Background: This study aimed to investigate groups of patients with a relatively homogenous health status to evaluate the degree to which use of the Australian hospital system is affected by socio-economic status, locational accessibility to services and patient payment classification.Method: Records of all deaths occurring in Western Australia from 1997 to 2000 inclusive were extracted from the WA mortality register and linked to records from the hospital morbidity data system (HMDS) via the WA Data Linkage System. Adjusted incidence rate ratios of hospitalisation in the last, second and third years prior to death were modelled separately for five underlying causes of death.Results: The independent effects of socioeconomic status on hospital utilisation differed markedly across cause of death. Locational accessibility was generally not an independent predictor of utilisation except in those dying from ischaemic heart disease and lung cancer. Private patient status did not globally affect utilisation across all causes of death, but was associated with significantly decreased utilisation three years prior to death for those who died of colorectal, lung or breast cancer, and increased utilisation in the last year of life in those who died of colorectal cancer or cerebrovascular disease.Conclusion: It appears that the Australian hospital system may not be equitable since equal need did not equate to equal utilisation. Further it would appear that horizontal equity, as measured by equal utilisation for equal need, varies by disease. This implies that a 'one-size-fits-all' approach to further improvements in equity may be over simplistic. Thus initiatives beyond Medicare should be devised and evaluated in relation to specific areas of service provision.