Meshing Biosurveillance and Climate Data to inform Chikungunya Disease Surveillance

Vector-borne and zoonotic pathogens comprise a substantial portion of the global disease burden causing ∼1.4 million deaths annually, and account for approximately 17% of the entire disease burden due to infectious diseases. Current Public Health and Department of Defense surveillance systems track...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of infectious diseases Vol. 116; p. S97
Main Authors:	Anyamba, A., Bishnoi, B., Tubbs, H., Small, J., Damoah, R., Thiaw, W., Cinnamond, A.R., Linthicum, K.J.
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01-03-2022 Elsevier
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Vector-borne and zoonotic pathogens comprise a substantial portion of the global disease burden causing ∼1.4 million deaths annually, and account for approximately 17% of the entire disease burden due to infectious diseases. Current Public Health and Department of Defense surveillance systems track individual infectious disease cases to report disease trends across populations, but these are retrospective and do not provide predictive information that could identify high-risk areas to better protect public health as well as deploying military personnel. We have built an integrated system which ingests (1) historical outbreak data from ProMED, PAHO and other sources; (2) global climate data sourced from NOAA and NASA including rainfall, land surface temperature etc; (3) Population data from the Socioeconomic Data and Applications Center (SEDAC) at Columbia University, and (4) Chikungunya vectors (Aedes aegypti and Aedes albopictus) location data from VectorMap (WRBU) and VectorBase (NIAID-BRC). We employ Machine Learning techniques to assess the relationship between locations of chikungunya outbreaks, climate variables and ancillary data. The Random Forested model was selected as the best performing model and used in deriving current and forecast risk maps globally. We derive both current and forecast risk maps at admin level 2 globally. Validation results for 2019, indicate that 80% of reported locations with chikungunya activity were predicted to be at risk by the current risk maps and ∼70 % of reported locations with chikungunya activity were predicted to be at risk by the forecast risk maps. This information can be visualized in CHIKRisk App at https://vbd.usra.edu/ and is updated on a monthly basis. Global Chikungunya Monitoring and Forecasting System (CHIKRisk App) presents progress towards monitoring and forecasting on vector-borne diseases by utilizing publicly available climate and outbreak data. This system is now being utilized by the Armed Forces Health Surveillance Division - Global Emerging Infections Surveillance Branch to inform Force Health Protection (FHP) decisions and by PAHO for public health surveillance. We hope to use this framework to build other next generation early warning systems for vector-borne diseases of global public health significance.
ISSN:	1201-9712 1878-3511
DOI:	10.1016/j.ijid.2021.12.228