Simplified model of the number of Covid-19 patients in the ICU: update April 6, 2020

  1. A. Cividjian
  2. F Wallet
  3. C. Guichon
  4. O. Martin
  5. S. Couray-Targe
  6. T Rimmelé
  7. PF Wey

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

INTRODUCTION

Predicting the number of Covid-19 patients in the Intensive Care Units (ICU) could be useful to avoid the breaking point. We attempted to deduce a formula in order to model the number of the ICU patients in France from the official data and patient turnover in the ICU.

METHODS

The Covid-19 ICU patient turnover was calculated using a recurrence relation from the internal data provided by Hospices Civils de Lyon. The number of new Covid-19 cases detected daily was modelized to fit with the last known data in France and extrapolated for the coming days using two scenarios following the existing data in China (best scenario) and Italy (worst scenario). The number of daily admissions in ICU was calculated as the sum of 13.7% of the new Covid-19 cases detected on a given day and 7.8% of the average of the total new Covid-19 cases recorded in the last week. Approximately 39.7% of patients admitted to the ICU were non-intubated with an average ICU length of stay of 4 days. Conversely, 60.3% of patients were intubated and for those who died among them (14.44%) the ICU length of stay was of 4 days for 78.3% of them and of 15 days for 21.7% of them. For the intubated patients that were discharged alive, the ICU length of stay was of 6 days for 44.4% of them and of 20 days for 55.6% of them.

RESULTS

We predict a peak of 7072 – 8043 patients for the overall French territory.

CONCUSION

Despite a simplified mathematical model, the strength of our study is a narrow possible range of predicted total number of ICU patients.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.04.07.20056226: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board Statementnot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    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:
    This mathematical model presents several limitations. First, the spread of Covid-19 is not homogeneous in France, with a higher density in the Eastern part of France and around Paris. The presented model uses all the pooled data irrespective of the widespread. Second, all the proportions and coefficients used in the model can vary in time and between hospitals. Third, age and patient comorbidities were not considered. The variability of all these factors should be considered for a more accurate model. The formula used in the model is the same for both scenarios, the only difference being the curve of the new Covid-19 cases. The Chinese and the Italian scenarios were taken as references because the former corresponds to an ideal case with quasi-absolute containment, and the latter corresponds to one of the highest known cases of prevalence with containment conditions similar to the French ones. Thus, the strength of our study is a narrow possible range of predicted total number of ICU patients, namely a peak of 7072 – 8043 patients for the overall French territory. Such a prediction could help to avoid the ICU bed capacity reaching breaking point. Mixing our method with the already published methods such as (1) could yield more precise results. A last point, is that after the start of containment, only the severe cases of Covid-19 have been tested in France and consequently the ratios kd = 13.7% and kw = 7.8% estimated from the Hospices Civils de Lyon (4) were higher than the ...

    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 10.1101/2020.04.07.20056226v1 on medRxiv