Connect
MJA
MJA

Diabetes care in remote northern Australian Indigenous communities

Robyn A McDermott, Fiona Tulip and Barbara Schmidt
Med J Aust 2004; 180 (10): 512-516.
Published online: 17 May 2004

Abstract

Objective: To assess primary care processes and clinical characteristics of adults with diabetes in remote northern Australian Indigenous communities.

Design: Clinical audit from diabetes registers in 21 remote primary healthcare centres in the Torres Strait Health Service District (n = 921), three in Cape York, Queensland (n = 252), and three in the Northern Territory (n = 194), between September 2002 and February 2003.

Participants and setting: Aboriginal and Torres Strait Islander adults with diabetes who were receiving their routine diabetes care in these 27 centres.

Main outcome measures: Provision of regular checks for weight, blood pressure, glycaemia (HbA1c), proteinuria, lipid levels, renal function, eyes and feet, influenza and pneumococcal vaccination. Weight, blood pressure and glycaemic control.

Results: Most routine diabetes checks were delivered according to recommended schedules, except for eye and foot checks in the NT. There were uniformly high rates of appropriate treatment for hypertension and albuminuria, but low rates of insulin treatment and self-monitoring despite a high mean HbA1c level (8.9%). Vaccination rates were low in the NT. Torres Strait Islanders with diabetes were significantly heavier than Aboriginals, but had lower mean diastolic blood pressure (77.3 mmHg compared with 79.5 mmHg) and lower prevalence of albuminuria and smoking.

Conclusion: A high proportion of Aboriginals and Torres Strait Islanders requiring treatment for high blood pressure and proteinuria are receiving it. However, there is dissonance between the relatively high rates of routine checks and apparent lack of therapeutic action on glycaemia. More intensive management of glycaemia, including improved nutrition, exercise and (probably) insulin, is required to reduce microvascular complications.

Type 2 diabetes is increasing rapidly in Australia, following sharp rises in overweight and obesity in the general population.1 It is contributing to a steep escalation in demand for expensive tertiary healthcare services which produce relatively small marginal health gains.2,3 Diabetes is also frequently accompanied by chronic conditions such as depression, which significantly reduce quality of life.

Indigenous Australians have the highest prevalence of diabetes in the country and an excess of avoidable complications and early death.4 Many of these complications can be reduced with appropriate community-based primary healthcare interventions. There is growing evidence that a structured approach to caring for people with chronic conditions can significantly improve patient satisfaction and quality of life, clinical outcomes and cost to the community.5 However, structured approaches require a considerable change in the way primary care is organised, moving from a reactive, acute-care model to one in which early detection is important, care is planned with patients, registers and recall systems are linked to appropriate action, and there is a way of measuring quality and outcomes of care.6

We examined diabetes care processes and patient characteristics in 27 primary healthcare centres in far north Queensland and the Northern Territory.

Methods
Sample selection

Twenty-seven remote Indigenous communities participated in a clinical audit, which was voluntary and formed the basis of an ongoing quality improvement process for diabetes care. This included all 21 primary care centres in Torres Strait, three in Cape York and three in the Northern Territory. The six mainland centres were included as they were already participating in a chronic disease quality improvement program. They are not different from other centres in their districts with respect to ethnicity, prevalence of diabetes, service model or remoteness.

The audit included all adults (aged 15 years and older) with diabetes who were on existing clinic-based registers in each participating community. Excluded were people receiving dialysis and people not receiving most of their care from that centre. The patients’ clinical records formed the basis of the audit, and data were abstracted from the clinic file by a registered nurse according to a standard protocol, which has been reported elsewhere.7 The audit was performed between August 2002 and February 2003.

The audit also attempts to describe any differences in patient characteristics between those attending for regular checks and those not attending.

Comparison data

For comparison with diabetes care in non-Indigenous people, we compared our results with data from the National Association of Diabetes Centres (NADC) Audit and Benchmarking study.8 The NADC report summarises diabetes care for 2077 adults with type 2 diabetes from 24 diabetes centres around Australia during 2002. It is a convenience sample and is not necessarily representative of all people with diabetes in Australia. In particular, the NADC sample is more likely to include patients whose diabetes is not managed well, or those referred with more complex or complicated diabetes.8 We also compared the rate of glucose self-monitoring with data from the Queensland Health Diabetes Survey conducted in 2000,9 as this information was not available from the NADC report.

