Outbreak detection in Harar town and Kersa district, Ethiopia using phylogenetic analysis and source attribution

Foodborne diseases (FBDs) represent a significant risk to public health, with nearly one in ten people falling ill every year globally. The large incidence of foodborne diseases in African low- and middle-income countries (LMIC) shows the immediate need for action, but there is still far to a robust...

Full description

Saved in:
Bibliographic Details
Published in:BMC infectious diseases 2024-08, Vol.24 (1), p.864-11
Main Authors: Thystrup, Cecilie, Hald, Tine, Belina, Dinaol, Gobena, Tesfaye
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Animal human relations
Animals
Clusters
Cross-sectional studies
Denmark
Diarrhea
Disease
Disease Outbreaks
Epidemics
Epidemiology
Ethiopia - epidemiology
Food products
Food safety
Food sources
Foodborne diseases
Foodborne Diseases - epidemiology
Foodborne Diseases - microbiology
Genetic aspects
Genomes
Households
Humans
Infections
Low- and middle-income countries
Machine learning
Medical research
Medicine, Experimental
Outbreak detection
Outbreaks
Pathogens
Phylogenetics
Phylogeny
Physiological aspects
Public health
Robustness
Salmonella
Salmonella - classification
Salmonella - genetics
Salmonella - isolation & purification
Salmonella Infections - epidemiology
Salmonella Infections - microbiology
Salmonellosis
Sequences
Social aspects
Surveillance
Surveillance systems
Testing
Typhoid
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Foodborne diseases (FBDs) represent a significant risk to public health, with nearly one in ten people falling ill every year globally. The large incidence of foodborne diseases in African low- and middle-income countries (LMIC) shows the immediate need for action, but there is still far to a robust and efficient outbreak detection system. The detection of outbreak heavily relies on clinical diagnosis, which are often delayed or ignored due to resource limitations and inadequate surveillance systems. In total, 68 samples of non-typhoidal Salmonella isolates from human, animal and environmental sources collected between November 2021 and January 2023 were analyzed using sequencing methods to infer phylogenetic relationships between the samples. A source attribution model using a machine-learning logit-boost that predicted the likely source of infection for 20 cases of human salmonellosis was also run and compared with the results of the cluster detection. Three clusters of samples with close relation (SNP difference 
ISSN:1471-2334
1471-2334
DOI:10.1186/s12879-024-09800-4