EFFECTS OF MARITAL STATUS AND
RACIAL DISPARITIES IN LOW BIRTHWEIGHT
INFANTS OF MEDICAID WOMEN
By
Lori House

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Human Development and Family Studies – Doctor of Philosophy
2014

ABSTRACT
EFFECTS OF MARITAL STATUS AND
RACIAL DISPARITIES IN LOW BIRTHWEIGHT
INFANTS OF MEDICAID WOMEN
By
Lori House
Low birthweight (less than 2500 grams) is a leading cause of death in the first 28
days of life in the United States; in 2012, eight percent of infants born in the United States
were low birthweight. These infants are more likely to have long-term health sequealae or
die during the first year of life than are infants of normal weight. This makes it important
to further the understanding of low birthweight so prevention efforts are evidence-based
and can lower the infant mortality rate in the United States; the United States has
consistently higher infant mortality rates than most industrialized countries.

Infant

mortality is a multi-faceted problem, with poverty being a known risk factor for both infant
mortality and low birthweight.
Weathering is a sociologic concept which posits that the longer one lives in poverty,
the faster the aging process and onset of health conditions occur. Studies have been done
applying the weathering concept to low birthweight in pregnancy, and comparisons of
Black and White populations have been done; Hispanics have not been included in these
studies.
Marital status has been studied for its effect on numerous health conditions, such
as cardiovascular disease and cancer, and being married has been shown to frequently be

protective against disease complications and mortality. There has been very little research
to date investigating the effect of being married on low birthweight.
This dissertation is a secondary analysis of the 2012 Natality Public Use Dataset,
compiled annually by the Centers of Disease Control and Prevention. Black, White, and
Hispanic mothers having singleton first-births in the US in 2012 were included in this study
(N=499,323). Medicaid status was used as a proxy for living in poverty. The three
racial/ethnic groups were analyzed for weathering by examining rates of hypertensive
disorders (both chronic and pregnancy-induced). Marital status was included to investigate
protective effects of being married.
Results showed that the weathering hypothesis was supported only within each
racial/ethnic group. This may be due to all women in this study already being in poverty;
comparison against the general population was not included.
Marital status was found to be significantly protective against LBW regardless of
the presence or absence of hypertensive disorders for all three racial/ethnic groups, with
Black women receiving the strongest protection. Hispanic women had the highest odds
of LBW in the presence of both chronic and pregnancy-induced hypertension. These are
both new findings.
Implications for healthcare providers and policymakers are discussed. In this time
of rapidly evolving healthcare insurance in America, the impetus is great to make evidencebased decisions on creating healthcare agencies and policies to help lower America’s infant
mortality rate.

This dissertation is dedicated to the late Dr. Tom Luster. His love of teaching, collegial
approach to students, and demonstration of true mentorship live on in my approach to
students. He continually encouraged students to look at things differently, develop
strategies to improve conditions for people at risk, and demonstrated how to think globally
but act locally. You are truly missed, Dr. Luster.

iv

ACKNOWLEDGMENTS

The completion of this dissertation was dependent on the mentorship and
encouragement of multiple people. First, Dr. Barbara Ames, who had the unenviable task
of stepping into Dr. Tom Luster’s shoes as my chair. You not only fulfilled that role, but
taught me the true meaning of academic leadership and resilience after the loss of Dr.
Luster. Dr. Sue Grady inspired me to continue working on this data, despite numerous
difficulties; your support was truly foundational for finalization of this research.
My children, Courtney (and Joe), Brandon, Morgan, and Eric, continually had faith
in my ability to complete this degree. Your faith in me was more important than you will
ever know. I have that same faith in each of you; follow your dreams!
My special friend, Kim, lifted me up frequently. Your proof-reading, suggestions,
and emotional support were instrumental in seeing this project through to the end.
There are a plethora of words to express my gratitude, but they all seem inadequate
to convey the depth and breadth of my appreciation. Thank you, each of you, for all your
encouragement, love, and support. You made this possible.

v

TABLE OF CONTENTS

LIST OF TABLES

viii

LIST OF FIGURES

x

CHAPTER 1
INTRODUCTION
Purpose Statement
Statement of the Problem
Significance of the Study
Research Questions
Hypotheses
Theoretical Framework
Dissertation Format

1
1
4
4
6
7
8
11
15

CHAPTER 2
REVIEW OF THE LITERATURE
Poverty and Pregnancy
Poverty and Hypertension
Low Birthweight
Preterm Birth
Small for Gestational Age
Hypertension
Smoking
Weathering
Marital Status

16
16
16
18
18
19
22
24
25
25
27

CHAPTER 3
RESEARCH DESIGN
Data Set
Definitions

29
29
29
30

CHAPTER 4
32
STATISTICAL ANALYSIS
32
Descriptive Statistics
32
Results of the Logistic Regression
38
Question #1: What is the effect of increasing maternal age on the odds of having
a LBW infant controlling for marital status and smoking (Path A)?
38
Question #2: What is the effect of increasing maternal age on the odds of having
hypertension (CHTN or PIH) while controlling for marital status and smoking
(Path B)?
40

vi

Question #3: What is the effect of hypertensive disorders on the odds of having
LBW, controlling for smoking and marital status (Path C)?
43
Question #4: What is the effect of increasing maternal age on the odds of having
LBW, controlling for hypertensive disorder, marital status and smoking
(Path D)?
45
CHAPTER 5
48
CONCLUSIONS
48
Discussion of the Results of the Research Questions
51
Question #1: What is the effect of increasing maternal age on the odds of having
a LBW infant controlling for marital status and smoking (Path A)?
52
Question #2: What is the effect of increasing maternal age on the odds of having
hypertension (CHTN or PIH) while controlling for marital status and smoking
(Path B)?
54
Question #3: What is the effect of hypertensive disorders on the odds of having
LBW, controlling for smoking and marital status (Path C)?
56
Question #4: What is the effect of increasing maternal age on the odds of having
LBW, controlling for hypertensive disorder, marital status and smoking
(Path D)?
57
Summary and Implications
59
Limitations
62
Future Research
66
REFERENCES

67

vii

LIST OF TABLES

Table 1: Birth Certificate Variables Excluded in this Study

34

Table 2: Frequency of Racial/Ethnic Groups Included in this Study

34

Table 3: Rate of LBW by Racial/Ethnic Group (Rate per 100 live births)

35

Table 4: Descriptive Statistics on Maternal Age in Years by
Racial/Ethnic Group

35

Table 5: Descriptive Statistics of Marital Status and Smoking by
Racial/Ethnic Group

36

Table 6: Percentage of Births with CHTN and PIH by Racial/Ethnic Group

36

Table 7: Cross Tabulation of Frequency of Presence/Absence of CHTN and
Smoking Within CHTN Variable

37

Table 8: Cross Tabulation of Frequency of Presence/Absence of CHTN and
Smoking Within Smoking Variable

37

Table 9: Cross Tabulation of Frequency of Presence/Absence of PIH and
Smoking Within PIH Variable

38

Table 10: Cross Tabulation of Frequency of Presence/Absence of PIH and
Smoking Within Smoking Variable

38

Table 11: Effect of Increasing Age on the Odds of Having LBW
Controlling for Marital Status and Smoking

40

Table 12: Effect of Increasing Age on the Odds of CHTN
Controlling for Marital Status and Smoking

41

Table 13: Effect of Increasing Age on the Odds of Development of PIH
Controlling for Marital Status and Smoking

42

Table 14: Effect of CHTN on the Odds of Having LBW
Controlling for Marital Status and Smoking

44

Table 15: Effect of PIH on the Odds of Having LBW
Controlling for Marital Status and Smoking

45

viii

Table 16: Effect of Increasing Age on the Odds of Having LBW
Controlling for CHTN, Marital Status, and Smoking

47

Table 17: Effect of Increasing Age on the Odds of Having LBW
Controlling for PIH, Marital Status, and Smoking

47

ix

LIST OF FIGURES

Figure 1: Diagram: Research model

10

Figure 2: Diagram: Application of Bronfenbrenner’s Model

15

Figure 3: Bar Graph: U.S. Preterm Birth percents by Weeks Gestation and
Maternal Race

23

x

Chapter 1
INTRODUCTION
Infant mortality is often viewed as an important overall measure of the health of a
nation, as it includes maternal health, public health practices, socioeconomic conditions,
and the quality of and access to medical care. The world-wide infant mortality rate (number
of infants who die by the age of 12 months per 1000 live births), was 42 when last reported
as 2005-2010 figures by the United Nations Population Division (2013). Among the 195
countries reporting, the United States ranked 27th, with a rate of 6.1. This is a slight
improvement from being ranked 33rd in 2005 with the rate of 6.8 (United Nations
Population Division, 2007), but still demonstrates a continuation of the decades-long gap
between the countries with the lowest infant mortality rates and the United States. This
has long been a point of embarrassment and frustration for the United States, as the level
of health care services available in America is far more advanced than many countries with
significantly lower infant mortality rates. Access to those services and equity of services
within geographic areas and sub-population groups (ethnicity/race) remains a challenge in
America.
Infant mortality rates (IMR) vary widely within the United States. In the latest
reported figures from 2008 births, the rate ranged from a low of 5.1 in California to a high
of 11.9 in the District of Columbia (Mathews & MacDorman, 2010). This variation is due
in part to the differences in high-risk population subgroups within each state (e.g., rural,
urban, racial/ethnic, poverty). In addition, the disparity between the races continues. The
IMR for Black infants born in the United States in 2011 was 11.4 compared to White
1

infants 5.1, (Black/White rate ratio [RR] = 2.2), and the IMR for Hispanics was 4.6 (Hoyert
& Xu, 2012).
Disorders related to short gestation (infants born < 37 weeks gestation) and low
birthweight (infants born < 2,500 grams) account for 17.0% of all infant deaths, second
only to the grouping of congenital malformations, deformations, and congenital anomalies
(Mathews & MacDorman, 2010). Aside from rare teratogen exposures, the congenital
malformation group is largely unpreventable.

This makes the category of short

gestation/low birthweight the most common cause of infant deaths which may be amenable
to improvement through intervention strategies.
Living in poverty is a known maternal risk factor for having a preterm or low
birthweight infant (Borders, Grobman, Amsden, & Hall, 2007; Braveman et al., 2010;
Collins, Wamback, David, & Rankin, 2009; DeFranco, Lian, Muglia, & Schootman, 2008;
Geronimus, 1996; Goldenberg et al., 1996; Janevic et al., 2010; Kramer & Hogue, 2008;
Sims, Sims, & Bruce, 2007). Mothers who live in poverty are at higher risk of having low
birthweight infants because of the lack of amenities and resources available to maintain
and promote preconceptual and maternal health. However, even among women living in
equivalent poverty, there are significant differences in birth outcomes between
races/ethnicities (Grady, 2006). One explanation of this disparity is that of “weathering,” a
concept first proposed by Geronimus (1996). The premise of weathering is that there is
accelerated aging when living in detrimental conditions such as poverty, and the more one
weathers, the higher the incidence and/or severity of premature and chronic medical
conditions. These medical conditions may create secondary effects, such as adverse birth
outcomes. Geronimus found that black women weathered faster than white women, which

2

accounted for some of the disparity in birth outcomes as maternal age increased; this has
been supported by later studies (Geronimus, Bound, & Waidmann, 1999; Collins et al.,
2008; Holzman et al., 2009; Love, David, Rankin, & Collins, 2010; Lu & Chen, 2004; Lu
& Halfon, 2003; Rauh, Andrews, & Garfinkel, 2001).
The weathering research above was limited to Black and White populations. The
exclusion of Hispanics may be due to the “epidemiologic paradox.” This well-studied
phenomenon states that while most Hispanics live in poverty, they have health outcomes,
including infants with low birthweight, that are comparable or better than those of their
White counterparts, and considerably better than their Black counterparts to whom they
were more comparable from a socioeconomic viewpoint (Franzini, Ribble, & Keddie,
2001; Fuentes-Afflick, Hessol, & Perez-Stable, 1999; Markides & Coreil, 1986;
Rosenberg, Raggio, & Chiasson, 2005). The lack of inclusion of Hispanics within the
context of weathering is a gap in the literature.
Many health conditions affect pregnancy outcomes. Hypertension is one of the
most prevalent conditions, and a known risk factor for adverse birth outcomes, including
low birth weight, abruption placentae, and iatrogenic prematurity (Perry, Hockenberry,
Lowdermilk & Wilson, 2010). There are two classifications of hypertensive disorders in
relation to pregnancy: pre-existing and pregnancy-induced. Pre-existing (also known as
chronic hypertension [CHTN]) must be diagnosed prior to pregnancy, while pregnancyinduced hypertension is first evident after 20 weeks gestation. Both conditions have the
potential of having profound effects on the mother and fetus, up to and including death due
to the narrowing of the vasculature, causing poor perfusion of the placenta, which
subsequently causes decreased perfusion of oxygen to the fetus.

3

Being married also has been found to improve birth outcomes across various
populations (Barrington, 2010; Hohmann-Marriott, 2009; Masho, Chapman, & Ashby,
2010; Nkansah-Amankra, Dhawain, Hussey, & Luchok, 2010). It is acknowledged that
marital status is not a direct equivalent of financial and/or psychosocial support, but it is
an indirect measure that is easily obtained on several demographic forms (including birth
certificates), and has been found to have a positive influence on birth outcomes. Marriage
is a mechanism by which mothers receive social/emotional and other types of support
during and after pregnancy, and that support appears to mitigate some of the detrimental
effect of living in poverty. This study will attempt to further explain the relationships
between weathering and poverty with marital status, to explore racial differences in birth
outcomes.
PURPOSE STATEMENT
The overall purpose of this study is to investigate racial differences (Black, White,
Hispanic) in low birthweight for single mothers on Medicaid. The variables that will be
studied to explain racial differences in low birthweight are hypertensive disorders to assess
racial differences in “weathering” and marital status to assess racial differences in the
potential of family protection.

Statement of the Problem
As the leading cause of infant mortality that is potentially preventable (or at least
amenable to prevention strategies), low birthweight is an issue of great importance in the
United States. Aside from mortality, the morbidity consequences of low birthweight are
both immediate and long-lasting. In the best situation, the infant will have to spend a few

4

extra days in the hospital to ascertain the ability to properly feed and demonstrate
thermoregulation. In the more severe cases, a lengthy Neonatal Intensive Care Unit
(NICU) stay ensues, with likely complications and set-backs along the way. The family
living in poverty may be ill-equipped to deal with the NICU stay (e.g. costs, travel), and
may be overwhelmed with care of a special-needs infant upon discharge. The health
concerns of these low birthweight infants do not end upon discharge, or even upon reaching
one year of age; they are at high-risk for life-long morbidity (Barker, 2006). It has been
shown that low birthweight is associated with increased rates of adult coronary heart
disease, stroke, diabetes, and hypertension (Barker, 2006). These morbidities are thought
to be due to the negative effect on the plasticity of vasculature and organs during in-utero
stress.
A reduction in the rate of low birthweight infants is critical to addressing the
problem, but has remained a medical mystery for decades (Berkowitz & Papiernik, 1993;
Goldenberg et al., 1996; Tucker et al., 1991; Tucker & McGuire, 2004). Berkowitz and
Papiernik even concluded, “It is unlikely that there will be further substantial improvement
in infant survival in the United States unless a reduction in births of preterm low birth
weight infants can be accomplished” (p. 414); this has yet to happen. It is critical that the
underlying predisposing factors be identified to address effective reduction or prevention
strategies. A holistic approach to investigation of these predisposing factors is tantamount
to a solution; decades of biological studies have failed to identify an etiology for the
majority of cases.

