- Stigmatizing language used in patient EHRs can potentially influence the attitudes of physicians-in-training with regard to patients, possibly impacting physicians’ medication prescribing behavior, according to research published in the Journal of General Internal Medicine.
This can also cause clinician bias to be propagated from one clinician to the next, wrote Goddu et al. Providers should pay attention to the language used in patient EHRs to ensure that patient-centered care is promoted and healthcare disparities for stigmatized populations are reduced.
“Clinicians may acquire implicit bias towards patients from one another when communicating verbally or when writing or reading medical records; physicians-in-training may absorb these attitudes as part of the ‘hidden curriculum’ of medical training,” researchers explained.
“We hypothesized that resident physicians and medical students who read a chart note using stigmatizing language to describe a patient would have more negative attitudes about the patient and would treat the patient’s pain less aggressively than those who read a chart note using neutral language,” the team continued.
The researchers randomly read chart notes from residents and medical students, which presented medically identical information about a hypothetical patient with sickle cell disease (SCD). One chart had stigmatizing language and the other had neutral language.
“After reading the chart note, participants completed a survey assessing their attitudes regarding the patient (residents and students) and treatment decisions (residents only) for the patient,” the team explained. “We focused on a patient with SCD, as there is significant evidence that these patients are stigmatized by clinicians and experience inadequate pain management, conflicts with staff, and lack of respect.”
The medical students and residents were in emergency medicine and internal medicine because they often care for patients with SCD who present with pain crises.
Researchers added that they used a vignette study design for the hypothetical charts because it is “a robust and efficient method to systematically assess variation in opinion and decision-making, without confounding by other patient characteristics.”
The vignettes described Mr. R, a 28-year-old man with SCD. The patient’s chief complaint, history of present illness, and physical exam findings were then described in two different ways.
“Stigmatizing language in patient charts was characterized by three linguistic features, as elaborated in prior work: casting doubt on the patient’s pain (e.g. insisting that his pain is ‘still a 10’ vs. still has 10/10 pain), portraying the patient negatively (with irrelevant or unnecessary indicators of lower socioeconomic status such as hanging out with friends outside McDonald’s), and implying patient responsibility with references to uncooperativeness (e.g. he refuses his oxygen mask vs. he is not tolerating the oxygen mask),” the research team said.
Eighty-eight percent of participants who read the neutral language chart note thought the physician had neutral or positive attitudes towards the patient, while 45 percent who read the stigmatizing chart note thought the same.
Researchers also found that exposure to the stigmatizing language chart note negatively influenced resident and medical student attitudes towards the hypothetical patient.
“Resident physicians who read the stigmatizing language chart note prescribed pain medication less aggressively than those who read the neutral language chart note,” the report said.
The comfort in dosing pain medication correlated with more aggressive pain management for respondents who were exposed to the neutral language chart note. That same correlation was not present in respondents who read the stigmatizing language chart note.
“Our results suggest that language used in medical records to describe patients can directly influence subsequent physicians-in-training who read the notes, in terms of both their attitudes towards the patient and their medication-prescribing behavior,” the team wrote. “This is an important and overlooked pathway by which bias can be propagated from one clinician to another, further impairing healthcare quality for the individual patient as well as entrenching healthcare disparities overall for those who are stigmatized.”
Moving forward, it is important to consider what should be documented in the patient EHR, researchers recommended. Physicians should be taught to not perpetuate bias, but some of the information that was described in stigmatizing notes should also be shared with other members of the healthcare team.
“Physicians-in-training may feel both positive and negative emotions about their patients, but value judgments informed by feelings about a patient should not appear in the medical record,” the team stated. “Clinicians have a responsibility to record data that are relevant and as objective as possible, as well as to recognize that using certain phrases (e.g. substance abuse vs. substance use) opens the possibility for other clinicians to make or interpret judgment.”
Researchers acknowledged that the study results may not be generalizable to other healthcare settings. For example, the data collection was at a single center located at an urban medical center with a large black patient population.
The vignette study also had a defined set of answer choices specific to assess pain management attitudes.
However, the stigmatizing language used was abstracted from real patient medical records.
The research team noted that a medical record may be the only source of information for a new clinician, which is why having an objective record can be so critical.
“Clinicians must be vigilant to guard against contributing to bias as they write chart notes about their own patients and as they read chart notes written by others,” the study concluded. “The language in medical records should be more carefully considered to avoid perpetuating clinician biases and the healthcare disparities that may arise from them.”