Comorbidities contribute to the risk of cancer death among Aboriginal and non-Aboriginal South Australians: Analysis of a matched cohort study. (February 2018)

• Record Type: Journal Article
• Authors: Banham, David; Roder, David; Brown, Alex
• Is Part Of: Cancer epidemiology. Volume 52(2018:Feb.)
• Page Start: 75
• Page End: 82
• Publication Date: 2018-02
• Subjects: Aboriginal -- Indigenous -- Cancer -- Disparity -- Comorbidity -- Cancer stage -- Geographic remoteness -- Survival; Cancer -- Epidemiology -- Periodicals; Cancer -- Prevention -- Periodicals; Cancer -- Diagnosis -- Periodicals; Carcinogenesis -- Periodicals; 616.994005
• DOI: 10.1016/j.canep.2017.12.005
• Languages: English
• ISSNs: 1877-7821

Abstract

Highlights: Hospital administrative records improve insight into the prevalence of comorbid disease. Three times more Aboriginal cases were diagnosed with four or more comorbidities on the Elixhauser Comorbidity Index. Heightened levels of comorbid disease at the time of diagnosis increased the risk of cancer death. The analysis broadens existing aetiological models of stage at diagnosis and remoteness to include comorbidity's effect. Findings draw attention to comorbidity prevalence and added risk among a relatively young, but very disadvantaged population. Abstract: Background: Aboriginal Australians have poorer cancer survival than other Australians. Diagnoses at later stages and correlates of remote area living influence, but do not fully explain, these disparities. Little is known of the prevalence and influence of comorbid conditions experienced by Aboriginal people, including their effect on cancer survival. This study quantifies hospital recorded comorbidities using the Elixhauser Comorbidity Index (ECI), examines their influence on risk of cancer death, then considers effect variation by Aboriginality. Methods: Cancers diagnosed among Aboriginal South Australians in 1990–2010 (N = 777) were matched with randomly selected non-Aboriginal cases by birth year, diagnostic year, sex, and primary site, then linked to administrative hospital records to the time of diagnosis. Competing risk regression summarised associations of Aboriginal status, stage, geographic attributes and comorbidities with risk of cancer death. Results: A threshold of four or more ECI conditions was associated with increased risk of cancer death (sub-hazard ratio SHR 1.66, 95%CI 1.11-2.46). Alternatively, the presence of any one of a subset of ECI conditions was associated with similarly increased risk (SHR = 1.62, 95%CI 1.23-2.14). The observed effects did not differ between Aboriginal and matched non-Aboriginal cases. However, Aboriginal cases experienced three times higher exposure than non-Aboriginal to four or more ECI conditions (14.2% versus 4.5%) and greater exposure to the subset of ECI conditions (20.7% versus 8.0%). Conclusion: Comorbidities at diagnosis increased the risk of cancer death in addition to risks associated with Aboriginality, remoteness of residence and disease stage at diagnosis. The Aboriginal cohort experienced comparatively greater exposure to comorbidities which adds to disparities in cancer outcomes.