Visualizing the Chain of SARS-CoV-2 Infection in Luxembourg

Context: The SARS-CoV-2 positive cases had an unabated near-exponential rise in a matter of days across almost all of Europe, starting at some point in October 2020, and eventually topped 2 to 4 times the maximun cases recorded in individual countires (first wave), and Luxembourg was no exception. These recent high numbers with the ‘background data’ during first wave may have an interesting outlook, one being to find the ‘chain-of-infections’ and if it has deviated from the first-wave or not. It is NOT about finding the first cases but if we could manage to compartmentalize the recent cases into different variables, like ‘sectors’, ‘age-groups’, ‘regions (for bigger countries)’, etc. and then make a chart of daily cases across these variables, theoretically it is possible to check if there was a ‘chain-of-infections’ between the variables or not.

Data: We have limited data for different variables available publicly and to test the theory, here we have used the variable ‘age-groups’ (Luxembourg resident population) and them plotted against the weekly number of positive cases (Mon-Sun). Since the ‘age-groups’ are made up of different ‘size’ (number of people in each age-group), we normalized the cases by per 1,000 of each age-group for comparison purposes.

Assumptions: Before the observations are presented below, please note that certain assumptions are necessary but may not be 100% accurate. First assumption would be that every week the number of tests done are evenly distributed (pro-rata) amongst all age-groups. While, if the number of tests in each groups is available, then we could have normalized the number of tests and had a more accurate assumption, but without this data, we have to assume that the tests were evenly distributed amongst the age-groups. A second assumption would be that majority of each age-group ‘socializes’ with the same age group. Without this assumption, it would be hard to follow the chain within the same age-group and which slowly progresses to other age-groups.
Both these assumptions are critical for the abservations below.

Observations: It is a great pleasure to note that this week, 17 – 23 May 2021, the cases per thousand residents (cumulative, all ages) is below 1. It has been for 35 weeks that the weekly number of new cases were above 1 per 1,000 residents, last noted in October 2020. It is a ‘positive improvement’ and hope we continue the trend.

If you are using a smartphone or a small-display, please rotate the device to ‘landscape view’ for better visualization

first published : 28.11.2020
data updated: 26.05.2021

✏️ for comments or suggestion, message us on facebook page.