a study of older Australians with multiple chronic ... - CSIRO Publishing

CSIRO PUBLISHING

Australian Journal of Primary Health, 2013, 19, 144–149 http://dx.doi.org/10.1071/PY12035

Research

Multimorbidity is associated with higher out-of-pocket spending: a study of older Australians with multiple chronic conditions Ian McRae A,E, Laurann Yen B, Yun-Hee Jeon C, Pushpani M. Herath A and Beverley Essue D A

Australian Primary Health Care Research Institute, Ian Potter House, Corner Marcus Clarke and Gordon Streets, The Australian National University, Acton, Canberra, ACT 0200, Australia. B Menzies Centre for Health Policy, Ian Potter House, Corner Marcus Clarke and Gordon Streets, The Australian National University, Acton, Canberra, ACT 0200, Australia. C Faculty of Nursing and Midwifery, The University of Sydney, 88 Mallett Street, Camperdown, NSW 2006, Australia. D The George Institute for Global Health, Level 10, King George V Building, Missenden Road, Camperdown, NSW 2050, Australia. E Corresponding author. Email: [email protected]

Abstract. Most older Australians have at least one chronic health condition. The management of chronic disease is associated with potentially severe economic consequences for patients and their households, partially due to the financial burden associated with out-of-pocket costs for medical and health-related care. A questionnaire was mailed to a crosssectional sample of older Australians in mid-2009, with 4574 responding. Multivariate logistic regression models were developed to investigate the relationships between multimorbidity and out-of-pocket spending on medical and health-related expenses, including the factors associated with severe financial stress among older Australians. We found a positive relationship between number of chronic conditions and out-of-pocket spending on health and that people with multiple chronic conditions tend to be on lower incomes. People with five or more chronic conditions spent on average five times as much on their health as those with no diagnosed chronic conditions and each additional chronic disease added 46% to the likelihood of a person facing a severe financial burden due to health costs. While health policy may minimise out-of-pocket spending for individual conditions, costs compound rapidly for patients with multiple conditions and this burden falls most heavily on those with the lowest incomes. Received 27 March 2012, accepted 24 April 2012, published online 24 May 2012

Background In Australia, over 70% of the burden of disease is attributable to chronic conditions (Australian Institute of Health and Welfare 2010), and 87.5% of total recurrent health expenditure can be attributed to the 12 major chronic disease groups (Australian Institute of Health and Welfare 2006). Previous studies conclude not only that the prevalence of chronic conditions is increasing (Goss 2008) but that the greatest growth is in the prevalence of people with multiple complex chronic diseases (Hwang et al. 2001; Paez et al. 2009). Almost all older Australians have at least one long-term condition and over 80% have three or more longterm conditions (Australian Bureau of Statistics 2009a). The management of multiple conditions increases the economic impact chronic illnesses have on patients and their households due in part to the out-of-pocket health care costs associated with greater health service use. The financial stress caused by these costs in some cases can be severe, leading to an inability to afford necessary health care services, which may compound the health and financial pressures faced by the patients. Journal compilation La Trobe University 2013

In Australia out-of-pocket health costs are substantially mitigated by the health care system, through for example, the Medical and the Pharmaceutical Benefits Schemes and free public hospital services. Analysis of the Australian Household Expenditure Survey of 2003–04 (NATSEM 2008) estimated that health expenses, excluding private health insurance, constituted 3.8% of average household weekly expenditure for households headed by people aged 55 and over. However, overall costs of health care can still be significant for some patients (Doggett 2009). Studies based on the Household Expenditure Survey identify those households that had high total out-of-pocket costs as including: those headed by older Australians; those in the poorest income and wealth quintiles; those without private health insurance; those living in a more socioeconomically disadvantaged area or living in a rural or remote area; and those holding a concession card (Jones et al. 2008; NATSEM 2008). People with multiple chronic conditions with a high out-ofpocket payment burden relative to their incomes are at risk of insufficient health care (Piette et al. 2004; Lehnert et al. 2011). www.publish.csiro.au/journals/py

