4'1. .z

:1}... . I.
x. .-

fume,

.1363.

13:4,...
. ‘ .

H
awn ,

“fizmmmmwwa
f3

9.

3

Hi. >-
. i
’iflW-Ia

3 :
... an r
refit“. a
31.... if

‘5‘...““3’

”Nan

P o
9 1 1:311
. » 3...... :6:

 

 

2993 LIBRARY
Michigan State

I "JI pa:
Ul IIVCI alty

 

 

 

This is to certify that the
thesis entitled

MATERNAL AND INFANT HEALTH OF IMMIGRANTS IN
THE CAPITAL TRl-COUNTY AREA IN MICHIGAN

presented by

Yu-Ying Chu

has been accepted towards fulfillment
of the requirements for the

MS. degree in Geography

 

 

{7 . 4 j?
_ Vii/LU? (I 1533a (fig; , PA 0 I {I 91“;

 

Major Professor‘s Signature

771041,}, Q0, 510/0
U

 

Date

MSU is an Affirmative Action/Equal Opportunity Employer

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K2lProlecc8PrelelRCIDateDue.Indd

 

MATERNAL AND INFANT HEALTH OF IMMIGRANTS IN THE CAPITAL TRI-
COUNTY AREA IN MICHIGAN

By

Yu-Ying Chu

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Geography

2010

ABSTRACT

MATERNAL AND INFANT HEALTH OF IMMIGRANT IN THE CAPITAL
TRI-COUNTY AREA IN MICHIGAN

By
Yu-Ying Chu

Previous studies have shown that many groups of immigrant mothers have
improved birth outcomes compared to mothers born in the United States, which is
referred to as the epidemiological paradox. This thesis research was designed to study
how individual- and neighborhood—level risk factors affect the adverse birth outcomes
of U.S.-bom and foreign-bom mothers in the capital tri-county area in Michigan from
the year 1995 to 2007. There were 73,682 women in total, including 67,515 U.S.-bom
and 5,628 foreign-bom mothers. The results indicated that foreign-bom women were
less likely to contribute to both low birth weight and preterm birth than U.S.-bom
women, and the birth outcomes varied considerably within different foreign-bom
groups of mothers. I also found that neighborhood-level risk factors affect U.S.-bom
women more while individual-level risk factors were more important for foreign-bom

WOan.

Keywords: immigrants, epidemiological paradox, low birth weight, preterm birth,
Michigan

This thesis is dedicated to my family

ACKNOWLEDGEMENTS

I wish to express my greatest thanks to Dr. Sue Grady, who has mentored,
advised, assisted, and encouraged me a lot during the past two years at graduate
school and led me to explore and understand more about medical geography and
public health.

I also wish to acknowledge Dr. Guo Chen and Dr. Leo Zulu, who provided
invaluable help and advice and always encouraged me warmly.

I wish to thank to Kristie Socia and Jordan Howell for their encouragement and

assistance.

TABLE OF CONTENTS

LIST OF TABLES .............................................................................. vii
LIST OF FIGURES .............................................................................. ix
1.0 Introduction ................................................................................... l
2.0 Literature Review ............................................................................ 6
2.1 Individual Level Risk Factors for Adverse Birth Outcomes ...................... 6
2.1.1 Age ........................................................................... 6

2.1.2 Education ................................................................... 7

2.1.3 Marital Status ............................................................... 8

2.1.4 Socioeconomic Status ..................................................... 9

2.1.5 Personal Behaviors ....................................................... 10

2.1.6 Medical Risks ............................................................. 11

2.1.7 Parity ....................................................................... 12

2.1.8 Prenatal Care .............................................................. 13

2.1.9 Mother’s Birthplace ...................................................... 15

2.1.10 Duration of Residence .................................................. 16

2.1.11 Acculturation ............................................................ 17

2.2 Neighborhood Level Risk Factors for Adverse Birth Outcomes ............... 19
2.2.1 Income Inequality ........................................................ 20

2.2.2 Immigrant Density ....................................................... 21

2.2.3 Housing .................................................................... 22

2.3 Societal Level Risk Factor for Adverse Birth Outcomes ........................ 23

2.4 Thesis Significance ................................................................... 24
2.4.1 Goal ........................................................................ 25

2.4.2 Objectives .................................................................. 25

2.4.3 Hypothesis ................................................................. 26

3.0 Data and Methods ........................................................................ 27
3.1 Data ..................................................................................... 27
3.1.1 Birth Data .................................................................. 27

3.1.2 Neighborhood-level Data ................................................ 29

3.2 Methods ................................................................................. 29
3.2.1 Join Tables ................................................................. 29

3.2.2 Descriptive Statistics ..................................................... 30

3.2.3 Spatial Statistics Analyses ............................................... 31

4.0 Results ....................................................................................... 34
4.1 Individual and Neighborhood Level Risk Factors ................................ 34
4.1.1 Frequencies ............................................................... 34

4.1.2 Spatial Distribution ...................................................... 35

V

4.1.3 Spatial Statistics — Hot Spot Analyses ................................. 36

4.1.4 Spatial Statistics — Geographically Weighted Regressions ......... 37

4.2 Adverse Birth Outcomes ............................................................. 38
4.2.1 Frequencies ............................................................... 38

4.2.2 Spatial Distribution ...................................................... 39

4.3 Epidemiological Paradox ............................................................ 41

5.0 Discussion ................................................................................... 49
6.0 Conclusion ................................................................................... 53
Appendix A ...................................................................................... 55
Appendix B ...................................................................................... 86
Reference ....................................................................................... 121

vi

LIST OF TABLES

Table 1: Characteristics of U.S.-bom and foreign-bom mothers and infants in the
capital tri-county area in Michigan, 1995-2007 ............................................. 56

Table 2: Birth outcomes of U.S.-bom and foreign-born mothers and infants in the
capital tri-county area in Michigan, 1995-2007 ............................................. 58

Table 3: Birth outcomes within different foreign-bom groups by country/region of
origin in the capital tri-county area in Michigan, 1995-2007 ............................. 59

Table 4: Binary logistic regression analyses of low birth weight in the capital
tri-county area in Michigan, 1995-2007 ...................................................... 60

Table 5: Binary logistic regression analyses of low birth weight of U.S.-bom mother
group in the capital tri-county area in Michigan, 1995-2007 .............................. 62

Table 6: Binary logistic regression analyses of low birth weight of foreign-born
mother group in the tri-county area in Michigan, 1995-2007 ............................. 64

Table 7: Binary logistic regression analyses of preterm birth in the capital tri—county
area in Michigan, 1995-2007 .................................................................. 66

Table 8: Binary logistic regression analyses of preterm birth of U.S.-bom mother
group in the capital tri-county area in Michigan, 1995-2007 .............................. 68

Table 9: Binary logistic regression analyses of preterm birth of foreign—bom mother
group in the capital tri-county area in Michigan, 1995-2007 .............................. 70

Table 10: Binary logistic regression analyses of low birth weight of both
individual-level and neighborhood-level characteristics in the capital tri-county area
in Michigan, 1995-2007 ........................................................................ 72

