A Dynamic Bayesian Model for Identifying High-Mortality Risk in Hospitalized COVID-19 Patients

  1. Amir Momeni-Boroujeni
  2. Rachelle Mendoza
  3. Isaac J. Stopard
  4. Ben Lambert
  5. Alejandro Zuretti

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

As Coronavirus Disease 2019 (COVID-19) hospitalization rates remain high, there is an urgent need to identify prognostic factors to improve patient outcomes. Existing prognostic models mostly consider the impact of biomarkers at presentation on the risk of a single patient outcome at a single follow up time. We collected data for 553 Polymerase Chain Reaction (PCR)-positive COVID-19 patients admitted to hospital whose eventual outcomes were known. The data collected for the patients included demographics, comorbidities and laboratory values taken at admission and throughout the course of hospitalization. We trained multivariate Markov prognostic models to identify high-risk patients at admission along with a dynamic measure of risk incorporating time-dependent changes in patients’ laboratory values. From the set of factors available upon admission, the Markov model determined that age >80 years, history of coronary artery disease and chronic obstructive pulmonary disease increased mortality risk. The lab values upon admission most associated with mortality included neutrophil percentage, red blood cells (RBC), red cell distribution width (RDW), protein levels, platelets count, albumin levels and mean corpuscular hemoglobin concentration (MCHC). Incorporating dynamic changes in lab values throughout hospitalization lead to dramatic gains in the predictive accuracy of the model and indicated a catalogue of variables for determining high-risk patients including eosinophil percentage, white blood cells (WBC), platelets, pCO2, RDW, large unstained cells (LUC) count, alkaline phosphatase and albumin. Our prognostic model highlights the nuance of determining risk for COVID-19 patients and indicates that, rather than a single variable, a range of factors (at different points in hospitalization) are needed for effective risk stratification.

Article activity feed

  1. Version published to 10.3390/idr13010027
  2. ScreenIT

    SciScore for 10.1101/2021.02.02.21251023: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board StatementIRB: Case Selection and Data Extraction: Approval for the study was obtained from the State University of New York, Downstate Medical Center Institutional Review Board (IRB#1595271-1).
    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:
    Given the reactive nature of the pandemic response, prognostic modelling efforts have suffered from a number of limitations, including lack of external validation and a high risk of bias (16). We fit our model to data from a single hospital in New York, where our study population mostly consists of black African American patients, which have been shown to be disparately affected by the COVID-19 pandemic (31, 32). Socio-economic variables can impact in-hospital mortality risk (9, 33) and external validation of ours and other models is therefore required. Finally, any clinical implementation requires an assessment of the impact of the prognostic model on clinicians’ behavior, patient health and associated costs (15). Our model allows for risk stratification and triage. At the patient level, the model allows for individualization of care for each hospitalized COVID-19 patient and identification of need for additional levels of care (34); the main advantage of our approach is the incorporation of dynamic variables which allows for daily adjustments to the patient’s risk level. In addition, identification of high-risk patients and determining the surge capacity needed for advanced intensive care should allow for early resource allocation and ultimately improved outcomes for patients (6, 35). However, with insufficient surge capacity, triage should consider both patient prognosis and ethical considerations to avoid health inequities (36, 37). A number of vaccines are in the process...

    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.

  3. Version published to 10.1101/2021.02.02.21251023v1 on medRxiv