Multimorbidity and out-of-pocket spending

The Commonwealth Fund has estimated that 18% of the Australian adult population had not seen a doctor when sick or had not received recommended care due to cost, and 12% of Australians had not filled prescriptions or had skipped doses of prescribed medications due to cost (Schoen et al. 2010). A 2010 study of Australian adults also showed a strong relationship between financial stress and failure to purchase medical services and medicines, for example showing 28% of those under financial stress skipping medical tests or treatment compared with 16% of those with no financial stress (Menzies Centre for Health Policy and Nous Group 2009). Consistent with these studies, Australians with chronic obstructive pulmonary disease whose households experienced an out-of-pocket payment burden greater than 10% of household income were significantly more likely to report an inability to pay for basic living or medical expenses (Essue et al. 2011). A recent systematic review (Lehnert et al. 2011) showed that the average health-related expenditure of people with three conditions was 3.3 times that of those with no conditions, and people with five or more conditions spent 3.6 times as much as those with no conditions, leading to concerns that those with many chronic conditions may be facing problems in paying for medical services. While there have been some studies of the costs of chronic illness undertaken in Australia for individual illnesses (Lapsley et al. 2002; Dewey et al. 2004; Gordon et al. 2009) there is a lack of evidence on the relationship between the number of chronic conditions and out-of-pocket costs in older individuals. Similarly, while there has been considerable discussion in the international literature of catastrophic out-of-pocket costs (Wagstaff and van Doorslear 2003; Xu et al. 2003), there are few studies in Australia and those studies (Essue et al. 2011) have tended to be disease specific. The present study had three aims. First, to determine the average amount spent out-of-pocket on medical and healthrelated expenses by older Australians. Second, to determine the relationships between multiple morbidity and out-of-pocket spending in this population. Third, we measured the financial burden associated with out-of-pocket spending and the factors associated with a severe financial burden. Methods Setting and participants The study population included all members of National Seniors Australia, a nation-wide organisation with 285 000 members aged over 50 years. An opt-in invitation and a study questionnaire were mailed to a representative cross-section of their membership base (n = 10 000) during mid-2009. The sample was stratified by age, rurality and state of residence, with those aged 75 years or older over-represented to permit analysis of this older cohort. The survey and study were approved by the Australian National University Human Research Ethics Committee (no. 2009/309). Questionnaire The study questionnaire was a structured instrument that was adapted from one used previously (Essue et al. 2011) and included questions on demographic information, self-reported chronic illness and disability, health service use and out-of-pocket

Australian Journal of Primary Health

145

spending, household economic circumstances and quality of life. The questions were drawn from existing validated tools, including: the Australian 45 and Up study (45 and Up Study Collaborators 2008), the Household, Income and Labour Dynamics in Australia survey (Wooden et al. 2002) and the Quality Metric Short Form Version 12 (SF-12) survey of healthrelated quality of life (Ware et al. 1996). In addition, we measured self-assessed health using the standard question, which is part of the SF-12 set. While information was collected on all conditions that lasted more than 6 months, this study focussed on a sub-set comprising the most common serious chronic diseases in Australia (Australian Bureau of Statistics 2009a), including: cancer, heart disease, diabetes, arthritis and depression and anxiety. Measurement of out-of-pocket spending Respondents were asked to report their personal out-of-pocket expenses over the previous 3 months for five main categories of health-related services, including medication, medical services, transport, home care, medical equipment and other expenses. Respondents reporting ‘Don’t know’ to any category were omitted from calculations of total costs. The ‘Other health-related expenses’ category encompassed a variety of expenses, including physiotherapy, dental care and podiatry. As extreme expenses such as those for housing modifications (one observation over $20 000) and very expensive hearing aids had the potential to significantly influence estimates, observations with quarterly costs of $5000 or over were excluded when estimating costs (removing 26 observations or ~1% of observations reporting total expenditure). Measurement of income Income was adjusted for household size using the modified Organisation for Economic Co-operation and Development equivalence scales to give ‘household equivalent income’ by applying a scale of 1 to the first adult in a household, 0.5 to the second and later adults, and 0.3 to children (Australian Bureau of Statistics 2007a). Midpoints of the ranges of annual household income were used, except for ‘under $20 000’ where a level of $15 000 was used, which approximates the single aged pension, and for the category ‘$150 000 and over’ where $175 000 was used. Measurement of financial burden We measured financial burden as the proportion of equivalised annual household income (adjusted to give quarterly comparison) expended on health-related goods and services. Those who expended more than 10% were considered to face a moderate financial burden from health care costs, and those who expended more than 20% were considered to face a severe financial burden. These percentages are necessarily somewhat arbitrary, and if the information was available it would be desirable to explore the percentage of uncommitted income rather than total income (Xu et al. 2003; Ataguba 2011). Further, proportions of total income do not accommodate the fact that outlaying 10% of income on health costs may be much more ‘catastrophic’ for low-income earners than for a high-income earners (Ataguba 2011). For this