Definitions

Diabetes was confirmed according to the 1991 World Health Organization (WHO) criteria,10 or the American Diabetes Association criteria.11 Patients were classified as hypertensive if they had blood pressure > 140/90 mmHg or had been prescribed antihypertensive medication. Patients were classified as having dyslipidaemia if they had total cholesterol level > 5.5 mmol/L or triglyceride level ≥ 2.1 mmol/L or they had been prescribed lipid-lowering treatment. Macroalbuminuria was defined as urinary albumin/creatinine ratio > 34. Eye examination included retinal photography or funduscopy by a non-ophthalmologist.

Unless specified otherwise, current guidelines recommend these checks at least annually.12

Ethical approval

Ethical approval was obtained from the Cairns Base Hospital Ethics Committee and the Northern Territory Human Research Ethics Committee, with support from the relevant Indigenous bodies.

Statistical analysis

Data were abstracted from clinical records to MS Access files in the field. Univariate and multivariate analyses were performed using SPSS for Windows.13 Means and proportions were compared using Student’s t tests and χ2 tests for normally distributed data, or F statistics otherwise. Stepwise logistic regression was used to estimate adjusted odds ratios and 95% confidence intervals.

Results
Patient characteristics

There were 1367 adults with confirmed diabetes on registers in the 27 centres. The mean patient age was 52 years, and 59% were women. Reported mean duration of diabetes was significantly shorter in the NT (6.8 years) than in Queensland (7.9 years). Torres Strait Islanders were significantly heavier than Aboriginals. Reported smoking was lower among Torres Strait Islanders than for Aboriginal patients (Box 1).

Diabetes care processes

Patients who had been weighed in the previous 12 months were more likely to be older (mean difference, 2.17 years; 95% CI, 0.5–3.8 years; P = 0.01), female (OR, 1.3; 95% CI, 1.09–1.72) and had a slightly longer duration of known disease than those not weighed in the previous 12 months.

Patients who had had their HbA1c level measured in the previous 6 months were similar in age and sex distribution to those who had not, but had been diagnosed for longer (mean duration of diabetes, 8.2 versus 6.9 years; mean difference, 1.2 years; 95% CI, 0.62–1.94 years; P = 0.004). Female and older patients were more likely to have had blood pressure checked in the previous 12 months (OR for women v men, 1.9; 95% CI, 1.32–2.78). There was no difference in age, sex or duration of diabetes in the likelihood of having urinary albumin/creatinine ratio measured in the previous 12 months.

Routine checks were generally done according to the recommended schedule, with lower rates in the NT, including apparently poorer access to specialist ophthalmology services (Box 2). In particular, there was good attention to blood pressure checks, and most patients with high blood pressure had been prescribed antihypertensive medication, with correspondingly relatively high rates of acceptable blood pressure control (< 140/90 mmHg). Similarly, nearly all patients with albuminuria had been prescribed angiotensin-converting enzyme (ACE) inhibitors. Vaccination rates were lowest in the NT.

Few patients self-monitored for glycaemia and HbA1c level. Very few patients appeared to have been prescribed insulin treatment, either alone or combined, for hyperglycaemia, despite the very high HbA1c levels (Box 2).

Intermediate clinical outcomes

Glycaemic control was poor overall (mean HbA1c level, 8.9%) and the Torres Strait Islanders had the lowest proportion with good glycaemic control (HbA1c level, < 7%). Diastolic blood pressure and albuminuria were significantly lower among Torres Strait Islanders (Box 3).

Despite having much lower mean weight and body mass index (BMI) than Torres Strait Islanders, Aboriginals had higher risk of hypertension and albuminuria, as well as higher rates of tobacco smoking. The excess risk for albuminuria among Aboriginals remained significant after controlling for age, sex, duration of diabetes, blood pressure, HbA1c level, smoking and weight (adjusted OR, 2.14; 95% CI, 1.36–3.36; P = 0.001). In this Indigenous population, albuminuria was also positively associated with male sex, blood pressure, HbA1c level and duration of diabetes.

