An exploratory analysis of housing and the distribution of COVID-19 in Sweden

The impact of COVID-19 on various aspects of our life is evident. Proximity and close contact with individuals infected with the virus, and as well as the extent of such contact, contribute to the intensity of the spread of the virus. Healthy and infected household members, who both require sanctuary and quarantine space, come into proximate and extended contact in housing. In other words, housing and living conditions can impact the health of occupants and the spread of COVID-19. This study investigates the relationship between housing characteristics and variations in the spread of COVID-19 per capita across Sweden's 290 municipalities. For this purpose, we have used the number of infected COVID-19 cases per capita during the pandemic period, February 2020 through April 2021, per municipality. The focus is variables that measure housing and housing conditions in the municipalities. We have used exploratory and principal components analysis to reduce highly correlated variables into a set of linearly uncorrelated variables. We then use the generated variables to estimate direct and indirect effects in a spatial regression analysis. The results indicate that housing and housing availability are important explanatory factors for the geographical spread of COVID-19. Overcrowding, availability, and quality are all critical explanatory factors