Language Reveals Implicit Bias In Healthcare

Clinicians depend on patient characteristics, such as age, sex, and symptoms, to construct an accurate and compelling clinical picture. However, there exists a regrettable tendency to employ derogatory expressions when describing patients.

Terms like “poor historian,” “noncompliant,” or “no-show” are frequently utilized in clinical medicine, perpetuating and reinforcing negative stereotypes. Recent research has unveiled the surprisingly widespread use of stigmatizing language across various clinical settings, including outpatient clinics, emergency rooms, and inpatient hospital stays.

This prompts an inquiry into the prevalence of stigmatizing language in home health care, an increasingly popular form of outpatient care. Furthermore, how does the use of such language impact patient care?

Use of language

To shed light on this matter, researchers from the University of Pennsylvania’s Leonard David Institute of Health Economics conducted a novel study utilizing machine learning techniques to identify patterns of “judgment language” in the notes of urban home health care clinicians.

By leveraging data collected by a private company, the authors examined clinicians’ perceptions of a patient’s reliability, as reflected in phrases such as the patient “states,” “claims,” or “admits.”

This pioneering study marks the first of its kind to explore the extent of stigmatizing language in the realm of home health care, an outpatient setting experiencing rapid growth and catering to the needs of over 5 million Americans each year.

Within a cohort of 45,384 patients, the researchers meticulously scrutinized over 260,000 patient notes. Applying their algorithmic approach, they discovered that 10% of all notes contained judgment language.

More notably, this type of language was found to be most prevalent among Hispanic and Black patients, followed by white patients, and then Asian patients. In fact, Black and Hispanic patients were 14% more likely to encounter judgment language in their notes compared to their white counterparts.

Affecting care

Importantly, the researchers discovered that the length of a home health care visit was reduced by 21 minutes when judgment language was used.

“This is concerning,” they write, “since shorter home health care visits are associated with a higher risk for poor outcomes,” such as a higher risk of hospitalizations.

The study sheds light on a concerning correlation between negative language and a patient’s race and ethnicity. Robust evidence has revealed that Black patients, across a range of medical settings, are 25-50% more likely to encounter stigmatizing language in their electronic health records compared to their white counterparts.

Furthermore, similar stigmatizing words have been commonly associated with patients grappling with substance use disorders and chronic conditions, such as diabetes.

Exploratory investigations have begun to delve into the relationship between stigmatizing language and the quality of care provided. Notably, one study uncovered a noteworthy correlation between clinicians exposed to such language and their tendency to administer less aggressive pain management for patients.

Implicit biases

The study effectively demonstrates the pervasiveness of implicit racial biases prevailing among healthcare professionals and their tangible influence on clinical practices. Consequently, the authors have proffered four policy recommendations to mitigate clinician bias:

Develop more inclusive and impartial guidelines for clinical documentation, with particular emphasis on eliminating or providing justifications for the usage of certain judgmental terms or expressions in patient notes.

  • Implement targeted training programs for clinicians who frequently employ stigmatizing language in their clinical encounter notes.
  • Offer counseling and educational interventions to clinicians aimed at reducing the overall use of stigmatizing language.
  • Provide training for clinicians to allocate more constructive time to patients who struggle to adhere to instructions or exhibit resistance to self-management.

These proposed solutions hold the potential to mitigate clinician bias and contribute to a more equitable healthcare system.

Facebooktwitterredditpinterestlinkedinmail