Regional variation in the incidence of end-stage renal disease in Indigenous Australians

Alan Cass, Joan Cunningham, Zhiqiang Wang and Wendy Hoy
Med J Aust 2001; 175 (1): 24-27.
Published online: 2 July 2001

Indigenous health

Regional variation in the incidence of end-stage renal disease in Indigenous Australians

Alan Cass, Joan Cunningham, Zhiqiang Wang and Wendy Hoy

MJA 2001; 175: 24-27

Abstract - Methods - Results - Discussion - Acknowledgements - References - Authors' details
- - More articles on Aboriginal health


Objective: To evaluate regional variation in the incidence of end-stage renal disease (ESRD) in Indigenous Australians, and to examine the proximity to ESRD treatment facilities of Indigenous patients.
Design: Secondary data review, with collection of primary data regarding patients' place of residence before beginning ESRD treatment.
Participants: Indigenous ESRD patients who commenced treatment in Australia during 1993-1998.
Methods: We obtained data from the Australian and New Zealand Dialysis and Transplant Registry regarding 719 Indigenous patients who started ESRD treatment between 1 January 1993 and 31 December 1998. We obtained primary data from the treating renal units to determine the place of residence before beginning renal replacement therapy. We calculated the average annual incidence of ESRD for each of the 36 Aboriginal and Torres Strait Islander Commission regions using population estimates based on the 1996 Census, and calculated standardised incidence ratios with 95% confidence intervals for each region. We compared the number of cases with the treatment facilities available in each region.
Main outcome measure: Regional standardised ESRD incidence for Indigenous Australians referenced to the total resident population of Australia.
Results: Standardised ESRD incidence among Indigenous Australians is highest in remote regions, where it is up to 30 times the national incidence for all Australians. In urban regions the standardised incidence is much lower, but remains significantly higher than the national incidence. Forty-eight per cent of Indigenous ESRD patients come from regions without dialysis or transplant facilities and 16.3% from regions with only satellite dialysis facilities.
Conclusions: There is marked regional variation in the incidence of ESRD among Indigenous Australians. Because of the location of treatment centres, there is inequitable access to ESRD treatment services for a significant proportion of Indigenous patients.

In Australia, geographical differences in morbidity and mortality have been demonstrated.1-4 In general, people living in rural and remote areas have higher death and hospital separation rates, which have been attributed to differences in socioeconomic status,1,3-5 access to health services,2,6,7 ethnicity4 and racial discrimination.8 Indigenous Australians constitute a disproportionate number of new patients commencing end-stage renal disease (ESRD) treatment:9 in 1997, the incidence rate (adjusted for age and sex) for Indigenous Australians beginning ESRD treatment was nearly nine times that of non-Indigenous Australians.10 Although epidemics of renal disease among Indigenous Australians in defined areas11 have been documented, there have been no systematic reports of the regional patterns of ESRD incidence in Australia. In this study we have attempted to determine these patterns and to examine the accessibility of ESRD treatment facilities for Indigenous people.



The Australia and New Zealand Dialysis and Transplant Registry (ANZDATA) maintains a database of patients treated by maintenance dialysis or renal transplantation in Australia. All renal units that provide ESRD treatment in Australia participate in the Registry. Postcode of residence at the start of treatment is collected for all new patients entered into the ANZDATA Registry. ANZDATA maintains a list of hospital renal transplant services, tertiary referral units and satellite dialysis units. Satellite units are defined as dialysis facilities, generally staffed by specialist nurses, that are geographically separate from hospital nephrology services.

Data validity

Postcode of residence at the start of treatment is an imperfect indicator of the usual place of residence before starting treatment. In remote areas of Australia a single postcode may apply to many communities across a vast area. Furthermore, patients may be required to relocate to a major regional centre to access dialysis services; thus, their postcode at the start of treatment may not reflect their previous usual place of residence. To determine the usefulness of postcode data, we reviewed 104 Indigenous patients who commenced ESRD treatment from 1993 to 1998 at Royal Darwin Hospital, Northern Territory. For these patients the previous usual place of residence was known. Fifty-one patients (49%) had postcodes in the Darwin region, but only nine of these 51 patients previously lived in this region. The other 42 had relocated to Darwin to commence dialysis and were from communities across the "Top End", extending from the Torres Strait in Queensland to Geraldton in Western Australia. As a result of this audit, we decided to collect primary data from each treating renal unit regarding the previous place of usual residence of their Indigenous patients.

