EHR Notes Reveal Clinical Documentation Implicit Bias Concerns

Black patients were 2.54 times more likely to have at least one negative descriptor in their EHR notes compared to White patients, according to a study published in Health Affairs that raises concerns about stigmatizing language in clinical documentation and its potential to worsen racial health disparities.

Researchers analyzed a sample of history and physical notes from 18,459 patients for sentences containing a negative descriptor of the patient or the patient’s behavior.

Negative descriptors included “resistant” or “noncompliant.” The study authors said that while these and similar descriptors are not explicitly stigmatizing terms, that verbiage may convey a negative connotation in the context of describing a patient.

The research team noted that their findings are especially alarming because their analysis was limited to the history and physical notes of patient EHRs, which are often drawn upon and used by other providers.

“Negative descriptors written in the admission history and physical may be likely to be copied into subsequent notes, recommunicating and amplifying potential biases,” the researchers explained. “This practice underscores the responsibility of providers who document the initial patient encounter to do so in an aware and sensitive manner.”

“Subsequent providers may read, be affected by, and perpetuate the negative descriptors, reinforcing stigma to other healthcare teams,” they wrote.

Additionally, contrary to expectations, the researchers found that EHR notes written after COVID-19 began were less likely to have a negative descriptor.

The onset of COVID-19 coincided with a national response to racialized state violence prompted by the police murders of George Floyd and others. COVID-19 infection and mortality rates also highlighted racial health disparities.

“These social pressures may have sensitized providers to racism and increased empathy for the experiences of racially minoritized communities,” the study authors wrote.

They emphasized that future research is needed to investigate the longitudinal consequences of a negative descriptor in a patient’s EHR notes.

“Additional investigation may use a similar machine learning approach to examine EHR data over a longer period of time for repeated use of negative descriptors and for potential effects on health outcomes,” the researchers noted.

The study authors said the research has several implications for policy. First, they said that medical institutions need to better address the introduction of implicit bias of all forms, but most significantly racial bias.

To address implicit bias is to address the underlying mechanisms that prompt the use of negative descriptors to describe patients, they explained.

“This includes preventing the introduction of biased language by providers, preventing the perpetuation of biased language by members of the healthcare team, and increasing awareness of the effects of providers’ language on the patient relationship,” the authors wrote.

Interventions such as provider bias training can help educate clinicians on the use of nonstigmatizing language in clinical documentation.

“Better education on race and racism may help equip providers with the understanding needed to identify, prevent introduction of, and discontinue use of negative descriptors in the EHR,” the researchers noted.

However, as more institutions begin to share EHR data, a broader shift may be necessary to catalyze evolution in the language of healthcare, they suggested.

Regulatory bodies, like the Accreditation Council for Graduate Medical Education and its counterparts, maintain training standards for professional communication, nondiscrimination, and internal biases.

While these guidelines affirm the importance of nonstigmatizing, patient-centered language, specific recommendations may be necessary to align professional standards with practices, the au