Decision support models for public health informatics

Ith an increasing concern over emerging infectious diseases, efficient and reliable public health monitoring is critical. The prototype models described in this article were built to aid public health officials in monitoring the health of their communities by increasing situational awareness and reducing false-positive identification of disease outbreaks. This comprehensive capability is needed to bolster public health acceptance of biosurveillance systems by making the complex information environment more manageable and by achieving performance that is more robust. The models introduced in this article were built to recognize and differentiate influenza outbreaks from the other seasonal respiratory activities. The models were tested with historical data collected by the Electronic Surveillance System for the Early Notification of Community-based Epidemics (ESSENCE) in the National Capital Region. Results show significant improvement in both the sensitivity and specificity of the detections compared with the ESSENCE algorithms.