Building on existing evidence showing that implicit bias among clinicians can result in disrespectful attitudes toward patients based on their race and ethnicity, this project will (1) examine language used by doctors and nurses in medical records that may convey disrespectful attitudes towards patients, (2) attempt to develop algorithms using natural language processing (NLP) to efficiently detect such language, (3) evaluate the impact that stigmatizing language has on the quality of care as reflected in clinician attitudes and clinical decision-making, and (4) make recommendations to guide medical record documentation and health system strategies to reduce stigmatizing language. The goal of this project is to improve respect, and thus the quality and equity of services, for persons who are stigmatized in health care.