In Australia, despite having a universal health insurance system (Medicare), there are variations in access to healthcare services.1 One potential barrier to the use of these services is the personal out-of-pocket cost and the requirement for "up-front" payment of fees. After adjusting for a range of health and socioeconomic variables, higher out-of-pocket cost per consultation has been shown to be associated with lower use of general practice consultations by Australian women.2 Nationally, there has been a lack of data to examine out-of-pocket costs for subgroups of the population, particularly those who are less able to manage on their incomes and who have a greater need for services. Some people cannot budget for even modest healthcare costs on a regular basis.3
Australia has no legislation restricting how much a general practitioner can charge for a consultation. The patient is required to pay the difference between the charge for the consultation and the Medicare rebate, which, for most services, is set at 85% of the schedule fee fixed by the federal government (the remaining 15% is referred to as the "gap"). Under bulk-billing arrangements, patients may assign their Medicare rebate directly to the treating GP. The GP accepts the rebate as full payment and cannot charge the patient an additional fee. There are policies in place that attempt to protect patients from substantial out-of-pocket costs. Under the Medicare Safety Net scheme, once the cumulative gap payments for non-inpatient services for an individual or family exceed a set amount in a calendar year, the patient may receive the full Medicare schedule fee as the rebate for all services for the remainder of the calendar year. However, any amounts charged above the schedule fee by the GP must still be paid by the patient.
The greater access to bulk-billing and lower out-of-pocket costs in urban areas have been well documented.4 In the financial year 1998–99, non-specialist medical services delivered in capital cities had the lowest average out-of-pocket expenditures of $1.61 per service. In contrast, the figures for rural areas and remote centres were $3.72 and $5.73, respectively.4 However, the Medicare database used in these analyses does not include measures of health or socioeconomic status for individuals, so it has not been possible, in previous studies, to establish whether out-of-pocket costs are related to patients' health status or socioeconomic factors.
In contrast, access to linked data in our study has allowed us to examine some of these associations. Our analysis was conducted as part of the Australian Longitudinal Study on Women's Health (ALSWH). Data on a range of health, social and demographic factors have been collected since 1996 on a random sample of more than 40 000 women, of whom more than half live in rural areas. In addition, almost 23 000 of the women have consented to their survey data being linked to their Medicare data. Record linkage can reduce the effect of recall bias and respondent burden, and also allows analysis of relationships between the use of healthcare services and health and socioeconomic characteristics.
In April 1996, women from three age groups (18–23, 45–50 and 70–75 years) were sampled from the Medicare database, with oversampling of women living in rural and remote areas of Australia to provide adequate statistical power for analyses.5
Survey 1 was conducted in 1996. Survey 2, the first follow-up study, was conducted for each group over three successive years: 1998 (mid-age cohort), 1999 (older cohort) and 2000 (young cohort). Response rates at the recruitment stage cannot be exactly specified, as some women selected in the sample may not have received the invitation (eg, if they had died, or had changed their address without notifying Medicare). An estimated 41%–42% of the young women, 53%–56% of the mid-age women and 37%–40% of the older women agreed to participate in the longitudinal study.6 The respondents were broadly representative of the national population of women in the target age groups.6
The response rates for Survey 2 were 92% (mid-age cohort), 91% (older cohort) and 71% (young cohort). In the young cohort, most non-respondents to Survey 2 were lost to follow-up because of high mobility in this age group, but a representative sample was retained.
In 1997, and again in 1999, consent was sought from the participants for the Health Insurance Commission (HIC) to release Medicare and Department of Veterans' Affairs (DVA) medical claim details to the research team. Details of the methods used have been previously reported.7 Initially, 19 700 women consented to record linkage of their Medicare/DVA and survey data relating to 1995–1996. Consenters tended to have a higher education level and, among the older cohort only, were in better health than non-consenters.7 Following the second request for consent, in 1999, there were 22 633 consenters: 6219 young women, 8883 mid-age women and 7531 older women.
The Rural, Remote and Metropolitan Areas Classification8 was used to define the area of residence of the woman at the time of each consultation as urban (capital city, other metropolitan area) or rural (large rural centre, small rural centre, other rural areas, remote centre and other remote areas). If a woman lived in both urban and rural areas during a calendar year, the area where she had most of her general practice consultations was defined as her area of residence for that year.
Self-rated health was measured in Survey 2 by responses to the question "In general, would you say your health is 'excellent', 'very good', 'good', 'fair', or 'poor'?".
Education level was determined by asking participants about the highest qualification they had completed. Responses were grouped into two categories: school education only, or post-school qualifications (including a trade qualification, certificate/diploma or university degree). Level of education has been shown to be a good measure of socioeconomic status.9
For each year in the period 1995–2001, the average out-of-pocket cost per consultation for each woman was summarised by age cohort and area of residence. To allow for the effects of inflation, costs for all years were adjusted to 2001 dollar values using the Consumer Price Index published by the Australian Bureau of Statistics for the June quarter of each year.
