Brazilian Modeling of COVID-19(BRAM-COD): a Bayesian Monte Carlo approach for COVID-19 spread in a limited data set context

  1. Samy Dana
  2. Alexandre B. Simas
  3. Bruno A. Filardi
  4. Rodrigo N. Rodriguez
  5. Leandro da Costa Lane Valiengo
  6. Jose Gallucci-Neto

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

1
1.1

Background

The new coronavirus respiratory syndrome disease (COVID-19) pandemic has become a major health problem worldwide. Many attempts have been devoted to modeling the dynamics of new infection rates, death rates, and the impact of the disease on health systems and the world economy. Most of these modeling concepts use the Susceptible-Infectious-Susceptible (SIS) and Susceptible-Exposed-Infected-Recovered (SEIR) compartmental models; however, wide imprecise outcomes in forecasting can occur with these models in the context of poor data, low testing levels, and a nonhomogeneous population.

1.2

Objectives

To predict Brazilian ICU beds demand over time and during COVID-19 pandemic “peak”.

1.3

Methods

In the present study, we describe a Bayesian COVID-19 model combined with a Hamiltonian Monte Carlo algorithm to forecast quantitative predictions of infections, number of deaths and the demand for critical care beds in the next month in the Brazilian context of scarce data availability. We also estimated COVID-19 spread tendency in the state of São Paulo and forecasted the demand for critical care beds, as São Paulo is the epicenter of the Latin America pandemic.

1.4

Results

Our model estimated that the number of infected individuals would be approximately 6.5 million (median) on April 25, 2020, and would reach 16 to 17 million (median) by the end of August 2020 in Brazil. The probability that an infected individual requires ICU-level care in Brazil is 0.5833%. Our model suggests that the current level of mitigation seen in São Paulo is sufficient to reach R t < 1, thus attaining a “peak” in the short term. In São Paulo state, the total number of deaths is estimated to be around 9,000 (median) with the 2.5% quantile being 6,600 deaths and the 97.5% quantile being around 13,350 deaths. Also, São Paulo will not attain its maximum capacity of ICU beds if the current trend persists over the long term.

1.5

Conclusions

The COVID-19 pandemic should peak in Brazil between May 8 and May 20, 2020 with a fatality rate lower than that suggested in the literature. The northern and northeastern regions of Brazil will suffer from a lack of available ICU beds, the southern and central-western regions appear to have sufficient ICU beds, finally, the southeastern region seems to have enough ICU beds only if it shares private beds with the publicly funded Unified Health System (SUS). The model predicts that, if the current policies and population behavior are maintained throughout the forecasted period, by the end of August 2020, Brazil will have around 7.6% to 8.2% of its population immune to COVID-19.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.04.29.20081174: (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: An explicit section about the limitations of the techniques employed in this study was not found. We encourage authors to address study limitations.

    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 10.1101/2020.04.29.20081174v3 on medRxiv
  3. Version published to 10.1101/2020.04.29.20081174v2 on medRxiv
  4. Version published to 10.1101/2020.04.29.20081174v1 on medRxiv