During the COVID-19 pandemic, severe inequalities of contracting and dying from the virus have been observed across populations and social groups. Multiple factors can underlie those differences, including demographic ones. Julien Giorgi (Insee, Department of Economic Studies) and Diederik Boertien (Centre d’Estudis Demogràfics) focus on the impact of household composition and within-household transmission on COVID-19 mortality inequalities by education level and migratory background. Inhabitants of the same dwelling indeed expose each other to the risk of becoming infected with the virus. Hence, different household composition can shape socio-economic inequalities in COVID-19 mortality through infections within the household. Two overriding factors are the size of households — limiting the maximum number of people to whom the virus can be transmitted following an initial infection— and the age of household members — given the higher mortality rates for older adults.
Using a representative sample of 19.6 million observations from French Census data, the authors model an individual measure of ‘vulnerability’. This measure estimates the number of deaths that would be caused by the transmission of the virus to other household members following a random initial infection of each individual studied. Vulnerability modelling factors in the household structure of all sampled individuals, the age-specific likelihood of transmitting the virus within the household and the age-specific likelihood of dying following an infection. As co-residential patterns differ throughout the life cycle, vulnerability indexes are averaged across 10-year age groups of initially infected people and, within each age-group, across different categories of the same socio-demographic variable. Socio-demographic variables are assigned at the household level using the information from the head of the household.
Infections of young people are simulated to lead to more deaths among less educated and foreign-born populations. The greatest dispersion occurs between 30 and 39 for the education level. In this age group, a person living in a household with the lowest level of education is estimated to cause 7.5 times more deaths through within-household transmission following their own infection than a person of the same age group in a household where the education level of the head of the household is a university degree. The relative difference peaks between 20 and 29 years of age for the place of birth. At that age, infected foreign-born household inhabitants are expected to cause on average 2.4 times more deaths through within-household transmission. These gaps then shrink and eventually reverse as the originally infected person gets older. In households with a university degree as the education level, the infection of someone in their 80s will cause 44 % more deaths compared to the infection of someone in this age group living in a household with the lowest education level. The same result holds true but is less pronounced for the birthplace variable.
Patterns in household size and composition explain these discrepancies. Before 60, large households are more prominent among the lower educated and migrant communities, who also live more regularly with at least one person who is one or two generations older. More educated and native-born people also live alone more often between 20 and 40 due to leaving their parents’ home earlier and marrying later. After age 60, the higher educated are less likely to live alone than lower educated persons because they still live with their partner. Foreign-born people more often live in larger households at older ages but live with fewer vulnerable people. Whereas native-born people predominantly live with their partner or alone at later ages, the foreign-born people less often live with a person from their own generation such as their partner.