How high and long will the COVID-19 wave be? A data-driven approach to model and predict the COVID-19 epidemic and the required capacity for the German health system

  1. Thomas Klabunde
  2. Clemens Giegerich

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background and objective

In March 2020 the SARS-CoV-2 outbreak has been declared as global pandemic. Most countries have implemented numerous “social distancing” measures in order to limit its transmission and control the outbreak. This study aims to describe the impact of these control measures on the spread of the disease for Italy and Germany, forecast the epidemic trend of COVID-19 in both countries and estimate the medical capacity requirements in terms of hospital beds and intensive care units (ICUs) for optimal clinical treatment of severe and critical COVID-19 patients, for the Germany health system.

Methods

We used an exponential decline function to model the trajectory of the daily growth rate of infections in Italy and Germany. A linear regression of the logarithmic growth rate functions of different stages allowed to describe the impact of the “social distancing” measures leading to a faster decline of the growth rate in both countries. We used the linear model to predict the number of diagnosed and fatal COVID-19 cases from April 10 th until May 31 st . For Germany we estimated the required daily number of hospital beds and intensive care units (ICU) using clinical observations on the average lengths of a hospital stay for the severe and critical COVID-19 patients.

Results

Analyzing the data from Germany and Italy allowed us to identify changes in the trajectory of the growth rate of infection most likely resulted from the various “social distancing” measures implemented. In Italy a stronger decline in the growth rate was observed around the week of March 17 th , whereas for Germany the stronger decline occurred approximately a week later (the week of March 23 rd ). Under the assumption that the impact of the measures will last, the total size of the outbreak can be estimated to 155,000 cases in Germany (range 140,000-180,000) and to 185,000 cases in Italy (range 175,000-200,000). For Germany the total number of deaths until May 31 st is calculated to 3,850 (range 3,500-4,450). Based on the projected number of new COVID-19 cases we expect that the hospital capacity requirements for severe and critical cases in Germany will decline from the 2 nd week of April onwards from 13,500 to ∼2500 hospital beds (range 1500-4300) and from 2500 to ∼500 ICU beds in early May (range 300-800).

Conclusions

The modeling effort presented here provides a valuable framework to capture the impact of the “social distancing” measures on the COVID-19 epidemic in European countries and to forecast the future trend of daily COVID-19 cases. It provides a tool for medical authorities in Germany and other countries to help inform the required hospital capacity of the health care system. Germany appears to be in the middle of the (first) COVID-19 outbreak wave and the German health system is well prepared to handle it with the available capacities.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.04.14.20064790: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Evidently projections in the future come with certain limitations. First the projection for new COVID-19 cases assume a continuation of (similar) control measures of social distancing having the same impact on the growth rate of new cases. They can only provide a trend, can’t be applied to project further out into the future and obviously are not able to predict the size of a second wave (or whether there would be one at all). In addition, for the calculations used above to estimate the risk for the overwhelming of the German health care system, several assumptions are based on clinical data from the epidemic in China. Evidently due to differences in the age distribution of patients and in the health system between both countries these can only be estimates for the situation in Germany. Nevertheless, it appears that several of these assumptions translate into projections of death rate and capacity needs for ICU beds that are in-line with the observed clinical findings in Germany. We estimate the total rate of fatalities per diagnosed COVID-19 case to be 2.5% for Germany, resulting from the assumption that 20% of the diagnosed cases are severe, 25% of the severe cases get critical and 50% of the critical cases are fatal. On April 12th for 2.2% of diagnosed COVID-19 cases in Germany a fatal outcome has been reported [11]. While the epidemic is still ongoing, this formula to calculate the final death rate can be misleading and may lead to a too low number as of today the outcome...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.04.14.20064790v1 on medRxiv