Objective
The objective of the study was to compare the mortality risk of expectant management with the risk of delivery at each week of term pregnancy in 4 racial/ethnic groups.
Study Design
This was a retrospective cohort study of all nonanomalous, term deliveries in California from 1997 to 2006 among white, black, Hispanic, and Asian women. In each racial/ethnic group, we compared the risk of infant death at each week with a composite risk representing the mortality risk of 1 week of expectant management.
Results
The risk of stillbirth and infant death is highest in black women (stillbirth risk: 18.0 per 10,000, infant death: 24.4 per 10,000, compared with 9.4 per 10,000 and 10.8 per 10,000 in white women, respectively; P < .001). Although absolute risks differ by race/ethnicity, the composite risk of expectant management does not surpass the risk of delivery until 39 weeks in any group. At 39 weeks these absolute risk differences are low, however, with a number needed to deliver to prevent 1 death ranging from 751 (among black women) to 2587 (among Asian women).
Conclusion
The mortality risk of expectant management exceeds the risk of delivery at 39 weeks in all racial/ethnic groups, despite variation in absolute risks.
Stillbirth and infant death are two adverse perinatal outcomes that have been shown to demonstrate substantial disparities among different racial/ethnic groups. In particular, non-Hispanic black women have elevated risks of both stillbirth and infant death, with rates of both more than double that seen in non-Hispanic whites. Hispanic and Asian women have rates of stillbirth similar to non-Hispanic white women with lower rates of infant death. Although the complications of preterm delivery do play a large role in some of these disparities, 29% of all stillbirths after 24 weeks and 63% of infant deaths occur in term pregnancies, with similar rates of disparities seen.
Risk factors for both stillbirth and infant death are similar, and there are some causes of death that can result in either stillbirth or infant death (eg, congenital anomaly or hydrops), depending on the gestational age (GA) at delivery. Other causes are not similar. For example, infant death related to prematurity is less likely to occur if the pregnancy is prolonged, but the risks of placental insufficiency or cord accident increase as the gestation increases and disappear once delivery is achieved.
Early delivery decreases the risk of stillbirth, but neonatal complications are higher in babies born before 39 weeks, leading to efforts to decrease elective early-term deliveries. Recently an American College of Obstetricians and Gynecologists Committee Opinion restated that women should not be delivered without medical indication prior to 39 weeks’ gestation. However, there may be populations of women (for example, black women) whose increased risk of stillbirth merits a delivery earlier than 39 weeks’ gestation. Determining the ideal time to deliver a pregnancy at term must consider the mortality risks to the fetus/infant in both the intrauterine and postnatal environment.
We have previously demonstrated that a composite metric of stillbirth and infant mortality rate can be useful for quantifying the mortality risks faced by pregnant women at each week of gestation at term in both low- and high-risk groups. In this study, we compared the mortality risk of expectant management with the risk of infant death at term across 4 racial/ethnic groups.
Materials and Methods
We conducted a retrospective cohort study of California births that occurred between 1997 and 2006, using deidentified data from the California Vital Statistics Birth Certificate Data, California Patient Discharge Data, Vital Statistics Death Certificate Data, and Vital Statistics Fetal Death File. The California Office of Statewide Health Planning and Development, as part of the California Health and Human Services Agency, maintains linked data sets that include maternal antepartum and postpartum hospital records as well as birth records and all infant admissions occurring within the first year of life.
The goal of this project was to compare the mortality risks between delivery at a particular gestational age with that of expectant management (ie, continuing the pregnancy for another week and then delivering 1 week later) in 4 common racial/ethnic groups. More specifically, the mortality risk of delivery at a given week was defined as the infant mortality rate among those infants born at that week of gestation.
The mortality risk of a week of expectant management was defined as the risk of stillbirth during that week plus the mortality risk experienced by infants born in the subsequent week of gestation. This composite risk of expectant management beyond each given week of gestation was then compared with the risk of infant death for infants born in the given week of gestation.
The number needed to deliver was calculated as an analogous measure to the number needed to treat by taking the reciprocal of the absolute risk difference between delivery and expectant management. When the 95% confidence interval (CI) of the absolute risk difference included zero, the 95% CI of the number needed to deliver approaches infinity and is not well defined, thus was not reported. Infant death, rather than neonatal death, was chosen as the preferred metric to examine because neonatal death (death within the first 28 days of life) does not capture all of the mortality that can be attributed to management of term pregnancies. Specifically, gestational age has been associated with infant death in prior studies.
Women were divided into 4 major racial/ethnic categories based on their categorization on their infant’s birth certificate: white (non-Hispanic), black (non-Hispanic), Hispanic, and Asian. Women with other racial identities or missing information were excluded from this analysis. We compared the absolute risks of stillbirth and infant death among these 4 groups as well as the gestational age at which the overall mortality is the lowest.
