Background: Kidney disease affects approximately 13% of adults worldwide and markedly increases an individual’s risk of cardiovascular disease. While the incidence of dialysis treatment for kidney failure is decreasing in higher income countries, the incidence rate is increasing among Māori in Aotearoa. Existing evidence suggests disparity in health outcomes may be related principally to lower socioeconomic opportunity, although epidemiological analysis have not adjusted fully for community deprivation when exploring determinants of health disparities in kidney disease. In this study, survival during dialysis treatment was assessed among Māori and non-Māori patients while accounting for baseline differences in geographical location, socioeconomic deprivation, and clinical comorbidity.
Methods: A kaupapa Māori propensity score matched cohort study was conducted involving all adults who commenced dialysis in Aotearoa between 2002 and 2011. Data were drawn from ANZDATA, a quality clinical registry, linked to the NZ National Health Index to obtain deprivation statistics. Propensity score methods were used to assemble a cohort of Māori patients who were matched 1:1 on clinical and demographic characteristics with non-Māori patients at dialysis start. Risk of all-cause mortality were estimated using Cox proportional hazards modelling in matched cohorts stratified by geographical location.
Results: The balance of characteristics between Māori and non-Māori patients was substantially improved after propensity matching with small differences in age, gender, deprivation score, type 2 diabetes, smoking, and body mass index. Among the matched patients, Māori had a higher risk of mortality (hazard ratio 1.46, 95% CI 1.31-1.64). Māori were more likely than non-Māori to experience a non-tunnelled catheter for dialysis access (47% versus 30%, standardised difference 0.35) despite socio-demographic and clinical matching.
Interpretation: Māori experience higher mortality rates during dialysis treatment than non-Māori, which is not fully accounted for by socioeconomic deprivation, geographical location, or comorbidity. This finding has implications for policy and health providers as it suggests health interventions that insufficiently consider Indigenous status, as a marker of exposure to other risks for clinical outcomes, independent of social and clinical indicators may not effectively address sustained disparities in Aotearoa.