Bulk‐billing rates and out‐of‐pocket costs for general practitioner services in Australia, 2022, by SA3 region: analysis of Medicare claims data

Affordable and equitable access to health care is a pillar of the Australian universal health care system,1 but out‐of‐pocket costs for patients comprise a substantial proportion of all health care spending.2 As the cost of living increases, concern is growing about the decline in bulk‐billing by general practitioners and the rise in out‐of‐pocket costs for non‐bulk‐billed services. Medicare subsidises fees for patients, but doctors set their own fees. Out‐of‐pocket payments (the gap between fee and Medicare rebate) consequently vary substantially between doctors.

Information about geographic variation in costs for general practitioner care is important because general practitioners are the main gatekeepers for health care in Australia. Further, variation by broad area‐level characteristics is important because people with lower incomes and those living in rural and remote areas are more likely to use general practitioner care than specialist care.3 Greater transparency in general practitioner fees, particularly at the small area level, would also inform people about the out‐of‐pocket costs they are likely to pay.

Geographic differences in out‐of‐pocket costs have been reported for specific Medicare‐subsidised services and conditions4 and for general practitioner services aggregated by geographic remoteness,5 or at the primary health network6 or state and territory levels.7 The federal government has increased price transparency for different types of health care services, including general practitioner services, at the primary health network level with its Medical Costs Finder.8 However, as it depends on providers reporting their own fees, the information provided by this tool has been limited. Area‐level information is crucial for informing people about the costs to expect in their communities, particularly as they are likely to seek care near where they live.9

To provide this information, we used newly available whole‐of‐population administrative data to derive rich, small area‐level information about costs in Australia. We examined bulk‐billing rates and out‐of‐pocket costs for non‐bulk‐billed general practitioner services at the Statistical Area 3 (SA3) level. We also explored differences in these outcomes by area‐level socio‐economic disadvantage and remoteness.

Methods

For our retrospective analysis, we used the Person Level Integrated Data Asset (PLIDA), an individual‐level, linked administrative dataset that includes information on Medicare claims and patient demographic characteristics for Australia.10 Given administrative delays in claims processing, we analysed data for claims reported during the 2022 calendar year, the most recent year for which claims data were complete at the time of our analysis. We extracted data for all claims for all non‐referred general practitioner attendances based on Broad Type of Service category 101.11 Claims data include information on whether a service was bulk‐billed (that is, the total service cost is covered by Medicare), the postcode in which the service was provided, the fee charged by the provider, and the benefit paid by Medicare. The out‐of‐pocket cost for non‐bulk‐billed services was calculated by subtracting the Medicare benefit paid from the provider fee; all costs were inflated to 2024 dollars using the consumer price index.12 Using the linked demographic information in PLIDA, we then calculated the number of services, the number of bulk‐billed services, and the mean out‐of‐pocket costs per non‐bulk‐billed service for each postcode, age group (0–15, 16–40, 40–65, 65 years or older), and sex (male, female).

We generated weighted means at the SA3 level (the smallest spatial level permitted by data custodians and PLIDA user agreements) using 2021 Australian Bureau of Statistics correspondences.13 SA3 regions include populations of 30000–130000 people, and include clusters of related suburbs in larger cities and regional towns, and their larger surrounding areas in less densely populated parts of Australia.14 We excluded all SA3 regions with an effective design population of zero (for example, large national parks) and, in accordance with PLIDA use agreements, all SA3s with fewer than ten general practitioner services. We calculated the mean proportion of bulk‐billed services and the mean out‐of‐pocket costs per non‐bulk‐billed general practitioner service in each SA3 region. Given the substantial differences in the case mix of services in different regions, we adjusted for differences in general practitioner service use by age group (0–15, 16–40, 40–65, 65 years or older) and sex (male, female) (Supporting Information, table 1). We depict these differences in SA3‐level maps by level quartiles for each outcome. Using the Greater Capital City Statistical Area correspondences,13 we also provide separate maps for Greater Melbourne, Greater Adelaide, Greater Brisbane, Greater Perth, Greater Sydney, Greater Darwin, Greater Hobart, and the Australian Capital Territory.

We linked SA3‐level outcomes information with classification by remoteness and area‐level socio‐economic resources. For remoteness, we used three categories based on those of the Modified Monash Model (MMM):15 metropolitan, including metropolitan areas (MMM1) and regional centres (MMM2); rural, including large rural towns (MMM3), medium rural towns (MMM4), and small rural towns (MMM5); and remote, including remote communities (MMM6) and very remote communities (MMM7). Area‐level socio‐economic resource status was based on the 2021 SA3‐level Socioeconomic Indexes for Areas (SEIFA) Index of Relative Socioeconomic Disadvantage (IRSD),16 by quintile.

