Development and Performance of Text-Mining Algorithms to Extract Socioeconomic Status From De-Identified Electronic Health Records

Abstract

Socioeconomic status (SES) is a fundamental contributor to health, and a key factor underlying racial disparities in disease. However, SES data are rarely included in genetic studies due in part to the difficultly of collecting these data when studies were not originally designed for that purpose. The emergence of large clinic-based biobanks linked to electronic health records (EHRs) provides research access to large patient populations with longitudinal phenotype data captured in structured fields as billing codes, procedure codes, and prescriptions. SES data however, are often not explicitly recorded in structured fields, but rather recorded in the free text of clinical notes and communications. The content and completeness of these data vary widely by practitioner. To enable gene-environment studies that consider SES as an exposure, we sought to extract SES variables from racial/ethnic minority adult patients (n=9,977) in BioVU, the Vanderbilt University Medical Center biorepository linked to de-identified EHRs. We developed several measures of SES using information available within the de-identified EHR, including broad categories of occupation, education, insurance status, and homelessness. Two hundred patients were randomly selected for manual review to develop a set of seven algorithms for extracting SES information from de-identified EHRs. The algorithms consist of 15 categories of information, with 830 unique search terms. SES data extracted from manual review of 50 randomly selected records were compared to data produced by the algorithm, resulting in positive predictive values of 80.0% (education), 85.4% (occupation), 87.5% (unemployment), 63.6% (retirement), 23.1% (uninsured), 81.8% (Medicaid), and 33.3% (homelessness), suggesting some categories of SES data are easier to extract in this EHR than others. The SES data extraction approach developed here will enable future EHR-based genetic studies to integrate SES information into statistical analyses. Ultimately, incorporation of measures of SES into genetic studies will help elucidate the impact of the social environment on disease risk and outcomes.

Editorial: The Importance of Diversity in Precision Medicine Research

“…While this observation is widely accepted, few genetic association studies incorporate important measures of lifestyle, environmental exposures, or social determinants of health associated with disease risk and health disparities. Hollister et al address this challenge by applying their recently validated algorithm that defines socioeconomic status using electronic health records (Hollister et al, 2017 ) to a large clinical population of African American patients. All patients were clinically screened for hypertension, a complex condition disproportionately prevalent in African Americans (Fryar et al, 2017 ) that is independently associated with many common genetic variants and environmental exposures such as diet and socioeconomic status (Aburto et al, 2013 ; Giri et al, 2019 ; de las Fuentes et al, 2020 ; Glover et al, 2020 ; Hollister et al ).…”

Using Electronic Health Records To Generate Phenotypes For Research

“…Other EHR data collection trends attempt to address poor documentation of social determinants of health such as socioeconomic status(Hollister, Restrepo et al 2016, The National Academies of Sciences 2017). Zipcodes, census blocks, and geocoded addresses can be linked to various public repositories of community-level environmental data such as walkability maps, air pollution monitors, food desert maps(King and Clarke 2015, Pike, Trapl et al 2017, Xie, Greenblatt et al 2017).…”

Leveraging the learning health care model to improve equity in the age of genomic medicine

“…Social determinants of health―such as income, financial resource strain, education, and health literacy―are especially important in providing a social context for understanding a patient's clinical phenotype. Some EHRs capture such data as geolocation and employment status, as well as alcohol and smoking history, which may be used as proxies to measure the effects of certain social determinants on health and wellness . Income and education are strongly correlated with health and life expectancy .…”

“…68 As outlined in the report: as proxies to measure the effects of certain social determinants on health and wellness. 69 Income and education are strongly correlated with health and life expectancy. 70 More than the effect such factors may have on health directly, they may also impinge on patients' access to genomic and precision medicine and to new medical technologies.…”