The broad project objective is to establish a robust technical link in the process of automated outbreakdetection to complement and backup the traditional sentinel surveillance system. Means to this objectiveinclude development and efficient combination of data-driven statistical alerting algorithms andimplementation of higher level decision-support tools for fusing algorithmic results with external informationand epidemiologist judgment.
Specific aims are to establish and exercise a context-sensitive testbed forstandardized algorithm development and evaluation, to develop an information-sharing methodology forjurisdictional situations that preclude data-sharing, and to create decision-support tools by combiningheuristic methods used by experience health monitors with Bayesian Belief Net representation. Researchwithin the testbed will advance the state of the art in detection algorithms, stressing the adaptations requiredto make them relevant and effective for monitoring on a daily or, depending on input data rates, a near-realtimebasis. Algorithms will broadly include univariate, multivariate hypothesis tests and data miningtechniques more general automated learning and we will seek to determine the appropriate niche for eachapproach found useful. For univariate data, hypothesis test research will investigate means of combiningdata modeling and process control for optimal detection performance depending on the data background.Multivariate algorithm research will include both fully multivariate methods and multiple univariate methodsand will blend the two for optimal monitoring capability. Structured testbed design and development willestablish standards for algorithm evaluation and comparison based on health monitoring effectiveness,measured by sensitivity, specificity, and timeliness of event detection. These standards will make this publichealth research area more accessible and relevant to local heath department users. Potentially, thedecision-support tools may make complex data scenarios understandable to users of these systems.Acceptance of potential of automated surveillance systems will stimulate increased cooperation of healthdepartments, data providers, and policymakers, further improving public health monitoring capability.
