valuation: what carers themselves feel they should be paid.
The self-valuation approach has been found elsewhere to be empirically the most conservative
(eg, for carers of people with dementia in O’Shea, 2000). The opportunity cost valuation is
implicitly more conservative than the replacement valuation approach and is used where
possible (cancers). Where there is an absence of data, however, the replacement valuation
approach is used in the relevant sections of this chapter (diabetes) or, failing any robust data,
the costs of informal care were excluded and only the formal care was costed (osteoarthritis).
Extension of the next SDAC to a larger sample size may enable robust carer estimates using
opportunity cost valuation to be calculated for all obesity related (and other) conditions. The
best estimates would require a sample of adequate size to confidently estimate the number of
carers, by gender and broad age group, and average hours of care, by the main condition(s) of
the person to whom care is provided.
Certainly the value of informal care is large by any measure. The imputed value of unpaid care
by other adult family members, friends or neighbours was estimated by the AIHW as
$19.3 billion in 2000-01, greater than the total welfare expenditure incurred by governments
55
and non-government community service organisations combined (AIHW, 2003a:76). Access

Economics (2005c) estimated that the value of informal care in 2005 in Australia was worth, in
total, between $4.9 billion (using an opportunity cost approach) up to $30.5 billion (using a
replacement valuation approach).
The estimate of the replacement value of informal community care is sensitive to changes in
the estimate of the wage parameter for alternate formal sector care. In this analysis, the unit
cost used has been based on the wage of moderately skilled formal sector carers (supervised
employees). In May 2004, full-time carers and aides employed in the formal sector received
an average wage of $17.20 per hour, or $650.30 for a 37.8 hour week (ABS, 2005b). This
average includes payment of overtime for after hours work. However, the hourly rate received
by employees does not account for on-costs such as superannuation incurred by employers,
the wages of supervisors, managers or administrative support staff or other capital overheads.
Loadings are added for each of these additional costs, and for average wage growth between
May 2004 (when the survey was last undertaken) and February 2005. The 15.6% loading of
on-costs comprises superannuation, workers compensation, payroll and Fringe Benefits
Taxation allowances. Loadings for capital (3.6%) and administrative (16.3%) overheads are
based on the relative shares of capital expenditure and administration costs to other areas of
recurrent spending in Australia’s formal health sector. When all these loadings are added, the
hourly cost of employing a carer in the formal sector to replace an informal carer is $25.01 in
2005 (Table 5-2).
TABLE 5-2: REPLACEMENT VALUATION OF INFORMAL CARE, UNIT COST COMPONENTS
% Loading Hourly rate

Base rate per hour – May 2004 $17.20
Loading for growth in AWE May 2004 to Feb 2005 4.9% $0.85
Loading for on-costs 15.6% $2.82
Loading for capital 3.6% $0.75
Loading for supervision and administration 16.3% $3.40
Total hourly rate inc. overheads $25.01
5.1.3 TRANSFERS AND DEADWEIGHT COSTS

Recalling the distinction between real and transfer payments in Section 1.2.2, the taxation
losses calculated in this chapter represent transfers. There are two sources of lost tax
revenue that result from the lower earnings; the potential income tax foregone and the
potential indirect (consumption) tax foregone. The latter is lost because, as income falls, so
does consumption of goods and services. The average personal income tax rate used is
21.20% and the average indirect taxation rate used is 15.51%, based on parameters for 2005
from the Access Economics macroeconomic model.
As discussed earlier, transfer payments (Government payments/services and taxes) are not a
net cost to society, as they represent a shift of consumption power from one group of
individuals to another in society. If the act of taxation did not create distortions and
inefficiencies in the economy, then transfers could be made without a net cost to society.
However, through these distortions, taxation does impose a DWL on the economy.
DWL is the loss of consumer and producer surplus, as a result of the imposition of a distortion
to the equilibrium (society preferred) level of output and prices. Taxes alter the price and
quantity of goods sold compared to what they would be if the market were not distorted, and
thus lead to some diminution in the value of trade between buyers and sellers that would
56
otherwise be enjoyed. The principal mechanism by which a DWL occurs is the price induced
reduction in output, removing potential trades that would benefit both buyers and sellers. In a
practical sense, this distortion reveals itself as a loss of efficiency in the economy, which
means that raising $100 dollars of revenue requires consumers and producers to give up more
than $100 of value (Figure 5-2).
FIGURE 5-2: DWL OF TAXATION
Price ($)
Supply
Deadweight Loss (cost
of raising taxation
revenue)
Price plus Tax
Taxation Revenue
Price
Demand
Output
Actual Quantity Potential Quantity
Supplied Supplied
The rate of DWL used in this report is 27.5 cents per $1 of tax revenue raised plus 1.25 cents
per $1 of tax revenue raised for Australian Taxation Office administration, based on
Productivity Commission (2003) in turn derived from Lattimore (1997), ie. 28.75% overall. The
total extra tax dollars required to be collected include:
the taxation revenue lost as a result of obesity and its impacts;
the additional induced social welfare payments required to be paid; and
the value of government services provided (including the Government-funded
component of health system costs).
5.1.4 OTHER NON-HEALTH FINANCIAL COSTS

A broad range of other costs may be incurred due to the health impacts of obesity, including
various Government program expenditures as well as out-of-pocket expenses for the PWO
and their family on items for personal care, home help, aids and home modifications and travel
costs, for example.
Aids and home modifications can be a relatively substantial component of this item, but assist
with enabling the PWO to continue living at home safely. These items may relate to
communication; bathing, toileting and continence; leisure and recreation; mobility, seating,
lifting/transfers, transport (eg, ramps, hoists); nursing (eg, pressure-relief mattresses); and
safety (eg, grab rails, lighting).
57
Assistance is available to cover some of the costs of these items. For example, some of the
bigger-ticket items can be borrowed from a community organisation; the Program of
Appliances for Disabled People scheme can assist with access to aids and equipment, and the
Community Options Program can assist with the cost of home modifications.
Travel and accommodation costs are particularly burdensome for regional and remote patients
travelling to metropolitan areas for treatment. Sometimes the patient may choose to commute
on a daily basis, while other times the patient may choose to stay for part or all of their
duration of the treatment (perhaps travelling home on weekends or more irregularly).
However, even if the medical treatment is available locally, travel costs can still be substantial
in terms of both distance and time.
5.2 DIABETES
5.2.1 PRODUCTIVITY LOSSES
In previous studies, a significant productivity loss associated with diabetes was found. For
2002, the American Diabetes Association (2003) estimated the indirect costs as US$40 billion
(US$3,289 per person per annum), which included lost workdays at a daily rate of US$168,
restricted activity days at a daily rate of US$67, mortality at US$116,928 per annum and
permanent disability due to diabetes at a rate of US$42,462 per annum. Tunceli et al (2005)
found that the probability of working was 4.4% less for women and 7.1% less for men relative
to people without diabetes. In contrast, weekly hours worked was not statistically significantly
associated with diabetes, but women with diabetes had two more work-loss days per year.
Robinson et al (1989) looked at the link between diabetes and unemployment and found that
22% of males and 12% of females with diabetes were unemployed compared with 8% of
males and 5% of females without diabetes. They also found that a greater percentage of
people with diabetes were economically inactive (retired, unable to work, ill or “housewives”)
compared with the control group who participated more in the workforce. Von Korff et al
(2005) added that complications associated with diabetes and overall work disability status
increased unemployment rates of people with diabetes further. Among subjects experiencing
both major depression and three or more diabetes complications, more than 50% were
unemployed (compared to an average 12% of people with diabetes). A study looking at
diabetes in Queensland (Queensland Health, 2000) also substantiated the hypothesis that
diabetes is linked to lower employment rates with only 44% of people with diabetes being
employed full time compared to 56% of the general population.
Colagiuri et al (2003) estimated that income lost from days in which people (aged 40 years or
older) were unable to work averaged $35 per person per year. Lost income was higher for
people with complications, particularly macrovascular complications. However, the study
included many elderly, so consequently relatively few were employed.
5.2.1.1 EMPLOYMENT IMPACTS
Table 5-3 reports on employment outcomes for people with Type 2 diabetes with data drawn
from the ABS (2003b) SDAC data. Of people with diabetes aged 15–64 years, 52.1%
reported being employed compared with 73.7% of people in the total population. While the
unemployment rate for people with diabetes was only 3.2%, the participation rate was a low
53.8%. While employment opportunities may be affected by gender, age and by increasing
levels of disability, only overall numbers were available for people with diabetes. The lack of
58
data on age may make the productivity cost estimate more generous, while the lack of data on
gender may make it more conservative.
TABLE 5-3: EMPLOYMENT OUTCOMES FOR PEOPLE WITH TYPE 2 DIABETES, 15-64 YEARS
Labour force status % Population with Type 2 diabetes
Employed working full-time 36.7 346,204
Employed working part-time 15.4 145,273
Total employment rate 52.1 491,477
Participation rate 53.8 507,514
Unemployment rate 3.2 30,187
Total 100.0 943,334
Source: ABS (2003b)
The proportional difference in employment for people with diabetes at each age is thus
estimated as (73.7-52.1)/73.7= 29.3%, applied to the number of people with diabetes by age
and gender. The cost of this lower employment was estimated using age- and gender-specific
AWE estimates from ABS data. The annual cost of lost earnings due to workplace separation
and early retirement from Type 2 diabetes is thus estimated as $3.96 billion in 2005. This
equates to 0.44% of GDP. This is multiplied by the AF of 10.8% to estimate the proportion of
the lower employment impacts due to obesity.
The annual cost of lost earnings for people with Type 2 diabetes as a result of
obesity was $429.1 million in 2005 (0.14% of GDP).
5.2.1.2 ABSENTEEISM
In addition to workforce separation, people with Type 2 diabetes may be absent from work
more often as a result of their condition. Our literature search was unable to locate published
works on such absenteeism rates for people with Type 2 diabetes in Australia. Consequently,
the average work-loss days per year were estimated using the results from Tunceli et al’s
(2005) United States study that weekly hours worked was not statistically significantly
associated with diabetes, while women with diabetes had two more work-loss days per year.
Based on a share of 57% of males and 43% of females in the 15-64 age group for people with
Type 2 diabetes, this gives an average of 0.86 work-loss days per year.
For all those who have Type 2 diabetes and are employed ie, 943,389*52.1% (recall Section
5.2.1.1) = 491,477 people, the absenteeism cost was estimated using age- and gender-
specific AWE estimates. In 2005, the annual absenteeism cost due to Type 2 diabetes was
$53.1 million. Of this, $8.2 million is incurred by employees and $44.9 million is incurred by
employers (based on the proportion of time taken as sick leave17 from ABS data). On average,
the absenteeism cost per employed person with Type 2 diabetes is $108 per annum. This is
multiplied by the AF of 10.8% to estimate the proportion of the lower employment impacts due
to obesity.
17
The proportion of paid sick leave for time off work was estimated as 75.9% for males and 69.7% for females
(ABS, 2000).
59
For people with Type 2 diabetes as a result of obesity the annual costs of
absenteeism are $5.8 million ($0.9 million incurred by employees and $4.9 million
incurred by employers) in 2005.
5.2.1.3 PREMATURE DEATH
The production losses arising from premature mortality associated with Type 2 diabetes were
estimated based on mortality statistics published by the ABS (2006b). In 2005, there were an
estimated 3,092 deaths (1,570 males and 1,522 females) due to Type 2 diabetes, based on
the 2004 deaths data from Section 3.1, population growth and the relative prevalence of Type
2 compared to total diabetes in males and females. Using RR of mortality estimates of 1.124
for males and 1.117 for females (derived from the mortality estimates relative to population
mortality rates), it was estimated that 249 of those deaths (180 males, 69 females) occurred in
the ‘under 65’ age group. The present value of the lost earnings of the share of people who
died but would otherwise have been employed was estimated using the employment rates
discussed in the Section 4.2.1.2 and estimates of average lifetime earnings for the different
age groups. In total, the productivity loss from premature mortality from Type 2 diabetes was
estimated as $64.7 million in 2005.
For people aged 15-64 with Type 2 diabetes as a result of obesity the annual cost
due to premature death was $7.0 million in 2005.
Premature death also leads to additional search and hiring costs for replacement workers. In
2005, these costs are estimated as $159,405 for people with Type 2 diabetes, based on the
present value of bringing forward by three years the average cost of staff turnover (26 weeks
at AWE). The estimate is conservative because search and hiring costs may also apply to
premature retirement due to diabetes, but this is conservatively excluded as there may be
greater scope to reduce handover costs when the exit is more planned.
Additional search and hiring costs for people with Type 2 diabetes as a result of
obesity were estimated as $17,267 in 2005.
5.2.1.4 SUMMARY OF PRODUCTIVITY COSTS
Total productivity costs in 2005 due to diabetes ($4,078.8 million) were thus estimated as
the sum of the losses due to lower employment rates ($3,960.8 million), absenteeism
($53.1 million) and premature death, including the search and hiring bring-forward
($64.8 million). Of the total, $2,552.9 million (62.6%) was borne by the people with diabetes,
$28.5 million was borne be employers (0.7%) and $1,497.3 million (36.7%) was borne by
government, in the form of lost taxation revenues.
In relation to the productivity costs for people with diabetes due to obesity, the total
costs in 2005 were $441.9 million, the sum of the losses due to lower employment rates
($429.1 million), absenteeism ($5.8 million) and premature death, including the search and
hiring bring-forward ($7.0 million). Of the total, $276.6 million (62.6%) was borne by the
people with diabetes due to their obesity, $3.1 million was borne be employers (0.7%) and
$162.2 million (36.7%) was borne by government.
60
5.2.2 CARER COSTS

Informal carer costs associated with diabetes have been analysed in several studies. Langa
et al (2001) estimated that people with diabetes taking no medications received 10.5 hours per
week of informal care, while those with diabetes taking oral medications received 10.1 hours
and those with diabetes taking insulin received 14.4 hours (in comparison, people without
diabetes received an average of 6.11 hours per week). Disabilities related to heart disease,
stroke and visual impairment were found to be particularly important predictors of diabetes-
related care. The total costs of informal caregiving for elderly people in the US was estimated
as US$3 billion to US$6 billion per year.
In Australia, Colagiuri et al (2003) found that 10% of people with Type 2 diabetes (aged 40
years and older) had a carer an average of 36 hours per week due to their diabetes. Most of
the carers were informal carers (ie care was provided by family and friends of a person with
Type 2 diabetes at no monetary cost), though 11.7% received Carer Payment and 14.2%
received Carer Allowance (based on Centrelink payments and allowances). Carers did not
include services provided by local authorities, but did include people paid by the person or
family to provide the caring; 1% of carers nominated themselves as ‘professional paid carers’.
On average, people with Type 2 diabetes received 3.6 hours per week of care.
While the dollar value of some carer cost can be easily measured through Carer Payment,
Carer Allowance and wages for the professional paid carers, there is no monetary cost for
other informal care. However, informal care still has an economic cost, as the caregiver
cannot spend that time doing other activities, including paid work or leisure activities.
Presently, there are no data available on the costs of caring for people with Type 2 diabetes in
Australia from the ABS SDAC due to the very small number of responses and consequent
very high standard errors. Hence, Colagiuri et al’s (2003) estimate of 3.6 hours per week was
used as the best estimate of the average rate of informal care required due to Type 2 diabetes
in Australia. This parameter (3.6) is consistent with Langa et al’s (2001) finding of a minimum
of 4 hours per week difference between the diabetes group and the control. For the 943,334
Australians with Type 2 diabetes for 52 weeks per year, this sums to 176.6 million care hours
per year.
Access Economics has adopted the replacement valuation approach in this section, due to the
lack of information about the demographic characteristics of carers of Australians with Type 2
diabetes. Based on the estimated replacement wage rate of $25.01 derived above
(Section 5.1), the total replacement value of family and other informal care provided to
Australians for Type 2 diabetes is estimated as $4.42 billion in 2005. Of this, the
Government paid the Carer Payment ($215.65 per week) and Carer Allowance ($63.98 per
week) for those who received it, and is assumed also to pay for the 1% of paid carers. In total
the Government share was thus $212.5 million (4.8%) while family and friends bore
$4.20 billion.
61
For Australians with Type 2 diabetes as a result of obesity (i.e. 10.8% of Type 2
diabetes sufferers), the replacement value of informal care is $478.5 million of
which family and friends bore $455.5 million and the Government $23.0 million in
2005.
5.2.3 TRANSFER COSTS AND DWL

While people with Type 2 diabetes may receive employment support benefits and other
welfare payments, no data were available in the timeframes to estimate the value of those
welfare payments and the DWL associated with them (ie the DWL caused by the taxation
needed to finance those payments).
Access Economics estimates that for Type 2 diabetes, $700.5 million in DWL was incurred in
2005, due to the additional taxation required to replace that foregone due to the health system
costs borne by Government ($726.6 million), the lost productivity of people with diabetes
($1.5 billion), welfare payments (only the Carer Payment and Carer Allowance have been
included, to total $168.3 million) and other costs borne by Government (only the 1% of paid
carers have been included to total $44.1 million). Summing these and multiplying by 28.75%
(recall Section 5.1.3) provides the estimate of the DWL (Table 5-4).
TABLE 5-4: LOST EARNINGS AND TAXATION REVENUE DUE TO TYPE 2 DIABETES, $ MILLION, 2005
Health System Costs Borne by Government 726.6
Lost Taxes 1,497.3
Welfare Payments 168.3
Other Costs Borne by Government 44.2
DWL (28.75% of the total) 700.5
Source: Access Economics
For Type 2 diabetes as a result of obesity, the additional DWL was $75.9 million in
2005.
5.2.4 COSTS OF AIDS AND HOME MODIFICATIONS

