In this work, we establish a methodological framework to analyze the care demand for elderly citizens in any area with a large proportion of elderly population, and to find connections to the cumulative incidence of COVID-19. Thanks to this analysis, it is possible to detect deficiencies in the public elderly care system, identify the most disadvantaged areas in this sense, and reveal convenient information to improve the system. The methods used in each step of the framework belong to data analytics: choropleth maps, clustering analysis, principal component analysis, or linear regression. We applied this methodology to Barcelona to analyze the distribution of the demand for elderly care services. Thus, we obtained a deeper understanding of how the demand for elderly care is dispersed throughout the city. Considering the characteristics that were likely to impact the demand for homecare in the neighborhoods, we clearly identified five groups of neighborhoods with different profiles and needs. Additionally, we found that the number of cases in each neighborhood was more correlated to the number of elderly people in the neighborhood than it was to the number of beds in assisted living or day care facilities in the neighborhood, despite the negative impact of COVID-19 cases on the reputation of this kind of center.
Keywords: COVID-19; care system; data analytics; elderly population.