As we discuss in Parts II and IV, a system designer has a large selection of surveillance data from which to construct a biosurveillance system. This abundance is a result of the increasing amounts of data collected electronically about the health of individuals and their purchasing, travel, attendance, and other behaviors (Sweeney, 2001).
For a designer, the task of data selection may be simple or complex. It is simple if the organization planning a system requests that the designer automate the collection of data that the organization is already receiving. It may be quite challenging if the organization specifies an engineering goal of earlier detection and leaves it to the designer to select data. In part, the designer's difficulty stems from the current lack of full understanding of the value of many of these types of data. The designer must weigh whether data can meet functional requirements for timeliness and accuracy of detection, the availability of the data, and the cost and efforts to acquire the data. Cost and effort may be dominant factors for data that either require effort on the part of many individuals or organizations to obtain the data, or data that are distributed among many computer systems, especially if the biosurveillance system will monitor at the national or international level. The chapters in Parts II and IV of this book provide detail about data availability and the cost and effort to obtain different types of data.
The complexity of the design process is further increased by the amount of data that must be specified, especially for a system that supports outbreak characterization. The data may include data about human health, animal health, environmental data, locations, and relationships of people. Just the data potentially needed to diagnosis a single sick human are staggeringly large-there are more than 4,000 symptoms and laboratory tests that bear on the diagnosis of disease in humans.
Appendix A provides tables that may be useful to system designers. This appendix contains tables of surveillance data that have been pivotal in detecting and characterizing outbreaks in the past, data that are required to satisfy published case definitions, and data currently being collected routinely for biosurveillance by governmental public health. The tables also include types of data that may be very early indicators of diseases, identified by a review of the literature on health psychology, especially the subliterature relevant to behaviors of ill individuals between the onset of symptoms and presentation (if ever) for medical care. The tables also identify data systems that routinely collect the data either for the purposes of biosurveillance or, more typically, for other purposes.
Was this article helpful?