While people who have Type 2 diabetes and no complications do not require any aids or home
modifications, people who have a disability due to diabetes (ie excluding people who have a
disability and happen to have diabetes) require a variety of devices, special equipment and
home modifications to function adequately and to enhance their quality of life. As noted in
Section 3.1, one third of people with diabetes experience complications such as eye problems,
kidney damage, foot or leg ulcers, heart attack and amputation. However, from SDAC data,
only 14% (132,765) of people with Type 2 diabetes have disabling complications due to the
diabetes (ie greater than population average rate of disabilites). Of those people, 60%
(79,927) use aids and equipment such as self care aids (eating aids, showering/bathing aids,
dressing aids, toileting aids and continence aids), mobility aids (canes, crutches, walking sticks
and frames, wheelchairs or scooters and car aids), communication aids (hearing and other
communication aids) and home modifications (provision of ramps and handrails and
modification to bathrooms and toilets).
Cost estimates for various products are based on prices provided by the Independent Living
Centre NSW, the Victorian Aids and Equipment Program and previous studies undertaken by
Access Economics. While some equipment and modifications require large outlays but are
amortised over a number of years, other devices need to be replaced more regularly. It was
62
assumed that devices in heavy use (eating, dressing and continence aids and batteries) need
to be replaced on an annual basis, while most other devices with a cost range of between $30
and $200 (showering and toileting aids and most mobility aids such as canes, crutches,
walking sticks and frames) have a lifespan of three years, while larger expenses such as
wheelchairs ($5,000) and hearing aids ($2,500) were depreciated over five years. Home
modifications ($7,500) tend to be one-off investments, so their lifespan was assumed to be 20
years (Table 5-5).
Overall, the cost for aids and equipment for people with Type 2 diabetes was an
estimated $52.8 million in 2005. As it is not known how much of this cost is subsidised by
governments, paid for by the person with diabetes or their family and friends, or paid for
through community programs, the amount is allocated to the individual with diabetes.
The annual cost for aids and equipment for people disabled by Type 2 diabetes as
a result of obesity was $5.7 million in 2005.
63
TABLE 5-5: DIABETES, AIDS AND EQUIPMENT PRICES, ESTIMATED PRODUCT LIFE AND TOTAL COSTS
Device Base Product Unit cost Number of Total cost
price life ($ per devices ($ per
($) (years) annum) used annum)
1
Self care Eating aids $100 1 $100 2,270 $226,980
Showering or $85 3 $28 24,596 $696,880
2
bathing aids
1
Dressing aids $20 1 $20 5,544 $110,886
2
Toileting aids $80 3 $27 12,837 $342,331
Managing $1,200 1 $1,200 8,372 $10,046,677
1
incontinence
6
Total Self care $401 28,466 $11,423,754
2
Mobility aids Canes $30 3 $10 3,907 $39,070
2
Walking stick $30 3 $10 16,782 $167,817
Crutches $50 3 $17 1,637 $27,287
1
Walking frame $300 3 $100 10,865 $1,086,529
Wheelchair or $5,000 5 $1,000 7,814 $7,814,082
1
scooter
Specially modified $200 3 $67 558 $37,210
2
car or car aid
4
Other mobility aids --- 3 $201 5,805 $1,164,174
6
Total Mobility aids $327 31,628 $10,336,170
Communication Communication aids $2,500 5 $500 35,759 $17,879,364
aids (electronic, non-
electronic and other
hearing and
3
communication aids)
Batteries
3
$137 1 $137 35,759 $4,898,946
Total Communication aids $637 35,759 $22,778,309

Home Home modifications $7,500 20 $375 22,028 $8,260,601
modifications (incl structural
changes, ramps,
bath modifications,
doors widened,
5
handrails, etc)
Total Home modifications $375 22,028 $8,260,601
People using Aids and Equipment $661 79,927 $52,798,834
People not using Aids and Equipment 52,838

People with disability 132,765
Sources: ABS (2003b), 1 Victorian Aids and Equipment Program, 2 Independent Living Centre NSW, 3 Access
Economics (2006a) 4 average of mobility aids, 5 Access Economics (2006b) 6 average, people may use multiple
devices
5.2.5 FUNERAL COSTS

The ‘additional’ cost of funerals borne by family and friends of cancer patients is based on the
likelihood of death in the five years due to cancer. However, some patients (particularly older
patients) would have died during this time anyway, and eventually everyone must die, and
64
thus incur funeral expenses – so the true cost is the cost brought forward (adjusted for the
likelihood of dying anyway). The BTRE (2000) calculated a weighted average cost of a funeral
across all States and Territories, to estimate an Australian total average cost of $3,200 per
person for 1996, or $3,949 per person in 2005. Funeral costs associated with premature
death due to Type 2 diabetes are estimated as $12.2 million in 2005 - $6.2 million for males
and $6.0 million for females. These costs are paid for by family or friends.
For individuals with Type 2, funeral costs due to premature death were $1.3 million
in 2005).
5.2.6 SUMMARY OF NON-HEALTH FINANCIAL COSTS

The total real financial costs (other than health costs) of Type 2 diabetes are estimated
as $9.3 billion in 2005, summarised in Table 5-6 and Figure 5-3.
The cost of carers to individuals with Type 2 diabetes is the greatest cost, accounting for
almost half (47.7%) of other financial costs ($4.4 billion).
Lost earnings to individuals with Type 2 diabetes is second at 44.0% of the total
($4.1 billion).
The deadweight costs from losing taxation revenue and having to find alternative
sources of taxation to fund increased welfare and health services, cost over $700 million
in 2005 (7.6% of total costs).
Aids and home modifications account for 0.6% ($53 million), while indirect costs
comprise the remaining 0.1% ($12 million).
Annual costs per person with Type 2 diabetes are $9,817, $455 for every Australian and
1.04% of GDP each year.
TABLE 5-6: TYPE 2 DIABETES, NON-HEALTH FINANCIAL COST SUMMARY, 2005, $M

Cost element Real cost, Real cost,
Type 2 diabetes Type 2 diabetes as
a result of obesity
Employment impact $3,960.8 $429.1
Absenteeism $53.1 $5.8
Premature death (including additional Search $64.8 $7.0
and Hiring Costs)
Subtotal, productivity costs $4,078.8 $441.9
Carer costs (replacement valuation) $4,416.8 $478.5
DWLs from transfers $700.5 $75.9
Cost of aids, modifications and funerals $65.0 $7.0
Total, non-health financial costs $9,261.1 $1.003.4
Per person with disease $9,817 $9,817
Per capita (population) $455 $49
% of GDP 1.04% 0.11%
65
FIGURE 5-3: TYPE 2 DIABETES, NON-HEALTH FINANCIAL COST SUMMARY, 2005 (% TOTAL)
0.6% 0.1%
7.6%
Productivity
Carers
DWLs
44.0%
Aids and Equipment
Funerals
47.7%
Source: Table 5-6
For individuals with Type 2 diabetes as a result of obesity, real financial costs
(other than health costs) are estimated as just over $1.0 billion (measuring informal
care using replacement valuation). Annual costs per person with Type 2 diabetes
as a result of obesity are $9,817, $49 for every Australian and 0.11% of GDP in
total.
5.3 CVD
Indirect financial costs for CVDs are primarily productivity losses from reduced workforce
participation and carer costs (Access Economics 2005a).
5.3.1 LOSS OF PRODUCTIVITY

CVD has a negative impact on productivity in terms of lower rates of employment than healthy
people of the same age, reduced activity and elevated absenteeism of people with CVD
relative to Australian averages, and lost human capital from premature death.
5.3.1.1 LOWER EMPLOYMENT RATES
From ABS NHS data, the estimated age and gender standardised difference in employment
between PWO related CVD and those in the general population was 3.5% in 2005. If PWO
related CVD aged between 15 and 65 years achieved the same employment rate as in the
general population, there would be an extra 48,042 people in the workforce in 2005, with AWE
of $805.40 per week (recall Section 5.1.1), all other things being unchanged. This would
have generated an estimated $2.02 billion in extra production and income in the
Australian economy over the course of 2005. This figure may be a conservative estimate
of the true loss as many people may reduce their workload rather than stop work completely,
as a result of the health impacts of CVD.
66
5.3.1.2 ABSENTEEISM
In the two weeks prior to the 2001 NHS, 15.7% of people with all CVDs had days of reduced
activity compared with 10.8% for the average Australian. The days of reduced activity for
people with hypertension were 1.3 times the average, with 2.3 times the average for CHD.
There were an estimated 420,551 people employed aged 15-65 years with obesity related
CVD in 2005, earning an estimated $17.6 billion. The difference in absenteeism rates (11.6%
for people with CVD compared to 11.2% on average) and length of absence (3.9 days for
people with CVD compared to 3.1 days on average) together imply that people with CVD lose
0.35% more time off work than the average Australian. The cost of absenteeism would
thus be a further $60.9 million.
This may be a conservative estimate of the cost of absenteeism as average absenteeism rates
and length of absence for all CVDs is used as an estimate for obesity related CVDs (due to
lack of data for stroke-related absenteeism). AIHW (2004) observes that:
On average, those hospitalised for stroke tended to stay twice as long as those
hospitalised for CHD; and
Those disabled by stroke were twice as likely to need ongoing assistance with [activities
of daily living] as those disabled by CHD.
These statements suggest that the severity and absenteeism impact of strokes may be greater
than the average for CVDs, which presumably would have an upward impact on the cost of
absenteeism of obesity related CVDs.
In total then, the loss of earnings from loss of employment and absenteeism is
estimated for 2005 as $2.08 billion for obesity related CVDs.
In addition to the income foregone due to those with obesity related CVD in the community
who are unable to work due to illness, there is also the income foregone of those who have
died prematurely due to adverse CVD events. Deaths from CVD, made up primarily of deaths
from CHD and stroke, in 2002 are shown in Table 5-7, from which a cost estimate of the
mortality burden can be derived.
TABLE 5-7: DEATHS FROM CVD, AUSTRALIA, BY AGE AND GENDER, 2002
Age group (yrs) Males Females Persons

1-14 17 13 30
1
15-24 64 28 92
25-34 100 54 154
35-44 409 156 565
45-54 1,107 339 1,446
55-64 2,212 794 3,006
65-74 4,662 2,449 7,111
75-84 8,643 8,446 17,089
85+ 6,773 14,026 20,799
Total deaths 23,987 26,305 50,292
Source: ABS Causes of Death Cat No 3303.0 data series
1 The 15-24 year age group also contains a few deaths from the under 1 year category, due to
limitations of the published data.
67
Deaths between 2002 and 2005 were factored up by population growth only (3.4%), to
balance the impacts of demographic ageing and expected declining mortality rates at each
age-group. Assuming that if those who died from CVD prior to retirement age in 2005 (an
estimated 5,474 people under 65 years) instead lived and were well and employed at the
same rate as the age-standardised general population (62.2%), then an estimated 3,407
people under 65 years would be employed rather than dying from obesity related CVD in
2005. The average age of death for those people under 65 years is estimated from the ABS
mortality data as 54.3 years (ie with 10.7 years to nominal retirement, see Table 5-8) and the
income stream is discounted at 1.55% per annum (recall Section 1.2.4) giving a net expected
average earnings till retirement of $417,120 per person.
This yields the Net Present Value (NPV) of the mortality burden as $1.42 billion in
2005.
TABLE 5-8: AVERAGE AGE AT DEATH AND AGE-STANDARDISED EMPLOYMENT RATES

2 3 4
Age Midpoint Age- Mortality weights Employment rates
1
group weight
1-14 8.0 0.05 male female total male female total
15-24 21.0 0.37 1.6% 2.0% 1.7% 62.9% 62.4% 62.7%
25-34 31.0 0.90 2.6% 3.9% 2.9% 86.7% 66.5% 76.5%
35-44 41.0 4.38 10.5% 11.4% 10.7% 87.7% 68.1% 77.8%
45-54 51.0 13.93 28.4% 24.7% 27.5% 84.4% 70.3% 77.3%
55-64 61.0 34.64 56.8% 57.9% 57.1% 62.1% 40.3% 51.3%
15-64 54.3 100.0% 100.0% 100.0% 77.6% 62.8% 70.2%
5
Age-standardised employment rate for mortality burden) 71.8% 52.4% 62.2%
1 The 15-24 year age group also contains a few deaths from the under 1 year category, due to limitations of
the published data from ABS (2003).
2 Midpoints are slightly above the strict middle of the ranges reflecting the distribution of deaths. If midpoints
were at the middle of the ranges, the average age at death would be 53.3 years; if highest (ie 23, 33, 43 etc)
the average age at death would be 56.3 years.
3 Mortality weights are the proportion of deaths at each age relative to total deaths for those aged 15-64.
4 Employment rates were derived from AusStats data for April 2004.
5 Age-standardised rates multiply the mortality weights by the published employment rates to derive
expected employment rates for the population who died under the aged of 65 from CVD.
Note that CVD mortality is a key driver of life expectancy at these ages, so these estimates are
again conservative.
5.3.2 CARER COST

Most people with CVD receive care at home at least initially, although some may be
transferred to residential care depending on the disability caused by the illness, other
comorbidities and the availability of carers. Post-operative care at home is especially
important. Society, and our public sector health and welfare budget, relies increasingly on the
support that families and carers provide.
However, there is a paucity of reliable data in Australia providing good quantitative estimates
of the average care hours required by people with various CVDs or relating the care to
disability levels. This might be an area of future investigation. As a consequence, carer
estimates here are based on UK data, utilising the ratio of care to productivity costs from
employment and absenteeism impacts (Petersen et al, 2003), which was nearly 1.1:1
(productivity losses from morbidity of people with CVD were £2,207.5m while informal carer
costs were £2,416.5m; productivity losses from premature mortality were not included). The
68
ratio is then applied to the Australian estimate of employment losses (morbidity burden) of
$2.08 billion from Section 5.3.1.2.
As such, the value of care for people with CVD was estimated as $2.27 billion in
2005.
While Australian Governments contribute to community care through various programs, the
lion’s share of community care is borne by informal carers themselves (Access Economics,
2005c). Partial compensation for the burden is offered through federal Carer Payment and
Carer Allowance and the DWLs from these welfare payments are discussed in Section 5.2.3
below.
5.3.3 TRANSFER COSTS AND DWL

5.3.3.1 LOSS OF TAX REVENUE
Table 5-9 shows income tax lost, calculated by multiplying lost earnings by the average
personal income tax rate. Indirect tax revenue lost is calculated based on the consumption
lost (estimated based on the difference between 90% of average earnings – the savings rate
may well be lower than 10% – and income from the disability pension). The same methods
are used to derive the NPV of taxation revenue lost due to premature death, and in relation to
the carers of PWO-related CVD. Tax revenue sacrificed is again included as a transfer
payment (not a real economic cost).
TABLE 5-9: POTENTIAL EARNINGS AND TAX REVENUE LOST DUE TO OBESITY RELATED CVD, 2005
Potential Earnings Lost $2,076.8 million
Average personal income tax rate# 21.20%
Potential personal income tax lost $440.3 million
Average indirect tax rate# 15.51%
Potential indirect tax lost $167.2 million
Total potential tax revenue lost $607.5 million
# Source: Access Economics macroeconomic model
Tax revenue foregone on potential earnings lost were estimated as $608 million in
2005 for obesity related CVDs, comprising $440 million of personal income tax and
$167 million of indirect tax.
The NPV of taxation revenue sacrificed due to premature death was $416 million.
And, the opportunity cost of work for informal carers that could take place in the
marketplace rather than in unpaid work, results in $665 million of tax foregone.
5.3.3.2 WELFARE TRANSFERS
Some people living with obesity related CVD receive welfare benefits. In most cases, this is
the means-tested age pension, paid to eligible men over 65 and eligible women aged over 60-
65, depending on their birth date (by 2014 the age will be 65 for everyone). Since the age
pension would be paid to eligible elderly regardless of CVD, it is not included in modelling
here.
69
People under retirement age with obesity related CVD, especially those suffering the effects of
stroke, may be eligible for the DSP and in some cases, Sickness Allowance and Mobility
Allowance. There are also entitlements to concession cards – Pensioner Concession Card
and Health Care Card, which may result in concessional transfers such as prescription
medicines, transport fares, rates, power bills and car registration – and to Rent Assistance, for
people who get a payment such as the Carer Payment and pay rent for private
accommodation.
Although insufficient data preclude a firm estimate of many of these transfer payments, an
estimate of welfare payments for some of the main items, based on Centrelink parameters
outlined in more detail in Access Economics (2005a), is provided in Table 5-10 below, totalling
$407.1 million.
TABLE 5-10: COST OF WELFARE PAYMENTS

weekly receiving total cost
payment benefit calculation $m
DSP $215.93 8,353 0.61% of people with CVD aged 16-65 94.0
Carer payment $215.93 14,071 0.5% of people with CVD 158.3
Carer allowance $45.05 56,283 2% of people with CVD 132.1
Pharmaceutical
allowance $2.90 8,353 0.61% of people with CVD aged 16-65 1.3
Rent assistance $51.41 2,506 0.18% of people with CVD aged 16-65 6.7
Mobility allowance $34.00 8,353 0.61% of people with CVD aged 16-65 14.8
Total 407.1
Source: Access Economics estimates utilising Centrelink rates of November 2004
5.3.3.3 DWL
The real DWLs associated with taxation and welfare transfers were 28.75% of the value of the
transfers and of Government financed payments requiring Federal taxation funding (health
system expenditures only in this case). The health system costs borne by Government were
calculated as (67.8%*$3,902.5=) $2,645.9 million; the lost tax of people with CVD and their
carers was estimated as ($608+$416+665=) $1,688.2 million; and the welfare transfers from
the section above were $407.1 million. Summing these and multiplying by 28.75% provides
an estimate of the DWL of $1.36 billion in 2005 (Table 5-11).
TABLE 5-11: CVD, SUMMARY OF TRANSFERS AND DWL, 2005
$ million
Health System Costs Borne by Government 2,645.9
Lost Taxes 1,688.2
Other Costs Borne by Government 0
DWL 1,363.1
Total real DWLs are estimated as $1.36 billion in 2005, for obesity related CVDs.
5.3.4 OTHER FINANCIAL COSTS

Access Economics (2005a) estimated the cost of aids and modifications for people with CVD
based on survey work by Frisch (2001:18), with an average unit cost imputed for ‘aids and
appliances’ of $174.20 per person, while including home modifications and consumables
increased this average to $738.40 per person. Some 20% of people with CVD were estimated
70
to have disability levels consistent with those in the survey. If 20% of the 2.8 million PWO
related CVD in 2005 have such disability and spend these average amounts on aids and home
modifications, then the estimate is the product of 562,833 and $738.40. The total cost in 2005
of aids and modifications would be estimated as $416 million, and sensitivity analysis would
suggest that each 10% equates to around $200 million. However, since there are no data to
provide robust evidence for the 20% parameter, the entirety of this cost is conservatively
treated as being included in the estimate for unallocated health costs ($546.3 million).
5.3.5 NON-HEALTH FINANCIAL COSTS ATTRIBUTABLE TO OBESITY

Table 5-12 shows that obesity related CVD non-health costs were $7,134 million in total in
2005, to which the obesity AF of 13.5% is applied.
TABLE 5-12: CVD NON-HEALTH FINANCIAL COSTS DUE TO OBESITY, 2005, $M
Indirect Cost Item Total ($m) Obesity AF ($m)*