While these medical studies must continue, so must they be

supplemented with psychosocial evidence-based practice implications.

5

Racial (black vs. white) disparities in low birthweight are a well-known yet
unexplained phenomenon. Even when controlling for known biological risk factors, there
remains an inexplicable, very significant gap between black and white low birth weight
births. This becomes a greater issue when looking at population projections, as the
percentage of “white” is declining. The US Census Bureau (as cited in Martinez & Ariosto,
2011) reported that based on the 2010 Census, Hispanics accounted for more than half of
the nation’s expansion from the last census, making them currently the second most
populous subgroup in the United States. It is crucial, therefore, to include Hispanics in this
research, as they have been largely ignored in this research area.
Hypertension is one of the most common pregnancy complications, and has low
birthweight and/or iatrogenic (caused by treatment) prematurity as common outcomes. In
2007, the US rate of chronic (pre-existing) hypertension (CHTN) was 11.0 per 1,000 live
births. This compared to 2000 when the rate was 7.6, which is an increase of 45 percent
from 2000 to 2007 (Martin et al., 2010). Conversely, the rate of pregnancy-induced
hypertension (PIH; hypertension that does not occur until at least half-way through the
gestation) has remained stable during that same time span, fluctuating between 37.4 (2003)
and 39.9 (2005), with the latest reported rate being 38.8 in 2007.
Significance of the Study
Weathering within the context of pregnancy, although a relatively new concept, has
received support in the literature (Borders et al., 2007; Collins et al., 2009; Geronimus,
1996; Holzman et al., 2009; Love et al., 2010; Lu & Chen, 2004; Lu & Halfon, 2003;
Nkansah-Amankra, Luchok, et al., 2010; Sims et al., 2007). A very similar yet distinct
body of literature has investigated the birthweight effects of living in poverty (Colen, 2006;

6

DeFranco et al., 2008; Grady, 2006; Grady & Ramirez, 2008; Janevic et al., 2010;
Nkansah-Amankra, Dhawain, et al., 2010; Reagan, Salsberry, & Olsen, 2007). All of these
studies except Sims et al. have looked at either solely black outcomes or black-white
differences; none have included Hispanics. This is understandable, as the black population
has had the worst rates for infant mortality within America, and comparison to whites has
highlighted that disparity. However, as the largest growing ethnic group in the United
States (Martinez & Ariosto, 2011), it is crucial to include Hispanics in this research.
Additionally, as compared to other racial/ethnic groups in America, they have the highest
fertility rates (Martinez & Ariosto); the obvious potential effect on overall birth outcomes
is significant.
Marital status, as a measure of support, has been shown to be protective against
adverse birth outcomes (Barrington, 2010; Hohmann-Marriott, 2009; Masho et al., 2010).
However, it has not been included in the weathering studies to date. This study will
contribute to the literature by compiling and analyzing the interplay of weathering and
marital status for the three most populous (non-Hispanic black, non-Hispanic white,
Hispanic) racial/ethnic groups. As aforementioned, Hispanics are the fastest growing
subculture within America, and have yet to be included in major studies.

RESEARCH QUESTIONS
The following research questions are posed to gain a better understanding of the
effects of weathering and marital status on the incidence of low birthweight, for nonHispanic Black women, non-Hispanic White women, and Hispanic women who are on
Medicaid in the United States. In this study Medicaid is used as a proxy for poverty. For

7

each of these questions, the three racial/ethnic groups will be analyzed individually; marital
status and smoking will be entered as controls but racial differences in their impacts on low
birthweight will also be evaluated.
1. What is the effect of increasing age on the odds of having a low birthweight infant
controlling for marital status and smoking (Path A)?
2. What is the effect of increasing age on the odds of having hypertension
(pre-existing [chronic] and pregnancy-induced) controlling for marital status and
smoking (Path B)?
3. What is the effect of hypertensive disorders (chronic and pregnancy-induced) on
the odds of having a low birthweight infant, controlling for marital status and
smoking (Path C)?
4. What is the effect of increasing age on the odds of having a low birth weight infant,
controlling for hypertensive disorders, marital status and smoking (Path D)?

Hypotheses
Path A:
1. The strongest effect of increasing age on the odds of having a low birthweight infant
will be seen in Black women.
2. Smoking will increase the odds of low birthweight for all racial/ethnic groups.
3. Being married will decrease the odds of low birthweight for all racial/ethnic groups.

8

Path B:
4. The strongest effect of increasing age on the odds of having both chronic and
pregnancy-induced hypertension will be seen in Black women.
5. Smoking will increase those odds for all three racial/ethnic groups.
Path C:
6. The strongest effect of CHTN on the odds of having LBW will be seen in Black
women.
7. Being married will decrease the odds of having low birthweight for all three
racial/ethnic groups with CHTN or PIH.
Path D:
8. The strongest effect of increasing age on the odds of having low birthweight will
be seen in Black women.
9. All three racial groups will have increased odds of LBW with increasing age in the
presence of CHTN.
10. All three racial groups will have decreased odds of LBW if they are married.

9

Figure 1: Research Model

Hypertensive
Disorders
(both CHTN & PIH)

Smoking

Marital
Status

Path A

Marital
Status

Path B

Path C

Age

Marital
Status

Path D: Age

LBW

Smoking

LBW, controlling for hypertension, smoking, and marital status

: mediator

: moderator

LBW: low birthweight
CHTN: chronic hypertension
PIH: pregnancy-induced hypertension

10

Smoking

THEORETICAL FRAMEWORK
This study will utilize the human ecological framework, to investigate the
interactions between mothers living in poverty and their environmental systems. Human
ecology theory focuses on the interaction of humans, both as biological and social beings,
with their environment (Bubolz & Sontag, 1993). Individuals, as well as families, are seen
as dynamic and adaptive to changes within the internal and external systems with which
they are interdependent (Bronfenbrenner, 1986). The framework can be used with many
different populations of interest at the core (e.g., individual, family, neighborhood), as the
interactions and adaptations within systems are applicable and adjustable to any core unit.
For purposes of this study, the new mother and low birthweight (herein referred to as LBW)
infant living in poverty will be the unit of analysis. There may or may not be others in the
household, including a potential spouse.
Bronfenbrenner (1979) developed a conceptual model of the ecological
environment, which is a “nested arrangement of concentric structures, each contained
within the next. These structures are referred to as the micro-, meso-, and macrosystems”
(p. 22). The innermost structure, the microsystem, is “a pattern of activities, roles, and
interpersonal relations experienced by the developing person in a given setting with
particular physical and material characteristics” (p. 22). An example of a microsystem is
an individual’s or family’s home; the activities, roles, and relationships within that home
comprise the critical components of the microsystem. Each home must be evaluated for
its individual occupants, activities, roles, and relationships; similar family compositions
(e.g., traditional married couple with two preschool aged children) will have both shared
characteristics and unique features.

11

The other systems move outward from the microsystem. The next concentric layer
is the mesosystem. Described by Bronfenbrenner (1979) as those settings beyond the
immediate household in which the household occupants regularly interact, this would
include entities such as work and school. This is followed by the exosystem, which
includes those settings in which the household occupants do not directly interact, but still
have a direct effect on the household. An example would be a non-live-in boyfriend’s
workplace; his job allows him to be able to contribute financially to the care of his child
who lives in the household of interest.

The most outer layer is the macrosystem.

Bronfenbrenner describes this as “consistencies, in the form and content of lower-order
systems (micro-, meso-, and exo-) that exist, or could exist, at the level of the subculture
or the culture as a whole, along with any belief systems or ideology underlying such
consistencies” (p. 26). Family and health professionals’ beliefs that all pregnant women
deserve access to proper prenatal care and resources, regardless of race or ethnicity, is an
exemplar of a macrosystem consistency. Other examples are religion and neighborhood
mores.
Bronfenbrenner (1986) later added a fifth system to his model. This is not another
concentric circle, but rather an “over-lay” on the model. Labeled the “chronosystem,” this
refers to “time,” and represents the fluidity and dynamic status of families and individuals.
It can be used simplistically to represent a current snapshot of a certain timeframe in a
family’s life, or in more complex application, can be used to look at family life transitions
over a time period. This complexity is echoed by another well-respected family scientist’s
statement that “Studying families isn’t rocket science, it’s harder. They are continually

12

changing, day to day, and always fluid” (Dr. Joanne Keith, personal communication, July
2003).
In later years, Bronfenbrenner (1995) further refined his model to focus on ProcessPerson-Context-Time.

Process refers to the interactions the person has with their

environment, which is key to development; “it is by engaging in these activities and
interactions that individuals come to make sense of their world and understand their place
in it” (Tudge, Mokrova, Hatfield, & Karnik, 2009).
Human ecology theory supplements the concept of weathering. As women adapt
and interact with their environment (ecology theory), there are simultaneous and
subsequent physical adaptations occurring (weathering). Looking through the lens of
human ecology allows for identification of the various factors within different levels
(immediate or removed) of a mother’s environment that can either directly or indirectly
affect her health. A representative theoretical model is in Figure 2, with the factors
contained in the research questions highlighted (page 15).
This study will focus on key factors within the microsystem, mainly, the presence
of a spouse and the neighborhood of poverty. It must be understood that the many
exemplars within the other systems (meso-, exo- macro-) are an ever-present influence on
the mother and LBW infant (as demonstrated in Figure 2), but will not be included in this
study.

13

Figure 2: Application of Bronfenbrenner’s Model

14

DISSERTATION FORMAT
The following outlines the intended format of this dissertation:
Chapter One outlines the purpose and significance of this dissertation, with an
overview of low birthweight, weathering, smoking, and marital status. A discussion of the
ecological framework is included. Measures and variables to be extracted from the birth
certificates are described.
Chapter Two is a detailed literature review of the key concepts of low birthweight,
hypertension in pregnancy, weathering, and marital status.
Chapter Three describes the research questions, hypotheses, and research design,
including the statistical methods that will be used.
Chapter Four lists the results of the statistical analysis.
Chapter Five is a discussion and integration of the results. This includes the
contribution to the literature from these studies.

Of import are implications for

professionals working with families in poverty, health professionals engaged in the care of
pregnant women, and policymakers who are responsible for designation of funding for
health and family support programs, especially for those living in poverty.

15

Chapter Two
Review of the Literature
It is important to look only at singleton births when examining preterm birth rates,
due to the rapid increase in the use of assisted reproduction techniques resulting in higher
rates of twins and higher-order multiple gestations over the last several decades. Therefore,
this dissertation is limited to singleton births, which account for approximately 96 percent
of births in the US (Martin et al., 2010).
Poverty and Pregnancy
Poverty and its ill-effects have been thoroughly documented. The effects of living
in poverty have special implications for pregnant women, as the repercussions for the fetus,
then child, can be life-long (further discussed in the following sections).
The relationship of poverty to low birthweight has been well established in the
literature, to the point where poverty is listed as a risk factor in maternity nursing textbooks
(London, Ladewig, Ball, Bindler, & Cowen, 2011; Perry, Hockenberry, Lowdermilk, &
Wilson, 2010).

It is accepted that the connections between poverty and health are

multifaceted, and include substandard housing, inadequate nutrition, lack of access to
healthcare, and more environmental stressors with fewer material resources to cope with
those stressors (Braveman et al., 2010).

Physiologic responses to stress are well-

documented, and involve many body systems, including the immune and vascular systems.
These two physiologic systems are key to proper fetal development, as pregnancy already
is an immunocompromised state, and the maternal vascular system must supply the
placenta with all the fetal oxygen and nutritional needs.

16

Prenatal nutrition will affect both the mother and fetus. During pregnancy, if the
body senses nutritional deficits, they will be shunted and/or pulled from the mother’s
nutritional stores to supply the growing fetus. These needs especially include calcium,
protein, and iron. If there is inadequate maternal calcium intake, it will be pulled from
maternal bones to supply the fetus, leaving the mother with increased risk of early and
pronounced osteoporosis. Sources of protein tend to be expensive, and can be unattainable
to those struggling economically. Red meat also is the main source of iron, which has
increased demand during pregnancy. If the pregnant woman becomes anemic, it places her
at high risk for extreme fatigue and hemorrhage, while placing the fetus at further risk for
restricted growth and oxygen deprivation. Briley, Flanagan, and Lewis (2002) found that
an in-home nutritional education program for black pregnant women living in poverty was
effective in increasing iron intake, and women in the intervention group had significantly
higher infant birthweights. LeBlanc and Rioux (2007) found a nutritional intervention
program to be effective in lowering rates of maternal anemia for a low-income group of
Caucasian women.in Canada.
Psychosocial/emotional stressors of living in poverty also are well-documented.
Worry about the basic needs of survival (food, shelter, clothing, safety) cause chronic
stress, which has been linked to multiple physiologic adversities, including hypertension
and cancer. Stress during pregnancy (antenatal stress), however, has been studied very
little as a separate entity. Associations have been found between antenatal stress and
domestic violence (Curry, 1998; Dunn & Oths, 2004; Heaman, 2005), substance use (Jesse
& Reed, 2004; Orr, James, & Miller, 1996), depression (Jesse & Swanson, 2007; Jesse,

17

Walcott-McQuigg, Mariella, & Swanson, 2005; Zuckerman, Amaro, Bauchner, & Cabral,
1989), poor weight gain (Orr et al.), and presence of a chronic medical disorder (Orr et al.).
Poverty and Hypertension
The association between poverty and hypertension is well-documented and
multifactorial. Lack of access to healthcare and the inability to afford medications are
common issues.

Negative lifestyle choices which complicate hypertension, such as

smoking and poor nutrition, also are more prevalent among lower income populations
(Baumann, Chang, & Hoebeke, 2002; Kaplan et al., 2005; Nagahawatte & Goldenberg,
2008; Rothberg, Magriples, Kershaw, Rising, & Ickovics, 2011; Seligman, Laraia, &
Kushel, 2010), especially if Black (Morenoff et al., 2007; Tanaka et al., 2007).
Less research has been done specific to pregnancy-associated hypertension and
poverty, but studies are concordant in their findings that there is increased risk of PIH when
living in poverty (Laraia, Siega-Riz, & Gundersen, 2010; Nagahawatte & Goldenberg,
2008; Tanaka et al., 2007).
Low Birthweight
Low birthweight (LBW) is defined as those infants weighing less than 2500grams
at birth (5½ pounds), regardless of gestation. The US rate of LBW in 2007 was 8.2 percent,
which has been on a slow but steady increase since the reported rate of 6.7 percent in 1984
(Martin et al., 2010), and data from 2012 shows the rate to be relatively stable at 8.0 (Martin
et al., 2013). The increased health risks associated with LBW are well-documented. In
2006, 25 percent of all infant deaths were attributed to LBWs (Martin et al., 2010).
Additionally, these infants have a multitude of chronic conditions, which will be further
discussed later.