146


reason we report on people who expend more than both 10% and 20% of income on their health.

I. McRae et al.

Table 1.

Demographic characteristics of the study population and the Australian general population Percentage of Percentage of responses (weighted Australian estimates) population

Statistical analysis Univariate analysis of the population shows mean out-of-pocket costs and proportions of the population facing moderate or severe financial stress and their relationship to multiple morbidity. Logistic regression was used to determine the relationship between the likelihood of facing severe financial stress and the number of chronic conditions, allowing for likely confounding variables such as demographic and family background measures, socioeconomic and employment measures and self-assessed health. Household income is not included in the reported regressions. While income is clearly a determinant of whether a person faces financial stress it is also a mathematical component of the dependent variable. Analysis was undertaken using Stata 10 (StataCorp, College Station, TX, USA). All estimates (including regressions) were weighted to adjust totals to the state, age or sex structure of the Australian population and to align with the socioeconomic structure of Australia (Australian Bureau of Statistics 2008). Results Questionnaire response and characteristics of the study population The overall response rate to the survey was 45.7%, with little difference between male and female (45.1% to 46.3%). A higher response was achieved for those aged 65 years and over (48.3%) than for those aged under 65 years (38.9%). Table 1 shows a comparison of the study population with the Australian population. Respondents were better educated, reported better health and were more likely to have private insurance coverage than the average Australian in their age range. This study sample is comparable to the Australian population on all other demographic characteristics. The estimated prevalence of chronic disease in the study population was similar to the Australian population prevalence in this age group, although we found a higher prevalence of high blood pressure and lower prevalence of arthritis. The prevalence of cancer was much higher in the study population if ‘cancers ever diagnosed’ were measured, but similar to national figures for those ‘treated in the last 3 months’. Table 2 shows that 82% of respondents reported at least one chronic disease and over 56% reported more than one chronic disease. Of those aged over 75 years, this increased to 93% with at least one chronic disease and 73% with more than one chronic condition. Out-of-pocket payments and income The mean out-of-pocket expenditure on health care over the previous 3 months was $353 (Table 3), with a median expenditure of $150. Medication and medical services expenses stood out as the major costs, but substantial costs also applied for equipment and other expenses. Out-of-pocket expenditure on medical and health-related purchases increased with the number of chronic conditions experienced, and as the average household income declined (Table 4). The proportion of household equivalised income

EducationA Total without non-school qualifications Total certificate/diploma University degree or higher

44.2 34.7 21.1

63.5 24.8 11.8

Occupation statusB Retired (fully or partially) Working (full-time or part-time) Other

51.3 37.2 11.5

45.1 42.1 12.8

Private health insuranceC Yes No

80.9 19.1

56.5 43.5

Living areaD Major cities Inner regional Australia Outer regional Australia Remote Australia Very remote Australia

62.1 24.1 11.6 1.8 0.4

66.0 22.1 10.1 1.3 0.5

Self assessed health (comparison group aged 65 or over)E Excellent/very good 53.3 Good 32.5 Fair/poor 14.2

36 32 33

ConditionE Cancer Cancer (treated within last 3 months) Heart disease High blood pressure Diabetes Stroke Asthma/hay fever Emphysema/bronchitis Arthritis Osteoporosis Depression/anxiety

5.0 5.0 17.7 29.8 12.1 4.0 9.6 5.2 43.4 12.0 11.8

17.9 4.7 12.3 43.1 12.8 3.2 18.2 3.4 32.2 9.3 15.3

Population age groups 50 years and over or 55 years and over depending on the publication from which they were drawn: AAustralian Bureau of Statistics 2007b; BAustralian Bureau of Statistics 2009b, 2009c; CPrivate Health Insurance Administration Council 2009; DAustralian Bureau of Statistics 2007c; EAustralian Bureau of Statistics 2006.

