The US Black–White Infant Mortality Gap: Marker of Deep Inequities

The infant mortality rate has long been considered a summary measure of the health of a population.^1,2 The differential between infant mortality rates for babies born to White and Black mothers in the United States has been a focal point in public health for many years.³ Although infant mortality has continually declined over the past century, this gap has been persistent, even growing at times in recent decades. The infant mortality gap has been resistant to local, state, and national efforts including expanded health insurance coverage for pregnant women and infants, the national Healthy Start initiative, and other public health programs.

WONDER DATABASE

In their article in this issue of AJPH, Brown Speights et al.⁴ examined the recent experience with Black and White infant mortality across US states. They used publicly available data from the Centers for Disease Control and Prevention’s (CDC’s) WONDER database (https://wonder.cdc.gov) to examine infant mortality rates for the period 1999 to 2013, using three-year rolling average estimates, excluding those states with fewer than 10 annual infant deaths in each racial group and classifying deaths without regard to Hispanic ethnicity. The focus only on race of mother differs from the typical classification by maternal race/ethnicity, and may render comparisons with other state and national studies more difficult. No sociodemographic characteristics were included in the analysis, nor were other infant outcomes (birth weight, gestational age) or infant age at death (hebdomadal, neonatal, postneonatal mortality).

The authors’ rationale for the study design was based in part on their desire to give a simple, straightforward look at the patterns of the Black–White infant mortality gap across the states, but also had a pragmatic basis. Not only has it become more difficult for researchers to obtain nationwide individual-level data sets from the National Center for Health Statistics with place-specific identifiers, but also the time frame of interest cuts across the years when all states used the 1989 revision of the national standard certificates of live birth and infant death, through the gradual implementation of the 2003 revision, resulting in some analytical complexities the authors may have wished to avoid. However, obtaining data from CDC WONDER with a focus only on maternal race limited the scope of the state-level analysis. In the absence of individual-level records, incorporation of risk factors well-known to influence population-level infant mortality rates into the trend analyses was rendered extremely difficult. Data for examination of at least some covariates could have been obtained through CDC WONDER, and may provide a more nuanced assessment of the recent progress of each state in reducing the infant mortality gap.

STATE-LEVEL VARIATIONS

Most analyses of trends in disparities in Black–White infant mortality take a national^5,6 or regional perspective⁷; the focus on state-level variation in trends and predictions for the future is refreshing. Many common factors influence the likelihood of infant death, but the social, economic, political, and health care contexts of states differ substantially. Brown Speights et al.⁴ classified recent patterns of Black–White infant mortality as converging or parallel based on linear regression analyses of each state’s recent trends. Although many states show convergence based on their recent experience, few are likely to achieve equality in their White and Black infant mortality rates by midcentury, and many of the states identified as showing converging trends have relatively small numbers of annual births to Black women.

BIG PICTURE

In interpreting the state-by-state results, readers should focus on the big picture, rather than specific estimates of the year when that state might achieve equality based on “optimal” or “equitable” improvements in the years to come. The 90% or 95% confidence intervals around these point estimates (not provided) would show their degree of imprecision, which is in part a reflection of the relative sizes of the Black and White birth populations in each state and random variability in the incidence of infant death. Early in my career, I was the data analyst at the Arkansas Department of Health tasked to explain to then-Governor Bill Clinton why the infant mortality rate for the state had increased from the previous year. Although the year-to-year change involved a very small increase in infant deaths (fewer than 10 in all), because the birth population of the state was relatively small, it was sufficient to increase the infant mortality rate by several tenths of a point. Random variability is addressed to some extent by using three-year moving averages, yet some vestiges remain in the estimates reported by Brown Speights et al.⁴

CONTRIBUTION OF OTHER FACTORS

The focus on overall crude infant mortality rates and Black–White rate ratios overlooks the contribution of other factors. Persistent disparities may be rooted at least to some degree in residential segregation, housing policies, quality of public education, differences in access to quality health care, food security, and the structure of the built environment. The relative contributions of preterm and very low birth weight births or distributions of neonatal and postneonatal death have been studied extensively. In the final analysis, the stark fact remains that, even as the overall infant mortality rate of the United States exceeds that of most other developed nations, infant mortality rates for Black infants are unacceptably high in almost every state. The disparities identified by Brown Speights et al.⁴ are likely to continue unless we make a concerted effort to address their root causes.

MARKER OF LARGER ISSUES

Researchers, policymakers, and concerned citizens should remember that racial/ethnic disparities in birth outcomes are markers or symptoms of larger social, economic, and political issues facing our nation. Too often, as public health practitioners, we think treatment of symptoms when faced with major public health concerns. The answer to the infant mortality gap does not lie in universal access to women’s and perinatal health services, nor in higher quality systems of health care for high-risk pregnancies, though these solutions would help. Rather, we need policies that value all children and treat their needs holistically, regardless of their race, ethnicity, immigrant status, or social class, and support their parents in their efforts to nurture them on the tortuous path to full participation in American society. As we think, as a nation, about how to reduce the disparity between Black and White infant mortality rates, we must always be cognizant that lasting solutions involve social and economic equity; access to affordable, quality health and social services; educational opportunities; and, perhaps most importantly, not merely honoring but reveling in the diversity of the peoples who constitute our nation.