Medicare/DVA data and Survey 2 responses were linked according to the calendar year in which Survey 2 was conducted, as this varied for each age cohort. Logistic regression analysis was performed using the SAS "Logistic" procedure10 to determine the extent to which health and socioeconomic variables were associated with bulk-billing. The outcome variable was defined for each woman as whether she had all her general practice consultations for that year bulk-billed or not. The explanatory variables for the analysis were area of residence, frequency of attendance, education and self-rated health.
There were almost one million general practice consultations over the seven-year period for the women who gave consent for their records to be linked. The percentage of women who had all their general practice consultations bulk-billed (and hence had no out-of-pocket costs) was about 20% higher in urban areas than rural areas, for all age groups (Box 1). Furthermore, mean out-of-pocket costs increased over time in rural areas as the use of bulk-billing decreased.
Women in the older age group were most likely to have all their general practice consultations bulk-billed, particularly those living in urban areas (Box 2). For all ages and areas, women with lower levels of education were more likely to have all their visits bulk-billed. However, there were only slightly higher rates of bulk-billing for older women in rural areas with a lower education level (61%) compared with women the same age with a higher education level (55%). These rates were considerably lower than for older women in urban areas (where 82% of women with lower education and 71% with higher education were bulk-billed).
For all age groups, women who lived in urban areas were more than twice as likely to have all their general practice consultations bulk-billed than women living in rural areas (Box 3). Women with lower levels of education were significantly more likely to receive bulk-billing for all their consultations. Being a frequent attender was also associated with bulk-billing for mid-age and older women (but not young women). After adjusting for other variables, poorer self-rated health was associated with having all visits bulk-billed for young and mid-aged women, but not for older women.
Our study presents new findings about access to bulk-billing and changes over time in costs for general practice consultations for women in Australia. Our results demonstrate geographical inequities in access to bulk-billing, although there is some consideration given to women in poorer health and with lower socioeconomic status. The major finding of our analysis of linked data is that not only are the rates of bulk-billing lower, and declining, in rural areas, but the out-of-pocket costs are increasing. These costs are increasing as women age, as this is a longitudinal study of the same women over time. It might be expected that, among the older cohort, out-of-pocket costs would decrease as women age and their health deteriorates, but this was not the case, especially in rural areas.
Women living in urban areas who were in good health and had better education were more likely to be bulk-billed than their rural counterparts in poor health or of lower socioeconomic status. A strength of the longitudinal design of the study is that these trends can be monitored over the next few years to evaluate whether the situation is improving or not. In addition, the women in our study also provide self-reported ratings of their access to bulk-billing and medical services, which can be linked to their actual health service utilisation data from the HIC. There are no other comparable published studies in Australia that examine out-of-pocket costs from the perspective of individuals and their medical and social circumstances.
A strength of our study is that the results are based on a national random sample of women rather than a sample of practice attenders, and so can be more readily generalised to the population. A limitation is that our results are based on women who consented to record linkage, thus creating a socioeconomic bias in the sample. Women who consented to linkage had a higher level of education and, among the older cohort, were in better health. However, as there was no bias among consenters in terms of urban or rural area of residence, the geographical inequities demonstrated in our study are real. The proportion of women who have all their consultations bulk-billed, as reported here, may be an underestimate of the proportion in the population, as the consenters in this study tended to have higher socioeconomic status and so may be less likely to be bulk-billed. However, the trends over time showing the decreasing use of bulk-billing and increasing out-of-pocket costs in rural areas would be unaffected by any socioeconomic bias in the sample of consenters.
The increased out-of-pocket expenditure in rural areas may in part reflect increased costs of healthcare service delivery in rural areas. Policy changes are required to enable women in rural and remote areas to have access to affordable healthcare services.
1: Mean out-of-pocket cost per general practice consultation per woman, 1995–2001, by age group and area of residence, adjusted to 2001 dollar values
2: Percentage of women who had all their general practice consultations bulk-billed, by age group, area of residence, and measures of health and socioeconomic status
|
Women aged 22–27* |
Women aged 47–52† |
Women aged 73–78‡ |
||||||||
|
Urban |
Rural |
Urban |
Rural |
Urban |
Rural |
|||||
Self-rated health |
|
|
|
|
|
|
|||||
Excellent, very good or good |
50% |
30% |
44% |
24% |
79% |
60% |
|||||
Fair or poor |
56% |
34% |
55% |
32% |
81% |
59% |
|||||
Education |
|
|
|
|
|
|
|||||
Post-school qualifications |
49% |
29% |
40% |
23% |
71% |
55% |
|||||
School only |
55% |
33% |
49% |
26% |
82% |
61% |
|||||
Frequency of consultations |
|
|
|
|
|
|
|||||
Below 95th percentile |
50% |
30% |
44% |
24% |
79% |
60% |
|||||
95th percentile and above |
49% |
36% |
59% |
44% |
83% |
68% |
|||||
* Data for year 2000. † Data for year 1998. ‡ Data for year 1999. |
|||||||||||
3: Factors associated with having all general practice consultations bulk-billed, analysed by multiple logistic regression, by age cohort
|
Women aged 22–27* |
Women aged 47–52† |
Women aged 73–78‡ |
||||||||
Area of residence |
|
|
|
||||||||
Rural |
1.00 |
1.00 |
1.00 |
||||||||
Urban |
2.40 (2.12–2.72) |
2.52 (2.28–2.78) |
2.62 (2.34–2.94) |
||||||||
Frequency of consultations |
|
|
|
||||||||
Below 95th percentile |
1.00 |
1.00 |
1.00 |
||||||||
95th percentile and above |
0.91 (0.71–1.18) |
1.89 (1.55–2.30) |
1.42 (1.09–1.86) |
||||||||
Education |
|
|
|
||||||||
Post-school qualifications |
1.00 |
1.00 |
1.00 |
||||||||
School only |
1.24 (1.10–1.42) |
1.30 (1.17–1.44) |
1.49 (1.30–1.71) |
||||||||
Self-rated health |
|
|
|
||||||||
Excellent, very good or good |
1.00 |
1.00 |
1.00 |
||||||||
Fair/poor |
1.27 (1.05–1.53) |
1.30 (1.12–1.51) |
0.98 (0.87–1.11) |
||||||||
* Data for year 2000. † Data for year 1998. ‡ Data for year 1999. OR = odds ratio. |
|||||||||||
- Anne F Young1
- Annette J Dobson2
- 1 Research Centre for Gender and Health, University of Newcastle, Callaghan, NSW.
