How much do we spend on prescription medicines? Out-of-pocket costs for patients in Australia and other OECD countries

Additional keywords: international comparison, pharmaceutical expenditure, private expenditure.

Across industrialised countries, pharmaceutical expenditures have increased rapidly over recent decades.^1–3 These increases have been attributed to ageing populations, the development of new and expensive drugs, and the increasing use of pharmaceuticals in managing chronic disease.^4,5 In Australia, public expenditure under the Pharmaceutical Benefits Scheme (PBS) rose by ~11% per annum throughout the late 1990s and 2000s, reaching AU$7 billion in 2007–08.⁶ There has been much public discussion about whether this scheme is becoming too expensive and how to best ensure that the scheme remains affordable for the community, and less about out-of-pocket costs for individual patients and whether these costs are affordable.^7–9 In recent years, a series of PBS policy changes has increased the proportion of medicine costs borne by patients. Patient co-payments^A were increased by 24% in January 2005 and the safety-net threshold^B has increased by the equivalent of eight prescriptions since January 2006. The 2006 introduction of the safety net ‘20-day rule’^10,C has reduced the number of patients eligible for cheaper medicines towards the end of the calendar year.¹¹ There is growing evidence that many Australian patients are struggling to afford their prescribed medicines.^3,11,12

Comparisons are often made with other Organisation for Economic Co-operation and Development (OECD) countries when determining whether public and patient medicine expenditure in Australia is sustainable.^5,13,14 Patients’ out-of-pocket spending (‘private expenditure’) on prescription medicines is affected by the pharmaceutical subsidy systems which provide coverage. Subsidy arrangements vary considerably between countries, with public, private and mixed systems operating across OECD nations. Despite the large differences in national wealth across OECD countries, all have special pharmaceuticals subsidy arrangements in place for their low income and other vulnerable populations (Table 1).

Table 1.  Characteristics of pharmaceutical subsidy schemes across Organisation for Economic Co-operation and Development (OECD) countries

This study examines private expenditure on prescription medicines in Australia over the previous three decades, and compares the current out-of pocket prescription medicine expenditure of Australian patients with that in other OECD countries.

Data were sourced from: (i) the Australian Institute of Health and Welfare (AIHW) data cubes¹⁵; (ii) annual reports from the AIHW’s ‘Health Expenditure Australia’ series^13,14,16; and (iii) the 2008 release of the OECD’s Health Data statistics.¹⁷ The AIHW data cubes contain Australian public and private health expenditure statistics dating back to 1960. These data do not capture private expenditure for all prescription medicines in Australia, only publicly subsidised prescription medicines. Public subsidies are paid for PBS-listed medicines which cost more than the patient contribution (co-payment) for a given beneficiary group; currently $5.40 for concessional beneficiaries and $33.30 for general beneficiaries.¹⁸ PBS medicines priced below these co-payments are not subsidised, and the expenditure data related to these prescriptions are not captured in the ‘benefit-paid’ data. The completeness of these data have changed over time as the general co-payment amount, and the number of patients eligible for concessional benefits, has changed.⁵ Prior to 1986, the co-payment for general beneficiaries was sufficiently low that virtually all prescriptions were subsidised and therefore captured in the benefit-paid data.¹⁹ In November 1986 the general co-payment was doubled from $5 to $10. Following this, and all subsequent co-payment increases, some PBS medicines were no longer captured in the benefit-paid data.¹⁹ Between 1990 and 2007 the benefit-paid statistics have only captured an average of 81% of the PBS prescriptions dispensed nationally.^20–23

The AIHW data also provide statistics on ‘other’ pharmaceuticals, which is an aggregate expenditure for over-the-counter (non-prescription) medicines and private (non-PBS) prescriptions as well as prescription medicines below the PBS co-payments.¹⁵ Annual AIHW reports provide expenditure estimates specifically for under co-payment and private prescriptions. These reports allow total private expenditure to be calculated for all prescription medicines dispensed in Australia for the years 2004–07.^13,14,16

The 2008 OECD Health Data statistics include health system, healthcare consumption and economic indicators for the 30 OECD member countries. Data are intermittent for the years 1960–70 but were collected annually from 1971. Some data are not available for all countries in all years. The primary indicator of interest in this study was private expenditure on prescription medicines. These data are not collected in all countries, and some countries’ records do not distinguish between private expenditure on prescription and over-the-counter medicines.