Table 11: Binary logistic regression analyses of low birth weight of both
individual-level and neighborhood-level characteristics of U.S.-bom mother group in
the capital tri-county area in Michigan, 1995-2007 ....................................... 74

vii

Table 12: Binary logistic regression analyses of low birth weight of both
individual-level and neighborhood-level characteristics of foreign-bom mother group
in the capital tri-county area in Michigan, 1995-2007 ..................................... 76

Table 13: Binary logistic regression analyses of preterm birth of both individual-level
and neighborhood-level characteristics in the capital tri-county area in Michigan,
1995-2007 ....................................................................................... 78

Table 14: Binary logistic regression analyses of preterm birth of both individual-level
and neighborhood-level characteristics of U.S.-born mother group in the capital
tri-county area in Michigan, 1995-2007 ...................................................... 80

Table 15: Binary logistic regression analyses of preterm birth of both individual-level
and neighborhood-level characteristics of foreign-bom mother group in the capital
tri-county area in Michigan, 1995-2007 ...................................................... 82

Table 16: The correlation between low birth weight and neighborhood-level
characteristics among different foreign-bom groups of mothers and the U.S.-bom
mothers in the capital tri-county area in Michigan, 1995-2007 ........................... 84

Table 17: The correlation between preterm birth and neighborhood-level
characteristics among different foreign-bom groups of mothers and the U.S.-bom
mothers in the capital tri-county area in Michigan, 1995-2007 ........................... 85

viii

LIST OF FIGURES

Figure 1: Reference map of the capital tri-county area in Michigan .................... 87

Figure 2: The hot/cold spots of the percentage of poor in the capital tri-county area in
Michigan, 1995-2007 ........................................................................... 88

Figure 3: The hot/cold spots of the percentage of extremely rich in the capital
tri-county area in Michigan, 1995-2007 ..................................................... 89

Figure 4: The spatial distribution of median household income in the capital
tri-county area in Michigan, 1995-2007 ..................................................... 90

Figure 5: The spatial distribution of the immigrant density in the capital tri-county
area in Michigan of 1999 ...................................................................... 91

Figure 6: The hot/cold spots of U.S.-bom Caucasians in the capital tri-county area in
Michigan, 1995-2007 ........................................................................... 92

Figure 7: The hot/cold spots of foreign.-bom Caucasians in the capital tri-county area
in Michigan, 1995-2007 ........................................................................ 93

Figure 8: The hot/cold spots of U.S.-bom African Americans in the capital tri-county
area in Michigan, 1995-2007 .................................................................. 94

Figure 9: The hot/cold spots of foreign.-bom African Americans in the capital
tri-county area in Michigan, 1995-2007 ...................................................... 95

Figure 10: The hot/cold spots of U.S.-bom Asians in the capital tri-county area in
Michigan, 1995-2007 ........................................................................... 96

Figure 11: The hot/cold spots of foreign-bom Asians in the capital tri-county area in
Michigan, 1995-2007 ........................................................................... 97

Figure 12: The hot/cold spots of U.S.-bom Hispanic in the capital tri-county area in
Michigan, I995-2007 ........................................................................... 98

Figure 13: The hot/cold spots of foreign-bom Hispanic in the capital tri-county area
in Michigan, 1995-2007 ........................................................................ 99

Figure 14: The spatial distribution of predicted low birth weight and its residual of
U.S.-bom mothers in the capital tri-county area in Michigan, 1995-2007 ............ 100

Figure 15: The spatial distribution of predicted preterm birth and its residual of
U.S.-bom mothers in the capital tri-county area in Michigan, 1995-2007 ............ 101

Figure 16: The spatial distribution of predicted low birth weight and its residual of
foreign.-bom mothers in the capital tri-county area in Michigan, 1995-2007. ........102

Figure 17: The spatial distribution of predicted preterm birth and its residual of
foreign.-bom mothers in the capital tri-county area in Michigan, 1995—2007. 103

Figure 18: The hot/cold spots of low birth weight of U.S.-born and foreign-bom
mothers in the capital tri-county area in Michigan, 1995-2007 ......................... 104

Figure 19: The hot/cold spots of preterm birth of U.S.-bom and foreign-bom mothers
in the capital tri-county area in Michigan, 1995-2007 .................................... 105

Figure 20: The spatial distribution of the percentage of low birth weight in U.S.-bom
women in the capital tri-county area in Michigan, 1995-2007 .......................... 106

Figure 21: The spatial distribution of the percentage of preterm birth in U.S.-bom
women in the capital tri-county area in Michigan, 1995-2007 .......................... 107

Figure 22: The spatial distribution of the standardized percentage of low birth weight
in U.S.-bom women in the capital tri-county area in Michigan, 1995-2007 .......... 108

Figure 23: The spatial distribution of the standardized percentage of preterm birth in
U.S.-bom women in the capital tri-county area in Michigan, 1995-2007 ............. 109

Figure 24: Numbers of low birth weight and preterm birth of the foreign-bom women
from Eastern Europe in the capital tri-county area in Michigan, 1995-2007 .......... 110

Figure 25: Numbers of low birth weight and preterm birth of the foreign-bom women
from Western Europe in the capital tri-county area in Michigan, 1995-2007 ......... lll

Figure 26: Numbers of low birth weight and preterm birth of the foreign-bom women
from North Africa in the capital tri-county area in Michigan, 1995-2007 ............. 112

Figure 27: Numbers of low birth weight and preterm birth of the foreign-bom women
from Sub Sahara in the capital tri-county area in Michigan, 1995-2007 ............... 113

Figure 28: Numbers of low birth weight and preterm birth of the foreign-bom women
from Canada in the capital tri-county area in Michigan, 1995-2007 ................... 114

Figure 29: Numbers of low birth weight and preterm birth of the foreign-bom women
from Mexico in the capital tri-county area in Michigan, 1995-2007 ................... 115

Figure 30: Numbers of low birth weight and preterm birth of the foreign-born women
from Central/ South America in the capital tri-county area in Michigan, 1995-2007.
................................................................................................... 116

Figure 31: Numbers of low birth weight and preterm birth of the foreign-bom women
from South Asia in the capital tri-county area in Michigan, 1995-2007 ............... 117

Figure 32: Numbers of low birth weight and preterm birth of the foreign-bom women
from Southeast Asia! Oceania in the capital tri-county area in Michigan, 1995-2007.
................................................................................................... 118

Figure 33: Numbers of low birth weight and preterm birth of the foreign-bom women
from Eastern Asia in the capital tri-county area in Michigan, 1995-2007 ............. 119

Figure 34: Numbers of low birth weight and preterm birth of the foreign-bom women
from Central Asia/ Middle East in the capital tri-county area in Michigan, 1995-2007.
................................................................................................... 120

Images in this thesis are presented in colors.

xi

1.0 Introduction

The World Health Organization (WHO) adopted in 1950 the standard of less than

2,500 grams (or 5 pounds, 8 ounces) as a universal definition of low birth weight and

this threshold has been used in studies of the subject in the decades since (WHO,

1950). According to Paneth (1995) both low birth weight and its major antecedent,

preterm birth (referring to the delivery of infants prior to 37 completed weeks of

gestation), are more common in the United States than in most Westem European

nations. Low birth weight is also the most important predictor of neonatal mortality

(deaths that occur in the first week of life) in the United States.

