Most epidemiologic studies have shown that the risk of adverse birth outcomes is twice as high among blacks as among whites. However, many of the clinical studies designed to explore adverse birth outcomes were limited to data from one or a few hospitals. Thus, generalizability of their results is questionable. Moreover, a small sample size makes modeling of a large number of risk factors of adverse birth outcome difficult. Conversely, most large-scale data sets often have limited information on potential risk factors. In contrast, National Maternal and Infant Health Survey (NMIHS) is a nationally representative large data set, which contains substantial information on determinants of birth outcomes. The NMIHS links U.S. vital statistics to information derived from extensive questionnaires completed by women who gave birth in 1988. The NMIHS data was utilized to examine the extent of variation by race in the prevalence of compromised birth outcomes and whether the racial differentials persisted after simultaneous control for a large number of risk factors. Also, the study examined whether the direction and strength of the associations varied by the type of compromised birth outcome. The analysis was restricted to 8,424 live births of at least 500g and to infants born at 22 through 50 weeks of gestation in order to limit errors in misclassifying stillbirths, in recording birthweight and also to limit miscalculation of gestational age. The three compromised birth outcomes were intrauterine growth retardation, prematurity, and heavy preemie. Intrauterine growth retarded infants are those whose with a birthweight less than 85% of the mean birthweight for gestational age of a sex-specific fetal distribution. Premature infants are those with a birthweight is < 2500g and gestation is < 37 weeks, but are not classified as intrauterine growth retarded. Heavy preemies are defined as infants with gestation age <36 weeks and birthweight 2500g or more. Heavy preemie is considered an adverse outcome since the mortality risk for heavy preemies may be twice the risk for normal infants.

Approximately 82% of Mexican-American and 84% of non-Hispanic white infants were classified as normal, while 73% of black infants were. Teen mothers represented 22% of the black mothers, 16% of the Mexican-Americans, and only 10% of the non-Hispanic white mothers. While over 95% of black and white women were born in the U.S., only 57% of the Mexican-American women were. Also, the blacks and Mexican-American women had similar parity, while the non-Hispanic white women had lower parity. Maternal education, annual household income, and receipt of government were used as measures of socioeconomic status. In every instance, low socioeconomic status characterized the two minority groups. Despite this, Mexican-American women were more likely than either blacks or non-Hispanic whites to pay for delivery costs out of their own pocket.

The list of determinants in the analysis was extensive and included sociodemographic variables, socioeconomic status factors, biomedical variables, the quality of prenatal care, maternal smoking, and vaginal bleeding during pregnancy. The risk of intrauterine growth retardation was 40% higher among black infants than among non-Hispanic white infants, while the risk among non-Hispanic whites and Mexican-Americans were comparable. Among blacks, unadjusted and adjusted odds ratios for premature infant than among the non-Hispanic infants were respectively 2.8 and 2.2. Although both the unadjusted and adjusted odds ratios for prematurity among Mexican-Americans than among the non-Hispanics was respectively 0.10, the odds ratios were not statistically significant. Furthermore, the unadjusted and adjusted odds ratios for heavy preemie among blacks than among non-Hispanic whites were 3.7 and 2.8 respectively. The unadjusted and adjusted odds ratios for heavy preemie among Mexican-Americans than among non-Hispanic whites were 1.9 and 1.7 respectively.

Among the three compromised birth outcomes, intrauterine growth retardation, prematurity, and heavy preemie, African-Americans were at greater risk than non-Hispanic whites and Mexican-Americans. For intrauterine growth retardation and prematurity, Mexican-Americans were comparable to non-Hispanic whites. However, their odds of heavy preemies was approximately twice the odds among the non-Hispanic whites, which probably reflects the higher prevalence of diabetes among Mexican-American women. Although, the net effects of education, income, and receipt of welfare, proxies for socioeconomic status, were not statistically significant, their influence on more proximate risk factors, that directly affect birth outcome, can not be discounted. The only statistically significant variables that reduced the odds of the compromised birth outcomes were women who gained 41or more pounds during pregnancy and received Special Supplemental Food Program for Women, Infants, and Children (WIC) support. These two findings are “mutually reinforcing with regard to the importance of nutrition, and suggest that WIC is cost-effective, especially when considered in relation to the cost of the medical and social intervention necessary to prevent the death and improve the health of infants and children who experience birth outcomes.” The most disturbing finding of the study is that, even after adjustment for an extensive list of determinants, African-American still experienced very high rates of compromised birth outcomes relative to the non-Hispanic white population. But, it is even more distressing that risk among African-Americans is quite high relative to the Mexican-Americans, a minority group that is socioeconomically more disadvantaged than the African –Americans economically.