Subjectively-Assigned versus self-reported race and ethnicity in US healthcare

Abstract

Documenting patient “race” descriptors in clinical medicine, epidemiology, and public health data and analysis has been routine in the US. However, patient race has historically been and is still most often subjectively-assigned rather than self-identified. Even when self-identification is allowed, persons must often self-deny parts of their ancestry by adhering to restrictive race categories. In contrast, most other countries ignore so-called race and may use other ancestral background information including family and geographical histories, language(s) and/or ethnic group(s) membership. We performed two studies involving 160 patients to investigate subjectively-assigned versus self-reported race using a verbal questionnaire in a New Orleans medical clinic. Results revealed that the subjectively-assigned race recorded by the hospital administration/physician was incomplete and therefore inaccurate. Clinicians and researchers must make more accurate and respectful ancestral inquiries in order to derive useful information about individual and population health risks and disease conditions, while also being mindful of potentially erroneous race data previously gathered and conclusions inferred in healthcare literature.