According to the US. Census Bureau, the term foreign born refers to anyone who

is not a US. citizen at birth. This includes naturalized US. citizens, lawful permanent

residents (immigrants), temporary migrants (such as foreign students), humanitarian

migrants (such as refugees), and people illegally present in the United States.

Over the last three decades in the United States the neonatal mortality rate has

declined primarily due to the increase in available neonatal care facilities, yet the

incidence of low birth weight has continued to rise. While there are many known

maternal and environmental risk factors for low birth weight, these still do not

completely explain why the overall incidence is increasing. This increase in incidence

however, is not seen across all populations. Previous research has shown that the

incidence of low birth weight is increasing among U.S.-bom mothers, while the rate

for immigrant mothers is still relatively low (Gould et al., 2003; Madan et al., 2006;

El Reda et al., 2007). This phenomenon of immigrant mothers having improved birth

outcomes compared to U.S.-bom mothers is referred to in the literature as the

“epidemiological paradox” (Gould et al., 2003).

The epidemiological paradox was first observed in the US. Latino population in

the 19705 (Markidas and Coreil, 1986) and was characterized by favorable birth

outcomes among Latino mothers compared to other mothers of similar low

socioeconomic status (SES) (Acevedo-Garcia et al., 2003). A large number of studies

have subsequently evaluated the association between immigrant status, SES, health

behaviors, birth outcomes and other types of morbidity and mortality (Palloni and

Morenoff, 2001; Peak and Weeks, 2002; Gould et al., 2003). For instance, unlike

other socioeconomically disadvantaged minority groups, Mexican Americans were

found to have very low rates of psychiatric service utilization compared to other

ethnic groups (Markides and Coreil, 1986). Furthermore, the concept of

epidemiological paradox applies to health behaviors such as smoking as

Acevedo-Garcia et al (2004) reported that rates of tobacco use were lower among

certain foreign-bom groups than among their U.S.-bom ethnic counterparts,

controlling for socioeconomic position. Also, Singh and Siahpush (2002) used data

from the National Longitudinal Mortality Study (1979-1989) and found that compared

with U.S.-bom Caucasians of equivalent socioeconomic and demographic

background, foreign-born African Americans had a 48% lower risk of mortality.

Similar findings were also observed for foreign-born Hispanics (-45%), foreign-bom

Asians/Pacific Islanders (APIs) (-43%), U.S.-bom Hispanics (-26%), U.S.-bom APls

(-32%) and foreign-bom Caucasians (-16%). While American Indians did not differ

significantly from U.S.-bom Caucasians, U.S.-bom African Americans had an 8%

higher mortality risk. African American and Hispanic immigrants experienced,

respectively, 52% and 26% lower mortality risks than their U.S.-bom counterparts.

The purpose of this thesis research is to improve our understanding of the

geography of the epidemiological paradox by studying the incidence of low birth

weight and preterm birth across and within different groups of foreign-bom mothers

living in the capital tri-county area in Michigan (Ingham, Eaton, and Clinton). I

studied the likelihood that foreign-bom mothers would have fewer low birth weight

and preterm birth babies compared to U.S.-bom mothers who are also living in the

capital tri-county area. I tried to explain the population-geographic disparities in low

birth weight and preterm birth rates by studying the individual characteristics of

mothers, including age, education level, marital status, medical risk factors and

origin of birth as well as the neighborhood environment in which mothers live and

infants are born. This thesis would draw on theoretical and methodological

approaches from the fields of population and medical geography to answer these

important and timely questions.

This study took place in the capital tri-county area in Michigan where there are a

large number of foreign-bom mothers of reproductive age. From 2002 to 2008

immigrants have comprised approximately 7.8% (range, 7.2% to 9.7%) of the

population in Ingham county, 3.3% (range, 2.8% to 4.4%) of the population in Eaton

county, and 1.6% (range, 1.3% to 1.8%) of the population in Clinton county (US.

Bureau of the Census). Therefore, the capital tri-county area is an ideal place to study

maternal differences in low birth weight by origin of birth. Over the last three decades

the United States has also received a substantial number of immigrants and there have

been few studies to-date that have attempted to understand the role of origin of birth

in the rising incidence of low birth weight in the capital tri-county area. The findings

from this study will also help to inform similar trends in the United States.

The objectives in this thesis research are:

(1) To assess whether foreign-bom groups of mothers living in the capital

tri-county area have lower rates of low birth weight and preterm birth compared to

U.S.-bom mothers also living in this area (i.e., to see if an epidemiological paradox

exists in the capital tri-county area);

(2) To determine the variation in incidence of low birth weight and preterm birth
among different foreign-bom groups of mothers to see if the epidemiological paradox
is stronger among some groups compared to others;

(3) To describe the individual and neighborhood level characteristics of
foreign-born groups and U.S.-bom mothers in the capital tri-county area
neighborhoods; and

(4) To identify differences in individual and neighborhood level risk factors for
low birth weight and preterm birth incidence in different foreign-bom groups and
U.S.-bom mothers to better understand underlying factors that may contribute to the

epidemiological paradox.

2.0 Literature Review

The following literature review examines known individual and neighborhood

level risk factors for low birth weight.

2.1 Individual Level Risk Factors for Adverse Birth Outcomes

Several determinants influence birth outcomes. The following are the main

individual characteristics of mothers that may contribute to low birth weight and

preterm birth as indicated in previous research.

2.1.1 Age

Valero de Bembé et al (2004) indicated the incidence of low birth weight

increases in the young and old extremes of women’s reproductive life; that is, between

15 and 19 years and between 35 and 40 years of age. Reichman and Pagnini (1997)

showed that both African American and white mothers in their 305 were significantly

more likely to deliver a low birth weight baby than women aged 25 to 29 years of age

of the same race. However, other studies have found that even though older maternal

age is associated with increased risk of low birth weight among singleton births, this

effect was significant for African American women only (Collins and David, 1990;

Starfield et al., 1991). Rauh et al (2001) further found that the extreme age-related

effects observed for African American women in relation to low birth weight were

largely concentrated among poor women.

2.1.2 Education

Previous studies have found a powerful connection between health and education.
Education and knowledge of appropriate health behaviors are important determinants
of health and the education of a child’s mother is an important predictor of the health
of a child (Skilnik, 2008). One study conducted in the Philippines illustrated how
higher educated mothers were able to keep their children healthy, even in locations
without a safe water supply (Glewwe, 1997).

Auger et al (2008) indicated that among Canadian-bom mothers, all levels of
education less than university were associated with a greater likelihood of all three
adverse birth outcomes. The strongest associations were seen for mothers having “no
high school diploma” relative to “university” education, for small for gestational age
births (infants born below the 10% percentile of a standard population) odds ratio (OR)
= 2.03 (95% CI 1.84 to 2.22), low birth weight OR = 2.03 (95% CI 2.61 to 3.91), and
preterm birth OR = 1.67 (95% CI 1.49 to 1.87).

