Is ethnicity an independent predictor of health need? Linked cohort logistic regression analysis to predict amenable mortality

AIM: This study examines whether ethnicity is an independent marker of health or if ethnic disparities are fully explained by age, sex, rurality, socio-economic position and morbidity. METHOD: Using the Stats NZ Tatauranga Aotearoa Integrated Data Infrastructure, we identified the resident population of Aotearoa New Zealand each year from 2009 to 2018, establishing 10 cohorts that were followed up with at 12 months for amenable mortality, i.e., deaths from conditions responsive to healthcare in under-75-year-olds. Age, sex, ethnicity, rurality, small area deprivation, personal income and morbidity of cohort members were described. Logistic regression analyses and likelihood ratio tests were used to assess the independent association of these variables with amenable mortality. RESULTS: Ethnicity, socio-economic position and morbidity, along with age, sex and rurality, made significant independent contributions to predicting amenable mortality. Ethnicity predicted amenable mortality after adjusting for other variables. Compared with Europeans, the odds of amenable mortality were 1.46 (95% confidence interval [CI] 1.43-1.50) times greater in Māori and 1.18 (95% CI 1.14-1.23) times greater in Pacific and half as likely in Asian (0.54, 95% CI 0.52-0.57). CONCLUSION: Māori and Pacific ethnicity, and also socio-economic position and morbidity, are independent markers of health need relevant to the distribution of health resources and targeting of health services.