Duke Medicine Initiative Integrates EHR, Geographic Data
July 19, 2014 in News
Duke Medicine has announced an initiative to integrate its electronic health record data with its geographic information system to enable clinicians to predict patients’ diagnoses, Health Data Management reports.
Duke Medicine includes Duke University’s School of Medicine, the School of Nursing and Health System.
Details of the Initiative
The integrated system will allow clinicians to use EHR and geographical data to predict in real time which people within a given population are likely to be diagnosed with certain ailments.
The system works by:
- Using an automated geocoding system to verify addresses with a U.S. Postal Service database;
- Passing the addresses through a commercial mapping database to geocode them; and
- Importing all U.S. Census Bureau data with a block group ID.
The result is an environmental assessment of socioeconomic indicators to individual patients or groups of patients.
According to Health Data Management, Duke is working to develop a proof of concept and algorithms for map locations and patient proximities. In addition, researchers are developing on-demand models to be used for tracking food-borne illness outbreaks, leading patients to the most cost-effective treatment facilities and other purposes.
Sohayla Pruitt, a senior geospatial scientist at Duke and head of the initiative, called the automated system “a way of taking the bias out of traditional analysis,” by letting “statistical models tell you what’s going on and where the true correlations exist.”
Pruitt said, “[O]nce you know how geography is influencing events and what they have in common, you can project that to other places where you should be paying attention because they have similar probability” (Ericson, Health Data Management, 7/16).