Characteristics of COVID-19 patients admitted to a tertiary care hospital in Pune, India, and cost-effective predictors of intensive care treatment requirement

  1. Urvi Bhooshan Shukla
  2. Sharvari Rahul Shukla
  3. Sachin Bhaskar Palve
  4. Rajiv Chintaman Yeravdekar
  5. Vijay Madhusoothan Natarajan
  6. Pradeep Tiwari
  7. Chittaranjan Sakerlal Yajnik

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

Maharashtra is one of the worst affected states in this pandemic. 2 As of 30th September, Maharashtra has in total 1.4 million cases with 38,000 deaths. Objective was to study associations of severity of disease and need for ICU treatment in COVID-19 patients.

Methods

A retrospective study of clinical course in 800 hospitalized COVID-19 patients, and a predictive model of need for ICU treatment. Eight hundred consecutive patients admitted with confirmed COVID-19 disease.

Results

Average age was 41 years, 16% were <20 years of age, 55% were male, 50% were asymptomatic and 16% had at least one comorbidity. Using MoHFW India severity guidelines, 73% patients had mild, 6% moderate and 20% severe disease. Severity was associated with higher age, symptomatic presentation, elevated neutrophil and reduced lymphocyte counts and elevated inflammatory markers. Seventy-seven patients needed ICU treatment: they were older (56 years), more symptomatic and had lower SpO2 and abnormal chest X-ray and deranged hematology and biochemistry at admission. A model trained on the first 500 patients, using above variables predicted need for ICU treatment with sensitivity 80%, specificity 88% in subsequent 300 patients; exclusion of expensive laboratory tests did not affect accuracy.

Conclusion

In the early phase of COVID- 19 epidemic, a significant proportion of hospitalized patients were young and asymptomatic. Need for ICU treatment was predicted by simple measures including higher age, symptomatic onset, low SpO2 and abnormal chest X-ray. We propose a cost-effective model for referring patients for treatment at specialized COVID-19 hospitals.

Key Messages

  • Of 800 patients, 73% had mild, 6% moderate and 20% had severe disease.

  • Seventy-seven patients (9.6%) required ICU treatment, 25 (3%) died.

  • ICU treatment was predicted by higher age, more symptomatic presentation, lower SpO2 and pneumonia on chest X-ray at admission.

  • A machine learning model features in first 500 patients accurately predicted ICU treatment in subsequent 300 patients.

  • A good clinical protocol, SpO2 and chest X-ray are adequate to predict and triage COVID-19 patients for hospital admissions in resource poor environments.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.11.26.20239186: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board StatementConsent: All patients signed a written informed consent at the time of admission which permitted use of anonymized data for research.
    IRB: The Institutional Research Committee (IRC) of Symbiosis Medical College for Women gave necessary approvals.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Analyses were carried out using IBM SPSS Statistics for Windows, version 23.0 (IBM Corp.,
    SPSS
    suggested: (SPSS, RRID:SCR_002865)
    The model was built with 10,000 trees using randomForest package6 in R statistical programming language.7
    randomForest
    suggested: (RandomForest Package in R, RRID:SCR_015718)

    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.

    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 10.1101/2020.11.26.20239186v1 on medRxiv