Aims: To examine whether different comorbidity, stage and cancer treatment contribute to an overall cancer survival difference between Indigenous and non-Indigenous Queenslanders.
Methods: Cohort study of 956 Indigenous and 869 non-Indigenous people (frequency-matched on cancer type, age, sex and location) diagnosed with cancer during 1998-2004 and treated in Queensland public hospitals. Survival after cancer diagnosis, stratified by stage at diagnosis, treatment, and comorbidities (presence and severity) were examined.
Results: Indigenous people had less localised cancer at diagnosis, more comorbidities (22% vs 14% had a Charlson index score > 2, P<0.001), and received less treatment (any cancer treatment, 75% vs. 86%, P=0.001) than non-indigenous people. However, for those who did receive treatment, time to commencement or duration were comparable. As diabetes was the most common comorbidity for Indigenous patients, we attempted to examine if diabetes control (HbA1C measurements used as a proxy) was comparable between the two groups. Although there was no difference in HbA1c levels between the groups (P=0.59), information was available for only 35% of patients. Crude cancer survival was 30% worse (HR 1.30 95%CI 1.15, 1.48) and non-cancer deaths was over twice more common among Indigenous patients (HR 2.39 95%CI 1.57, 3.63). When stage at diagnosis, socioeconomic status, comorbidities and treatment were taken into account, the risk of cancer death was not significantly different for Indigenous people (HR 1.11 95%CI 0.98, 1.27).
Conclusion: Findings suggest that treatment, comorbidities and stage at diagnosis explain most of the poorer cancer outcomes among Indigenous patients. There is a need for a greater understanding of the reasons for delays in diagnosis, and to explain why less cancer treatments are provided (reduced uptake or access to treatment, or this is confounded by un/measured comorbidity). The effect of comorbidity on survival differences need to be fully quantified.