This report provides projections of expenditure by disease for Australia for the period 2003 to 2033. The projections revise and update those in an earlier report prepared for the United Nations (Vos et al. 2007).

The projection model used here combines demographic factors of population ageing and population growth, and non-demographic factors of changes in disease rates, volume of services per treated case, treatment proportions (the proportion of cases that receive treatment) and excess health price inflation (i.e. the difference between health inflation and general inflation). Changes in these factors are applied to current health and high-care residential aged care expenditure (hereafter referred to as 'health and residential aged care expenditure') to project health and residential aged care expenditure for each disease and by the different areas of expenditure, such as hospitals, out-of-hospital medical services and pharmaceuticals. In most cases, disease trends are based on what has happened with a particular disease in the past. In some instances, this has been supplemented with expert judgement about likely changes in future treatment practices and the impact this may have on disease trends.

Under the central set of assumptions used in this study, total health and residential aged care expenditure is projected to increase by 189% in the period 2003 to 2033 from $85 billion to $246 billion-an increase of $161 billion (Table 3). This is an increase from 9.3% of gross domestic product (GDP) in 2002-03 to 12.4% in 2032-33. Increases in volume of services per treated case are projected to account for half of this increase (50%) (Table 4).

The two demographic growth factors-population ageing (23%) and the absolute increase in population (21%)-are projected to contribute almost a quarter each.

Other non-demographic factors contributing to the increase include excess health price inflation (5% of the increase). Changes in disease rates overall reduce expenditure projections by about 1.5% or $2.3 billion (Table 4).

The projection model shows that the causes of increased health and residential aged care expenditure vary greatly depending on the disease being considered.

Diabetes has the greatest projected increase (436%) between 2003 and 2033, followed by dementia (364%) (Table 3). The projected increase in expenditure for injuries (116%), neonatal and maternal services (88% and 84%) is low in comparison, these last two because changes in the age structure of the population mean those giving birth will be a smaller proportion of the population. The projected growth in diabetes expenditure of 436% is due to multiple reasons, particularly the projected impact on diabetes prevalence rates of expected increases in obesity.

Given the difficulties in making predictions about what non-demographic factors will be in 30 years time, these projections of health and residential aged care expenditure in 2033 may differ significantly from what actually transpires. To gain some insight into the levels of uncertainty around some of the variables, the report includes some sensitivity analyses which was conducted by varying the estimated growth in volume of services per treated case and excess health price inflation to 20% lower than the central case, and 20% higher than the central case. These analyses show that the 20% lower assumptions for the volume per case and excess health price inflation variables gave a projection in 2032-33 for health and residential aged care expenditure of 10.8% of GDP (Table 15). The 20% higher assumptions gave an estimate of 14.4% of GDP. In other words, the uncertainty around the central estimate of 12.4% of GDP or $246 billion is about 15% (tables 14 and 15). Despite these uncertainties, projection work is important, not so much for the accuracy of the numbers produced, but because it assists us to understand the drivers of health and residential aged care expenditure, and to anticipate changes.