Comparison with non-Indigenous people

Our sample of Indigenous people were on average more than 10 years younger, more likely to be female, had similar or better blood pressure control, but much poorer glycaemic control and higher prevalence of albuminuria, than non-Indigenous people with diabetes reported from Australian diabetes centres.8 These Indigenous people were much less likely to have been prescribed insulin, either alone or combined with oral treatment, despite much higher mean HbA1c level (Box 4). In addition, they were much less likely to be self-monitoring for glucose level compared with a Queensland non-Indigenous sample in 2000.9

Discussion

Our audit showed that diabetes checks are being routinely conducted in remote Aboriginal and Torres Strait Islander communities. Hypertension and albuminuria are being treated appropriately, but glycaemia is not well controlled in these populations. Torres Strait Islanders are heavier than Aboriginal people with diabetes and have poorer glycaemic control, but are more likely to be self-monitoring and receiving insulin treatment. Aboriginals are more likely to have albuminuria than Torres Strait Islanders. Those more likely to attend for routine checks in this sample were female, older, and had diabetes for longer than those who did not regularly attend.

Although all adults with diabetes on clinic registers in the Torres Strait District were counted in the audit, we included only three communities from Cape York and three from the NT, thus limiting generalisability for Aboriginal communities. However, the Aboriginal communities in the study were not different from others in their respective districts demographically or in the type of healthcare service model provided. Also, as the NADC sample is probably biased more towards people with severe and complex diabetes than those who are not seen in diabetes centres, the differences between the Indigenous and national samples reported here would probably be greater if a more representative (and therefore probably more healthy) national comparison were used.

There is dissonance between the relatively high rates of routine checks and apparent lack of therapeutic action on glycaemia among Indigenous patients compared with general practice in New South Wales.14,15

Observational studies suggest that the level of glycaemic control is related to the progression of renal disease.16 High rates of albuminuria have been reported in other Australian Indigenous groups, with and without diabetes, and are associated with rapid progression to end-stage renal disease. The variation in prevalence of albuminuria in different ethnic populations remains poorly explained by conventional risk factors and has been identified as a priority area for more research.8 Treatment with ACE inhibitors has been shown to be effective in slowing progression of nephropathy,17,18 and we found uptake of appropriate blood pressure and proteinuria treatment in the populations studied is quite high.

In contrast, hyperglycaemia is relatively neglected and may well now be the single most important contributor to continuing excessive rates of microvascular complications.

The evidence for better glycaemic control is compelling19 and is highly cost-effective.20,21 Recent trials suggest that regimens combining oral and insulin treatment are more effective than either agent alone, and newer insulin delivery devices make such options more attractive, even for people in remote areas.22-24 However, uptake of recommendations for tighter glycaemic control by doctors has been slow.25 Improving quality in diabetes care should emphasise earlier therapeutic changes, rather than more frequent testing in this population.

Responses from local treating clinicians in Queensland, when presented with these results to date, have tended to concentrate on compliance issues with oral hypoglycaemic agents. The problem of non-compliance with long-term treatments is not confined to Indigenous patients. It has been estimated that compliance with long-term therapy is less than 50% in developed countries.26 In settings where Western-trained doctors work with Indigenous patients, cross-cultural communication issues further cloud the complex area of treatment negotiation; for example, there may not be a shared understanding of the need for ongoing medication.27 Considerable debate has accompanied provider claims of non-compliant behaviour among people with diabetes, especially those with poor glycaemic control. Although we did not look directly at compliance with prescribed treatments, our data suggest that Indigenous patients are probably taking their blood pressure tablets regularly, perhaps with better compliance than their non-Indigenous counterparts. Therefore, we need to look at explanations other than non-compliance with treatment to explain high rates of glycaemia in these populations.

