Covidcare: Covid-19 Patient Number Predictor [EN]

Go to German Version/Deutsche Version

UPDATE August 2020: We haven taken the calculation module offline because the logic and data does not apply to the current developments of the pandemic any more.


The number of COVID-19 diseases (caused by the “Corona virus SARS-CoV-2”) is developing very dynamically in Germany as well as in other countries and actual groth parameters vary even from region to region. In particular, the crisis management teams of the federal states, districts and cities as well as the strategic operations management of the clinics, emergency services and health care providers are constantly striving to optimally plan the available resources. In particular, the emergency room availability, inpatient provisions as well as intensive care and ventilation resources are at the forefront.

This project provides tools with which, among other things, the expected number of patients of the above-mentioned areas can be estimated in advance.

Based on your daily updated data of confirmed regional SARS-CoV-2 infections, the cumulative capacity requirements are calculated in addition to the chronological progression of patient admissions.

COVIDcare is also unique in that, in addition to inpatient resources, the number of patients that you can discharge from emergency rooms, inpatient and intensive care units is also estimated on a daily basis. COVIDcare enables you to make a cumulative calculation of the actual occupancy of your care units and thus to calculate material and personnel requirements in advance.

IMPORTANT TO NOTE #1: This software does not do the thinking for you!

The underlying algorithm is super simple and documented here. You as the user decide, with the quality of your input, how good the results are (“Garbage in, garbage out”). Most of the input parameters are preset with best-estimate values, but they could be different for your individual case. You can find a detailed description of the parameters in the online documentation.

IMPORTANT TO NOTE #2: We do not guarantee or take responsibility for the results

This online service was created with great care but also under extreme time pressure (less than 72 hours from the idea’s conception to its go-live on the website). We are very confident that the calculations will at least provide useful results, but we cannot guarantee this. The assumptions formulated in the calculation model are also essentially based on the technical literature published to date and are subject to expected further changes. Therefore, please check the plausibility of all input data for your calculated results.

Online service “COVIDcare”

Notice: We are developing the online service right now. It will probably not be released until March. Until then, please use the interim solution from Friday’s blog article.

After launching the service, you must at least enter or adjust the following input parameters for your region:

Recommendation NameExplanation
?Start dateDay from which the most recent cumulative number of SARS-CoV-2 cases is available (probably from yesterday or today)
?Number of confirmed casesCurrent cumulative number of confirmed cases in the desired region. Current figures per county/municipality are available in the COVID-19 Dashboard of the RKI (Germany) or of Johns Hopkins University .
30%% growth of case numbers in 24hIn many regions the growth is currently about 30% from one day to the next. You have to check whether this also applies to your region.
5-Apr-2020Number of new infections starts decreasingDate estimate: At what point will the lockdown stop the spread? We expect this to happen about 14 days after an effective regional lockdown.
12%After that: decrease by %/dayDecrease in new infections per day (% vs. the previous day) from the date the number of new infections began to decrease. This is a very simple model of the decline in new infections.

Click here to go to the COVIDcare online service, which makes the algorithms documented here available for free online use. The service will be extended in the coming days.

How it all began

This project was initiated by Prof. Dr. Harald Dormann, president of a German emergency medicine foundation (Deutsche Stiftung Akut- und Notfallmedizin) from the City Hospital of my hometown Fürth, and quickly made available online by Dirk Paessler and many volunteers from the Paessler AG team.

Prof. Dr. Harald Dormann, head physician of urgent care at Clinic Fürth, had contacted me and asked me if I could write a predictor tool to estimate the expected flow of patients to Clinic Fürth by COVID-19 – with about 2-4 weeks visibility. The first version of the predictor went online after 2 days on March 20, 2020.

Then we, a small team of software specialists from the Paessler AG environment, provided a corresponding online service as part of a hackathon in only 3 days:

Documentation of the calculation

A detailed documentation is in the works. Until then please read the introductory article (March 20, 2020, in German): COVID-19: A simple predictor tool for case numbers and number of patients for hospitals and the update 1 (March 21, 2020, in German): Update: COVID-19 case number predictor tool for hospitals

Documentation of clinical parameters

Recommendation NameExplanation
3Time to emergency room visit (days)The time it takes from recognition of a “case” to first outpatient visit to the emergency room  
5Time to inpatient admission (days)The time it takes from recognition of a “case” until admission to the ER and inpatient treatment
2Time from inpatient to intensive care unit (days)The time it takes from inpatient admission to the intensive care unit
5.5Duration of inpatient stay (days)Average minimum number of days a patient is in the hospital
10Length of stay of patient in intensive care (days)Average number of days an intensive care patient is in intensive care (after beginning on the ward)
10Aftercare intensive-care inpatient (days)Average number of days an intensive-care patient stays on the ward after intensive care
5%% Outpatient infections in the emergency roomPercentage of (confirmed) cases that visit the ER once and go home again (without inpatient admissions)
10%Inpatient infections in hospitalPercentage of (confirmed) cases requiring inpatient treatment (4.4% is optimistic, 10% is pessimistic)
3%% Infections in intensive care unitsPercentage of (confirmed) cases requiring intensive care

Further updates are available here in the blog and on Twitter

If you want to be informed about further updates on this project, please “subscribe” to this blog (top right) or follow me on Twitter @dpaessler.


Please send an email to or follow me on Twitter @dpaessler.