18

Within the LBW classification, there are two distinct groups: those who were born
early and are therefore small, and those who are smaller than expected for that gestation.
To be able to differentiate those infants who are smaller than expected for their given
gestation, the term “SGA” (small for gestational age) is utilized. An SGA infant must be
less than the 10th percentile for weight at that gestation. There are typically different
etiologies involved in SGA births than LBWs, as SGAs have had an issue with in-utero
growth. This is different from a normally-growing fetus who is then subjected to preterm
birth for whatever reason, and would then be classified as LBW, but not SGA. Conversely,
it is possible to be both LBW and SGA if the infant was born early and is still smaller than
expected for that gestation. Therefore, the outcome of LBW will be discussed in terms of
both preterm birth (PTB) and SGA, to facilitate full understanding.
Preterm Birth
Preterm birth occurs prior to 37 weeks of gestation. It is often argued that preterm
labor is the most complex, perplexing pathology within obstetrics today. There are
multiple known causal factors such as trauma, infections, cocaine use, and uterine
overdistention (e.g., twins), as well as iatrogenic causes.

Despite all these known

etiologies, approximately 45 percent of all preterm labor remains unexplained, which
constitutes the largest percentage of all preterm labor causes (Goldenberg, Culhane, Iams,
& Romero, 2008). Without a known cause for the majority of cases, prevention is difficult
if not impossible. In fact, for decades, the question has remained as to whether it is feasible
to reduce preterm birth rates with the limited knowledge of etiology. Tucker et al., as early
as 1991, posited that further reduction in the preterm birth rate was not logical until further
etiologies were identified; to date, these “further” etiologies remain unknown.

19

The overall rate of prematurity in America for 2012 was 11.6 percent of all births
(Martin et al., 2013).

Prematurity remains the leading cause of infant morbidity and

mortality, with the severity of outcomes being positively correlated with the degree of
prematurity. The degrees of prematurity are grouped into late preterm (34-37 weeks),
preterm (28-33 weeks), and extreme preterm (<28 weeks).
The percentage of singleton births which were late preterm was 8.03 in 2007, down
slightly from 8.14 in 2006; however, from 1990 (6.7 percent) to 2006 there had been a
continual overall increase of 20 percent (Martin et al., 2010). Late preterm births comprise
the majority (73 percent) of all preterm births (Martin et al.), and have the best outcomes
of all preterm infants. However, this group still has increased rates of long-term morbidity
and higher mortality than infants born at term (Mathews & MacDorman, 2010). Latepreterm infants have higher frequencies of “neonatal and postneonatal morbidities, such as
respiratory distress, temperature instability, hypoglycemia, kernicterus, apnea, and feeding
problems” (Reddy, Chia-Wen, Tonse, & Willinger, 2009, p. 235), and higher rates of postneonatal rehospitalization, neonatal and post-neonatal mortality (Tomashek, ShapiroMendoza, Davidoff, & Petrini, 2007). Furthermore, Moster, Lie, and Markested (2008),
in a study of preterm infants who were followed for over 20 years, found that late preterm
infants had significantly increased rates of mental retardation, cerebral palsy, and other
major disabilities as compared to term infants. These more recent findings have led to a
shift in clinical practice; because of inherent errors in establishing due dates, induction of
labor prior to 39 weeks must now have a medical indication (American College, 2009).
Prior to that practice statement, it was common for elective inductions to occur as early as
37.5 weeks, since that was technically no longer preterm.

20

The remaining categories of degree of prematurity constitute the minority of all
preterm births. “Preterm” accounts for 21 percent, and “extreme preterm” is 6 percent of
all preterm births. The percent of all singleton births which were extreme preterm has
remained essentially unchanged from 1990 (0.61 percent) to 2007 (0.62 percent) (Martin
et al., 2010). These infants have the highest risk of mortality and long-term morbidity, with
a first-year mortality rate of greater than 40 percent (Martin et al.).
Another complexity with the pathology of preterm labor is the significant racial
disparity in rates. In 2007, the total preterm birth percent for non-Hispanic whites was
11.5, for Hispanic mothers was 12.3, but for non-Hispanic blacks was 18.3 (Martin, et al.,
2010). This gap for non-Hispanic blacks continues across the subcategories of prematurity
(Figure 1); and in fact is more than triple (1.9) that of non-Hispanic whites and Hispanics
(0.6 each) for extreme preterm births (Figure 3). This racial gap has persisted and even
widened across decades, and there is no evidence to show expectations of a narrowing of
that gap. Additionally, there has been no overarching explanation for the disparity.

21

Figure 3: U.S. Preterm Birth percents (number per 100 births in the specified category) by
Weeks Gestation and Maternal Race, 2007 (Martin et al., 2010)

Total under 37

34-36
Non-Hispanic Black
Hispanic
Non-Hispanic White

28-33

<28

0

5

10

15

20

In a large study by Sparks (2009), seven different racial/ethnic groups were
included in an investigation of preterm birth disparities. The groups included non-Hispanic
white, non-Hispanic black, US-born Mexican-origin Hispanic, foreign-born Mexicanorigin Hispanic, other Hispanic, Native American, and Asian mothers. Results showed
that “only non-Hispanic black mothers have a LBW disadvantage compared to nonHispanic white mothers” (p. 769).
Small for Gestational Age
Small for gestational age (SGA) is the descriptor given to an infant who is less than
the 10th percentile for weight for that gestational age. It is necessary to go as low as the
10th percentile to allow for genetic variations in size, especially bone length, which is one
22

of the key ultrasound determinants of fetal size. Next to prematurity, fetal growth
restriction is the second leading cause of infant mortality (Baschat, Galan, Ross, & Gabbe,
2007). Unfortunately, the National Vital Statistics Reports does not record rates of SGA,
but only LBW.
The etiologies of SGA are numerous, and are typically able to be identified, either
pre- or post-natally. The restriction in growth can be inherent (e.g., two very short-statured
parents), due to an exposure to a noxious agent during the pregnancy (e.g., cocaine,
cytomegalovirus) or a maternal medical condition (e.g., chronic hypertension, severe
anemia). These conditions result in a decrease in fetal growth due to an impairment at
some level of oxygen and nutrient delivery to the fetus. The origin of that deficiency can
be maternal (as in hypertensive disorders), placental (as in partial placenta previa), or fetal
(as in chromosomal anomalies). Compared to appropriate-for-gestational age infants,
“perinatal mortality rates in growth restricted neonates are 6 to 10 times greater; mortality
rates as high as 120 per 1000 for all cases of (SGA)… have been reported” (Baschat et al.,
2007, p. 771).
Barker (2006) published an oft-cited study in which adults who had been SGA were
identified and assessed for medical conditions.

Barker found that these adults had

increased incidence of coronary heart disease, stroke, hypertension, and Type 2 diabetes.
The study also found that “slow growth during infancy and rapid weight gain after the age
of two years exacerbated the effect of slow fetal growth…The associations are thought to
reflect the body’s plasticity during development, by which its structure and function can
be permanently changed by the intrauterine and early postnatal environment” (p. 270). The
profound life-long effects of SGA are clear.

23

Hypertension
Hypertension is the constriction of blood vessels from any of numerous causes,
resulting in higher intravascular pressure and decreased blood flow due to the narrowed
diameter of the blood vessels. Overtime, this causes intravascular damage from the
increased pressure, and end-organ damage from decreased delivery of oxygen and other
nutrients.
The degree of perinatal risk is directly proportionate to the duration of the
hypertension and the degree of hypertension; i.e., the woman who has had hypertension
the longest with the highest blood pressures is the most at risk for poor outcomes. This is
mainly due to vascular and organ damage from the elevated intravascular pressures.
Vascular damage compromises placental perfusion; the woman’s arteries provide
suboptimal perfusion to the placenta, which directly affects fetal oxygenation and
nutritional status. Maternal organ damage is found mainly in the kidneys and liver. The
kidneys are an integral part of maintaining normal hydration and fluid/electrolyte balances,
both of which are crucial to pregnancy outcomes. The liver is responsible for maintenance
and/or conversion of multiple metabolic compounds, including bilirubin, amino acids,
carbohydrates, fats, and thyroid hormones, all of which are necessary to achieve optimal
pregnancy outcomes. Hypertensive damage to the kidneys and/or the liver will initially be
temporary, but will become permanent with continued exposure to the elevated
intravascular pressure. The central nervous system (CNS) also is affected, especially in
pregnancy. In the presence of elevated blood pressures, the CNS becomes irritable and
hypersensitive to stimuli; this can lead to seizures (also known as “eclampsia” during

24

pregnancy). Seizures during pregnancy are of major concern, not only for the trauma
potential (falling, striking the abdomen/head), but also due to the lack of breathing during
an eclamptic episode which can cause hypoxia in the fetus and premature labor.
Smoking
Use of cigarettes during pregnancy has long been established as a cause of poor
pregnancy and postnatal outcomes.

In the US in 2002, prenatal cigarette use was

accountable for 5 to 8 percent of preterm deliveries, 5 to 7 percent of preterm-related
deaths, 13 to 19 percent of term infants with growth restriction, and 23 to 34 percent of
sudden infant death syndrome (Tong et al., 2013). It also increases the risk of placenta
previa, placental abruption, and premature rupture of the membranes (Centers for Disease
Control and Prevention, 2010).
There are over 7000 chemicals in cigarette smoke (Centers for Disease Control and
Prevention, 2010), many of which cause inflammation and cellular damage to lung and
vascular tissue. This creates reduced blood flow to the placenta, which is then exacerbated
by intra-placental damage from the chemicals, further reducing blood flow and nutrient
delivery to the fetus. These chemicals make the mother more at risk for hypertensive
disorders from vascular damage, and lung damage from the direct exposure to inhaled
chemicals; the outcome of each damaged system results in decreased blood flow and
oxygen delivery to the fetus.
Weathering
The concept of weathering was first introduced by Geronimus, and was described
as “the effects of social inequality on the health of populations [that] may compound with
age, leading to growing gaps in health status through young and middle adulthood that can

25

affect fetal health” (Geronimus, 1996, p. 590). In her research on birth outcomes among
blacks and whites living in poverty, increasing maternal age (from 15 years old forward)
was significantly positively related to the rates of LBW for black women, but not for
whites. It also was found that a “notable share of this interaction effect [is] explained by
the measured maternal health characteristics…consistent with the theoretical perspective
that among the socioeconomically disadvantaged, black women’s health deteriorates more
rapidly over the young adult ages” (Geronimus, p. 594). These same findings have been
replicated by several subsequent studies (Collins et al., 2008; Geronimus, Bound, &
Waidmann, 1999; Holzman et al., 2009; Lu & Halfon, 2003; Rauh, Andrews, & Garfinkel,
2001). However, all of these studies have investigated the application of the weathering
concept to only black and white populations; Hispanics have not been included.
Goldenberg et al. (1996) studied almost 1500 singleton pregnancies in black and
white women with lower socioeconomic status, analyzing multiple medical, psychosocial,
and behavioral risk factors and their association with the LBW racial disparity. This study
found that “many of the risk factors for low birth weight were more common among white
women than black women. Nevertheless, black women had more infants born preterm,
with growth restriction, and with low birthweight than did white women” (p. 1317). The
concept of weathering, although not discussed in the study, could easily be applied and
tested to help explain these findings.
Support of a racial difference in weathering is not universal. Borders, Bryant,
Grobman, Amsden, and Holl (2007), in a study of chronic psychosocial stress and LBW in
a low-income population, found the only demographic factor which was associated with
LBW was increasing maternal age, regardless of race. They also found multiple

26

psychosocial factors (food insecurity, a child with chronic illness in the home, increased
crowding in the home, unemployment, poor coping skills) were all significantly associated
with LBW even after adjusting for maternal age.
Marital Status
The presence of a support system is a desired entity for anyone, especially a
pregnant woman. While marital status has not been extensively studied, there are research
findings which support its importance in birth outcomes.
Marital status can be used as a measure of parental relationship; it is not as
descriptive as qualitative data, but has been used to infer a support system. Unmarried
women are less likely than married women to report they and their partner both desired the
pregnancy (Williams, 1994), and are more likely to be lower-income, lower-educated, have
their relationship with their significant other dissolve, (Carlson, McLanahan, & England,
2004; Smock, 2000), and have low birthweight infants (Barrington, 2010; Bird, Chandra,
Bennett, & Harvey, 2000; Hohmann-Marriott, 2009; Masho et al., 2010; Sullivan, Raley,
Hummer, & Schiefelbein, 2012). Women who reported their pregnancy as unintended also
were more likely to have inadequate prenatal care (Hohmann-Marriott).
None of these studies included the combination of race/ethnicity, maternal age, and
poverty status to investigate variations in the above findings. This is a gap in the literature.
Marriage has been shown to be associated with improved overall physical and
mental health (Koball, Moiduddin, Henderson, Goesling, & Besculides, 2010; Schoenborn,
2004; Wood, Goesling, & Avellar, 2007). Higher rates of cancer survival (Baine et al.,
2011; Mahdi et al., 2013; Wang, Wilson, Stewart, & Hollenbeak, 2011), and better control
of diabetes (Lister, Fox, & Wilson, 2013; Seidel, Franks, Stephens, & Rook, 2012) are

27

examples of health benefits experienced by people who are married. Specifically of import
to this study, being married has been shown to be protective against cardiovascular
complications (Baker et al., 1999; Floud et al., 2014: Idler, Boulifard, & Contrada, 2012;
Kiecolt-Glaser & Newton, 2001; King & Reis, 2012; Robles & Kiecolt-Glaser, 2003;
Zhang & Hayward, 2006). King and Reis also included marital satisfaction for those who
were married, and found those who were not only married, but in a highly satisfying
marriage, had the best long-term survival rates after cardiac surgery. They cited the
combination of spousal support and motivation to adopt healthy lifestyle behaviors as key
factors in this improved outcome.
Without qualitative data to describe the level of support in any relationship, marital
status is used in many studies as a measure of support. Marriage may be another
mechanism by which mothers receive social/emotional and other types of support during
and after the pregnancy, and that support may mitigate some of the detrimental effects of
living in concentrated poverty. Marriage rates are changing rapidly in America, making
timely application of research findings difficult.

28

Chapter Three
Research Design
Data Set
The data set being used for this secondary analysis is the 2012 Natality Public Use
Dataset, compiled annually by the Centers for Disease Control and Prevention (CDC) from
recorded birth certificate data from all 50 states and the District of Columbia. Each state’s
registration area transmits those data electronically to the CDC’s National Center for
Health Statistics.
The data are limited to those US residents who give birth within the 50 states and
the District of Columbia. Data from US residents who give birth outside the US are not
included.
Inclusion criteria for this study are first-time live birth mothers, who are having a
singleton birth (i.e., no twins, triplets, or higher order multiples). They are listed as having
Medicaid as their primary payment source, and are between the ages of 15-44 years. Their
race/ethnicity is listed as non-Hispanic Black, non-Hispanic White, and Hispanic. From
that pool, any listed fetal anomaly or birth defect was cause for exclusion, due to altered
growth issues.
Application for approval from Michigan State University’s IRB was submitted.
The IRB application #i043927 was determined to be exempt from human subject research
criteria.