Lost earnings from workplace separation $2,016 $272
Lost earnings from absenteeism $61 $8
Lost earnings from premature death $1,421 $192
Carers $2,273 $306
DWL $1,363 $184
Total Indirect Costs $7,134 $962
*Calculated by applying an overall obesity AF of 13.5% (weighted across all ages and genders)
For individuals with obesity related CVD as a result of obesity, real financial costs
(other than health costs) are estimated as $962 million.
5.4 OSTEOARTHRITIS
5.4.1 LOSS OF PRODUCTIVITY
The productivity losses for osteoarthritis include only the employment impacts and the
absenteeism impacts. Productivity losses due to premature death are (conservatively)
excluded but, in any event, are likely to be relatively low due to the age and gender distribution
of people with osteoarthritis. However, the cost estimate for employment impacts was
generous since due to data constraints it could not categorically be ruled out that some portion
of the lower employment rate of people with osteoarthritis might be due to the influence of
other socioeconomic factors or the impact of other comorbidities. Hence the two influences
will tend to counter each other in deriving the overall estimate of productivity losses due to
osteoarthritis.
Based on Access Economics (2005b) and data from the 2001 NHS, it is estimated that the
employment rate for people with osteoarthritis is 5.9% lower than the age-standardised rate for
all Australians implying that, if people with osteoarthritis were employed at the same rate as
average Australians of the same age, then in 2005 an estimated additional 94,450 people
would be employed. With AWE of $805.40, the annual cost of lost earnings from workplace
separation due to osteoarthritis was $3.96 billion.
71
5.4.1.2 ABSENTEEISM
As well as those who do not work at all, some people with osteoarthritis will still be employed
but may reduce the number of hours worked or take a greater number of days off due to their
illness. The NHS also asked respondents whether they had taken days off from work either
for their own illness or to care for another. In the 1995 survey, 8,201 people with arthritis
reported that they had taken time off from work or school in the fortnight prior to the survey
due to illness, for an average of 2.9 days. This is equivalent to 0.31% of all people with
arthritis at the time. Assuming a similar proportion of osteoarthritis sufferers were absent from
work in 2005, there would have been the equivalent of 75,022 weeks of work lost through
osteoarthritis-connected absenteeism. Multiplying by AWE, the lost earnings from
absenteeism among people with osteoarthritis in 2005 was $60.4 million.
5.4.2 DWL FROM TRANSFERS

Potential tax revenue foregone associated with these sources of lost production was
$1,179 million (Table 5-13). Of the total tax foregone, $853 million (72%) relates to personal
income tax and $326 million (28%) to indirect taxation. Lost taxation revenue is estimated as
a transfer payment, not a real economic cost.
TABLE 5-13: POTENTIAL EARNINGS & TAX REVENUE LOST, OBESE AUSTRALIANS WITH
OSTEOARTHRITIS, 2005
$ million
Potential earnings lost 4,024
Potential personal income tax lost 853
Potential indirect tax lost 326
Total potential tax revenue lost 1,179
DWL from additional taxation 339
Note: Based on average personal income tax rate of 21.2% and indirect tax rate of 15.51%
Table 5-13 also shows the estimated real DWL arising from this lost revenue, noting:
no data were available on welfare payments to people with osteoarthritis due to their
condition so these potential associated DWLs have, conservatively, been excluded: and
total real DWLs from taxation revenue raising were estimated as $339 million in 2005.
5.4.3 COST OF CARERS, TRAVEL, AIDS & MODIFICATIONS

These costs are combined for osteoarthritis since they are based on common data. Walsh
and Chappell (1999) conducted a survey on behalf of the Department of Family and
Community Services of 409 recipients of DSP who had a musculoskeletal impairment. The
study estimated the additional expenditure of these people on personal care, home help, and
other aids and appliances. Based on these data, Access Economics estimated (Table 5-14)
that, for Australian osteoarthritis sufferers in 2005:
the cost of paid carers was $110.7 million;
the cost of aids and modifications was at least $23.4 million; and
the cost of travel associated with their condition was $43.3 million.
The gross cost of aids and modifications, $54.4 million, includes out-of-pocket expenditure on
consumables, aids and appliances (including wheelchairs, communication aids, special
72
clothing and furniture). In the calculation of net costs, we exclude cost items 6 and 7 to
(conservatively) avoid double counting of health costs.
TABLE 5-14: COST OF CARERS, AIDS & OTHER FINANCIAL COSTS, 2004
Cost Item 2005 Total cost

$ pa $m
1 Care, inc. personal care, bathing, travel assistance 1,094.5 61.9
2 Home tasks, inc. house cleaning, gardening, house 862.3 48.8
maintenance
3 Travel, inc. MV modifications, taxis, community transport, 765.9 43.3
personal travel expenses
4 Uncapped prescriptions 310.6 17.6
5 Housing modifications – amortised 307.6 17.4
6 Consumables, inc. dressings, ointments, batteries, 346.7 19.6
incontinence sheets, pads
7 Health practitioners 355.8 20.1
8 Aids and appliances inc. wheelchairs, special clothing, 202.0 11.4
communication aids
9 Furniture – amortised 105.5 6.0
Total 4,350.9 246.1
Sum 1, 2 (carers) 1,956.9 110.7
Sum 5, 9 (aids & modifications) – net 413.1 23.4
Sum 5, 6, 8, 9 (aids & modifications) – gross 759.8 54.4
The cost of care is a very conservative estimate as it does not include informal care, which is
able to be estimated for the other obesity related conditions. To date, there has not been a
comprehensive study into the informal care needs of Australians with osteoarthritis upon which
to base a robust estimate of the cost of this informal care, including the lost income of carers
(who could otherwise have been involved in paid work) and SDAC data were also unable to be
obtained and analysed in the timeframe and scope for this project. This informal care cost
could probably be quite substantial. A Dutch study (Brouwer et al, 2004) of rheumatoid
arthritis patients found that approximately 50% of all patients reported that they received
informal care from their partner. These informal caregivers spent, on average, 27.4 hours per
week providing care, comprising around 15 hours per week on household tasks such as
shopping, cleaning and other household chores and 12.4 hours per week assisting the patient
with the activities of daily living. Informal care was supplemented with formal assistance with
household tasks in 24% of cases (average 4.5 hours a week) and for activities of daily living in
3.9% of cases (average 2.5 hours a week). In addition, 6.1% of patients receiving informal
care were on a waiting list for formal care. A similar epidemiological study in Australia would
be valuable.
5.4.4 NON-HEALTH FINANCIAL COSTS ATTRIBUTABLE TO OBESITY

Table 5-15 shows that osteoarthritis non-health costs were $4,540 million in total in 2005, to
which the obesity AF of 14.0% is applied.
73
TABLE 5-15: PREVENTABLE INDIRECT COSTS DUE TO OBESITY, 2005, $M
Indirect Cost Item Total ($m) Obesity AF ($m)*

Lost earnings from workplace separation $3,963 $553
DWL $339 $47
Paid carers $111 $15
Aids and home modifications $23 $3
Associated travel $43 $6
*Calculated by applying an overall obesity AF of 14.0% (across all ages and genders)
For individuals with osteoarthritis as a result of obesity, real financial costs (other
than health costs) are estimated as $633 million (excluding the cost of informal
care).
5.5 CANCER
5.5.1 PRODUCTIVITY LOSSES

For cancer, the productivity cost pathways are slightly different and more complex relative to
the other obesity related conditions, since cancer can be a more acute illness while diabetes,
osteoarthritis and CVD tend to be more chronic (non-remitting) in nature. Moreover, since
carer costs for people with cancer were able to be measured on an opportunity cost basis,
these are effectively another form of productivity loss, so are summarised in conjunction with
the productivity losses.
Access Economics undertook a literature review for studies examining employment rates of
cancer patients. Analysis of the mean employment rates from Australian studies revealed:
The 2003 SDAC - cancer reduces the probability of employment by 41% in males
and 17% in females (28% overall); however this affect is mainly driven by reduced
employment in the 45-64 year olds when decisions about the age of retirement become
important.
The 2004-05 NHS - cancer reduces the probability of employment by 16% in males
and 14% in females (15% overall).
The NSW Cancer Survival Study by the Centre for Health Research and Psycho-
oncology (CHeRP) - cancer reduces the probability of employment by 29% in males
and 24% in females (which is between the results for SDAC and the NHS).
The impact of cancer on the probability of employment did not vary significantly
between the 50-59 year olds and those aged 60 years or older.
Excluding melanoma, the impact of cancer on the probability of employment18 did
not vary significantly by cancer type.
18
Estimated as the difference in employment rates.
74
The Australian Longitudinal Study on Women’s Health (ALSWH) - ever being

diagnosed with (non skin) cancer reduces the probability of employment on
average by 4.2%, but has no significant negative impact on hours worked and earnings
per hour worked. This study more likely reflects the long-term impact of cancer.
Given the survey methodology, SDAC is more likely to capture people actively being treated
for cancer (comparable to Short et al, 2005, which estimates a 50% reduction in employment
rates in the first six months since diagnosis). In comparison, the NHS is more likely to have
captured mainly longer-term survivors.
Consequently this study assumes that the short-term impact of cancer on the
probability of employment for people while they have active cancer is -41% for
males and -17% for females (based on SDAC).
ALSWH focused on people who have ever been diagnosed with cancer, and thus is most likely
to reflect the long-term impact of cancer. Furthermore as the analysis controlled for most
other factors that influence employment rates, it is probably the most accurate in attributing
changes in employment to cancer. As many of the surveys identified found significant
differences between males and females, the impact on productivity for females from the
ALSWH was inflated by the ratio between males and females from SDAC to estimate the long-
term impact on males.
Consequently this study assumes that the long-term impact of cancer on the
probability of employment for people ever diagnosed with cancer is -10.3% for
males and -4.2% for females, with people gradually returning to work at a linear
rate until five years after diagnosis.
Short et al (2005) examined the employment pathways of cancer survivors after diagnosis.
The study surveyed 1,763 cancer survivors aged between 25 and 62 years, and found no
significant difference by age between 45 to 62 years old. Furthermore, the Cancer Survival
Study found no significant difference by age between the 50-59 year olds and the 60+ year
olds. There is not enough information to determine any significant difference between these
age groups and the younger ages.
Consequently for adult survivors of cancer the impact on employment rates by

age group will not be varied for this study.
Finally this study will also assume that cancer has no long-term impact on hours
worked or on income per hour (based on the ALSWH analysis).
TABLE 5-16: SUMMARY OF IMPACT ON EMPLOYMENT RATES FOR CANCER PATIENTS

Survey Males Females
Active Prevalence -41.0% -17.0%
nd
2 Year -34.9% -14.4%
rd
3 Year -28.7% -11.9%
th
4 Year -22.6% -9.3%
th
5 Year -16.4% -6.8%
Long-term Impact -10.3% -4.2%
While many adult survivors are able to return to similar productivity levels after recovering from
cancer, reduced educational attainment through reduced cognitive functioning and memory
75
from radiotherapy and chemotherapy, and frequent and sometimes extended non-attendance
at school will lead to longer-term impacts on future productivity (Stam et al, 2005:228 and
Eiser 1998:625). Ongoing health problems as a result of treatment can further reinforce these
impacts on productivity. Moreover, an econometric analysis of the ALSWH found that, all else
being equal, at least seven years after diagnosis/treatment, childhood cancer survivors were
12% less likely to attain a university degree or higher – resulting in a 2.2% reduction in
employment rates, 1.7% reduction in hours worked, and 5.9% reduction in earnings per hour
worked. Overall earnings of childhood cancer survivors were 9.6% lower.
As seven years (or more) after diagnosis/treatment is a significant lapse in time

and the impact appear to be ongoing and increasing (rather than decreasing), this
study assumes that, on average, the impact of childhood cancer on educational
attainment opportunities permanently reduces lifetime earnings by 9.6% (based
on the ALSWH analysis).
In total, in 2005 an estimated $414.3 million is incurred through long-term reductions in

employment, hours worked and earnings per hour worked due to obesity-related
cancer.
5.5.1.2 ABSENTEEISM
Two studies were identified that estimate average days absent from work of employed people
with cancer.
A US study interviewed 445 employed people with breast (or prostate) cancer and found
that 93% of people with breast cancer missed at least one day of work over a period of
six months from diagnosis, and on average they missed 44.5 days from work,
respectively (Bradley, 2005).
The NHS estimated that, of employed people with cancer, 17.2% of males and 15.5% of
females had days away from work due to their illness in the past two weeks, and on
average they lost 4.8 days from work – equivalent to 20.5 days over 12 months per
employed person. While this estimate includes people with cancer taking time off for
work due to non-cancer related reasons, this estimate is more likely to underestimate
days absent as the survey is more likely to capture cancer survivors rather than people
actively being treated for cancer– thus lowering the average days absent.
In addition, data collected from the AIHW show that the average length of stay in hospital in
2002-03 for a person with cancer was 14.3 days, which represents 70% of days absent from
work compared to the NHS. Estimating days absent from work by cancer type by inflating
hospital days by 43% (1/70%) will provide a more accurate distribution of likely absenteeism,
as well as being a conservative estimate since:
the NHS is likely to underestimate days absent for people actively being treated for
cancer,
if the hospital days excluded days in hospital on weekends then the ratio between
hospital days and days absent would decrease (thus increasing the inflation rate), and
for some cancers, notably breast, chemotherapy with recovery in the home, rather than
treatment in hospital, is increasingly used.
Some people may use sick, annual, or long service leave when they are temporarily absent
from work. It is estimated that 71.2% of females and 78.1% of males are paid for the days
taken off paid work (ABS Cat No 6342.0 data series) and the employer incurs wages, on-costs
76
and an overtime premium relating to the paid days off work, while the worker incurs the lost
wages relating to the remaining unpaid days off work.
Furthermore, each day a patient is absent from work it is estimated that 30 minutes of
management time is lost processing those absent workers19. This includes the time of line
managers in rearranging work and the time of back-office personnel. The cost of managers’
time in 2004 is estimated to be $1,286 for an average working week of 40.2 hours (ABS Cat
No 6310.0 data series) plus 15.5% on costs (ABS Cat No 6348.0.55.001 data series).
TABLE 5-17: HOSPITAL DAYS AND DAYS ABSENT FROM WORK, AUSTRALIA, 2002-03
Days
Total Hospital
Total Active Separations Absent
Hospital ALOS Days per
Separations Prevalence per Person from
Days Person
Work
Colorectal 27,338 228,502 8.4 17,717 1.5 12.9 18.5
Breast 11,161 47,015 4.2 16,464 0.7 2.9 4.1
Kidney 11,398 50,073 4.4 3,260 3.5 15.4 22.1
Uterine 8,464 50,318 5.9 4,450 1.9 11.3 16.3
All four
Cancers
58,361 375,908 5.0 41,891 2.9 14.3 20.5
ALOS = Average Length of Stay. Australian Active Prevalence for 2002-03 is based on estimates for 2005,
deflated based on growth in cancer incidence from 2002-03 to 2005. Sources: AIHW (2003b) and ABS 4364.0,
special request.
In total, in 2005 an estimated $21.8 million is incurred through temporary absenteeism

from work (including management time) due to obesity-related cancer.
Premature death from cancer results in a long-term reduction in the productive capacity of the
labour force. The value of each person’s remaining expected earnings is allocated to the year
that they die.
People who prematurely died from cancer who were employed results in turnover costs to the
employer. As with the turnover costs for diabetes, this cost is estimated to be equal to
26 weeks salary of the incumbent worker (Access Economics, 2004a). However this cost is
merely ‘brought forward’ a number of years because there would be some normal turnover of
cancer patients – approximately 15% per annum (which implies that people change jobs, on
average, approximately once every 6.7 years (Access Economics, 2004b).
In total, in 2005 an estimated $1,092.7 million is incurred through loss of remaining

lifetime earnings due to premature death and an additional $3.8 million is incurred
through additional search, hiring and training costs due to obesity related cancer.
5.5.2 CARER COSTS

For cancer, the opportunity cost method was thus used as good quality data about the age
and sex of the carers of people with cancer were available from SDAC for adults and from
other sources for children.
19
The HSE (1999) assume that administrative costs associated with dealing with absences (such as the calculation
and payments of benefits, processing of sick leave and extra management time) equates to an average of 30
minutes per day of absence.
77
5.5.2.1 ADULTS
SDAC reported that in 2003 in Australia there were 18,800 carers of people whose main
condition was cancer, of which 4,300 were primary carers. By applying a point prevalence
weighting factor, the total carers for obesity related cancers were estimated (see Table 5-18).
Thus, on average, there were 3 carers per 100 people with active obesity related cancer20
(after indexing to population growth).
TABLE 5-18: CARERS OF PWO RELATED CANCERS, AUSTRALIA, 2005

