Longitudinal Changes in COVID-19 Associated In-Hospital Mortality

  1. Frederick Warner
  2. Matt D.T. Hitchings
  3. Derek A.T. Cummings
  4. Jacob McPadden
  5. Harlan M. Krumholz
  6. Albert I. Ko
  7. Wade L. Schulz

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Objective

As the COVID-19 pandemic has evolved, a key question for health care systems is whether in-hospital mortality has changed over time and if so, what factors contributed to these changes. Our goal was to leverage real-world data spanning two COVID-19 surges over the first year of the pandemic to determine the temporal trend of in-hospital mortality.

Design

This was an observational, retrospective study based on real-world data for patients admitted with COVID-19. Generalized additive models (GAM) were used to evaluate the association of covariates with the composite outcome over time.

Setting and Population

We identified a retrospective cohort of all patients who were hospitalized within the Yale New Haven Health (YNHH) system with an admission diagnosis of COVID-19 between March 1, 2020 and February 28, 2021.

Main outcome

The primary outcome for the study was a composite endpoint of in-hospital mortality, defined as death during the index hospitalization or discharge to hospice.

Results

Among 6,477 discharges over the study period, the mean age was 66.2 years (SD=17.6), 52.5% (n=3,401) were male and the overall composite mortality was 14.2% (n=920). Composite in-hospital mortality was significantly associated with increased age, comorbidity index, respiratory rate, and heart rate; decreased systolic blood pressure; male sex; and admission from a long-term care facility (LTCF). The significant temporal decrease in mortality that was observed for patients admitted from a location other than a LTCF was not seen in those admitted from a LTCF.

Conclusions

We found that the adjusted in-hospital mortality rate declined over the first year of the pandemic, despite a second surge in COVID-19-related hospitalizations. Importantly, the decrease in mortality appeared to be driven by declines in risk in those not admitted from a LTCF. The observed decrease in mortality over time suggests that improved outcomes may be due to progressive, incremental learning and continuous evolution in hospital practice and policy over the course of the pandemic.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.05.04.21255938: (What is this?)

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

    Table 1: Rigor

    Ethicsnot detected.
    Sex as a biological variableOf discharged patients, the mean age was 66.2 years (SD=17.6), 52.5% (n=3,401) were male, 56.3% (n = 3,647) had a recorded race of White, 21.2% (n=1,374) had a recorded race of Black, 22.3% (n=1,446) had a recorded ethnicity of Hispanic or Latino, and 10.1% (n=656) were admitted from a LTCF.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot 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: 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.

    Results from scite Reference Check: We found no unreliable references.

    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 10.1101/2021.05.04.21255938v2 on medRxiv
  3. Version published to 10.1101/2021.05.04.21255938v1 on medRxiv