P48 Risk of COVID-19 infection among men and women during the lockdown of spring 2020 in France

BackgroundIn the context of the Covid-19 pandemic, several factors such as age, chronic disease or obesity have been associated with adverse outcomes and mortality from Covid-19. However, the social distribution of Covid-19 infection among men and women was largely neglected in France, mainly due to a lack of data. The aim of this study is to describe and analyse the risk of Covid-19 infection in relation to sex, and the influence of other social factors, specifically occupation, in this association.MethodsWe used data from the citizen science initiative ‘Barometre Covid-19’. Each week, an internet survey was administered to a sample of 5,000 people representative of the French mainland population aged 18 and over, using the quota method. A total of 25,001 participants were interviewed between 7 April and 11 May 2020. We used multivariable nested logistic regression modelling to study the relationship between sex, occupation and Covid-19 infection. Confounders included age, region of residence, population density, whether you worked outside of home during the lockdown, house overcrowding, comorbidities and body mass index.ResultsWomen reported a medical diagnosis of Covid-19 infection more often than men (4% vs. 3.2%). In a model adjusted for confounders, women were 23% more likely to report a medical diagnosis of Covid-19 infection than men (OR=1.23 [95%-CI=1.06–1.42]). Controlling for sex and socioeconomic variables (occupation), the risk of infection for women was reversed (OR=0.84 [95%-CI=0.59–1.19]). While most men, other than executives, were less likely to report the infection, this association was not observed amongst women.ConclusionOccupation was found to influence the relationship between sex and Covid-19 infection suggesting a gender effect. The differences in the risk of infection between men and women require exploration with regard to socioeconomic factors. The social roles of women and men are associated with a non-random distribution of the virus, potentially reflecting structural societal inequalities.