expended on medical and health-related costs increased with the number of chronic conditions, and contributed to the financial burden and reduced purchasing capacity for other living expenses faced by people with chronic diseases. The overall mean proportion of income expended on health was 4.8% of household income. The 2003–04 Household Expenditure Survey provides estimates of 3.8% of household income expended on health for this age group (NATSEM 2008). Given the growth in health expenditure over the period to 2009 and the somewhat different definitions this is not inconsistent. Financial burden of out-of-pocket costs Table 5 shows the proportions of the study population with different levels of self-assessed health and different numbers of



147

Table 2. Pattern of prevalence (%) of chronic diseases according to demographic characteristics of the study population For presentation purposes confidence intervals are omitted from this table; however, for comparison purposes most differences of more than 3 percentage points are significant and most less than 2 percentage points are not Number of chronic diseases

0

1

2

3

4

5 or more

Any chronic disease

More than one

Total

17.8

26.4

23.4

15.8

8.8

7.8

82.2

55.8

Age groups 50 64 years 65 74 years >75 years

23.0 14.4 7.4

29.8 24.0 20.0

20.3 28.0 27.0

13.8 16.2 21.0

6.9 9.4 12.9

6.2 8.0 11.7

76.6 86.6 92.6

47.7 61.6 72.6

Sex Male Female

20.3 15.5

28.0 25.1

23.5 23.4

14.9 16.6

7.0 10.3

6.2 9.1

79.7 85.0

51.7 59.4

Table 3. Mean out-of-pocket expenditure by medical and healthrelated categoryA

Table 5. Financial burden of out-of-pocket costs

Mean total expenditure per quarter ($ per quarter, (95% CI)) Medication Medical services Transport Home care Medical equipment Other expenses Total expenses A

132 (116–148) 117 (103–131) 12 (9–15) 5 (3–7) 37 (8–67) 50 (39–62) 353 (312–395)

These do not add due to differential non-response to different categories. Table 4. Income and expenditure by chronic disease ever reported Average equivalent income ($’000 pa (95%CI))

No. of conditions diagnosed 0 47 (43–52) 1 43 (40–46) 2 40 (37–43) 3 42 (36–47) 4 35 (30–40) 5 33 (28–38) Overall average 42 (40–44) A

Average total health expenditure ($ per quarter (95%CI))

Average percentage of income spent on healthA (% (95%CI))

156 (105–206) 260 (214–306) 381 (319–443) 459 (383–537) 492 (362–623) 882 (456–1307) 353 (312–395)

1.9 (1.2–2.6) 2.9 (2.3–3.5) 5.0 (4.1–5.9) 6.7 (4.9–8.4) 6.4 (4.6–8.1) 16.3 (8.1–24.6) 4.8 (4.1–5.6)

Calculated by assuming quarterly income is 25% of annual income.

chronic conditions who faced moderate or severe financial burdens from their health care costs. Those with poor health measured either by self report or as numbers of chronic conditions were more likely to face financial burdens related to their health expenditure. Around one-third of those reporting poor health faced moderate financial stress and around one-quarter severe financial stress due to health care costs. Determinants of a severe financial burden among older individuals Table 6 shows the results of a logistic regression model for the correlates of severe financial burden associated with out-ofpocket expenditure.

Self assessed health Excellent Very good Good Fair Poor

Percentage expending over 10% of income on health

Percentage expending over 20% of income on health

4.7 (2.5–6.9) 8.4 (6.6–10.2) 17.4 (14.7–20.0) 19.3 (14.1–24.4) 31.0 (11.0–50.9)

2.2 (0.7–3.8) 2.9 (1.8–4.0) 7.5 (5.7–9.4) 9.1 (5.4–12.9) 25.8 (6.9–44.6)

Number of chronic diseases 0 3.7 (1.9–5.6) 1 7.3 (5.3–9.3) 2 15.2 (12.3–18.2) 3 13.9 (10.4–17.3) 4 19.9 (14.3–26.7) 5 or more 30.5 (23.2–37.9) Total

11.8 (10.5–13.1)