Total Carers for all Cancers Carers Per 100 people with Active
Obesity related Cancer
Primary Non- Total Primary Non- Total
Primary Primary
Males
0-64 200 900 1,100 0.1 0.4 0.5
65+ 200 500 700 0.1 0.2 0.3
Total 400 1,400 1,800 0.2 0.6 0.7
Females
0-64 900 3,200 4,000 0.4 1.3 1.7
65+ 300 1,000 1,300 0.1 0.4 0.5
Total 1,200 4,200 5,300 0.5 1.7 2.2
Persons
0-64 1,100 4,000 5,100 0.5 1.7 2.1
65+ 500 1,500 1,900 0.2 0.6 0.8
Total 1,600 5,500 7,000 0.7 2.3 2.9
Source: ABS Cat No 4430.0 data series. Note: Totals may not sum due to rounding.
As SDAC reports hours of informal care provided per week for primary carers only, it is
necessary to impute the average number of hours of care given per week21 by primary (42
hours on average over all age groups and genders) and non-primary carers (5 hours). Overall
in 2005, around 254,900 hours of informal care were provided to people with cancer
(either colorectal, breast, kidney or uterine) in Australia, equivalent to 6 hours per
person with active obesity related cancer per year (Table 5-19).
20
Not a one-to-one relationship mainly due to the high presence of other conditions also causing disability in people
with cancer.
21
10 hours, 29.5, and 50 hours per week was imputed in <20 hours, 20-39 hours, and 40+ hours per week groups,
respectively and 5 hours per week was imputed for the non-primary carers (Access Economics, 2005c).
78
TABLE 5-19: AVERAGE HOURS OF INFORMAL CARE PROVIDED TO PEOPLE WITH CANCER
Active Non-primary Annual Carer
Prevalence Primary Carers Carers hours (‘000)
Males
0-64 4,658 42 155 132.9
65+ 7,624 29 89 90.3
Total 12,283 71 244 218.9
Females
0-64 16,310 153 565 186.7
65+ 14,568 56 172 44.7
Total 30,878 209 736 191.5
Persons
0-64 20,969 196 719 187.0
65+ 22,192 85 261 67.9
Total 43,161 281 980 254.9
Source: ABS Cat No 4430.0 data series
The opportunity cost method is calculated by multiplying the total number of informal hours of
care by the average employment rate and AWE and then dividing by the average hours
worked (adjusted for the demographic profile of carers).
TABLE 5-20: HOURS AND COST OF INFORMAL CARE OF PWO RELATED CANCER, 2005
Total Hours of Employment AWE ($) Opportunity
Care Provided Rate (%) Cost ($m)
(‘000)
Males
0-64 132.9 63% 681 $1.4
65+ 90.3 8% 578 $0.1
Total 218.9 $1.5
Females
0-64 186.7 52% 468 $1.6
65+ 44.7 3% 389 $0.0
Total 191.5 $1.6
Persons
0-64 187.0 $3.0
65+ 67.9 $0.1
Total 254.9 $3.1
In 2005 males worked on average 39.3 per week, and females worked 28.9 hours per week. Note: Totals may not
sum due to rounding
Sources: ABS Cat No 6105.0 and 6310.0 data series (indexed to $2005)
5.5.2.2 CHILDREN
Evidence suggests that there may be a greater impact on carers of children with cancer
(namely their parents) compared to carers of adults with cancer. Children often require
someone to accompany them to medical appointments, whereas many adult patients are able
to attend these appointments by themselves. Many children with cancer also have siblings
that require care, resulting in the shifting of parenting responsibilities or ‘tag-team parenting’
(Cohn et al 2003:855). Consequently carers of children with cancer may be more likely to
experience additional health impacts such as anxiety, depression, and sleep problems.
Sloper (1996) surveyed 181 parents in the UK six months post-diagnosis and found that, of
mothers who were employed when the child was diagnosed, 15% resigned or were dismissed
(20 out of 136 mothers). Similarly Dockerty et al (2000) surveyed 397 parents in New Zealand
79
(most were in the category 12 to 24 months post diagnosis) and found that that mothers
(average age was 35.7 years) had a 15.8% lower probability of being in a paid job compared
to the control group. Based on AWE for women aged 35-39 in 2005 of $1,06522 and a 15%
reduction in employment, this corresponds to an expected loss of $8,307 in income per year
per child diagnosed with cancer.
Cohn et al (2003) interviewed 100 parents in Australia around 3.4 years post-diagnosis, on
average, and asked them to estimate the financial impact on the family for various expenses
(see Table 5-21). They found that 28% of families experienced a loss of income through
reduced paid hours (including resigning from employment) at an average cost of $2,505 in
income per year per child diagnosed with cancer – a greater impact on employment than
Sloper (1996) and Dockerty (2000). However some families also incurred leave without pay,
used annual or sick leave, and closed/suspended the family business. Based on Cohn et al
(2003), the total expected loss in income is $8,031 per child with active cancer per year.
TABLE 5-21: IMPACT OF CHILDHOOD CANCER ON FAMILY INCOME
% of Families Incurring Average Cost Cost per Year

Cost ($2001*) ($2005)
Reduced Paid Hours** 28% 25,588 8,945
Leave Without Pay 39% 12,361 4,321
Close/suspend Business 21% 41,636 14,556
Use Annual/Sick Leave 35% 6,411 2,241
Expected Cost 22,973 8,031
* No time-frame for the survey was given, however part of the paper was presented in October 2001 so the survey
must have been conducted in 2001 or earlier
** Includes resigning
Source: Cohn et al (2003)
5.5.2.3 ALL CARERS
In total, in 2005 an estimated $24.0 million is incurred through the opportunity cost of
carer time due to cancer.
Overall in 2005, informal care provided to PWO related cancer in Australia cost
$3.4 million, or $78 per person with active cancer per year.
5.5.3 SUMMARY OF PRODUCTIVITY AND CARER COSTS

The expected productivity and carer costs (on an opportunity cost basis) in 2005 of colorectal,
breast, kidney or uterine cancer in Australia is around $1,556.6 million. Of the total
productivity costs:
$414.3 million is incurred through long-term reductions in employment, hours worked
and earnings per hour worked;
$21.8 million is incurred through temporary absenteeism from work (including
management time);
22
ABS Cat No 6310.0 data series (indexed to $2005).
80
$1,092.7 million is incurred through lost of remaining lifetime earnings due to premature
death;
$3.8 million is incurred through additional search, hiring and training costs; and
$24.0 million is incurred through the opportunity cost of carer time due to cancer.
The expected productivity cost of all four cancers combined is $47,970 per person (point
prevalence). Figure 5-4 illustrates that costs differ significantly by age of the person with
cancer.
Children: Child prevalence of the cancers being examined is very small therefore the
estimates should be treated with caution. The significant costs associated with child
cancers are made up largely of the productivity cost of premature mortality among
children and the high ongoing impact on earnings if they survive.
Working Age Population: The most costly cancers are colorectal ($149,300 per
person) and kidney ($160,350 per person). These costs are made up largely of the
productivity cost of premature mortality however employment reduction costs are also
significant.
Older Population: Per person productivity costs range between $1,510 (breast cancer)
and $5,840 (kidney cancer). These costs are made up largely of the productivity cost of
premature mortality.
FIGURE 5-4: PRODUCTIVITY AND CARER COSTS PER PERSON BY AGE AND CANCER TYPE ($)
$300,000
$250,000
$200,000
$150,000
$100,000
$50,000
$0
Colorectal Cancer Breast Cancer Kidney Cancer Uterine
Children Working Age Older
The expected productivity and carer costs (on an opportunity cost basis) in 2005 of
colorectal, breast, kidney or uterine cancer due to obesity in Australia is estimated
as $221.5 million.
81
5.5.4 DWL FROM TRANSFERS

5.5.4.1 LOSS OF TAX REVENUE
Cancer patients and their carers in paid employment, who are out of the workforce temporarily
due to sickness or caring responsibilities, or permanently due to premature retirement or
death, will contribute less tax revenue to the Government. This lost value in wages and firm
output was calculated in Section 5.5.3 and is distributed in the same manner as for diabetes in
relation to who bears the costs. Pre-tax (ie of the $1556.6 million total):
cancer patients lost around $1,510.3 million in wage income due to temporary
absenteeism, long-term lost earnings, and premature death;
carers lost around $24.0 million in wage income due to caring for the patient; and
employers lost around $22.3 million in production value on account of absenteeism of
the patient or carer, lost management productivity in managing the absenteeism, and
direct worker hiring/retraining costs.
With an average personal income tax rate of 21.2% and an average indirect tax rate of 15.51%
in 2005, the expected total loss of tax revenue from obesity related cancers was
approximately $571.4 million in 2005.
5.5.4.2 WELFARE PAYMENTS
A number of welfare payments are available to cancer patients and their carers, and can
sometimes be back-dated to when costs began to be incurred (ie, when the person stopped
working). The cost of welfare payments in this study relate to those directly attributable to
cancer.
Patients
Data from Centelink was used to establish that 33% of all working aged people who had active
cancer in 2005 received welfare payments (DSP, Newstart, or Sickness Allowance) due to
cancer (Access Economics, 2006c). However some of these people would have ordinarily
received welfare payments which must be netted out to estimate the additional welfare
payments due to cancer, using a Melbourne University study (Tseng and Wilkins, 2002) about
the ‘reliance’ of the general population (aged 15-64 years) on income support. Based on this
information, each working age person in Australia with active obesity related cancer (20,900
people) receives $1,007 in additional welfare payments per annum, on average. $21.3 million
was spent on additional welfare payments to Australians with obesity related cancers in
2005.
Carers
In 2005 around $2,012.7m was paid to carers in the form of the Carer Payment and Carer
Allowance and 1.221 billion hours of care was provided (Access Economics, 2005c) equivalent
to $1.68 per hour. Based on the total amount of hours of care provided by informal cares of
PWO related cancer and assuming that carers these cancer patients receive the Carer
Payment or Allowance at the same rate as the general carer population for the period for
which care is required (ie, while the person has active cancer), then $2.3 million was
provided to carers of Australian PWO related cancers in 2005.
82
5.5.4.3 DWL
For the four obesity related cancers, the DWL of the total extra tax dollars required to be
collected were $340.8 million in 2005 comprising (Table 5-22):
the lost taxation revenues from patients, carers and employers;
the additional induced social welfare payments required to be paid; and
the value of Government services provided (eg. health system costs, counselling etc).
TABLE 5-22: CANCER, SUMMARY OF TRANSFERS AND DWL, 2005
$m
Health System Costs Borne by Government 576.5
Lost Taxes 571.4
Other Costs Borne by Government 13.9
DWL 340.8
For cancer costs that were attributable to obesity, the estimated DWL was
$50.7 million (14.9% of $340.8 million).
5.5.5 OTHER NON-HEALTH FINANCIAL COSTS

This section estimates the main costs of respite and palliative care programs used by PWO
related cancers as well as various out-of-pocket expenses, community programs and the
bring-forward of funeral costs.
5.5.5.1 NATIONAL RESPITE FOR CARERS PROGRAM
The National Respite for Carers Program provides services for at-home carers of people who
are unable to look after themselves due to frailty, disability, or chronic illness. Funding data
presented in Productivity Commission (2006) combined with distributional analysis from
Australian National Audit Office (2005) enabled an estimate of the total expenditure in 2005 on
respite for carers of cancer patients of $5.8 million, or $169 per death. For PWO related
cancers in 2005, the cost of respite for carers was $1.7 million.
5.5.5.2 PALLIATIVE CARE
Palliative care is the specialised care provided for people who are dying from active,
progressive and far-advanced diseases, with little or no prospect of cure. The aim of palliative
care is to achieve the best possible quality of life, both for the person who is dying and for their
family. Services are provided by both government and non-government organisations,
delivered by family and friends, GPs and palliative care specialists. State and Territory
governments have the primary responsibility for management of palliative care services in
Australia, while the Federal government has an oversight role (responsible for planning and
strategy). Cancer is the primary diagnosis for which most (89.7% of) palliative care services
are provided (Palliative Care Australia, 1998). Total Federal government funding for the
National Palliative Care Program in 2005 was approximately $60 million, of which an
estimated $53.6 million was for the palliative care of cancer patients or $1,570 per death. This
estimate would underestimate the total funding for palliative care, as it does not include
contributions from State and Territory governments and private sources. However, the AIHW
estimates of health costs for cancer in Section 4.5 already includes the costs of palliative care
83
provided in hospitals and hospices, but not community palliative care services. Consequently
to avoid double counting, only 78% of these costs are attributed to community based services
(Department of Health and Ageing, 2005), or $41.8 million.
In addition to funded services, Palliative Care Australia (1998) estimated that approximately
78% of palliative care services used volunteers, with each service receiving approximately
35.7 hours per week of volunteer time. Thus for the 254 services registered on the Register of
Palliative Care Services held by Palliative Care Australia, around 367,790 hours of volunteer
service was provided, or 10.7 hours per annum per death. Thus in 2005 around 110,500
volunteer hours were provided to palliative care services, worth $2.8 million (valued at a
replacement cost of $25.01) of which $2.5 million was provided to PWO related cancer, or
$268 per death.
Consequently the total value of community based palliative care services was $1,494 per
death, which equates to a cost of $14.9 million for Australians with obesity related
cancers in 2005.
5.5.5.3 OUT-OF-POCKET EXPENSES
Estimates of out-of-pocket expenses were based on Arozullah et al (2004) who surveyed 156
women with breast cancer in the US between October 1999 and November 2002. Overall the
additional financial costs for adult cancer were $2,648 for females and $2,120 for males per
year, including counselling/support, special equipment, childcare, transport, housekeeping,
meals, telephone and accommodation. In addition, various studies have found that a large
proportion of people with cancer use at least one complementary or alternative therapy after
their diagnosis. The most commonly practised therapies include meditation/relaxation, diet,
multivitamins, herbal medicines and antioxidants. Using results from one (conservative)
Australian study (Miller et al, 1998), who estimated 40% usage of such services in Australia,
the expected cost of complementary or alternative therapies is $422 per patient for the entire
course of the cancer.
For some out-of-pocket expenses, childhood cancer may be more costly than adult cancer.
Based on Cohn et al (2003), the additional financial costs for childhood cancer are $1,318 per
year per child diagnosed with cancer.
Overall, the expected out-of-pocket costs (aids, equipment and modifications, formal
care, travel and accommodation, communication, and complementary or alternative
therapies) were $125.2 million for obesity related cancers in 2005.
5.5.5.4 COMMUNITY PROGRAMS
Many non-profit organisations provide programs for people with cancer paid for with funding
from government grants and donations from the wider community. Counselling and support
programs (such as Look Good…Feel Better, CanTeen and, Make-a –wish Foundation) provide
emotional support for people with cancer and their families. In 2005 at least $2.7 million was
spent on these programs and 7,600 hours of volunteer time was provided – worth a further
$73,000 (based on AWE, employment rates and hours worked for people aged 15-65 years).
This equates to around $65 per person with active cancer.
To assist rural and remote families stay together during treatment, many organisations also
provide accommodation close to treatment facilities. In 2005 at least $2.5 million was spent on
providing accommodation to people with cancer by non-profit organisations, or $59 per person
with active cancer.
84
While the following costs are mainly educational programs and paper-based educational
material and support packs, patients can also find a wealth of information about cancer on the
internet. In 2005 at least $1.4 million was spent on providing educational material and
programs to people with cancer by non-profit organisations and 700 hours of volunteer time
was provided – worth a further $7,000 (based on AWE, employment rates and hours worked
for people aged 15-65), or $34 per person with active cancer.
Overall, the cost of community programs (such as counselling and support,

accommodation and educational programs) for obesity related cancers is estimated as
at least $6.8 million in 2005.
5.5.5.5 FUNERAL EXPENSES
The bring-forward of funeral expenses is measured in the same manner as for diabetes
(Section 5.2.5), based on the average cost of a funeral of $3,949 per person in 2005. Overall,
the cost of funeral expenses brought forward is estimated as at least $11.5 million for
PWO related cancers in 2005.
5.5.5.6 SUMMARY OF OTHER FINANCIAL COSTS
Apart from the productivity and carer costs and DWL, other non-health financial costs
associated with obesity related cancers in 2005 in Australia are estimated as $160.1 million,
of which:
$1.7 million is government funding for respite;
$14.9 million is government funding and volunteer time for community-based palliative
care;
$125.2 million is out-of-pocket expenses such as aids, equipment and modifications,
formal care, travel and accommodation, communication, and complementary or
alternative therapies;
$6.8 million is community programs run by non-profit organisations; and
$11.5 million is funeral costs brought forward.
However these costs are an underestimation of the total cost of these items as volunteer time
and funding by non-profit organisations is often unreported.
Of these expenses, the most costly related to colorectal cancer ($68.4 million) and breast
cancer ($60.1 million), as illustrated in Figure 5-5.
85
FIGURE 5-5: SUMMARY OF OTHER FINANCIAL COSTS BY CANCER TYPE AND GENDER
60
50
40
Cost ($m)
30
Column 2 - Breast Column 3 - Kidney
Column 4 - Uterine
20
Column 1 - Colorectal
10
0
Respite Care Palliative Care Out-of-pocket Expenses Community Programs Funeral Costs
Male female
5.5.6 SUMMARY OF NON-HEALTH FINANCIAL COSTS ATTRIBUTABLE TO

OBESITY
TABLE 5-23: NON-HEALTH FINANCIAL COSTS, CANCERS DUE TO OBESITY, 2005, $M
Indirect Cost Item Total ($m) Obesity AF ($m)

Lost earnings from workplace separation $414 $59
Lost earnings from premature death $1,093 $155
Search, hiring & training $4 $1
Carers’ productivity losses $24 $3
DWL $341 $51
Out-of-pocket expenses $125 $19
Other financial costs $35 $5
For individuals with cancer as a result of obesity, real financial costs (other than
health costs) are estimated as $296 million (measuring informal care on an
opportunity cost basis).
86
6. BURDEN OF DISEASE
6.1 METHODOLOGY – VALUING LIFE AND HEALTH
Since Schelling’s (1968) discussion of the economics of life saving, the economic literature
has properly focused on willingness to pay (willingness to accept) measures of mortality and
morbidity risk. Using evidence of market trade-offs between risk and money, including
numerous labour market and other studies (such as installing smoke detectors, wearing
seatbelts or bike helmets etc), economists have developed estimates of the Value of a
‘Statistical’ Life (VSL).
The willingness to pay approach estimates the value of life in terms of the amounts
that individuals are prepared to pay to reduce risks to their lives. It uses stated or
revealed preferences to ascertain the value people place on reducing risk to life
and reflects the value of intangible elements such as quality of life, health and
leisure. While it overcomes the theoretical difficulties of the human capital
approach, it involves more empirical difficulties in measurement (BTE, 2000, pp20-
21).
Viscusi and Aldy (2002) summarise the extensive literature in this field, most of which has
used econometric analysis to value mortality risk and the ‘hedonic wage’ by estimating
compensating differentials for on-the-job risk exposure in labour markets, in other words,
determining what dollar amount would be accepted by an individual to induce him/her to
increase the possibility of death or morbidity by a given percentage. They find the VSL ranges
between US$4 million and US$9 million with a median of US$7 million (in year 2000 US
dollars), similar but marginally higher than the VSL derived from US product and housing
markets, and also marginally higher than non-US studies, although all in the same order of
magnitude. They also review a parallel literature on the implicit value of the risk of non-fatal
injuries.
A particular life may be regarded as priceless, yet relatively low implicit values may
be assigned to life because of the distinction between identified and anonymous
(or ‘statistical’) lives. When a ‘value of life’ estimate is derived, it is not any
particular person’s life that is valued, but that of an unknown or statistical individual
(Bureau of Transport and Regional Economics, 2002, p19).
Weaknesses in this approach, as with human capital, are that there can be substantial
variation between individuals. Extraneous influences in labour markets such as imperfect
information, income/wealth or power asymmetries can cause difficulty in correctly perceiving
the risk or in negotiating an acceptably higher wage.
Viscusi and Aldy (2002) include some Australian studies in their meta-analysis, notably
Kniesner and Leeth (1991) of the ABS with VSL of US2000 $4.2 million and Miller et al (1997)
of the National Occupational Health and Safety Commission (NOHSC) with quite a high VSL
of US2000$11.3m-19.1 million (Viscusi and Aldy, 2002, Table 4, pp92-93). Since there are
relatively few Australian studies, there is also the issue of converting foreign (US) data to
Australian dollars using either exchange rates or purchasing power parity and choosing a
period.
87
Access Economics (2003) presents outcomes of studies from Yale University (Nordhaus,
1999) – where VSL is estimated as $US2.66m; University of Chicago (Murphy and Topel,
1999) – US$5m; Cutler and Richardson (1998) – who model a common range from US$3m to
US$7m, noting a literature range of $US0.6m to $US13.5m per fatality prevented (1998 US
dollars). These eminent researchers apply discount rates of 0% and 3% (favouring 3%) to the
common range to derive an equivalent of $US 75,000 to $US 150,000 for a year of life gained.
6.1.1 DISABILITY ADJUSTED LIFE YEARS (DALYS) AND QUALITY ADJUSTED