The incidence of stillbirth at a given gestational age was calculated as the number of stillbirths at that GA per 10,000 ongoing pregnancies minus half of the births in the given week. This correction factor (subtracting half of the deliveries during the week of investigation from the denominator of total ongoing pregnancies) was described by Smith as a way to correct for the censoring of pregnancies that are delivered during the week, assuming they occurred on average halfway through the week. Infant mortality at each GA was calculated as the number of infants born at that gestational age who die within 1 year of life per 10,000 live births at that same GA.
These birth certificate data use the last menstrual period (LMP) as the basis for gestational age dating; we included gestational ages between 37 weeks 0 days and 42 weeks 6 days. We excluded mother/infant pairs if the LMP was missing or nonsensical. Exclusion criteria included multiple gestations and complications such as diabetes mellitus (preexisting or gestational), chronic hypertension, and congenital anomalies or genetic causes of death based on the International Statistical Classification of Diseases and Related Health Problems , revision 10 codes.
Statistical calculations were performed with Excel and Stata (version 12; StataCorp, College Station, TX), including proportions, relative risks, and 95% CIs. χ 2 tests were performed to compare proportions of categorical variables and analysis of variance was performed to compare means. Statistical significance was reached with a value of P < .05 or if the 95% CIs did not include one. We assumed that the binomial probability distributions of both mortality risks approximated the normal distribution and derived the CI of the composite risk using the sum of the variances plus twice the covariance of the estimates of infant death and stillbirth.
We obtained institutional review board approval from the Committee on Human Research at the University of California, San Francisco, and the California Committee for the Protection of Human Subjects. Because the data are deidentified and part of the public record of vital statistics, informed consent was not required.
Results
Between 1997 and 2006, there were 5,396,680 deliveries available for analysis. Our cohort of term deliveries excluded 396,311 because of missing or nonsensical LMP and 653,961 deliveries that occurred before 37 weeks or after 42 weeks. We then excluded pregnancies complicated by congenital anomalies (n = 238,879), multiple gestations (n = 59,930), diabetes (n = 202,211), and chronic hypertension (n = 24,419) and women with race/ethnicity coded as other (n = 56,878) or missing (n = 4611), leaving a final cohort for analysis of 3,759,300 deliveries.
Hispanic women comprised the largest ethnic group, with 45% of all deliveries (n = 1,689,195). The next most populous group is white non-Hispanic women (n = 1,441,804; 38%), followed by Asian (n = 424,383; 11.2%) and black women (n = 203,918; 5%).
Demographic characteristics of these women varied among the racial/ethnic groups, and because of the large population size, even small differences were statistically significant. Compared with white and Asian women, black and Hispanic women were younger, less likely to have a college education, more likely to have Medicaid insurance, less likely to be nulliparous, and more likely to have preeclampsia ( Table 1 ).
| Characteristic | White (n = 1,441,804) | Black (n = 203,918) | Hispanic (n = 1,689,195) | Asian (n = 424,383) | P value a |
|---|---|---|---|---|---|
| Maternal age, y | 29.0 (±6.2) | 26.1 (±6.3) | 26.1 (±5.9) | 30.1 (±5.4) | < .001 |
| More than 12 y of education | 787,849 (61) | 77,309 (43) | 286,178 (19) | 260,987 (70) | < .001 |
| Public insurance | 257,267 (24) | 82,930 (55) | 891,506 (65) | 74,249 (22) | < .001 |
| Nulliparous | 616,408 (43) | 79,875 (39) | 602,464 (36) | 200,697 (47) | < .001 |
| Preeclampsia | 26,017 (1.8) | 5109 (2.5) | 33,832 (2.0) | 5444 (1.3) | < .001 |
| Gestational age, wks | 39.3 (±1.3) | 39.2 (±1.3) | 39.3 (±1.3) | 39.2 (±1.3) | < .001 |
| Birthweight, g | 3494 (±466) | 3300 (±480) | 3425 (±459) | 3304 (±437) | < .001 |
Black women had the highest rate of stillbirth (18.0 per 10,000 deliveries), followed by Hispanic women (10.9 per 10,000) and then Asian and white women (9.6 and 9.4 per 10,000, respectively). When examined by gestational age, the risk of stillbirth increased with gestational age in all groups except for black women, who had a slight decrease in the risk at 40 weeks ( Table 2 and Figure 1 ).
