Diabetes among aboriginal people in central Australia: a high prevalence based on health service attendance

Diabetes and diabetic retinopathy in indigenous Australians

The health status of indigenous Australians is significantly poorer than that of other Australians. One of the major health problems for this population is type 2 diabetes which has been reported to have much higher prevalence rates than those in the rest of the Australian population. The significance of diabetes as a health issue has recently been reinforced by its inclusion as one of the six National Health Priorities. The recently developed National Diabetes Strategy is now in its implementation phase. Diabetic retinopathy is an important complication of diabetes. However, its prevalence within the indigenous population of Australia is not well documented. In this paper, we give a brief overview pertaining to this topic and include two case reports to illustrate some of the specific issues related to the management of diabetic retinopathy in this population.

Mobile retinal photography. A means of screening for diabetic retinopathy in aboriginal communities

BACKGROUND

Mobile fundus photography undertaken by Aboriginal health workers has been suggested as a practical and effective method of screening for diabetic retinopathy among Aboriginal and Torres Strait Islander communities.

METHODS

A pilot study was undertaken to demonstrate that interpretable colour slides of the retina could be produced by Aboriginal health workers trained over a two-week period.

RESULTS

Of 47 subjects, 36 (76.6%) had interpretable retinal photographs.

CONCLUSIONS

These findings warrant further evaluation as to the benefits and cost effectiveness of mobile retinal photography.

High mortality from renal disease and infection in Aboriginal central Australians with diabetes

Few studies have examined the consequences of the high prevalence of diabetes in Aboriginal communities. We aimed to determine the rates and causes of mortality in all Aboriginal central Australians with diagnosed diabetes, identified by a previous study (n =374). Cohort members were followed from 1 January 1984, or the date of diagnosis (to 31 December 1986), to 31 December 1991 or death. Death certificates, medical notes and autopsy reports were examined for cause of death. There were 130 deaths in 2280.7 person-years of follow-up. Standardised mortality ratios for Aboriginal people with diabetes, compared to the Northern Territory Aboriginal population, were 209 (95 per cent confidence interval (CI) 158 to 273) for men and 169 (CI 129 to 218) for women. The difference in ratios for men and women was not statistically significant when adjusted for age (P = 0.2). The eight-year survival rates for men and women diagnosed between 1984 and 1986 were 55.8 per cent (CI 32.6 to 73.7) for men and 80.3 per cent (CI 64.8 to 89.5) for women. Renal disease was the direct cause of death in 22.3 per cent. Infection accounted for 20.8 per cent of deaths and ischaemic heart disease for 13.8 per cent. Forty-four per cent of death certificates made no mention of diabetes. Diabetes confers an additional risk of death on a population whose mortality is already markedly worse than that of other Australians. Unlike Western diabetic populations, infections and renal disease were more common causes of death than macrovascular disease. Diabetes amplifies the effect of the community prevalence of infection and renal disease.

A community-based cervical screening program in a remote Aboriginal community in the Northern Territory

We established a culturally appropriate, community-based women's health service in Yuendumu, Northern Territory, to improve women's health and to remedy the low rate of cervical screening. During the 16 months of the program 419 cervical smears were taken, increasing coverage of the women eligible from 51 to 78 per cent. Acceptance of the program was excellent, with only 2 per cent of the women approached refusing to have a smear. Over 70 per cent of the Pap smears were done by the nursing staff in the clinic; quality control was good, with 9 per cent of smears reported as having no endocervical cells. Sixty-four per cent of screened women had normal smears and 0.9 per cent showed evidence of cervical intraepithelial neoplasia (CIN). Three women were referred for treatment of CIN, two for CIN I and one for CIN III. The program illustrates how a combination of community involvement, staff stability, teamwork, and cross cultural understanding can achieve a comprehensive and successful cervical screening service in a remote Aboriginal community.

Food-purchasing behaviour in an Aboriginal community. 1. Results of a survey

Attempts to improve the nutritional status of Aboriginal people through nutritional education programs should be informed by an understanding of contemporary patterns of food procurement, preparation and distribution. This paper describes the results of a survey of food-purchasing behaviour in a central-Australian Aboriginal community. Every transaction occurring in each food outlet in the community over a two-week period was recorded and the data analysed. The results show that women play a much greater role than men in food purchasing, that there is a significant recourse to takeaway foods, that there is a cycle of expenditure determined by distribution of pension and Community Development Employment Project cheques, and that children have sufficient disposable income to be able to provision themselves from the food outlets, so that much of their food consumption is not determined by adult members of their family.

