Study finds bias against Black patients written into medical charts

Black patients were more than two-and-a-half times as likely as white patients to have negative descriptors about them in their electronic health record, according to a study published Wednesday in Health Affairs.

Why it matters: The study is further evidence of bias in the U.S. health care system, which can ultimately result in worse care and disparately poor outcomes.

Details: University of Chicago researchers used machine learning to analyze more than 40,000 notes in the EHRs of more than 18,000 patients at an urban academic medical center.

• The study looked for sentences that included negative descriptors such as "resistant," "challenging" or "noncompliant."
• Even when controlling for individuals' sociodemographic and health characteristics, Black patients were 2.54 times as likely to have a negative descriptor than white patients.

What they're saying: "Negative descriptors written in the admission history and physical [notes] may be likely to be copied into subsequent notes, recommunicating and amplifying potential biases," the authors wrote.

• "This practice underscores the responsibility of providers who document the initial patient encounter to do so in an aware and sensitive manner," they wrote.