Indigenous identification was based upon self-identification and discussion with the treating physician. There is often significant concern about the quality of Indigenous identification in morbidity, mortality and demographic data sets. However, we believe that racial identification in the ANZDATA registry is good. A survey form is completed every six months for all patients on maintenance dialysis or with functioning renal transplants. In this survey, question five is about "Racial origin" and includes a prompt regarding Indigenous status. ESRD patients have regular contact with renal services from the time of diagnosis, through intensive maintenance therapy until death. There is heightened awareness of renal disease in Indigenous Australians and multiple opportunities exist to reconfirm data accuracy.


From 1 January 1993 to 31 December 1998, 719 Indigenous patients started treatment for ESRD in Australia. We determined the previous place of usual residence for 680 (94.6%). For 38 patients (5.3%) we used the postcode at entry as an indicator of previous place of usual residence. No geographical data were obtainable for one patient.


We used the 36 Aboriginal and Torres Strait Islander Commission (ATSIC) regions (Boxes 1 and 2) as our geographic units for analysis. These are legally prescribed administrative areas and the smallest geographical areas for which accurate resident population estimates for the Indigenous population are available.12 We assigned the place of usual residence for Indigenous patients to the appropriate ATSIC regions. We assigned the 38 patients for whom we could not obtain exact information on previous place of residence to ATSIC regions according to their postcode at the time of beginning ESRD treatment.

Statistical analysis

Using population estimates based on the 1996 Census, we calculated the average annual incidence of ESRD in the 36 ATSIC regions. We used ABS estimates of the Indigenous population, derived using Census information on place of usual residence. These estimates are adjusted for net Census undercount and non-response to the Census question about Indigenous status.13 We used indirect standardisation to calculate an age- and sex-standardised incidence ratio with 95% confidence intervals for each region. Rates for the total Australian resident population were used as the reference (standardised incidence ratio equals incidence in the Indigenous population divided by incidence in the total Australian population, after adjusting for differences in the age and sex composition of both populations). Statistical analysis was performed using Stata.14

Ethical approval

We obtained ANZDATA approval to analyse geographic data for Indigenous patients starting treatment for ESRD between 1 January 1993 and 31 December 1998. We also obtained approval for the study from the joint institutional ethics committee of Royal Darwin Hospital and the Menzies School of Health Research. The head of each renal unit gave written consent for us to access potentially identifying patient data in order to determine the previous place of usual residence for Indigenous ESRD patients.


Mapping reveals significant variation in the incidence of ESRD among Indigenous Australians. The areas of highest incidence (up to 1300 cases per million per year) were the remote regions of Tennant Creek, Aputula and Jabiru in the Northern Territory, Warburton and Kalgoorlie in Western Australia, and Ceduna in South Australia (Box 1). The areas of lowest incidence (less than 100 per million per year) were the regions of Rockhampton and Brisbane in Queensland, Sydney and Queanbeyan in NSW/ACT, Wangaratta (which includes much of eastern Victoria) and Hobart (which encompasses all of Tasmania) (Box 1). The standardised incidence ratio for ESRD (compared with the total national population incidence) ranged from less than two in Rockhampton, Sydney, Queanbeyan and Wangaratta to more than 25 in Aputula, Kalgoorlie and Tennant Creek (Box 2). There were no ESRD patients identified as Indigenous in Tasmania in the six-year period.

Tertiary renal services, particularly transplant services, are located within significant population centres such as capital cities. Three hundred and forty-five (48.0%) of the 719 Indigenous ESRD patients lived in ATSIC regions without ESRD treatment facilities (Box 2). A further 117 (16.3%) lived in regions with only satellite dialysis facilities (Box 2). Most Indigenous patients must travel hundreds of kilometres to access transplant services, which are located in Perth, Adelaide, Melbourne, Sydney, Newcastle and Brisbane.


In this study we have demonstrated a large gradient in Indigenous ESRD incidence from urban to remote regions and highlighted inequitable access for remote patients to treatment facilities. However, even in urban regions, the Indigenous ESRD incidence was high after age and sex standardisation. Poor Indigenous health outcomes are not confined to the most disadvantaged or most remote regions, but exist across the Indigenous population.

The quality of Indigenous identification is a potential concern in our study. ANZDATA relies upon self-identification and discussion with the treating physician. Self-identification is the method used by the Australian Bureau of Statistics in census collections and is generally used in health-related data collection. We believe that the quality of identification in our study is high owing to the ongoing intensive interaction of ESRD patients with medical and nursing staff, Indigenous status being a prominent question in the six-monthly survey form, and the strong awareness of Indigenous ESRD among nephrologists. The most likely error would be the failure to identify all urban Indigenous ESRD patients. This would result in an underestimate of the true Indigenous ESRD incidence in urban areas and an overestimate of the gradient from urban to remote Indigenous ESRD incidence. Yet, as this gradient is so large, representing an almost 20-fold variation in standardised ESRD incidence, it can not be entirely explained by problems with Indigenous identification. The very high standardised incidence ratios for Indigenous people in remote areas would not change.

