| von Sascha Mertens

Analytical ad-hoc evaluations using the example of COVID-19

This article uses the "Corona Dashboard" as an example to show what can be achieved with self-service visualizations

Please note: The dashboard "Utilization of hospital beds" does not represent the actual data on the pandemic, but rather specially derived key figures based on data from the RKI.

In the context of the Corona pandemic, I keep noticing that the data and metrics provided are not particularly strong. Many Corona-Dashboards focus only on the total number of infections, deaths, and recoveries, without at least giving the data a time trend or relations. The Robert Koch Institute's dashboard is a bit more granular. There you can analyze regional developments and get time courses important for ad-hoc evaluations. 

In my opinion, however, the crucial key figures are those relating to the utilization of the healthcare system: How much capacity is there in terms of hospital beds? I have developed a dashboard for this and would like to briefly present how I approached the building process. 

Merge, process and visualize data

In addition to the RKI figures on the incidence of infections, I used data from the German hospital register as a basis for the data, from which I obtain the number of beds of each hospital. Furthermore, I narrowed down the selection of hospitals slightly: Among other things, only hospitals with intensive care beds and/or pulmonary specialization and more than 100 beds were considered. I have grouped directly adjacent hospitals together. 

For the ad-hoc evaluation, I merged this data from the German hospital directory and the RKI (see animated figure above). First, I defined a geographic perimeter for each hospital ("catchment area"). After further data preparation and geocoding, I mapped the RKI data to the geography. This allows me to specifically assign infection numbers to the catchment areas of the respective hospitals in in consideration of their county. 

The result is a data visualization, including a chronological progression, which quickly shows the regional burden on hospitals. To the right of the map of Germany, other important key figures on the pandemic are also visualized, for example the course of new infections and the ratio of the number of patients seen and the number of deaths in relation to the total number of infections. In the lower right-hand corner of the dashboard, you will also find the overall utilization of hospitals over time. 


Teile diesen Artikel mit anderen


Über den Autor

Mit seiner langjährigen Erfahrung in den Themen Unternehmensplanung, Reporting und Data Preparation ist Sascha bei Woodmark Ansprechpartner für die Erschließung neuer Lösungswege und innovativer Konzepte. Im Mittelpunkt seiner Kunden-Tätigkeit stehen Workshops, in denen maßgeschneiderte Lösungen für komplexe Fachanforderungen konzipiert und entwicelt werden.