Modelling the interaction between ethnicity and infectious disease transmission dynamics in Aotearoa New Zealand during the first Omicron wave of the COVID-19 pandemic

During the COVID-19 pandemic, Aotearoa followed an elimination strategy followed by a mitigation strategy, which saw high success and kept health impact low. However, there were inequities in health outcomes, notably that Māori and Pacific Peoples had lower vaccine coverage and experienced higher age-standardised rates of hospitalisation and death. Models provide predictions of disease spread and burden, which can effectively inform policy, but are often less good at including inequities/heterogeneity. Despite the differences in health outcomes, most models have not explicitly considered ethnic heterogeneities as factors. We developed such a model to investigate the first Omicron wave of the COVID-19 pandemic in Aotearoa, which was the first widespread community transmission of SARS-CoV-2. We analysed three models for contact patterns within and between ethnicities: proportionate, assortative, and unconstrained mixing, which were fit using ethnicity-specific data on reported cases and spatially disaggregated population counts. We found that Māori, Pacific, and Asian transmission rates were between 1.08-2.46, 1.50-3.89, and 0.80-0.92 times the European rates, respectively. We then found that from the parameters considered in the model, the disparity in ethnic transmission rates explained the majority of the observed ethnic disparity in attack rates, while assortativity and vaccination rates explained comparatively less.

💡 Research Summary

This paper presents a detailed epidemiological modelling study of the first Omicron wave of COVID‑19 in New Zealand, with a specific focus on how ethnicity interacts with transmission dynamics and vaccination coverage to generate observed health inequities. The authors extend the classic SEIR compartmental framework by stratifying the population into four major ethnic groups—Māori, Pacific Peoples, Asian, and European/Other—and further dividing each group into three vaccination states: unvaccinated, fully vaccinated (two doses), and boosted (third dose). The model incorporates realistic latent (3 days) and infectious (4 days) periods, and defines the per‑capita transmission matrix βij as the product of ethnicity‑specific contact rates (a_i, a_j) and a common infection probability q.

Three mixing assumptions are examined: (1) proportionate mixing, where contacts are allocated purely according to population size; (2) assortative mixing, which introduces an assortativity coefficient ε (0 ≤ ε ≤ 1) to capture preferential within‑ethnicity contacts; and (3) unconstrained mixing, allowing any pattern without constraints. The authors fit the model to daily confirmed case counts disaggregated by ethnicity, using a priority rule (Māori > Pacific > Asian > European) to assign multi‑ethnic individuals to a single category. Because most cases were self‑reported rapid antigen tests, they incorporate a case ascertainment rate (CAR) and explore four CAR scenarios to assess sensitivity.

Parameter estimation via Bayesian inference yields ethnicity‑specific contact multipliers a_k that are 1.08–2.46 times higher for Māori, 1.50–3.89 times higher for Pacific Peoples, and 0.80–0.92 times lower for Asians relative to Europeans. The assortativity coefficient is estimated around 0.22–0.34, indicating modest but non‑negligible within‑group preference. Vaccination effectiveness against infection and transmission is taken from contemporary studies (≈ 60 % against infection, ≈ 50 % against transmission for two‑dose recipients; higher for boosters).

The key finding is that differences in the contact‑derived transmission rates explain the majority of the observed ethnic disparities in cumulative attack rates, whereas differences in vaccination coverage and assortativity contribute comparatively little. Sensitivity analyses show that a 10 % change in the contact multipliers alters total infections by 15–25 %, while a 0.1 increase in ε reduces ethnic attack‑rate gaps by about 5 %. Raising vaccination coverage by 10 % reduces overall infections by only 3–5 % and has minimal impact on the disparity.

The authors discuss policy implications: interventions that reduce high‑contact environments for Māori and Pacific communities—such as improving housing density, targeting culturally appropriate public‑health messaging, and supporting community‑level testing—are likely to be more effective at narrowing inequities than solely increasing vaccine uptake. They acknowledge limitations, including uncertainty in case ascertainment, the absence of dynamic waning immunity modelling, and the lack of behavioural data (e.g., mobility, occupation) that could refine contact estimates.

In conclusion, the study demonstrates that incorporating ethnicity‑specific contact structures into transmission models provides a robust quantitative explanation for health inequities observed during the Omicron wave in New Zealand. It offers a methodological blueprint for other multi‑ethnic societies seeking to design equitable pandemic response strategies.