Introduction Highly pathogenic avian influenza (HPAI) outbreaks pose significant threats to animal and human health. Risk assessment and prediction modelling are essential for improving disease control and enabling timely intervention strategies. Methods HPAI outbreak data collected over 16 years (2005–2020) were integrated with meteorological data, wild bird nest proximity, and confirmed outbreak locations. Logistic regression and machine learning tools were used to predict HPAI outbreaks. Results The model demonstrated strong predictive performance, achieving a balanced accuracy score of 0.79 and a ROC AUC of 0.83. The study identified Kuwait metropolitan City and the coastline as the most vulnerable locations for HPAI outbreaks. Discussion This study provides a foundation for developing spatially targeted HPAI control strategies in Kuwait. Risk prediction and mapping can support early response efforts and surveillance prioritization during high-risk periods.
Alshatti et al. (2026) studied this question.