- 2 School of Population Health, University of Queensland, Herston, QLD.
The Australian Longitudinal Study on Women's Health, which was conceived and developed by groups of inter-disciplinary researchers at the universities of Newcastle and Queensland, is funded by the Commonwealth Department of Health and Ageing. We thank all participants for their valuable contribution to this project.
None identified.
- 1. Young AF, Dobson AJ, Byles JE. Access and equity in the provision of general practitioner services in Australia. Aust N Z J Public Health 2000; 24: 474-480.
- 2. Young AF, Dobson AJ, Byles JE. Determinants of general practice use among women in Australia. Soc Sci Med 2001; 53: 1641-1651.
- 3. Loftus-Hills A. The costs of care: disadvantaged consumers' financial access to health care. Melbourne: Health Issues Centre, 1993.
- 4. Commonwealth Department of Health and Aged Care. General practice in Australia: 2000. Canberra: Commonwealth Department of Health and Aged Care, 2000.
- 5. Brown WJ, Bryson L, Byles JE, et al. Women's Health Australia: recruitment for a national longitudinal cohort study. Women Health 1998; 28: 23-40.
- 6. Brown WJ, Dobson AJ, Bryson L, Byles JE. Women's Health Australia: on the progress of the main study cohorts. J Womens Health Gend Based Med 1999; 8: 681-688.
- 7. Young AF, Dobson AJ, Byles JE. Health services research using linked records: who consents and what is the gain? Aust N Z J Public Health 2001; 25: 417-420.
- 8. Department of Primary Industries and Energy and Department of Human Services and Health. Rural, remote and metropolitan areas classification: 1991 census edition. Canberra: Australian Government Publishing Service, 1994.
- 9. Mishra GD, Ball K, Dobson AJ, et al. Which aspects of socioeconomic status are related to health in mid-aged and older women? Int J Behav Med 2002; 9: 263-285.
- 10. SAS Institute Inc. SAS/STAT User's guide. Version 8. Cary, NC: SAS Institute Inc., 1999.
Abstract
Objective: To describe the changes in bulk-billing and out-of-pocket costs for Australian general practice consultations over the period 1995–2001.
Design: Retrospective analysis of 1996–2001 survey data from the Australian Longitudinal Study on Women's Health (ALSWH), linked with Medicare and Department of Veterans' Affairs (DVA) data on general practice consultations from 1995 to 2001.
Participants: 22 633 women who gave consent to linkage of their ALSWH data with Medicare/DVA records. In 1996, women in the "young" cohort (n = 6219) were aged 18–23 years, those in the "mid-age" cohort (n = 8883) were aged 45–50 years, and those in the "older" cohort (n = 7531) were aged 70–75 years.
Outcome measures: Out-of-pocket costs paid by patients for general practice consultations, by calendar year, urban/rural area of residence, age, frequency of attendance, self-rated health, and education level.
Results: For each age group and year studied, the use of bulk-billing was lower in rural areas than in urban areas. For example, in 2000, the percentage of women in rural and urban areas, respectively, who had all their general practice consultations bulk-billed was 31% v 52% (young women), 24% v 45% (mid-age women) and 58% v 79% (older women). There has been a steady decline in bulk-billing for general practice consultations in rural areas since 1995. The average out-of-pocket cost per consultation for women in rural areas was higher than the cost for women living in urban areas. After adjusting for age, health and socioeconomic factors, women living in urban areas were more than twice as likely to have all their consultations bulk-billed as women living in rural areas: odds ratio (OR), 2.4 (95% CI, 2.1–2.7) (young women); OR, 2.5 (95% CI, 2.3–2.8) (mid-age women); OR, 2.6 (95% CI, 2.3–2.9) (older women).
Conclusions: In Australia, the geographic differential in the cost of general practice consultations is widening. Policy changes are required to enable women in rural and remote areas to have access to affordable healthcare services.