Paying the price – out-of-pocket payments for mental health care in Australia

Keywords: access, Australia, Better Access Program, equity, Medicare, mental health, out-of-pocket payments.

This paper presents research into out-of-pocket payments (OOP) for mental health care, drawing on secondary analysis of publicly available Medicare data, covering an 8 year period from 2013–14 to 2020–21, for services provided under the Better Access Program¹ by general practitioners (GPs), psychiatry, clinical psychology and other psychology.

An OOP is the difference between the amount a doctor charges for a medical service and what Medicare and any private health insurer pays. OOPs are also called gap or patient payments.²

Health care in Australia is often associated with principles such as equity and universality. However, OOPs are a significant part of overall funding.^3,4 In 1980, the Organisation for Economic Cooperation and Development (OECD) ranked Australia the third highest out of 11 countries reporting OOPs. By 2018, Australia reported the tenth highest OOPs of 49 OECD countries.⁵ Other research confirms high out-of-pocket payments in Australia compared to other comparable countries.^6,7

While there has been research regarding OOPs in relation to primary and chronic care,^8,9 understanding their impact on mental health care specifically has been very limited.¹⁰ OOPs are a barrier to accessing treatment for people with chronic health conditions, making people with mental health conditions more likely to skip care and that this effect has been found to be stronger in Australia than other countries.¹¹

The exploratory analysis presented here is timely. An evaluation of the Better Access Program is currently underway.¹² Several recent inquiries and reports describe this Program as poorly targeted and recommended improvements or enhancements.^13,14

The Australian Institute of Health and Welfare (AIHW) publishes tables of Medicare-subsidised general practitioners (GP), allied health and specialist health care¹⁵ across 340 discrete geographic areas covering Australia with boundaries defined by the Australian Bureau of Statistics, known as statistical local areas (SLAs).¹⁶

We merged these tables to cover the period 2013–14 to 2020–21 and five services provided under the Better Access Program as follows:

1. GP mental health – plans and reviews

2. GP focused psychological strategies

3. Psychiatry

4. Clinical psychology

5. Other psychology

Apart from Table 1 which provides total expenditure on OOPs, all the other tables present the average OOPs per service based on the following equation: Total provider fees minus total Medicare benefit paid/No of services = average OOPs per service.

Table 1.  Total OOPs by profession 2013–21 (A$m).

Expenditure is reported as the AIHW publishes the data, unadjusted for inflation (current prices), reflecting the eligible claims made against Medicare each year. Across the period studied, the average annual inflation rate was 1.6%, meaning that a service costing A$100 in 2013–14 would be expected to cost A$111.90 in 2020–21.¹⁷

We used the resulting data to investigate three key research questions:

1. How much are OOPs for mental health in Australia?

2. What trends in OOPs can be seen?

3. What variations exist in OOPs across Australia?

Using this source data, it is impossible by SLA to extract the proportion of services provided at zero cost to the consumer (bulkbilled), versus those which incurred OOPs. The data therefore necessarily includes all services, including those provided at zero cost. We did find that OOPs were charged by psychiatrists and psychologists in every SLA area and in every year of the study (no SLA had zero OOPs for these services).

This limitation implies that the average OOPs presented here may be under-estimates, given inclusion of bulkbilled services serve to lower the average out-of-pocket costs for the mental health services discussed in this paper.

Private health insurance payments or costs are not included in the AIHW dataset and are therefore out of scope for the purposes of this analysis. This limitation implies that our estimates of OOPs may be over-estimates, particularly for SLAs with high private health insurance coverage. This bias would tend to lead to an under-estimate of the differences in access to services in disadvantaged areas discussed below.

Table 1 shows total OOPs paid for Medicare mental health services under the Better Access Program by Australians across the four professional groups between 2013–14 and 2020–21.

It shows steep growth, from A$1.068 bn in 2013–14, to A$1.827 bn in 2020–21, including both the Medicare and out-of-pocket funding components. This equates to A$35 m weekly, of which more than A$8 m are OOPs.

While the Program grew by 70% as a whole over this period, the out-of-pocket component grew by 164%, rising from A$163.09 to A$430.64 m. OOPs are an increasing element of overall Better Access Program costs, accounting for 15.26% of total costs at the start of the period and 23.57% at the end.

Table 2 summarises the year-on-year growth in total OOPs payments and services over the period (Fig. 1). The annual rate of growth of OOPs is considerably higher than for services.

Table 2.  Annual growth in Medicare mental health OOPs and services 2013–21.

Fig. 1.  Average OOP per service, by profession.

Table 3 shows the average OOP for each profession, per service. It shows that both psychology categories have nearly doubled over the period. OOPs for psychology and psychiatry have increased appreciably during the pandemic, at the same time as demand for services was very strong, particularly new telehealth services.¹⁸

Table 3.  Average OOP (A$) per service, by profession

The final element of our analysis was to consider regional variations in out-of-pocket spending. These variations impact the equity of the provision of Medicare benefits in Australia.¹⁹ We analysed average out-of-pocket spending by 340 SLAs to explore regional variations in OOPs.

Table 4 presents the highest and lowest average OOPs for services by SLA. We have not edited the dataset or considered service utilisation data. Table 4 therefore includes some SLAs where very few services are provided.