However, these results were reversed for foreign-bom mothers. FOr these mothers
all levels of education lower than university education were less strongly associated
with small for gestational age birth and preterm birth and not significantly associated
with low birth weight. For example, mothers having no high school diploma relative

to university-educated mothers had a higher likelihood of small for gestational age

OR = 1.26 (95% Cl 1.07 to 1.49) , an OR substantially smaller than the equivalent OR

for Canadian-bom mothers OR = 2.03 (95% CI 1.84 to 2.22). The two educational

levels “no high school diploma” versus university-education OR = 1.36 (95% Cl 1.11

to 1.66) and “high school diploma” versus university-education OR = 1.37 (95% CI

1.11 to 1.70) were associated with preterm birth among foreign-bom mothers but

these associations were not substantially different from those of Canadian-bom

mothers. Auger et al (2008) therefore concluded that the “healthy migrant” effect may

be present in mothers with lower education but not in other educational categories.

This finding was disputed however, by other researchers who studied only higher

educated women. The mechanisms by which higher educated foreign-bom mothers

are more likely to experience adverse birth outcomes than Canadian-born highly

educated women are stress and psychosocial factors (Dejin-Karlsson and Ostergren,

2004; Frank, 2005). After entry into Canada, immigrant women of higher education

could conceivably experience greater stress adapting to a new living environment.

For example, the challenge of finding employment comparable to what they may have

had in their own countries was considered very stressful.

2.1.3 Marital Status

Holt et al (1997) found that women who were married during their first

pregnancy had a lower incidence of low birth weight than single mothers but if they

were separated during the second pregnancy, the relative risk (RR) of low birth weight
increased RR = 1.4 in comparison to those who remained married. Conversely, among
women whose marital status changed fi'om single to married between pregnancies, the
risk of low birth weight decreased RR = 0.8. Also, Madan et al (2006) found that a
household with less familial and social support may contribute to poorer perinatal
outcomes. Further, Nothnagle et a1 (2000) pointed out that more women in the late
care group (women who received care only in the third trimester) reported being
unmarried, or having no supportive person available during pregnancy than did
women with earlier care.
2. I. 4 Socioeconomic Status

Socioeconomic status level is one of the factors most closely related with the
health status of populations, and it is shown that unfavorable socioecondmic
conditions increases the incidence of low birth weight (Valero de Bembé et al., 2004).
Other studies have consistently shown that racial or socioeconomic differences in
morbidity and mortality are most pronounced in young and middle-aged adults
(House et al., 1990; E10 and Preston, 1996).

Nevertheless, Madan et al (2006) illustrated that foreign-bom Asian-Indian
women have a low-risk sociodemographic profile but a paradoxically higher

incidence of prematurity, low birth weight, and small for gestational age birth infants.

The odds of low birth weight compared with white women were significantly higher

in both foreign-bom, OR = 2.37 (95% CI 2.3 to 2.4) and U.S.-bom, OR = 2.18 (95%

CI 1.95 to 2.18) Indian women. Markides and Coreil (1986) also concluded that the

health status of Hispanics in the Southwest is much more similar to the health status

of other Caucasians than that of African Americans although socioeconomically, the

status of Hispanics is closer to that of African Americans. These authors suggest that

the extended family support that Hispanics receive may protect them from

stress-related morbidity.

Moreover, Uretsky and Mathiesen (2007) had a different perspective toward

socioeconomic status among foreign-bom populations. These authors showed that

advances in educational attainment, economic status, and English proficiency were all

significantly related to improved health, but this effect was muted among the

foreign-bom as the number of years living in the United States increased. This result

suggests that along with improvement in key socioeconomic factors there is a

deterioration of some unmeasured indicators that appear to have an overwhelming and

negative influence on immigrant health.

2.1.5 Personal Behaviors

Maternal smoking, alcohol, caffeine, and drug consumption are the main

behavioral risk factors that may contribute to having low birth weight babies.

10

(McFarlane et al., 1996; Smeriglio and Wilcox, 1999). Even though the relationship

between caffeine consumption and low birth weight remains a subject of some debate,

Wilborg et a1 (1996) observed the risk of preterm birth in women who consumed large

doses of caffeine (> 400 mg per day) and also smoked was three times higher than

that of women who did not consume caffeine. Furthermore, the consumption of illicit

drugs had been associated with a lower birth weight, and it is estimated that up to 25

to 30% of women who consume cocaine during pregnancy will give birth to a small

for gestational age birth infant (Valero de Bembé et al., 2004). Kliegman et al (1994)

also indentified cocaine as the drug with the strongest association with preterm birth

and low birth weight.

2.1.6 Medical Risks

A variety of diseases may relate to adverse birth outcomes. Madan et al (2006)

pointed out that diabetes, depending on the type and severity, may increase the risk of

adverse birth outcomes including, macrosomatia (exceptionally large baby at birth),

low birth weight, prematurity, congenital anomalies and fetal death. In general, more

severe stages of diabetes are associated with vascular compromise and smaller than

expected birth weights. However, treatment of a diabetic pregnant woman with insulin

and diet decreases prenatal mortality and the incidence of macrosomy in the infant,

but may also increases the frequency of grth retardation due to iatrogenic

11

hyperinsulinism (an above normal level of insulin in the blood of a person or animal)
and excessive caloric reduction (Valero de Bembé et al., 2004).

Valero de Bembé et a1 (2004) also described that chronic hypertension may
provoke alterations in fetal grth as a result of reduced uteroplacetal fluid. Moreover,
hypertension, induced by pregnancy is defined as the development of blood pressure
values higher than 140/90 mm/Hg alter the 20th week of pregnancy, leads to an
increased risk of preterm birth and of low birth weight (Leung et al., 1998; Zeitlin et
al., 2001). Plus, lupus is the most frequent autoimmune disease in the pregnant
woman. Lupus increases the frequency of low birth weight and preterm birth 30 to
50%, especially when the disease involves the kidneys and hypertension (Valero de
Bembé et al., 2004).

2.1. 7 Parity

Parity refers to the number of times a woman has giving birth. Short intervals
between births constitute one of the main risk factors for prematurity and low birth
weight, although researchers are still debating about this (Valero de Bembé et al.,
2004). Ferraz et al (1988) showed that short birth intervals, varying fi'om 3 to 6
months in developing countries and from 1 to 2 years in developed countries may lead
to an increased tendency toward low birth weight and prematurity in subsequent

pregnancies.

12

Also, Roth et a1 (1998) indicated that second and third children weigh more than
the first because of improved intrauterine conditions, such as uterine structures and
vascular structures, which permit greater placental development, and consequently,
improved fetal nutrition (Valero de Bembe’ et al., 2004). However, the risk of low
birth weight will increase again with the fourth and subsequent child (Silva et al.,
1998). Further, a history of low birth weight in previous pregnancies is also an
important predictor of risk in the current pregnancy (Bratton et al., 1996).

