The U.S. Infant Mortality Rate: International
Comparisons, Underlying Factors, and Federal
Programs
Elayne J. Heisler
Analyst in Health Services
April 4, 2012
Congressional Research Service
7-5700
www.crs.gov
R41378
CRS Report for Congress
Pr
epared for Members and Committees of Congress
The U.S. Infant Mortality Rate
Summary
The infant mortality rate (IMR)—the number of deaths occurring in the first year of life per 1,000
live births—is a widely used proxy for the health status of a nation, and is commonly used for
international comparisons. As of 2008, the U.S. IMR was 6.6, a historically low rate of the United
States, but a rate that is still higher than the Organization for Economic Cooperation and
Development (OECD) average of 4.6. The relatively high U.S. rate—and the number of infant
deaths it indicates—concerns some policymakers. In addition, there is concern that the U.S. IMR
has leveled off after four decades of decline. Reducing the U.S. IMR has been—and continues to
be—a recognized public health objective.
Researchers and policymakers debate the various factors that may explain the high U.S. IMR
relative to other developed countries and its recent stagnation. Potential factors include
international differences in the recording of live births, different rates of low birthweight and
short gestational age births, and racial and ethnic disparities. Researchers conclude that
international recording differences do not explain the relatively high U.S. IMR. In addition, the
data suggest that racial disparities may only partially explain the relatively high U.S. IMR.
Instead, researchers suggest that higher U.S. rates of low birthweight and short gestational age
births may explain the relatively high U.S. IMR.
This report examines the U.S. IMR. It identifies the top three causes of U.S. infant death—
congenital malformations, disorders related to low birthweight and short gestational age, and
sudden infant death syndrome (SIDS). The report focuses on low birthweight and short
gestational age, because the United States has relatively high and increasing rates of these births,
and research has found that these births can be reduced through policy interventions.
The U.S. IMR varies geographically and is influenced by a number of factors, including the
mother’s demographic characteristics (e.g., education, income, or age) and health and health
system characteristics. In general, southern states have the highest IMRs and states in New
England and the Pacific Northwest have the lowest. The higher IMRs in southern states may be
partially explained by higher rates of low birthweight and short gestational age births in these
states. In addition, the racial and ethnic composition of a state’s population affects its IMR
because of higher IMRs among certain racial and ethnic groups. The IMR is also influenced by
health and health system characteristics, including the mother’s health behaviors, such as drinking
and smoking, and her access to and use of prenatal care.
A number of federal programs that aim to improve the health status, and the economic and social
circumstances, of low-income women and children may reduce the U.S. IMR. These programs
include Healthy Start, Maternal and Child Health Services Block Grants, Medicaid, and the State
Children’s Health Insurance Program (CHIP). Evaluating whether a particular program reduces
the IMR is challenging because individuals may be eligible for multiple programs and because
programs target those with IMR risk. Given this, it is difficult to determine the effectiveness of a
single program, and it is difficult to determine whether findings that a program does not reduce
the IMR are due to characteristics of the program or to characteristics of its participants.
The Patient Protection and Affordable Care Act (ACA, P.L. 111-148) either establishes new or
expands existing programs to reduce the IMR. For example, “Strong Start,” funded by the new
Center for Medicare & Medicaid Innovation created in the ACA, focuses on reducing the rate of
pre-term births, which may also reduce the IMR once implemented.
Congressional Research Service
The U.S. Infant Mortality Rate
Contents
Introduction...................................................................................................................................... 1
U.S. Infant Mortality........................................................................................................................ 3
International IMR Comparisons ...................................................................................................... 4
International IMR Comparisons and Trends ............................................................................. 4
Factors That May Contribute to a Relatively High U.S. IMR................................................... 6
Inconsistent Recording of Live Births................................................................................. 6
Rates of Low Birthweight and Short Gestational Age Births.............................................. 8
Racial Disparities ................................................................................................................ 8
Geographic Variation in U.S. Infant Mortality ................................................................................ 9
Factors That May Contribute to State Variation in IMR.......................................................... 10
State Variation in Low Birthweight and Short Gestational Age Births ............................. 10
State Variation in Population Composition ....................................................................... 13
Factors That Influence the U.S. IMR............................................................................................. 13
Mother’s Demographic Characteristics ................................................................................... 14
Race and Ethnicity ............................................................................................................ 14
Education........................................................................................................................... 15
Economic Status................................................................................................................ 15
Age .................................................................................................................................... 16
Health and Health System Characteristics............................................................................... 16
Assisted Reproductive Technology ................................................................................... 16
Health and Health Behaviors............................................................................................. 17
Prenatal and Preconception Care....................................................................................... 18
Quality of Care.................................................................................................................. 19
Access to Care................................................................................................................... 19
Federal Programs That May Reduce Infant Mortality................................................................... 20
Income and Material Support Programs.................................................................................. 20
Health Programs ...................................................................................................................... 21
Healthy Start...................................................................................................................... 21
Maternal and Child Health Services Block Grants ........................................................... 22
Health Education Efforts................................................................................................... 22
Prevention Programs ......................................................................................................... 23
Research Efforts ................................................................................................................ 24
Health Care Access Efforts................................................................................................ 24
Do Federal Programs Reduce Infant Mortality?...................................................................... 25
Evidence of Effects on Infant Mortality............................................................................ 25
Challenges Assessing Program Effectiveness ................................................................... 27
New Federal Initiatives.................................................................................................................. 28
Conclusion ..................................................................................................................................... 29
Figures
Figure 1. U.S. IMR (Infant Deaths/1,000 Live Births), by State, 2008 ......................................... 10
Congressional Research Service
The U.S. Infant Mortality Rate
Tables
Table 1. Ten Leading Causes of U.S. Infant Death, 2008................................................................ 3
Table 2. Infant Mortality Rates (Infant Deaths/1,000 Live Births) and Rankings for
OECD and Other Selected Countries, 2008 ................................................................................. 5
Table 3. Live Birth Recording Requirements, Selected Countries, 2004 ........................................ 7
Table 4. Percent and Rank of State IMR (Infant Deaths/1,000 Live Births) and Short
Gestational Age Births, 2008...................................................................................................... 11
Contacts
Author Contact Information........................................................................................................... 30
Acknowledgments ......................................................................................................................... 30
Congressional Research Service
The U.S. Infant Mortality Rate
Introduction
The infant mortality rate (IMR)—the number of deaths occurring in the first year of life per 1,000
live births—is a widely used proxy for the health status of a nation, and is commonly used for
international comparisons.1 International comparisons in health status are often used by
policymakers to formulate and guide the development of appropriate policy options and effective
programmatic efforts. These comparisons are relied upon to help frame and assess the magnitude
of a given health concern and the potential for remediation. For these reasons, policymakers may
find it useful to understand the potential strengths and weaknesses of a specific indicator,
including characteristics of its data sources and contextual factors that may affect the indicator.
The IMR has received the attention of policymakers in the United States for a number of reasons.
As of 2008,2 the U.S. IMR was 6.6,3 which is a historically low rate for the United States but is
higher than the Organization for Economic Cooperation and Development (OECD)4 average of
4.6.5 This relatively high rate concerns some policymakers because of the number of infant deaths
it indicates. In addition, policymakers are concerned that the U.S. IMR appears to have leveled
off after four decades of decline. The U.S. IMR declined by almost 75% from 1960 to 2000 (from
26.0 to 6.6); however, since 2000, the U.S. IMR has remained relatively stable. From 2000 to
2008, the U.S. IMR declined by 4% (from 6.9 to 6.6).
In partial response to these concerns, policymakers have recognized reducing the U.S. IMR as an
important public health objective as signified by including it as one of the goals of Healthy
People 20206—a set of national health objectives developed by governmental and
nongovernmental scientists to identify the most significant preventable threats to health and
establish national goals for their reduction. The Healthy People 2020 goal is to reduce the U.S.
IMR to 6.0 by 2020.
Researchers and policymakers debate various factors that may explain the high U.S. IMR. Some
researchers have found that a number of social, economic, and demographic characteristics of the
mother are associated with infant death. These characteristics include the mother’s educational
attainment, economic status, and age. Some have argued that the high U.S. IMR reflects racial
and ethnic disparities, as evidenced by racial differences in IMRs.7 Others suggest that the high
1 See for example, Central Intelligence Agency, “The World Fact Book.” https://www.cia.gov/library/publications/the-
world-factbook.html or the World Health Organization, “World Health Statistics, 2011” http://www.who.int/gho/
publications/world_health_statistics/en/index.html. There are a number of other health indicators used such as the
overall mortality rate; life expectancy at birth; and rates of preventable infectious diseases.
2 This report uses 2008 data, the most recent year of final U.S. data.
3 Ariadi M. Minino, et al., National Vital Statics Reports: Deaths: Final Data for 2008, National Center for Health
Statistics, Vol. 59, No. 10, Hyattsville, MD, December 7, 2011. Hereinafter Deaths: Final Data 2008.
4 The Organization for Economic Cooperation and Development (OECD) is an intergovernmental economic
organization in which the 30 member countries discuss, develop, and analyze economic and social policy; see CRS
Report RS21128, The Organization for Economic Cooperation and Development, by James K. Jackson.
5 Organization for Economic Cooperation and Development, OECD Health Data, 2010, http://www.oecd.org/dataoecd/
4/36/46796773.pdf.
6 See Healthy People 2020 Topics & Objectives, Maternal, Infant, and Child Health, at http://www.healthypeople.gov/
2020/topicsobjectives2020/overview.aspx?topicid=26.
7 James W. Collins, Jr. and Richard J. David, “Racial Disparities in Low Birth Weight and Infant Mortality,” Clinical
Perinatology, vol. 36, no. 1 (March 2009), pp. 63-73.
Congressional Research Service
1
The U.S. Infant Mortality Rate
U.S. IMR may reflect variation in a number of health system characteristics, such as the adequacy
of public health services and the availability of health care for women and infants.8
Interventions that seek to improve the IMR often do so by attempting to lower a particular cause
of infant death. However, some causes of death, such as congenital malformations, the leading
cause of U.S. infant death in 20089 may be less amenable to policy intervention.10 Therefore, this
report focuses on the second-leading cause of infant death—disorders related to low birthweight
and short gestational age—because the United States has relatively high rates of these births, and
research has found that these births can be reduced through policy intervention.11 These two
conditions are also linked because shorter gestation provides less time to attain a healthy
birthweight.
Definitions of Low Birthweight and Short
This report examines the U.S. IMR. In
Gestational Age
doing so, it first identifies a number of
Low Birthweight: Infants born at less than 2,500 grams
causes of U.S. infant mortality. The report
(i.e., less than 5.5 lbs).
then examines international IMR
Short Gestational Age: Infants born prior to 37
comparisons and discusses geographic
completed weeks of gestation.
variation in state IMRs. Next, it examines
Note: Short gestational age births may also be referred to
the mother’s demographic characteristics
as premature or pre-term.
and various health system characteristics
that may influence the U.S. IMR. The
Source: Joyce A. Martin, et al., Births: Final Data for 2008,
National Center for Health Statistics, National Vital Statistics
discussion of these factors is not
Reports vol. 59, no.1, Hyattsville, MD, December 8, 2010,
exhaustive, and the report does not attempt
and United States Department of Health & Human Services,
to assess the relative contribution of each
“CDC Features: Premature Birth” December 9, 2009,
factor. The report then describes a number
http://www.cdc.gov/Features/PrematureBirth/.
of federal programs that may indirectly
reduce the IMR. Federal programs dedicated explicitly to reducing the IMR are rare; however,
several programs target economically disadvantaged pregnant women. These women are more
likely to experience an infant death; therefore, programs that aim to improve the economic or
health status of this population may reduce infant mortality. Finally, the report summarizes
federal initiatives included in the ACA that may reduce infant mortality.
