2 tables, 1 figure omitted
Fetal deaths at ≥20 weeks' gestation account for 49% of all deaths
that occur between the 20th week of pregnancy and the first year of life.
Although the risk for fetal death has declined substantially since the 1950s,
disparities in the risk for fetal death by race/ethnicity exist.1,2 One
of the national health objectives for 2010 is to reduce deaths among fetuses
of ≥20 weeks' gestation to 4.1 deaths per 1,000 live births plus fetal
deaths for all racial/ethnic populations (objective no. 16-1a).3 To
assess progress toward meeting this objective, CDC analyzed 1990-2000 data
from the National Vital Statistics System (NVSS). The findings indicate substantial
reductions in fetal deaths, primarily because of reductions in late fetal
deaths (≥28 weeks' gestation) compared with early fetal deaths (20-27 weeks'
gestation). Despite these reductions, racial/ethnic disparities in fetal deaths
persist, particularly among non-Hispanic blacks. Prevention strategies should
recognize fetal deaths as a public health problem, improve fetal death surveillance
and reporting, target etiologic research, and educate practitioners in identifying
women at risk.
During 1990-2000, NVSS data included all U.S. fetal deaths that were
reported at ≥20 weeks' gestation. A fetal death was defined as an involuntary
loss in which the fetus showed no evidence of life (i.e., no heartbeat or
respiration) on delivery. Infant death was defined as the delivery of a live-born
infant who subsequently died by age 1 year.4 Perinatal
deaths included fetal and infant deaths (i.e., ≥20 weeks' gestation to
age 1 year). The fetal mortality rate (FMR) was calculated as the number of
fetal deaths in a specified group per 1,000 live births plus fetal deaths.
Fetal deaths with presumed gestation of ≥20 weeks but with unknown specific
gestational ages (2% of all fetal deaths at ≥20 weeks) were redistributed
proportionately to calculate rates for early and late fetal deaths. Race/ethnicity
was based on the self-reported race of the mother. Plurality (i.e., the number
of fetuses delivered in a pregnancy) was categorized into deaths in singleton
and multiple deliveries. All differences in mortality rates (based on Z scores)
are statistically significant (p<0.05) unless otherwise noted.
In 1990, of 69,737 perinatal deaths reported, 29,345 (42%) were fetal
deaths; of these, 12,554 were early fetal deaths, and 16,791 were late fetal
deaths. In 2000, of 54,964 perinatal deaths reported, 27,003 (49%) were fetal
deaths, including 13,497 early fetal deaths and 13,506 late fetal deaths.
A comparison of fetal deaths in 1990 and 2000 by more detailed gestational
age groups indicated a shift in the distribution toward deaths at earlier
gestational ages (Figure). During 1990-2000, late fetal deaths decreased;
however, early fetal deaths at 20-23 weeks increased from 27% to 34%.
During 1990-2000, FMR for all pluralities decreased 12%, from 7.5 per
1,000 live births plus fetal deaths to 6.6 (Table 1). However, contrasting
trends were observed for early and late FMRs. Early FMRs increased slightly,
from 3.2 to 3.3, whereas late FMRs declined 23%, from 4.3 to 3.3.
Among singleton deliveries (91% of all fetal deaths in 2000), FMRs decreased
13% overall (early FMR increased slightly, and late FMR decreased 22%). Among
multiple deliveries, larger declines were observed; overall FMRs decreased
22% (early and late FMRs decreased 11% and 35%, respectively). Despite improvements
in FMRs for multiple deliveries, in 2000, the risk for fetal mortality for
multiples was approximately three times that for singletons (FMR = 18.5 versus
Declines in overall fetal mortality were observed for all groups except
Asians/Pacific Islanders (A/PIs) (Table 2). FMRs declined 27% for American
Indians/Alaska Natives (AI/ANs), 16% for Hispanics, 10% for non-Hispanic whites,
and 5% for non-Hispanic blacks. Increases in early fetal mortality were observed
for all groups except AI/ANs and A/PIs, whereas decreases in late fetal mortality
were observed for all racial/ethnic populations.
Total, early, and late FMRs were substantially higher among non-Hispanic
blacks in 1990 and 2000 than among other racial/ethnic populations. In 2000,
FMR for non-Hispanic blacks was 12.1, compared with a total U.S. FMR of 6.6.
The gap between FMRs for non-Hispanic blacks and other racial/ethnic populations
remained wide during during 1990-2000; the FMR ratio between non-Hispanic
blacks and the U.S. total was 1.7 for 1990 and 1.8 for 2000.
