Public health practice uses a wide variety of data types, data sources, and data management techniques. While the data necessary for many public health processes can be provided solely by data generated during routine clinical care, supplemental data and improved information extraction techniques that could better inform public health processes are either inconsistently present or are typically absent from clinical systems. For example, while clinical transactions serve immediate patient care needs, they are often incomplete and lack specific patient, provider, or clinical information necessary for informed public health decisions.

Many public health processes need data not only from clinical systems, but also require nonclinical information to more accurately identify and characterize public health trends and events in order to identify clinical and nonclinical correlates, and to predict future public health outcomes. Nonclinical information can include a patient’s geospatial location, socioeconomic status, school affiliation, and proximity to risk factors such as elevated soil lead levels within a community.

To fully inform public health processes and to improve public health outcomes, clinical data must be augmented with additional, nonclinical data sources. Additionally, clinical systems often lack sophisticated information extraction techniques and case detection algorithms needed to identify clinical data needed for public health processes. These techniques and strategies may include natural language processing, rules engines, and machine learning algorithms; these techniques can substantially improve case identification. Finally, because clinical and nonclinical data are often stored in separate databases as separate islands of information, public health often lacks efficient access to integrated population-level health data, which hinders the ability to identify and manage the specific public health needs of a community. Efficient data integration strategies are needed.

We will design and implement informatics solutions to address these shortcomings by leveraging one of the Nation’s most technologically sophisticated, standards-based, comprehensive and longest-tenured health information exchanges combined with one of the Nation’s leading spatially enabled community information systems. The long-term goal of our efforts is to improve the overall community health by informing and improving public health practice through innovative standards-based public health informatics initiatives. Our objective in this project is to demonstrate the technical feasibility and value of enhancing data routinely captured in a health information exchange using additional data sources, novel information extraction techniques and data integration strategies. Our hypothesis is that enhancing the information infrastructure is both technically feasible and will add substantial value to public health practice.