Inequalities in COVID-19 severe morbidity and mortality by country of birth in Sweden

The COVID-19 pandemic disproportionately affected migrant populations globally, including Sweden. Existing studies primarily focused on specific pandemic phases, overlooking the evolution of inequalities across different waves and viral variants (Delta, Omicron). This study aimed to address this gap by examining how inequalities in COVID-19 ICU admission and mortality between migrants and Swedish-born individuals changed throughout the pandemic. The researchers hypothesized that inequalities would be significant early in the pandemic, potentially influenced by social conditions and later mitigated by vaccination rollout. Understanding these determinants is crucial for developing effective preventive measures in future health crises. The study's importance lies in its comprehensive examination of the entire pandemic period, allowing for an analysis of the dynamic interplay between migrant status, socio-economic factors, vaccination access, and COVID-19 outcomes.

Numerous studies documented higher risks of COVID-19 infection, mortality, and severe morbidity among ethnic minorities and migrants in Sweden and other countries. These studies often focused on specific pandemic phases, primarily the initial wave. Prior research suggested that adverse social and living conditions, such as socioeconomic status and crowded housing, might contribute to higher risks among migrant populations. While later waves showed lower vaccination uptake among migrants, the impact of vaccination on migrant-Swedish born inequalities remained unexplored. This study builds upon this existing research by providing a more comprehensive temporal analysis across multiple pandemic waves and incorporating vaccination status as a crucial factor.

This population-based cohort study utilized linked data from multiple Swedish registers, including the Total Population Register, the Longitudinal Integrated Database for Health Insurance and Labor Market Studies (LISA), the Dwelling Units Register, the national notifiable disease registry (SmiNet), the Swedish Intensive Care Register, the Cause of Death Register, and the National Vaccination Register. The study population comprised all adults residing in Sweden at the end of 2019 (n=7,870,441). The primary outcome measures were COVID-19 related ICU admissions and deaths. Exposure was defined by country of birth, categorized into nine groups: Sweden, Nordics (excluding Sweden), EU28/EEA (excluding Nordics), non-EU28/EEA European countries, Middle East, Africa, Asia, North America, and South America. Covariates included age, sex, education level, income, occupation, household type, housing characteristics (type of accommodation, living area per person), neighborhood population density, region of residence, and vaccination status. Poisson regression analyses were used to estimate incidence rate ratios (IRRs) for ICU admissions and mortality, adjusting for various combinations of covariates. The analysis was stratified by four pandemic waves, defined by periods of increased and decreased COVID-19 cases, hospitalizations, and deaths. Model comparisons were performed using the Akaike Information Criterion (AIC).

The study revealed substantially higher risks of COVID-19 ICU admission and mortality among migrants compared to Swedish-born individuals, particularly during the first wave. Migrants from Africa, the Middle East, Asia, South America, and non-EU28/EEA European countries exhibited significantly elevated risks. For example, in the unadjusted model, the crude rate for ICU admissions was highest for individuals born in the Middle East (118.5 per 100,000), followed by Europe excluding EU28/EEA (111.1) and South America (111.0). After adjusting for age and sex, the highest standardized rate was in the Middle East (177.8), with Africa (166.7) and Asia (157.2) showing next highest. The age and sex adjusted relative risks (RRs) for ICU admission in the first wave were substantial: 8.2 (95% CI: 6.7, 10.2) for those born in Africa, 6.2 (95% CI: 5.2, 7.4) for those born in the Middle East, and 5.8 (95% CI: 4.6, 7.3) for those born in South America. These high risks were partially explained by socioeconomic factors and living conditions. Across the four waves, disparities in ICU admissions decreased, becoming insignificant after adjustment for socio-economic factors, living conditions, and vaccination status in the fourth wave. The pattern for mortality was more complex, with disparities being largest in the first and third waves. Vaccination uptake significantly reduced inequalities in the fourth wave, explaining much or all of the excess risk in many groups.

The study’s findings highlight the significant disparities in COVID-19 outcomes experienced by migrants in Sweden. The higher risks observed in the initial pandemic phase likely reflect a combination of factors: increased vulnerability due to existing health conditions and risk factors, poor living conditions, and possibly inadequate public health messaging and access to healthcare. The reduction of inequalities in subsequent waves, especially after vaccination, demonstrates the critical role of equitable access to vaccination in mitigating health disparities. The residual excess risk in some groups despite adjustments suggests the need for further investigation into other potential factors, such as health behaviors, access to healthcare, and social determinants of health.

This study confirms substantially higher COVID-19 ICU admission and mortality risks among several migrant groups in Sweden, especially during early pandemic waves. Socioeconomic disadvantages and living conditions partially explain these inequalities, but the availability of vaccines proved decisive in reducing disparities. Future research should explore the roles of underlying health conditions, social interactions, access to information, and potential healthcare discrimination in shaping COVID-19 outcomes among migrants. The results underscore the need for comprehensive strategies focusing on improving social and living conditions for migrants and ensuring equitable access to vaccines during future pandemics.

The study acknowledges several limitations. The inability to directly measure underlying comorbidities and health behaviors might underestimate their full impact. Furthermore, the lack of data on social interaction patterns, language proficiency, and adherence to guidelines limits the complete explanation of the observed disparities. The use of Poisson regression, while suitable for descriptive purposes, does not fully capture the complexity of infectious disease transmission dynamics. The potential underestimation of COVID-19 mortality due to incomplete ascertainment is also noted, though unlikely to affect the identified disparities.