Institutional Racism: A Key Contributor to Perinatal Health Inequity

The experience of childbirth and its outcomes in the United States varies by race: Black and Hispanic infants face higher risks of neonatal morbidity and mortality even after accounting for clinical risk. Although perinatal disparities are rooted in a complex set of causes, there is increasing recognition that differential quality of care contributes to disparities.^1–3  This has been most clearly demonstrated among preterm infants receiving care in the NICU. Minority infants and families are more likely to receive care in poor quality NICUs^3,4  and within a NICU are more likely to receive worse care.¹ 

In this issue of Pediatrics, Glazer et al⁵  found that in New York City, compared with white infants, Black and Hispanic term infants had increased risk of complications. The study confirms the grim reality of racial and ethnic disparities in infant outcomes in a low-risk population that accounts for 90% of all births. Disparities within individual hospitals were attenuated when adjusted for sociodemographic factors. However, the disparities measured across hospitals were significant. Prevalence of complications differed by race and ethnicity and by hospital. Infection rates demonstrated the widest variation: a sixfold difference between highest- and lowest-performing tertile of hospitals. Although other factors, such as data validity, case mix, and obstetric and neonatal practice variation may account for observed disparities between hospitals, and further investigation into underlying drivers of disparity is needed, this study adds to a growing body of literature that highlights that infants born to mothers from vulnerable populations are being sorted into lower-performing hospitals.^1,3,4  Barriers that prevent racial and ethnic minority and vulnerable populations from accessing high-quality care are a form of institutional racism that perpetuates disadvantage.

Although local changes can ameliorate this disadvantage, we think this untenable situation requires systemic solutions. Several policy options can be applied to remedy the problem of segregation. The promise of these policy options and the potential drawbacks need to be carefully considered (Fig 1), because each may have strengths and weaknesses. First, we can support hospitals that predominantly care for vulnerable populations by improving their capacity to conduct effective quality improvement. Specific quality gaps can be addressed through individual efforts or through collaborative quality improvement networks that are tailored to the needs of hospitals that serve vulnerable populations. Collaborative quality improvement sustained by a peer-learning network of hospitals has been a highly effective strategy to improve population level outcomes^6–8  but has not been applied in the context of hospitals that serve vulnerable populations. Regulatory oversight would be needed, but incentives could be provided to enable hospitals to participate in peer-learning networks or to create them in the first place. Partnerships with academic medical institutions and with other high-performing hospitals can provide the knowledge and resources to improve quality. Additional investments may be required to address resource limitations, which prevent hospitals from matching the needs of vulnerable populations with their clinical and social needs. Resources could also help hospitals work closely with community partners and state agencies, to help them overcome the social disadvantage faced by their patient population. Ultimately, to address structural, systemic, and interpersonal racism and help rebuild trust with the medical system, hospital care systems and practices may require redesign to purposefully include partnerships with community-based organizations.

Second, policy makers could steer patients toward higher-performing hospitals. Unfortunately, although this may sound like an attractive option, multiple factors could degrade its effectiveness. It is unclear whether other hospitals could handle the additional volume, and although higher volume is generally beneficial in the setting of preterm birth,⁹  there may be an inflection point at which overcrowding may denigrate quality.¹⁰  In addition, this option removes choice from vulnerable populations who often choose care in hospitals serving their community. It may also create care deserts with long commutes for populations that often struggle with transportation. Finally, given that Medicaid reimbursement rates are often below cost,^11,12  a worsening payer mix may, over time, generate financial distress at receiving hospitals and denigrate care there.

Third, policy makers could explore models of co-ownership and joint ventures to strengthen quality capacity and available resources. These models might lead to a more efficient regionalized care system. However, they need to be structured and monitored carefully to avoid unintended consequences, such as a further deterioration in availability of ancillary resources or inappropriate patient transfer agreements. Ultimately, we need to deploy all available tools to improve quality at low-performing hospitals. Given the potential caveats of moving patients to higher-performing hospitals, experimentation should likely focus on strengthening quality-improvement capacity and ownership structures that permit a more equitable distribution of funds. Financial incentives should reward such activities.

Access to high-quality care for vulnerable populations is the fundamental question that this study forces us to reckon with. A key knowledge gap that needs to be addressed is whether the pattern of differential quality highlighted in the study varies by geography. It is possible that the magnitude and the source of disparities between birth hospitals varies by region. We do not have definitive answers to the generalizability question. A study of maternal outcomes among New York City hospitals by some of the same authors found that hospital quality of care appeared to drive 48% of the substantial Black disparity in severe maternal morbidity.¹³  However, when the same analytic approach was used to assess the contribution of hospital quality of care to maternal morbidity disparities among California facilities, the results were much more modest, and only 8% was accounted for.¹⁴  This underscores the need to replicate this study in other geographies. However, specific to unexpected newborn complications, wide variation driven by hospital-related factors has been shown in other studies^15,16  Although these studies did not break down results by race and ethnicity, the high proportion of variation explained by hospital factors suggests both the importance of the delivery hospital and the potential opportunities to improve care.

Segregated and unequal care contributes to racial and ethnic disparities in perinatal outcomes and sustains disadvantage for infants and their families. The critical need is to identify the primary drivers of disparity and act to eliminate them. The study by Glazer et al makes an important contribution: in an understudied population that accounts for the vast majority of births, the study not only shows variation on an important composite measure of quality by race and ethnicity, but also points the way forward by identifying the primary sources of variation, highlighting the important role of the delivery hospital.