1.6 (0.4–2.9) 2.8 (1.5–4.1) 5.5 (3.6–7.2) 6.7 (4.2–9.2) 6.6 (3.1–10.1) 18.6 (12.4–24.8) 5.1 (4.2–5.9)

Poor physical and mental health as reflected in the SF12 measures had a significant impact on the likelihood of facing a severe financial burden due to health costs. Those aged 65–74 years spent significantly more of their income on health than the younger and older age groups. However, women were significantly less likely to face a high financial burden due to health expenditure. Women on average reported spending around $100 per quarter less on health-related costs than men. The least-educated group had the lowest likelihood of experiencing substantial financial burden due to health costs. Despite having the lowest average incomes, they also had the lowest average out-of-pocket expenditures of $245 per quarter. Even after conditioning on all the available health and demographic variables, the higher the number of chronic conditions the higher the likelihood that health costs were a severe financial burden. With an odds ratio of 1.46, on average each additional chronic disease added 46% to the likelihood of facing a severe financial burden. Similar results applied to those with moderate financial burdens (equation not shown). Discussion Older Australians expended on average $353 per quarter on health-related goods and services, with median expenditure of

148


I. McRae et al.

Table 6. Adjusted logistic regression results for the determinants of severe financial burden For simplicity of presentation no detail is provided for the following variables that were not significant: self-assessed health, current employment status, government concession card, Australian born Dependent variable – indicator of severe financial burden (odds ratio) SF12 mental health score SF12 physical health score

0.96** 0.94***

Sex Male Female

1.00 0.44***

Qualifications No school certificate or other qualification School or intermediate certificate Higher school or leaving certificate Trade/apprenticeship Certificate/diploma University degree or higher

1.00 3.61** 2.20 1.20 2.32* 0.69

Region Major cities Inner regional areas Outer regional areas Remote areas

1.00 0.68 2.15** 0.19

Age group 50–64 years 65–74 years 75 years or older

1.00 2.01** 1.26

Number of chronic diseases n McFadden pseudo R-square

1.46*** 2230 0.16

Significant at: *10%, **5%, ***1%.

$150 per quarter. Those with five or more chronic conditions were estimated to expend $882 per quarter on average. After adjusting for socioeconomic conditions and general health, the levels of financial burden due to costs of health care were significantly higher for those with more chronic diseases. Consistent with international studies (Lehnert et al. 2011), this study of older Australians shows substantial health care costs paid by older individuals with multiple chronic conditions. These data demonstrate that on average those with five or more conditions spend on average five to six times the amount spent by those with no chronic conditions, in contrast to the international studies where those with large numbers of health conditions spent three to four times more than those with no chronic conditions (Lehnert et al. 2011). Apart from potential sampling variability, there are several possible reasons for this difference. First, our study explicitly includes factors like transport not generally included in the other studies. Second, the nature of financial arrangements and safety nets in Australia are different to other systems. Third, the proportions of people spending less than they ‘should’ spend on their health because of the financial stresses they face may differ under different systems. We cannot separate these effects, but what is clear is that the Australian system with all its virtues is no better than other systems when it