Explanations for the low rate of self-monitoring and insulin use rest in the notion of “therapeutic nihilism” on the part of some practitioners. Therapeutic nihilism occurs when there is a lack of confidence by doctors that existing treatment will be effective, so the doctors do not consider or offer potentially effective treatments.28 The apparent slowness of doctors to follow clinical guidelines has been described elsewhere as “clinical inertia”, and has been generally interpreted as a sign of low quality of care. Reasons given for clinical inertia include doctors’ overestimating the care they actually provide, lack of time, and lack of appropriate training.29

Other complicating factors that may lead to the low rate of self-monitoring and insulin use are the persisting poor socioeconomic conditions of most Indigenous patients, for whom the availability and affordability of glucose monitors, good food and physical exercise is low, and competing demands of family circumstances make self-care difficult.30 There is evidence that the fruit and vegetable intake among Indigenous adults in remote communities is particularly low; this is reflected in low serum red cell folate levels.31

Despite all these problems, it is urgent that we find more effective approaches to improving diabetes care, including self-care, for Indigenous patients in remote areas. This will most likely be a combination of better medical management, including more use of insulin in a form that can be conveniently taken by patients, and action to improve opportunities for self-care, including smoking cessation, nutrition and physical activity.

1: Characteristics of people on diabetes registers in Northern Territory, Cape York and Torres Strait remote communities, 2002–2003

Northern Territory

Cape York

Torres Strait


Number of patients

194

252

921

Number (%) women

115 (59%)

139 (55%)

553 (60%)

Number (%) Indigenous

186 (96%)

250 (99%)

909 (98%)

Mean (95% CI) duration of known diabetes (years)

6.8 (6.3–7.3)

8 (7.4–8.6)

7.9 (7.5–8.3)

Mean (95% CI) age (years)

51.8 (50.9–52.8)

50.9 (49.2–52.6)

52.4 (50.3–54.5)

Mean (95% CI) weight, men (kg)

88.9 (85.9–91.9)

85.7 (83.3–88.1)

97.4 (96.1–98.7)

Mean (95% CI) weight, women (kg)

79.7 (77.0–82.4)

76.8 (74.4–79.2)

87.6 (86.4–88.8)

Mean (95% CI) body mass index (kg/m2)

31.1 (30.2–32.1)

29.9 (29.1–30.7)

32.9 (32.5–33.3)

Number (%) smokers

64 (33%)

74 (29%)

227 (25%)

2: Diabetes-related checks and preventive care practices in Northern Territory, Cape York and Torres Strait remote communities, 2002–2003

Northern Territory (= 194)

Cape York (n = 252)

Torres Strait (n = 921)


Weight recorded (%) in previous 12 months

107 (55%)

216 (86%)

602 (65%)

Height recorded (%) at any time

118 (61%)

177 (72%)

701 (76%)

Blood pressure recorded (%) in previous 12 months

126 (65%)

231 (92%)

711 (77%)

High blood pressure recorded (%)

110 (87%)

199 (79%)

631 (69%)

Antihypertensive medication prescribed

101 (92%)

188 (95%)

577 (91%)

Hypertension with good blood pressure control

63 (62%)

134 (71%)

402 (70%)

Lipid levels checked (%) in previous 12 months

122 (63%)

212 (84%)

642 (70%)

With dyslipidaemia (high cholesterol or triglyceride levels)

97 (50%)

172 (68%)

525 (57%)

Prescribed lipid-lowering medication

40 (41%)

117 (68%)

301 (57%)

Good lipid control (cholesterol level < 5.5 mmol/L)

24 (60%)

79 (67%)

138 (46%)

Urinary albumin/creatinine ratio checked (%) in previous 12 months

119 (61%)

209 (83%)

588 (64%)

Macroalbuminuria (albumin/creatinine ratio > 34)

42 (35%)

72 (34%)

134 (24%)*

Angiotensin-converting enzyme inhibitor prescribed

41 (98%)

64 (89%)

117 (85%)

Serum creatinine level checked in previous 12 months

149 (77%)

227 (90%)

706 (77%)

HbA1c level checked in previous 6 months

108 (56%)

201 (80%)

519 (56%)

Self-monitoring glucose level

1 (0.5%)

16 (6.3%)

130 (14%)

Prescribed oral hypoglycaemic treatment only

149 (77%)

191 (76%)

543 (59%)

Taking insulin only

4 (2.1%)

11 (4.4%)

54 (6%)

Taking combined oral and insulin treatment

13 (7%)

19 (7.5%)

89 (10%)

Eye examination in previous 12 months

27 (14%)

185 (73%)

584 (63%)

Ophthalmologist review in previous 12 months

43 (22%)

122 (48%)

373 (41%)

Foot check in previous 12 months

67 (35%)

129 (51%)

544 (59%)

Influenza vaccination in previous 12 months

53 (27%)

180 (71%)

624 (68%)

Pneumococcal vaccination in previous 5 years

93 (48%)

208 (83%)

677 (74%)


*P < 0.05 compared with other two groups.

