How closely is COVID-19 related to HCoV, SARS, and MERS? : Clinical comparison of coronavirus infections and identification of risk factors influencing the COVID-19 severity using common data model (CDM)

  1. Yeon Hee Kim
  2. Ye-Hee Ko
  3. Sooyoung Kim
  4. Kwangsoo Kim

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

South Korea was one of the epicenters for both the 2015 Middle East Respiratory Syndrome and 2019 COVID-19 outbreaks. However, there has been a lack of published literature, especially using the Electronic Medical Records (EMR), that provides a comparative summary of the prognostic factors present in the coronavirus-derived diseases. Therefore, in this study, we aimed to evaluate the distinct clinical traits between the infected patients of different coronaviruses to observe the extent of resemblance within the clinical features and to identify unique factors by disease severity that may influence the prognosis of COVID-19 patients.

Methods

We utilized the common data model (CDM), which is the database that houses the standardized EMR. We set COVID-19 as a reference group in comparative analyses. For statistical methods, we used Levene’s test, one-way Anova test, Scheffe post-hoc test, Games-howell post-hoc test, and Student’s t-test for continuous variables, and chi-squared test and Fisher’s exact test for categorical variables. With the variables that reflected similarity in more than two comparisons between the disease groups yet significantly different between the COVID-19 severity groups, we performed univariate logistic regression to identify which common manifestations in coronaviruses are risk factors for severe COVID-19 outcomes.

Findings

We collected the records of 2840 COVID-19 patients, 67 MERS patients (several suspected cases included), 43 SARS suspected patients, and 87 HCoV patients. We found that a significantly higher number of COVID-19 patients had been diagnosed with comorbidities compared to the MERS and HCoV groups (48.5% vs. 10.4 %, p < 0.001 and 48.5% vs. 35.6%, p < 0.05) and also that the non-mild COVID-19 patients reported more comorbidities than the mild group (55.7% vs. 47.8%, p < 0.05). There were overall increases in the levels of fibrinogen in both sets of disease and severity groups. The univariate logistic regression showed that the male sex (OR: 1.66; CI: 1.29-2.13, p < 0.001), blood type A (OR: 1.80; CI: 1.40-2.31, p < 0.001), renal disease (OR: 3.27; CI: 2.34-4.55, p < 0.001), decreased creatinine level (OR: 2.05; CI: 1.45-2.88, p < 0.001), and elevated fibrinogen level (OR: 1.59, CI: 1.21-2.09, p < 0.001) are associated with the severe COVID-19 prognosis, whereas the patients reporting gastrointestinal symptoms (OR: 0.42; CI: 0.23-0.72, p < 0.01) and increased alkaline phosphatase (OR: 0.73; CI: 0.56-0.94, p < 0.05) are more less likely to experience complications and other severe outcomes from the SARS-CoV-2 infection.

Interpretation

The present study observed the highest resemblance between the COVID-19 and SARS groups as clinical manifestations that were present in SARS group were linked to the severity of COVID-19. In particular, male individuals with blood type A and previous diagnosis of kidney failure were shown to be more susceptible to developing the poorer outcomes during COVID-19 infection, with a presentation of elevated level of fibrinogen.

Article activity feed

  1. Version published to 10.1101/2020.11.23.20237487v4 on medRxiv
  2. Version published to 10.1101/2020.11.23.20237487v3 on medRxiv
  3. ScreenIT

    SciScore for 10.1101/2020.11.23.20237487: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board StatementIRB: The study was approved by the Seoul National University Hospital Institutional Review Board (Seoul, South Korea) on April 2nd, 2020. II. Study Cohort: The follow-up period was from October 15th of 2004 to July 31st of 2020.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.
    Cell Line Authenticationnot detected.

    Table 2: Resources

    Experimental Models: Cell Lines
    SentencesResources
    , HCoV-NL63, HCoV-OC43, HCoV-HKU1 (HCoVs), SARS-CoV-1, MERS-CoV, and SARS-CoV-2 infections using International Classification of Diseases, Tenth Revision (ICD-10)17.
    HCoV-NL63
    suggested: RRID:CVCL_RW88)

    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:
    Our study had several limitations; first, there existed missing data, which may explain a disparity in our creatinine analyses. Considering the relative nature of logistic regression, we speculate the model to presume the contradictory correlations between the disease severity and creatinine level—only 16% of the individuals with high creatinine level belonged to the non-mild group, yet it should be noted that the number of admitted severe patients in Seoul National University Hospital was relatively low. Further, there were missing PCR data available on our database— although we utilized the diagnoses that existed on the system, there may involve the suspected cases, especially within the SARS group. Therefore, this study provides the limited clinical observations that represent the disease groups. Second, the study population was limited to the patients at a single national hospital due to the limited study period. Thus, the results in this study should not be generalized fully to other populations and ought to be considered with a caution. Yet, we maximized the benefit of utilizing the anonymized EMR data integrated into CDM, which now is expanding in its scope globally for the vitalization of medical research67. Upon choosing a database, we ensured that the domains from the CDW were equally available in our CDM and that there were no gaps between the amount of data in each database with the frequent extract, transform, and load (ETL) process. Thus, through the use of CDM,...

    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.

  4. Version published to 10.1101/2020.11.23.20237487v2 on medRxiv
  5. Version published to 10.1101/2020.11.23.20237487v1 on medRxiv