W Barfield, MD, Div of Reproductive Health, National Center for Chronic
Disease Prevention and Health Promotion; J Martin, MPH, D Hoyert, PhD, Div
of Vital Statistics, National Center for Health Statistics, CDC.
The findings in this report indicate substantial declines in late fetal
mortality for all populations (i.e., singleton and multiple deliveries in
all racial/ethnic populations) during 1990-2000. The 23% decline observed
in late fetal mortality is similar to that reported for neonatal mortality
during the same period.2
The findings also indicate that mortality among fetuses delivered at
20-27 weeks' gestation increased slightly during 1990-2000. This finding was
consistent for all populations except multiples and AI/ANs. To determine whether
the lack of improvement in early fetal mortality was restricted to earlier
and less viable gestations, FMRs were calculated separately for fetal deaths
at 20-23 and 24-27 weeks. No improvement in FMRs was reported for either group.
In contrast, outcomes among live-born infants delivered at 24-27 weeks have
Trends in the risk for early and late fetal death suggest that changes
in perinatal technologies (e.g., fetal imaging, prevention of perinatal infections,
effective treatment of maternal medical conditions such as diabetes and chronic
hypertension, and more aggressive management of labor and delivery)6 might have had more of an impact on fetal survival
at later rather than earlier gestational ages. In addition, rates of prenatal-care
use increased substantially during the 1990s,7 and
the subsequent improved access to care also might have had more impact on
late rather than early fetal mortality (e.g., through the detection of maternal,
fetal, or placental abnormalities that might lead to a live-born delivery).
The lack of progress in reducing fetal mortality at earlier gestational ages
might be related to (1) poor understanding of the factors associated with
premature delivery and (2) limited understanding of the causes of fetal death
and the role of maternal, fetal, and placental pathology.8
Racial/ethnic disparities in fetal mortality persist. The risk for early
fetal mortality among non-Hispanic blacks was more than double that for other
racial/ethnic populations; the risk for late fetal mortality for non-Hispanic
blacks was approximately two thirds higher. Despite substantial reductions
in late fetal mortality among non-Hispanic blacks during 1990-2000, greater
reductions must occur during the next decade if the 2010 national health objective
is to be achieved.
CDC, in conjunction with state partners, has revised the U.S. Standard
Report of Fetal Death and is collaborating with the National Association for
Public Health Statistics and Information Systems and the Social Security Administration
to improve the quality of fetal death surveillance.4,9 Several
states are linking additional data sources (e.g., maternal delivery and hospital
records) to fetal death records to enhance understanding of the relationship
between maternal disease and fetal death.9 The
National Institutes of Child Health and Human Development recently established
a national research agenda on stillbirths (fetal deaths at ≥20 weeks' gestation)8 and granted awards to five sites to conduct population-based
studies of stillbirths. The March of Dimes also has launched a 5-year national
prematurity campaign focusing on education and research that includes fetal
The findings in this report are subject to at least two limitations.
First, fetal deaths might be underreported, particularly at earlier gestational
ages. Considerable variability was observed among reporting areas in the quality
and completeness of data elements such as Hispanic ethnicity.4 The
completeness of reporting of gestational age improved from 93.6% to 96.4%
for all records during 1990-2000. Because reporting tends to be less complete
at earlier gestational ages,4 the likely
impact of this change, if any, would be to increase the number of early fetal
deaths. Thus, the increase in early fetal mortality (but not overall mortality)
might be less than reported here. Second, live-born infants who died shortly
after birth, particularly infants born at early gestational ages or low birthweights,
might have been misclassified as fetal deaths.
In the United States, the number of fetal deaths nearly equals that
of infant deaths during the perinatal period. Mechanisms that contribute to
fetal and infant deaths might be related in certain circumstances, but targeting
infant deaths is insufficient to appropriately address adverse pregnancy outcomes.
Prevention efforts must first acknowledge fetal deaths as a public health
problem. Prevention strategies should begin with accurate surveillance and
reporting of fetal death to provide researchers with population-based information
that provides a better understanding of the factors associated with fetal
death. Racial/ethnic disparities also must be addressed in research on fetal
deaths, including preconceptional and maternal medical conditions, fetal and
placental pathology, and social and environmental factors. These findings
can be used by health-care professionals to (1) improve the identification
of women at risk for fetal death and (2) educate practitioners on improving
the health of women and their infants.
Racial/Ethnic Trends in Fetal Mortality—United States, 1990-2000. JAMA. 2004;292(5):559-561. doi:10.1001/jama.292.5.559