29

Definitions
Low birthweight (LBW): infant whose birthweight is <2500grams (5 pounds 8 oz.),
regardless of gestation
Mother’s Age: 15-44 years of age included in this study
Pre-existing hypertension (chronic hypertension): hypertension which has been diagnosed
prior to conception of this pregnancy
Pregnancy-induced hypertension (also known as pre-eclampsia): a condition which is
newly diagnosed during this pregnancy and is due to pregnancy.
Marital Status: data classified into “married” or “not married”
Smoking: use of cigarettes at any time during the pregnancy
Poverty: for this study, those mothers receiving Medicaid insurance

This dissertation is a secondary analysis of birth certificate data from the fifty states
and the District of Columbia. Birth certificates are required for every infant born alive,
regardless of gestation. Since the vast majority of births in America occur within hospitals,
the completion of the birth certificate occurs in the hospital prior to discharge. There are
two parts to the completed birth certificate: the abbreviated form and the complete form.
The mother (and father, if present) are given the abbreviated form to complete, which lists
parental birth dates, ethnicity/race, and the name for the newborn. The birth attendant
completes the remaining portions, which include pertinent maternal medical history, and
specific birth information (e.g., complications with labor and/or delivery, newborn
complications). The hospital then files the completed birth certificate with the state, where

30

it is normally housed by the state’s Department of Community Health, Division for Vital
Records and Health Statistics. Each state then electronically transmits that data to the
CDC, where it is compiled into a national data set.
This study will include only first-time births, since a history of preterm labor,
preterm birth, or LBW is the largest single predictor of recurrence. The racial/ethnic groups
to be included are non-Hispanic black, non-Hispanic white, and Hispanic. Multiple
gestations will be excluded as preterm labor is expected as well as the subsequent LBW,
and those infants with congenital anomalies or malformations.

Maternal hypertension

will be divided into those conditions which were present prior to the pregnancy, and those
conditions which developed during the pregnancy or are classified as pregnancy-related.
Chronic hypertension places pregnancies at risk for alterations in placental perfusion due
to damage to the maternal vascular system prior to conception, while pregnancy-induced
hypertension is a serious acute condition which can cause sudden and severe alterations in
blood flow to the placenta and subsequently to the fetus. Hypertensive conditions will be
analyzed with increasing age, to test the concept of weathering.
Marital status will be included. Birth certificates list single, married, divorced, and
unknown as options. This study sorted those options into “married” and “not married” for
purposes of statistical analyses.
Descriptive statistics will be compiled separately for Blacks, Whites and Hispanics.
SPSS programming will be used for the descriptive and statistical analyses. Binary logistic
regression will be utilized to study racial disparities in LBW. Confidence intervals will
be set at 95 percent.

31

Chapter Four
STATISTICAL ANALYSIS
This chapter starts with the discussion of the descriptive statistics of the study
participants. The second part describes the results of the statistical analysis of the five
research questions.
Descriptive Statistics
The

2012

Natality

File,

found

on

the

CDC

website

(http://www.cdc.gov/nchs/data_access/VitalStatsOnline.htm), contains the records on all
birth certificates from the year 2012 that were filed in the fifty states and the District of
Columbia. From that entire database (N= 1,569,943), cases were initially narrowed to
those of mothers between the ages of 15-44 who were having their first live birth, and were
using Medicaid as their primary payment source (N=542,455) (Table 1). The cases were
then narrowed to include the races/ethnicities Non-Hispanic White (hereafter referred to as
“White”), Non-Hispanic Black (“Black”), and Hispanic, as the three target groups in this
study. Additional eliminations were then done for plurality (a pregnancy with more than
one fetus) and those birth defects which are listed on the birth certificate. It is necessary
to exclude plurality due to the high risk of premature labor and premature delivery, thus
resulting in a low birth weight infant which may have nothing to do with the effects of
living in poverty. Similarly, birth defects and chromosomal anomalies often result in
overall lack of normal growth, resulting in a low birth weight infant regardless of
environmental resources. This left a final N of 499,323 for inclusion in this study. These
final cases included Blacks (N= 116,732), Whites (N= 222,938), and Hispanics

32

(N=159,653) (Table 2), which are self-designated by the mother when she completes the
birth certificate after the birth of her child.
Table 1: Birth Certificate Variables Excluded in Study
BIRTH CERTIFICATE VARIABLE

N

Total Births in US, 2012

1,569,943

Eliminate other than Maternal Age 15-44, First -1,027,488
Live Birth, On Medicaid
Eliminate other than Non-Hispanic White, -34,004
Non-Hispanic Black, Hispanic
Eliminate Plurality
-6450
Eliminate Spina Bifida & Anencephaly

-1671

Eliminate Oomphalocele & Gastroschisis

-426

Eliminate Cleft Lip &/Or Palate

-357

Eliminate Diaphragmatic Hernia

-71

Eliminate Down Syndrome

-153

FINAL N OF CASES TO BE INCLUDED

499,323

Table 2: Frequency of Racial/Ethnic Groups Included in Study
RACIAL/ETHNIC CATEGORY

N

%

Non-Hispanic White

222,938

44.6%

Hispanic

159,653

32.0%

Non-Hispanic Black

116,732

23.4%

TOTAL

499,323

100%

33

The overall rate of LBW for women in this study is consistent with national birth data,
which has consistently shown Black women to have higher rates (12.5 per 100 live births)
than either Whites (7.3) or Hispanics (7.0). True to the epidemiologic paradox, Hispanics
have the lowest rate of LBW in this study despite living in increased poverty.
Table 3: Rate of LBW by Racial/Ethnic Group (Rate per 100 live births)
GROUP
Black
White
Hispanic

LBW
12.5
7.3
7.0

For all mothers, the age range was 15-44 years, and the mean, median, and mode for each
group were very similar, with Whites having just a slightly higher average age at first birth
(approximately 5 months older than the Hispanic average age, and 6 months older than the
Black average age).
Table 4: Descriptive Statistics on Maternal Age in Years by Racial/Ethnic Group
GROUP
Black
White
Hispanic

MEAN
22.10
22.74
22.20

MEDIAN
21
22
21

MODE
20
20
19

However, the other key variables in this study, marital status and smoking, showed large
discrepancies between groups. Specifically, Blacks were far less likely (9.6%) to be
married than either Whites (26.0%) or Hispanics (24.0%). Also, Whites were substantially
more likely to be smokers (23.1%) than either Blacks (5.4%) or Hispanics (2.1%).

34

Table 5: Descriptive Statistics of Marital Status and Smoking by Racial/Ethnic Group
GROUP
Black
White
Hispanic

MARRIED
9.6%
26.0%
24.0%

SMOKER
5.4%
23.1%
2.1%

Chronic hypertension (CHTN) (hypertension which is diagnosed and present prior
to pregnancy) and pregnancy-induced hypertension (PIH) (high blood pressure first
diagnosed during pregnancy) are both important complications of pregnancy. Blacks are
more likely than either Whites or Hispanics to have CHTN or PIH (Table 6). True again
to the epidemiologic paradox, Hispanics have the lowest pre-existing hypertension when
becoming pregnant, and are still less likely than either Whites or Blacks to have PIH.
Table 6: Percentage of Births with CHTN and PIH by Racial/Ethnic Group
GROUP
% with CHTN % with PIH
Black
1.9
7.2
White
1.1
6.4
Hispanic
0.6
4.6

Smoking is a well-known risk factor for hypertensive disorders. To analyze the
number of women in this study who smoked and the frequency of having either CHTN or
PIH, a cross-tabulation was run. Within the CHTN variable (Table 7), White women who
smoked were the most likely to have CHTN; approximately 6 times more likely than
Hispanics who smoked, and approximately 3 times more likely than Black women who
smoked. These percentages are similar to the results in Table 4, which showed the much
higher frequency of smoking amongst White women in this study when compared to

35

Blacks and Hispanics. Smoking was more apt to be associated with the presence of CHTN
in all three racial/ethnic groups.
Table 7: Cross Tabulation of Frequency of Presence/Absence of CHTN and Smoking
Within CHTN Variable
WITHIN CHTN
BLACK
WHITE
HISPANIC

CHTN
NO CHTN
CHTN
NO CHTN
CHTN
NO CHTN

% NON-SMOKER
91.3
94.3
74.2
75.3
95.6
97.9

% SMOKER
8.7
5.7
25.8
24.7
4.4
2.1

When looking within the smoking variable (Table 8), racial/ethnic differences were
seen. Of all women who smoked, Black women had the highest percentage of CHTN
(almost 3%), more than double that of Whites (1.2%) and Hispanics (1.3%). Among the
non-smoking women, Hispanics had the lowest percentage of CHTN (0.6%), compared to
non-smoking Whites (1.1%) and non-smoking Blacks (1.9%).
Table 8: Cross Tabulation of Frequency of Presence/Absence of CHTN and Smoking
Within Smoking Variable
WITHIN SMOKE
BLACK
CHTN
NO CHTN
WHITE
CHTN
NO CHTN
HISPANIC
CHTN
NO CHTN

% NON-SMOKER
1.9
98.1
1.1
98.9
0.6
99.4

36

% SMOKER
2.9
97.1
1.2
98.8
1.3
98.7

When cross-tabulation was utilized to look within the PIH variable for frequencies of
smoking, similar results were obtained. White women were the most likely to be smokers,
whether PIH was present (21.7%) or not (25.0%), which is more than three times higher
than rates of smoking amongst Blacks, and approximately 10 times higher than rates of
smoking amongst Hispanic women.
Table 9: Cross Tabulation of Frequency of Presence/Absence of PIH and Smoking
Within PIH Variable
WITHIN PIH
BLACK
WHITE
HISPANIC

PIH
NO PIH
PIH
NO PIH
PIH
NO PIH

% NON-SMOKER
93.7
94.3
78.3
75.0
97.7
97.9

% SMOKER
6.3
5.7
21.7
25.0
2.3
2.1

Within the smoking variable, smoking Black women had the highest percentage of
PIH (7.9%), compared to smoking White women (5.7%) and smoking Hispanic women
(5.1%). Interestingly, PIH was less common among White women who smoked (5.7%)
vs. White women who did not smoke (6.7%).
Table 10: Cross Tabulation of Frequency of Presence/Absence of PIH and Smoking
Within Smoking Variable
WITHIN SMOKING
BLACK
WHITE
HISPANIC

PIH
NO PIH
PIH
NO PIH
PIH
NO PIH

% NON-SMOKER
7.3
92.7
6.7
93.3
4.6
95.4

37

% SMOKER
7.9
92.1
5.7
94.3
5.1
94.9

RESULTS OF THE LOGISTIC REGRESSION
(Application of the Research Model)
Logistic regression was used to analyze the research questions. The models
estimated (Paths A, B, C, and D) are described below. To reiterate, only first time singleton
births for Black, White, and Hispanic women on Medicaid are included in this study, and
each racial/ethnic group were run as a separate model.

Question #1: What is the effect of increasing maternal age on the odds of having a
LBW infant controlling for marital status and smoking (Path A)?
The results from the first research question are presented in Table 11.
Interestingly, the odds of having an LBW infant with every year of age increase was almost
the same for Blacks (OR 1.025, 95% CI 1.021-1.029) and for Whites (OR 1.024, 95% CI
1.021-1.028) controlling for marital status and smoking. This finding does not support the
weathering hypothesis, but will be discussed in Chapter Five. Hispanic women showed a
significant but lower risk (OR 1.018, 95% CI 1.014-1.022). This finding corresponds to
the epidemiologic paradox for Hispanics. Being married was protective for all racial/ethnic
groups with Black mothers receiving the greatest protection (OR .774, 95% CI .724-.827).
For all three groups, smoking also significantly increased the odds of having a LBW infant
with White mothers having the highest odds (OR 1.807, 95% CI 1.745-1.872).
findings were significant at p=.000.

38

All

Table 11: Effect of Increasing Age on the Odds of Having LBW
Controlling for Marital Status and Smoking

AGE

SMOKING

MARITAL

OR
CI
p
OR
CI
p
OR
CI
p

BLACK
1.025
1.021, 1.029
.000
1.606
1.503, 1.716
.000
.774
.724, .827
.000

WHITE
1.024
1.021, 1.028
.000
1.807
1.745, 1.872
.000
.911
.875, .948
.000

HISPANIC
1.018
1.014, 1.022
.000
1.749
1.568, 1.952
.000
.896
.854, .940
.000

Hypothesis #1: Black women will have the strongest effect of increasing age on the odds
of having a LBW infant. This hypothesis is rejected, as there is almost no difference
between Black and White women.
Hypothesis #2: Smoking will increase the odds of LBW for all racial/ethnic groups. This
hypothesis is accepted.
Hypothesis #3: Being married will decrease the odds of LBW for all racial/ethnic groups.
This hypothesis is accepted.

39

Question #2: What is the effect of increasing maternal age on the odds of having
hypertension (CHTN or PIH) while controlling for marital status and smoking
(Path B)?
The results (Table 12) show that for all three groups, as age increased, so did the
odds of CHTN, controlling for marital status and smoking. The greatest odds were for
Blacks to develop CHTN with increasing age (OR 1.133, 95% CI 1.124-1.141). In other
words, for every year of age increase, Black women were 13 percent more likely to have
chronic hypertension, Hispanic women were 12 percent more likely, and White women
were 10 percent more likely. Being married was significantly protective against CHTN
for every year of increasing age for both Black (p=.000) and White (p=.013) mothers.
Again Black mothers were the most protected by being married. Being married does not
significantly protect Hispanic women from CHTN. This may be due to the epidemiologic
paradox. Smoking was found to be significantly associated with CHTN for Hispanic and
Black mothers at p=.000, and still significant for Whites but slightly less so at p=.010.
Table 12: Effect of Increasing Age on the Odds of CHTN
Controlling Marital Status and Smoking
CHTN
AGE

SMOKING

MARITAL

OR
CI
p
OR
CI
p
OR
CI
p

BLACK
1.133
1.124, 1.141
.000
1.404
1.203, 1.639
.000
.710
.620, .813
.000

WHITE
1.101
1.093, 1.108
.000
1.132
1.030, 1.243
.010
1.120
1.024, 1.225
.013

40

HISPANIC
1.122
1.111, 1.133
.000
2.251
1.652, 3.067
.000
.953
.827, 1.098
NS

For PIH, the results (Table 13) show that for all three groups, as age increased, so
did the odds of PIH, controlling for marital status and smoking. The greatest odds were
for Blacks (OR 1.023, 95% CI 1.017-1.028) and Hispanics (OR 1.024, 95% CI 1.0191.029) to develop PIH with increasing age, with slightly lower odds for Whites (OR 1.017,
95% CI 1.013-1.021). In other words, for every year of age increase, Black and Hispanic
mothers were more than 2 percent more likely to have PIH, while Whites were slightly
more than 1.5 percent more likely. Importantly, being married was protective of PIH for
Black mothers only. Smoking was significantly associated with increased odds of PIH for
White mothers only. This finding needs further investigation, as physiologically, smoking
is a known risk factor for vasoconstriction and hypertensive disorders including CHTN as
above.
Table 13: Effect of Increasing Age on the Odds of Development of PIH
Controlling for Marital Status and Smoking
PIH
AGE

SMOKING

MARITAL

OR
CI
p
OR
CI
p
OR
CI
p

BLACK
1.023
1.017, 1.028
.000
1.076
.979, 1.183
NS
.885
.815, .960
.003

WHITE
1.017
1.013, 1.021
.000
.841
.806, .878
.000
1.039
.998, 1.082
NS

HISPANIC
1.024
1.019, 1.029
.000
1.126
.961, 1.321
NS
1.017
.961, 1.077
NS

Hypothesis #4: Black women will have the strongest effect of increasing age on the odds
of having both chronic and pregnancy-induced hypertension. This hypothesis is rejected.