LIFE YEARS (QALYS)
In an attempt to overcome some of the issues in relation to placing a dollar value on a human
life, in the last decade an alternative approach to valuing human life has been derived. The
approach is non-financial, where pain, suffering and premature mortality are measured in
terms of DALYs, with 0 representing a year of perfect health and 1 representing death (the
converse of a QALY where 1 represents perfect health). This approach was developed by the
World Health Organization, the World Bank and Harvard University and provides a
comprehensive assessment of mortality and disability from diseases, injuries and risk factors
in 1990, projected to 2020 (Murray and Lopez, 1996). Methods and data sources are detailed
further in Murray et al (2001).
The DALY approach has been adopted and applied in Australia by the AIHW with a separate
comprehensive application in Victoria. Mathers et al (1999) from the AIHW estimate the BoD
and injury in 1996, including separate identification of premature mortality; Years of Life Lost
due to Premature Mortality (YLL), and morbidity; Years of Healthy Life Lost due to Disability
(YLD) components. In any year, the disability weight of a disease (for example, 0.18 for a
broken wrist) reflects a relative health state. In this example, 0.18 would represent losing 18%
of a year of healthy life because of the inflicted injury.
The DALY approach has been successful in avoiding the subjectivity of individual valuation
and is capable of overcoming the problem of comparability between individuals and between
nations, although nations have subsequently adopted variations in weighting systems. For
example, in some countries DALYs are age-weighted for older people although in Australia the
minority approach is adopted – valuing a DALY equally for people of all ages.
The main problem with the DALY approach is that it is not financial and is thus not directly
comparable with most other cost measures. In public policy making, therefore, there is always
the temptation to re-apply a financial measure conversion to ascertain the cost of an injury or
fatality or the value of a preventive health intervention. Such financial conversions tend to
utilise “willingness to pay” or risk-based labour market studies described above.
The Department of Health and Ageing (based on work by Applied Economics) adopted a very
conservative approach to this issue, placing the value of a human life year at around A$60,000
per annum, which is lower than most international lower bounds on the estimate.
“In order to convert DALYs into economic benefits, a dollar value per DALY is
required. In this study, we follow the standard approach in the economics
literature and derive the value of a healthy year from the value of life. For
example, if the estimated value of life is A$2 million, the average loss of healthy life
is 40 years, and the discount rate is 5% per annum, the value of a healthy year
would be $118,000.23 Tolley, Kenkel and Fabian (1994) review the literature on
23 2 40
In round numbers, $2,000,000 = $118,000/1.05 + $118,000/(1.05) + … + $118,000/(1.05). [Access Economics
comment: The actual value should be $116,556, not $118,000 even in round numbers.]
88
valuing life and life years and conclude that a range of US$70,000 to US$175,000
per life year is reasonable. In a major study of the value of health of the US
population, Cutler and Richardson (1997) adopt an average value of US$100,000
in 1990 dollars for a healthy year.
Although there is an extensive international literature on the value of life (Viscusi,

1993), there is little Australian research on this subject. As the Bureau of
Transport Economics (BTE) (in BTE, 2000) notes, international research using
willingness to pay values usually places the value of life at somewhere between
A$1.8 and A$4.3 million. On the other hand, values of life that reflect the present
value of output lost (the human capital approach) are usually under $1 million.
The BTE (2000) adopts estimates of $1 million to $1.4 million per fatality, reflecting
a 7% and 4% discount rate respectively. The higher figure of $1.4 million is made
up of loss of workforce productivity of $540,000, loss of household productivity of
$500,000 and loss of quality of life of $319,000. This is an unusual approach that
combines human capital and willingness to pay concepts and adds household
output to workforce output.
For this study, a value of $1 million and an equivalent value of $60,000 for a
healthy year are assumed.24 In other words, the cost of a DALY is $60,000. This
represents a conservative valuation of the estimated willingness to pay values for
human life that are used most often in similar studies.25” (DHA, 2003, pp11-12).”
As the citation concludes, the estimate of $60,000 per DALY is very low. The Viscusi (1993)
meta-analysis referred to reviewed 24 studies with values of a human life ranging between
$US 0.5 million and $US 16m, all in pre-1993 US dollars. Even the lowest of these converted
to 2003 Australian dollars at current exchange rates, exceeds the estimate adopted ($1m) by
nearly 25%. The BTE study tends to disregard the literature at the higher end and also adopts
a range (A$1-$1.4m) below the lower bound of the international range that it identifies (A$1.8-
$4.3m).
The rationale for adopting these very low estimates is not provided explicitly. Certainly it is in
the interests of fiscal restraint to present as low an estimate as possible.
In contrast, the majority of the literature as detailed above appears to support a higher
estimate for VSL, as presented in Table 6-1, which Access Economics believes is important to
consider in disease costing applications and decisions. The US dollar values of the lower
bound, midrange and upper bound are shown at left. The ‘average’ estimate is the average of
the range excluding the high NOHSC outlier. Equal weightings are used for each study as the:
Viscusi and Aldy meta-analysis summarises 60 recent studies;
ABS study is Australian; and
Yale and Harvard studies are based on the conclusions of eminent researchers in the
field after conducting literature analysis.
24
The equivalent value of $60,000 assumes, in broad terms, 40 years of lost life and a discount rate of 5 per cent.
[Access Economics comment: More accurately the figure should be $58,278.]
25
In addition to the cited references in the text, see for example Murphy and Topel’s study (1999) on the economic
value of medical research. [Access Economics comment. Identical reference to our Murphy and Topel (1999).]
89
Where there is no low or high US dollar estimate for a study, the midrange estimate is used to
calculate the average. The midrange estimates are converted to Australian dollars at
purchasing power parity (as this is less volatile than exchange rates) of USD=0.7281AUD for
2003 as estimated by the OECD.
Access Economics concludes the VSL range in Australia lies between $3.7m and $9.6m26,
with a mid-range estimate of $6.5m. These estimates have conservatively not been inflated to
2004 prices, given the uncertainty levels.
TABLE 6-1: INTERNATIONAL ESTIMATES OF VSL, VARIOUS YEARS

US$m A$m
Lower Midrange Upper 0.7281
Viscusi and Aldy meta- 4 7 9 9.6
analysis 2002
Australian: ABS 1991 4.2 5.8
NOHSC 1997 11.3 19.1
Yale (Nordhaus) 1999 2.66 3.7
Harvard (Cutler and 0.6 5 13.7 6.9
Richardson) 1998
Average* 2.9 4.7 7.4 6.5
* Average of range excluding high NOHSC outlier, using midrange if no data; conservatively not inflated
A$m conversions are at the OECD 2003 PPP rate
Discounting the VSL of $3.7m from Table 6-1 by the discount rate of 3.3% (from Section 1.2.4)
over an average 40 years expected life span (the average from the meta-analysis of wage-risk
studies) provides an estimate of the Value of a Life Year (VLY) of $162,561.
6.2 DIABETES
6.2.1 YEARS OF LIFE LOST DUE TO PREMATURE DEATH

Recalling Section 5.2.1.3, there were an estimated 3,092 deaths (1,570 males and 1,522
females) due to Type 2 diabetes in 2005. In comparison, there were an estimated 25,640
deaths (12,616 males and 13,023 females) due to other causes in the population affected by
Type 2 diabetes. Based on those estimates, the RR of mortality due to Type 2 diabetes was
calculated as 1.124 for males and 1.117 for females. Although this is much lower than the RR
of 2 estimated by the International Diabetes Institute (2006), it might be explained by including
all people with Type 2 diabetes and not just those individuals with previously diagnosed
diabetes (ie there would be a larger proportion of people with relatively mild or early diabetes).
YLL was then calculated from the age-gender distribution of deaths by the corresponding YLL
for the age of death in the Standard Life Expectancy Table (West Level 26) with a discount
rate of 3.3% and no age weighting. For the age-gender distribution of deaths, the total YLL for
diabetes in 2005 was estimated as 23,030 DALYs (Table 6-2).
26
Calculated from the non-indexed studies themselves. Converting the Access Economics average estimates from
USD to AUD at PPP would provide slightly higher estimates - $3.9 million and $10.2m, with the same midrange
estimate.
90
TABLE 6-2: DEATHS AND YLL DUE TO TYPE 2 DIABETES, 2005

0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+ Total
Deaths, Males 0 0 0 5 28 148 329 1,061 1,570
Deaths, Females 0 0 0.1 2 13 53 154 1,300 1,522
Deaths, Persons 0 0 0.1 7 41 201 483 2,361 3,092
YLL, Males 0 0 0 107 542 2,307 3,632 5,909 12,499
YLL, Females 0 0 3 51 266 907 1,932 7,372 10,531
YLL, Persons 0 0 3 158 808 3,214 5,564 13,281 23,030
Source: Access Economics
6.2.2 YEARS OF LIFE LOST DUE TO DISABILITY

The disability weight used in this study is based on the AIHW estimate for Type 2 diabetes of
0.070 (Mathers et al, 1999). The number of people experiencing loss of wellbeing due to
disability from Type 2 diabetes is estimated by gender as shown in Table 6-3, totalling 66,037
DALYs.
TABLE 6-3: ESTIMATED YLD FOR TYPE 2 DIABETES, 2005

Number of people with Disability weight YLD
Type 2 diabetes
Males 488,706 0.070 34,207
Females 454,628 0.070 31,830
Persons 943,334 0.070 66,037
6.2.3 TOTAL DALYS DUE TO TYPE 2 DIABETES

Figure 6-1 illustrates YLD and YLL due to Type 2 diabetes, which total 89,065 DALYs. The
greatest impact of Type 2 diabetes is in the 75+ age group due to the disability burden for
women, while the DALYs lost in the 55-64 year group are mainly due to mortality in men.
The estimated gross cost of lost wellbeing due to Type 2 diabetes is $14.5 billion in 2005
(calculated by multiplying 89,065 DALYs by the VLY of $162,561).
For Type 2 diabetes as a result of obesity, the estimated gross cost of lost
wellbeing is $1.6 billion.
91
FIGURE 6-1: LOSS OF WELLBEING DUE TO TYPE 2 DIABETES (DALYS), BY AGE AND GENDER, 2005
35,000
30,000 Female YLL
25,000 Male YLL
Female YLD
DALYs
20,000
15,000 Male YLD
10,000
5,000
0
25-34 35-44 45-54 55-64 65-74 75+
Age Groups
6.2.4 NET VALUE OF HEALTHY LIFE LOST

Bearing in mind that the wage-risk studies underlying the calculation of the VSL take into
account all known personal impacts – suffering and premature death, lost wages/income, out-
of-pocket personal health costs and so on – the estimate of $14.5 billion should be treated as
a ‘gross’ figure. However, costs specific to Type 2 diabetes that are unlikely to have entered
into the thinking of people in the source wage/risk studies should not be netted out (eg,
publicly financed health spending, care provided voluntarily). The results after netting out are
presented in Table 6-4.
TABLE 6-4: NET COST OF LOST WELLBEING, TYPE 2 DIABETES, $M, 2005
Individual
Gross cost of lost wellbeing 14,479
Minus production losses net of tax 2,553
Minus health costs borne out-of-pocket 211
Minus aids and modification costs 53
Net cost of lost wellbeing 11,662
The net cost of lost wellbeing due to Type 2 diabetes was thus $11.66 billion in 2005.
The net cost of lost wellbeing due to diabetes caused by obesity was $1.26 billion
in 2005.
6.3 CVD
Throughout the last 25 years, there have been substantial declines in the mortality burden of
CVD. The era 1981 to 1996 saw a decline in mortality burden of 30%-40% for both CHD and
stroke, a reflection of some successful primary prevention methods (such as reductions in the
level of tobacco smoking, better control of hypertension and dietary changes) and of
92
improvements in treatment (AIHW, 2004). However recent obesity issues in Australia, if not
managed effectively, may potentially retard or even reverse these declining mortality trends.
Table 6-5 shows the total disease burden, in terms of YLLs and YLDs, from obesity related
CVDs in 2005, by disease. There are two offsetting factors in projecting to 2005 from the 1996
Australian data. While prevalence is increasing due demographic ageing in the population
growth, there have been falls in mortality rates and morbidity due to better treatments. Thus
only an increase in line with population growth is modelled.
On this basis, obesity related CVD is estimated to be responsible for 532,334

years of healthy Australian life lost in 2005.
Of the total obesity-related CVD DALYs lost, 81.5% (433,821) were lost due to the
premature death (YLL) of people with CVD. The remaining 18.5% (98,513) of healthy
life years (DALYs) are lost due to disability (YLD).
CHD is the source of 65.0% of the total DALYs, with stroke a further 28.5%.
Males bear 54.3% of the disease burden.
TABLE 6-5: BOD OF OBESITY RELATED CVDS, AUSTRALIA, 2005

DALYs Total Males Females
CHD 345,791 200,624 145,166
Stroke 151,696 71,451 80,245
PAD 20,362 11,276 9,087
Hypertensive disease 14,485 5,553 8,932
Total 532,334 288,904 243,430
YLL
Ischaemic heart disease 306,304 175,909 130,394
Stroke 109,428 46,497 62,932
PAD 5,527 2,507 3,020
Hypertensive disease 12,562 5,139 7,424
Total 433,821 230,052 203,769
YLD
Ischaemic heart disease 39,487 24,715 14,772
Stroke 42,267 24,954 17,313
PAD 14,835 8,769 6,067
Hypertensive disease 1,923 414 1,508
Total 98,513 58,852 39,661
Source: Access Economics based on Mathers et al (1999)
Applying the VLY to the DALYs associated with obesity related CVD, Access Economics
estimates the gross cost of lost wellbeing was $86.5 billion in 2005. Distributional detail is
provided in Figure 6-2 and Note: The columns refer to the DALYs (right axis) and the lines to the gross costs
(left axis)
Table 6-6.
93
FIGURE 6-2: CVD OBESITY RELATED BOD AND GROSS COST OF LOST WELLBEING, 2005
$35,000 400,000
$30,000 350,000
300,000
$25,000
Total Cost ($m)
250,000
Life Years
$20,000
200,000
$15,000
150,000
$10,000
100,000
$5,000 50,000
$0 0
Ischaemic heart Stroke Peripheral arterial Hypertensive
disease disease disease
Male - YLD Male - YLL Female - YLD Female - YLL Male Female
Note: The columns refer to the DALYs (right axis) and the lines to the gross costs (left axis)
TABLE 6-6: GROSS COST OF LOST WELLBEING FROM CVD, 2005 ($ BILLION)
Male Female Total

Gross YLL cost 37.4 33.1 70.5
Gross YLD cost 9.6 6.4 16.0
Gross DALY Cost 47.0 39.6 86.5
After adjusting for the share of costs already borne by individuals, Table 6-7 shows that the
net cost of lost wellbeing due to obesity-related CVD was $83.4 billion in 2005.
TABLE 6-7: NET COST OF SUFFERING FROM OBESITY RELATED CVD, 2005
Total ($m) Obesity AF ($m)
Gross cost of suffering $86,537 $11,665

less lost earnings after tax 2,475 334
less health costs borne personally 769 84
plus welfare payments received by individual 117 16
Net cost of suffering $83,410 $11,263
94
The net cost of pain and suffering for obesity related CVD in 2005, after lost
earnings and the out-of-pocket personal health costs of individuals are removed, is
$83.4 billion, as shown in Table 6-7, of which $11.3 billion can be attributed to
obesity.
6.4 OSTEOARTHRITIS
The calculations of the BoD from osteoarthritis are based on extrapolations of the Mathers et
al (1999) estimates for osteoarthritis to 2005, based on the growth in the number of people
with osteoarthritis over this period.
FIGURE 6-3: OSTEOARTHRITIS, BOD, 2005, BY AGE & GENDER
4,000 16,000
3,500 14,000
3,000 12,000
Column 2 - Females
Gross Cost ($m)
2,500 10,000
Life Years
2,000 Column 1 - Males 8,000
1,500 6,000
1,000 4,000
500 2,000
0 0
0-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
YLL YLD YLL Cost Burden YLD Cost Burden
Figure 6-3 displays both the mortality and morbidity burden by age group and gender.
The mortality burden of osteoarthritis pales in comparison to the morbidity burden (the
YLL is 748 DALYs compared to 68,330 DALYs for the YLD).
Due to the demographic prevalence of the condition, female DALYs (41,326 years) are
about one and half times higher than male DALYs (27,753 years). Indeed, osteoarthritis
was the third leading contributor for females of disease burden in 1996, responsible for
5.7% of total YLD. Amongst the male population, osteoarthritis made up 3.9% of the
total YLD.
As would be expected due to the greater prevalence of arthritis in older age groups, the
BoD is also borne largely by people over the age of 45 years.
Applying the VLY of $162,561 to the DALYs associated with osteoarthritis, Access Economics
estimates the gross cost of lost wellbeing associated with osteoarthritis was $11.2 billion in
2005 (Table 6-8), of which $1.6 billion can be attributed to obesity.
95
TABLE 6-8: GROSS COST OF LOST WELLBEING FROM OSTEOARTHRITIS, 2005
Item Male ($m) Female ($m) Total ($m) Obesity AF($m)

Gross YLL cost 36 86 122 13
Gross YLD cost 4,476 6,632 11,108 1,624
Gross DALY cost 4,512 6,718 11,230 1,637
After adjusting for the share of costs already borne by individuals, Table 6-9 shows that the
net cost of lost wellbeing due to osteoarthritis was $7.9 billion in 2005.
The net cost of lost wellbeing due to osteoarthritis caused by obesity was
$1.2 billion in 2005.
TABLE 6-9: NET COST OF LOST WELLBEING, OSTEOARTHRITIS, $M, 2005

