Objective: This study compares prevalence of obesity, hypertension and diabetes in two groups of Aboriginal adults: those living in homelands versus centralised communities in central Australia. It also compares weight gain. incidence of diabetes, mortality and hospitalisation rates between the groups over a seven-year period.
Methods: Baseline survey of 826 Aboriginal adults in rural central Australian communities in 1987-88 with a fol1ow up survey of 416 (56% response rate, excluding deaths). Each time, they had a 75 g oral glucose tolerance test (OGTI, and blood pressure and anthropometry measurement. Deaths and hospitalisations for all of the original cohort were recorded for the seven-year period.

Results: Homelands residents had a lower baseline prevalence of diabetes (risk ratio [RRI=On, 0.59-1.00), hypertension (RR=0.66, 0.54-0.80) and overweigh obesity (RR=0.70, 0.59-0.83). The incidence of diabetes was lower among homelands residents (RR=0.70, 0.46-t.06). They were less likely to die than those living in centralised communities (RR=0.56. 0.370.85) and less likely to be hospitalised for any cause (RR=0.79, 0.71-0.87), particularly infections (RR=0.70, 0.61-0.80). injury involving alcohol (RR=0.61, 0.470.79) and other injury (RR=0.75, 0.60-0.93). Mean age at death was 58 and 48 years for residents of homelands and centralised communities respectively.

Conclusion: Aboriginal people who live in homelands communities appear to have more favourable health outcomes with respect to mortality, hospitalisation, hypertension. diabetes and injury, than those living in more centralised settlements in Central Australia. These effects are most marked among younger adults.