Estimating alcohol-attributable mortality among Indigenous Australians: towards Indigenous-specific alcohol aetiologic fractions.

INTRODUCTION AND AIMS

Past estimates of Indigenous alcohol-attributable health in Australia have been based on drinking prevalence estimates from the general population, rather than prevalence figures from the Indigenous population. The purpose of this paper is to demonstrate the efficacy of using Indigenous-specific drinking prevalence to estimate alcohol-attributable deaths among Indigenous Australians.

DESIGN AND METHODS

Estimates of Indigenous alcohol-attributable deaths between 2000 and 2004 were obtained using both (i) national general-population drinking prevalence estimates and (ii) national Indigenous-specific drinking prevalence. Estimates were calculated using the 'aetiologic fraction' method.

RESULTS

By using national general-population drinking prevalence figures, past reports on Indigenous health have underestimated alcohol-attributable deaths for the national Indigenous population. Female deaths due to alcohol-attributable haemorrhagic stroke were estimated to be approximately four times higher and alcohol-attributable suicides among men were estimated to be 30% higher than was previously held, when Indigenous-specific drinking prevalence figures were used.

DISCUSSION AND CONCLUSIONS

By substituting Indigenous-specific alcohol consumption prevalence estimates for general-population drinking prevalence, the accuracy of estimates of alcohol-related harm among Indigenous Australians can be significantly improved.