Gross cost of lost wellbeing 11,230 1,637
Less lost earnings after tax 2,845 397
Less paid carers costs 111 15
Less aid & modification costs 23 3
Less travel costs 43 6
Less health costs borne personally 338 44
Net cost of lost wellbeing 7,869 1,172
6.5 CANCER
For people with the four obesity related cancers in 2005, the total loss of wellbeing is
38,750 YLDs and 119,900 YLLs. YLLs are based on the most current survival data.
96
FIGURE 6-4: CANCER, BOD, 2005, BY AGE AND GENDER

$5,000 25,000
Column 1 - Colorectal
$4,500
$4,000 20,000
$3,500
Column 2 - Breast
$3,000 15,000
Total Cost ($m)
Life Years
$2,500
Column 3 - Kidney
$2,000 10,000
$1,500
$1,000 Column 4 - Uterine 5,000
$500
$0 -
0-14 15-24 25-34 35-44 45-54 55-64 65-74 75+
Age Group
Male - YLD Male - YLL Female - YLD Female - YLL Male Female
Figure 6-4 displays both the mortality and morbidity burden by age-group and gender for
obesity related cancers.
Breast cancer is the greatest source of disability burden (51% of total YLD for obesity
related cancers) with colorectal cancer second (35% of total YLD).
Colorectal cancer has the greatest mortality burden (50% of total YLL for obesity related
cancers), with breast cancer second (24% of total YLL).
Overall, obesity related cancers cost 158,650 DALYs, of which colorectal cancer was
47%, breast cancer 31%, uterine cancer 13% and kidney cancer 10%.
Based on the VLY estimate of $162,561, and after adjusting for the share of costs already
borne by individuals, Table 6-10 shows that the net cost of lost wellbeing due to obesity
related cancer was $24.6 billion in 2005.
The net cost of lost wellbeing due to cancer caused by obesity was $3.5 billion in
2005.
TABLE 6-10: NET COST OF LOST WELLBEING FROM CANCER, 2005
Cancer Type DALY ($m) Obesity AF($m)

Colorectal 11,457 1,526
Breast 7,627 1,203
Kidney 2,320 305
Uterine 3,217 507
Total cost 24,621 3,542
97
7. COST SUMMARY AND INTERVENTIONS

This final chapter first draws togethers the costs of obesity calculated in the previous chapters
(Section 7.1). The economic cost of diabetes has not previously been calculated by Access
Economics so these costs are also summarised in Section 7.2. The remaining sections
summarise discusses the microeconomics of obesity, briefly outlining the economic arguments
for intervention by the public sector as well as market solutions. A range of public sector
interventions have been mooted and there is stronger evidence for the efficacy of some
interventions rather than others. Problems with taxation options and potential regulatory
responses are discussed as well as the cost effectiveness of public health strategies targeting
physical activity and/or diet, pharmacotherapies and surgical inteventions.
7.1 SUMMARY OF THE COST OF OBESITY

The economic cost of obesity in Australia is summarised in Table 7-1 and Figure 7-2.
The total financial cost of obesity in 2005 was estimated as $3.767 billion.
Of this, productivity costs were estimated as $1.7 billion (45%), health system
costs were $873 million (23%) and carer costs were 804 million (21%).
DWL from transfers (taxation revenue foregone, welfare and other Government
The net cost of lost wellbeing was valued at a further $17.2 billion, bringing the total cost
of obesity in 2005 to $21.0 billion.
FIGURE 7-1: FINANCIAL COSTS OF OBESITY BY TYPE OF COST, 2005 (% TOTAL)
1.1%
9.5%
23.2% Health system
Productivity
Carers
21.3%
DWL
Other indirect
44.9% Total = $3.767billion
98
TABLE 7-1: COST SUMMARY, OBESITY ($M) 2005

Type 2 diabetes
BoD 1,269 0 0 0 0 0 1,269
Health System 23 0 54 25 0 15 116
Productivity 277 0 162 0 3 0 442
Carers 0 456 23 0 0 0 479
DWL 0 0 0 0 0 76 76
Transfers 0 -18 18 0 0 0 0
Total financial 305 439 257 25 3 90 1,119
Total inc. BoD 1,574 439 257 25 3 90 2,389
CVD
BoD 11,263 0 0 0 0 0 11,263
Health System 84 0 198 93 0 54 428
Productivity 334 0 138 0 0 0 472
Carers 0 217 90 0 0 0 306
DWL 0 0 0 0 0 184 184
Transfers -16 -39 55 0 0 0 0
Total financial 402 178 480 93 0 237 1,390
Total inc. BoD 11,665 178 480 93 0 237 12,653
Osteoarthritis
BoD 1,172 0 0 0 0 0 1,172
Health System 44 0 102 48 0 28 221
Productivity 164 0 397 0 0 0 561
Carers 15 0 0 0 0 0 15
DWL 0 0 0 0 0 47 47
Transfers 0 0 0 0 0 0 0
Total financial 233 0 499 48 0 75 855
Total inc. BoD 1,405 0 499 48 0 75 2,027
Cancer
BoD 3,542 0 0 0 0 0 3,542
Health System 21 0 50 23 0 13 107
Productivity 136 0 80 0 2 0 218
Carers 0 2 1 0 0 0 3
DWL 0 0 0 0 0 51 51
Transfers -21 -2 24 0 0 0 0
Total financial 154 2 157 23 2 66 403
Total inc. BoD 3,696 2 157 23 2 66 3,945
Total
BoD 17,246 0 0 0 0 0 17,246
Health System 172 0 403 189 0 109 873
Productivity 911 0 777 0 5 0 1,693
Carers 15 674 114 0 0 0 804
DWL 0 0 0 0 0 358 358
Transfers -37 -60 97 0 0 0 0
Total financial 1,095 618 1,393 189 5 468 3,767
Total inc. BoD 18,340 618 1,393 189 5 468 21,013
Of total obesity costs, 87.3% are borne by the individual, 2.9% by family and friends, 6.6% by
Federal Government, 0.9% by State Government, 0.0% by employers and 2.2% by the rest of
society (Figure 7-2).
99
FIGURE 7-2: TOTAL COSTS OF OBESITY BY BEARER, 2005 (% TOTAL)

0.0%
0.9% 2.2%
6.6%
Individuals
2.9%
Family/Friends
Federal Government
State Government
Employers
Society/Other
87.3% Total = $21.0 billion
However, of the financial costs only, the respective shares are 29.1%, 16.4%, 37.0%, 5.0%,
0.1% and 12.4% (Figure 7-3).
FIGURE 7-3: FINANCIAL COSTS OF OBESITY BY BEARER, 2005 (% TOTAL)
12.4%
Individuals
0.1%
5.0%
Federal Government
State Government
Employers
Society/Other
37.0%
16.4%
Total = $3.767billion
7.2 SUMMARY OF THE COST OF DIABETES

The economic cost of diabetes is summarised in Table 7-2 and Figure 7-4.
The total financial cost of diabetes in 2005 was estimated as $10.3 billion (of which
$1.1 billion was due to obesity).
Of this, carer costs (at replacement valuation) were estimated as $4.4 billion
(43%), productivity losses were $4.1 billion (40%) and health system costs were
$1.1 billion (10%).
100
DWL from transfers (taxation revenue foregone, welfare and other Government
The net cost of lost wellbeing was valued at a further $11.7 billion, bringing the total cost
of diabetes in 2005 to $21.3 billion.
TABLE 7-2: COST SUMMARY, TYPE 2 DIABETES ($M) 2005

BoD 11,714 0 0 0 0 0 11,714
Health System 211 0 495 231 0 134 1,072
Productivity 2,553 0 1,497 0 29 0 4,079
Carers 0 4,204 213 0 0 0 4,417
DWL 53 12 0 0 0 0 65
Transfers 0 -168 168 0 0 0 0
Total financial 2,817 4,048 2,373 231 29 834 10,333
Total inc. BoD 14,531 4,048 2,373 231 29 834 21,282
FIGURE 7-4: FINANCIAL COST OF DIABETES BY TYPE, 2005 (% TOTAL)
6.8% 10.4%
0.6%
Health system
Productivity
Carers
DWL
Other indirect
42.7%
39.5%
Total = $10.33 billion
Of total diabetes costs, two thirds are borne by the individual, 18% by family and friends, 11%
by Federal Government, 1% by State Government, 0.1% by employers and 3.8% by the rest of
society (Figure 7-5).
101
FIGURE 7-5: TOTAL COST OF DIABETES BY BEARER, 2005 (% TOTAL)

0.1% 3.8%
1.0%
10.8% Individuals
Family/Friends
Federal Government
State Government
Employers
18.4%
Society/Other
65.9%
Total = $21.3 billion
However, of the financial costs only, the respective shares are 27.3%, 39.2%, 23.0%, 2.2%,
0.3% and 8.1% (Figure 7-6).
FIGURE 7-6: FINANCIAL COST OF DIABETES BY BEARER, 2005 (% TOTAL)
0.3% 8.1%
2.2% Individuals
Federal Government
State Government
23.0%
Employers
Society/Other
39.2% Total = $10.33 billion
7.3 THE MICROECONOMICS OF OBESITY

Economic theory suggests that market failure can occur when socially optimal outcomes differ
from privately optimal ones. In the case of obesity, social costs exceed private costs due to
economic “externalities” (factors that the private individual does not take into account when
assessing costs and benefits).
People have a ‘positive time preference’, preferring current consumption and discounting
the future impacts.
102
People may have imperfect information, being unaware of future risks and not
necessarily behaving rationally, even when aware.
Many costs are not borne by the individual (individuals bear only 29% of the financial
costs – recall Figure 7-3), so these are not included in private planning.
Thus at any given cost of obesity, private individuals will be more obese than socially
desirable. The flipside of this is that the marginal social benefits of obesity prevention exceed
the private benefits, so at any given cost of obesity prevention, society would choose more
prevention than private individuals. Economic theory suggests that when this occurs there are
essentially three options:
tax or otherwise disincentivise obesity (or its causes) where it imposes social costs;
subsidise obesity prevention activities to bring private costs in line with social ones
(eg, awareness campaigns); and/ or
regulate.
FIGURE 7-7: EXTERNALITIES AND THE CASE FOR INTERVENTION
Left panel: Taxation of obesity sources Right: Subsidy / funding of obesity prevention activities
Before looking at these options, is should be noted that if the sources of externalities can be
removed, the private sector would adjust to a more socially optimal equilibrium. This might
occur, for example, if people bore more of their own health system costs, which may eventuate
more in the future as public health dollars become increasingly scarce, although the many
issues in relation to this will not be addressed further in this report.
Providing better information may also help to redress externalities. For example, food labelling
such as the National Heart Foundation ‘healthy red tick’ can assist in identifying foods lower in
fat or sugar, or higher in fibre.
A final point is that the private sector can also be influenced, through discussions with public
policy makers, to embark on voluntary codes or independent initiatives to help switch demand
patterns (see Section 7.4.2.). For example, McDonalds Australia has responded favourably to
a Federal Government call for labelling with nutritional information on packaging in 2002 by
including nutritional information on in-store packaging, producing a range of healthy alternative
options (including salads) and cutting back children’s TV advertising by 40% (Barnett, 2005).
Indeed, in many cases the private sector is leading the way. Weight loss programs such as
Jenny Craig and healthier food and beverage alternatives such as SubWay, Pepsi Max, Coke
Zero, juice bars and sushi stands are gaining market share, without public subsidy, driven by
demand for healthy alternatives. There may be great scope for innovative initiatives in this
vein.
103
This report has noted the substantial DWLs associated with any government intervention, so
options for pursuing partnerships with the private sector or intervening in light-handed ways
(where the cost effectiveness of interventions is established) are likely to be more optimal than
poorly conceived (costly or ineffective) interventions for the sake of being seen to be ‘doing
something’ or ‘starting somewhere’. Calling for taxation of certain foods and beverages is a
case in point, as is demonstrated in the next section.
7.4 TAXATION, REGULATION AND SUBSIDISED PROGRAMS
7.4.1 TAXATION
One suggestion that is periodically floated in relation to addressing obesity is the proposal for
an (additional) indirect tax imposed on selected food products considered to have unhealthy
characteristics – such as a high content of fat, sugar and/or salt. Proponents of such a ‘fat tax’
also support it as a revenue-raising measure, with the revenue to be used to prevent obesity in
other ways (eg healthy diet promotion). At least to a significant degree, these twin objectives
are internally inconsistent due to the nature of demand elasticities: if a tax is effective in
switching consumption away from a product, it will not raise much revenue; alternatively, if it is
a good revenue raiser, this is because it fails to switch demand. Excise on cigarettes, petrol
and other goods with inelastic demand curves, are primarily useful for revenue raising.
Access Economics does not rule out the possibility that differential or discriminatory taxation
has a role to play in modern economies. But it does start from the three premises that
deviations from the ‘broad base/low uniform rate’ design ideal for taxation of goods and
services must:
1 be justified by demonstrating a clear case in principle for such deviations;
2 in practice, be an efficient and effective way of promoting the specified objectives; and
3 that the side-effects of the deviations are an acceptable cost to the community.
The ‘fat tax' does not meet any of these taxation design criteria. The proposal is flawed in a
number of ways.
A fat tax would be an ill-targeted, blunt instrument for improving health outcomes, since
it is not targeted at obese people, but rather at food products consumed in some degree
by all people. Implicit in this prescription is the assumption that the type of food
consumed is a significant contributor to the obesity problem, rather than the amount.
A fat tax implicitly assumes that higher taxation of such products will substantially shift
the pattern of food consumption away from them and towards other, more healthy,
foods. However, if the demand for these goods is relatively inelastic (ie, quantities
consumed change only in minor ways for any unit price change), then this will not be the
case, and the only or main impact will be to increase the price of the product.
An example of this principle is that tap water is an essentially free and healthy
beverage in Australia. However, its close-to-zero price does not necessarily
induce people to drink water rather than (more expensive) carbonated soft drinks.
People’s preferences may not be greatly affected by taxation or other interventions
to change prices.
Moreover, concessional GST-free status is already extended to certain fresh food
while more processed food including most snack food, takeaway and restaurant
food is taxed at the normal rate of 10%. There is no evidence that this
concessional treatment is effective in switching demand towards GST-free items.
104
Given that demand for food generally is relatively insensitive to price movements, as
shown in various econometric analyses (eg, Huang and Lin, 2000; You et al, 1998;
Department for Environment, Food and Rural Affairs, 2000); the fat tax is more likely to
raise revenue than change consumption patterns.
As such, a fat tax would need to be imposed at very high rates to change
consumption patterns substantially and to have any significant impact on health
outcomes. The cost in terms of equity and tax system simplicity would be
unacceptably high.
Moreover, a fat tax would be regressive, hurting the poor proportionately more than the
rich.
Poor people who are not obese would be disadvantaged by the tax, even though
their moderate consumption of ‘bad’ foods is not causing them to be obese,
particularly if they exercise responsibly also.
Such a tax would be extremely complex to design and administer, imposing significant
compliance costs on the community and administrative costs on the Australian Taxation
Office. This was also experienced with the GST, creating numerous anomalies.
A 20% ad valorem fat tax on cheesecake would mean that a $10 cheesecake with
20 grams of fat would attract a tax of $2 but a $20 cheesecake with 2 grams of fat
would attract a tax of $4.
If a fat excise, sugar excise and salt excise were all introduced then there would be
up to 15 different tax rates potentially applicable to any food item.
Any new public health campaigns would be best funded from the budget (and out of
general revenue) rather than through the imposition of a fat tax.
Consider, in contrast, a tax on luxury cars aimed at reducing imports for balance of payments
reasons. The tax would be efficient and effective, since luxury cars have elastic demand
curves; progressive, taxing the rich proportionately more than the poor; and relatively simple to
administer (That said, there might be other reasons to avoid such a tax).
7.4.2 REGULATION
Australia’s regulatory environment in relation to advertising of food products is very similar to
that of other OECD countries. The Australian Broadcasting Tribunal’s Children’s Television
Standards that came into effect on 1 January 2006 require television stations, in addition to
program content requirements, to:
1 Not broadcast more than 13 minutes per hour of advertising during children’s viewing
time;
2 Not broadcast any advertisements or offer any prizes during preschool children’s viewing
periods;
3 Not broadcast an advertisement for a food product that contains any misleading or
incorrect information about the nutritional value of a product; and
4 Not broadcast advertisements for any alcoholic drink during children’s viewing times.
In addition, the Australian Association of National Advertisers established a voluntary Code of

Practice for Advertising to Children which came into effect in 2005 and which aims to “not
encourage or promote an inactive lifestyle combined with unhealthy eating or drinking habits”
as well as undertaking that advertising “must not contain any misleading or incorrect
105
information about the nutritional value of that product”27. Further guidelines (draft as of July
2006 on the AANA website28) states that advertisements directed towards children for food
and/or beverage products shall:
be designed to be understood by those children, and shall not mislead or deceive, nor
employ ambiguity or a misleading sense of urgency, nor feature inappropriate price
minimisation;
not seek to exploit children’s imagination in ways which might reasonably be regarded
as an intent to mislead about nutritional values which might be associated with
consumption of the product/s;
not state nor imply that a product makes children who own or enjoy it superior to their
peers;
not seek to undermine parents and/or other adults responsible for a child’s welfare in
their role of guiding diet and lifestyle choices;
not include any direct appeal to children to urge parents and/or other adults responsible
for a child’s welfare to buy particular products for them;
not use broadcast or print media personalities (live or animated) to sell products,
premiums or services without clearly distinguishing between commercial promotion and
programme or editorial content.
The accompanying practice note inclusions state that:
Although children may be expected to exercise some preference over the food
they eat or drink, advertisements must be prepared with a due sense of
responsibility and should not directly advise or ask children to buy or ask their
parents or other adults to make enquiries or purchases. Nothing in an
advertisement may seem to encourage children to pester or make a nuisance of
themselves. This extends to behaviour shown: for example, a child should not be
shown asking for a product or putting it into the parent’s trolley in the supermarket.
Phrases such as “Ask Mummy to buy you” are not acceptable. Advertisements
must not encourage or condone damaging oral health care practices.
The only exceptions to this current level of regulation in the OECD are Sweden’s Radio and
Television Act, which prohibits commercials that intentionally attract the attention of children
under 12 years and Quebec’s Consumer Protection Act, which was passed in 1978 says that,
“…no person may make use of commercial advertising directed at persons under thirteen
years of age.” (Clause 248)29
Health Minister Tony Abbott doubts the effectiveness of such regulation:
“Advertisements do have some influence on behaviour, otherwise people wouldn’t

pay for them. Still, banning food ads to children is a tokenistic pseudo-solution
that’s been proven not to work. Quebec banned food advertising to children 25
years ago and Sweden banned it 12 years ago without any appreciable impact on
27
http://www.aana.com.au/pdfs/A2CCode.pdf accessed 29 September 2006.
28
http://www.aana.com.au/pdfs/New_Food_Code_V6-5.pdf#search=%22AANA%20Child%20Code%20of%20Ethics%22
accessed 29 September 2006.
29
http://www.canlii.org/qc/laws/sta/p-40.1/20050513/whole.html accessed 29 September 2006.
106
obesity rates. In this area, bans are the soft option for governments more
interested in looking good than doing good…
“I'm a reluctant regulator. Regulation is something we do when absolutely

necessary as a last resort, when there is a clear benefit, when the benefits of doing
something fairly clearly outweigh the potential cost, including all the transitional
costs, then you consider new governmental programs, new governmental
regulations. (Abbott, 2006)
Proponents of regulation are unable to establish high level evidence of the efficacy or cost
effectiveness of the intervention (eg, Denniss, 2006) and, clearly, crude prevalence rates from
a sample size of two would not be a valid indicator. That said, it has been well established
that advertising is an effective marketing tool; the problems with this approach lie more in its
red tape costs and interventionist nature, the movement of the industry towards self-regulation,
as well as implementation issues, such as ensuring effective banning of advertising including
through other electronic media such as the internet or video games.
Another similarly ‘heavy handed’ approach is the introduction of implicit disincentives or

quantitative restrictions on the provision of services. For example, treating obesity as a risk
factor penalised in access to public health services (eg, hospital queues) or in private health
insurance premiums. However, the continuous risk and aetiology of obesity is more complex
than just being a lifestyle issue, and such regulation may lead to harsh thresholds from an
equity perspective, potentially forbidding hospital access to a child with BMI just in excess of a
threshold but who may also have Type 1 diabetes.
7.4.3 SUBSIDISED PROGRAMS – A LIGHTER HAND

