Migrant status, ethnicity and COVID-19: more accurate European data are greatly needed.

In the past months, systematically higher levels of severe COVID-19 illness and death among individuals belonging to migrant or ethnic minority groups have been reported in several countries, including the USA, the UK, Sweden, Brazil, Spain and South Africa. In particular, racialized populations tend to have less access to testing, higher rates of severe disease, higher mortality rates and worse sequalae when they survive the infection [[1], [2], [3], [4], [5]]. The interacting social, behavioural and biological pathways underlying these disparities are doubtlessly complex, but the hierarchy of social advantage in racialized societies drives health status, exposure, housing, employment and access to healthcare, and undoubtedly plays the dominant role [6].

Yet, with the notable exception of the UK, European countries do not report COVID-19 statistics according to migrant status or ethnicity (i.e. individuals' country of origin, nationality or membership of an ethnic minority group), faced with the complexity of collecting and using such information for public health purposes [7]. To exemplify the issues at stake and suggest a way forward, we wish to take the case of France, which has a substantial immigrant population and is to date among the European countries hardest hit by the COVID-19 epidemic.

In France, immigrants make up 10% of the population and descendants of immigrants 12% (https://www.insee.fr/fr/statistiques/4238373?sommaire=4238781#:~:text=En%202018%2C%207%2C5%20millions,le%20m%C3%AAme%20pays%20d′origine.), with recent immigrants most frequently originating from an African country (47%). Data show that, compared with the native population, persons who are foreign-born had on average double the rates of all-cause mortality between March and April 2020, some of which were due to COVID-19; however, that exact proportion is not known [8]. Specifically, while mortality increased by 22% among the native-born, it rose by 54% among individuals born in North Africa, 91% among those born in Asia and 114% among those born in sub-Saharan Africa. This trend is also reflected in geographic inequalities in mortality, with a 118% increase compared with the preceding year in Seine-Saint-Denis, a district North of Paris which is the poorest in France and where 30% of the population is immigrant, compared with a 96% increase in Paris itself [9]. However, the full extent of inequalities between migrants and members of ethnic minority groups and the rest of the population can only be assessed if population-wide data on morbidity and healthcare are collected by ethnic and migrant status. The stark reality of the available statistics, in conjunction with a climate of condemnation of structural racism following the death of George Floyd and associated worldwide mobilizations, has given rise to a renewed public discussion about the risks and benefits of “ethnic statistics”. Similar debates in this area are taking place in other European countries (http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=SDD/DOC(2018)9&docLanguage=En).

The European regulations regarding data protection such as the General Data Protection Regulation (GDPR) restrict the collection of research or administrative data making it possible to identify individuals' “racial or ethnic origins” at the same level as information on political opinions, labour union membership, sexual orientation and genetic background, unless specific law or adequate safeguards are in place (https://ec.europa.eu/info/law/law-topic/data-protection/reform/rights-citizens/how-my-personal-data-protected/how-data-my-religious-beliefs-sexual-orientation-health-political-views-protected_en). Likewise, for instance, in the UK, data on an individual's migrant status and ethnicity are regularly collected in national surveys, making it possible to ascertain disparities with regard to health as well as in other domains, particularly as regards the BAME (Black, Asian, and Minority Ethnic) groups (https://dataingovernment.blog.gov.uk/2019/06/27/how-were-helping-people-understand-ethnicity-data/). In France, on the other hand, a deliberate choice has been made to implement these regulations in a strict sense, regulatory authorities being directly called upon each time a decision upon the inclusion of any information on migrant/ethnic background is to be made. Indeed, for each survey employing sensitive data (ethnicity and country of birth, religion, sexual orientation), special authorizations are to be exceptionally delivered. These special rules are the legacy of a strict interpretation of Article 1 of the French Constitution, which states the prohibition of distinction of citizens on the bases of race, origin or religion, and is meant to protect individuals from possible data misuse and discrimination.

However, there are several flaws to the above argument. First, it is obvious that racism and discrimination exist even in the absence of reliable statistics to measure them. In France, as in other European countries, researchers and the Ombudsman's office have repeatedly documented disparities in domains such as access to education, housing and employment (https://www.defenseurdesdroits.fr/fr/etudes-et-recherches/2020/06/inegalites-dacces-aux-droits-et-discriminations-en-france). However, discrimination with regard to health and healthcare have to date received little attention. Second, while it seems reasonable to limit collection of data on an individual's geographical, racial and religious background for administrative purposes, absence of such information in health surveys makes it impossible to quantify disparities, barring the possibility of reducing sources of discrimination and lost opportunities for members of minority groups [10]. For instance, demonstration of an increased risk of HIV among immigrants originating from sub-Saharan Africa in several European countries, including France, was used to inform the design of specific prevention strategies and adapt screening and treatment procedures, directly involving members of communities most at risk to improve their effectiveness. This type of benefit has led international institutions such as the WHO and ECDC to provide recommendations on ways of collecting information on individuals' health according to migrant status, with regard to the risk of COVID-19 as well as other health problems (https://www.who.int/migrants/en/; https://www.ecdc.europa.eu/en/methods/specific-populations/migrant-and-ethnic-groups). Third, in the absence of a balanced public discussion about the pros and cons of assessing the health of migrant and ethnic minority groups, the taboo relative to measurement of geographical origins is such that even for research projects specifically designed to address health disparities, obtaining authorization to collect information on participants' geographical region of origin can prove impossible.

The demonstration of elevated rates of health problems—particularly infectious diseases—among migrant or racialized populations does carry risks, the greatest likely to be that of stigmatization. Such stigmatization can of course be deliberate, and if it leads to actual discrimination, consistent with the European Union's Charter of Fundamental Rights, it is prohibited by law. Lack of contextualization can lead to racialized characterization of behaviour instead of visualizing and detecting the structural and economic inequalities and the structural racism of society and of health systems that are the fundamental drivers of the disparities. However data on health disparities can also be misinterpreted, as when observed differences are mistakenly attributed to genetic makeup [6]. There is therefore need not only to collect data making it possible to observe and monitor disparities, but also accompany such data collection and analysis with thoughtful interpretation. One way of moving in that direction without doing wrong is most likely to get members of migrant and ethnic minority groups directly involved in research which pertains to their health. Data should also not only measure components late in the pathways of disease, morbidity and mortality, but capture upstream disparities in exposure and mechanisms contributing to vulnerability, which include social determinants of health and disease.

In France, the production of national mortality statistics according to immigration status during the COVID-19 epidemic is a first step towards a broader discussion about the necessity to measure and address inequalities in health between migrants—first, second, and third generation—and the majority population. More generally, in Europe, further progress will necessarily require deliberate consideration of discrimination with regard to health and healthcare, clarification of terms and indicators (migrant status, ethnicity) to help physicians and public health professionals understand how to interpret disparities based on social mechanisms of action and without falling into the ‘genetic trap’. The COVID-19 epidemic may be an opportunity for a renewed public discussion about the relevance of collecting and analysing data on migrant status and ethnicity and health in Europe, and a time to move forward with regard to politically and socially acceptable indicators to be used in public health research and surveillance.

Funding

None.

Transparency declaration

J.M. reports being an employee of bioMérieux. Other authors report no conflict of interest.

Author contributions

M.M., G.D.D., J.M., M.C. and J.S.K. had the original idea for the manuscript. M.M. drafted an initial version and all co-authors contributed to the literature review and writing of the text.