2.1.8 Prenatal Care

Prenatal care has long been endorsed as a mean to identify mothers at risk of
delivering a preterm or growth-retarded infant while also providing them with an
array of available medical, nutritional, and educational interventions to reduce the
risks associated with low birth weight and other adverse pregnancy conditions and
outcomes (Alexander and Korenbrot, 1995). Hence, adequacy of prenatal care use
could be an indicator of a myriad of health-enhancing maternal attitudes and
behaviors as well as a measure of the prenatal care received.

The most targets for prenatal interventions to prevent low birth weight, according
to Kramer (1990), are “(1) smoking (aimed at reduction or cessation); (2) nutrition
(aimed at increasing pre-pregnancy weight and/ or ensuring adequate weight gain

during pregnancy); and (3) medical care (aimed at reducing overall morbidity).

13

Nevertheless, the determinants of prenatal care use are varied and range from obvious

financial, geographic, and support barriers to more subtle cultural and attitudinal

characteristics.” Those who received the least prenatal care cited finances as the most

important reason for not having prenatal care earlier in the pregnancy or more often

during the pregnancy (Alexander and Korenbrot, 1995).

Nothnagle et al (2000) had the same observation in California. These authors

found that a higher percentage of women in the late and no care groups had income

under the poverty line compared with women in the earlier care group. In addition,

although the majority of women in each group (earlier care, late care, and no care)

had Medi-Cal as their primary insurance during pregnancy, over two-fifths (41.5%) of

women who received no prenatal care were uninsured throughout pregnancy,

compared with only about 1% of women in the earlier care and late care groups;

women in the late care and no care groups appeared less likely to have private

insurance than women in the earlier care group.

Moreover, Gavin et a1 (2004) indicated the racial and ethnic disparities relate to

the use of a range of prenatal care among Medicaid-covered women as well.

Compared with white non-Hispanic women, minority women were less likely to

receive health services that the woman initiates, discretionary services, and services

potentially requiring specialized follow-up care, whereas they were more likely to

14

receive screening tests for diseases related to high-risk behaviors. The authors also
proposed that these results may be explained by the markedly different composition of
the Hispanic and Asian/Pacific Islander populations with respect to country of origin
and hence cultural beliefs and practices.
2.1.9 Mother 3' Birthplace

Mother’s birthplace has also been highlighted as an important predictor of birth
outcomes among immigrant subgroups in either Europe (Vahratian et al., 2004) or the
United States (El Reda et al., 2007). Urquia et a1 (2009) pointed out that the risk of
low birth weight varied considerably according to the region of origin of the
immigrant mother; the country of origin appears to be a much more important factor
in low birth weight among children of recent immigrants than the neighborhood in
which they currently live. For example, infants of North African immigrants,
compared to infants of Belgian women, were less likely to be born preterm, despite
their lower socioeconomic status. Also, foreign-born Hispanic women, despite a
high-risk demographic and socioeconomic profile, experienced birth outcomes
superior to those of their U.S.-bom counterparts (Crump et al., 1999). Another
observation was among foreign-bom Asian and Asian-bom Indian women. Despite
the fact that they have a better socioeconomic status profile, foreign-bom Asian and

Asian-bom Indian women experienced a higher incidence of low birth weight and

15

preterm birth than their U.S.-bom counterparts (Acevedo-Garcia et al., 2003; Gould et
al., 2003; Tore et al., 2006 ).

One similar trend has also been observed in Michigan, which is hometo about
490,000 persons of Arab ancestry, one of the largest populations of Arab immigrants
outside of the Middle East (Arab American Institute Foundation, 2003). El Reda et a1
(2007) reported that even though foreign-bom Arab women in Michigan have a
higher-risk maternal demographic profile and being at a considerable socioeconomic
disadvantage (having less education, being more likely to report Medicaid as the
expected payer source, and more likely to receive no prenatal care) than that of their
U.S.-bom white counterparts, their prevalence of preterm birth is significantly lower,
which is consistent with the epidemiologic paradox reported among foreign-born
Hispanic women.

2.1.10 Duration of Residence

Previous studies indicate that the favorable birth and health outcomes of
foreign-bom migrant women might be explained by the “health migrant effect” and
by the relatively healthy life styles that they maintained from the country of origin.
However, a change to an unhealthier life style could contribute to the decreased health
outcomes of native-bom migrants and migrants with longer residence duration (Tore

et al., 2006).

16

Crump et al (1999) shared the same observation. Their study of Mexican
Americans found that longer residence in the United States resulted in increased risk
of preterm birth among foreign-bom Americans, indicating that acculturation plays a
major role in reversing the effects of traditionally protective social and cultural
factors.

Keams (I993) illustrated that what occurs in a place (in terms of the relations
between people and elements of their environment) has profound importance to health.
Previous research also suggested that birth outcomes may either improve or
deteriorate with length of residence among first-generation immigrants, depending on
the migrant group or the receiving environment or a combination of both (Urquia et
aL,2009)

Some studies have shown that immigrants generally arrive in the US. healthier
than the general population, but as time passes their health status converges towards
the levels found in the US. (Singh and Miller, 2001; Singh and Siahpush, 2002).
Importantly, Uretsky and Mathiesen (2007) showed that foreign-bom populations
with improved health seem to decrease uniformly with years living in the US. and
after about 10 years immigrant health becomes roughly equal to the level found
among the U.S.-bom population.

2.1.11 Acculturation

17

Culture assimilation, or acculturation, is usually the first and the easiest in the
series of stages of assimilation by which immigrants become theoretically integrated
into US. society (Gordon, 1964). Acculturation is also a multidimensional
phenomenon by which language components, dietary intake and smoking are
important indicators of birth outcomes (Tore et al., 2006).

As a matter of fact, active smoking is highly associated with birth outcomes;
maternal smoking during pregnancy increases the relative risk of low birth weight
considerably. Dejmek et a1 (2002) found that mothers who smoked moderately had a
higher risk of low birth weight OR = 2.81 (95% CI 2.21 to 3.71) than mothers who
did not smoke and mothers who smoked heavily had a significantly higher risk of
having low birth weight babies compared to mothers who did not smoke OR = 4.95
(95% Cl 4.95 to 8.06). Also, the association between birth weight and maternal
smoking was weaker when they used data about smoking during early pregnancy,
stronger with data characterizing smoking habits in the first trimester, and even
stronger if based in smoking in the second and third trimester. 1f the mothers continue
to smoke even during the second trimester, the adjusted weight reduction for infants
was -152 grams (95% Cl -117 grams to -185 grams) in moderate and -259 grams
(95% C1 -175 grams to -342 grams) in heavy active smoking mothers.

Research in The Netherlands showed that higher infant mortality of Turkish

18

 

migrants who are more integrated into Dutch society (i.e. Dutch-bom and Turkish

migrants with younger age at immigration) might be due to adoption of unhealthy

western life styles. This suggestion is supported by a Dutch report showing a rising

trend of tobacco use especially among younger Turkish women. Meanwhile, the

opposite trend was observed among Surinamese mothers, in which infant mortality

risk decreased with younger age at immigration. This result implies that increased

acculturation and social integration could result in improving health outcomes of their

children as well (Tore et al., 2006).

