PM2096: The Use of Stigmatizing Language in Electronic Medical Record of Critically Ill Children

Background: Disparities in health outcomes may result in part from the attitudes and behaviors of health care providers. Black patients are more likely to be blamed for their health outcomes, described as less personally responsible for their health agency, and less likely to receive appropriate treatment when compared to white patients.. Similar data has been collected with respect to provider perceptions of patients labeled as "difficult" or “angry”. A simulation of case scenarios using stigmatizing language for sickle cell patients in pain crisis was directly associated with treatment decisions which deviated from standard protocols. Little is known about stigmatizing language ad its impact in pediatric patients.

Methods: We performed a retrospective cross-sectional study of patients admitted to the pediatric intensive care unit at a large, urban, quaternary care center who had been referred to palliative care in 2019. Demographic data were collected including patient age, sex, race, ethnicity, insurance. Number of problems served as a proxy for medical complexity. Sixty-six charts were reviewed and instances of negative and positive stigmatizing language were counted. The type of provider writing the note was recorded. Odds ratios were calcutated to determine if patients were more likely to be labeled with negative or positive stigmatizing language depending on their race, ethnicity, insurance status, and illness severity.

Results: Twenty-nine charts contained negative stigmatizing language. 69% were white, 21% black, and 10% other. 34% of those labeled negatively were hispanic. 69% were on Medicaid. Physicians wrote 80% of the negative labels in the EMR with 10% by social workers and 3% by nurses.

Sixty-five charts contained at least one positive patient label., 71% were white, 17% black, and 12% other. 38% were hispanic. 62% were on Medicaid. Doctors wrote 34% of the positive language, 28% by social workers, and 14% by nurses.

Odds ratios are pending at the time of submission. Analyses of patient descriptors as a function of medical complexity is pending at the time of submission.

Conclusion: Preliminary results from our small, retrospective study seem to indicate that there is no difference in the use of negative or positive stigmatizing language between racial or ethnic groups or between payer groups. There does seem to be a difference in who is using the stigmatizing language, with physicians overwhelmingly being the ones who use negative patient descriptors but the use of positive patient descriptors is much more balanced between physicians, social workers, and nurses. These results indicate that further studies may need to be completed with a larger sample size to detect differences in the use of stigmatizing language between races, ethnic groups, and patients from different socio-economic backgrounds. These data could be use to support targeted implicit bias training in social workers and nurses as well as physicians.