3: Intermediate clinical outcomes for patients with diabetes in Northern Territory, Cape York and Torres Strait remote communities, 2002–2003

Northern Territory (n = 194)

Cape York (n = 252)

Torres Strait (n = 921)


Mean (95% CI) HbA1c level

8.7% (8.4%–9.2%)

8.7% (8.5%–9.1%)

9.0% (8.9%–9.1%)

HbA1c level < 7%

28 (26%)

53 (26%)

116 (22%)

HbA1c level < 9.5%

75 (69%)

129 (64%)

232 (45%)

Mean (95% CI) systolic BP (mmHg)

128.4 (125.8–131.0)

129.4 (127.2–131.6)

130.1 (128.8–131.4)

Mean (95% CI) diastolic BP (mmHg)

79.1 (77.4–80.8)

79.6 (78.2–81.0)

77.3 (76.6–78.0)

Median (interquartile range) urinary albumin/creatinine ratio

14 (2.9–86)

13 (2.2–62)

6.8 (1.7–32)


BP = blood pressure.

4: Characteristics of three groups of patients with type 2 diabetes

Aboriginals (Northern Territory and Cape York) (n = 446)

Torres Strait Islanders (n = 921)

Non-Indigenous Australians* (n = 2077)


Mean age (years)

51.5

52.4

62.5

Proportion female

57%

60%

47.6%

Mean body mass index (kg/m2)

30.9

32.9

31.5

Mean HbA1c level

8.7%

9%

7.8%

“Good glycaemic control”

26%

22%

38%

Blood pressure < 140/90 mmHg

66%

70%

40%

Macroalbuminuria

34.4%

24%

6.9%

Self-monitoring for glucose

4%

14%

58%

Insulin treatment, alone or combined

10.5%

16%

34.4%


* Data for non-Indigenous patients are from the National Association of Diabetes Centres (NADC) Audit and Benchmarking study,8 except for glucose self-monitoring rate, which are from Queensland Health Diabetes Survey 2000.9 † “Good glycaemic control” is defined as HbA1c level < 7% in our study, and as HbA1c level ≤ 1% above upper limit of normal in the NADC study. ‡ Blood pressure results for the NADC sample are for patients older than 60 years.

  • Robyn A McDermott1
  • Fiona Tulip2
  • Barbara Schmidt3

  • Department of Public Health and Tropical Medicine, James Cook University, Cairns, QLD.

Correspondence: 

Acknowledgements: 

This study was supported by NHMRC Grant number 219109 and the Australian Department of Health and Ageing. Thanks to Phillip Mills for supporting the project, Fiona Cannon for data collection and entry, and Ashim Sinha, Christine Connors and four anonymous reviewers for comments on the manuscript.

Competing interests:

None identified.