El Reda et al (2007) also mentioned that behaviors of Arab women are being

altered by residing in the United States, as evidenced by the higher rates of selected

characteristics among U.S.-bom Arabs than among their foreign-bom counterparts.

Specifically, more U.S.-bom Arab mothers than foreign-bom Arab mothers report

tobacco use during pregnancy and list only one named parent on their infants’ birth

certificate. It is very likely that this higher tobacco use among U.S.-bom Arabs is due

to acculturation because previous studies have documented that smoking rates among

Arab women are significantly lower than those of non-Arabs in Michigan (Akbar,

1994).

2.2 Neighborhood Level Risk Factors for Adverse Birth Outcomes

Women’s health is influenced not only by behavior and culture, but also by the

19

social, economic, and political contexts in which women live: “people’s health both

shapes and is shaped by the places in which their lives unfold” (McLafferty and

Tempalski, 1995; Dyck and Keams, I995). The neighborhood or community context,

according to McLafferty and Tempalski (1995), encompasses the “local social

networks of neighbors and friends, geographical access to jobs and services, housing,

and environmental quality.” For example, because women often use prenatal care

service in their neighborhoods, the locations of services and transportation can be

important determinants of birth outcome (Hoagberg et al., 1990).

The following are some neighborhood characteristics which may increase the risk

of adverse birth outcomes for mothers.

2.2.1 Income Inequality

O’regan and Wiseman (1990) suggested that low-income neighborhoods

consistently have high rates of infant mortality and low birth weight, with rates often

several times higher than those in affluent neighborhoods. Also, poverty and

unemployment emerge as important predictors of infant health. Huynh et a1 (2005)

reported an adverse influence of income inequality on preterm birth in a study of US.

counties, and found that the influence of income inequality depends on race.

Moreover, another study on cumulative exposure to income inequality reported an

association with preterm birth for Hispanic but not African American or Caucasian

20

ethnicity (Reagan and Salsberry, 2005).

Auger et a1 (2009) tried to examine the association between birth outcomes and

area income and income inequality across social makers in Québec, Canada. These

authors found that both preterm birth and small gestational for age birth were

positively related to area poverty and inversely related to income inequality. However,

high area poverty was associated with preterm birth among Canadian-bom, adjusted

OR = 1.07 (95% CI 1.00 to 1.14), but not foreign-bom mothers, adjusted OR = 0.95

(95% Cl 0.83 to 1.09). There was a strong association between high area poverty and

small gestational age birth among Canadian-bom OR = 1.13 (95% CI 1.06 to 1.20),

but the association was not significant among foreign-bom mothers, adjusted OR =

1.00 (95% CI 0.88 to 1.13). These authors concluded that income inequality might be

a pathway through which “area” exerts its effects on birth outcomes. Also, income

inequality could be a proxy for other unrelated neighborhood factors favorably

associated with birth outcomes.

2. 2.2 Immigrant Density

Besides examining the relationship between birth outcomes and income

inequality, Auger et a1 (2009) also had the observation that low immigrant density was

associated with preterm birth in the fully adjusted model for Canadian-bom mothers

OR = 1.14 (95% Cl 1.07 to 1.21), and the association between preterm birth and

21

foreign-bom mothers was OR = 0.79 (95% CI 0.63 to 1.00). That is, high immigrant
density was protective against preterm birth for Canadian-bom mothers, but
unfavorably associated with preterm birth for foreign-bom mothers. These authors
also suggested that high immigrant density may be associated with conditions that
reduce stress in native-bom mothers, but increase stress in foreign-bom mothers. Such
conditions might arise if, for instance, employment opportunities were greater for
native- than for foreign-bom individuals due to prejudice or network integration.
Several other studies confirm this point of view by addressing the result that the
psychosocial stress associated with balancing home and work heightens the risk of ill
health and low birth weight for some women (Pritchard and Teo Mpfan, 1993; Elliott,
1995).
2.2.3 Housing

Besides using different individual-level risk factors as indicators of low birth
weight, Shiono et a1 (1997) introduced the concept of level of living as well. It
includes housing density (had two or more people per room during pregnancy), stable
housing (lived three or more years in current residence), moved (moved two or more
times in the past year), and housing problems (had two or more major housing
problems in need of repair during pregnancy). After controlling for level of poverty

and the other known correlates of birth, they found out that living in public housing

22

was associated with an 83-gram decrease in birth weight. Nevertheless, having a
stable residence was associated with a 76-gram increase in birth weight. Importantly,
these authors concluded that living in public housing had an independent negative
relationship with birth weight while having a stable residence was positively related to
birth weight.

2.3 Societal Level Risk Factors for Adverse Birth Outcomes

Societal (the broadest scale) processes “influence the allocation and quality of
goods and services, the distribution of wealth, and legal and institutional constrains”
(McLafferty and Tempalski, 1995). Side] (1992) mentioned that the social, economic,
and political context in which people live strongly influences the health of both
individuals and of populations. The research done by McLafferty and Tempalski
(1995) showed that in New York City “changing political, social, and economic
relations have profoundly affected the urban landscape, with corresponding impacts
on women’s reproductive health.”

Furthermore, Dyck (1990) illustrated that although women play an active role in
shaping communities, some elements are beyond individual control. Those processes
that operate beyond the community scale have distinct impacts on individuals and
places. For example, Fisher et a1 (1995) showed a deterioration of birth outcomes

during periods of recession and high unemployment.

23

These studies highlight the different risk factors for low birth weight and preterm

birth. Among all the risk factors listed above and the birth data that I am available for,

1 will select mother’s birthplace, maternal race, maternal age, maternal education,

marital status, personal behavior (tobacco use during pregnancy), parity, numbers of

prenatal care visits, and payment of insurance as individual-level risk factors for low

birth weight and preterm birth. As for neighborhood-level risk factors, in addition to

using immigrant density in this research, I will study local racial residential

segregation in order to understand the clustering of racial groups. Also, 1 will use the

two economic measures of income inequality and area-level poverty to understand the

socioeconomic status of the neighborhoods in which mothers live and infants are born.

Societal- level risk factor will not be analyzed in this research on neighborhood

impacts on birth outcomes.

2.4 Thesis Significance

With the higher percentage of immigrants in the United States year by year,

more studies have been focusing on immigrant mothers and their birth outcomes.