41

Hypothesis #5: Smoking will increase the odds of having LBW for all racial/ethnic groups
for both CHTN and PIH. This hypothesis is accepted for the CHTN model, and rejected
for the PIH model.

42

Question 3: What is the effect of hypertensive disorders on the odds of
having LBW, controlling for smoking and marital status (Path C)?
Not unexpectedly, these results showed a significant increase in having a LBW
infant in the presence of CHTN for all three groups (Table 14). Both Blacks (OR 2.566,
95% CI 2.326-2.831) and Whites (OR 2.611, 95% CI 2.339-2.913) had greater than a 2.5
increase in odds of having a LBW infant in the presence of CHTN; Hispanics (OR 3.456,
95% CI 2.956-4.041) had more than a three-fold increase. Marital status was significantly
protective against LBW in the presence of CHTN for Blacks (OR .871, 95% CI .817-.928;
p=.000) and Hispanics (OR .940, 95% CI .897-.985; p=.009), and still significant but less
so for Whites (OR .962, 95% CI .925-1.000; p=.050). When CHTN was present, smoking
significantly (p = .000) increased the odds of LBW across all three racial/ethnic groups.
Table 14: Effect of CHTN on the Odds of Having LBW Controlling for Marital Status
and Smoking
CHTN
CHTN

SMOKING

MARITAL

OR
CI
p
OR
CI
p
OR
CI
p

BLACK
2.566
2.326, 2.831
.000
1.625
1.521, 1.737
.000
.871
.817, .928
.000

WHITE
2.611
2.339, 2.913
.000
1.807
1.744, 1.871
.000
.962
.925, 1.000
.050

HISPANIC
3.456
2.956, 4.041
.000
1.728
1.548, 1.930
.000
.940
.897, .985
.009

In the PIH path (Table 15), all three groups were statistically significantly at
increased odds for having a LBW infant when PIH was present. Blacks (OR 2.807, 95%
CI 2.660-2.962) had increased odds of more than 2.5; Whites (OR 2.818, 95% CI 2.682-

43

2.961) had almost the same odds as Blacks. Hispanics (OR 3.469, 95% CI 3.257-3.694)
had the largest risk, with increased odds of 3.5 of having an LBW infant when PIH was
present. Being married was protective for Blacks (p=.000) and Hispanics (p=.005), but
less significant for Whites (p=.035). Smoking was again shown to have significant (p =
.000) effect on the odds of LBW when PIH was present for all three groups, demonstrating
the importance of smoking reduction interventions.
Table 15: Effect of PIH on the Odds of Having LBW Controlling for Marital Status and
Smoking
PIH
PIH

SMOKING

MARITAL

OR
CI
p
OR
CI
p
OR
CI
p

BLACK
2.807
2.660, 2.962
.000
1.642
1.535, 1.756
.000
.879
.825, .937
.000

WHITE
2.818
2.682, 2.961
.000
1.845
1.781, 1.912
.000
.958
.921, .997
.035

HISPANIC
3.469
3.257, 3.694
.000
1.745
1.562, 1.950
.000
.936
.893, .981
.005

Hypothesis #6: Black women with CHTN will have the strongest effect of CHTN on the
odds of having LBW. This hypothesis is rejected.
Hypothesis #7: Being married will decrease the odds of having low birthweight for all three
racial/ethnic groups with CHTN or PIH. This hypothesis is accepted.

44

Question #4: What is the effect of increasing maternal age on the odds of having
LBW, controlling for hypertensive disorder, marital status and smoking (Path D)?
When Path A was re-analyzed by adding the presence of a hypertensive disorder
(CHTN or PIH) to the controls, there were only very slight changes in the odds of
increasing age on LBW (Tables 16 and 17) suggesting that hypertensive disorders are not
strong mediators in the age and LBW relationship for Black, White and Hispanic mothers
on Medicaid. Hispanic women had the least increase with each year of increasing age,
which supports the epidemiologic paradox. As expected, the presence of CHTN or PIH
increased the odds of LBW with every year of increase in maternal age. Hispanic women
had the highest odds of LBW with the presence of CHTN (Hispanic OR 3.265, 95% CI
2.790-3.820; Black OR 2.393, 95% CI 2.167-2.643; White OR 2.460, 95% CI 2.203-2.747)
or with the presence of PIH (Hispanic OR 3.439, 95% CI 3.229-3.663; Black OR 2.782,
95% CI 2.636-2.936; White OR 2.794, 95% CI 2.660-2.936). In other words, for every
increased year of age, Hispanic women who had CHTN or PIH had more than a three-fold
increase in the odds of having LBW, which are higher odds than Black or White women
with either hypertensive disorder.
Being married did continue to significantly decrease the odds of having a LBW
infant in every group when CHTN was present (Black OR .781, 95% CI .730-.835; White
OR .909, 95% CI .873-.946; Hispanic OR .898, 95% CI .856-.942) as well as when PIH
was present (Black OR .779, 95% CI .728-.833; White OR .905, 95% CI .869-.942;
Hispanic OR .894, 95% CI .856-.942).

These are new findings, which have not been

reported in the literature.

45

Smoking remained significantly associated with increased odds of LBW in both
CHTN and PIH models. All findings were significant at p=.000.
Table 16: Effect of Increasing Age on the Odds of Having LBW,
Controlling for CHTN, Marital Status, and Smoking
CHTN
AGE

SMOKING

MARITAL

CHTN

OR
CI
p
OR
CI
p
OR
CI
p
OR
CI
p

BLACK
1.021
1.017, 1.025
.000
1.592
1.489, 1.702
.000
.781
.730, .835
.000
2.393
2.167, 2.643
.000

WHITE
1.022
1.019, , 1.026
.000
1.805
1.743, 1.870
.000
.909
.873, .946
.000
2.460
2.203, 2.747
.000

HISPANIC
1.015
1.011, 1.019
.000
1.723
1.543, 1.924
.000
.898
.856, .942
.000
3.265
2.790, 3.820
.000

Table 17: Effect of Increasing Age on the Odds of Having LBW,
Controlling for PIH, Marital Status, and Smoking
PIH
AGE

SMOKING

MARITAL

PIH

OR
CI
p
OR
CI
p
OR
CI
p
OR
CI
p

BLACK
1.023
1.019, 1.027
.000
1.603
1.499, 1.714
.000
.779
.728, .833
.000
2.782
2.636, 2.936
.000

WHITE
1.023
1.019, 1.026
.000
1.842
1.778, 1.909
.000
.905
.869, .942
.000
2.794
2.660, 2.936
.000

46

HISPANIC
1.015
1.011, 1.019
.000
1.740
1.557, 1.944
.000
.894
.856, .942
.000
3.439
3.229, 3.663
.000

Hypothesis #8: All three racial/ethnic groups will have increased odds of LBW with
increasing age in the presence of chronic hypertension. This hypothesis is accepted.
Hypothesis #9: Black women with CHTN will have the strongest effect of increasing age
on the odds of having low birthweight. This hypothesis is rejected.
Hypothesis #10: All three racial groups will have decreased odds of LBW if they are
married. This hypothesis is accepted.

There were both expected findings (e.g., the increased odds of LBW in the presence
of CHTN) and unexpected findings (e.g., Black women not having higher odds of LBW
than White women) in these results. It is important to note that the findings on marital
status and the results on Hispanics are both are contributions to the literature which have
not yet been reported in this subject area. All findings are further discussed in the following
chapter.

47

CHAPTER FIVE
CONCLUSIONS
This chapter includes the discussion and implications of the research findings,
theoretical applications, limitations of this study, and implications for future research. It
also includes implications for health care professionals and legislative officials who control
budgets and make policies regarding health care.
This study sought to explore the relationship of living in poverty –i.e., being on
Medicaid, and the odds of having a LBW infant, through a secondary analysis of all birth
certificates filed in the United States in 2012 and correlated into a public data set owned
by the CDC (http://www.cdc.gov/nchs/data_access/VitalStatsOnline.htm). Included in
this research were first-time mothers living in poverty (as defined by having Medicaid for
insurance) who were non-Hispanic White (45% of total N in this study), non-Hispanic
Black (23%), and Hispanic (32%). There are multiple confounding factors listed on birth
certificates which raise the risk of having an LBW infant that have nothing to do with living
in poverty per se; these factors (e.g., congenital anomalies, twins, chromosomal
abnormalities) were cause for elimination from the study. The final number of cases
included in this study was just under one-half million (N=499,323), which allows for
generalizability based on the large sample size.
The frequency of LBW was highest in the Black women (13% of Black women in
this study, vs. 7% of included White and Hispanic women), which corresponds to national
statistics and speaks to the disturbing continued disparity in Black and White birth
outcomes. These LBW infants face increased odds of poor outcomes such as death in the
first year of life and adult co-morbidities such as diabetes and heart disease. The need for
further understanding and prevention of LBW has obvious significance for not only these
48

infants as individuals, but also for healthcare costs and budgetary considerations for
America in general. A significant reduction in the rate of hypertension and diabetes would
cause a profound reduction in healthcare costs in America.
Weathering, as a sociologic construct, is defined as a more rapid aging process
brought on by the chronic stress of living in poverty. Decreased access to amenities (e.g.,
heat, proper nutrition, clean water), as well as limited access to health care (geographic and
financial availability) can cause exacerbation of current health issues, failure to recognize
the onset of new health conditions, and lack of treatment for diagnosed conditions. Work
by previous researchers has shown that Black women weathered faster than White women
living in similar poverty, and that there were disparate pregnancy outcomes related to that
weathering process. To date, Hispanic women had not been included in any of that
research. As the fastest growing racial/ethnic group in America, it is important to include
Hispanics in such comparative research.
Human ecology theory studies the interaction of humans and their environment.
Family ecology, then, looks specifically at families as the unit of interest. In this study, the
family was defined as the mother, her newborn infant, and a potential spouse (if listed on
the birth certificate). The environment included in this study was that of living in poverty.
Having Medicaid insurance was used as the proxy for living in poverty, since there are no
income data on a birth certificate.
Marriage has been shown to be associated with better health in the literature. This
study included marital status as a variable, to see if it provided a protective effect against
the risk of having a LBW infant for these women living in poverty. Of the women in this

49

study, 10 percent of Blacks, 26 percent of Whites, and 24 percent of Hispanic women were
listed as “married” on the birth certificate.
With the increase in acceptance of inter-racial/mixed ethnicity marriages, the issue
of racial/ethnic homogenicity becomes complex. Some of the women in this study, are no
doubt not homogeneously Black or White or Hispanic; but that is a limitation of the data
available on the birth certificates currently used. The implications, then, for interpretation
of racial/ethnic results and comparisons becomes blurred, and will become more so with
increased societal acceptance and frequency of heterogeneous couples.
In the descriptive statistics, it was found that the average age at first birth was only
slighter (approximately six months) older for Whites than for Blacks or Hispanics. This
defies public perception of many conservatives that low SES Blacks are having babies at a
much younger age than similar SES Whites. Instead, it speaks directly to the need to have
pregnancy and women’s health services universally available to all women, especially
those living in poverty.
There is a striking difference in the large percentage of White mothers (23%) vs.
Black (5%) or Hispanic (2%) who smoke. Smoking has been shown repeatedly, for
decades, to have deleterious effects on a fetus and newborn, not to mention the woman
herself. Qualitative research could help here, to investigate why this large disparity exists,
and to see why Black and Hispanic women have lower rates. Did they smoke prior to the
pregnancy and quit with the affirmation of pregnancy- and if so, what influenced them to
stop smoking? Are cessation programs and therapies being equally offered and made
available to these women?

The American College of Obstetricians and Gynecologists

(American College, 2010) recommends that all pregnant women be extensively counseled

50

regarding the dangers of smoking during pregnancy and that on-going assessment and
encouragement be given throughout the pregnancy and post-partum period, since 50-60%
of those women who quit smoking during pregnancy will resume smoking within the first
year post-partum (American College). This has implications for the next pregnancy, and
the goal of being smoke-free prior to becoming pregnant the next time. Women must be
effectively counseled, encouraged, and coached in order to increase the success of
cessation programs. Too often, in a busy obstetric practice, the lengthy time necessary to
counsel women on smoking is not available, and the opportunity to effect change is lost.
“Patient Education” within the context of an office visit is not a reimbursable event. If
patient education is done as a separate event, such as a class on smoking cessation, it can
then be billed but it is rare to receive third-party reimbursement for such a class. Typically,
these type of classes are funded by the patient having to pay cash to be able to enroll, which
automatically eliminates this as an option for women living in poverty.

Insurance

companies must place more value on this preventative teaching, and develop
reimbursements specific to smoking cessations to allow healthcare providers the fiscal
availability of time spent on this educational efforts.
DISCUSSION OF THE RESULTS OF THE RESEARCH QUESTIONS
This study looked at first-time mothers living in poverty and the relationships
between increasing maternal age, hypertensive disorders, smoking, and marital status, and
how those factors affected the odds of having an LBW infant. The results are discussed in
order of the research questions.

51

Question #1: What is the effect of increasing maternal age on the odds of having
LBW, controlling for marital status and smoking (Path A)?
The effect of increasing maternal age on the odds of having a LBW infant were
consistent with current literature. While controlling for marital status and smoking,
increasing age did increase the odds of LBW in each racial/ethnic group. This supports the
weathering theory in general; women living in poverty (even younger women of
reproductive age) have higher odds of health issues for every year they age, and that would
include higher odds of having a LBW infant. What was unexpected was the lack of
difference between those increased odds for Black and White women. According to the
weathering research, Blacks have been found to weather faster, and therefore it was
unexpected to find that Blacks had a 2.5 percent increase in LBW for every year they aged
when compared to Whites who had a 2.4 percent increase. Hispanics had an expected
result of lower odds than either Whites or Blacks, with a 1.8 percent increase in LBW for
every year of increased maternal age. While the Hispanic finding is supported by the
epidemiologic paradox, the lack of difference between White and Black is not supported
by the weathering concept and is unexplained. This may be due to all women in the study
already being in poverty, and not compared to the general population rates of LBW for
each specific race/ethnicity. Further investigation is warranted to explore this finding.
Smoking was significantly associated with increased odds of LBW for all three
racial/ethnic groups. This is an expected finding based on the physiologic effects of
smoking, causing potent vasoconstriction and resulting in decreased maternal blood supply
to the placenta, thereby reducing oxygenation and nutritional supply to the fetus. The
combination of increasing maternal age (aging causes increased chance for

52

vasoconstriction due to loss of elasticity in the blood vessels) in the presence of smoking
should increase odds of LBW, and that was shown in these results. It is important to reiterate the profound difference in rates of smoking amongst the three groups; White women
were more than four times more likely than Black women, and more than ten times more
likely than Hispanic women to smoke during the pregnancy. Targeting smoking cessation
strategies for areas with higher White populations would be most cost-effective based on
these results, but is important for all women of childbearing age.
Being married was significantly protective against having a LBW for all three
racial/ethnic groups. Within the limitations of secondary analyses, no further details can
be gleaned from what benefit “married” is truly providing. This is congruent with prior
research which has shown decreased rates of health issues such as hypertension and
cardiovascular diseases for married people (see Chapter 2). It is postulated here that the
benefits obtained from being married are multifactorial- financial, emotional, and logistic
(e.g., having someone to help with errands, care of the home), all of which would lower
stress, thereby decreasing the risk for high blood pressure and other stress-mediated
illnesses. However, qualitative research is needed to go beyond the legal label of “married”
and further define what is truly gained from that relationship. Certainly there are many
couples who co-habitate without legal coupling and receive those same benefits, but
without qualitative research or other designations in a research model, they are still
considered “single” in studies bound by data which does not include some type of
“supportive relationship” designation as an option.

