Physician-patient racial concordance and disparities in birthing mortality for newborns

Introduction

Prior research has documented that sharing demographic characteristics increases a sense of understanding or empathy between two people. This in-group bias has been found to influence the decisions made by leadership teams, inspection enforcers, teachers, and judges to favor those with shared racial or gender traits (further exasperated by white racial power). Greenwood, Hardeman, Huang, and Sojourner posit that the effect of this racial concordance — a shared racial identity — may exist in medicine; yet, previous studies have been limited and have not sought to demonstrate the effect of racial concordance on actual health outcomes. By looking at the health outcomes of 1.8 million hospital births in Florida over 23 years, this study attempted to bridge this gap. Additionally, by using newborns as a test subject, the authors eliminate a possible confounder: the communication skills of the infant patient.

One of the most significant medical racial disparities in health outcomes concerns Black newborns: Black newborns are more than twice as likely as white newborns to die before their first birthday. Given that all newborns are presumably born at the same level of nonverbal communication, the researchers assert that any differences in health outcomes between racially concordant and racially discordant patient-doctor pairs must involve factors other than the ability of the infant patient to self-advocate or provide additional information. The authors hypothesize that the effect of racial concordance on health outcomes may occur because Black doctors are more aware of the racial and socioeconomic barriers that Black mothers and newborns face, and thus may be better equipped to treat, and more attentive to, their patients’ social determinants of health.

Dr. Brad N. Greenwood is an Associate Professor of Information Systems and Operations Management at George Mason University who researches the intended and unintended effects of innovation on welfare. Dr. Rachel R. Hardeman is a tenured Associate Professor of Health Policy & Management at the University of Minnesota School of Public Health and a reproductive health equity researcher. Dr. Laura Huang is an Associate Professor of Organizational Behavior at Harvard Business School who researches interpersonal relationships and implicit bias in the workplace. Dr. Aaron Sojourner is a labor economist and Associate Professor at the University of Minnesota Carlson School of Management who studies labor-market institutions, early childhood and K-12 education, and behavioral economics.

Methods and Findings

The authors examined data from Florida’s Agency for Healthcare Administration from 1992 to 2015, including information about: mother and child’s race, comorbidities, and outcomes; hospital where birth occurred; and attending doctor’s name, specialty certifications, and date of licensure. The authors then tracked the attending doctor’s race through a public photo search. The authors used statistical analysis to estimate the interaction effect of infant race and doctor race, controlling for insurance providers, comorbidities, and factors related to the quarter and year, specific hospital, and specific doctor. This methodology allowed them to investigate whether a newborn’s mortality risk changed based on their doctor’s race.

The authors note four key findings related to Black infancy and medical care:

Finding 1: Overall, Black infants have three times worse health outcomes than white infants regardless of the doctor’s racial identity. However, compared to Black infants treated by white doctors, Black infants treated by Black doctors have half the mortality risk. When accounting for all controls, racial concordance lowers mortality risk for Black infants by 39%. The gap in mortality risk for Black infants is smaller among board certified pediatricians, regardless of race, compared to non-pediatricians, but the mortality gap persists among Black infants treated by white pediatricians compared to Black pediatricians.

Finding 2: As note, as the number of newborn comorbidities rises (representing more medically complicated cases), Black doctors produce increasingly better health outcomes for their Black newborn patients compared to white doctors.

Finding 3: Hospitals that deliver more Black newborns see an even greater benefit of racially concordant patient-doctor pairs. The mortality risk for Black newborns treated by Black patients is fairly stable across hospitals that deliver more Black newborns and those that deliver few Black newborns. However, the gap in infant mortality risk between Black newborns treated by racially concordant and racially discordant patient-doctor pairs is larger at hospitals that deliver more Black newborns than those that deliver few Black newborns, suggesting this disparity in health outcomes is primarily due to the underperformance of white doctors rather than significantly better performance by Black doctors at hospitals that deliver more Black newborns. The authors also demonstrated in supplementary analysis that white doctors’  underperformance does not differ based on the overall experience that hospitals have in treating newborns, or in treating white newborns, but only in their level of experience in treating Black newborns.

Finding 4: There is no statistically significant effect of patient-doctor racial concordance on the mother’s mortality, but the authors note that Black mothers treated by white doctors experience triple the mortality risk of white mothers treated by white doctors. 

Conclusions

This article highlights the critical inequity in health outcomes for Black mothers and their children, in addition to the significance of Black professionals in healthcare. The authors effectively demonstrate that the Black-white newborn mortality gap is smaller for Black newborns treated by Black doctors than white doctors, and that the benefit of racially concordant newborn care is even greater for Black infants with complex medical conditions.

The researchers’ data indicates why Black families giving birth may prefer to seek care from Black doctors. However, the authors note that because the physician-level workforce in the US is disproportionately white, it is not always possible for a Black newborn to receive care from a Black doctor. 

Additionally, doctor performance varies widely regardless of the doctor’s race. The findings presented highlight the need for additional research to understand the drivers of performance among doctors (and their teams) and why Black doctors outperform their white colleagues when caring for Black newborns. Such findings could be used to improve the care that white doctors offer for Black newborns.

Greenwood, Hardeman, et al.’s  work underscores the need to diversify the physician workforce and for further investment in reducing the effects of institutional racism on health outcomes. The authors also point out the need for education for health professionals about the prevalence and effect of racial and ethnic disparities in health outcomes and the need to include other actors, such as nurses, hospital administrators, and policymakers, in examining organizational policies and processes to reduce racial bias.