However, previous studies were mainly conducted in areas with higher immigrant

density, such as California and Toronto, Canada. As a matter of fact, the capital

tri-county area in Michigan receives a substantial number of immigrants each year

from many different countries of the world, yet to-date there have been no studies on

24

the maternal and infant health of immigrants mothers and children in this area.
Therefore, there is a need for more studies that focus on immigrant health in general
and maternal and infant health of immigrants in particular to better understand if an
epidemiological paradox exists in Michigan and how it is different from the
epidemiological paradox in other regions of the world.
2.4.] Goal
The goals of my thesis are to compare the incidence of low birth weight and
preterm birth in foreign-bom and U.S.-bom mothers to determine if an
epidemiological paradox exists in the capital tri-county area in Michigan and to
explore both individual- and neighborhood- level risk factors that may contribute to
these differences.
2. 4.2 Objectives
The objectives of this thesis are (l) to assess whether immigrant groups of
mothers living in the capital tri-county area have lower rates of low birth weight and
preterm birth compared to U.S.-bom mothers also living in this area (i.e., to see if an
epidemiological paradox exists in the capital tri-county area); (2) to determine the
variation in incidence of low birth weight and preterm birth among different
immigrant groups of mothers to see if the epidemiological paradox is stronger among

some groups compared to others; (3) to describe the individual and neighborhood

25

level characteristics of immigrant groups and U.S.-bom mothers in the capital

tri-county area neighborhoods; and (4) to identify differences in individual and

neighborhood level risk factors for low birth weight and preterm birth incidence in

different immigrant groups and U.S.-bom mothers to better understand underlying

factors that may contribute to the epidemiological paradox.

2. 4. 3 Hypothesis

Following the objectives, 1 hypothesize that:

a. Foreign-bom mothers will have a lower incidence of low birth weight births

and preterm birth than U.S.-bom mothers;

b. The incidence of low birth weight and preterm birth will differ from different

groups of foreign-bom mothers;

c. Individual-level risk factors for low birth weight and preterm birth will be

more important for foreign-bom mothers compared to U.S.-bom mothers;

d. Neighborhood-level risk factors have stronger affect on U.S.-bom mothers

than foreign-bom mothers.

26

3.0. Data and Methods

The following section describes the data and methods that I used in my thesis

research to address the goal and objectives and to test the hypotheses.

3.1 Data

3.1.1 Birth Data

The data of all live singleton births in the capital tri-county area were obtained

from the Vital Statistics Office at the Michigan Department of Community Health for

the years 1995-2007. Vital statistics data include information taken directly off the

birth certificate that is routinely collected on all live births in Michigan. All U.S.-bom

and foreign-bom women who gave birth in the capital tri-county area during this

period were included in the study population. There were 73,682 women in total,

including 67,515 U.S.-bom, 5,628 foreign-bom, and 529 missing data.

All of the matemal-level variables used in this research were from the vital

statistics birth data set. There were 9 independent variables in total. These variables

were selected because they have been previously shown to be related to adverse birth

outcomes as mentioned at the end of my Literature Review Section. These variables

and the form in which they were analyzed included origin of birth (ORIGIN),

U.S.-bom = 0 and foreign-bom = 1; racial/ ethnic group (RACE), African American =

l and others = 0, Asian = l and others = 0, American Indian = 1 and others = 0,

27

Hispanic = 1 and others = 0, and Hawaiian/ Pacific Islanders = 1 and others = 0;
maternal age (MAGE) in years, less than 20 = l and others = 0, and greater than 34 =
l and others = 0; educational level (EDU), no high school diploma = 1 and others = O,
and some college or more = l and others = 0, representing less than 12 and 16+ years
of education; marital status (MARITAL), one parent (single) = 1 and others = O, and
acknowledgment of paternity = l and others = 0; insurance converge (INSURE),
Medicaid = 1 and others = O, and self pay and other = 1 and others = 0; parity
(PARITY), O = 1 and > 1 = 0; trimester of prenatal care (CARE), none = 1 and others
= 0, second = 1 and others = 0, and third = 1 and others = 0; and smoking during
pregnancy (SMOKE), yes = 1 and no = 0.

Overall statistical models included U.S.-bom and foreign-bom and more specific
models included U.S.-bom and foreign-bom by origin of birth. Foreign-bom mothers
and infants were grouped by origin of birth after exploring the birth data and
determining an adequate and similar N for each group. Also, I aimed to group those
foreign-bom populations by geographical proximity and cultural similarity. These
groups represented Eastern Europe (N = 263), Western Europe (N = 551), North
Africa (N = 114), Sub-Saharan Africa (N = 407), Canada (N = 229), Mexico (N =
528), Central/ South America (N = 569), Eastern Asia (N = 1029), South Asia (N =

461), Southeast Asia/ Oceania (N = 904), and Central Asia/ Middle East (N = 522).

28

Infant-level risk factors included birth weight (birth weight < 2,500 grams = 1
and birth weight >= 2,500 grams = 0; birth weight continuous, grams), preterm birth
(gestation < 37 weeks = 1 and gestation >=37 weeks = 0; gestation, continuous,
weeks).

3. 1.2 Neighborhood-level Data

All the neighborhood-level variables used in this research were obtained from the
US. Bureau of the Census, SF] and SF3 data files at the census tract level and
included median household income (INCOME), used as an indicator of neighborhood
poverty and economic deprivation; ratio of income to poverty threshold by household
type (POVERTY), used to describe the poverty level of households of immigrants;
immigrant density by census tract (DENSITY), used to understand the density of
immigrant populations in the neighborhoods; and Anselin’s local Moran’s 1
(LMiZscore) was used to measure the clustering of racial groups as an indicator of
local racial residential segregation (Anselin, 1995).

3.2 Methods
3.2.] Join Tables

In order to visualize the spatial patterns of both individual- and

neighborhood-level risk factors and analyze the effect of these risk factors on adverse

birth outcomes for U.S.-bom and foreign-bom mothers and infants these data were

29

joined to county boundary files for the capital tri-county area using the common
identifier “GEO_ID” in GIS analysis. There were 117 census tracks in the capital
tri-county area.

3. 2. 2 Descriptive Statistics

To assess the individual- and neighborhood-level risk factors and adverse birth
outcomes for U.S.-bom and foreign-bom mothers and infants, these data were first
explored for missing values and general descriptive statistics such as frequencies and
summary statistics were estimated. The rate of low birth weight and preterm birth
were also calculated to compare the birth outcomes of U.S.-bom and foreign—bom
women. These rates represented the number of low birth weight or preterm birth per
1,000 live births.

Moreover, in order to compare the adverse birth outcomes of U.S.-bom mothers
in the capital tri-county area and the state of Michigan, I calculated the standardized
rates. The standardized rates were obtained by multiplying the rate of low birth weight
(8.4%) and preterm birth (12.5%) of the state of Michigan in the year 2006 to estimate
the expected number of U.S.-bom births. Then 1 divided the estimated numbers by the
observed numbers of low birth weight and preterm birth infants of U.S.-bom mothers
for each census tract to calculate standardized morbidity ratios (SMR) by census tract.

Three different standardized morbidity ratios could be observed: SMR >1 meaning

30

that the observed numbers of low birth weight or preterm birth was greater than
expected, SMR=1 meaning that the observed and the expected numbers were
relatively equal; or SMR <1 meaning that the observed numbers of low birth weight
or preterm births were less than expected.

3 .2. 3 Spatial Statistics Analyses

Due to the rapid development of Geographic Information Systems (GIS) in recent
years, spatial data analysis has received considerable attention and played an
important role in social science. Spatial means that each individual record has a
geographical reference that is important in understanding the local environment in
which mothers and infants are exposed. Using spatial statistics, 1 was able to identify
where clusters of immigrant populations reside in the capital tri-county area in order
to better integrate individual- and neighborhood-level risk factors.

