The EHealth team is currently partnering with the University of North Carolina at Chapel Hill (UNC). This partnership involves linking electronic health records (EHRs) to restricted Census microdata. EHRs have the potential to shed light on disparities in health with their up-to-date and detailed information on patient visits, lab reports, diagnoses, and medications for a large sample size. However, there are some limitations, including missing race and ethnicity, limited social characteristics, and questionable population representation.

To overcome these limitations, this project aims to strategically repurpose already available data (EHRs and Census microdata) to provide a unique and innovative way in which to study population health. To utilize the potential of EHRs more fully, this partnership has two overarching aims. First, we aim to evaluate the representativeness of EHRs and correct for potential bias in EHR-based studies. Second, we aim to expand the utility of EHRs through the creation of a unique integrated dataset that expands information on social determinants of health that are missing from clinical databases.

Our most recent collaboration was a feasibility study assessing the linkage of EHRs with Census microdata. This work was recently published in the American Journal of Public Health (see link below). Moving forward, we plan to link EHR data to Census Bureau data to further evaluate the opportunities an integrated dataset would bring to both parties.

Through this collaboration, we demonstrate that linking EHRs and Census microdata has the potential to provide a comprehensive view of population health as well as a novel examination of social determinants of health and health disparities that is not possible in either source alone.

Concordance Rates in Racial and Ethnic Identification: Insights Based on Linked Electronic Health Records and the American Community Survey

April 2024
Authors: Rocio Rosa-Lebron, Aubrey Limburg, Timothy S. Carey, Victoria Udalova, and Barbara Entwisle

Addressing Health Disparities by Integrating Electronic Health Records with Census Data: First Steps

September 2022
Author: Aubrey Limburg, Joanna Motro, Lucinda Dalzell, and Victoria Udalova, Barbara Entwisle, Jordan Young, Tim Carey, and Paul Chelminski