8 Federal Interagency Forum on Child and Family Statistics, America’s Children: Key National Indicators of Health
and Well-Being, 2011, Washington, DC, July 2011, http://www.childstats.gov/americaschildren/health2.asp.
9 Deaths: Final Data 2008.
10 The United States reported a 46% decline in the number of infant deaths due to congenital malformations from 1980
to 2000. Consequently, some researchers believe that much of the reduction in death from reducing congenital
malformations has already been achieved. For example, see Arnold Christianson, Christopher P. Howson, and
Bernadette Modell, Global Report on Birth Defects: The Hidden Toll of Dying and Disabled Children, March of
Dimes, White Plains, NY, 2006.
11 United States Department of Health & Human Services, “Preventing Infant Mortality,” press release, January 13,
2006, http://www.hhs.gov/news/factsheet/infant.html. Hereinafter, Preventing Infant Mortality.
Congressional Research Service
2
The U.S. Infant Mortality Rate
U.S. Infant Mortality
Infant mortality refers to deaths that occur during infancy—the first year of life, or from a live
birth to age one.12 It is measured as a rate of infant deaths per 1,000 live births. In 2008, the U.S.
IMR was 6.6. A number of sources collect infant mortality data and use those data to calculate the
IMR. This report presents the most recent year of final data available (2008) from the National
Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC).
NCHS data are generally considered to be the most
authoritative because they are drawn directly from birth
Calculating the U.S. IMR
and death records through the National Vital Statistics
[28,059 infant deaths/ 4,247,694 live births
System, an interagency collaboration that collects birth
x 1,000 = 6.6]
and death certificate data obtained from state vital
statistics offices.13
A variety of conditions can cause infant death. In 2008, there were 28,059 infant deaths. Of these,
68.9% were due to the 10 leading causes of infant deaths shown in Table 1. One-fifth, or 20.1%,
of U.S. infant deaths (5,638 deaths) were attributed to congenital malformations, the leading
cause of infant death in 2008. The second-most common cause of infant death was disorders
related to the infant being born at either a low birthweight, a short gestational age, or both. This
accounted for 16.9% (4,754 deaths) of infant deaths. Sudden Infant Death Syndrome (SIDS) was
the third-most frequent cause of death, accounting for 8.4% (2,353 deaths) of infant deaths.
Table 1. Ten Leading Causes of U.S. Infant Death, 2008
Percent of
Total Infant
Rank
Cause of Death
Deaths
1
Congenital malformations, deformations and chromosomal abnormalities (congenital
20.1
malformations)
2
Disorders related to low birthweight and short gestational age
16.9
3
Sudden infant death syndrome (SIDS)
8.4
4
Newborn affected by maternal complications of pregnancy (maternal complications)
6.3
5
Accidents (unintentional injuries)
4.7
6
Newborn affected by complications of placenta, cord and membranes (cord and
3.8
placenta complications)
7
Bacterial sepsis of newborn
2.5
8
Respiratory distress of newborn
2.2
9
Diseases of the circulatory system
2.1
10 Neonatal
hemorrhage
2.0
Total
Ten Leading Causes of U.S. Infant Death
68.9
Source: Ariadi M. Minino, et al., National Vital Statics Reports: Deaths: Final Data for 2008, National Center for
Health Statistics, Vol. 59, No. 10, Hyattsville, MD, December 7, 2011.
Note: The 10 leading causes of infant death have been relatively stable since 2006; however, there have been
changes in the ranking of some causes of death.
12 David Yaukey, Demography: The Study of Human Populations (Prospect Heights, IL: Waveland Press, 1985), pp.
126-127.
13 See “National Vital Statistics System” at http://www.cdc.gov/nchs/nvss.htm.
Congressional Research Service
3
The U.S. Infant Mortality Rate
International IMR Comparisons
This section compares the U.S. IMR to rates in other selected developed countries. It describes
how the U.S. IMR compares to rates in these countries in 2008, how comparisons have changed
since 1960, and some factors that may explain why the U.S. IMR is higher than rates in a number
of other developed countries. These factors include international differences in how live births are
recorded, differences in rates of low birthweight and short gestational age births, and racial and
ethnic IMR disparities.
International IMR Comparisons and Trends
In the last year for which comparable data were available (2008) the U.S. IMR was 6.6 (see Table
2) and ranked 31st among OECD countries.14 The U.S. IMR is higher than the rates found in many
Western European and East Asian countries. In addition, the 2008 U.S. IMR of 6.6 was higher
than the 2008 OECD average of 4.6.15
The U.S. IMR rank has declined over time and, in doing so, has fallen below the OECD average.
Compared to the countries in Table 2, the U.S. rank has declined since 1960. In 1960, the United
States was ranked 12th among the group. In 1980, the United States ranked 19th; fell to 30th in
2005; and 31st in 2008.16 Prior to 1997, the U.S. IMR was lower than the OECD average;
however, between 1997 and 2008, the U.S. IMR remained stable, while the OECD IMR declined.
Given this, the U.S. IMR is now higher than the OECD average.17
14 This ranking is relative to the OECD countries. Beginning in 2010, NCHS began comparing the U.S. IMR to OECD
countries. Prior to that time, NCHS compared the U.S. to the rates in other developed countries that had comparable
data available from 1960. In general, these countries are members of the OECD; however, NCHS excluded small
OECD countries or those that do not have comparable data over time. For countries that had been included in
comparisons, see Marian F. MacDorman and T.J. Mathews, Behind International Rankings of Infant Mortality: How
the United States Compares with Europe, U.S. Department of Health and Human Services: Centers for Disease Control
and Prevention: National Center for Health Statistics, NCHS Data Brief No.23, Hyattsville, MD, November 2009.
Hereinafter, Behind International Rankings of Infant Mortality.
15 Organization for Economic Cooperation and Development, OECD Health Data, 2010, http://www.oecd.org/
dataoecd/4/36/46796773.pdf.
16 U.S. Department of Health and Human Services: Centers for Disease Control and Prevention: National Center for
Health Statistics, Health United States, 2008, Table 24, http://www.cdc.gov/nchs/data/hus/hus07.pdf#listtables
17 Organization for Economic Cooperation and Development, Health at a Glance 2009: OECD Indicators, Paris,
France, December 8, 2009, http://www.oecd.org/health/healthataglance.
Congressional Research Service
4
The U.S. Infant Mortality Rate
Table 2. Infant Mortality Rates (Infant Deaths/1,000 Live Births) and Rankings for
OECD and Other Selected Countries, 2008
Country IMR
Rank
Luxembourg 1.8
1
Slovenia
2.1
2
Iceland 2.5
3
Sweden 2.5
3
Japan 2.6
5
Finland 2.6
5
Norway 2.7
7
Greece 2.7
7
Czech Republic
2.8
9
Ireland 3.0
10
Portugal 3.3
11
Belgium 3.4
12
Germany
3.5
13
Spain 3.5
13
Austria 3.7
15
Italy 3.7
15
France 3.8
17
Israel 3.8
17
Netherlands 3.8
17
Denmark 4.0
20
Switzerland 4.0
20
Australia 4.1
22
OECD Average
4.6
—
Korea 4.7
23
United Kingdom
4.7
23
New Zealand
4.9
25
Estonia 5.0
26
Hungary 5.6
27
Poland 5.6
27
Canada 5.7
29
Slovak Republic
5.9
30
United States
6.6
31
Chile 7.0
32
Mexico 15.2
33
Turkey 17.0
34
Source: Adapted from Organization for Economic Cooperation and Development, OECD Health Data, 2010,
http://www.oecd.org/dataoecd/4/36/46796773.pdf. U.S. 2008 data from Ariadi M. Minino, et al., National Vital
Congressional Research Service
5
The U.S. Infant Mortality Rate
Statics Reports: Deaths: Final Data for 2008, National Center for Health Statistics, Vol. 59, No. 10, Hyattsville, MD,
December 7, 2011.
Notes: Countries with the same IMR receive the same rank. The country with the next highest IMR is assigned
the rank it would have received had the lower-ranked countries not been tied, (i.e., it skips a rank). For example,
Iceland and Sweden were both assigned a rank of 3. Japan, which has the next lowest IMR, is assigned the rank of
5.
Factors That May Contribute to a Relatively High U.S. IMR18
Analysts have suggested a number of reasons why the U.S. IMR may be higher than the rates
found in other developed nations. This section examines three potential reasons: (1) inconsistent
recording of live births, (2) different rates of low birthweight and short gestational age births, and
(3) racial and ethnic IMR disparities. Racial and ethnic disparities are discussed briefly here and
more fully later in this report (see “Factors That Influence the U.S. IMR”/“Mother’s
Demographic Characteristics”/“Race and Ethnicity”).
Inconsistent Recording of Live Births19
Some suggest that international IMR differences, and the higher U.S. IMR, are caused by
inconsistent recording of live births, because this is the only component of the IMR with
international variation.20 These analysts suggest that some countries record very small infants
who die or infants who do not survive the first 24 hours as stillbirths, while other countries record
these events as live births and include these deaths in the IMR.21 To minimize recording
differences and create comparable international infant mortality data, the United States and most
developed countries have agreed to use the World Health Organization (WHO) definition of live
births:
Live birth is the complete expulsion or extraction from its mother of a product of conception,
irrespective of the duration of the pregnancy, which, after such separation, breathes or shows
any other evidence of life—e.g., a beating of the heart, pulsation of the umbilical cord or
definite movement of voluntary muscles—whether or not the umbilical cord has been cut or
the placenta is attached. Each product of such a birth is considered liveborn.22
Despite the use of the WHO definition, there is still some variation in the recording of live births
among some European countries (see Table 3). Although 12 European countries record all live
18 This discussion draws from U.S. and international IMR rates of various years. Neither the causes of infant death in
the U.S. nor measurement of the IMR changed over this time period; therefore, the conclusions drawn based on
research using prior years of IMR data should also apply to more current years of IMR data.
19 Unless otherwise noted, this section is drawn from Behind International Rankings of Infant Mortality.
20 There are three components of IMR calculation: (1) the numerator: the number of infant deaths; (2) the denominator:
the number of live births; and (3) the way the rate is calculated: infant deaths/1,000 live births. The specifications for
developing the numerator and the calculation do not vary internationally. However, there is some variation in the
denominator—recording of live births—this variation may impact international comparisons of the IMR to a larger
extent than would measurement differences in the numerator.
21 Conrad F. Meier, “Lessons from Cuba on Infant Mortality,” The Heartland Institute, January 13, 2005, or Editorial
“Canada’s Single-Prayer Health care,” Investor’s Business Daily, June 30, 2009.
22 World Health Organization, International Statistical Classification of Diseases and Health Related Problems, 10th
ed. (1993), see http://www.who.int/whosis/indicators/compendium/2008/3mr5/en/.
