Is A Nationwide Health-Data Network Possible?

The elements are in place, but challenges remain, says the co-director of the Biomedical Security Institute.
InformationWeek senior writer Marianne Kolbasuk McGee spoke with Dr. Michael Wagner, director of the Real-Time Outbreak and Disease Surveillance Systems (RODS) Laboratory at the University of Pittsburgh and co-director of the Biomedical Security Institute, about the potential for building a nationwide health-data network that could be used to provide health officials with early alerts about disease outbreaks.

InformationWeek: Will the RODS laboratory serve as the model for a nationwide health-data network?

Wagner: The concept of a nationwide health-data network is an interesting one because there are already elements of it in place in the form of data networks built by health system and public health [departments] in various regions that run over the Internet or over leased lines. RODS itself isn't a model for a national health-data network but rather an example of a public-health application that's running on top of such a network. In our large-scale demonstration projects in Utah and Pennsylvania, we've utilized both the Internet and leased lines.

InformationWeek: RODS was implemented in Utah (in time for the Olympics) and in Western Pennsylvania. What sort of work needs to be done by other states (such as public-health departments, as well as local hospitals, physicians' offices, and pharmacies) to build their own RODS?

Wagner: To implement a system like RODS requires (1) that a set of parties enter into data-sharing agreements of one form or another under the sponsorship of a public-health authority, and (2) a certain amount of technical work at the network, data, and application levels.

InformationWeek: How long would it take to build a nationwide RODS?

Wagner: That's a difficult question to answer without first some qualifications. If the goal is collection of part--say 50%--of the full data and to do it for a relatively small number of variables such as chief complaints and microbiology cultures, it could be done relatively quickly. The technical part of the Utah deployment, which covers a minimal data set with approximately 70% coverage in the state, was completed in several weeks. Utah had a high level of commitment and technical sophistication at the participating health systems, but we have found similar levels of commitment and sophistication in the six health systems that we have worked with in Pennsylvania.

InformationWeek: What are the goals for the Biomedical Security Institute?

Wagner: The mission of the institute is to improve the nation's capacity to detect and respond to outbreaks of disease, whether they be naturally occurring or terrorist.

InformationWeek: What other projects will the institute work on?

Wagner: Researchers at the Biomedical Security Institute are working on basic and applied research in simulation, automatic detection of outbreaks, economic modeling, decision making, mental health, and data modeling.

InformationWeek: How are institute projects being divided between the University of Pittsburgh and Carnegie Mellon?

Wagner: The institute is a researcher-focused organization, which means that researchers develop ideas, seek funding, and conduct the work. The projects divide among the campuses based on researcher effort.

InformationWeek: How many people are involved with the Biomedical Security Institute at each university?

Wagner: The universities are highly committed to the institute concept so there's administrative participation up to the Senior Vice Chancellor-Provost level as well as researcher interest. The total number of involved individuals probably exceeds 100.

Close this window