These results have significant implications for the delivery of services to Indigenous people with ESRD. Satellite facilities opened in the Jabiru region in 1999 and the Katherine region in 2000 (after the patients in this study commenced treatment). Of the 16 regions with the highest Indigenous ESRD incidence, at the beginning of 2001 only Kalgoorlie, Jabiru (Tiwi Islands), Geraldton, Katherine and South Hedland had satellite dialysis units. A satellite unit is scheduled to open soon in Broome and recommendations have been accepted to establish a satellite haemodialysis service in the Torres Strait as part of the recent Queensland Renal Strategy.15 We recognise the significant difficulties related to the establishment and maintenance of renal treatment facilities in remote locations. These include high construction costs, poor reliability of electricity and water supply, variable water quality, difficulties in training and retaining specialised nursing staff, infrequent access to medical staff and provision of housing for patients returning to live in their local community. Despite these difficulties, treatment facilities have been established in some of the most remote communities in Australia.

Even with the availability of satellite units, initiation of ESRD treatment usually requires a prolonged stay in a major urban centre. During this stay, vascular or peritoneal access for dialysis is created, the patient starts and is stabilised on treatment and learns skills required for self-care in order to return to a remote satellite dialysis unit. We should develop more innovative methods of patient education, training for self-care and delivery of treatment to allow patients to remain within their communities whenever possible. Improving prevention and treatment services in high-incidence areas should be a priority.

Indigenous people living in remote communities demand more equitable access to dialysis services,16 regardless of practical problems related to the establishment of remote treatment facilities. The need to relocate to distant urban areas to access treatment affects the patient, patient's family and community. A recent study of ESRD among Aboriginal people of central Australia concluded: "This level of illness and death [due to ESRD] represents Aboriginal family trauma and loss on a shocking scale, described without exaggeration as sorrows nearly every year [because] the young and the old are dying". 17


The data reported here have been supplied by the Australia and New Zealand Dialysis and Transplant Registry. The interpretation of these data is the responsibility of the authors and should not be seen as an official policy or interpretation of the Australia and New Zealand Dialysis and Transplant Registry. Dr Alan Cass receives postgraduate research scholarship funding from the Colonial Foundation. Dr Joan Cunningham is supported by a fellowship from the Menzies Foundation. We thank Dr Mark Thomas, Dr Paul Snelling, Dr Meshak Kirubakaran, Dr Tim Furlong, Dr Peter de Jersey and the heads of renal units for providing information regarding the place of usual residence for their patients.


  1. Glover J, Harris K, Tennant S. A social health atlas of Australia. 2nd ed. Adelaide: Public Health Information Development Unit, University of Adelaide, 1999.
  2. Sexton PT, Sexton TL. Excess coronary mortality among Australian men and women living outside the capital city statistical divisions. Med J Aust 2000; 172: 370-374 [see comments Med J Aust 2000; 172: 360-361].
  3. National Health Strategy. Enough to make you sick: how income and environment affect health, Research Paper No. 1. Melbourne: National Health Strategy Unit, 1992.
  4. Taylor R, Chey T, Bauman A, Webster I. Socio-economic, migrant and geographic differentials in coronary heart disease occurrence in New South Wales. Aust N Z J Public Health 1999; 23: 20-26.
  5. Turrell G, Mathers CD. Socioeconomic status and health in Australia. Med J Aust 2000; 172: 434-438.
  6. Heller RF. Mortality from cardiovascular disease is too high outside capital cities [editorial]. Med J Aust 2000; 172: 360-361.
  7. McLaren B. Renal failure in Arnhem Land: missed opportunities for prevention and treatment. Aust J Rural Health 1996; 4: 61-66.
  8. Lowe M, Kerridge IH, Mitchell KR. 'These sorts of people don't do very well': race and allocation of health care resources. J Med Ethics 1995; 21: 356-360.
  9. Disney A, Russ G, Walker R, et al, editors. ANZDATA Registry Report 1999. Adelaide: Australia and New Zealand Dialysis and Transplant Registry, 1999.
  10. Cass A, McDonald SP, Wang Z. Australians with renal disease: a new national survey [letter]. Med J Aust 1999; 171: 444.
  11. Spencer JL, Silva DT, Snelling P, Hoy WE. An epidemic of renal failure among Australian Aboriginals. Med J Aust 1998; 168: 537-541 [see comments Med J Aust 1998; 168: 532-533 and Med J Aust 1999; 170: 191-192].
  12. Australian Bureau of Statistics. Population issues, Indigenous Australians. Canberra: Australian Bureau of Statistics, 1999.
  13. Australian Bureau of Statistics. Experimental estimates of the Aboriginal and Torres Strait Islander population. Canberra: Australian Bureau of Statistics, 1998.
  14. Stata version 6 [computer software]. College Station, TX: Stata Corporation, 1999.
  15. Schmidt B. Northern zone renal services plan 2000-2010, 2000.
  16. Devitt J, McMasters A. On the machine: Aboriginal stories about kidney troubles. Alice Springs: IAD Press, 1998.
  17. Devitt J, McMasters A. Living on medicine: a cultural study of end-stage renal disease among Aboriginal people. Alice Springs: IAD Press, 1998.

