A risk prediction model for HPAI outbreaks in Kuwait

Area of Science:

• Veterinary Epidemiology
• Disease Modeling
• Public Health Surveillance

Background:

• Highly pathogenic avian influenza (HPAI) poses significant risks to animal and human health.
• Effective disease control relies on accurate risk assessment and prediction modeling for timely interventions.

Purpose of the Study:

• To develop a predictive model for HPAI outbreaks in Kuwait.
• To identify high-risk areas for HPAI occurrence.

Main Methods:

• Integrated 16 years of HPAI outbreak data (2005-2020) with meteorological and wild bird nest proximity data.
• Employed logistic regression and machine learning algorithms for outbreak prediction.
• Utilized confirmed outbreak locations for model training and validation.

Main Results:

• The predictive model achieved a balanced accuracy of 0.79 and ROC AUC of 0.83.
• Kuwait metropolitan City and the coastline were identified as the most vulnerable areas for HPAI outbreaks.
• Spatially explicit risk prediction was achieved.

Conclusions:

• The study provides a foundation for developing spatially targeted HPAI control strategies in Kuwait.
• Risk prediction and mapping enhance early response and surveillance prioritization during high-risk periods.
• Informed public health and veterinary authorities on HPAI risk areas.