Modelling the transmission and impact of Omicron variants of Covid-19 in different ethnicity groups in Aotearoa New Zealand

TL;DR

This study extends a Covid-19 transmission model to include ethnicity-specific factors in New Zealand, revealing that disparities in severe disease risk significantly contribute to unequal health outcomes among different ethnic groups.

Contribution

It introduces an ethnicity-stratified Covid-19 model incorporating vaccination, clinical severity, and contact patterns, improving understanding of disparities in health outcomes.

Findings

Differences in vaccination rates partially explain disparities.

Per-infection risk of severe disease is a key factor.

Model helps predict inequitable impacts of future pandemics.

Abstract

Previous pandemics, including influenza pandemics and Covid-19, have disproportionately impacted M\=aori and Pacific populations in Aotearoa New Zealand. The reasons for this are multi-faceted, including differences in socioeconomic deprivation, housing conditions and household size, vaccination rates, access to healthcare, and prevalence of pre-existing health conditions. Many mathematical models that were used to inform the response to the Covid-19 pandemic did not explicitly include ethnicity or other socioeconomic variables. This limited their ability to predict, understand and mitigate inequitable impacts of the pandemic. Here, we extend a model that was developed during the Covid-19 pandemic to support the public health response by stratifying the population into four ethnicity groups: M\=aori, Pacific, Asian and European/other. We include three ethnicity-specific components in the model: vaccination rates, clinical severity parameters, and contact patterns. We compare model results to ethnicity-specific data on Covid-19 cases, hospital admissions and deaths between 1 January 2022 and 30 June 2023, under different model scenarios in which these ethnicity-specific components are present or absent. We find that differences in vaccination rates explain only part of the observed disparities in outcomes. While no model scenario is able to fully capture the heterogeneous temporal dynamics, our results suggest that differences between ethnicities in the per-infection risk of clinical severe disease is an important factor. Our work is an important step towards models that are better able to predict inequitable impacts of future pandemic and emerging disease threats, and investigate the ability of interventions to mitigate these.