Congressional Research Service
6
The U.S. Infant Mortality Rate
births,23 other countries use weight, gestational age, or both in their definition of a live birth. For
example, the Czech Republic records live births as infants born above 500 grams (1 pound and
1.63 ounces) or an infant at any birthweight that survives the first 24 hours.24
Table 3. Live Birth Recording Requirements, Selected Countries, 2004
Recording Requirements
Country
Differences by Gestational Age
All live births
Austria, Denmark, England and Wales, Finland, Germany,
Hungary, Italy, Northern Ireland, Scotland, Slovak
Republic, Spain, Sweden, United Statesa
Live births at 12 weeks of gestation or more
Norwayb
Differences by Birthweight
Live births at 500 grams birthweight or more, and
Czech Republic
births at any weight if the infant survives for 24 hours
Live births at 500 grams birthweight or more
Poland, Ireland
Differences by Birthweight and Gestational Age
Live births at 22 weeks of gestation or more, or 500
France
grams birthweight or more
Live births at 22 weeks of gestation or more, or 500
Netherlands
grams birthweight or more if gestational age is
unknown
Sources: Marian F. MacDorman and T.J. Mathews, Behind International Rankings of Infant Mortality: How the United
States Compares with Europe, U.S. Department of Health and Human Services: Centers for Disease Control and
Prevention: National Center for Health Statistics, NCHS Data Brief No.23, Hyattsville, MD, November 2009.
a. In addition to the countries noted in the table, Canada and Japan record al live births regardless of
gestational age or weight. See Organization for Economic Cooperation and Development Health Data, 2009
at http://www.stats.oecd.org.
b. Norway records live births at 12 weeks of gestation or more, which the NCHS concludes is effectively the
same as recording all live births since no live births occur before 12 weeks of gestation.
Differences in how live births are recorded may affect international IMR comparisons; however,
it is unlikely that these recording differences would entirely explain the high U.S. IMR or the
variation between the U.S. IMR and those of some European countries. This is because of both
the widespread use of the WHO definition of live births and the small number of births that fall
outside the WHO definition.25 Researchers at NCHS conclude that for recording differences to
completely explain the high U.S. IMR, European countries would have to misreport one-third of
their infant deaths, which these researchers conclude is unlikely.
23 A 13th European country, Norway, counts live births at 12 weeks of gestation or more, which NCHS concludes is
effectively the same as counting all live births since no live births occur before 12 weeks of gestation.
24 See European Perinatal Health Report at http://www.europeristat.com/bm.doc/european-perinatal-health-report.pdf,
page 40.
25 For example, infants born below 500 grams are rare and account for 0.16% of all U.S. live births (e-mail from
Reproductive Statistics Branch: National Center for Health Statistics, July 21, 2009).
Congressional Research Service
7
The U.S. Infant Mortality Rate
After adjusting for recording differences, NCHS researchers found that the U.S. IMR was still
higher than those in most European countries. Specifically, these researchers excluded deaths that
occurred prior to 22 weeks of gestation. They found that excluding these very short gestational
age births lowered the U.S. IMR to 5.8 (in 2004), but also lowered the IMR of European
countries; therefore, it changed the U.S. ranking compared to European countries by only a small
amount.26 Given this finding, the NCHS researchers conclude that recording differences can
explain only a small percentage of international IMR variation and do not entirely explain why
the U.S. IMR is higher than the rates of a number of European countries.
Rates of Low Birthweight and Short Gestational Age Births
Researchers from the NCHS found that low birthweight and short gestational age births—a
leading cause of infant death internationally27—were more common in the United States than in
Europe.28 Although outcomes for these births were similar, they found that the U.S. rate of low
birthweight and short gestational age births was double that of Finland and Sweden.29 After
adjusting for potential recording differences by excluding very short gestational age births, they
found that the U.S. IMR would be 3.9 if the United States had the same rate of low birthweight
and short gestational age births as Sweden. Given this estimate, the researchers conclude that
reducing the rate of these births would lower the U.S. IMR.30 OECD researchers have also
suggested that the increasing rates of low birthweight births may explain the recent stagnation in
the U.S. IMR.31
Racial Disparities
The U.S. IMR differs among racial and ethnic groups, and these differences may contribute to the
high U.S. IMR. Specifically, in 2008, the IMR for infants born to black mothers was 12.7, more
than double the white IMR of 5.5. This difference has the effect of increasing the U.S. IMR, as
births to black mothers make up 16% of U.S. births, but 30.4% of U.S. infant deaths in 2008.32 In
contrast, the U.S. IMR for white infants was 5.5.33 This rate is closer to the Canadian IMR of 5.6;
however, it is still higher than the OECD average of 4.6 and the IMRs of other English-speaking
countries such as the United Kingdom and New Zealand (IMR of 4.7, rank of 23), and Australia
(IMR of 4.1, rank of 22).34
26 In the NCHS recalculated rates, the U.S. IMR was higher than most European countries with the exception of
Hungary, Poland, and Slovakia. The NCHS analysis excluded European countries that had unadjusted IMRs that were
higher than the United States.
27 See Table 1 and Christopher P. Howson et al., White Paper on Preterm Birth: The Global and Regional Toll, March
of Dimes, White Plains, NY, September 21, 2009.
28 Behind International Rankings of Infant Mortality.
29 Ibid.
30 Ibid.
31 Organization for Economic Cooperation and Development, OECD Health Data, 2010, http://www.oecd.org/
dataoecd/4/36/46796773.pdf. Data from 2008 suggest that rates of low birthweight births have decreased; a change
from prior years. See Joyce A. Martin, et al., Births: Final Data for 2008, National Center for Health Statistics, National
Vital Statistics Reports vol. 59, no.1, Hyattsville, MD, December 8, 2010; hereinafter, Births: Final Data for 2008.
32 Deaths: Final Data for 2008 and Births: Final Data for 2008.
33 Deaths: Final Data for 2008. Non-Hispanic white infants make up 53% of U.S. births and 45% of U.S. infant deaths.
34 See Table 2.
Congressional Research Service
8
The U.S. Infant Mortality Rate
These comparisons provide some evidence that racial disparities raise the U.S. IMR. Eliminating
these disparities would likely lower the U.S. IMR, but would not likely lower it below the OECD
average, or below those countries with the lowest IMRs (those in Scandinavia).
Geographic Variation in U.S. Infant Mortality
There is large variation in IMRs among U.S. states. This variation may reflect differences in
demographic and health system factors associated with infant death (as will be discussed further
below; see “Factors That Influence the U.S. IMR”). Specifically, factors that may explain
geographic variation in IMRs include (1) rates of low birthweight and short gestational age births
and (2) the racial composition of the state’s population. This section examines state-level
geographic variation in the U.S. IMR.
In 2008, state IMRs ranged from a low of 3.95 in New Hampshire to a high of 10.84 in the
District of Columbia (see Table 4).35 Figure 1 shows that 26 states (and the District of Columbia)
are above the national average IMR of 6.61. Infant mortality rates are generally highest in the
southern states, including Mississippi, Alabama, and Louisiana.36 The higher IMRs in these states
may be explained, in part, by demographic and health system characteristics of these states. For
example, southern states have high poverty and uninsurance rates.37 The opposite is generally true
in states with low IMRs, such as those in the New England and the Pacific Northwest.38
35 Deaths: Final Data for 2008.
36 Deaths: Final Data for 2008, and Charles E. Menifield and Jacob Dawson, “Infant Mortality in Southern States: A
Bureaucratic Nightmare,” Journal of Health and Human Services, vol. 31, no. 3 (Winter 2008), pp. 385-402.
37 Ibid.
38 Kaiser Family Foundation State Health Facts, http://www.statehealthfacts.org. California appears to be an exception
to this general trend, because the state has a low IMR, but a high poverty rate. This may be explained by the state’s
large Hispanic population because the IMR for infants born to Hispanic mothers is lower than the overall U.S. IMR
(see “Race and Ethnicity”).
Congressional Research Service
9
The U.S. Infant Mortality Rate
Figure 1. U.S. IMR (Infant Deaths/1,000 Live Births), by State, 2008
Source: Adapted by CRS from Ariadi M. Minino, et al., National Vital Statics Reports: Deaths: Final Data for 2008,
National Center for Health Statistics, Vol. 59, No. 10, Hyattsville, MD, December 7, 2011.
Factors That May Contribute to State Variation in IMR
Geographic variation in the U.S. IMR results from, among other factors, variation in rates of
specific causes of infant death and from variation in demographic factors associated with infant
death. This section discusses variation in rates of low birthweight and short gestational age births
and how different state rates of these births relate to IMR variation. It then discusses how
variation in the racial composition of the state’s population—specifically, variation in the relative
size of a state’s black39 population—affects state IMR variation.
State Variation in Low Birthweight and Short Gestational Age Births
Geographic variation in the U.S. IMR results from different rates of infant death by state. As
discussed above, low birthweight and short gestational age are a leading cause of infant death in
the United States. Rates of low birthweight births (data not presented) vary by state, ranging from
11.8% in Mississippi (where the IMR is 10.0) to 6.0% in Alaska (where the IMR is 5.9). Rates of
39 The NCHS data used compare the U.S. “black” and U.S. “white” populations.
Congressional Research Service
10
The U.S. Infant Mortality Rate
low birthweight births influence state IMRs, as states with high IMRs, such as Mississippi and
Alabama, generally have high percentages of low birthweight births. The opposite is also true, as
states with low percentages of low birthweight births, such as New Hampshire and Utah have low
IMRs.40
Table 4 shows state variation in rates of short gestational age births and its impact on state IMRs.
With relatively few exceptions,41 states that have high percentages of short gestational age births
have high IMRs. The percentage of short gestational age births relative to total births in a state
varies from 18.0% in Mississippi (where the IMR is 10.0) to 9.5% in Vermont (where the IMR is
4.6).
Table 4. Percent and Rank of State IMR (Infant Deaths/1,000 Live Births) and Short
Gestational Age Births, 2008
Short
Gestational
Age Births as
Percentage of
State IMR Rank
Total Births
Rank
New
Hampshire
4.0 1 9.6 2
Vermont
4.6 2 9.5 1
Utah 4.8 3
11.0 12
California 5.1
4
10.5
9
Massachusetts 5.1
4
10.8
11
Oregon 5.2
6
10.1
5
Nebraska 5.4
8
11.8
23
Nevada 5.3
7
13.5
41
Washington 5.4
8
10.7
10
Maine 5.5
10
10.3 6
New York
5.5
10
12.0
25
Hawai 5.5
12
12.8 33
New Jersey
5.6
13
12.5
29
New Mexico
5.6
13
12.3
26
Iowa 5.7 15 11.5 20
North Dakota
5.8
16
11.1
13
Alaska 5.9
17
10.3 6
Rhode Island
5.9
17
11.2
15
Idaho
5.9 19 9.8 3
Connecticut 6.0
20
10.4
8
40 Births: Final Data for 2008, data are not shown in Table 4.
41 The exceptions are generally states with small populations, such as Alaska and Iowa.
Congressional Research Service
11
The U.S. Infant Mortality Rate
Short
Gestational
Age Births as
Percentage of
State IMR Rank
Total Births
Rank
Minnesota 6.0
20
10.0
4
Colorado 6.2
22
11.4
19
Texas 6.2
22
13.3 38
Arizona 6.4
24
12.9
34
U.S.
