A new study published in JMIR Medical Informatics has uncovered concerning evidence of racial bias in how doctors describe patients in electronic health records. Researchers analyzed records from 281 physicians and found that doctors were more likely to use negative language when writing about Black and Hispanic patients compared to white patients.
The study revealed that summaries for Black and Hispanic patients contained significantly more negative adjectives like “unkind” or “stupid,” as well as more words related to fear and disgust. Notes for Hispanic patients also included fewer positive descriptors and words expressing trust or joy.
“Previous studies have shown that care providers’ biases may be part of the reason for racial disparities in health,” said Eden King, professor at Rice University and lead author. “We wanted to know whether we could detect such biases in the language providers use in health records, and we did.”
These linguistic differences could potentially result in biased treatment and unequal care for minority patients. The researchers hope their work will help identify and mitigate bias in medical interactions.
“Understanding that providers’ language may indicate bias points to an opportunity to interrupt it,” King stated. “If we can perfect algorithms to detect such bias, we can raise awareness in the moment of the patient–provider conversation. That awareness may be enough to encourage more equitable health care.”
The findings underscore the need to address unconscious biases in healthcare to reduce racial and ethnic health disparities.
