Racial and Ethnic Differences in Psychiatry Resident Prescribing: a Quality Improvement Education Intervention to Address Health Equity.

Abstract

Quality improvement (QI) tools can identify and address health disparities. This paper describes the use of resident prescriber profiles in a novel QI curriculum to identify racial and ethnic differences in antidepressant and antipsychotic prescribing. The authors extracted medication orders written by 111 psychiatry residents over an 18-month period from an electronic medical record and reformatted these into 6133 unique patient encounters. Binomial logistic models adjusted for covariates assessed racial and ethnic differences in antipsychotic or antidepressant prescribing in both emergency and inpatient psychiatric encounters. A multinomial model adjusted for covariates then assessed racial and ethnic differences in primary diagnosis. Models also examined interactions between gender and race/ethnicity. Black (adjusted OR 0.66; 95% CI, 0.50-0.87; p < 0.01) and Latinx (adjusted OR, 0.65; 95% CI, 0.49-0.86; p < 0.01) patients had lower odds of receiving antidepressants relative to White patients despite diagnosis. Black and Latinx patients were no more likely to receive antipsychotics than White patients when adjusted for diagnosis. Black (adjusted OR 3.85; 95% CI, 2.9-5.2) and Latinx (adjusted OR 1.60; 95% CI, 1.1-2.3) patients were more likely to receive a psychosis than a depression diagnosis when compared to White patients. Gender interactions with race/ethnicity did not significantly change results. Our findings suggest that racial/ethnic differences in antidepressant prescription likely result from alternatively higher diagnosis of psychotic disorders and prescription of antipsychotics in Black and Latinx patients. Prescriber profiles can serve as a powerful tool to promote resident QI learning around the effects of structural racism on clinical care.