Table 4.  Highest and lowest average OOPs, by SLA and profession.

However, variable service use does not explain everything. For example, in 2020–21, the highest average OOPs for GP focused psychological strategies was in the Manly SLA while the lowest was in Innisfail. Both these locations recorded small numbers of services overall, but the Innisfail service provided them all bulkbilled (or zero OOPs).

Table 4 demonstrates significant, regional cost variations. For example, in 2020–21, the OOPs for seeing a psychiatrist in Weston Creek (Canberra) were on average A$120.55 per service, but A$4.12 in the Eyre Peninsula. Looking at the 2014–15 data it is possible to see large variations even between two remote areas, with the Goldfields SLA average OOPs for clinical psychology services at A$73.56 per service while in Bourke/Cobar it was A$2.15.

As stated earlier, we found no SLAs recording zero OOPs for services provided by psychologists or psychologists. We found that for GP focused psychological strategies, they are either not available, rare, or often bulk-billed. For example (Table 4), the average OOPs in Manly in 2020–21 was A$99.56 per service but in Rockhampton, this service was only provided to a few dozen patients, bulkbilled, and therefore incurring zero OOPs.

Table 5 attempts to even out some of these exaggerated regional variations by grouping together regions with similar characteristics, using the Socio-Economic Status (SES) variable provided in the Medicare data. Table 5 groups SLAs into three categories: High SES, Low SES and Inner Regional SLAs.

Table 5.  Average variations in OOPs by SLA and SES 2020–21.

Despite grouping similar places, strong variation in OOPs is evident. In South Canberra the average OOPs payment is A$26.92 per GP mental health service but in Blacktown North SLA you are likely to be bulkbilled. Another Canberra SLA (High SES), Weston Creek, faces average OOPs of A$120.55 for every psychiatry service while in Mitcham SLA (Adelaide, High SES) OOPs on average are just A$38.88. Psychiatry in regional areas also varies appreciably with average OOPs of A$99.69 in the Wheat Belt North SLA (WA) but only A$14.77 in Devonport (Tasmania). More than 20 000 other psychologist services were provided in both Newcastle and Merrylands-Guildford SLAs in 2020–21, yet the former recorded average OOPs of A$56.79 per service, the latter A$15.45.

OOPs remain a relatively minor component of the costs associated with mental health services provided by GPs, presumably reflecting a relatively high rate of bulkbilling for these services.

This is not the case for the other professions, which have seen very considerable increases in OOP for their services. For example, total expenditure on OOPs to see ‘other psychologists’ has tripled. While total out-of-pocket spending has increased, so of course has the size of the program itself with more services being provided year on year. In 2013–14, around 8.8 million Medicare mental health services were provided, increasing by 2020–21 to more than 12 million. While overall volume may be up, the average number of services provided to each person in the Better Access Program has been in decline – it was 5.1 services per person in 2007–08, down to 4.5 in 2021.²⁰

In 2019–20 the average OOPs charged by ‘other psychologists’ overtook those set by clinical psychologists, perhaps to compensate for lower Medicare payments for their services.

The OOPs Australians are charged for mental health care vary considerably depending on where they live, raising issues about equity of access and fairness, even between SLAs of similar socio-economic status. When one community is paying two, three or even four times as much in OOPs for mental health care as another community of similar socio-economic status, it is reasonable to ask if such variations are either justifiable or desirable.

We already know that Tudor Hart’s Inverse Care Law²¹ applies to the Better Access Program, so that services are most available where they are needed least, and vice versa.²² Further research to marry this OOPs data with service utilisation data may well confirm the aggregation of mental health services in the SLAs with most capacity to pay.

The Productivity Commission inquiry recommended a greater role for regions in planning and funding mental health care.²³ Further investigation of low OOPs in high service volume SLAs might yield useful information about local models of service delivery and payment which do not generate OOPs while providing acceptable remuneration to professionals.

The data presented here is an initial, exploratory interrogation of a very large Medicare mental health data set. But is it clear that OOPs are becoming a bigger component of overall mental health expenditure. Perhaps with the exception of GPs, mental health professionals are increasing their charges for services at a faster rate than overall growth in mental health spending under Medicare. Average OOPs per service almost doubled over the period 2013–21 for all psychologists.

The Federal Government sanctioned the removal of regulation of OOPs when the Program shifted in 2006 from its smaller predecessor, Better Outcomes, to Better Access, partly in response to psychologists’ concern about remuneration.²⁴

The Government has recently expanded the number of subsidised sessions of psychology available under Medicare, but more sessions mean more OOPs, making ongoing treatment less affordable for poorer people.

Not only is the Better Access Program currently under evaluation, but the direction of national mental health reform is also under scrutiny by a new Federal Government. Data presented here on OOPs can help shape this work, to create a fairer and more sustainable environment for the next phase of national mental health reform.

The tables and figures presented here are all derived from source data publicly available from the Australian Institute of Health and Welfare, here: https://www.aihw.gov.au/reports/mental-health-services/mental-health-services-in-australia/report-contents/medicare-subsidised-mental-health-specific-services.

The authors declare no conflicts of interest.

This research did not receive any specific funding.

Dr Rosenberg prepared the manuscript, Dr Park collated the dataset and Professor Hickie contributed to the analyis.