comes to assisting older Australians with large numbers of chronic conditions. If we apply the estimated percentages of the community expending more than 10% or more than 20% of their equivalised income on health to the Australian population aged 55 years or over (10.8% and 5.1% respectively from Table 5), around 570 000 people would be expending more than 10% of their income on health, and around 250 000 expending more than 20%. As people with multiple chronic conditions tend to be on lower incomes (Table 4), the impact of chronic conditions on the proportion of equivalised income expended on health is more extreme than the impact on the level of expenditure. It is clear that the cost burden falls most heavily on those least able to bear it, both in terms of their health and in terms of their income. The results in Tables 5 and 6 amplify this view, showing that the number of chronic conditions is a major determinant of the probability of facing severe financial stress due to health costs, with an estimated 18% of those with five or more chronic conditions expending more than 20% of their equivalised income on out-of-pocket costs. Within some parts of the Australian health care system it has been argued that copayments should be used to minimise overuse of services (Doran et al. 2005). However, higher costs of health care are faced by those with the greatest number of chronic diseases who on average also have lower incomes (Table 4), suggesting that increasing financial barriers would further disadvantage those who are already the most needy. Limitations The main limitations of this study are the possible ‘selection’ biases, as the National Seniors Australia sample is better educated than the whole aged population although their health conditions broadly reflect those of the wider population, and that those who responded to the income and expenditure questions may also be biased due to unmeasured (and unmeasurable) factors. Nevertheless we have a reasonably large sample, weighted to reflect population age, sex, socioeconomic and geographic structures, with plausible and relevant results. Another limitation is that the expenditure data is based on recall, and perhaps more importantly a relatively short recall period leading to high variability for conditions that have less frequent but expensive events. As dental costs are not identified in our lists of costs it is also possible that some respondents did not identify their dental service costs, which suggests that the levels of out-of-pocket spending may be under-reported here. Both the Australian Medical and Pharmaceutical Benefits Schemes include safety nets that limit total out-of-pocket expenditure on goods and services covered by these schemes by a family over a calendar year. Measures of medical and pharmaceutical costs may be minimally overstated as the survey was conducted in the third quarter of the year and many people reach the thresholds late in the year. Conclusions This study shows that the likelihood of facing a substantial financial burden becomes significantly higher for each additional chronic disease experienced. This has significant implications for older Australians, with over 80% of respondents in this



study having at least one chronic disease and 56% having more than one condition. This study has provided some indicative estimates of the outof-pocket costs of health care paid directly by older Australians. It is clear that for sicker older Australians, even with the protection of Medicare, costs can be significant and are associated with a substantial financial burden. As noted in other studies (Kemp et al. 2010; Schoen et al. 2010) such financial burdens can themselves lead to reduced use of medical services and hence overall poorer health. The main conclusion of this study is that there are substantial levels of financial burden for those with multiple chronic health conditions. The combination of complex comorbidity, the costs of these conditions, and low incomes can create a ‘perfect storm’ that affects those with the least resources to deal with it. Taking the broader picture, the solution to this issue is either to introduce a form of wider safety net that addresses costs for those with multiple conditions, or in the longer term, greater endeavours to prevent chronic disease to reduce the prevalence of multiple chronic conditions. Conflicts of interest None declared. References 45 and Up Study Collaborators (2008) Cohort profile: the 45 and Up Study. International Journal of Epidemiology 37(5), 941–947. Ataguba JE (2011) Reassessing catastrophic health-care payments with a Nigerian case study. Health Economics, Policy and Law (Feb 11), 1–18. doi:10.1017/S1744133110000356 Australian Bureau of Statistics (2006) ‘National health survey, 2004–05.’ Cat. no. 4364.0. (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2007a) ‘Household income and income distribution, Australia, 2005–06.’ Cat. no. 6523.0. (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2007b) ‘Census of population and housing, non-school qualification, level of education by age and sex.’ Cat. no. 2068.0. (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2007c) ‘Census of population and housing, CDATA 2006.’ (Australian Bureau of Statistics: Canberra.) Australian Bureau of Statistics (2008) ‘Census of population and housing: socio-economic indexes for areas (SEIFA), Australia data only 2033.0.55.001.’ (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2009a) ‘National health survey: summary of results, 2007–2008 (reissue).’ Cat. no. 4364.0. (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2009b) ‘Australian labour market statistics, July 2009.’ Cat. no. 6105.0. (Australian Bureau of Statistics: Canberra) Australian Bureau of Statistics (2009c) ‘Australian social trends 2009.’ Cat. no. 4102.0. (Australian Bureau of Statistics: Canberra) Australian Institute of Health and Welfare (2006) ‘Chronic disease and associated risk factors in Australia 2006.’ Cat. no. PHE 81. (Australian Institute of Health and Welfare: Canberra) Australian Institute of Health and Welfare (2010) ‘Australia’s health 2010.’ (Australian Institute of Health and Welfare: Canberra) Dewey HM, Thrift AG, Mihalopoulos C, Carter R, Macdonell RA, McNeil JJ, Donnan GA (2004) ‘Out of pocket’ costs to stroke patients during the first year after stroke results from the North East Melbourne Stroke Incidence Study. Journal of Clinical Neuroscience 11(2), 134–137. doi:10.1016/S0967-5868(03)00148-6 Doggett J (2009) ‘Out of pocket: thinking health copayments.’ (Center for Policy Development: Canberra)