We examined changes in public and private expenditure on prescription medicines in Australia between 1971 and 2007. To adjust for changes in population size and inflation, we compared per capita real expenditure converted to 2007–08 gross domestic product (GDP) price levels.¹⁵ We calculated the proportion of all national prescription medicine expenditure paid privately, and the proportion of household expenditure directed towards prescription medicines between 1971 and 2007.

We compared private expenditure on prescription medicines in Australia with that in 15 other OECD countries; Canada, Czech Republic, Denmark, Finland, France, Germany, Japan, Republic of Korea (South Korea), Luxembourg, Poland, Slovak Republic, Spain, Sweden, Switzerland and the United States (US) in 2005, the most recent year when data from all countries were available. To allow comparison between currencies, per capita expenditure on prescription medicines were converted into $US using the OECD’s purchasing power parity (PPP) exchange rates.²⁴ PPPs account for price differences across a range of goods and services between countries and are more reliable and stable than exchange rate conversions.²⁴ The OECD caution that PPPs are constructs rather than exact values²⁵ and that small differences in per capita expenditure between countries should not be interpreted as economically or statistically significant. They recommend that countries be grouped according to their per capita wealth rather than strictly ranked.²⁵ Consistent with OECD recommendations on groupings, we compared the per capita prescription medicine expenditure of high income (Luxembourg, Switzerland and the US), high–middle income (Australia, Canada, Denmark, Finland, France, Germany, Japan and Sweden), low–middle income (Czech Republic, South Korea and Spain), and low income countries (Poland and the Slovak Republic).²⁵

We also compared private expenditure on prescription medicines in Australia with 15 other countries as a proportion of total national prescription medicine expenditure and as a proportion of household consumption expenditure in 2005. Household consumption expenditure literally refers to total household spending of goods and services and is included in the OECD Health Statistics.¹⁷

Prescription medicine expenditure in Australia has increased substantially over the past three decades, even after accounting for inflation and population growth (Fig. 1). Private spending on benefit-paid prescription medicines increased steadily in real terms throughout the 1970s and 1980s, but has tripled since 1991, rising from an average of AU$19 to $62 per person in 2007. Private spending on all prescription medicines reached $134 per person by 2007.

Fig. 1.  Public and private expenditure per capita on prescription medicines in Australia (2007 GDP price levels). (a) Data sourced from the Australian Institute of Health and Welfare (AIHW) data cubes.¹⁵ (b) Data sourced from the AIHW’s ‘Health Expenditure Australia’ reports.^13,14,16

Patient spending accounted for ~22% of benefit-paid expenditure in the 1970s and 1980s and ~16% through the 1990s and 2000s (Fig. 2). These data do not capture all PBS medicines after 1986, when some medicines began falling below the general co-payment. When expenditure on private prescriptions and those under co-payment is included, patient contributions account for 28% of national prescription expenditure. Fig. 3 shows private expenditure on prescription medicines as a proportion of household consumption expenditure, in 2007 GDP prices. In the 1970s and 80s Australian patients used ~0.1% of their household expenditure for benefit-paid medicines. This proportion increased rapidly after the late 80s, despite some medicines falling out of data-capture during this time, and by 2006 reached 0.2%. In 2006, the total proportion of household expenditure used for all prescription medicines in Australia was 0.43%.

