Abstract
In recent years the focus has been to explore the linkages between Indigenous health and institutional arrangements for health care delivery and expenditure. The consequences of low socio-economic status for health status are examined at both the community level and at a wider national and comparative international level.
This chapter examines the amount of money spent, the data supply, issues of equity, and income and health status. Attention is drawn to the fact that spending on the environmental health infrastructure is overlooked.
While the broad profile of ill health and excess mortality among Indigenous Australians is well documented, research on determinants remains relatively undeveloped. Nonetheless, sufficient insight exists to place an emphasis in explanation on the compound effects of overall low socio-economic status, including low income. This is in line with theoretical trends generally in social epidemiology in which biological pathways between psychosocial stress and ill health are seen as explanatory (Berkman & Kawachi 2000; Kawachi, Kennedy & Wilkinson 1999; Marmot & Wilkinson 1999). Within this paradigm, persistently low levels of life expectancy among Indigenous Australians would be viewed as a product of their entrenched position at the bottom of the socio-economic hierarchy.[1]
One aspect of this focus in recent years has been to explore the linkages between Indigenous health and institutional arrangements for health care delivery and expenditure. Thus, in considering Indigenous health development, issues to do with interactions between minority Indigenous and majority non-Indigenous institutions are increasingly to the fore. Conceptually, these fall within a framework of what has been described as the ‘new public health’, which stresses the contribution of social justice, social action, power and access to resources as key components of Indigenous health outcomes.
To date, this redirection has generated insights at three levels. The first is at the scale of discrete communities where the focus has been on measuring the effects of improved environmental health infrastructure and identifying institutional impediments to achieving this (Torzillo & Kerr 1991). At a more macro-level, the consequences of control over health policy and its delivery have been explored in the context of Australian federalism, and at a comparative international scale (Bartlett & Legge 1994; Hogg 1992; Kunitz 1990). Finally, the consequences of low socio-economic status for health status are increasingly being examined at both the individual and group levels (Deeble et al. 1998; Gray & Broughton 2001; Hogg 1990; Hunter 1999, 2000a, 2001). The present analysis falls firmly within the last category, although some overlap with the first is also achieved through examination of the relationship between income and health expenditure using micro-level data.
Previous analysis of the amounts spent on health services for and by Aboriginal and Torres Strait Islander people found that the per capita level was about eight per cent higher than that spent for and by other Australians. Government health expenditure on Aboriginal and Torres Strait Islander people was found to be 47 per cent higher than on other Australians (Deeble et al. 1998). To assess whether this represented an equitable allocation, it was noted that relative expenditure fell well short of implied levels of Indigenous need, given that death rates for the Indigenous population were around three times the national average. By comparing government expenditure on Indigenous and other Australians in the lowest income group, the conclusion was drawn that Indigenous people were in receipt of expenditure equivalent to others in a similar economic position, but their worse health status was not adequately reflected. In further pursuit of a ‘needs-based’ formula for resource allocation, comparative crude death rates have been applied as a proxy for morbidity to argue for an additional 27 per cent increase in total expenditures on Indigenous health (NCEPH 2000).
While excess mortality provides a rough indication of need, it offers no guidance as to the cost-effectiveness of resource allocation for treating differentially prevalent morbidity. Such an approach would focus on ‘capacity to benefit’ and, ideally, a formulation would be based on equal expenditures for the same medical conditions. However, as Deeble et al. (1998: 52) point out, while such a calculation should be possible, it would not be the only criterion for allocating government expenditures since equality by medical need would be sufficient only if all services were publicly provided to all people without charge. This is not the case.
The amount of money spent on the health of each individual is comprised of expenditure by government (public health expenditure) and private health expenditure. It is an observed fact that the balance of these expenditures on individuals varies such that a positive correlation exists between income levels and the proportion of the population with private health insurance (Deeble et al. 1998: 57). Furthermore, people with higher income tend to have more out-of-pocket payments through schemes like the Pharmaceutical Benefit Scheme (PBS). The significance of this observation is seen in the quite different income distributions of the Indigenous and non-Indigenous populations. In 1996, Indigenous family incomes were on average 32 per cent lower than non-Indigenous family incomes. Ipso facto, dependence on public expenditure for access to health services is greater among Indigenous people.
This report asks the question—what is the relationship between income, health expenditure and health status for the Indigenous and non-Indigenous populations? The analysis draws out differences in expenditure between the Indigenous and non-Indigenous population holding income level constant. This is important to the extent that income is seen as an indicator of ability to address the need for health expenditure.
Unlike the analysis for the Australian population as a whole presented in Deeble et al. (1998), changes in the questions on the 1995 NHS mean that it is not possible to separate expenditure into private and public components, and therefore the analysis in this report deals only with total health expenditure. In spite of this limitation, the analysis remains of policy value since there is no existing analysis of health expenditure by income for the Indigenous population.
[1] The health–social status correlation is hypothesised to occur because: (1) social position impacts health through access to health care, nutrition, working conditions etc; (2) health determines social status (the health selection hypothesis); or (3) common factors determine both social position and health (a variation of the health selection hypothesis)—that is, the social epidemiology literature tends to discount the health selection hypothesis.