149

Doran E, Robertson J, Henry D (2005) Moral hazard and prescription medicine use in Australia the patient perspective. Social Science & Medicine 60(7), 1437–1443. doi:10.1016/j.socscimed.2004.08.005 Essue B, Kelly P, Roberts M, Leeder S, Jan S (2011) We can’t afford my chronic illness! The out-of-pocket burden associated with managing chronic obstructive pulmonary disease in Western Sydney, Australia. Journal of Health Services Research & Policy 16(4), 226–231. doi:10.1258/jhsrp.2011.010159 Gordon LG, Ferguson M, Chambers SK, Dunn J (2009) Fuel, beds, meals and meds: out-of-pocket expenses for patients with cancer in rural Queensland. The Cancer Forum 33(3), 206–212. Goss J (2008) ‘Projection of Australian health care expenditure by disease, 2003 to 2033.’ Cat. no. HWE 43. (AIHW: Canberra) Hwang W, Weller W, Ireys H, Anderson G (2001) Out-of-pocket medical spending for care of chronic conditions. Health Affairs 20(6), 267–278. doi:10.1377/hlthaff.20.6.267 Jones G, Savage E, Van Gool K (2008) The distribution of household health expenditures in Australia. The Economic Record 84(Special Issue), S99–S114. doi:10.1111/j.1475-4932.2008.00487.x Kemp A, Roughead E, Preen D, Glover J, Semmens J (2010) Determinants of self-reported medicine underuse due to cost: a comparison of seven countries. Journal of Health Services Research & Policy 15(2), 106–114. doi:10.1258/jhsrp.2009.009059 Lapsley H, March LM, Tribe K, Cross M, Courtenay B, Brooks P (2002) Living with rheumatoid arthritis: expenditures, health status, and social impact on patients. Annals of the Rheumatic Diseases 61, 818–821. doi:10.1136/ard.61.9.818 Lehnert T, Heider D, Leicht H, Heinrich S (2011) Health care utilization and costs of elderly persons with multiple chronic conditions. Medical Care Research and Review 68, 387–420. doi:10.1177/1077558711399580 Menzies Centre for Health Policy and Nous Group (2009) 2008 report 2: financial stress and health. Sydney. Available at http://www. menzieshealthpolicy.edu.au/mn_survey/index.htm [Verified 10 May 2012] NATSEM (2008) ‘Distribution of expenditure on health goods and services by Australian households.’ (National Centre for Social and Economic Modelling: Canberra) Paez KA, Zhao L, Hwang W (2009) Rising out-of-pocket spending for chronic conditions: a ten year trend. Health Affairs 28(1), 15–25. doi:10.1377/ hlthaff.28.1.15 Private Health Insurance Administration Council (2009) ‘Quarterly statistics, September 2009.’ (Private Health Insurance Administration Council: Canberra) Piette JD, Heisler MH, Wagner TH (2004) Cost related medication underuse among chronically ill adults: the treatments people forgo, how often and who is at risk. American Journal of Public Health 94(10), 1782–1787. doi:10.2105/AJPH.94.10.1782 Schoen C, Osborn R, Squires D, Doty MM, Peirson R, Appelbaum S (2010) How health insurance design affects access to care and costs, by income, in eleven countries. Health Affairs 29(12), 2323–2334. doi:10.1377/ hlthaff.2010.0862 Wagstaff A, van Doorslear E (2003) Catastrophe and impoverishment in paying for health care: with applications to Vietnam 1993 1998. Health Economics 12(11), 921–934. doi:10.1002/hec.776 Ware J, Kosinski M, Keller S (1996) A 12-Item short form health survey: construction of scales and preliminary tests of reliability and validity. Medical Care 34(3), 220–233. doi:10.1097/00005650-19960300000003 Wooden M, Freiden S, Watson N (2002) The Household Income and Labour Dynamics in Australia (HILDA) survey: wave 1. Australian Economic Review 35, 339–348. Xu K, Evans D, Kawabata K, Zeramdini R, Klavus J, Murray C (2003) Household catastrophic health expenditure: a multicountry analysis. The Lancet 362, 111–117.

www.publish.csiro.au/journals/py