Predicting prognosis in COVID-19 patients using machine learning and readily available clinical data

  1. Thomas W. Campbell
  2. Melissa P. Wilson
  3. Heinrich Roder
  4. Samantha MaWhinney
  5. Robert W. Georgantas
  6. Laura K. Maguire
  7. Joanna Roder
  8. Kristine M. Erlandson

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

No abstract available

Article activity feed

  1. Version published to 10.1016/j.ijmedinf.2021.104594
  2. ScreenIT

    SciScore for 10.1101/2021.01.29.21250762: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board StatementIRB: The study was reviewed and approved by the Colorado Multiple Institutional Review Board.
    Randomizationnot detected.
    BlindingDeidentified data for the validation cohort were transferred to Biodesix for blinded test classification generation.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Patient information for attributes deemed potentially useful at the time was extracted from the EHR by medical students and stored in a REDCap database (21).
    REDCap
    suggested: (REDCap, RRID:SCR_003445)
    Statistical analyses were performed using SAS Enterprise Guide 8.2 (SAS 9.4) (SAS Institute, Cary, NC).
    SAS Institute
    suggested: (Statistical Analysis System, RRID:SCR_008567)

    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:
    One limitation of this study is the relatively small and geographically restricted validation cohort. While patients excluded from the development cohort due to missing data were generally similar to those included, in validation, patients with complete data (only 15% of the total available) exhibited higher rates of severe disease and generally worse prognostic factors (laboratory and vital signs) than those without. Further validation of the test in larger cohorts derived from other health systems and geographic areas is necessary. In summary, we have developed and validated a suite of tests able to assess the risk of a poor outcome for patients hospitalized with COVID-19 based on information easily and routinely collected at time of hospital admission. Additional validation, preferably in a prospective setting, is required to further demonstrate the clinical utility of this risk assessment tool beyond clinical assessment alone. However, with readily-derived and quickly-available EHR data, a risk assessment at or near the time of admission can inform prognosis, guide discussions on the risks and benefits of treatments (including intubation), or identify low or high-risk patients for limited resources or enrollment in clinical trials. Furthermore, the methods here may be implemented in the care of future patients with novel viral infections.

    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 10.1101/2021.01.29.21250762v1 on medRxiv