(Received 13 Sep 2000, accepted 22 Mar 2001)

Authors' details

Menzies School of Health Research, Casuarina, NT.
Alan Cass, GradDipClinEpi, FRACP, PhD Student and Nephrologist.
Joan Cunningham, ScD, Epidemiologist and Menzies Fellow.
Zhiqiang Wang, PhD, Biostatistician.
Wendy Hoy, MB BS, FRACP, Head of Renal Unit.

No reprints will be available from the authors.
Correspondence: Dr Alan Cass, Menzies School of Health Research, PO Box 41096, Casuarina, NT, 0811.

Make a comment


Map of Australia
Map numbers refer to Aboriginal and Torres Strait Islander Commission regions as specified in Box 2
Back to text
2: End-stage renal disease among Indigenous Australians from 1993 to 1998
ATSIC region
(map references)
Standardised incidence
ratio* (95% CI)

Tennant Creek (35) 30 31.05 (20.96 - 44.33)
Kalgoorlie (27) S 23 27.75 (17.60 - 41.64)
Aputula (33) 58 25.03 (19.01 - 32.36)
Warburton (23) 20 22.77 (13.91 - 35.17)
Ceduna (18) 10 22.48 (10.78 - 41.34)
Jabiru (31) 45 21.87 (15.95 - 29.26)
Geraldton (28) S 25 18.20 (11.78 - 26.86)
Mount Isa (11) 33 17.74 (12.21 - 24.91)
Kununurra (22) 22 16.85 (10.56 - 25.51)
Katherine (32) 30 15.64 (10.56 - 22.33)
Torres Strait (15) 28 14.99 (9.96 - 21.66)
South Hedland (25) S 18 14.75 (8.74 - 23.30)
Derby (26) 16 13.40 (7.66 - 21.76)
Nhulunbuy (34) 21 11.74 (7.27 - 17.94)
Cooktown (12) 21 11.61 (7.19 - 17.75)
Broome (21) 11 11.47 (5.73 - 20.53)
Port Augusta (19) S 17 10.45 (6.09 - 16.74)
Bourke (2) S 21 10.16 (6.29 - 15.53)
Townsville (16) T, S 35 9.41 (6.55 - 13.08)
Cairns (10) T, S 35 8.71 (6.07 - 12.12)
Alice Springs (30) T, S 11 8.55 (4.27 - 15.30)
Narrogin (24) S 13 8.20 (4.37 - 14.02)
Darwin (36) T, S 17 7.02 (4.09 - 11.24)
Perth (20) Tx, T, S 29 6.70 (4.48 - 9.61)
Adelaide (17) Tx, T, S 15 4.62 (2.58 - 7.61)
Tamworth (5) T, S 12 4.18 (2.16 - 7.30)
Roma (14) T, S 8 3.70 (1.60 - 7.28)
Coffs Harbour (3) Tx, T, S 24 3.68 (2.36 - 5.47)
Ballarat (8) Tx, T, S 10 3.42 (1.64 - 6.28)
Wagga Wagga (6) T, S 14 2.98 (1.63 - 5.00)
Brisbane (9) Tx, T, S 17 2.51 (1.46 - 4.02)
Rockhampton (13) T, S 5 1.78 (0.58 - 4.16)
Sydney (4) Tx, T, S 16 1.77 (1.01 - 2.88)
Queanbeyan (1) T, S 4 1.75 (0.48 - 4.48)
Wangaratta (7) Tx, T, S 4 1.39 (0.38 - 3.55)
Hobart (29) T, S 0 0.00 (0.00 - 1.03)

* Indirectly standardised to the rates for the total Australian resident population.
†Tx = transplant service, T = tertiary renal unit, S = satellite dialysis unit.
Geographical data were unobtainable for one patient.
Back to text
  • Alan Cass
  • Joan Cunningham
  • Zhiqiang Wang
  • Wendy Hoy



remove_circle_outline Delete Author
add_circle_outline Add Author

Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Responses are now closed for this article.