Information Architecture

Abstract

Public health information today is spread across many programs, mostwith their own method for identifying people and related data elements,a program-level data model, nonstandard vocabulary, program specificvertical workflow, and customized reports. A single office may usemultiple processes to manage workflow information, including paper, fax,spreadsheets, electronic documents, and proprietary databases. If neededdata resides in another database silo, the process may include a custominterface that then must be maintained as another step in the process tocapture and manage public health information.These disparate data silos translate to fragmented and redundant publichealth data and workflow and the inability to present a complete pictureof an individual or population, further distancing care coordination andthe achievement of best outcomes. An effective information architecturemust address the issue of interoperability between data silos and serveas a guide for transitioning to shared data and optimized workflow. Ashared data model using standards-based metadata tags and attributes isthe best option for public health to begin this transition.Adopting a shared metadata model enables exchange of information withother standards-based systems, such as electronic health records,without redundant data entry. It enables care to be coordinated acrossprograms and agencies according to best practice evidence and reinforcedwith alerts and reminders to individuals and providers. A sharedmetadata model is extensible to other disciplines in the public sector,such as education and child welfare, by populating the model withcontent from those domains. Standard terminology practices arefacilitated and the model becomes scalable to large populations. Givenpublic health funding and workload realities, moving toward a sharedmetadata model will occur in a phased approach with programs joining theprocess over time, and according to priority assigned by a governancebody of stakeholders.