Compared with white mothers born in the United States, Asian Indian mothers who have immigrated to the country have a lower prevalence of several social and demographic risk factors for poor birth outcomes; nevertheless, according to a population-based study conducted in California, they are more likely to have a low-birth-weight or premature infant, or to experience fetal death.1 Foreign-born Mexican mothers have a higher prevalence of risk factors than white mothers, but fare better on some birth outcomes. U.S.-born black mothers also have a higher prevalence of risk factors than white mothers, but their rates of most poor birth outcomes are the highest. Factors associated with reduced odds of white and black mothers' having a low-birth-weight infant do not show a similar pattern of associations among Asian Indian and Mexican mothers.
Researchers analyzed data from California's infant birth and death certificate files for the years 1995-1997. Ethnicity was self-reported. The birth outcomes assessed were low birth weight (less than 2,500 g); very low birth weight (less than 1,500 g); preterm birth (before 37 weeks' gestation); intrauterine growth retardation (birth weight less than the third percentile); and fetal and infant mortality. Analyses were based on about one million births, of which 48% were to U.S.-born non-Hispanic white women, 41% to foreign-born Mexican women, 10% to U.S.-born black women and 1% to foreign-born Asian Indian women.
Risk factors differed significantly, and often dramatically, by ethnic group. Larger proportions of Mexican and black mothers than of white mothers were teenagers (11% and 19%, respectively, vs. 8%), began prenatal care in the third trimester or had no prenatal care (7% and 6% vs. 4%), had deliveries that were paid for by Medicaid (72% and 55% vs. 23%), had a high school education or less (69% and 21% vs. 10%), and had a partner who had no more than a high school education (61% and 11% vs. 7%). In contrast, smaller proportions of Asian Indian than of white mothers were teenagers (1% vs. 8%), had at most a secondary education (8% vs. 10%) and had deliveries that were paid for by Medicaid (18% vs. 23%).
During pregnancy, black women had a higher incidence of hypertension and a lower incidence of diabetes than white women. Among the immigrant groups, the incidence of diabetes was somewhat elevated, but Mexican mothers had a lower incidence of hypertension and placental bleeding problems than white mothers.
Rates of most poor birth outcomes were lowest in white mothers and highest in black mothers. Despite the sharply contrasting risk profiles, Asian Indian mothers' rates of poor birth outcomes were fairly similar to those of black mothers, and rates for Mexican mothers more closely resembled those of white mothers.
Larger proportions of Mexican and black mothers than of white mothers gave birth prematurely (18% and 25%, respectively, vs. 17%) and had an infant with intrauterine growth retardation (3.3% and 7.2% vs. 3.0%). Similarly, Mexican and black mothers experienced higher fetal death rates (5-7 per 1,000 births) than did white mothers (four per 1,000). A larger proportion of black than of white mothers had infants with a low birth weight (13% vs. 6%) or very low birth weight (3% vs. 1%), whereas a smaller proportion of Mexican than of white mothers had low-birth-weight infants (5% vs. 6%). In contrast, larger proportions of Asian Indian than of white mothers gave birth prematurely (20% vs. 17%) and had an infant with a low birth weight (9% vs. 6%), a very low birth weight (1.4% vs. 1.0%) or intrauterine growth retardation (7% vs. 3%). In addition, Asian Indian mothers experienced a higher fetal death rate than white mothers (seven vs. four per 1,000 births).
The rate of deaths in the neonatal period (i.e., by four weeks of age) was higher among black than among white infants (five vs. three per 1,000 live births), but rates were similar for Mexican and white infants. Somewhat unexpectedly, Asian Indian infants also had a neonatal death rate similar to that of white infants. This was explained by a lower neonatal death rate among Asian Indian infants who had a low or very low birth weight, which compensated for the larger proportion of infants born at these weights. In the postneonatal period (four weeks to one year of age), black infants had a markedly higher death rate than whites (4.5 vs. 1.9 per 1,000 infants alive at four weeks), whereas Mexican infants had a lower rate (1.6). Asian Indian infants appeared to have the lowest mortality rate in this period (1.3), but the difference between this rate and the rate for white infants was not statistically significant.
In multivariate analyses of singleton births, white and black mothers' odds of having a low-birth-weight infant were higher if they had not completed high school than if they had (odds ratio, 1.2 for each); the odds were reduced if they had completed college (0.8 for each) or had some college education (0.9 for each). White and Mexican mothers had an elevated likelihood of this outcome when the infant's father had not completed high school (1.2 and 1.1); the likelihood was reduced for all four ethnic groups when the father had completed college (0.8-0.9), and for black and white mothers when the father had some college education (0.9 for each).
Compared with their counterparts aged 20-34, Mexican and Asian Indian teenagers had increased odds of having a low-birth-weight infant (odds ratios, 1.4 and 3.4), but black teenagers had reduced odds of this outcome (0.8). Women aged 35 or older had increased odds in all four ethnic groups (1.3-1.6). The likelihood was elevated for black and white mothers when prenatal care started in the second trimester (1.1 for each), and for all mothers when this care began in the third trimester or did not occur (1.2-1.8). White, black and Mexican mothers had elevated odds when their delivery was covered by Medicaid (1.1-1.3) or they paid for it themselves (1.3-2.0).
The odds of having a low-birth-weight infant were elevated in black, white and Mexican mothers who had hypertension during pregnancy (odds ratios, 1.9-3.7), but were reduced in white and Mexican mothers who had diabetes during pregnancy (0.9 for each). Mothers of all ethnicities had sharply elevated odds if their pregnancy was complicated by preeclampsia or eclampsia (3.8-7.6) or by a placental bleeding disorder (4.3-9.7). The odds were also increased in each group for mothers whose infants were female (1.0-1.2).
The researchers contend that the findings reveal a new "epidemiologic paradox" in foreign-born Asian Indian women that, like the documented paradox in foreign-born Mexican women, cannot be fully explained by known risk factors. Maternal diet, social support, attitudes toward pregnancy and stress may be among the unidentified factors influencing birth outcomes, they note. The dual paradox highlights "the need for continued research directed at understanding the mechanisms by which social factors influence perinatal outcomes," they conclude.
-S. London