Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

Epidemiology and Regional Predictors of COVID-19 Clusters: A Bayesian Spatial Analysis Through a Nationwide Contact Tracing Data

Authors
Hong, KwanYum, SujinKim, JeehyunYoo, DaesungChun, Byung Chul
Issue Date
20-10월-2021
Publisher
FRONTIERS MEDIA SA
Keywords
COVID-19; cluster analysis; disease cluster; epidemiology; risk factors; spatial analysis
Citation
FRONTIERS IN MEDICINE, v.8
Indexed
SCIE
SCOPUS
Journal Title
FRONTIERS IN MEDICINE
Volume
8
URI
https://scholar.korea.ac.kr/handle/2021.sw.korea/136024
DOI
10.3389/fmed.2021.753428
ISSN
2296-858X
Abstract
Purpose: Revealing the clustering risks of COVID-19 and prediction is essential for effective quarantine policies, since clusters can lead to rapid transmission and high mortality in a short period. This study aimed to present which regional and social characteristics make COVID-19 cluster with high risk.</p> Methods: By analyzing the data of all confirmed cases (14,423) in Korea between January 10 and August 3, 2020, provided by the Korea Disease Control and Prevention Agency, we manually linked each case and discovered clusters. After classifying the cases into clusters as nine types, we compared the duration and size of clusters by types to reveal high-risk cluster types. Also, we estimated odds for the risk factors for COVID-19 clustering by a spatial autoregressive model using the Bayesian approach.</p> Results: Regarding the classified clusters (n = 539), the mean size was 19.21, and the mean duration was 9.24 days. The number of clusters was high in medical facilities, workplaces, and nursing homes. However, multilevel marketing, religious facilities, and restaurants/business-related clusters tended to be larger and longer when an outbreak occurred. According to the spatial analysis in COVID-19 clusters of more than 20 cases, the global Moran's I statistics value was 0.14 (p < 0.01). After adjusting for population size, the risks of COVID-19 clusters were related to male gender (OR = 1.29) and low influenza vaccination rate (OR = 0.87). After the spatial modeling, the predicted probability of forming clusters was visualized and compared with the actual incidence and local Moran's I statistics 2 months after the study period.</p> Conclusions: COVID-19 makes different sizes of clusters in various contact settings; thus, precise epidemic control measures are needed. Also, when detecting and screening for COVID-19 clusters, regional risks such as vaccination rate should be considered for predicting risk to control the pandemic cost-effectively.</p>
Files in This Item
There are no files associated with this item.
Appears in
Collections
College of Medicine > Department of Medical Science > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetrics

Total Views & Downloads

BROWSE