6.6
—
12.3
—
Montana 6.8
25
11.5
20
Indiana 6.9
26
12.4 28
Kentucky 6.9
26
14.0
46
Virginia 6.9
26
11.3
18
Wisconsin 7.0
29
11.1
13
Wyoming 7.0
29
11.2
15
Illinois 7.1
31
12.7 31
Florida 7.2
32
13.8 45
Missouri 7.2
32
12.3
26
Kansas 7.3
34
11.2 15
Oklahoma 7.3
34
13.4
39
Arkansas 7.4
36
13.5
41
Michigan 7.4
36
12.7
31
Pennsylvania 7.4
36
11.6
22
Ohio 7.7 39 12.6 30
West Virginia
7.7
39
13.7
44
Maryland 8.0
41
13.0
37
South Carolina
8.0
41
14.3
47
Georgia 8.1
43
13.4
39
Tennessee 8.1
43
13.5
41
North Carolina
8.2
45
12.9
34
Delaware 8.4
46
12.9
34
South Dakota
8.4
46
11.9
24
Louisiana 9.1
48
15.4
48
Alabama 9.5
49
15.7
50
Mississippi
10.0 50 18.0 51
District of
Columbia
10.8 51 15.5 49
Source: Joyce A. Martin, et al., Births: Final Data for 2008, National Center for Health Statistics, National Vital
Statistics Reports vol. 59, no.1, Hyattsville, MD, December 8, 2010, and Ariadi M. Minino, et al., National Vital
Congressional Research Service
12
The U.S. Infant Mortality Rate
Statics Reports: Deaths: Final Data for 2008, National Center for Health Statistics, Vol. 59, No. 10, Hyattsville, MD,
December 7, 2011. Hereinafter Deaths: Final Data 2008.
Notes: States with the same IMR or percentage short gestational age births receive the same rank. The state
with the next highest IMR/percentage of short gestational age births is assigned the rank it would have received
had the lower-ranked states not been tied, (i.e., it skips a rank). For example, Massachusetts and California were
both assigned a rank of 4. Oregon, which has the next lowest IMR, is assigned the rank of 6.
State Variation in Population Composition
Differences in state IMR may also, in part, reflect the racial composition of a state’s population
because of racial differences in the IMR and specifically, higher IMRs for infants born to black
mothers. State IMR for infants born to black mothers ranges from 17.1 in Arizona to 9.4 in
Massachusetts.42 The IMR range for infants born to white mothers among states ranges from 3.35
in Alaska to 7.6 in Alabama.43 Thus, states with a larger percentage of the population that is white
have lower IMRs, whereas the opposite is true in states, such as those in the south, that have
larger percentages of the population that are black.44 Specifically, in Washington, DC—the
jurisdiction with the highest IMR—51% of the population is black; in contrast, in New
Hampshire—the state with the lowest IMR—1% of the population is black45 (for further
discussion of race and ethnicity, see “Factors That Influence the U.S. IMR,” “Mother’s
Demographic Characteristics,” and “Race and Ethnicity”).
Factors That Influence the U.S. IMR
As previously discussed, the U.S. IMR is high due, in part, to high rates of low birthweight and
short gestational age births. The high rates of both of these conditions are also linked to other
factors that can be roughly grouped into two categories: (1) demographic characteristics of the
mother such as race, education, and economic status, and (2) health and health system
characteristics, such as the mother’s health and health behaviors, and the mother’s access to and
receipt of prenatal care. Other causes of infant death, such as the rate of SIDS—the third-leading
cause of infant death in 2008—may also vary by demographic characteristics. In addition,
demographic characteristics may be correlated with each other, and with health and health system
characteristics. For example, individuals who have low incomes may have inadequate access to
prenatal care. These correlations make it difficult to examine the relative contribution of specific
factors to the IMR. Consequently, though this report discusses these factors, it cannot
conclusively attribute causality to any single factor or group of factors. Examination of other
potential contributors to the IMR, such as environmental pollutants, is beyond the scope of this
report.
42 Deaths: Final Data for 2008, see Table 22. The national average black IMR is 12.7. A number of states do not have
sufficient data to calculate black IMR. Data in this table were aggregated into the racial categories of “black” and
“white.”
43 Ibid.
44 Charles E. Menifield and Jacob Dawson, “Infant Mortality in Southern States: A Bureaucratic Nightmare,” Journal
of Health and Human Services, vol. 31, no. 3 (Winter 2008), pp. 385-402.
45 Kaiser Family Foundation State Health Facts, http://www.statehealthfacts.org.
Congressional Research Service
13
The U.S. Infant Mortality Rate
Mother’s Demographic Characteristics
A number of demographic characteristics of expectant mothers may be correlated with the
likelihood of having an infant die within the first year of life. Among those that researchers often
consider are race, ethnicity, education, socioeconomic status, and age.
Race and Ethnicity46
Infant mortality rates vary by race and ethnicity. According to 2008 NCHS data, the IMR for
infants born to white mothers (5.5) is almost half of the rate for infants born to black mothers
(12.7). The overall IMR for infants born to Hispanic mothers (5.3) lower than the overall rate for
the United States; however, this overall IMR masks variation among Hispanic subgroups. The
IMR for infants born to Hispanics that are not classified (“other and unknown Hispanic”) is 13.5,
which is more than double the U.S. IMR average of 6.6. The IMRs for infants born to mothers of
Mexican (5.1), Puerto Rican (6.1), and Central and South American descent (3.3) are lower than
overall U.S. IMR.47 The lower IMR for infants born to Hispanic mothers is considered by some to
be an epidemiological paradox because some Hispanic groups (e.g., Mexican Americans)
experience the same disadvantaged socioeconomic conditions as many blacks, yet have lower
rates of infant mortality.48 These racial and ethnic disparities in the IMR may reflect demographic
or socioeconomic characteristics of the mother, health and health system characteristics, or other
characteristics of the mother or her environment.
The high average IMR for non-Hispanic blacks can be partly explained by higher rates of low
birthweight49 and short gestational age births,50 which, in turn, may be explained by demographic
characteristics of the mother. Demographic characteristics such as being in poverty, having less
than a high school education, or being unmarried are more common in the non-Hispanic black
population, and these characteristics are correlated with having a low birthweight or short
gestational age birth. Rates of SIDS are higher among non-Hispanic black and American
Indian/Alaska Native infants. Research suggests that low rates of preventive behaviors—such as
entering prenatal care early in pregnancy, taking folic acid supplements, and completing
childhood vaccinations—may explain the elevated IMR among infants born to non-Hispanic
46 This report uses the same racial and ethnic categories used by NCHS. How the federal government collects racial and
ethnic data has changed over time. As of 2003, federal agencies were required to collect data on race and ethnicity
separately and were required to collect data on individuals of one or more race. Office of Management and Budget,
“Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity,” 62 Federal Register 36873-
36946, July 9, 1997. In some cases, data on race and ethnicity are not collected separately. This is generally done to
maintain comparable data over time. This CRS report will use the same racial and ethnic categories as the source data
described, which may vary from the Office of Management and Budget requirements.
47 Deaths: Final Data for 2008.
48 Robert A. Hummer et al., “Paradox Found (Again): Infant Mortality Among the Mexican-Origin Population in the
United States,” Demography, vol. 44, no. 3 (August 2007), pp. 441-457.
49 Births: Final Data for 2008.
50 See, for example, Anjel Vahratian, Pierre Buekens, and Greg R. Alexander, “State-Specific Trends in Preterm
Delivery: Are Rates Really Declining Among Non-Hispanic African American Across the United States,” Maternal
Child Health, vol. 10, no. 1 (January 2006), pp. 27-32, and Michael Kramer and Carol R. Hogue, “Place Matters:
Variation in the Black/White Very Preterm Birth Rate Across U.S. Metropolitan Areas, 2002-2004,” Public Health
Reports, vol. 123, no. 5 (September/October 2008), pp. 576-585.
Congressional Research Service
14
The U.S. Infant Mortality Rate
black women.51 Research also suggests that high rates of smoking and alcohol abuse may explain
the elevated IMR among infants born to American Indian/Alaska Native women.52
Education
Studies have shown that women with higher levels of education are less likely to give birth to an
infant who dies within the first year of life.53 Researchers at the Robert Wood Johnson Foundation
found that infants born to mothers with at least a high school education are less likely to die in
their first year of life than are infants born to mothers who have not completed high school. These
gaps are more dramatic in some states; for example, the researchers found the widest gap in South
Carolina, where infants born to mothers who were not high school graduates were twice as likely
to die as infants born to mothers who had graduated.54
Education may influence the IMR in a variety of ways. Education is often a pathway to better
employment opportunities and higher incomes, so it may indirectly affect health status through
income.55 It may also affect the IMR through marital status, as college-educated mothers are more
likely to be married, and the IMR is lower for infants born to married mothers. Lower educational
attainment may also be correlated with higher rates of certain negative health behaviors, such as
tobacco and alcohol consumption, which may increase the risk of having an infant die, or of
having conditions linked with infant death. Researchers have found that declining smoking rates
among less-educated women have reduced infant mortality. Conversely, increasing rates of
excessive weight gain among less-educated women during pregnancy have offset some of these
gains. More highly educated women are also less likely to smoke and are more likely to gain the
appropriate amount of weight during pregnancy.
Some recent trends may be altering the relationship between education and the IMR. More highly
educated women are more likely to be older mothers (over age 40), have multiple births, and use
fertility treatments. All of these factors may increase infant mortality, as discussed below (see
“Age”).
Economic Status
Women with higher incomes are less likely to have an infant die. The inverse relationship
between economic status and the IMR is part of the larger finding, in general, that people with
higher incomes are in better health.56 Economic status may also be related to access to health
51 U.S. Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child
Health Bureau 2006, Evidence of Trends, Risk Factors, and Intervention Strategies, A Report from the Healthy Start
National Evaluation 2006: Racial and Ethnic Disparities in Infant Mortality, Rockville, MD.
52 Ibid.
53 See, for example, Rebecca Din-Dzietham and Irva Hertz-Picciotto, “Infant Mortality Differences between Whites
and African Americans: The Effects of Maternal Education,” American Journal of Public Health, vol. 88 (1998), pp.
651-55.
54 Robert Wood Johnson Foundation, “New State-by-State Report Finds Shortfalls in Children’s Health Tied to
Parents’ Income and Education,” press release, October 8, 2008, http://www.rwjf.org/pr/product.jsp?id=35208.
55 This paragraph is drawn from John Mirowsky and Catherine E. Ross, Education, Social Status, and Health (New
York: Aldine De Gruyter, 2003) and Wanchaun Lin, “Why Has the Health Inequality Among Infants in the US
Declined? Accounting for the Shrinking Gap,” Health Economics, vol. 18 (September 24, 2008), pp. 823-841.
56 See, for example, Brian Karl Finch, “Early Origins of the Gradient: The Relationship Between Socioeconomic Status
and Infant Mortality in the United States,” Demography, vol. 40, no. 4 (November 2003), pp. 675-699.
Congressional Research Service
15
The U.S. Infant Mortality Rate
insurance and to the quality of health care received, and can affect infant health both directly and
indirectly through health and health system characteristics.57 Researchers have found that
increasing access to health care for women at all income levels has served to narrow the income
gap in infant mortality.58
Age
The changing age composition of women giving birth may increase the average U.S. IMR.59
Births to women ages 40 to 44 have increased since 1990, which may increase the IMR.60 Women
over age 30 are more likely to have multiple births. This occurs both because of the use of
fertility treatments and because the likelihood of twins increases with age.61 Infants of multiple
births are more likely to die and to be both of low birthweight and short gestational age.62 In
addition to the increased risk of multiple births, increasing maternal age may raise the IMR
because the risk of a number of health conditions, including birth defects and congenital
abnormalities, increases with maternal age.63
Health and Health System Characteristics
A number of characteristics of an expectant mother’s health and the health care she receives could
influence infant mortality. Such characteristics include whether the mother used fertility
treatments (called assisted reproductive technology, or ART) to conceive,64 the mother’s health
and health behaviors, and the amount and timing of prenatal care received.65 These characteristics
may be interrelated and may also be influenced by the expectant mother’s demographic
characteristics. In addition, health system characteristics may affect infant mortality and the IMR.
These factors include access to care and quality of care received.