While larger scale subsidies, such as for fruit and vegetables, raise many of the same (but
mirror image) issues as the fat tax, adopting a more gentle approach by funding awareness,
health promotion and prevention campaigns has been the preferred policy approach to date,
including:
GP management plans and team care plans;
health checks;
after-school exercise programs;
the Australian Better Health Initiative; and
joint partnerships (private sector provided, publicly subsidised) eg "Jump Rope for
Heart".
Subsidising participation in private weight loss or behavioural change programs, even with GP
prescription and co-payment arrangements are likely to be very expensive, potentially in the
order of $50 million per annum (Koutsoukis, 2005). Moreover, like the private health insurance
rebate, most of the money is spent on the ‘converted’, rather than on changing behaviour at
the margins.
As with all health interventions, a key consideration is a body of evidence for demonstrated
cost effectiveness of the intervention. A number of sport/exercise programs, in particular for
children, have been reviewed in this respect.
The Planet Health evaluation (Wang et al, 2003) was shown to cost less per QALY than
many adult treatments, providing evidence that early prevention and intervention are
better than later treatment. Planet Health was an interdisciplinary school-based
intervention to reduce obesity among middle school students (vs control schools) which
107
resulted in reductions in BMI and triceps skinfold among girls (although these outcome
measure reductions were not statistically significant in boys). The cost effectiveness of
US$4,305/QALY (similar to hypertension treatment, low-cholesterol-diet therapies,
diabetes screening programs and adult exercise programs) was notable in that it was the
first economic evaluation of an obesity-reduction program.
Although cost effectiveness analysis is relatively new in this field, efficacy studies have
been undertaken for many decades. Examples of other efficacious interventions for
obesity have included: a randomised controlled trial in relation to reduced TV viewing
(Robinson, 1999), a dance program (Flores, 1995), a school-based cardiovascular
program for Year 10 students (Killen et al, 1988), a daily physical activity program
(Dwyer et al, 1983) and a health education program (Lionis et al, 1991).
A number of the latter interventions were reviewed as part of a Cochrane review of 22 studies
(randomised controlled trials and controlled clinical trials with minimum duration of twelve
weeks) of interventions for preventing obesity in children (Summerbell et al, 2005), which
concluded that:
“some studies that focused on dietary or physical activity approaches showed a

small but positive impact on BMI status… There is not enough evidence from trials
to prove that any one particular programme can prevent obesity in children,
although comprehensive strategies to address dietary and physical activity
change, together with psycho-social support and environmental change may help.
There was a trend for newer interventions to involve their respective communities
and to include evaluations.”
Cost effectiveness criteria are also of fundamental importance in relation to other

interventions, to which we now turn.
7.5 PHARMACOLOGICAL AND SURGICAL INTERVENTIONS

The main current treatment of patients who are overweight or obese is a suitable diet and
exercise programme with appropriate support and encouragement. The previous section
provided some evidence that combined lifestyle interventions, increased physical activity and
behaviour therapy are relatively successful therapies for weight loss and weight maintenance.
However, for severely obese people, these interventions may be more limited in efficacy since
maximum weight loss rarely exceeds 10% of initial body weight, drop-out rates are often high
and a substantial proportion of people who lose weight regain it once dismissed from therapy
(Kravis, 2005). This section briefly reviews pharmacotherapies and surgeries.
7.5.1 PHARMACOTHERAPY
Pharmacotherapy is generally efficacious when combined with lifestyle therapy for obese
adults with BMI>30 kg/m² and no other risk factors or BMI>27 kg/m² with risk factors, showing
average weight loss (Kravis, 2005) of 3.9-10.3kg at 1 year and 5.0-7.4kg at 2 years; if
efficacious and free of side effects, a drug may be continued long-term. However, longer term
safety and efficacy are not yet well established.
Two common anti-obesity drugs (orlistat or sibutramine) can be used to aid weight-loss in
people with BMI and comorbidities as above and who have managed to lose some weight on
diet alone. Orlistat prevents absorption of fat, while sibutramine hydrochloride inhibits re-
uptake of non-adrenaline and serotonin, promoting the feeling of having eaten enough.
108
The UK’s National Horizon Scanning Centre (2004) reviewed rimonabant (Acomplia), a
selective cannabinoid receptor blocker, following its Phase III clinical trials in late 2004, for
weight loss and maintenance of weight loss, as an aid to smoking cessation and maintenance
of abstinence, and for the management of metabolic disorders related to being overweight or
obese. Rimonabant is the first drug to target factors governing the body's appetite,
metabolism and energy use. Randomised controlled trial data are from the RIO-Lipids
international multi-centre, double-blind study of 1,036 overweight or obese patients with
dyslipidaemia and a BMI between 27 and 40 kg/m2 (Dale and Anthenelli, 2004; Despres and
Sjostrom, 2004). Patients were randomised to either rimonabant 5mg or 20mg or placebo with
a reduced calorie diet for one year.
Patients treated for one year with rimonabant 20mg daily lost 8.6kg versus 2.3kg on
placebo (p<0.001).
The trial found that 58.4% of patients on 20mg lost over 5% of their body weight when
treated for one year with rimonabant 20mg (p<0.001 versus placebo) as compared to
30% of patients on rimonabant 5mg (p=0.001 versus placebo), and 19.5% of patients in
the placebo group.
Thirty-three percent (p<0.001 versus placebo) of patients lost more than 10% of their
body weight with rimonabant 20mg versus 10.6% on rimonabant 5mg, and 7.2% on
placebo.
Rimonabant had positive effects on abdominal obesity, lipid profiles, insulin sensitivity,
adipokines (adiponectin) and inflammatory markers. 30
In 787 smokers, rimonabant at a dose of 20mg doubled the odds of quitting smoking
versus placebo, with an absence of weight gain in successful quitters.
The main adverse effects were nausea and vomiting, dizziness and headache which
affected 18.9% of patients on rimonabant 20 mg.
Rimonabant was launched in the UK in June 2006 and costs £55/ month.
Relative efficacy of weight loss compared to other drugs is summarised below (Kravis, 2005):
Sibutramine: 4.45 kg at 12 months
Orlistat: 2.89 kg at 12 months
Phentermine: 3.6 kg at 12 months
Diethylpropion:3.0 kg at 6 months
Fluoxetine: 3.15 kg at 12 months
Bupropion: 2.8 kg at 6 to 12 months
Rimonabant: 7.2 Kg at 12 months (European studies promising)
Zonisamide: 9.2 Kg at 9 months
30
On average there was a waist circumference reduction of 9.1cm in patients who completed treatment for a year
with rimonabant 20mg (p<0.001 versus placebo). There was an average increase of 23% in HDL-cholesterol
(p<0.001 versus placebo), and an average reduction of 15% in triglycerides in completers (p<0.001 versus
placebo). There was a positive shift in LDL particle size, with a reduction (p=0.002 vs. placebo) in the proportion of
smaller dense atherogenic LDL particles, and an increase (p<0.001 versus placebo) in the proportion of larger, less
atherogenic LDL particles. C-reactive protein and leptin fell significantly and adiponectin increased. Metabolic
syndrome criteria were met in 53% of the 20mg rimonabant groups, falling to 26% after 1 year (p<0.001 versus
placebo).
109
7.5.2 SURGERY
For patients with morbid obesity, defined as >40 kg/m2, or between 35 kg/m2 and 40 kg/m2 in
the presence of significant co-morbidities, surgery may aid weight reduction and or maintain it.
There are two main types of surgery, malabsorptive and restrictive. Malabsorptive surgery
involves bypassing part of the gastrointestinal tract to reduce absorption of food. Restrictive
surgery reduces the size of the stomach to give the person the experience of fullness.
Bariatric surgery may result in weight loss for carefully selected patients with clinically severe
obesity (BMI>35 with comorbid conditions), when less invasive methods of weight loss have
failed and the patient is at high risk for obesity associated morbidity/mortality. Bariatric bypass
surgery produces more weight loss than gastroplasty, has a mortality rate <1-5% but
complication costs are high (Macgregor and Rand, 1993). Nonetheless, in the US, the
American Society of Bariatric Surgeons note that gastric bypass surgery expenditure has
quadrupled in the last five years to $2.5 billion.
Craig and Tseng (2002) reviewed the cost effectiveness of gastric bypass vs. no treatment for
relatively healthy women and men aged 35-55 years (BMI 40-50 kg/m²). Their study did not
include severely obese patients with chronic medical conditions for whom the surgical risk as
well as the benefits of weight loss would be greater. The cost effectiveness ratio of bypass vs.
no treatment was favourable with ratios ranging from $5,000-$16,000/QALY for women and
$10,000-$35,000/QALY for men (depending on age and initial BMI). However, interestingly
because reduction in lifetime medical cost was not greater than the cost of treatment in any
subgroup, “gastric bypass was not cost saving from the payer perspective”.
A recent Cochrane review (Clegg et al, 2003) of clinical and cost effectiveness of surgery for
people with morbid obesity concluded:
Surgery resulted in a significantly greater loss of weight (23-37 kg more weight)

than non-surgical treatment, which was maintained to 8 years and led to
improvements in quality of life and comorbidities. The economic evaluation of
surgery compared with nonsurgical management suggested that surgery was cost
effective at £11,000 per QALY. Comparisons of the different types of surgery were
equivocal.
7.5.3 CONCLUDING COMMENTS

A number of factors make it difficult to evaluate the best interventions to address obesity. The
limited number of randomised controlled trials with long term follow up and the paucity of cost
effectiveness data constrain the evaluator’s ability to clearly identify and compare the relative
value and effectiveness of individual and combined weight loss programs over substantial
periods (five years or more). Combined lifestyle modification in severely obese adults can
achieve weight loss and a reduction in comorbidities in some patients but does not always
achieve sustained significant long-term weight loss, while many patients regain weight. There
is some evidence that workplace weight management programs are effective substitutes for
physician-directed programs. Pharmacotherapy combined with lifestyle modifications may
achieve weight loss and improve comorbidities, but longer term effectiveness has not yet been
fully established, although efficacy is very promising. Surgery for selected patients, with
appropriate follow-up care, has achieved weight loss and improvement of some comorbidities
and quality of life at five years, but risks are higher. Bariatric surgical technology has
improved, stimulating new consumer demand which is expected to increase.
110
REFERENCES
Abbott A (2006) Speech by the Minister for Health and Ageing, Tony Abbott, to the National
Press Club [online], Department of Health and Ageing, Available from:
http://www.health.gov.au/internet/ministers/publishing.nsf/Content/health-mediarel-
yr2006-ta-abbsp020806.htm?OpenDocument&yr=2006&mth=8 accessed 29 September
2006 [2 August].
Access Economics (2006a) Listen Hear! The economic impact and cost of hearing loss in
Australia, CRC HEAR and the Victorian Deaf Society (Vicdeaf), May, Canberra.
Access Economics (2006b) Cost Benefit Analysis of fortifying the food supply with folic acid,
Food Standards Australia New Zealand, May, Canberra.
Access Economics (2006c) The Economic Cost of Cancer in New South Wales, Cancer
Council NSW, forthcoming, Canberra.
Access Economics (2005a) The shifting burden of cardiovascular disease in Australia, The
National Heart Foundation of Australia, May, Canberra.
Access Economics (2005b) Arthritis – the bottom line: The economic impact of arthritis in
Australia, Arthritis Australia, January, Canberra.
Access Economics (2005c) The economic value of informal care in Australia, Carers Australia,
August, Canberra.
Access Economics (2004a) Costs Of Workplace Injury And Illness To The Australian
Economy: Reviewing The Estimation Methodology And Estimates Of The Level And
Distribution Of Costs, The National Occupational Health And Safety Commission.
Access Economics (2004b) The Cost of Domestic Violence to the Australian Economy.
American Diabetes Association (2003) “Economic Costs of Diabetes in the US in 2002”

Diabetes Care 26: 917-932.
Australian Bureau of Statistics (2006a) Diabetes in Australia: A Snapshot, 2004-05, Cat No

4820.0.55.001, Canberra.
Australian Bureau of Statistics (2006b) Causes of Death, 2004, Cat No 3303.0, Canberra (and
previous issues).
Australian Bureau of Statistics (2005a) National Health Survey: Summary of Results 2004-05,
Cat No 4364.0, Canberra.
Australian Bureau of Statistics (2005b) Employee Earnings and Hours, Australia, Cat No
6306.0, Canberra.
Australian Bureau of Statistics (2003a) Population Projections, Australia, 2002 to 2101, Cat No
3222.0, September, Canberra.
Australian Bureau of Statistics (2003b) Survey of Disability, Ageing and Carers, Cat No
4430.0.
111
Australian Bureau of Statistics (2002) National Health Survey: Summary of Results 2001, Cat
No 4364.0, Canberra.
Australian Bureau of Statistics (2000) Working Arrangements, Cat No 6342.0.
Australian Institute of Health and Welfare (2006) Australia’s Health 2006, AIHW.
Australian Institute of Health and Welfare (2005a) Health Expenditure Australia 2003-04,
AIHW, Cat No HWE 32, September, Canberra.
Australian Institute of Health and Welfare (2005b) Health system expenditure on disease and
injury in Australia, 2000-01, AIHW, 2nd edition, Cat No HWE 28, April, Canberra.
Australian Institute of Health and Welfare (2005c) Health system expenditures on cancer and
other neoplasms in Australia, 2000–01, AIHW, Cat No HWE 29, May, Canberra.
Australian Institute of Health and Welfare (2004) Heart, stroke and vascular diseases –
Australian Facts 2004, AIHW and National Heart Foundation of Australia
(Cardiovascular Disease Series No 22), Cat No CVD 27, Canberra:
Australian Institute of Health and Welfare (2003a) Australia’s Welfare 2003, AIHW, Cat No
AUS 41, November, Canberra.
Australian Institute of Health and Welfare (2003b) Australian Hospital Statistics 2002-03,
AIHW, Cat No HSE 32, Canberra.
Australian Institute of Health and Welfare (2002) Diabetes: Australian Facts 2002, AIHW,
Canberra.
Australian National Audit Office (2005) Helping Carers: the National Respite for Carers
Program 2004-05, Audit Report No.58, Performance Audit.
Barnett G (2005) Australia and the Obesity Epidemic – Past and Future, Australian Senator
Guy Barnett to the First International Congress on Pre-diabetes and the Metabolic
Syndrome, 14 April, Berlin.
Barr E, Magliano D, Zimmet P, Polkinghorne K, Atkins R, Dunstan D, Murray S, Shaw J (2006)

AusDiab 2005: The Australian Diabetes, Obesity and Lifestyle Study – Tracking the
accelerating epidemic: its causes and outcomes, International Diabetes Institute,
Melbourne.
Booth M, Okely AD, Denney-Wilson E, Yang B, Hardy L, Dobbins T (2006) NSW schools
physical activity and nutrition survey (SPANS) 2004, NSW Department of Health,
Sydney.
Booth M, Chey T, Wake M, Norton K, Hesketh K, Dollman J, Robertson I (2003) “Change in

the prevalence of overweight and obesity among young Australians, 1969-1997.”
American Journal of Clinical Nutrition, 77: 29-36.
Booth M, Wake M, Armstrong T, Chey T, Hesketh K, Mathur S (2001) “The epidemiology of

overweight and obesity among Australian children and adolescents, 1995-97” Australian
and New Zealand Journal of Public Health, 25(2): 162-9.
112
Bradley C, Oberst K, Schenk M (2005) “Absenteeism from Work: The Experience of Employed
Breast and Prostate Cancer Patients in the Months Following Diagnosis” Psycho-
Oncology, 15(8): 739-47.
Brameld K, Holman D, Threlfall T, Lawrence D, De Klerk N (2002) “Increasing ‘active

prevalence’ of cancer in Western Australia and its implications for health services”
Australian and New Zealand Journal of Public Health, 26(2): 164-9.
Brouwer W, van Exel N et al (2004) “Burden of Caregiving: Evidence of Objective Burden,

Subjective Burden, and Quality of Life Impacts on Informal Caregivers of Patients with
Rheumatoid Arthritis” Arthritis Care & Research, 51(4): 570-7.
Bureau of Transport Economics (2000) Road Crash Costs in Australia, Bureau of Transport
Economics, Report 102, Canberra.
Cameron A, Wellborn T, Zimmet P, Dunstan D, Owen N, Salmon J, Dalton M, Jolley D, Shaw