53

Question #2: What is the effect of increasing maternal age on the odds of having
hypertension (CHTN or PIH) controlling for marital status and smoking (Path B)?
Aging, as a physiologic process, increases the risk of hypertension. Overall, Blacks
have higher risks than Whites and Hispanics (National Heart, Lung, and Blood Institute,
2012). This model showed that for women in this study (15-44 years of age), every year
of increased age was significantly associated with increased odds of having CHTN or
developing PIH regardless of race/ethnicity, when controlling for smoking and no prenatal
care. Comparatively, Blacks had a slightly higher chance of CHTN (13%) than Hispanics
(12%) or Whites (10%) for every year of increasing age. These results are supportive of
the weathering hypothesis when comparing Blacks and Whites, showing that over multiple
years, Blacks are developing CHTN at faster rates than Whites.
The fact that Hispanics had almost the same odds as Blacks needs further
investigation. In this study, Hispanics were classified as a single group (as coded by the
CDC in this data set). The birth certificate does list different ethnicities within this group
that were not used in this study (e.g., Mexican, Puerto Rican). There are known differences
in these Hispanic ethnicities in regards to hypertensive disorders. For instance, Puerto
Ricans have higher death rates from hypertension than either Whites or Blacks, but Cubans
have lower death rates than either Whites or Blacks (National Heart, Lung, and Blood
Institute, 2012). The constitution of “Hispanics” in this study, as a group, is not known,
but it is acknowledged that it is a multi-ethnic group which confounds these findings and
their interpretation.
Smoking was significantly associated with increased odds of CHTN with
increasing age for all three groups as well. This corresponds to general population risk of

54

smoking and developing hypertensive disorders. As previously discussed, the chemicals
in cigarettes are potent vasoconstrictors; for anyone with a proclivity toward hypertension,
or in the presence of overt hypertension, smoking will exacerbate the narrowing of the
blood vessels and any outcomes of that constriction.
Being married conveyed significant protection against CHTN for Black mothers,
and only mild protection for White mothers. It was not a significant factor for Hispanic
mothers. This may be influenced by the much lower rate of marriage among Black women
when compared to White and Hispanic women. Again, many Black women receive
benefits from being in committed relationships without being able to check the “married”
box on the birth certificate.
In the PIH arm of this model, chances of developing PIH were almost equal across
the three groups. With every year of increased age, Blacks had a 2.3 percent higher chance
of PIH; Whites had a 1.7 percent higher chance, and Hispanics had a 2.4 percent increase.
An interesting finding in this PIH model, however, is that smoking was only a significant
finding for every year of increasing age in White women. This may be related to the much
higher rate of smoking amongst the White women in this study (23% were smokers) vs.
the Black (5%) and Hispanic (2%) women. In light of the much higher rate of smoking, it
is surprising that White women did not have higher odds of PIH than their counterparts.
More research is needed here which would include identification of additional factors
influencing these outcomes, as well as including quantification and timing of smoking in
the pregnancy. There are certain to be outcome differences between those that smoke 1-2
cigarettes per day than those that use 1 pack (20 cigarettes) per day, and between those that
quit in early pregnancy vs. those that smoke throughout the pregnancy.

55

Question 3: What is the effect of hypertensive disorders on the odds of having LBW,
controlling for marital status and smoking (Path C)?
This question was posed to assess the odds of having LBW when hypertensive
disorders were present, regardless of age. Results are consistent with healthcare literature.
When CHTN was present, Blacks and Whites were more than 2.5 times more likely to have
LBW than when CHTN was not present in their same racial group; Hispanics were more
than 3 times more likely than non-hypertensive Hispanics. Physiologically, this makes
sense since chronic hypertension has been present prior to the pregnancy, and vascular
damage has been ongoing for an undetermined amount of time prior to the added
physiologic stress of pregnancy. Once vascular damage becomes severe enough to cause
decreased oxygen delivery to tissues, a process known as shunting occurs. The body
recognizes the tissue hypoxia, and diverts a higher percentage of blood to what are
considered “vital” organs- the brain, heart, and kidneys. The uterus then becomes even
more depleted of available oxygen and nutrients to transfer to the fetus, resulting in LBW.
The occurrence of PIH also was significantly associated with higher odds of LBW.
When PIH was present, Black and White women were more than 2.5 times more likely to
have LBW (than Black or White women without PIH), and Hispanic women had the largest
increase at almost 3.5 times more likely (than Hispanic women without PIH). These
findings again do not support the weathering hypothesis, since White women with PIH
were equally apt to have LBW than Black women with PIH. This may be due to the
aforementioned situation of all women in this study already being in poverty, and not
comparing to the general population for each race/ethnicity.

56

And for PIH, the

epidemiologic paradox for Hispanic women was not protective, as they had the highest risk
of LBW in this situation.
For both the CHTN and the PIH arm, smoking was significantly associated with
increased odds of LBW. This is congruent with healthcare literature, where the association
between smoking and having a LBW infant are well-publicized. Further study which
includes quantification and duration of smoking in pregnancy could lend further insight
into critical levels of smoking which trigger the LBW outcome.
Being married was protective against CHTN and PIH for all three racial/ethnic
groups. However, Black women received the largest protection. This is a new finding,
and important enough to warrant further investigation to identify the true benefits these
women are receiving from being married.
Question 4: What is the effect of increasing maternal age on the odds of having
LBW when controlling for marital status, smoking, and hypertensive disorder,
(Path D)?
In the final question, all the variables included in the research questions were
included as controls to assess the effect of increasing maternal age on the odds of LBW.
This question was analyzed once controlling for CHTN along with the other
controls (smoking and marital status), and then re-analyzed substituting CHTN with PIH.
For each racial/ethnic group, there was essentially no difference in odds of LBW with
increasing maternal age whether CHTN or PIH was in the model.
All the variables used as controls were found to have significance in both the CHTN
and the PIH model. The presence of a hypertensive disorder and smoking were both
individually significantly (p=.000) associated with increased odds of LBW. And in both

57

the CHTN and PIH model, marriage was significantly (p=.000) protective against the odds
of LBW. This is an important finding, which has not been previously reported in the
literature.
Comparing the three racial/ethnic groups, marriage made the largest difference for
Black women (OR .781 in the CHTN model, .779 in the PIH model; White OR .909 CHTN
model, .905 PIH model; Hispanic OR .898 CHTN model, .894 PIH model). This is an
important finding, especially in view of the low percentage of Black women in this study
who were married (9.6%) vs the percentage of White (26.0%) and Hispanic (24%) women
who were married. This lower rate of marriage among Black women may be partially
explained by the higher rates of incarceration, joblessness, and mortality among Black men
when compared to the general population (Koball et al., 2010). This creates a higher ratio
of Black women to Black men during reproductive years, limiting the availability of
available partners. Another limitation to be considered is the fact that healthy people are
more desirable as a spouse (Koball et al., 2010; Wood et al., 2007), and therefore those
women who have risk factors for having LBW (such as CHTN) may be less likely to be
married.
Without further research into what benefits it is that marriage offers which conveys
this protection, it can only be postulated that the largest benefit is gained through increased
household income. This would offer potentially more access to food (quantity and quality),
better housing, and neighborhoods with more resources. The primary advice to women
with known LBW pregnancies is to increase their protein intake; however, the largest
source of protein is meat, which is very expensive and quickly decimates a food budget.

58

The benefit of emotional support, however, cannot be dismissed or minimized.
Studies have shown decreased rates of hypertension and cardiovascular disease for those
who are married, citing the benefit of emotional support as well as increased resources.
In this study, there was essentially no difference in odds of LBW with increasing
maternal age as compared to the results from Path A (effect of increasing maternal age on
LBW, controlling for smoking and marital status). The addition of hypertensive disorder
to the controls did not substantially change the effect of increasing age on LBW, for any
of the racial/ethnic groups. This finding suggests that CHTN and PIH have an independent
effect on increasing odds of LBW independent of the mother’s age.
SUMMARY AND IMPLICATIONS
The weathering hypothesis was not universally supported in this study. It was
found to be applicable to increased odds of LBW with increasing maternal age for all three
groups who were living in poverty, but did not show the racial difference between Blacks
and Whites found in previous studies. For the odds of hypertension with increasing age,
the weathering hypothesis again was applicable to all groups but did not show differences
between groups. This may be due to all women in this study living in poverty (using
Medicaid as a proxy for poverty), and not being compared to the general population.
Hypertensive disorders (CHTN, PIH) were shown to be significantly associated
with increased odds of LBW, consistent with healthcare literature. There were racial/ethnic
differences found here, with Hispanics having the highest odds of LBW when either
hypertensive disorder was present (approximately 3.5 times more likely), compared to
Blacks and Whites (each were more than 2.5 times more likely). This is an unexpected
finding, as Hispanics typically have better outcomes (i.e., the epidemiologic paradox).

59

Smoking was found to be associated with hypertensive disorders and LBW for all
groups except in one question. When assessing the effect of increasing maternal age on
PIH, smoking was not found to be a significant variable for the Black women and Hispanic
women; it was a significant variable for White women. This needs further investigation
into quantification and duration of smoking, which was not included in this study.
Being married was found to be significantly associated with lowering the odds of
LBW, for all three racial/ethnic groups but was especially true for Black women. These
are new findings to add to the literature, and need confirmatory testing as well as qualitative
investigation to get to the true benefits that marriage provides for these women. Further
exploration into the racial/ethnic differences for marital health benefits is also needed.
The interaction of families with their environment has been the focus of many
family scientists. As a dynamic, and ever-changing species, amid a continually evolving
environment, such study is challenging and difficult at best. The “environment” (i.e.,
poverty) and “family” (pregnant woman and presence of a spouse) have evolved even since
this study began. With the new healthcare system in America (the Affordable Care Act)
requiring every person to have some type of health insurance, the indicator for poverty
used in this study (Medicaid) would need to be modified to replicate the design. And the
current definition of “married” is changing by the minute, as more and more states amend
laws to recognize same-sex marriages. Human ecology theory has the ability to guide
research based on the interaction of families with their complex environment. This study
used an earlier version of Bronfenbrenner’s model, with limited inclusion of environmental
factors. Expanding that model in future research to include more of the process and context

60

of Bronfenbrenner's later models may yield further insight into the potential importance of
environmental factors which were not included in this study.
Healthcare professionals can utilize these findings not only to direct future research,
but to emphasize the importance of education for low SES pregnant women. While
smoking education is not a reimbursable event, the ethical impetus is present to implement
and follow up on cessation therapies in these women. Similarly, since the presence of a
spouse in this study was protective against LBW in all racial/ethnic groups, the designation
and inclusion of a significant support person could easily be integrated into prenatal care.
Once designated, that primary support person should not only be encouraged to participate
in prenatal visits, but welcomed and included by the healthcare staff. Older guidelines and
policies that only include the father of the baby as a designated visitor or able to participate
in visits are not only outdated, but potentially increasing the odds of a poor outcome (such
as LBW). Integrating the support from the healthcare team and the support from her
significant others will optimize the positive effects of each woman’s care. A perfect
example of such integration would be the development of a smoking cessation program for
pregnant women which includes their significant other. The presence of the significant
other would lend the support, especially if they smoked also, and they could encourage
each other in the cessation therapy.
Allocation of funding for such programs is typically under the direction of agency
officials and extrapolates to legislators who set state and national budgets. With the new
national healthcare insurance program (i.e., the Affordable Care Act) beginning its
implementation, it is an opportunity for legislators to implement evidence-based funding.
The literature continues to show improved outcomes when women receive prenatal care,

61

and racial disparities in LBW continue to exist as shown in this study. Developing
culturally sensitive AND accessible health clinics which deliver prenatal care are necessary
to improve rates of prenatal care, but funding must be allocated for such programs. Simply
requiring Americans to have health insurance, without an acceptable and culturally
sensitive health center to utilize that insurance, is a waste of valuable funding and a lost
opportunity to improve health for many Americans.

Additional funding to provide

smoking cessation programs is another example of where budgets could practice evidencebased funding.
It is important to revise outdated policies for these pregnant women which restrict
benefits and access to the historical role of “husband.” This study, concordant with other
research, has shown the improved health outcomes when a person is “married.” Current
policies should be updated to allow each woman to designate who her significant other is,
regardless of legal marital status and gender. This will need to extend to legislative policies
for insurance and financial benefits, visitation policies, and legal designation as “next of
kin.”
LIMITATIONS
Due to the constrains of a secondary analysis of a pre-existing data set, it is not
possible to get any other indicator of poverty than “Medicaid” as the mother’s source of
payment at the time of delivery. There are multiple limitations with use of this variable as
a poverty indicator. Many states will automatically qualify uninsured women who are
lower and middle class for Medicaid when they become pregnant, to encourage them to
seek prenatal care and hospital/facility- based births to avoid the need to have an unattended
pregnancy and birth. Also previously (prior to implementation of the Affordable Care Act

62

in 2014) included in eligibility for Medicaid were females who were in college but had
“aged-out” of their parents’ insurance, despite their income level. These young women do
not live in poverty; in fact, many of them were higher income families. These are examples
of women who received Medicaid but do not live in poverty, and therefore may skew these
statistics. Using additional research methodologies, such as GIS to tie birth certificate
addresses to neighborhoods can refine the sample to include only those truly living in
impoverished neighborhoods and those living in segregated poverty.
Low birthweight and small for gestational age (SGA) infants can have very
different pathologies. Any infant born prematurely will be LBW (birthweight less than
2500g); this includes those infants who are delivered early purposefully due to a
deteriorating maternal or infant condition, as well as those infants of mothers with
premature labor with no known etiology. The lower birthweight is then not a result of
restricted growth, but simply earlier-than-expected birth. However, SGA applies only to
those infants who are smaller than expected based on what week of gestation they are born.
These SGA infants are sometimes constitutionally small, but very often are smaller than
expected due to either serology insults (e.g., bacterial or viral infections) or decreased
blood flow due to conditions such as hypertension or smoking. Ideally, this study would
have looked at SGA infants, but birth certificates do not list SGA, only birthweight.
Being married has been shown in the literature to be protective against
complications from numerous health conditions, including cardiovascular disease, cancer,
and diabetes, as well as lower rates of LBW. There are obvious limitations to using the
basic “married” classification on the birth certificate. Qualitative research, to determine
the true presence of a supportive relationship, would yield results more indicative of the

63

advantage of an emotionally and financially supportive relationship with a significant
other, regardless of marital status. This also would be inclusive of gay women, who may
or may not be able to be legally married in their specific state as of the date of this study.
While the exact components of standard prenatal care are debatable, as is the ideal
number and timing of visits, the evidence supports the necessity of prenatal care in
improving perinatal outcomes. Identification and reduction and/or elimination of risks,
along with early detection and treatment of abnormalities, lowers the rates of undesirable
events. Research has shown the link between no prenatal care and higher rates of LBW
(Cox, et al., 2011; Krans & Davis, 2012; Partridge, et al., 2012; Taylor, et al., 2005;
Willems Van Dijk, et al., 2010), prematurity (Cox, et al., 2011; Taylor, et al., 2005; & .
Willems Van Dijk, et al., 2010), early and late neonatal death (Partridge, et al., 2012), as
well as stillbirth and infant death (Partridge, et al., 2012). In specific research related to
hypertension, prenatal care was found to be essential in proper identification and treatment
of the disorder (Hadwen, 2011). However, the interplay between not receiving prenatal
care and being diagnosed with either CHTN or PIH is complex, and will have irregularities
in reporting on the birth certificate. This is due to the definition of both CHTN and PIH.
In order to accurately diagnose either condition, it must be determined when the blood
pressure elevated in relationship to the pregnancy.