Assisted Reproductive Technology
Increased use of ART may raise the IMR through increasing rates of multiple births, which may
increase the percentages of infants born at low birthweight or short gestational age.66 ART may
also increase low birthweight births among single births. The CDC found, in a study of single
births, that a 2% increase in ART births resulted in a 7% increase in low birthweight births.67 ART
57 Ibid.
58 Wanchaun Lin, “Why Has the Health Inequality Among Infants in the US Declined? Accounting for the Shrinking
Gap,” Health Economics, vol. 18 (September 24, 2008), pp. 823-841.
59 More discussion of recent changes in fertility can be found at Births: Final Data for 2008.
60 For example, in 2008, the birthrate for women over age 40 for was the highest report in more than 40 years.
61 See http://www.marchofdimes.com/14332_1155.asp.
62 Births: Final Data for 2008.
63 See http://www.marchofdimes.com/14332_1155.asp.
64 Centers for Disease Control and Prevention, “Assisted Reproductive Technology and Trends in Low Birthweight—
Massachusetts, 1997—2004,” Morbidity and Mortality Weekly, vol. 58, no. 03 (January 30, 2009).
65 Charles E. Menifield and Jacob Dawson, “Infant Mortality in Southern States: A Bureaucratic Nightmare,” Journal
of Health and Human Services, vol. 31, no. 3 (Winter 2008), pp. 385-402.
66 See http://www.marchofdimes.com/professionals/14332_4545.asp and http://ww2w.marchofdimes.com/prematurity/
21326_1157.asp.
67 Centers for Disease Control and Prevention, “Assisted Reproductive Technology and Trends in Low Birthweight—
(continued...)
Congressional Research Service
16
The U.S. Infant Mortality Rate
may also increase short gestational age births, and one study by the March of Dimes Foundation
attributed part of the 25-year, 36% increase in short gestational age births in the United States to
the increasing use of ART.68 The increase in ART births may confound the relationship between
the mother’s demographic characteristics and the IMR, because ART is more commonly used by
women who would otherwise be less likely to have an infant die—that is, women who are older,
more educated, and more financially secure.69
Health and Health Behaviors
Since 1990, a number of maternal lifestyle and health characteristics that may raise the IMR have
increased. These include inappropriate weight gain during pregnancy,70 tobacco use, gestational
diabetes, and hypertension. There has been a 30% increase in women gaining too much weight
during pregnancy and a 50% increase in women gaining too little weight.71 Weight gains outside
of the recommended guidelines have been associated with a number of adverse outcomes,
including infant mortality.72 Research has found that tobacco use during pregnancy increases rates
of low birthweight and short gestational age births, and the rate of SIDS.73 Rates for gestational
diabetes have increased 3% per year on average since 1990, and rates of both pregnancy-
associated and chronic hypertension have increased by an average of 1% annually since 1990.
These conditions are associated with greater risk of pregnancy complications and adverse
outcomes.74
Health behaviors following pregnancy, such as breastfeeding and the duration between
pregnancies, can affect the IMR, and differences in these health behaviors may partially explain
racial and ethnic IMR differences. From 2000 to 2007, breastfeeding rates have increased;
however, there are racial and ethnic differences in breastfeeding rates. Non-Hispanic black
women have consistently lower rates of breastfeeding when compared to white or Hispanic
women. 75 Short duration between pregnancies (less than six months) increases the risk of infant
(...continued)
Massachusetts, 1997—2004,” Morbidity and Mortality Weekly, vol. 58, no. 03 (January 30, 2009).
68 The March of Dimes Foundation. Christopher P. Howson, Mario Merialdi, and Joy E. Lawn, et al., White Paper on
Preterm Birth: The Global and Regional Toll, March of Dimes, White Plains, NY, September 21, 2009. The 25-year
period was from 1985 to 2009. The authors cannot determine what percentage of the 36% increase was due to ART
because the time period corresponds with other factors that may increase short gestational age births. These factors
include advances in neonatal technology, increases in the percentage of women giving birth over age 35, and changes
in clinical practice that might contribute to early induction of labor or performing Cesarean sections prior to full-term.
69 Elizabeth Harvey Stephen and Anjani Chandra, “Use of Infertility Services in the United States: 1995,” Family
Planning Perspectives, vol. 32, no. 3 (May/June 2000).
70 Recommended weight gain during pregnancy is between 16 and 40 lb., see Births: Final Data for 2008.
71 Unless otherwise noted, material in this section is drawn from Births: Final Data for 2006.
72 National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health & Human Development,
“Pregnancy and Healthy Weight,” press release, April 7, 2010, http://www.nichd.nih.gov/news/resources/spotlight/
040710-pregnancy-healthy-weight.cfm.
73 TJ Mathews, “Smoking During Pregnancy—United States, 1990—2002,” MMWR, vol. 53, no. 39, pp. 911-915.
From 1990 to 2002, the rate of women smoking during pregnancy appears to have declined; however, between 2002
and 2006 the rate appears to have increased (from 11.4% to 13.2%). This apparent increase may be the result of
methodological changes in how these data were collected.
74 Elizabeth Baraban, Lucie McCoy, and Paul Simon, “Increasing Prevalence of Gestational Diabetes and Pregnancy-
Related Hypertension in Los Angeles County, California, 1991–2003,” Preventing Chronic Disease, vol. 5, no. 2 (July
2008), p. http://www.cdc.gov/pcd/issues/2008/.
75 Centers for Disease Control and Prevention, “Breastfeeding Among U.S. Children Born 2000-2008, CDC National
(continued...)
Congressional Research Service
17
The U.S. Infant Mortality Rate
death, of being born at a low birthweight or short gestational age, and of dying of SIDS. Duration
between pregnancy varies by racial and ethnic groups, with non-Hispanic blacks and American
Indian/Alaska Native women having shorter durations between pregnancies.76
Prenatal and Preconception Care
Prenatal care has been demonstrated to be an effective intervention to reduce infant mortality
because it includes health care as well as education and counseling about how to handle various
aspects of pregnancy, including nutrition, physical activity, and basic infant care skills.77 Women
who did not receive any prenatal care are more likely to have an infant die in the first month after
birth. In addition, research has found that prenatal care reduces the incidence of low birthweight
and short gestational age births, and of deaths due to accidents or SIDS.78
Delayed prenatal care may increase infant mortality. Prenatal care may be particularly important
at earlier points in the pregnancy because health education to encourage appropriate weight gain
or to discourage alcohol or tobacco use will have more effect on the fetus at early stages of the
pregnancy, when the fetus is developing most rapidly. For example, researchers found that young
women who initiated prenatal care early in pregnancy were less likely to use alcohol, cigarettes,
or marijuana.79 Prenatal care receipt may explain some observed racial differences in IMR, since
non-Hispanic black mothers are 2.5 times more likely to begin prenatal care in the third trimester,
or to receive no prenatal care.80
Some researchers and policymakers suggest that prenatal care should begin prior to pregnancy;
this is called “preconception care.” These researchers suggest that such care can ensure that the
to-be expectant mother receives the proper vitamins and vaccinations, and that she engages in
healthy behaviors such as exercising and abstaining from smoking, drinking, and drugs.81 Some
(...continued)
Immunization Survey,” August 1, 2011, http://www.cdc.gov/breastfeeding/data/NIS_data/index.htm.
76 U.S. Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child
Health Bureau, Evidence of Trends, Risk Factors, and Intervention Strategies, A Report from the Healthy Start
National Evaluation 2006: Racial and Ethnic Disparities in Infant Mortality, Rockville, MD.
77 Eunice Kennedy Shriver National Institute of Child Health and Human Development. “Care Before and During
Pregnancy—Prenatal Care.” http://www.nichd.nih.gov/womenshealth/research/pregbirth/prenatal_care.cfm.
78 See, for example, Prenatal Care: Reaching Mothers, Reaching Infants, ed. Sarah S. Brown (Washington, DC:
National Academy Press, 1988) and Greg R. Alexander and Milton Kotelchuck, “Assessing the Role and Effectiveness
of Prenatal Care: History, Challenges, and Directions for Future Research,” Public Health Reports, vol. 116
(July/August 2001), pp. 306-316.
79 Joyce C. Abma and Frank L. Mott, “Substance Use and Prenatal Care During Pregnancy Among Young Women,”
Family Planning Perspectives, vol. 23, no. 3 (May/June 1991), pp. 117-128.
80 U.S. Department of Health and Human Services, The Office of Minority Health, “Infant Mortality and African
Americans,” factsheet, July 22, 2011, http://www.minorityhealth.hhs.gov/templates/content.aspx?lvl=3&lvlID=8&ID=
3021. However, even in cases where non-Hispanic black mothers began primary care at the beginning of pregnancy,
the IMR is higher for babies born to these women. This suggests that increasing access to prenatal care for non-
Hispanic black pregnant women will not be sufficient to decrease racial disparities in the IMR. There are a number of
reasons why prenatal care may be less effective for non-Hispanic black pregnant women, such as differences in the
quality of health care received by non-Hispanic blacks as compared to other pregnant women. See Andrew J. Healy et
al., “Early Access to Prenatal Care: Implications for Racial Disparity in Perinatal Mortality,” Obstetrics & Gynecology,
vol. 107, no. 3 (March 2006), pp. 625-631.
81 Eunice Kennedy Shriver National Institute of Child Health and Human Development. “Care Before and During
Pregnancy—Prenatal Care,” http://www.nichd.nih.gov/womenshealth/research/pregbirth/prenatal_care.cfm.
Congressional Research Service
18
The U.S. Infant Mortality Rate
experts have suggested that prenatal care occurs too late to effectively improve the health of the
expectant mother or her infant. This may occur because many women have health conditions such
as obesity or asthma prior to pregnancy, or because women typically do not visit a doctor until 6
to 12 weeks after conception.82 Preconception care, in addition to prenatal care, may reduce infant
mortality by improving the health and health behaviors of the expectant mother, thereby reducing
the risks of a number of causes of infant mortality.83
Quality of Care
Variation in the IMR by race, education, and socioeconomic status may be related to both the
quality of prenatal care and the quality of care at delivery. Specifically, the hospital at which an
infant is delivered may affect health outcomes, particularly for low birthweight or short-
gestational-age infants, who may require specialized care such as neonatal intensive care.84
Researchers have found that black infants were more likely to be born at hospitals with higher
overall mortality rates. These researchers conclude that if black infants were to be born at the
same hospitals as white infants, mortality rates for black infants would be lower.85 Some
researchers have also found that hospital-level improvements, such as the adoption of health
information technology that may standardize treatment and improve monitoring, reduce the
number of infant deaths and improve the IMR.86 In addition, insurance status, which is related to
socioeconomic characteristics, may influence the quality of care received. For example, women
enrolled in Medicaid—a joint federal and state program that provides health insurance coverage
to certain low-income pregnant women and infants—are more likely to give birth at public
hospitals that may be of lower quality than the hospitals used by women with private insurance.87
Access to Care
A pregnant woman’s access to health care is also important for reducing infant mortality. The
availability of physicians, particularly those providing primary care, is associated with lower
infant mortality.88 In a study examining the factors associated with the higher rates of infant
mortality in southern states, the authors found that state IMR decreased when the number of
doctors and hospitals per capita increased. The authors found that states with larger total
uninsured populations had higher IMRs. A large uninsured population89 may increase the IMR
82 Jeffrey Levi, Marlene Cimons, and Kay Johnson, Healthy Women, Healthy Babies, An Issue Brief from Trust for
America’s Health, Washington, DC, June 2008.