Utilisation of health services by aboriginal Australians with diabetes

Diabetes is a major public health problem for Aboriginal Australians. We wished to determine the extent and pattern of health service utilisation by Aboriginal people with diabetes in central Australia. Medical records of all Aboriginal people known to have diabetes (n = 374), identified by a previous study, were examined for attendance to health services in central Australia. All had non-insulin-dependent diabetes. Between January 1984 and December 1986, Aboriginal adults with diabetes were admitted to hospital on 694 occasions, accounting for 10.8% of adult Aboriginal admissions. The crude admission rates were 0.78 and 0.84 per diabetes-year for men and women, respectively. The age-adjusted relative risks for admission, compared with Aboriginal non-diabetic patients, were 2.93 (95% C.I., 2.62-3.26) for men and 2.46 (95% C.I., 2.28-2.66) for women. If admission for conditions associated with diabetes are excluded, the admission rates were similar for the two groups. Infection was the most common reason for attendance to a health service, representing 41.7% and 39.8% of male and female admissions, and 21.8% and 26.3% of male and female outpatient attendances. Aboriginal patients with diagnosed diabetes suffer high morbidity and contribute disproportionately to health system costs.

Diabetes in aborigines and other Australian populations

The aetiology of diabetes mellitus is reviewed in the light of Australian epidemiological studies. The prevalence of diabetes has been studied in a wide variety of Australian settings. Comparisons of prevalences in Aborigines and other populations suggest that occurrence of this disease has both genetic and environmental determinants. Diet and obesity are associated with diabetes, but to increase physical exercise may be the most feasible intervention. Public health approaches are discussed.

Birth defects in the infants of aboriginal and non-aboriginal mothers with diabetes in Western Australia

OBJECTIVE

Because of the high prevalence of non-insulin-dependent diabetes in Australian Aborigines, and a suggestion that the prevalence of birth defects was high in the infants of Aboriginal mothers with gestational diabetes, this study was undertaken to determine the prevalence of birth defects in infants of Aboriginal and non-Aboriginal mothers with insulin-dependent, non-insulin-dependent, and gestational diabetes mellitus.

DESIGN

A retrospective cohort study of all births to diabetic and non-diabetic mothers in Western Australia, 1980-1984.

MAIN OUTCOME MEASURE

Birth defects diagnosed at any time up to the age of six years.

RESULTS

Compared with infants of non-diabetic, non-Aboriginal mothers, the prevalence ratio for birth defects in infants of non-Aboriginal insulin-dependent mothers was 2.08 (95% confidence interval, 1.2-3.7), and for infants of mothers with non-insulin-dependent diabetes the ratio was 3.64 (95% CI, 1.5-8.6). The corresponding ratios for infants of Aboriginal mothers were 4.85 (95% CI, 0.8-28.2) and 3.64 (95% CI, 1.3-10.4). For birth defects in infants of gestational diabetic mothers, the prevalence ratio was 1.07 (95% CI, 0.6-1.9) for the non-Aboriginal group and 3.65 (95% CI, 2.3-6.0) for the Aboriginal group. Diabetes could have accounted for 0.14% of birth defects in infants of non-Aboriginal mothers and for 4.62% in infants of Aboriginal mothers.

CONCLUSIONS

The excess of birth defects in infants of Aboriginal women with gestational diabetes may be due to non-insulin-dependent diabetes that predates the pregnancy but is only diagnosed during pregnancy. For Aboriginal infants, maternal diabetes may be the single most common known cause of birth defects that is amenable to change.

Frequent hospital admissions for bacterial infections among aboriginal people with diabetes in central Australia

OBJECTIVE

To determine hospital-based morbidity and mortality from bacterial infections among Aboriginal people with diabetes in central Australia.

DESIGN

Examination of medical records on diabetic inpatients between January 1984 and December 1986.

SETTING

Alice Springs Hospital in central Australia, covering an Aboriginal population of about 12,500 people.

PATIENTS

Subjects admitted with a bacterial infection (n = 165) were selected from the database of all Aboriginal people known to have diabetes in the region (n = 374; all had non-insulin dependent diabetes).

MAIN OUTCOME MEASURES

Number of admissions, type and site of infections, causative organisms, duration of hospital stay and cause of death.

RESULTS

The 281 admissions for infections accounted for 4.6% of adult Aboriginal admissions to hospital. Thirteen patients died. Eleven patients required amputation of a digit or limb. In 1986, 13/200 patients admitted with diabetes had bacteraemia compared with 14/1885 patients who did not have diabetes. One patient had Fournier's gangrene, one had malignant otitis externa, one community acquired Acinetobacter pneumonia, one pneumonia with Succinivibrio spp. found in blood cultures, and one meningitis and transverse myelopathy with Streptococcus milleri found in blood cultures.

CONCLUSIONS

The frequency, type and severity of infections in the relatively young patients result in high personal costs to the community and financial costs to the health care system. Standard preventive measures must be implemented to control severe bacterial infections among subjects with diabetes in this socially and economically marginalised population group.