  • 1. Cameron AJ, Welborn TA, Zimmet PZ, et al. Overweight and obesity in Australia: the 1999–2000 Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Med J Aust 2003; 178: 427-432. <MJA full text>
  • 2. McCarty D, Zimmet P, Dalton A, et al. The rise and rise of diabetes in Australia, 1996: a review of statistics, trends and costs. Canberra: International Diabetes Institute and Diabetes Australia, 1996.
  • 3. Colagiuri S, Colagiuri R, Conway B, et al. DiabCost Australia: assessing the burden of type 2 diabetes in Australia. Adelaide: Australian Diabetes Society and Australian Diabetes Educators Association, 2002.
  • 4. Australian Institute of Health and Welfare. The health and welfare of Australia’s Aboriginal and Torres Strait Islander peoples. Canberra: Australian Bureau of Statistics, 1999.
  • 5. Hodgson TA, Cohen AJ. Medical care expenditures for diabetes, its chronic complications and its comorbidities. Prev Med 1999; 29: 173-186.
  • 6. Wagner EH, Austin BT, Von Korff M. Improving outcomes in chronic illness. Manag Care Q 1996; 4: 12-25.
  • 7. McDermott R, Schmidt BA, Sinha A, Mills P. Improving diabetes care in the primary healthcare setting: a randomised cluster trial in remote Indigenous communities. Med J Aust 2001; 174: 497-502.
  • 8. National Association of Diabetes Centres. ANDIAB 2002. Australian national diabetes information audit and benchmarking: final report. Canberra: NADC, 2003. Available at: www.health.gov.au/pq/diabetes/pdf/andiabreptjan03.pdf (accessed Mar 2004).
  • 9. Queensland Health. Survey of diabetes care in Queensland, 2000. Brisbane: Queensland Health, 2003.
  • 10. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications: provisional report of a WHO consultation. Diabetes Med 1998; 15: 539-553.
  • 11. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2000; 23 (Suppl 1).
  • 12. Central Australian Rural Practitioners Association. CARPA standard treatment manual. Alice Springs: IAD Press, 1997.
  • 13. SPSS for Windows [computer program]. Version 10.0. Chicago: SPSS Inc, 2001.
  • 14. Harris MF, Pridden D, Ruscoe W, et al. Quality of care provided by general practitioners using and not using Division-based diabetes registers. Med J Aust 2002; 177: 250-252. <MJA full text>
  • 15. Overland J, Yue D, Mira M. The pattern of diabetes care in New South Wales: a five year analysis using Medicare occasions of service. Aust N Z J Public Health 2000; 24: 391-395.
  • 16. Phillips C, Molitch M. The relationship between glucose control and the development and progression of diabetic nephropathy. Curr Diabetes Rep 2002; 2: 523-529.
  • 17. Mogensen CE, Keane WF, Bennett PH, et al. Prevention of diabetic renal disease with special reference to microalbuminuria. Lancet 1995; 346: 1080-1084.
  • 18. Hoy WE, Baker PR, Kelly AM, Wang Z. Reducing premature death and renal failure in Australian Aboriginals. A community-based cardiovascular and renal protective program. Med J Aust 2000; 172: 473-478. <MJA full text>
  • 19. UK Prospective Diabetes Study Group. Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes mellitus. Lancet 1998; 352: 837-853.
  • 20. Sidorov J, Shull R, Tomcavage J, et al. Does diabetes disease management save money and improve outcomes? A report of simultaneous short-term savings and quality improvement associated with a health maintenance organization-sponsored disease management program among patients fulfilling health employer data and information set criteria. Diabetes Care 2002; 25: 684-689.
  • 21. Wagner EH, Sandhu N, Newton KM, et al. Effect of improved glycemic control on health care costs and utilization. JAMA 2001; 285: 182-189.
  • 22. Vaaler S. Optimal glycemic control in type 2 diabetic patients. Does including insulin treatment mean a better outcome? Diabetes Care 2000; 23(Suppl 2): B30-B34.
  • 23. Cefalu W. Evaluation of alternative strategies for optimizing glycemia: progress to date. Am J Med 2002; 113(Suppl 6A): 23S-35S.
  • 24. Chan J, Abrahamson M. Pharmacological management of type 2 diabetes mellitus: rationale for rational use of insulin. Mayo Clin Proc 2003; 78: 411-413.
  • 25. Brown J, Nichols G. Slow response to loss of glycemic control in type 2 diabetes mellitus. Am J Manag Care 2003; 9: 213-217.
  • 26. World Health Organization. Adherence to long-term therapies: evidence for action. Geneva: World Health Organization, 2003.
  • 27. Humphery K, Weeramanthri T, Fitz J. Forgetting compliance: Aboriginal health and medical culture. Darwin: Northern Territory University Press, 2001.
  • 28. Scrimgeour D, Rowse T, Lucas A. Too much sweet. The social relations of diabetes in central Australia. Darwin: Menzies School of Health Research, 1997.
  • 29. Phillips LS, Branch WT, Cook CB, et al. Clinical inertia. Ann Intern Med 2001; 135: 825-834.
  • 30. McDermott R. Evidence-based medicine and the health of Indigenous Australians. Med J Aust 2001; 175: 35-37.
  • 31. McCulloch B, McDermott R, Miller G, et al. Self-reported diabetes and health behaviors in remote Indigenous communities in northern Queensland, Australia. Diabetes Care 2003; 26: 397-403.

Author

remove_circle_outline Delete Author
add_circle_outline Add Author

Comment
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.