83 Ibid.
84 Elizabeth A. Howell, et al., “Black/White Differences in Very Low Birth Weight Neonatal Mortality Rates Among
New York City Hospitals,” Pediatrics, vol. 121, no. 3 (March 2008), pp. e407-e415 and Leo S. Morales et al.,
“Mortality Among Very Low-Birthweight Infants in Hospitals Serving Minority Populations,” American Journal of
Public Health, vol. 95, no. 12 (December 2005), pp. 2206-2212.
85 Ibid.
86 Amalia R. Miller and Catherine E. Tucker, “Can Health IT Save Babies?” MIT Sloan Research Paper, vol. 4686-08
(January 1, 2008).
87 Nancy E. Moss and Karen Carver, “The Effects of WIC and Medicaid on Infant Mortality in the United States,”
American Journal of Public Health, vol. 88, no. 9 (September 1998), pp. 1354-1361.
88 Unless otherwise noted, this paragraph is drawn from a literature review by Barbara Starfield, Leiyu Shi, and James
Macinko, “Contribution of Primary Care to Health and Health Systems,” The Milbank Quarterly, vol. 83, no. 3 (2005),
pp. 457-502.
89 See for example, data from the Health Care Cost & Utilization Project (HCUP) from Agency for Health Care
(continued...)
Congressional Research Service
19
The U.S. Infant Mortality Rate
because uninsured pregnant women are less likely to receive prenatal care and report having
greater difficulty obtaining prenatal care than do insured women.90 In general, insurance status,
which is related to access to care and the quality of care received, may affect infant mortality.
Researchers have found that IMRs are the lowest for infants born to women enrolled in private
insurance, that IMRs are higher for women enrolled in Medicaid, and that IMRs are highest for
infants born to women who were uninsured.
Researchers also have found that access to primary care can influence the national IMR. In
general, countries with more primary care services available have lower national IMRs. In
addition, countries that have implemented health reforms to increase primary care access have
lower IMRs after implementation.
Federal Programs That May Reduce Infant Mortality
Research has shown that some federal programs may reduce the IMR by improving the financial
status or the health of low-income women and infants, who, as discussed above, may be at higher
risk of having an infant die within the first year of life.91 It is also speculated that other programs
that focus on providing income and material support, improving health or health behaviors, or
increasing access to health services may also reduce infant mortality. This section describes some
of these federal programs; it then briefly summarizes some of the research evaluating the effects
of federal programs on reducing infant mortality, and discusses some of the challenges associated
with assessing program effectiveness.
Income and Material Support Programs
Income and material support programs may reduce the IMR by providing direct financial or
material resources to low-income families. These programs include the Temporary Assistance for
Needy Families program (TANF),92 which, through block grants to states, funds a wide range of
benefits (including cash), services, and activities; and two nutrition programs—the Women,
Infants, and Children program (WIC) and the Supplemental Nutrition Assistance Program (SNAP,
formerly the Food Stamps program).93 In addition, federal programs that provide material
(...continued)
Research and Quality (AHRQ), http://hcupnet.ahrq.gov/; findings from an analysis of births in Hawaii covered by
Medicaid in A. Schempf, D. Hayes, and L. Fuddy, Medicaid/QUEST Birth Outcome Fact Sheet: Hospital Discharge
Data, Hawaii Health Information Corporation, Hawaii State Department of Health, Family Health Services Division,
Honolulu, HI, October 2008 and Louie Albert Woodbright, Method of Payment for Delivery, Alabama Department of
Public Health, Montgomery, AL, July 2007, http://www.adph.org.
90 Jeffrey Levi, Marlene Cimons, and Kay Johnson, Healthy Women, Healthy Babies, An Issue Brief from Trust for
America’s Health, Washington, DC, June 2008.
91 See “Do Federal Programs Reduce Infant Mortality?”
92 CRS Report RL32760, The Temporary Assistance for Needy Families (TANF) Block Grant: Responses to Frequently
Asked Questions, by Gene Falk.
93 CRS Report R42353, Domestic Food Assistance: Summary of Programs, by Randy Alison Aussenberg and Kirsten J.
Colello and CRS Report R41354, Child Nutrition and WIC Reauthorization: P.L. 111-296, by Randy Alison
Aussenberg.
Congressional Research Service
20
The U.S. Infant Mortality Rate
resources to low-income families, such as access to low-income housing and financial assistance
with heating costs, may also improve infant health and reduce the IMR.94
Health Programs
Various agencies within the Department of Health and Human Services (HHS) sponsor programs
to reduce infant mortality. This section describes some of these programs, including programs that
sponsor public health education, promote access to prenatal and infant care, and sponsor research
to reduce the IMR.95 The majority of these programs do not focus on reducing infant mortality.
Rather, reducing infant mortality may be one of many program goals or may be included in an
overall goal of improving health. One program, Healthy Start, focuses explicitly on reducing
infant mortality, and another, Maternal and Child Health Services Block Grants,96 focuses on
improving maternal and child health.
Healthy Start97
The Healthy Start Program is the only federal program that specifically aims to lower the IMR by
reducing racial and ethnic IMR disparities. The program, administered by the Health Resources
and Services Administration (HRSA), an agency within HHS, awards grants to non-profit
organizations, state and local health departments, and other entities to support projects to improve
maternal and child health in communities with high IMRs. In addition to addressing the
community IMR, Healthy Start sites focus on intermediate goals that may also reduce the IMR,
such as decreasing the percentage of low birthweight births and increasing prenatal care use
among program recipients. The program seeks to achieve these goals by both direct services—
home visits and case management—to pregnant women and by reducing gaps in health care and
support services available community-wide. In order to address community-wide gaps in
services, Healthy Start programs are required to coordinate with other federal health programs
(described below), including Maternal and Child Health Services Block Grants, Medicaid, and
the State Children’s Health Insurance Program (CHIP). This coordination attempts to avoid
duplication and to ensure that Healthy Start clients are enrolled in the programs for which they
are eligible. Healthy Start sites may also offer other services, such as transportation to medical or
other appointments, smoking cessation services, and programs to increase male involvement in
pregnancy and parenting.
94 CRS Report R41654, Introduction to Public Housing, by Maggie McCarty and CRS Report RL33275, The LIHEAP
Formula: Legislative History and Current Law, by Libby Perl.
95 Preventing Infant Mortality. The discussion below provides examples of programs that HHS administers; other HHS
programs not discussed may also prevent infant mortality.
96 CRS Report R42428, The Maternal and Child Health Services Block Grant: Background and Funding, by Amalia K.
Corby-Edwards.
97 This section is based on the following sources, U.S. General Accounting Office, Healthy Start: Preliminary Results
from National Evaluation are Not Conclusive, HEHS-98-167, June 1998; Barbara Devaney et al., Reducing Infant
Mortality Lessons Learned from Healthy Start, Mathematica Policy Research, Inc., Final Report, Princeton, NJ, July
2000; White House, Office of Management and Budget, Detailed Information on the Healthy Start Assessment,
Washington, DC, http://www.whitehouse.gov/omb/expectmore/; United States Department of Health and Human
Services, Health Resources and Services Administration, Maternal and Child Health Bureau, A Profile of Health Start:
Findings From Phase I of the Evaluation 2006, Rockville, MD, 2006, http://mchb.hrsa.gov/healthystart/phase1report/;
and U.S. Department of Health and Human Services, Health Resources and Services Administration, Maternal and
Child Health Bureau, Evidence of Trends, Risk Factors, and Intervention Strategies, A Report from the Healthy Start
National Evaluation 2006: Racial and Ethnic Disparities in Infant Mortality, Rockville, MD.
Congressional Research Service
21
The U.S. Infant Mortality Rate
Maternal and Child Health Services Block Grants
HRSA also administers the Maternal and Child Health Services Block Grant program.98 States
receive block grant funds and have discretion in the specific activities they fund, although grants
are awarded to meet a number of maternal and child health goals, including reducing the state’s
IMR. Block grant funds are used to fund a variety of services, including health, health education,
and prevention services mentioned below (e.g., health screenings and vaccinations). Another
purpose of the program is to coordinate the maternal and child health grant services with the
state’s Medicaid program.99
Health Education Efforts
HHS sponsors a number of health education efforts designed to improve maternal and child
health and thereby reduce the IMR:100
• Toll-free prenatal care line: HHS sponsors a toll-free hotline—in English and
Spanish—that provides information on prenatal care, including referrals to local
clinics or providers. The Center for Medicare & Medicaid Services (CMS) has
also partnered with Text4Babies, which provides health information to pregnant
women to encourage enrollment in Medicaid and the State Children’s Health
Insurance Program (CHIP).101
• SIDS prevention: HHS sponsors a health education campaign to encourage
parents to place infants to sleep on their backs. This campaign—called “Back to
Sleep”—increased the percentage of infants sleeping on their backs from 13% to
76% from 1992 to 2006 and has reduced the number of deaths from SIDS from
1.40 deaths per 1,000 births in 1988 to 0.55 deaths per 1,000 births in 2006.102
• Folic acid campaign: The CDC, in conjunction with private organizations,
promotes the consumption of folic acid among women of child-bearing age in
order to reduce birth defects.
• Preventing mother-to-child HIV transmission: HHS provides information to
pregnant women and to providers on HIV testing and treatment options to reduce
mother-to-child HIV transmission.
98 Preventing Infant Mortality and CRS Report R42428, The Maternal and Child Health Services Block Grant:
Background and Funding, by Amalia K. Corby-Edwards. This program is authorized under Title V of the Social
Security Act.
99 CRS Report R42428, The Maternal and Child Health Services Block Grant: Background and Funding, by Amalia K.
Corby-Edwards.
100 Unless otherwise specified, the information in this section is drawn from Preventing Infant Mortality.
101 Department of Health and Human Services, “Obama Administration and Text4Baby Join Forces to Connect
Pregnant Women and Children to Health Coverage and Information,” press release, February 28, 2012,
http://www.hhs.gov/news/press/2012pres/02/20120228c.html.
102 See http://www.nichd.nih.gov/SIDS/upload/SIDS_rate_back_sleep_2006.pdf released by the Eunice Kennedy
Shriver National Institute of Child Health and Human Development.
Congressional Research Service
22
The U.S. Infant Mortality Rate
Prevention Programs
HHS sponsors a number of prevention programs, which may indirectly lower the IMR. For
example, the CDC, through the Preventive Health and Health Services Block Grant (PHHSBG)
program, provides funding to states to support state-determined prevention efforts.103 The Social
Services Block Grant (SSBG) program—a flexible source of funds for states to support social
service activities104—may also provide funds for prevention efforts. PHHSBG or SSBG funds
may be used to support prevention efforts such as those discussed below:
• Teen pregnancy: HHS supports a number of programs to reduce teen
pregnancy.105 These include health education programs, including both sex
education and abstinence-only education programs. Educational efforts to prevent
teen pregnancy may reduce the IMR, because infants born to teenage mothers are
more likely to die in the first year of life.106
• Family planning: HHS provides block grants to states to support family
planning efforts, such as providing contraception, screening, counseling, and
referral for treatment of sexually transmitted infections, including HIV/AIDS.107
This program may reduce the IMR because research has shown that women who
have unintended pregnancies are more likely to delay prenatal care and to have
adverse outcomes, including low birthweight births and infant deaths.108
• Programs to prevent child abuse and neglect: HHS sponsors a number of
programs to reduce child abuse and neglect. These include programs that provide
education and support to new low-income parents through home visiting by
nurses or other trained health professionals. These programs aim to improve
parenting skills and to reduce infant death by preventing accidents or SIDS.109 In
addition, programs sponsored by the Administration for Children and Families
seek to prevent, identify, and intervene in instances of child abuse and neglect.