Chronic hypertension must be

diagnosed prior to the beginning of the pregnancy, and PIH can only be diagnosed after 20
weeks gestation (half-way through the pregnancy). In the presence of not receiving
prenatal care, the only way either of these can accurately be diagnosed is the presence of
other medical records which have recorded the maternal blood pressure prior to or during
the first half of the pregnancy. Therefore, there may be substantially more cases of both

64

CHTN and PIH which have not been officially labeled as such on the birth certificate for
those women who did not receive prenatal care. As with any data set, the data are only as
reliable as the person inputting/classifying the data. In a hectic labor and delivery unit,
taking the extra time to truly reflect on each birth certificate’s questions, and ascertain an
accurate response, is often not seen as a necessity by the busy delivery attendant. Due to
these complicating factors, and the low frequency of not receiving prenatal care (less than
1.8% for all racial/ethnic groups), it was not included in this study.
Quantification and duration of smoking during pregnancy were not included in this
study.

This is an important limitation of the results.

As previously discussed,

differentiating the number of cigarettes smoked daily, along with segregating those who
quit in early pregnancy would no doubt provide more detailed information on the true
effects on LBW.
This study was the first to include Hispanics when utilizing the weathering concept,
but all “Hispanic” cultures were grouped together as a homogenous variable. As there are
differences in hypertension within different Hispanic sub-cultures, these groups need to be
included as distinct groups in future research. Separation of these sub-cultures will produce
more accurate, and perhaps different, results.
While this study benefitted from the large number of cases available in using
national level data, it is acknowledged that there are very different geographical
environments within the United States, and those areas can also contain very different
community cultures which can affect rates of smoking and marriage. Using state or even
neighborhood/community-level data may provide more accurate results which can be
utilized to build studies with more homogenous data to better infer meaning of results.

65

This novice researcher chose to use an earlier model of Bronfenbrenner’s work, and
limit the systems included to those in the microsystem. Utilization of his later models,
along with a more inclusive use of his process-person-context-time components may yield
different and more comprehensive results.
FUTURE RESEARCH
In addition to the limitations aforementioned, and their implications for future study
design, there are additional areas for future research. Emphasis should be on qualitative
components to further explore the effect of a supportive relationship related to birth
outcomes as a high priority, especially in light of the quickly growing married lesbian
population here in America. Differences between lesbian and heterosexual couples could
be investigated using a modified replication of this study.
For further investigation into application of the weathering concept, more
ethnicities and racial groups need to be included. There are many areas within America of
concentrated ethnic/racial communities, where GIS software and birth certificate linkage
could provide valuable insight into differences in birth outcomes for these populations.
Cultural-specific research could lead to insights as to how to best provide educational,
financial, and healthcare resources which are sensitive to and appropriate for a range of
cultures and ethnicities. Studies which compare and contrast these same issues for women
who do and do not live in poverty will give further insight into poverty-related conditions
versus genetic or cultural issues in regards to health.

66

REFERENCES

67

REFERENCES

American College of Obstetrics and Gynecology. (2009). ACOG Practice Bulletin
Number 107: Induction of Labor. Obstetrics & Gynecology, 114(2), 386-397.
Ayoola, A.B., Netleman, M.D., Stommel, M., & Canady, R.B. (2010). Time of
pregnancy recognition and prenatal care use: a population-based study in the
United States. Birth, 37(1), 37-43.
Baine, M., Sahak, F., Lin, C., Chakraborty, S., & Lyden, E. (2011). Marital Status and
Survival in Pancreatic Cancer Patients. Retrieved on April 15, 2014, from
www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0021052#pon
e-0021052-g004
Baker, B., Helmers, K., O’Kelly, B., Sakinofsky, I., Abelsohn, A., & Tobe, S. (1999).
Marital Cohesion and ambulatory blood pressure in early hypertension. American
Journal of Hypertension, 12(2), 227-230.
Barker, D.J. (2006). Adult consequences of Fetal Growth Restriction. Clinical Obstetrics
& Gynecology, 49(2), 270-283.
Barrington, S. (2010). The Increasing Protection of Marriage on Infant Low Birth
Weight Across Two Generations of African American Women. Journal of
Family Issues, 31(8), 1041-1064.
Baschat, A.A., Galan, H.L., Ross, M.G., & Gabbe, S.G. (2007). Intrauterine Growth
Restriction. In S.G. Gabbe, J.R. Niebyl, & J.L. Simpson (Eds.), Obstetrics:
Normal and Problem Pregnancies (pp. 771-814). Elsevier: Philadelphia.
Bird, S.T., Chandra, A., Bennett, T., & Harvey, S.M. (2000). Beyond marital status:
Relationship type and duration and the risk of low birth weight. Family Planning
Perspectives, 32, 281-287.
Borders, A.E.B., Grobman, W.A., Amsden, L.B., & Holl, J.L. (2007). Chronic Stress and
Low Birth Weight Neonates in a Low-Income Population of Women. Obstetrics
& Gynecology, 109(2), 331-338.
Braveman, P., Marchi, K., Egerter, S., Kim, S., Metzler, M., Stancil, T., & Libet, M.
(2010). Poverty, Near-Poverty, and Hardship Around the Time of Pregnancy.
Maternal Child Health Journal, 14, 20-35.

68

Briley, C., Flanagan, N.L., & Lewis, N.M. (2002). In-home prenatal nutrition
intervention increased dietary iron intakes and reduced low birthweight in lowincome African-American women. Journal of the American Dietetic Association,
102(7), 984-987.
Bronfenbrenner, U. (1979). The Ecology of Human Development. Cambridge, MA:
Harvard University Press.
Bronfenbrenner, U. (1986). Ecology of the family as a context for human development:
Research perspectives. Developmental Psychology, 22(6), 723-742.
Bronfenbrenner, U. (1995). Developmental ecology through space and time: A future
perspective. In P. Moen, G.H. Elder, Jr., & K. Luscher (Eds.), Examining lives in
context: Perspectives on the ecology of human development (pp.617-647).
Washington, DC: American Psychological Association.
Bubolz, M.M., & Sontag, M.S. (1993). Human Ecology Theory. In P.G. Boss, W.J.
Doherty, R. LaRossa, W.R. Schumm, & S.K. Steinmetz (Eds.), Sourcebook of
Family Theories and Methods: A Contextual Approach (pp. 419-435). New York:
Plenum Press.
Byrd, D.R., Katcher, M.L., Peppard, P., Durkin, M., & Remington, P.L. (2007). Infant
Mortality: Explaining Black/White Disparities in Wisconsin. Maternal Child
Health Journal, 11, 319-326.
Cai, J., Hoff, G.L., Dew, P.C., Guillory, V.J., & Manning, J. (2005). Perinatal Periods of
Risk: Analysis of Fetal-Infant Mortality Rates in Kansas City, Missouri. Maternal
and Child Health Journal, 9(2), 199-205.
Carlson, M., McLanahan, S., & England, P. (2004). Union formation in fragile families.
Demography, 41, 237-261.
Centers for Disease Control and Prevention. (2010). How tobacco smoke causes disease:
a report of the Surgeon General. US Department of Health and Human Services,
Atlanta, GA.
Colen, C.G., Geronimus, A.T., Bound, J., & James, S.A. (2006). Maternal Upward
Socioeconomic Mobility and Black-White Disparities in Infant Birthweight.
American Journal of Public Health, 96(11), 2032-2039.
Collins, J.W., Wamback, J., David, R.J., & Rankin, K.M. (2009). Women’s Lifelong
Exposure to Neighborhood Poverty and Low Birth Weight: A Population-Based
Study. Maternal and Child Health Journal, 13, 326-333.

69

Cox, R.G., Zhang, L., Zotti, M.E., & Graham, J. (2011). Prenatal Care Utilization In
Mississippi: Racial Disparities and Implications for Unfavorable Birth Outcomes.
Maternal and Child Health Journal, 15, 931-942.
Curry, M.A. (1998). The interrelationships between abuse, substance use, and
psychosocial stress during pregnancy. Journal of Obstetric, Gynecologic, and
Neonatal Nursing, 27, 692-699.
David, R.J., & Collins, J.W. (1997). Differing Birth Weight Among Infants of U.S.-Born
Blacks, African-Born Blacks, and U.S.-Born Whites. The New England Journal
of Medicine, 337(17), 1209-1214.
DeFranco, E.A., Lian, M., Muglia, L.J., & Schootman, M. (2008). Area-level poverty
and preterm birth risk: A population-based multilevel analysis. BMC Public
Health, 8, 316-325.
Dunn, L.L., & Oths, K.S. (2004). Prenatal predictors of intimate partner abuse. Journal
of Obstetric, Gynecologic, and Neonatal Nursing, 33, 54-63.
El-Sayed, A.M., & Galea, S. (2012). Temporal Changes in Socioeconomic Influences on
Health: Maternal Education and Preterm Birth. American Journal of Public
Health, 102(9), 1715-1721.
Finer, L.B., & Kost, K. (2011). Unintended Pregnancy Rates at the State Level.
Perspectives on Sexual and Reproductive Health, 43(2), 78-87.
Floud, S., Balkwill, A., Canoy, D., Wright F.L., Reeves, G.K., Green, J., …& Cairns, B.J.
(2014). Marital Status and Ischemic Heart Disease incidence and Mortality in
Women: A Large Prospective Study. BMC Medicine, 12(1), 42-45.
Geronimus, A.T. (1996). Black/White differences in the relationship of maternal age to
birthweight: a population-based test of the weathering hypothesis. Social Science
& Medicine, 42(4), 589-97.
Geronimus, A.T., Bound, J., & Waidmann, T.A. (1999). Health inequality and
population variation in fertility-timing. Social Science & Medicine, 49, 16231636.
Geronimus, A.T., Hicken, M., Keene, D., & Bound, J. (2006). “Weathering” and Age
Patterns of Allostatic Load Scores Among Blacks and Whites in the United
States. American Journal of Public Health, 96(5), 826-833.
Geronimus, A.T., Hicken, M.T., Pearson, J.A., Seashols, S.J., Brown, K.L., & Cruz, T.D.
(2010). Do US Black Women Experience Stress-Related Accelerated Biological
Aging? Human Nature: An Interdisciplinary Biosocial Perspective, 21(1), 19-38.

70

Goldenberg, R.L., Cliver, S.P., Mulvihill, F.X., Hickey, C.A., Hoffman, H.J., Klerman,
L.V., & Johnson, M.J. (1996). Medical, psychosocial, and behavioral risk factors
do not explain the increased risk for low birthweight among black women.
American Journal of Obstetrics & Gynecology, 175, 1317-1324.
Goldenberg, R.L., Culhane, J.F., Iams, J.D., & Romero, R. (2008). Epidemiology and
causes of preterm birth. The Lancet, 371, 75-84.
Goza, F.W., Stockwell, E.G., & Balistreri, K.S. (2007). Racial Differences in the
Relationship Between Infant Mortality and Socioeconomic Status. Journal of
Biosocial Science, 39, 517-529.
Grady, S. (2006). Racial disparities in low birthweight and the contribution of residential
segregation: a multilevel analysis. Social Science & Medicine, 63(12), 3013-3029.
Grady, S., & Ramirez, I. (2008). Mediating medical risk factors in the residential
segregation and low birthweight relationship by race in New York City. Health &
Place, 14(4), 661-677.
Hadwin, G. (2011). Antenatal: Blood Pressure: Evidence Summaries. Joanna Briggs
Institute, 2011-04-13.
Hamilton, B.E., Martin, J.A., & Ventura, S.J. (2011). Births: Preliminary Data for 2010.
National Vital Statistics Report, 58(24). Hyattsville, MD: National Center for
Health Statistics.
Heaman, M.I. (2005). Relationships between physical abuse during pregnancy and risk
factors for preterm birth among women in Manitoba. Journal of Obstetric,
Gynecologic, and Neonatal Nursing, 34, 21-731.
Hillemeier, M.M., Lynch, D., Harper, S., Raghunathan, T., & Kaplan, G.A. (2003).
Relative or Absolute Standards for Child Poverty: A State-Level Analysis of
Infant and Child Mortality. American Journal of Public Health, 93(4), 652-657.
Hohmann-Marriott, B. (2009). The Couple Context of Pregnancy and its Effects on
Prenatal Care and Birth Outcomes. Maternal and Child Health
Journal, 13(6), 745-754.
Holzman, C., Eyster, J., Kleyn, M., Messer, L.C., Kaufman, J.S., Laraia, B.A., O’Campo,
P., … & Elo, I.T. (2009). Maternal Weathering and Risk of Preterm Delivery.
American Journal of Public Health, 99(10), 1864-1871.
Hoyert, D.L., & Xu, J.Q. (2012). Deaths: Preliminary data for 2011. National Vital
Statistics Report, 61(6). Hyattsville, MD: National Center for Health Statistics.

