Health

How to identify and prevent systematic error in electronic health records

Then, using machine learning, they analyzed more than 40,000 case histories from 33,000 patient visits from the University of Chicago Academic Medical Center between January 2019 and October 2020. Of the 18,500 patients included in the study, approximately 61% were black, 30% were white, 6% were Hispanic or Latino, and 3.5% were ‘other’. A total of 8.2% of patients had a history of one or more negative descriptors.

Fifteen common patient descriptors were used to pinpoint stigmatizing language, including noncommittal, aggressive, agitated, angry, defiant, combative, uncompromising, confrontational, noncooperative, defensive, exaggerated, hysterical, unpleasant, rejection, and resistance. Sun said that the generality of these terms in electronic health records reflects the cultural norm of misrepresenting barriers to patient health.

“There is a pattern of words that we use that are labels and we are doing a disservice to our patients by not giving them the full context, their full story,” he said.

For example, a doctor may call a patient “naughty” when the patient really lacks health literacy and misunderstands what he should be doing. Understanding this difference can help a patient return to a treatment plan that works for them and improves outcomes.

According to Sun, the next step in the study will be to study the relationship of negative comments in a patient’s electronic health record to clinical outcomes. The report does not directly link poor medical outcomes as a consequence of implicit bias, but notes other studies, including a study that found doctors with high rates of implicit bias were more verbally dominant with black patients, and a report indicating that bias in health care is associated with lower levels of patient adherence.

Download the Modern Healthcare app to keep up to date with industry news.

The report also explains how electronic health records can perpetuate bias and stigma among physicians. The authors cited a 2018 study that found that healthcare professionals were more likely to perceive a patient’s pain negatively when presented with a chart with annotations containing stigmatizing language such as “frequent flyer”.

“It wouldn’t be hard to imagine the different types of interactions they could have,” he said. “This will certainly be an additional area of ​​research for us, but we expect these descriptors to have some impact on the doctor-patient relationship, as well as many of the provider-patient relationships that will occur during patient care. hospital stay.”

The researchers found that the use of stigmatizing language has decreased in 2020. Sun said that since the onset of the COVID-19 pandemic and amid a nationwide reckoning for the murder of Georgie Floyd, clinicians were less likely to use a negative descriptor in an electronic health record. He said the results show the ability of clinicians to test their biases and be swayed by the use of negative descriptors in their charts, especially when describing a patient with skin color or a marginalized identity. It may also reflect the growing interest of service providers in addressing cultural incompetence in their operations.

“At first it surprised us because we thought that the pandemic as a whole, as a stressful environment, would force people to use more cognitive labels or stereotypes, relying on prejudice or using prejudice a little more. that it has really decreased during the pandemic,” he said. “I hope people think of this as an opportunity to tell the patient the full story and provide more compassionate and empathetic care. It’s certainly within our power, just need a little more intentions.”


Source link

Leave a Reply

Your email address will not be published.

Back to top button