| GA, wks | Stillbirth rate per 10,000 ongoing pregnancies (95% CI) | Infant death rate per 10,000 live births (95% CI) | ||||||
|---|---|---|---|---|---|---|---|---|
| White | Black | Hispanic | Asian | White | Black | Hispanic | Asian | |
| 37 | 2.0 (1.8–2.2) | 3.9 (3.0–4.8) | 2.3 (2.0–2.5) | 2.2 (1.7–2.7) | 16.2 (13.9–18.5) | 36.5 (28.4–44.6) | 10.3 (8.6–11.9) | 9.2 (6.3–12.2) |
| 38 | 2.7 (2.4–3.0) | 6.2 (5.0–7.4) | 3.1 (2.8–3.4) | 3.0 (2.4–3.6) | 10.9 (9.6–12.1) | 27.0 (21.9–32.1) | 8.5 (7.5–9.5) | 7.9 (6.1–9.8) |
| 39 | 3.8 (3.4–4.2 | 8.2 (6.6–9.9) | 4.6 (4.2–5.0) | 3.8 (3.0–4.6) | 9.5 (8.6–10.4) | 18.7 (15.2–22.3) | 7.3 (6.5–8.0) | 6.8 (5.3–8.2) |
| 40 | 4.9 (4.3–5.6) | 8.1 (5.8–10.3) | 6.6 (5.9–7.4) | 6.5 (5.0–7.9) | 10.1 (9.0–11.1) | 23.9 (19.6–28.2) | 7.6 (6.8–8.4) | 6.8 (5.2–8.4) |
| 41 | 8.5 (7.1–9.9) | 15.5 (10.4–20.6) | 10.4 (8.9–11.8) | 10.0 (6.8–13.1) | 10.5 (9.1–11.9) | 23.3 (17.5–29.1) | 8.7 (7.5–10.0) | 10.8 (7.8–13.7) |
| 42 | 22.0 (16.9–27.1) | 32.4 (16.0–48.7) | 20.2 (15.7–24.7) | 22.1 (11.6–32.5) | 13.5 (10.7–16.3) | 24.9 (14.7–35.0) | 8.1 (6.1–10.1) | 9.1 (4.3–13.9) |
Infant death rates by gestational age at birth had an overall U-shaped curve, highest at 37 weeks, lowest at 39 weeks, and then increasing again until 42 weeks ( Table 2 and Figure 2 ). Black infants had the highest overall infant mortality, with a rate of 24.4 deaths per 10,000 live births, which is 3 times the rate seen in Hispanic babies (8.1 per 10,000). Asian babies have the lowest risk of infant death, with a rate of 7.8 per 10,000 live births, and white infants have a risk of 10.8 per 10,000.
Although white and black mothers have very different absolute risks of stillbirth and infant death, the risks of expectant management and delivery at each gestational age follow a similar trajectory; the risk of expectant management is lower than that of delivery at 37 weeks, is approximately equal at 38 weeks, and exceeds the risk of delivery at 39, 40, and 41 weeks ( Table 3 and Figure 3 ). In both the Asian and Hispanic women, the risk of expectant management is similar to that of delivery at 37 weeks but then seems to diverge earlier, at 38 weeks, although a statistically significant difference between expectant management and delivery is not seen until 39 weeks in all groups.
| GA | White | Black | Hispanic | Asian | ||||
|---|---|---|---|---|---|---|---|---|
| Risk of infant death per 10,000 live births (95% CI) | Risk of expectant management for 1 wk (95% CI) | Risk of infant death per 10,000 live births (95% CI) | Risk of expectant management for 1 wk (95% CI) | Risk of infant death per 10,000 live births (95% CI) | Risk of expectant management for 1 wk (95% CI) | Risk of infant death per 10,000 live births (95% CI) | Risk of expectant management for 1 wk (95% CI) | |
| 37 | 16.2 (13.9–18.5) | 12.9 (11.6–14.2) | 36.5 (28.4–44.6) | 30.9 (25.7–36.0) | 10.3 (8.6–11.9) | 10.7 (9.7–11.8) | 9.2 (6.3–12.2) | 10.1 (8.2–12.0) |
| 38 | 10.9 (9.6–12.1) | 12.2 (11.2–13.2) | 27.0 (21.9–32.1) | 24.9 (21.1–28.7) | 8.5 (7.5–9.5) | 10.4 (9.6–11.2) | 7.9 (6.1–9.8) | 9.8 (8.2–11.3) |
| 39 | 9.5 (8.6–10.4) | 13.8 (12.7–14.9) | 18.7 (15.2–22.3) | 32.1 (27.5–36.7) | 7.3 (6.5–8.0) | 12.2 (11.3–13.1) | 6.8 (5.3–8.2) | 10.6 (8.8–12.4) |
| 40 | 10.1 (9.0–11.1) | 15.4 (13.8–16.9) | 23.9 (19.6–28.2) | 31.4 (25.2–37.6) | 7.6 (6.8–8.4) | 15.4 (14.0–16.8) | 6.8 (5.2–8.4) | 17.2 (13.9–20.5) |
| 41 | 10.5 (9.1–11.9) | 22.0 (18.9–25.2) | 23.3 (17.5–29.1) | 40.4 (29.0–51.7) | 8.7 (7.5–10.0) | 18.5 (16.0–21.0) | 10.8 (7.8–13.7) | 19.1 (13.4–24.8) |
| 42 | 13.5 (10.7–16.3) | 24.9 (14.7–35.0) | 8.1 (6.1–10.1) | 9.1 (4.3–13.9) | ||||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