71

Idler, E.L., Boulifard, D.A., & Contrada, R.J. (2012). Mending Broken Hearts: Marriage
and Survival Following Cardiac Surgery. Journal of Health and Social Behavior,
53(1), 33-49.
Janevic, T., Stein, C.R., Savitz, D.A., Jaufman, J.S., Mason, S.M., & Herring, A.H.
(2010). Neighborhood Deprivation and Adverse Birth Outcomes among Diverse
Ethnic Groups. Annals of Epidemiology, 20(6), 445-451.
Jesse, D.E., & Reed, P.G. (2004). Effects of spirituality and psychosocial well-being on
health risk behaviors in Appalachian pregnant women. Journal of Obstetric,
Gynecologic, and Neonatal Nursing, 33, 739-747.
Jesse, D.E., & Swanson, M.S. (2007). Risks and resources associated with antepartum
risk for depression among rural southern women. Nursing Research, 56, 378-386.
Jesse, D.E., Swanson, M.S., Newton, E.R., & Morrow, E.R. (2009). Racial Disparities in
Biopsychosocial Factors and Spontaneous Preterm Birth Among Rural LowIncome Women. Journal of Midwifery & Women’s Health, 54, 35-42.
Jesse, D.E., Walcott-McQuigg, J., Mariella, A., & Swanson, M.S. (2005). Risks and
protective factors associated with symptoms of depression in low-income African
American and Caucasian women during pregnancy. Journal of Midwifery &
Women’s Health, 50, 405-410.
Kiecolt-Glaser, J.K. & Newton, T. (2001). Marriage and health: His and Hers.
Psychological Bulletin, 127(4), 472-503.
King, K.B. & Reis, H.T. (2012). Marriage and Long-Term Survival after Coronary
Artery Bypass Grafting. Health Psychology, 31(1), 55-62.
Koball, H.L., Moiduddin, E., Henderson, J., Goesling, B., & Besculides, M. (2010).
What Do We Know About the Link Between Marriage and Health? Journal of
Family Issues, 31(8), 1019-1040.
Kramer, M.R., & Hogue, C.R. (2008). Place Matters: Variation in the Black/White Very
Preterm Birth Rate Across U.S. Metropolitan Areas, 2002-2004. Public Health
Reports, 123(5), 576-585.
Krans, E.E., & Davis, M.M. (2012). Preventing Low Birthweight: 25 years, prenatal risk,
and the failure to reinvent prenatal care. American Journal of Obstetrics &
Gynecology, 206(5), 398-403.
Laughon, S.K., Reddy, U.M., Sun, L., & Zhang, J. (2010). Precursors for Late Preterm
birth in Singleton Gestations. Obstetrics & Gynecology, 116(5), 1047-1055.

72

LeBlanc, C.P., & Rioux, F.M. (2007). Iron Deficiency Anemia following Prenatal
Nutrition Interventions. Canadian Journal of Dietetic Practice and Research,
68(4), 222-225.
Leslie, J., Galvin, S., Diehl, S., Bennett, T., & Buescher, P. (2003). Infant mortality, low
birth weight, and prematurity among Hispanic, white, and African American
women in North Carolina. American Journal of Obstetrics & Gynecology,
188(5), 1238-1240.
Lister, Z., Fox, C., & Wilson, C.M. (2013). Couples and Diabetes: A 30-Year Narrative
Review of Dyadic Relational Research. Contemporary Family Therapy, 35(4),
613-638.
London, M.L., Ladewig, P.A.W., Ball, J.W., Bindler, R.C.M., & Cowen, K.J. (2011).
Maternal & Child Nursing Care (3rd Ed.). Pearson: Upper Saddle River, NJ.
Love, C., David, R.J., Rankin, K.M., & Collins, J.W. (2010). Exploring Weathering:
Effects of Lifelong Economic Environment and Maternal Age on Low Birth
Weight, Small for Gestational Age, and Preterm Birth in African-American and
White Women. American Journal of Epidemiology, 172(2), 127-134.
Lu, M.C. & Chen, B. (2004). Racial and ethnic disparities in preterm birth: the role of
stressful life events. American Journal of Obstetrics & Gynecology, 191(3), 691699.
Lu, M.C, & Halfon, N. (2003). Racial and ethnic disparities in birth outcomes: a lifecourse perspective. Maternal & Child Health Journal, 7(1), 13-30.
MacDorman, M.F., Hoyert, D.L., & Mathews, T.J. (2013). Recent declines in infant
mortality in the United States, 2005-2011. NCHS Data Brief, 120. Hyattsville,
MD: National Center for Health Statistics.
MacDorman, M.F., & Mathews, T.J. (2009). Behind international rankings of infant
mortality: How the United States compares with Europe. NCHS Data Brief, 23.
Hyattsville, MD: National Center for Health Statistics.
Mahdi, H., Kumar, S., Munkarah, A.R., Adalamir, M., Doherty, M. & Swensen, R.
(2013). Prognostic impact of marital status on survival of women with epithelial
ovarian cancer. PsychoOncology, 22(1), 83-88.
Martin, J.A., Hamilton, B.E., Sutton, P.D., Ventura, M.A., Mathews, T.J., Kirmeyer, S.,
& Osterman, M.J.K. (2010). Births: Final Data for 2007. National Vital Statistics
Report, 58(24). Hyattsville, MD: National Center for Health Statistics.

73

Martin, J.A., Hamilton, B.E., Ventura, M.A., Osterman, M.J.K., Wilson, E.C., &
Mathews, T.J. (2010). Births: Final Data for 2010. National Vital Statistics Report,
61(1). Hyattsville, MD: National Center for Health Statistics.
Martinez, M., & Ariosto, D. (2011, March 24). Hispanic population exceeds 50 million,
firmly nation’s No. 2 group. CNN. Retrieved from http://articles.cnn.com/201103-24/us/census.hispanics_1_hispanic-population-illegal-immigration-foreignborn?_s=PM:US
Masho, S.W., Chapman, D., & Ashby, M. (2010). The Impact of Paternity and Marital
Status on Low Birth Weight and Preterm Births. Marriage & Family Review,
46(4), 243-256.
Mathews, T.J., & MacDorman, M.F. (2010). Infant Mortality Statistics From the 2006
Period Linked Birth/Infant Death Data Set. National Vital Statistics Report,
58(17). Hyattsville, MD: National Center for Health Statistics.
McArthur, A. (2010). Antenatal: Routine Care: Evidence Summaries. Joanna Briggs
Institute, 2010-09-14.
Michigan Department of Community Health. (2009). Michigan Resident Birth File [Data
set]. Retrieved from http://www.michigan.gov/mdch/0,1607,7-1322944_4669_4681---,00.html
Moster, D., Lie, R.T., Markestad, T. (2008). Long-term medical and social consequences
of preterm birth. New England Journal of Medicine, 359(3), 262-273.
Nkansah-Amankra, S. (2010). Neighborhood Contextual Factors, Maternal Smoking, and
Birth Outcomes: Multilevel Analysis of the South Carolina PRAMS Survey,
2000-2003. Journal of Women’s Health, 19(8), 1543-1552.
Nkansah-Amankra, S., Dhawain, A., Hussey, J.R., & Luchok, K.J. (2010). Maternal
Social Support and Neighborhood Income Inequality as Predictors of Low Birth
Weight and Preterm Birth Outcome Disparities: Analysis of South Carolina
Pregnancy Risk Assessment and Monitoring System Survey, 2000-2003.
Maternal Child Health Journal, 14, 774-785.
Nkansah-Amankra, S., Luchok, K.J., Hussey, J.R., Watkins, K., & Liu, X. (2010).
Effects of Maternal Stress on Low Birth Weight and Preterm Birth Outcomes
Across Neighborhoods of South Carolina, 2000-2003. Maternal Child Health
Journal, 14, 215-226.
Olson, M.E., Diekema, D., Elliott, B.A., & Renier, C.M. (2010). Impact of Income and
Income Inequality on Infant Health Outcomes in the United States. Pediatrics,
126(6), 1165-1173.

74

Orr, S.T., James, S.A., & Miller, C.A. (1996). Psychosocial stressors and low birthweight
in an urban population. American Journal of Preventative Medicine, 12, 459-466.
Osborne, C., Berger, L.M., & Magnuson, K. (2012). Family Structure Transitions and
Changes in Maternal Resources and Well-being. Demography, 49, 23-47.
Papacek, E.M., Collins, J.W., Schulte, N.F., Goergen, C., & Drolet, A. (2002). Differing
Postneonatal Mortality Rates of African-American and White Infants in Chicago:
An Ecologic Study. Maternal and Child Health Journal, 6(2), 99-105.
Partridge, S., Balayla, J., Holcroft, C.A., & Abenhaim, H.A. (2012). Inadequate Prenatal
Care Utilization and Risks of Infant Mortality and Poor Birth Outcome: A
Retrospective Analysis of 28,729,765 U.S. Deliveries over 8 Years. American
Journal of Perinatology, 29(10), 787-794.
Perry, S.E., Hockenberry, M.J., Lowdermilk, D.L., & Wilson, D. (2010). Maternal Child
Nursing Care (4th. Ed). Mosby Elsevier: Maryland Heights, MO.
Rauh, V.A., Andrews, H.F., & Garfinkel, R.S. (2001). The Contribution of Maternal Age
to Racial Disparities in Birthweight: A Multilevel Perspective. American Journal
of Public Health, (91)11, 1815-1824.
Reagan, P.B., Salsberry, P.J., & Olsen, R.J. (2007). Does the measure of economic
disadvantage matter? Exploring the effect of individual and relative deprivation
on intrauterine growth restriction. Social Science and Medicine, 64, 2016-2029.
Reddy, U.M., Chia-Wen, K., Tonse, N.K.R., & Willinger, M. (2009). Delivery
Indications at Late-Preterm Gestations and Infant Mortality Rates in the United
States. Pediatrics, 124(1), 234-240.
Reichman, N.E., Hamilton, E.R., Hummer, R.A., & Padilla, Y.C. (2008). Racial and
Ethnic Disparities in Low Birthweight Among Urban Unmarried Mothers.
Maternal Child Health Journal, 12, 204-215.
Robles, T.F. & Kiecolt-Glaser, J.K. (2003). The physiology of marriage: pathways to
health. Physiology & Behavior, 79(3), 409-416.
Rosenthal, M.B., Li, Z., Robertson, A.D., & Milstein, A. (2009). Impact of Financial
Incentives for Prenatal Care on Birth Outcomes and Spending. Health Research
and Educational Trust, 44(5), 1465-1479.
Schoenborn, C.A. (2004). Marital status and health: United States, 199-2002. Advance
Data, No. 351.

75

Shapiro-Mendoza, C.K., Tomashek, K.M., Kotelchuk, M., Barfield, W., Nannini, A.,
Weiss, J., & Declercq, E. (2008). Effect of late-preterm birth and maternal
medical conditions on newborn morbidity risk. Pediatrics, 121(2), e223-232.
Seidel, A.J., Franks, M.M., Stephens, M.A., & Rook, K.S. (2012). Spouse Control and
Type 2 Diabetes Management: Moderating Effects of Dyadic Expectations for
Spouse Involvement. Family Relations, 61(4), 698-709.
Sims, M., Sims, T.L., & Bruce, M.A. (2007). Urban Poverty and Infant Mortality Rate
Disparities. Journal of the National Medical Association, 99(4), 349-357.
Smock, P.J. (2000). Cohabitation in the United States: An appraisal of research themes,
findings, and implications. Annual Review of Sociology, 26, 1-20.
Sparks, P.J. (2009). One Size Does Not Fit All: An Examination of Low Birthweight
Disparities Among A Diverse Set of Racial/Ethnic Groups. Maternal Child
Health Journal, 13, 769-779.
Sullivan, K., Raley, R.K., Hummer, R.A., & Schiefelbein, E. (2012). The Potential
Contribution of Marital-Cohabitation Status to Racial, Ethnic, and Nativity
Differentials in Birth Outcomes in Texas. Maternal and Child Health Journal,
16, 775-784.
Taylor, C.R., Alexander, G.R., & Hepworth, J.T. (2005). Clustering of U.S. Women
Receiving No Prenatal Care: Differences in Pregnancy Outcomes and
Implications for Targeting Interventions. Maternal and Child Health Journal,
9(2), 125-133.
Tomashek, K.M., Shapiro-Mendoza, C.K., Davidoff, M.J., & Petrini, J.R. (2007).
Differences in mortality between late-preterm and term singleton infants in the
United States, 1995-2002. Journal of Pediatrics, 151(5), 450-456.
Tong, V.T., Dietz, P.M., Morrow, B., D’Angelo, D.V., Farr, S.L., Rockhill, K.M., &
England, L.J. (2013). Trends in Smoking Before, During, and After PregnancyPregnancy Risk Assessment Monitoring System, United States, 40 Sites, 20002010. MMWR, 62, 1-19.
Torche, F. (2011). The Effect of Maternal Stress on Birth Outcomes: Exploiting a
Natural Experiment. Demography, 48, 1473-1491.
Tucker, J.M., Goldenberg, R.L., Davis, R.O., Copper, R.L., Winkler, C.L., & Hauth, J.C.
(1991). Etiologies of Preterm Birth in an Indigent Population: Is Prevention a
Logical Expectation? Obstetrics & Gynecology, 77(3), 343-347.

76

Tudge, J.R.H., Mokrova, I., Hatfield, B.E., & Karnik R.B. (2009). Uses and Misuses of
Bronfenbrenner’s Bioecological Theory of Human Development. Journal of
Family Theory & Review, 1(4), 198-210.
United Nations, Department of Economic and Social Affairs, Population
Division (2013). World Population Prospects: The 2012 Revision.
United Nations, Department of Economic and Social Affairs, Population
Division (2007). World Population Prospects: The 2006 Revision, Highlights,
Working Paper No. ESA/P/WP.202.
Vespa, J., & Painter II, M.A. (2011). Cohabitation History, Marriage, and Wealth
Accumulation. Demography, 48, 983-1004.
Victoria, C.G., Wagstaff, A., Schellenberg, J.A., & Gwatkin, D. (2003). Applying an
equity lens to child health and mortality: More of the same is not enough. The
Lancet, 362(9379), 233-245.
Wang, L., Wilson, S.E., Stewart, D.B., & Hollenbeak, C.S. (2011). Marital status and
colon cancer outcomes in US Surveillance, Epidemiology and End Results
Registries: Does marriage affect cancer survival by gender and stage? Cancer
Epidemiology, 35(5), 417-422.
Willems Van Dijk, J.A., Anderko, L., & Stetzer, F. (2011). The Impact of Prenatal Care
Coordination on Birth Outcomes. Journal of Obstetric, Gynecologic, and
Neonatal Nursing, 40(1), 98-108.
Williams, L.B. (1994). Determinants of couple agreement in US fertility decisions.
Family Planning Perspectives, 26, 169-173.
Wood, R.G., Goesling, B., & Avellar, S. (2007). The Effects of Marriage on Health: A
Synthesis of Recent Research Evidence. Mathematica Policy Research Inc.,
Washington, D.C.
Woods, S.M., Melville, J.L., Guo, Y., Fan, M.-Y., & Gavin, A. (2009). Psychosocial
Stress during Pregnancy. American Journal of Obstetrics & Gynecology, 202(1),
61.e1-61.e7.
Zachariah, R. (2004). Attachment, Social Support, Life Stress, and Psychological WellBeing in Pregnant Low-Income Women: A Pilot Study. Clinical Excellence for
Nurse Practitioners, 8(2), 60-67.
Zhang, Z. & Hayward, M.D. (2006). Gender, the Marital Course, and Cardiovascular
Disease in Late Midlife. Journal of Marriage and Family, 68(3), 639-657.

77

Zuckerman, B., Amaro, H., Bauchner, H., & Cabral, H. Depressive symptoms during
pregnancy: relationship to poor health behaviors. American Journal of Obstetrics
& Gynecology, 160, 1107-1111.

78