IUPUI Researchers to Study Feasibility of Using EHR Data for Public Health Planning
Indiana University-Purdue University Indianapolis (IUPUI) researchers have received a $200,000 grant to study the feasibility of using electronic health records (EHRs) to measure health outcomes of Marion County residents by neighborhood or even by census block for public health planning.
Current information on the health of people living in Marion County available to public health officials is based on samples representing county-wide data, Brian Dixon, Ph.D., an assistant professor of health informatics in the IU School of Informatics and Computing at IUPUI, said.
But the increasing availability and use of EHRs may permit public health officials to peer within the county at smaller and smaller groups of residents, he said. If that’s possible, public health officials could more readily determine which areas of the county are most in need of certain public health programs.
"For several years we have worked on connecting data on community-based indicators gathered by the Polis Center with data in our local health information exchange and geocoding it to look at some of the social determinants of health," Dixon explained. "So once we had this infrastructure in place, we started looking at what other types of questions we could ask of it."
"For instance, there is no mandate that providers share data on diabetic patients with public health departments," Dixon said. But if public health officials can see how many people in specific areas are diabetic, they can better focus their energy and improve screenings and outreach."
Researchers will look at aggregate data taken from the health information exchange for finite geographical regions within the county to measure the health of a particular neighborhood or set of neighborhoods and then use that information to identify where certain public health programs or interventions are most needed, he said.
Researchers will analyze the electronic records of smaller and smaller groups of county residents, including by ZIP code (average population of 8,000), neighborhoods (average population of 3,000 to 4,500) and census block (average population of 1,500), the smallest population grouping to be examined.
The first test is to come up health measurements at a sub-county level for things like how many people in a particular area receive flu shots, Dixon said. Then Dixon and other researchers will be able to take that data and link it to socio-economic factors, enabling them to tease out what determinants make one area healthier than another.
"One of the lessons learned from this study is going to be how fine-grained we can get with the geocoding as well as the limitations of the data sources we are looking at," Dixon said.
The project is one of 11 that received research awards, totaling $2.7 million, facilitated by the National Network of Public Health Institutes, with guidance from the National Coordinating Center for Public Health Services and Systems Research housed at the University of Kentucky’s College of Public Health. Support for this research was provided by a grant from the Robert Wood Johnson Foundation.
"There are community information systems just like the Polis Center in 35 other regions," Dixon noted. "It will be quite possible for them to work with emerging HIEs in their regions to emulate what we are doing here."
