For real time surveillance, detection of abnormal disease patterns is based on a difference between patterns observed, and those predicted by models of historical data. The usefulness of outbreak detection strategies depends on their specificity; the false alarm rate affects the interpretation of alarms.
