Observed Versus Expected Distribution of Patient Self-Reported Race and Ethnicity in Quality Improvement Review Processes at a Single Safety Net Academic Institution

Abstract

Background While there is a plethora of evidence describing racial and ethnic disparities in obstetric care and outcomes, little has been published evaluating potential inequities in departmental Patient Safety and Quality Improvement (PSQI) processes. Objective The study aims to describe the distribution of patient-reported race or ethnicity for safety events at a single safety net teaching hospital. We hypothesized that the observed versus expected case distribution for each racial or ethnic group would be similar, signifying proportional representation in the PSQI reporting and review process. Study design We performed a cross-sectional study including all Safety Intelligence (SI) events filed on obstetric and gynecologic patients and all cases reviewed at monthly PSQI multidisciplinary departmental meetings from May 2016 to December 2021. We compared the distribution of patients' self-reported race or ethnicity as documented in the medical record to our patient population's expected race or ethnicity distribution based on historical institutional data. Results Two thousand and five SI events were filed on obstetric and gynecologic patients. Of those, 411 cases were selected for review by the departmental multidisciplinary PSQI committee, which meets once monthly. Of the 411 cases reviewed by the PSQI committee, 132 met Severe Maternal Morbidity (SMM) criteria defined by the American College of Obstetricians and Gynecologists (ACOG). Fewer SI reports were filed on Asian patients and those who declined to provide race or ethnicity (observed 4.3% versus expected 5.5%, p=0.0088 and 2.9% versus expected 1%, p<0.0001, respectively). For cases reviewed by the departmental PSQI committee and for those which met SMM criteria, there was no significant difference in race/ethnicity distribution. Conclusions There was a disparity between fewer safety events filed for Asian patients and those not reporting race/ethnicity. It was reassuring that our process did not identify other racial/ethnic disparities. However, given the widespread systemic inequities in healthcare, further evaluation of our PSQI process, and PSQI processes beyond our institution, is needed.

Keywords: health disparity; health equity; health inequity; maternal ethnicity; maternal race; patient safety and quality improvement; safety intelligence; severe maternal morbidity.

Figure 1. Observed versus expected distribution of patient self-reported race or ethnicity in Patient Safety and Quality Improvement processes at a single safety net academic institution.