Such programs may reduce the IMR because they may prevent accidents or other
unintentional or intentional deaths.110
• Childhood immunization programs: HHS administers the National Vaccine
Program, a coordinated effort among several federal departments and agencies to
prevent infectious diseases through immunization, and to prevent adverse
reactions to vaccines.111 As a result of these efforts, childhood vaccination levels
have increased and deaths from preventable diseases have decreased.112
103See “Preventive Health and Health Services Block Grant” at http://www.cdc.gov/nccdphp/blockgrant/about.htm.
104 CRS Report 94-953, Social Services Block Grant: Background and Funding , by Karen E. Lynch.
105 CRS Report RS20301, Teenage Pregnancy Prevention: Statistics and Programs, by Carmen Solomon-Fears.
106 Births, Final Data, 2008.
107 CRS Report RL33644, Title X (Public Health Service Act) Family Planning Program, by Angela Napili.
108 Deanna L. Pagnini and Nancy E. Reichman, “Psychosocial Factors and the Timing of Prenatal Care Among Women
in New Jersey’s HealthStart Program,” Family Planning Perspectives, vol. 32, no. 2 (March/April 2002), pp. 56-64 and
Prakesh S. Shah et al., “Intention to Become Pregnant and Low Birth Weight and Preterm Birth: A Systematic
Review,” Maternal Child Health Journal, vol. 15 (2011), pp. 205-216.
109 CRS Report R40705, Home Visitation for Families with Young Children, by Emilie Stoltzfus and Karen E. Lynch.
110 CRS Report R40899, The Child Abuse Prevention and Treatment Act (CAPTA): Background, Programs, and
Funding, by Emilie Stoltzfus.
111 HHS, National Vaccine Program Office, http://www.hhs.gov/nvpo/. The National Vaccine Program activities
(continued...)
Congressional Research Service
23
The U.S. Infant Mortality Rate
Research Efforts
HHS also supports health research to prevent infant mortality through the National Institutes of
Health (NIH) and the CDC.113 The Eunice Kennedy Shriver National Institute of Child Health
and Human Development (NICHD) is the primary NIH institute that supports research to prevent
infant mortality. For example, NICHD has supported research demonstrating the effects of
progesterone injections in reducing short gestational age births and has supported projects to
identify risk factors for IMR, such as poor air quality.114 In addition to NICHD, another NIH
institute, the National Heart, Lung, and Blood Institute, has supported research that may reduce
the IMR by improving outcomes for pregnant women with hypertension and improving outcomes
for short gestational age infants with respiratory distress syndrome.115 The CDC also sponsors or
conducts research on infant mortality, such as a study on behavioral and environmental factors
related to short gestational age births among African American women.116
Health Care Access Efforts
HHS sponsors programs to increase access to health services for low-income pregnant women or
low-income infants. Programs may provide services directly or may expand access by increasing
health insurance coverage. The federal health center program117 is one example of a program that
provides services directly to low-income pregnant women and their infants. Health centers are
facilities located in health professional shortage areas or medically underserved areas that provide
primary health care, including prenatal care, at reduced or no cost to low-income and uninsured
individuals. Health centers may provide care to women who would otherwise forgo prenatal care
or may provide primary and preventive care to infants, which could reduce infant mortality.118
Health centers also aim to reduce the number of low birthweight births among the pregnant
women they serve. Health centers have generally been successful in doing so, because the rate of
low birthweight births among women served by health centers is 5% below the national
average.119
(...continued)
include, but are not limited to (1) assurance of a vaccine’s safety and effectiveness before licensing; (2) development of
expert recommendations regarding the use of approved vaccines; (3) financing of recommended vaccines for certain
low-income children; and (4) a compensation program for individuals who suffer vaccine related injuries.
112 Preventing Infant Mortality.
113 Other agencies, such as AHRQ, may also sponsor IMR-related research. For example, AHRQ awards grants to
support the Excellence Centers To Eliminate Ethnic/Racial Disparities (EXCEED) program. These centers examine
causes of health disparities for a number of health conditions including infant mortality. See http://www.ahrq.gov/
research/exceed.htm.
114 Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health,
“Infant Mortality: Reducing Infant Deaths,” http://www.nichd.nih.gov/news/resources/advances/2005/infant.cfm.
115 Preventing Infant Mortality.
116 Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health,
“Infant Mortality: Reducing Infant Deaths,” press release, September 9, 2006.
117 CRS Report R42433, Federal Health Centers, by Elayne J. Heisler.
118 Ibid and Jeffrey Levi, Marlene Cimons, and Kay Johnson, Healthy Women, Healthy Babies, An Issue Brief from
Trust for America’s Health, Washington, DC, June 2008.
119 Department of Health and Human Services, Health Resources and Services Administration, Justification of
Estimations for Appropriations Committees, FY2013, Rockville, MD.
Congressional Research Service
24
The U.S. Infant Mortality Rate
The CDC, in partnership with state and local health departments, offers preconception services to
women. Some of these programs target women who have recently given birth in order to lengthen
the duration between a subsequent pregnancy, which may reduce the risk of infant mortality as
discussed above (see “Prenatal and Preconception Care”).120
Medicaid and CHIP provide health insurance coverage to certain low-income pregnant women
and infants.121 These programs may reduce the IMR by improving access to health care services.
Medicaid provides health benefits that cover prenatal care, birth, and health care costs for low-
income infants in the first year of life. The CHIP program covers low-income children with no
health insurance in families with income above Medicaid eligibility levels.
Do Federal Programs Reduce Infant Mortality?
Researchers found ambiguous results when they examined the effects of federal programs on
infant mortality and the IMR. Some researchers found that some programs, such as WIC, reduce
infant mortality; however, others found inconsistent results when examining the effects of
Medicaid. Determining whether a given income and material support program or health program
prevents infant mortality—and lowers the IMR—is difficult for a number of reasons, for
example, because data are unavailable or participants are enrolled in more than one program. This
section briefly summarizes some related research and then discusses some of the challenges
associated with assessing program effectiveness.
Evidence of Effects on Infant Mortality
Research on the effects of federal programs on infant mortality has generally examined the effects
of three programs: WIC, Medicaid, and Healthy Start. Researchers have found that women who
participate in WIC were less likely to have an infant die in the first year of life when compared to
similar women who did not participate in the program.122 WIC program participation resulted in
declines in conditions associated with infant mortality, such as low birthweight births.123 Another
study found that infants born to mothers enrolled in WIC were less likely to die from genetic
conditions or delivery complications, but WIC participation did not lower the risk of dying from
SIDS or accidents.124
Researchers and policymakers have questioned the effectiveness of Medicaid for reducing the
IMR.125 Some researchers have found that Medicaid participation improves outcomes for low-
120 See “Preconception Care” at http://www.cdc.gov/ncbddd/preconception/.
121 CRS Report RL33202, Medicaid: A Primer, by Elicia J. Herz, and CRS Report R40444, State Children’s Health
Insurance Program (CHIP): A Brief Overview, by Elicia J. Herz and Evelyne P. Baumrucker.
122 Marianne P. Bitler and Janet Currie, “Does WIC Work? The Effect of WIC on Pregnancy and Birth Outcomes,”
Journal of Policy Analysis and Management, vol. 24, no. 1 (Winter 2005), pp. 73-91; Hillary W. Hoynes, Marianne E.
Page, and Ann Huff Stevens, “Is a WIC Start a Better Start? Evaluating WIC’s Impact on Infant Health Using Program
Introduction,” National Bureau of Economic Research, Working Paper Series, No. 15589 (December 2009); and Intisar
Khanani et al., “The Impact of Prenatal WIC Participation on Infant Mortality and Racial Disparities,” American
Journal of Public Health, vol. 100, no. S1 (September 1, 2010), pp. S204-S209.
123 Ibid.
124 Nancy E. Moss and Karen Carver, “The Effects of WIC and Medicaid on Infant Mortality in the United States,”
American Journal of Public Health, vol. 88, no. 9 (September 1998), pp. 1354-1361.
125 See, for example, review in Dhaval M. Dave, et al., “Re-Examining The Effects of Medicaid Expansions for
(continued...)
Congressional Research Service
25
The U.S. Infant Mortality Rate
income women and their infants,126and that increased Medicaid eligibility reduces infant
mortality.127 Others have found that Medicaid has little or not effect on infant mortality.128
Medicaid may reduce infant mortality by providing access to prenatal care and health care in the
first year of life.129 However, the evidence that Medicaid increases access to prenatal care is
mixed. NCHS reported that between 1990 and 2003, rates of prenatal care utilization increased.
These researchers found gains among groups that typically have low prenatal care utilization—
the poor and racial and ethnic minorities—and linked these gains to Medicaid expansions that
began in the late 1980s.130 Others also found that Medicaid expansion increased prenatal care
use,131 and lowered the rate of low birthweight babies born to women at high risk for these
births.132 However, some researchers found that enrolling in Medicaid has little or no association
with the timing or number of prenatal care visits.133 This may occur because, as some researchers
have found, physicians and other providers will not accept Medicaid for a variety of reasons,
including low reimbursement rates;134 or because women may enroll in Medicaid after the first
trimester or at the time of the infant’s birth.135 These delays, or lack of access to care, in turn,
would mitigate the effectiveness of Medicaid coverage on infant health and the IMR.136
Evaluations of the Healthy Start program have also found conflicting outcomes. In general, this
has occurred because the evaluations were preliminary or focused on intermediate outcomes such
as low birthweight or short gestational births.137 A 2000 evaluation of the Healthy Start program
(...continued)
Pregnant Women,” National Bureau of Economic Research Working Paper Series, No. 14591 (December 2008).
126 Nancy E. Moss and Karen Carver, “The Effects of WIC and Medicaid on Infant Mortality in the United States,”
American Journal of Public Health, vol. 88, no. 9 (September 1998), pp. 1354-1361.
127 Seung-Eun Song, “Black and White Differences in Infant Mortality Risk Focusing on the Impact of the States’
Income Inequalities and Medicaid Eligibility,” Annual Meeting of the American Sociological Association,
Philadelphia, PA, August 2005.
128 Dhaval M. Dave et al., “Re-Examining The Effects of Medicaid Expansions for Pregnant Women,” National Bureau
of Economic Research Working Paper Series, No. 14591 (December 2008) and Jennifer Hollowell et al., “The
Effectiveness of Antenatal Care Programmes to Reduce Infant Mortality and Preterm Birth in Socially Disadvantaged
and Vulnerable Women in High-Income Countries: A Systematic Review,” Biomed Center: Pregnancy & Childbirth,
vol. 11, no. 13 (2011), pp. http://www.biomedcrentral.com/1471-2392/11/13.
129 U.S. Government Accountability Office, Poverty in America: Economic Research Shows Adverse Impacts on
Health Status and Other Social Conditions as well as the Economic Growth Rate, 07-344, January 24, 2007.
130 Births: Final Data 2008.
131 Janet Currie and Jeffrey Grogger, “Medicaid Expansion and Welfare Contractions: Offsetting Effects of Prenatal
Care and Infant Health?” National Bureau of Economic Research Working Paper Series, No. 7667 (April 2000).
132 Laura-Mae Baldwin et al., “The Effect of Expanding Medicaid Prenatal Services on Birth Outcomes,” American
Journal of Public Health, vol. 88, no. 11 (November 1998), pp. 1623-1629 and Lisa Dubay et al., “Changes in Prenatal
Care Timing and Low Birth Weight by Race and Socioeconomic Status: Implications for the Medicaid Expansion for
Pregnant Women,” Health Services Research, vol. 362, no. 2 (June 2001), pp. 373-398.
