Predicting Hospital Utilization and Inpatient Mortality of Patients Tested for COVID-19

  1. Connor Davis
  2. Michael Gao
  3. Marshall Nichols
  4. Ricardo Henao

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Using structured elements from Electronic Health Records (EHR), we seek to: i ) build predictive models to stratify patients tested for COVID-19 by their likelihood for hospitalization, ICU admission, mechanical ventilation and inpatient mortality, and ii ) identify the most important EHR-based features driving the predictions. We leveraged EHR data from the Duke University Health System tested for COVID-19 or hospitalized between March 11, 2020 and August 24, 2020, to build models to predict hospital admissions within 4 weeks. Models were also created for ICU admissions, need for mechanical ventilation and mortality following admission. Models were developed on a cohort of 86,355 patients with 112,392 outpatient COVID-19 tests or any-cause hospital admissions between March 11, 2020 and June 4, 2020. The four models considered resulted in AUROC=0.838 (CI: 0.832-0.844) and AP=0.272 (CI: 0.260-0.287) for hospital admissions, AUROC=0.847 (CI: 0.839-855) and AP=0.585 (CI: 0.565-0.603) for ICU admissions, AUROC=0.858 (CI: 0.846-0.871) and AP=0.434 (CI: 0.403-0.467) for mechanical ventilation, and AUROC=0.0.856 (CI: 0.842-0.872) and AP=0.243 (CI: 0.205-0.282) for inpatient mortality. Patient history abstracted from the EHR has the potential for being used to stratify patients tested for COVID-19 in terms of utilization and mortality. The dominant EHR features for hospital admissions and inpatient outcomes are different. For the former, age, social indicators and previous utilization are the most important predictive features. For the latter, age and physiological summaries (pulse and blood pressure) are the main drivers.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.12.04.20244137: (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

    Software and Algorithms
    SentencesResources
    Hyperparameter tuning was performed by using the RandomizedSearchCV function in Sklearn (Phyton package)21 on the training and validation sets.
    Phyton
    suggested: None

    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 study has several limitations. First, though the sample size of our study is large compared to other published works16,17, our cohort, largely pulled from the Durham-Raleigh area in North Carolina, may not be representative of the characteristics of the whole US population. Second, though results are validated chronologically, they are representative of a single health system (DUHS), thus the models developed here will need to be properly validated before being applied to other health systems. Third, from a data source perspective, features being extracted from an EHR, are limited to the availability, correctness and completeness of fields being considered (see Table S3 in the Supplement). Further, some patients may have incomplete features and outcomes in scenarios where they were seen at facilities outside DUHS. Fifth, though we count with the first five months worth of patient data since the beginning of the pandemic, it is likely that patient distribution will drift due to changing infection rates, treatment strategies, novel therapeutics and yet to be approved vaccines, which may change the performance characteristics of the model and the subset of most important predictors. Given these circumstances, we advocate for models that are refined (or retrained) periodically to take advantage of the sample size gains and to account for rapid changes in the population distributions. As future work, we will aim to: i) expand the cohorts both in terms of sample size and featu...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We found bar graphs of continuous data. We recommend replacing bar graphs with more informative graphics, as many different datasets can lead to the same bar graph. The actual data may suggest different conclusions from the summary statistics. For more information, please see Weissgerber et al (2015).

    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 10.1101/2020.12.04.20244137v1 on medRxiv