Space-time covid-19 Bayesian SIR modeling in South Carolina

  1. Andrew B. Lawson
  2. Joanne Kim

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

The Covid-19 pandemic has spread across the world since the beginning of 2020. Many regions have experienced its effects. The state of South Carolina in the USA has seen cases since early March 2020 and a primary peak in early April 2020. A lockdown was imposed on April 6th but lifting of restrictions started on April 24th. The daily case and death data as reported by NCHS (deaths) via the New York Times GitHUB repository have been analyzed and approaches to modeling of the data are presented. Prediction is also considered and the role of asymptomatic transmission is assessed as a latent unobserved effect. Two different time periods are examined and one step prediction is provided. The results suggest that both socio-economic disadvantage, asymptomatic transmission and spatial confounding are important ingredients in any model pertaining to county level case dynamics.

Article activity feed

  1. Version published to 10.1371/journal.pone.0242777
  2. ScreenIT

    SciScore for 10.1101/2020.11.03.20225227: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board Statementnot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: Thank you for sharing your data.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    5.1 Further Modeling considerations and limitations: Currently available data are limited by a number of factors. Under-ascertainment of positive cases is a major concern. Asymptomatic or pre-symptomatic people are not counted. Further some patients may not attend for Covid-19 screening even if they are in fact positive for the disease. In addition the under-ascertainment could be spatially and temporally differentiated in that it could vary over time and between areas. For normal influenza seasons, health departments often use a scale factor to allow for the iceberg effect of not observing the true case load, when only lab confirmed cases are presented. This could be added for Covid-19 of course. As an alternative the true case load could be included within a Bayesian model as a latent effect. An additional consideration is whether to model the death counts. If the focus is on the spread of Covid-19 then the death count is only a secondary outcome. In fact it is dependent on the true case count as opposed to infection process. A Poisson or binomial model could be assumed for this purpose. However as the focus here was on infection spread and case counts it has been assumed that death count is given and assumed fixed. Another aspect of the data quality which has not been addressed here is the fact that the degree of testing carried out affects the number of positive cases found. Assume the following, conditioning on the number of tests:

    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 10.1101/2020.11.03.20225227v1 on medRxiv