133 See, for example, Dhaval M. Dave, et al., “Re-Examining The Effects of Medicaid Expansions for Pregnant
Women,” National Bureau of Economic Research Working Paper Series, No. 14591 (December 2008). This article
reviewed some prior research in this area and also includes new analyses.
134 Peter J. Cunningham and Ann S. O’Malley, “Do Reimbursement Delays Discourage Medicaid Participation by
Physicians?” Health Affairs, vol. 28, no. 1 (November 18, 2008), p. w17–w28.
135 Janet Currie and Jeffrey Grogger, “Medicaid Expansion and Welfare Contractions: Offsetting Effects of Prenatal
Care and Infant Health?” National Bureau of Economic Research Working Paper Series, No. 7667 (April 2000).
136 Jeffrey Levi, Marlene Cimons, and Kay Johnson, Healthy Women, Healthy Babies, Trust for America’s Health, An
Issue Brief from Trust for America’s Health, Washington, DC, June 2008.
137 Information regarding the effectiveness of the Healthy Start program is drawn from the following sources: U.S.
General Accounting Office, Healthy Start: Preliminary Results from National Evaluation are Not Conclusive, HEHS-
(continued...)
Congressional Research Service
26
The U.S. Infant Mortality Rate
found that although the IMR declined in the Healthy Start sites, the declines were similar to what
had occurred in matched comparison areas—areas that had similar IMRs and economic
conditions as the Healthy Start sites—and in the nation as a whole. However, a 2006 evaluation
conducted by the Office of Management and Budget found some preliminary evidence that
Healthy Start reduced the IMR.138 Healthy Start may be more effective at reducing potential
causes of infant mortality and encouraging healthy behaviors in pregnant women. A 2006 HRSA
evaluation found that Healthy Start participation lowered the rates of low birthweight and short
gestational age births and increased prenatal care use, but this evaluation did not assess whether
participating in the program lowered the IMR. Similarly, a 2011 evaluation of an Arizona Healthy
Start site found that participants were less likely to have a low birthweight birth than
demographically similar non-participants.139 An evaluation of eight Healthy Start sites found that
participation in the program was associated with a number of healthy behaviors in pregnant
women. Specifically, women who participated in Healthy Start were, when compared to pregnant
women with similar demographic and economic characteristics, more likely to receive early
prenatal care, put their infants to sleep on their back, and engage in healthy behaviors during
pregnancy, such as not smoking or drinking.140
Challenges Assessing Program Effectiveness
The main goal of the federal programs discussed above is to improve the economic well-being or
health of low-income individuals, not to reduce the IMR. Although improving the economic well-
being or the health of low-income individuals may, by consequence, reduce infant death,
measuring this outcome and attributing it to a specific program is difficult. First, the data required
to do so are limited. Second, it is methodologically difficult to determine indirect outcomes.
Third, individuals are generally eligible for, and enrolled in, more than one program, making it
difficult to evaluate the effectiveness of a single program.141 Fourth, these programs generally
target those with IMR risk—low-income women and their infants—making it difficult to assess
(...continued)
98-167, June 1998; Barbara Devaney et al., Reducing Infant Mortality Lessons Learned from Healthy Start,
Mathematica Policy Research, Inc., Final Report, Princeton, NJ, July 2000; White House, Office of Management and
Budget, Detailed Information on the Healthy Start Assessment, Washington, DC, http://www.whitehouse.gov/omb/
expectmore/; United States Department of Health and Human Services, Health Resources and Services Administration,
Maternal and Child Health Bureau, A Profile of Health Start: Findings From Phase I of the Evaluation 2006,
Rockville, MD, 2006, http://mchb.hrsa.gov/healthystart/phase1report/; and U.S. Department of Health and Human
Services, Health Resources and Services Administration, Maternal and Child Health Bureau, Evidence of Trends, Risk
Factors, and Intervention Strategies, A Report from the Healthy Start National Evaluation 2006: Racial and Ethnic
Disparities in Infant Mortality, Rockville, MD. Four evaluations have examined the program since its inception in
1991: (1) a 1996 General Accounting Office (GAO, now the Government Accountability Office) review; (2) a 2000
review conducted by the Mathematica Policy Research, Inc. under contract from HRSA; (3) a 2006 Office of
Management and Budget (OMB) program assessment; and (4) a HRSA evaluation of the program from 2001-2005.
Phase I of this evaluation was published in 2006; the Phase II evaluation has not yet been published. Some recent
research articles have evaluated some Healthy Start Sites since 2006, some of these evaluations are described above.
138 White House, Office of Management and Budget, Detailed Information on the Healthy Start Assessment,
Washington, DC, http://www.experctmore.gov.
139 Syed K. Hussaini, Paul Holley, and Douglas Ritenour, “Reducing Low Birth Weight Infancy: Assessing the
Effectiveness of the Healthy Start Program in Arizona,” Maternal Child Health J, vol. 15 (2011), pp. 225-233.
140 Margo Rosenback et al., “Characteristics, Access, Utilization, Satisfaction, and Outcomes of Healthy Start
Participants in Eight Sites,” Maternal Child Health Journal, vol. 14 (2010), pp. 666-679.
141 One goals of Healthy Start is connecting participants with other services such as WIC and Medicaid. Given this
goal, a successful program would mean that applicants are enrolled in a number of programs.
Congressional Research Service
27
The U.S. Infant Mortality Rate
whether inconclusive findings or findings that a program does not reduce infant mortality occur
because a program is ineffective or whether the findings are due to the low socioeconomic status
of the program’s participants.142 A 2006 HRSA evaluation of Healthy Start suggests another
reason why Healthy Start, or any program that targets pregnant women, may not reduce the IMR.
In doing so, they summarize the life course perspective of research, saying that
[because] a woman’s birth outcome and subsequent pregnancies are influenced by the
accumulated life exposure of her mother and female ancestors before her, a short period of
intervention during the pregnancy period may not be able to cause a dramatic decrease in the
risk of poor birth outcomes. According to this model, the racial/ethnic disparities in birth
outcomes we see today can be explained by many lifetimes of amassed racism, poverty, and
stress.143
According to this perspective, interventions during pregnancy may be too late or insufficient to
reduce the risk of infant mortality among high risk pregnant women.
New Federal Initiatives
The Patient Protection and Affordable Care Act of 2010 (PPACA, P.L. 111-148) aims to expand
access to health insurance, which could reduce the IMR by expanding access to health care to
some individuals who are now uninsured. The law also establishes new, or expands existing,
programs that may reduce the IMR.144 Relevant sections include:
• Section 2951 established a program to award grants to states, Indian tribes, and
other entities, such as non-profit organizations, to implement home visitation
programs, including programs that target pregnant women. The section requires
an assessment of the new program to examine a number of child health
outcomes, such as infant mortality and low birthweight.
• Section 2953 established a new grant program for states to establish “Personal
Responsibility Education” programs.145 These programs, which target
adolescents, include instruction in both abstinence and contraception for the
prevention of pregnancy and sexually transmitted infections. Among other things,
states are required to establish a plan to reduce teen pregnancy and the teen birth
rate within the state.
• Section 3021 established a new Center for Medicare and Medicaid Innovation
within the CMS. This new center has broad authority to test new approaches for
delivering care to Medicare and Medicaid beneficiaries. Under this authority, the
142 There may be further challenges associated with evaluating Medicaid because it is a joint federal-state program so
states have flexibility in designing and administering their Medicaid programs. This creates variation that may
complicate Medicaid evaluations.
143 U.S. Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child
Health Bureau, Evidence of Trends, Risk Factors, and Intervention Strategies, A Report from the Healthy Start
National Evaluation 2006: Racial and Ethnic Disparities in Infant Mortality, Rockville, MD, see page 29.
144 The ACA (P.L. 111-148) was signed into law on March 23, 2010, it was subsequently by the Health Care and
Education Reconciliation Act of 2010 (HCERA, P.L. 111-152).
145 HHS announced grant awards made under this program at http://www.hhs.gov/news/press/2010pres/09/
20100930a.html.
Congressional Research Service
28
The U.S. Infant Mortality Rate
Center launched “Strong Start,” which will test the effectiveness of various
strategies to deliver enhanced prenatal care to reduce short gestational age births
among Medicaid recipients.146
• Section 4101 expanded Medicaid coverage of smoking cessation services for
pregnant women by making counseling and pharmacotherapy to promote tobacco
cessation a mandatory benefit for pregnant women beginning October 1, 2010.147
• Sections 10211-10214 established and appropriate funding for a “Pregnancy
Assistance Fund” that creates and funds a new competitive grant program to
states to help pregnant and parenting teens and women. States, in turn, will award
funds to institutions of higher education or other entities to establish programs
and services to support pregnant or parenting students.148
Conclusion
The United States has a higher infant mortality rate (IMR) than most European countries, and the
U.S. IMR has remained relatively constant since 2000 despite declines in prior decades. Reasons
for the high U.S. IMR and its recent stability are difficult to discern. However, experts believe
that differences in how live births are recorded internationally cannot fully explain the high U.S.
IMR. Instead, experts suggest that it might be explained by higher U.S. rates of low birthweight
and short gestational age births. Within the United States, IMR varies by racial and ethnic group.
The disproportionately high rates among certain racial and ethnic groups may partially explain
the high U.S. IMR. The U.S. IMR also varies geographically. In general, southern states have the
highest IMRs and states in the West and in New England have the lowest. The higher IMRs in
southern states may be partially explained by higher rates of low birthweight and short gestational
age births in these states. In addition, the racial and ethnic composition of a state’s population
affects its IMR, as the IMR for infants born to black mothers is more than double the rate for
infants born to white mothers. A number of demographic characteristics of the mother, including
race, education, and age, may directly or indirectly influence the IMR. In addition, the IMR may
be affected by health and health system characteristics such as the mother’s health behaviors or
whether she receives prenatal care.
Research indicates that a number of federal programs may reduce the IMR. These programs
generally target low-income women and children, and may improve their economic
circumstances and health status, thereby reducing the IMR. However, the effectiveness of these
programs is debated, and is difficult to evaluate. The ACA also establishes new programs that
may reduce the IMR. For example, the law appropriates funding for home visitation programs for
146 Centers for Medicare & Medicare Services: Center for Medicare & Medicaid Innovation, “Strong Start for Mothers
and Newborns,” press release, March 20, 2012, http://innovations.cms.gov/initiatives/strong-start/.
147 CRS Report R41210, Medicaid and the State Children’s Health Insurance Program (CHIP) Provisions in ACA:
Summary and Timeline, by Evelyne P. Baumrucker et al.
148 These provisions are summarized in CRS Report R41278, Public Health, Workforce, Quality, and Related
Provisions in PPACA: Summary and Timeline, coordinated by C. Stephen Redhead and Erin D. Williams. For
information on grants awarded, see Department of Health and Human Services, “HHS Awards $27 Million for the
Support of Pregnant and Parenting Teens and Women,” press release, September 28, 2010, http://www.hhs.gov/news/
press/2010pres/09/20100928d.html.
Congressional Research Service
29
The U.S. Infant Mortality Rate
pregnant women or new parents.149 It also requires that Medicaid cover smoking cessation
services for pregnant women.
Author Contact Information
Elayne J. Heisler
Analyst in Health Services
eheisler@crs.loc.gov, 7-4453
Acknowledgments
This report benefitted from the review and guidance of Kirsten J. Colello, Sarah A. Lister, and
Amanda K. Sarata.
149 For information about funding allocated under this program, see http://www.hhs.gov/news/press/2010pres/07/
20100721a.html.
Congressional Research Service
30