In this edition of the Journal, Dr David Paul and colleagues report racial differences in the receipt of prenatal and intrapartum interventions among a cohort of Caucasian and African-American women delivering infants with a birth weight less than 1500 g. The study is from a state-wide referral center. African-American women were less likely to receive antenatal steroids, tocolytics and cesarean delivery. These differences persisted after controlling for confounding medical complications and even income. How can we explain these differences? Candidate explanations include patient factors, system factors and provider factors.

It has been observed that African Americans are less likely to get prenatal care and register later for care than Caucasians. These differences may be related to access, economic barriers, cultural differences or a lack of trust in the system. Women presenting later in the course of preterm labor and with advanced cervical dilatation are not candidates for tocolytics. In such situations, there may not be sufficient time to administer steroids. The authors did not examine time from admission to the hospital until delivery, so it is unclear if this patient factor contributed to the observed differences in this study. Could there have been racial differences in patient refusal of steroids, tocolytics or cesarean delivery? This seems unlikely and other studies have found that differences in refusal rates are generally small and do not fully explain health-care disparities.

System factors may include differences in referral patterns such that only those women at highest risk for preterm delivery were referred or that more affluent women were cared for elsewhere, biasing the study. The likelihood is that these potential biases are small and do not account for the results reported. It is likely that the patient population was representative of the State of Delaware and, given the completeness of the data set, the study reflects the clinical care received by women delivering very low birthweight infants.

Do physicians have a bias concerning racial differences in risk? We would all agree that our African-American citizens are at greatest risk for preterm birth, but are the offspring considered at lower risk for morbidity and mortality associated with very low birthweight? We must also ask: Is it possible that categorization including stereotyping, bias and prejudice on the part of educated health-care professionals contributes to the racial disparities in health care? The classic experiment by Schulman et al. demonstrated that health-care providers are influenced by a patient's race and possibly gender.1 As discussed by the Institute of Medicine (IOM) Committee in the publication, Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare, these processes do not necessarily involve awareness or conscious motivation to discriminate but rather are probably subconscious.2

The etiologies of the racial differences in the receipt of health care are probably multiple and complex. Given the limitations of the study, these authors very correctly did not try to draw any causal inferences from the data. They did not attempt to associate the differences in care with neonatal or perinatal outcomes, nor did they speculate as to the reasons for the discrepancies. Instead, they correctly appealed for the development of strategies to ‘decrease the racial gap in the care of mothers … delivering high risk premature infants’. So what do we do? The IOM outlined a series of interventions involving health policy, health systems, patient education and empowerment. We must also keep this disturbing issue on the radar screen of health-care professionals. Studies such as this one serve to raise our level of consciousness. We must each be vigilant to avoid subtle stereotyping, bias and discrimination. It is time to admit that evidence of racism and discrimination remains in many sectors of life within the United States and health care has not been exempt. By making a conscious effort to treat all of our patients equally without regard to race, gender or socioeconomic status, we as individuals can contribute to the elimination of disparities in health care. Eliminating disparities in health care will lessen disparities in health outcomes.