We separately compared outcomes (bulk‐billing rates and out‐of‐pocket costs for non‐bulk‐billed services) by SA3‐level characteristics (IRSD quintile and remoteness category); the statistical significance of differences was assessed in Wald tests. All analyses were completed in Stata 17.

Ethics approval

The University of Melbourne Research Ethics Committee approved this study (project reference, 12519).

Results

Bulk‐billing rates and out‐of‐pocket costs for non‐bulk‐billed general practitioner visits, by SA3 level

Mean bulk‐billing rates by SA3 level are shown in Box 1 and Supporting Information, table 2; mean out‐of‐pocket costs for non‐bulk‐billed general practitioner services are shown in Box 2 and Supporting Information, table 2. Overall, 82% (95% confidence interval [CI], 80–83%) of general practitioner services were bulk‐billed; the mean out‐of‐pocket cost for non‐bulk‐billed visits was $43 (95% CI, $42–44). The bulk‐billing rate was lowest in South Canberra (46%) and highest in St Marys (New South Wales) and Daly–Tiwi–West Arnhem (Northern Territory), where 99% of general practitioner services were bulk‐billed. Mean out‐of‐pocket costs for non‐bulk‐billed general practitioner visits were lowest in Mount Druitt (NSW; $16 per non‐bulk‐billed visit) and highest in Bourke–Cobar–Coonamble (NSW; $99 per non‐bulk‐billed visit).

Bulk‐billing rates and out‐of‐pocket costs, by SA3‐level socio‐economic disadvantage quintile

Bulk‐billing rates and out‐of‐pocket costs for non‐bulk‐billed general practitioner services by SA3‐level socio‐economic disadvantage quintile are shown in Box 3. Mean bulk‐billing rates were higher in regions in the most socio‐economically disadvantaged quintile (86%; 95% CI, 84–88%) than in those in the least socio‐economically disadvantaged quintile (73%; 95% CI, 70–76%; mean difference, 13 [95% CI, 10–16] percentage points) (Supporting Information, table 3).

Mean out‐of‐pocket costs for non‐bulk‐billed general practitioner services were lower in SA3 regions in the most socio‐economically disadvantaged quintile ($42 [95% CI, $40–45] per visit) than in those in the least socio‐economically disadvantaged quintile ($47 [95% CI, $45–49] per visit; mean difference, $4.82 [95% CI, $2.15–7.50]) (Supporting Information, table 3).

Bulk‐billing rates and out‐of‐pocket costs, by SA3‐level remoteness category

Bulk‐billing rates and out‐of‐pocket costs for non‐bulk‐billed general practitioner services by SA3‐level remoteness category are shown in Box 4. The mean bulk‐billing rate was not significantly different in remote (86%; 95% CI, 79–92%) and metropolitan regions (81%; 95% CI, 79–83%; mean difference, 5 [95% CI, −1 to 11] percentage points) (Supporting Information, table 4).

The mean out‐of‐pocket cost for non‐bulk‐billed general practitioner services was higher in remote regions ($56; 95% CI, $46–66) than in metropolitan regions ($43; 95% CI, $42–44; mean difference, $13.33; 95% CI, $9.14–17.51) (Supporting Information, table 4).

Discussion

We examined the distribution of bulk‐billing and out‐of‐pocket costs for general practitioner services in Australia by SA3 region. We found that bulk‐billing rates are generally high, but they differed considerably between regions. It is reassuring that mean bulk‐billing rates are higher in the most socio‐economically disadvantaged regions (86%) than in the least disadvantaged regions (73%). However, mean out‐of‐pocket costs for non‐bulk‐billed services are considerable, do not vary substantially by area‐level disadvantage, and are highest for people in remote areas.

Our findings have important policy implications. First, the substantial differences in out‐of‐pocket costs could indicate inequitable access to health care. In particular, our findings are concerning because Australians in rural and socio‐economically disadvantaged areas have poorer health outcomes, lower access to and use of primary health care, and higher rates of potentially preventable hospitalisations than other Australians.17 As we found that out‐of‐pocket costs for non‐bulk‐billed general practitioner services did not differ markedly by area‐level disadvantage, and that they are in fact higher in remote areas, our findings are consistent with other reports that people in remote and socio‐economically disadvantaged areas are likely to spend larger proportions of their income on out‐of‐pocket fees for health care than other people.2

To reduce out‐of‐pocket primary care costs for patients, the Australian Department of Health increased bulk‐billing incentives for general practitioners in rural and remote areas in January 2022,18 then tripled the incentive payments in November 2023.19 These incentives are explicitly for general practitioner care for Commonwealth concession card holders and children, and are 50–90% higher for general practitioners practising in rural regions than for those in metropolitan areas. For example, metropolitan general practitioners now receive a $20.65 bonus for bulk‐billing a concession card holder or child under 16 years of age; general practitioners in rural and remote areas receive $31.35 to $39.65. We found that people pay a mean $43 in out‐of‐pocket costs for non‐bulk‐billed general practitioner visits in Australia, more than the bonus for bulk‐billing. The current incentives are therefore unlikely to be sufficient for reducing out‐of‐pocket costs, particularly for people in metropolitan areas.