Fig. 2.  Percentage of total prescription medicine expenditure paid privately in Australia (2007 GDP price levels). (a) Data sourced from the Australian Institute of Health and Welfare (AIHW) data cubes.¹⁵ (b) Data sourced from the AIHW’s ‘Health Expenditure Australia’ reports.^13,14,16

Fig. 3.  Percentage of household consumption expenditure used for prescription medicines in Australia, 1971–2006. (a) Data sourced from the Australian Institute of Health and Welfare (AIHW) data cubes.¹⁵ (b) Data sourced from the AIHW’s ‘Health Expenditure Australia’ reports.^13,14,16

We compared private per capita expenditure on prescription medicines in Australia with 15 OECD countries (see Fig. 4). In 2005, Australian patients spent US$84 per capita on prescription medicines. This positioned Australia as equal 6th highest of 16 OECD countries with available private expenditure data. Per capita private expenditure was higher for patients in the United States ($487), Canada ($262), Finland ($120), France ($103) and the Slovak Republic ($95), compared with Australia.

Fig. 4.  Private expenditure on prescription medicines per capita across OECD countries with different national incomes, 2005. Data was sourced from the OECD¹⁷ for all countries except Australia, with data sourced from the AIHW.¹³

We compared the proportion of national prescription medicine expenditure paid by patients in each of the countries with available data (Fig. 5). More than one quarter (28%) of prescription medicine costs in Australia are privately financed, placing Australia in the middle of the 16 OECD countries and 4th highest of the countries with universal pharmaceutical subsidies. Australian patients pay for a higher proportion of national pharmaceutical expenditure than patients in South Korea (27%), Slovak Republic (26%), Sweden (22%), France, Luxembourg, Japan and Switzerland (17%), Germany (15%), Czech Republic (11%) and Spain (6%), but less than patients in Finland (36%), Denmark (33%) and Poland (34%).

Fig. 5.  Percentage of national prescription medicine expenditure paid privately across OECD countries with different national incomes, 2005. Data were sourced from the OECD¹⁷ for all countries except Australia, with data sourced from the AIHW.¹³

Compared to other high–middle income countries, Australians spend a moderate proportion of their household income on prescription medicines (0.4%) (Fig. 6). This proportion is consistent with the proportions of income spent in Denmark, Japan and Sweden. Proportionate to household income, spending was twice as high for patients in Finland (0.8%) and more than three times as high in Canada (1.4) and the US (1.7%). However, substantially lower portions of national expenditure were incurred by patients in the Czech Republic (0.3%), Germany (0.3%) and Spain (0.2%).

Fig. 6.  Proportion of household consumption directed towards prescription medicines across OECD countries with different national income. Data were sourced from the OECD¹⁷ for all countries except Australia, with data sourced from the AIHW.¹³

Patient spending on prescription medicines has increased substantially in Australia over the past three decades. Private expenditure on benefit-paid medicines tripled between 1991 and 2007. During the same period, expenditure on other essentials, such as food and electricity and gas, increased by much smaller margins (66 and 121% respectively).²⁶ Although the benefit-paid statistics are available for every year in Australia since 1971, these data are not complete and underestimate patient expenditure after 1986. Many high-volume medicine groups, including antibiotics, antidepressants, anxiety medications, metformin and oral contraceptives are priced below the general beneficiary co-payment, and are not captured in the benefit-paid statistics.²⁷ This represents a critical gap in data availability over the past 24 years. Although the benefit-paid component of private spending showed little increase after the January 2005 co-payment increase, private expenditure on all prescription medicines actually increased from $97 to $123 per person in that year. From 2004 to 2006, the proportion of all prescription medicine expenditure borne by patients increased from 25 to 28%, and the proportion of household expenditure used for medicines rose from 0.34 to 0.43%.