J (2003) “Overweight and obesity in Australia: the 1999-2000 Australian Diabetes,
Obesity and Lifestyle Study (AusDiab)” Medical Journal of Australia, 178: 427-32.
Carey VJ, Walters EE, Colditz GA, Solomon CG, Willett WC, Rosner BA (1997) “Body fat
distribution and risk of non-insulin-dependent diabetes mellitus in women. The Nurses’
Health Study’”, American Journal of Epidemiology, 145: 614–9.
Center for Disease Control (2004) Targeting Arthritis: The Nation’s Leading Cause of Disability
[online], National Center for Chronic Disease Prevention and Health Promotion,
Available on: www.cdc.gov/nccdphp/promising_practices/arthritis/.
Clegg A, Colquitt J, Sidhu M, Royle P, Walker A (2003) “Clinical and cost effectiveness of
surgery for morbid obesity: a systematic review and economic evaluation” International
journal of obesity, 27(10): 1167-77.
Cohn R, Goodenough B, Foreman T, Suneson J (2003) "Hidden Financial Costs in Treatment

for Childhood Cancer: An Australian study of Lifestyle Implications for Families
Absorbing Out-of-Pocket Expenses" Journal of Pediatric Hematology and Oncology,
25(11).
Colagiuri S, Colagiuri R, Conway B, Grainger D, Davey P (2003) DiabCo$t Australia:

Assessing the burden of Type 2 Diabetes in Australia, Diabetes Australia, Canberra.
Colditz GA, Willett WC, Rotnitzky A, Manson JE (1995) “Weight gain as a risk factor for clinical
diabetes mellitus in women” Annals of Internal Medicine, 122: 481–6.
Colditz GA, Willett WC, Stampfer MJ, Manson JE, Hennekens CH, Arky RA (1990) “Weight as
a risk factor for clinical diabetes in women” American Journal of Epidemiology, 132:
501–13.
Cole T, Bellizzi M, Flegal K, Dietz W (2000) “Establishing a standard definition for child
overweight and obesity worldwide: international survey” British Medical Journal, 320:
1240-3.
Craig BM, Tseng DS. (2002) “Cost effectiveness of gastric bypass for. severe obesity”
American Journal of Medicine, October14,113: 491-8.
113
Cutler DM, Richardson E (1998) The Value of Health: 1970-1990 [online], JCPR Working
Paper 28, AEA session “What we get for health care spending”, Available from:
www.jcpr.org/wpfiles/value.pdf.
Dale L, Anthenelli R (2004) Rimonabant as an Aid to Smoking Cessation in Smokers

Motivated to quit: Results from a US Multicenter Study – STRATUS-US trial, American
College of Cardiology, New Orleans, March.
Denniss R (2006) Protecting children or protecting profits? The role of television advertising
regulation in the prevention of childhood obesity [online], Australian Greens, Available
from: http://www.bobbrown.org.au/files/speeches/Protecting%20children%20or%20prote
cting%20profits.pdf#search=%22Quebec%20Sweden%20Greens%20obesity%20childh
ood% [29 September].
Department of Health and Ageing (2005) Overview of the National Palliative Care Initiatives,
Department of Health and Ageing, December.
Department of Health and Ageing (2003) Returns on investment in public health: An

epidemiological and economic analysis, Applied Economics.
Department for Environment, Food and Rural Affairs (2000) National Food Survey 2000:
Annual report on food expenditure, consumption and nutrient intakes, United Kingdom.
Despres J-P, Sjostrom L (2004) “Effect of rimonabant on Leptin and Adiponectin levels in
overweight/obese subjects with untreated dyslipidemia: The RIO-Lipids trial” Endocrine
Society’s 86th Annual Meeting, New Orleans, June.
Dockerty J, Williams S, McGee R, Skegg D (2000) “Impact of Childhood Cancer on the Mental
Health of Parents” Medical and Paediatric Oncology, 35: 475-83.
Dunstan D, Zimmet P, Welborn T, Sicree R, Armstrong T, Atkins R, Cameron A, Shaw J,

Chadban S (2001) Diabesity & associated disorders in Australia – 2000: the accelerating
epidemic, The Australian Diabetes, Obesity and Lifestyle Study (AusDiab), International
Diabetes Institute, Melbourne.
Dwyer T, Coonan WE, Leitch DR, Hetzel BS, Baghurst RA (1983) “An investigation of the
effects of daily physical activity on the health of primary school students in South
Australia” Int J Epidemiol, 12: 308-13.
Ebbeling CB, Pawlak DB, Ludwig DS (2002) “Childhood obesity: public-health crisis, common
sense cure” Lancet, 360: 473–82.
Eiser C (1998) “Practitioner Review: Long-term Consequences of Childhood Cancer” Journal

of Child Psychology and Psychiatry, 39(5): 621-33.
Ettinger WH, Burns R, Messier SP (1999) “A randomized trial comparing aerobic exercise and
resistance exercise with a health education program in older adults with knee
osteoarthritis. The Fitness Arthritis and Seniors Trial (FAST)” JAMA, 227(1): 1361-9.
Felson DT, Zhang Y, Anthony JM, Naimark A, Anderson JJ (1992) “Weight loss reduces the
risk of symptomatic knee osteoarthritis in women” Ann Intern Med, 116(7): 535-9.
Flores R (1995) “Dance for health: improving fitness in African American and Hispanic
adolescents” Public Health Rep, 110(1): 189-93.
114
Frisch J (2001) Towards a Disability Allowance: Offsetting the Costs of Disability - An Analysis,
Physical Disability Council of Australia, June.
Hands B, Parker H, Glasson C, Brinkman S, Read H (2004) Physical activity and nutrition
levels in Western Australian children and adolescents: report, Western Australian
Government, Perth.
Health and Safety Executive (1999) The Costs to Britain of Workplace Accidents and Work-
Related Ill Health in 1995-96, HSE Books, Second edition.
Hillman DR, Scott Murphy A, Antic R, Pezzullo ML (2006) “The Economic Cost of Sleep
Disorders” Sleep, 29(3): 299-305.
Howard BV, Ruotolo G, Robbins DC (2003) “Obesity and dyslipidemia. Risk of mortality”
Endocrinol Metab Clin North Am, 32(4): 855-67.
Huang K, Lin B-H (2000) Estimation of Food Demand and Nutrient Elasticities from Household
Survey Data, Food and Rural Economics Division, Economic Research Service, US
Department of Agriculture, Technical Bulletin No 1887, August.
International Diabetes Institute (2001) “Diabesity and Associated Disorders in Australia –

2000: The Accelerating Epidemic” The Australian Diabetes, Obesity and Lifestyle Study
(AusDiab), Melbourne.
International Diabetes Institute (2006) ‘Tracking the Accelerating Epidemic: Its Causes and
Outcomes’, The Australian Diabetes, Obesity and Lifestyle Study (AusDiab 2005),
Melbourne.
Killen JD, Telch MJ, Robinson TN, Maccoby N, Taylor CB, Farquhar JW (1988)
“Cardiovascular disease risk reduction for tenth graders. A multiple factor school-based
approach” JAMA, 260(12): 1728-33.
Koutsoukis J (2005) “Australia mulls picking up the Weight Watchers tab” The Age
13 September.
Kniesner TJ, Leeth JD (1991) “Compensating wage differentials for fatal injury risk in Australia,
Japan and the United States” Journal of Risk and Uncertainty, 4(1): 75-90.
Kravis TC (2005) Obesity and Weight Management: Programs and the Evidence of
Effectiveness, Society of Actuaries, Reden & Anders Ltd, June.
Laing L, Bobic N (2002) Economic Costs of domestic violence: Literature Review, Australian
Domestic and Family Violence Clearinghouse, University of New South Wales, April,
Sydney.
Langa KM, Vijan S, Hayward R, Chernew ME, Blaum CS, Kabeto MU, Weir DR, Katz SJ,
Willis RJ, Fendrick M (2001) “Informal Caregiving for Diabetes and Diabetic
Complications Among Elderly Americans” University of Michigan, Michigan Retirement
Research Centre, Working Paper No wp013.
Lattimore R (1997) Research and Development Fiscal Incentives in Australia: Impacts and
Policy Lessons, OECD Conference on Policy Evaluation in Innovation, 26-27 June,
Paris, 81: 574-7.
115
Lionis C, Kafatos A, Vlachonikolis J, Vakaki M, Tzortzi M, Petraki A (1991) “The effects of a

health education intervention program among Cretan adolescents” Prev Med, 20(6):
685-99.
Macgregor AM, Rand CS (1993) “Gastric surgery in morbid obesity. Outcome in patients aged
55 years and older” Arch Surg, 128(10): 1153-7.
Magarey AM, Daniels LA, Boulton TJC (2001) “Prevalence of overweight and obesity in
Australian children and adolescents: reassessment of 1985 and 1995 data against new
standard international definitions” MJA, 174: 561-4.
Mathers C, Vos T, Stevenson C (1999) The burden of disease and injury in Australia, AIHW,
Cat No PHE17, Canberra.
Miller M, Boywer M, Butow P, Gattellari M, Dunn S, Childs A (1998) “The use of unproven
methods of treatment by cancer patients: Frequency, expectations and cost” Support
Care Cancer, 6: 337-47.
Miller P, Mulvey C, Norris K (1997) “Compensating differentials for risk of death in Australia”
Economic Record, 73(223): 363-72.
Minor MA, Allegrante JP (1997) “Exercise in the management of osteoarthritis of the hip and
knee” Osteoarthritis: Public Health Implications for an Aging Population, Baltimore,
Johns Hopkins University Press.
Murphy KM, Topel R (1999) The Economic Value of Medical Research, University of Chicago
Business School.
Murray C, Lopez A (1996) The Global Burden of Disease: a comprehensive assessment of

mortality and disability from diseases, injuries and risk factors in 1990 and projected to
2020, Global Burden of Disease and Injury Series, Volume 1, Harvard: Harvard School
of Public Health.
Murray C, Lopez A, Mathers C, Stein C (2001) The Global Burden of Disease 2000 Project:
aims, methods and data sources, WHO, Discussion Policy Paper No. 36, November.
National Health and Medical Research Council (2005) National Evidence Based Guidelines for
the Management of Type 2 Diabetes Mellitus, Canberra.
National Health and Medical Research Council (2004) Clinical practice guidelines for the
management of overweight and obesity in adults, 19 March, Canberra.
National Horizon Scanning Centre (2004) “Rimonabant for smoking cessation weight loss and
cardiovascular risk factors of overweight/obesity” New and Emerging Technology
Briefing, The University of Birmingham, July.
Njolstad I, Arnesen E, Lund-Larsen PG (1998) “Sex differences in risk factors for clinical
diabetes mellitus in a general population: a 12-year follow-up of the Finnmark Study”
American Journal of Epidemiology 147: 49–58.
Nordhaus W (1999) The Health of Nations: The Contribution of Improved Health to Living
Standards [online], Lasker/Funding First, Department of Economics, Yale University,
Available from: www.laskerfoundation.org/reports/pdf/healthofnations.pdf [2 April 2003].
116
O’Shea E (2000) The costs of caring for people with dementia and related cognitive
impairments, National Council on Ageing and Older People, Centre for Ageing Studies,
University of Ireland, Galway.
Palliative Care Australia (1998) State of the Nation 1998: Report of National Census of
Palliative Care Services, November.
Petersen S, Peto V, Rayner M (2003) Coronary heart disease statistics: 2003 edition, British
Heart Foundation Health Promotion Research Group, Department of Public Health,
University of Oxford, UK.
Productivity Commission (2006) Report on Government Services 2006.
Productivity Commision (2003) “Evaluation of the Pharmaceutical Industry Investment

Program” Research Report, AusInfo, Canberra.
Queensland Health (2000) Chronic Disease Surveys: Diabetes Prevalence and Management
Report, Brisbane.
Robinson N, Yateman NA, Protopapa LE, Bush L (1989) “Unemployment and diabetes”
Diabetic Medicine, 6(9): 797-803.
Robinson TN (1999) “Reducing Children's Television Viewing to Prevent Obesity: A

Randomized Controlled Trial” JAMA, 282: 1561-7.
Schelling (1968) “The life you save may be your own” in SB Chase (ed) Problems in public
expenditure and analysis, Brookings Institution, Washington DC, 127-62.
Short P, Vasey J, Tunceli K (2005) “Employment Pathways in a Large Cohort of Adult Cancer
Survivors” Cancer, 103(6): 1292-300.
Sloper P (1996) “Needs and responses of parents following the diagnosis of childhood cancer”
Child: care, health and development, 22(3): 187-202.
Stam H, Grootenhuis M, Last B (2005) “The course of life of survivors of childhood cancer”
Psycho-Oncology, 14: 227-38.
Summerbell CD, Waters E, Edmunds LD, Kelly S, Brown T, Campbell KJ (2005) “Interventions
for preventing obesity in children” The Cochrane Database of Systematic Reviews,
Issue 3 Art No: CD001871.pub2. DOI: 10.1002/14651858.CD001871.pub2.
Tolley G, Kenkel D, Fabian R (1994) Valuing Health for Policy: An Economic Approach, The
University of Chicago Press, Chicago.
Tseng Y, Wilkins R (2002) Reliance on Income Support in Australia: Prevalence and

Persistence, Melbourne Institute of Applied Economic and Social Research, Working
Paper No 6/02, May.
Tunceli K, Bradley CJ, Nerenz D, Williams LK, Pladevall M, Lafata JE (2005) “The Impact of
Diabetes on Employment and Work Productivity” Diabetes Care, 28: 2662-7.
Vaska VL, Volkmer R (2002) “Increasing prevalence of obesity in South Australian 4-year-olds:
1995−2002” Journal of Paediatrics and Child Health, 40(7): 353.
117
Viscusi WK (1993) “The value of risks to life and health” Journal of Economic Literature, 13:
1912-46.
Viscusi WK, Aldy JE (2002) The value of a statistical life: a critical review of market estimates
throughout the world, discussion paper no. 392 [online], Harvard Law School,
Cambridge MA, Available from: www.law.harvard.edu/programs /olin_center/.
Von Korff M, Katen W, Lin EHB, Simon G, Ciechanowski P, Ludman E, Oliver M, Rutter C,
Young B (2005) “Work Disability Among Individuals With Diabetes” Diabetes Care,
28: 1326-32.
Walsh J, Chappell P (1999) Cost of Disability Survey Stages 2 and 3 - Demonstration of

Relationship: Severity of Disability v Cost, Department of Family and Community
Services, AGPS, Canberra.
Wang LY, Yang Q, Lowry R, Wechsler H (2003) “Economic Analysis of a School-Based

Obesity Prevention Program” Obesity Research, 11: 1313-24.
Waters EB, Baur LA (2003) “Childhood obesity: modernity's scourge” MJA, 178 (9): 422-3.
You Z, Epperson J, Huang C (1998) Consumer Demand for Fresh Fruits and Vegetables in
the United States, The Georgia Agricultural Experiment Stations, College of Agricultural
and Environmental Sciences, The University of Georgia, Research Bulletin Number 431,
January.
118

Acr 2 D0 F

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Acr 2 D0 F

Uploaded by

Copyright:

Available Formats

October 2006

The economic costs of obesity

Report by Access Economics Pty Limited to

7. Cost Summary and Interventions.............................................................................98

Figure 1-1: Incidence and prevalence approaches to measurement of annual costs 2

Figure 1-1: Incidence and prevalence approaches to measurement of annual costs 2

Figure 5-2: DWL of Taxation 57

Professor Stephen Colagiuri

GLOSSARY OF COMMON ABBREVIATIONS

ABS Australian Bureau of Statistics

PREVALENCE OF OBESITY, 2005 (AUSTRALIANS)

This report estimates that in 2005:

FINANCIAL COSTS OF OBESITY BY TYPE OF COST, 2005 (% TOTAL)

44.9% Total = $3.767billion

FINANCIAL COSTS OF OBESITY: BY WHO BEARS THEM, 2005 (% TOTAL)

COST SUMMARY, OBESITY ($M) 2005

Individuals Family/ Federal State Employ- Society/ Total

1.2 CROSS-CUTTING METHODOLOGICAL ISSUES

All costs should be expressed as present values relative to 2005.

FIGURE 1-1: INCIDENCE AND PREVALENCE APPROACHES TO MEASUREMENT OF ANNUAL COSTS

Annual prevalence costs in the base year = Σ(A + B + C);

1.2.2 CLASSIFICATION OF COSTS

TABLE 1-1: SCHEMA FOR COST CLASSIFICATION

1.2.3 CALCULATING PARAMETERS

1.2.4 DISCOUNT RATES

5.8 - 2.5 - 1.75 = 1.55%.

5.8 - (3.2 - 1.75) - 1.75 = 2.6%.

5.8 – 2.5 = 3.3%

2.1.1 BODY MASS INDEX

Source: ABS (2005a)

TABLE 2-1: AGE AND GENDER-SPECIFIC THRESHOLDS

Source: Cole et al (2000)

Self-reported vs Measured BMI

2.1.2 WEIGHT DISTRIBUTION MEASURES

TABLE 2-2: DEFINITIONS OF OVERWEIGHT AND OBESITY

Measures of weight distribution are technically measures of a subset of obesity known as

2.2.1 EVIDENCE FOR ASSOCIATED HEALTH RISKS

TABLE 2-3: NHMRC - DISEASES ASSOCIATED WITH OBESITY BY CATGORY OF RR

RR Associated with metabolic Associated with excess weight

2.2.2 RISKS FOR CHILDREN

Psychosocial problems are also identified in the literature as an immediate consequence of

2.2.3 THE CAUSAL LINKS

Insulin resistance is also a component of ‘metabolic syndrome’, or ‘syndrome X’,

Dyslipidaemia is a disorder of lipoprotein metabolism, including lipoprotein overproduction or

2.3.1 OBESITY IN THE INDIVIDUAL

Small changes, weighty consequences

According to Ebbeling et al (2002) an energy surplus equivalent to one serve of

2.3.2 OBESITY IN SOCIETY

Energy intake = diet

Energy expenditure = physical activity

2.4 PREVALENCE IN AUSTRALIA

2.4.1 PREVALENCE IN ADULTS

TABLE 2-4: AUSDIAB 1999-2000

Age Group Males Females

2.4.2 PREVALENCE IN CHILDREN AND ADOLESCENTS

TABLE 2-5: PREVALENCE OF OBESITY IN CHILDREN AND ADOLESCENTS, STUDIES

Study Age Number of respondents Prevalence rate of

NNS (Australia;1995) 2-18 1,515 1,447 4.5 5.3

CAPANS (WA; 2003) 7-16 959 876 4.9 4.9

SPANS (NSW; 2004) 5-165 3,128 2,893 7.7 6.1

2.4.3 BASELINE PREVALENCE ESTIMATES

TABLE 2-6: BASELINE PREVALENCE RATES

Age Group Males Females