Further, given the large differences between regions in fees for general practice consultations, the 2023 incentive payments could have unintended consequences. In areas where the bonuses are higher than local fees, they could drive up fees and even lead to higher out‐of‐pocket costs for people not covered by the incentive program. As an alternative, the government could reduce bonus payments to the local fee level in rural and remote areas and redistribute the money saved to ensure that people with lower incomes and children receive free general practitioner care, regardless of where they live.20 As financial incentives for specific patient groups (instead of regions) reduce differences in primary health care use,21 they could also be used to help other priority groups, including Indigenous people and people living with disability or long term health conditions.22

Our findings also have a practical application. Based on whole‐of‐population claims data, our findings supplement and improve upon the spatial information provided by the Medical Costs Finder about the cost of general practitioner services.8 Decision makers can use our approach to monitor differences between regions and people can use our findings to inform themselves about local costs. However, price transparency tools have had mixed success with respect to reducing patient out‐of‐pocket costs.23 While people may alter their choice of health care provider to receive less expensive services, the uptake and use of these tools is generally low because the search effort required is too great.23 Nevertheless, reducing information asymmetry could create a better environment in which people could discuss costs with providers. Using administrative data to routinely provide this information is highly feasible, and could be updated as new claims data become available. Should such transparency tools be introduced, the efficacy of price transparency on out‐of‐pocket costs in Australia will need to be investigated.

Limitations

First, our analyses did not take care complexity or patient needs into account; for example, people in remote areas are likely to have more complex health problems, and general practitioners consequently charge higher fees in these area.24 The limited supply of general practitioners in these regions may also lead to higher fees.25 Second, our analysis did not examine bulk‐billing rates or out‐of‐pocket costs for different types of general practitioner services (eg, levels A–E) or specialist care. The fees charged by specialists are substantially higher than those of general practitioners and are much more variable.26 Finally, our analyses are based on the total number of all general practitioner visits, not the proportion of patients who are always bulk‐billed. Broader patient‐, provider‐, and system‐level factors that affect fees should be investigated, including personal income and other health conditions, practice costs and size, perceived competition, and practitioner experience and skill level.27,28 Nevertheless, the large differences between SA3 regions after adjustment for age and sex suggest that a proportion of the variation in fees is probably unwarranted. Robust patient‐, provider‐, and practice‐level information, including information about patient–general practitioner encounters, will be required to investigate this question.

Conclusion

Out‐of‐pocket costs are an important determinant of access to health care. Although most general practitioner services in Australia are bulk‐billed, out‐of‐pocket costs for non‐bulk‐billed services are relatively high, particularly in remote and socio‐economically disadvantaged areas. Recent Medicare strategies for increasing bulk‐billing rates are unlikely to reduce out‐of‐pocket costs for people living in metropolitan areas not already receiving bulk‐billed care.

Box 1 – General practice services bulk‐billing rate, Australia, 2022, mean values by Statistical Area level 3 (SA3)*

---

* By quartile, adjusted by age group (0–15, 16–40, 40–65, 65 years or older) and sex (male, female). Maps for the eight capital cities are included in the Supporting Information, figure 1.

Box 2 – Out‐of‐pocket costs for non‐bulk‐billed general practitioner services, Australia, 2022, mean values by Statistical Area level 3 (SA3)*

---

* By quartile, adjusted by age group (0–15, 16–40, 40–65, 65 years or older) and sex (male, female). Maps for the eight capital cities are included in the Supporting Information, figure 2. Costs are reported as 2024 dollars.

Box 3 – Bulk‐billing rates and out‐of‐pocket costs, Australia, 2022, by SA3‐level socio‐economic disadvantage quintile*

---

* Adjusted by age group (0–15, 16–40, 40–65, 65 years or older) and sex (male, female). Costs in panel B are reported as 2024 dollars. Red dot: mean value. [Correction added on 20 December 2024, after first online publication: Image in Box has been replaced.]

Box 4 – Bulk‐billing rates and out‐of‐pocket costs, Australia, 2022, by SA3‐level remoteness category

---

MMM = Modified Monash Model.* Adjusted by age group (0–15, 16–40, 40–65, 65 years or older) and sex (male, female). Costs in panel B are reported as 2024 dollars. Red dot: mean value.