Study analyzes racial bias in EHRs

An analysis of electronic health records (EHRs) from an urban academic medical center found negative descriptors such as “resistant” or “noncompliant” appeared two and a half times more often in the records of Black than white patients, according to an article in the February issue of Health Affairs. This study used machine learning to analyze 40,113 history and physical notes collected on 18,459 patients between January 2019 and October 2020. The researchers said their findings “raised concerns about stigmatizing language in the EHR and its potential to exacerbate racial and ethnic health care disparities.” The February issue is dedicated to racism and health.