The international data suggest that patient out-of-pocket expenditure in Australia is now moderate to high. Australia ranks in the middle (6th of 16) of the OECD countries in terms of patient expenditure, and 4th of 14 countries with universal pharmaceutical subsidies. It is unsurprising that per capita private spending would be substantially higher in the US and Canada given their largely private pharmaceutical insurance systems. Although Canada’s GDP/per capita is similar to Australia’s, the lack of universal pharmaceutical cover in Canada likely accounts for why Canadian patients pay almost twice as much towards their national medicine costs compared to Australian patients (53 v. 28%).

Australian patients do not face the highest prescription costs in the OECD, but their expenditure is in the mid to upper range for countries with universal pharmaceutical coverage. Recent data suggests than many Australian patients are at the limit of what they can afford to pay for medicines. Two international surveys have reported that in 2007, 13%³ of Australian adults, and in 2008, 20%²⁸ of Australian adults in poor health, reported recently skipping doses of medicines or not filling a prescription because of cost (Fig. 7). The rates of underuse due to cost reported by Australian patients in these studies were only second to those reported by US patients and higher than those in Canada,^3,28 France²⁸ and Germany.^3,28 Australian studies have reported that dispensings of selected essential medicines including statins, proton-pump inhibitors, non-aspirin antiplatelets and osteoporosis treatments have decreased significantly following the 2005 co-payments increase and subsequent safety net policy changes.^11,29

Fig. 7.  The proportion of adults reporting they had recently skipped doses of medicine or not filled a prescription because of cost in five OECD countries.^3,28

The expenditures reported here are national-level and therefore represent estimated per capita expenditure for the ‘average’ patient. Clearly, prescription medicine expenditure will vary according to the health status, comorbidities and socio-economic status of individual patients. Some patients will face higher cost burdens than are reflected by these national estimates. There are also limitations with the use of PPPs to compare pharmaceutical expenditure across countries. Although PPPs are preferable to volatile exchange rates, they are based on the price of a range of goods and services.²⁵ Healthcare or pharmaceuticals specific PPPs would be a more accurate unit of exchange. These conversion rates are currently under development by the OECD.²⁵ A further limitation is the incompleteness of the OECD database. Data on private prescription medicine expenditure were not available from the countries with national wealth and public subsidy systems most similar to Australia (i.e. Italy, Iceland, the Netherlands, New Zealand and the United Kingdom).

Our analysis has not accounted for patient expenditure on non-prescription medicines. Available data indicated that Australians spent $146 per person on non-prescription medicines in 2005; however, these data do not differentiate between spending on over-the-counter medicines and medical non-durables (i.e. bandages and condoms).^13,17 Cross-country comparisons of non-prescription medicine expenditure are also difficult due to differences in the way non-prescription medicines data are captured and defined.² Given these data limitations, we have not reported non-prescription medicine expenditure. However, we acknowledge that the true cost burden of medicines is likely to be much higher than indicated by our analyses of prescription medicines alone. The burden of prescriptions would also be higher for many patients if the PBS safety net was not in place.

The cross-county data discussed here are based on current (2005) expenditures, and the relative ranking of countries may change in the future. Of the 16 countries we examined, 4 have fully proportional systems (Japan, South Korea, Spain and Switzerland), 4 have capped proportional co-payments (Denmark, Finland, Germany and Sweden,), 5 have proportional co-payments in conjunction with ‘free lists’ (Czech Republic, France, Luxembourg, Poland and Slovak Republic), 1 has fixed co-payments (Australia) and 2 have mixed systems (Canada and the US). These systems may respond differently to future increases in medicines expenditures, and the effects on patient expenditures across countries may be varied.

The findings from this study indicate that Australian patients have faced increasing prescription medicines costs over recent years and that patient expenditure is now in the middle to upper range with comparable countries. Further large increases in the portion of medicine costs borne by patients have the potential to substantially affect access to and use of prescribed medicines, with potential risks to patient health and wellbeing. Policy makers should consider the ongoing affordability of medicine to patients, as well as to the wider community, when reviewing pharmaceutical reimbursement policy.

The authors declare that no conflicts of interest exist.