To calculate the racial residential segregation indices 1 used Anselin’s local
Moran’s 1 (Anselin, 1995). This index measures the level of spatial autocorrelation for
each census tract. 1 used the hot spot analysis to calculate the Getis-Ord Gi” statistics
(Getis and 0rd, 1992), another measure of spatial autocorrelation to map the clusters
of adverse birth outcomes, poverty, and different racial/ethnic groups. in the capital
tri-county area. The G-statistic showed whether features with high values or features

with low values tend to cluster in my study area. If a feature's value is high, and the

31

values for all of the neighborhood features are also high; it is a part of the hot spots.

Furthermore, in order to understand how individual- and neighborhood-level risk
factors affect the adverse birth outcomes for U.S.-bom and foreign-born mothers, 1
estimated logistic regression models and geographically weighted regression (GWR)
models. The logistic regression models were estimated in SPSS (v 17) (SPSS, 2009)
and the GWR models were estimated using the GWR function in the Spatial Statistics
tool in ArcGIS (v 9.3.1) (ESRI, 2009).

In both of these analyses the dependent variables were the low birth weight and
preterm birth rates and the independent variables included matemal- and infant-level
variables and/or neighborhood-level variables as described in the previous sections.
Binary logistic regression models were used to estimate the association between the
dependent variable(s) and each independent variable separately. The GWR models
were estimated to examine geographical differences in individual- and neighborhood-
level risk factors for low birth weight and preterm birth and where those differences
are for U.S-bom and foreign-bom mothers and infants. All analyses were performed
using the software ArcGIS (v 9.3.1)(ESR1, 2009) and SPSS (v 17) (SPSS, 2009).

An example of these models may include but are not limited to: .

P. = a0 + alMAGE + or

I 2 MARTIAL + a

CARE + a7SMOKE + a

EDU + a ORIGIN + aSINSURE +

3 4
a6 8INCOME + a9POVERTY + a1 ODENSITY + 8’.

where Pi is the adverse birth outcome under the hypothesis that individual-level

32

risk factors, such as maternal age (MAGE), marital status (MARITAL), education

level (EDU), mother’s country of origin (ORIGIN), insurance coverage (INSURE),

trimester of prenatal care (CARE), smoke during pregnancy (SMOKE),

neighborhood-level risk factors, including median household income (INCOME),

ratio of income to poverty threshold by household type (POVERTY), and immigrant

density (DENSITY), and an error term may in part affect birth outcomes.

33

4.0 Results

There were 73,143 valid data in this thesis research. The total number of births of
U.S.-bom mothers was 67,515 while 5,120 (7.6%) counted as low birth weight and
6,675 (9.9%) counted as preterm birth. As for foreign-bom mothers, the total
population was 5,628, and 408 (6.6%) births were low birth weight and 486 (7.8%)
were preterm births.
4.1 Individual and Neighborhood Level Risk Factor
4.1.1 Frequencies

Table 1 presents the basic characteristics of the populations in the capital
tri-county area in Michigan on the individual and neighborhood levels. At the
individual level, I found a higher percentage of U.S.-bom mothers being pregnant
before the ages 20 while the percentage of maternal age greater than 34 years was
higher among foreign-bom mothers. This observation somehow could be explained
by the distribution of maternal education that more fore'ign-bom mothers had higher
education compared to U.S.-bom mothers (59.6% vs. 49.0%).

Marital status made a great difference between U.S.-bom and foreign-bom in my
study. The percentage of single foreign-bom mothers was 5.1% compared to 13.3% of
single U.S.-born mothers; however, the percentage of acknowledgment of paternity

was only 8.1% in foreign-bom but 22.4% in U.S-bom mothers. And the percentage of

34

tobacco use during pregnancy was significant as well: 11.4% in U.S.-bom and 1.8%
in foreign-bom mothers. As for prenatal care, the numbers were mainly the same for
the U.S-bom and foreign-bom mothers: less than 1% of the population did not receive
prenatal care at all while more than 85% of the population started the prenatal care
from the first trimester.

On the neighborhood level, 1 found that more foreign-bom mothers lived in
highly-segregated neighborhoods compared to U.S.-bom mothers (35.4% vs. 10.4%).
The percentage of foreign-bom living in poverty level was 25.5% while it was 10.3%
for U.S.-bom women. As for immigrant density, more than 70% of the U.S.-bom
women lived in low immigrant density neighborhoods, whereas the foreign-bom
women were relatively evenly distributed into low (38.5%), medium (24.4%), and
high (37.0%) immigrant density neighborhoods.

4.1.2 Spatial Distribution

Figure 1 is the map of the study area for this research. It is called the capital
tri-county area in Michigan. The county to the north is Clinton County; to the
southeast is Ingham County, and to the southwest is Eaton County. Michigan’s capital
city is Lansing, which is comprised and surrounded by all three counties but it
primarily in Ingham County.

Figures 2 and 3 show the poverty level in the capital tri-county area. It is

35

significant that the central capital and the campus areas were poor areas while the
areas northwestern and southeastern of the capital were regarded as rich areas. Figure
4 indicates that the median household income of the capital area was only $6,250 to
$20,271 compared to the suburban areas, where there were more extremely rich
populations clustered and the median household income were above $44,667.

Figure 5 shows the distribution of immigrant density, and I noticed that the
percentage of immigrant density in most of the areas outside of the capital was fairly
low whereas a range of immigrant density from low to high was found in the capital
area. Also, the percentage of immigrant density was much higher in the eastern and
northern areas of Michigan State University.

4.1.3 Spatial Statistics — Hot Spot Analyses

Figures 6 to 13 show the areas where different U.S.-bom and foreign-bom racial/
ethnic groups tend to cluster more in the capital tri-county area in Michigan. It is
significant that U.S.-bom Caucasians were more likely to live in the south of capital
tri-county area while the capital area was a significant cold spot for them to live.
Foreign-bom Caucasians, on the other hand, were found living closer to the areas of
Michigan State University more. U.S.-bom African Americans were more likely to
live in the southwestern parts of the capital area and some of the foreign-bom African

Americans were observed to live in those areas as well. However, other foreign-bom

36

African Americans lived closer to the campus too.

As for U.S.-bom Asians, they were most likely to live in Ingham County and

lived closer to campus and the result was similar for foreign-bom Asians. For

U.S.-born Hispanic, it was significant that the areas north of Michigan State

University were the cold spots, whereas they were more likely to live in the

southwestern parts of the capital area. There were some overlapped areas for

U.S.-born African Americans and U.S.-bom Hispanic, but apparently the latter groups

spread their clustered neighborhoods toward the direction of west and south more.

The result of foreign-bom Hispanic was similar to the result of both U.S.-bom and

foreign-bom Asians that they were found living closer to the campus.

4.1.4 Spatial Statistics — Geographically Weighted Regressions

Figures 14 to 17 indicate how the predicted adverse birth outcomes and the

residuals distributed for U.S.-bom and foreign-bom mothers in the capital tri-county

area by using the geographically weighted regression method after adjusting for some

individual- (percentage of African Americans, percentage of no prenatal care,

percentage of no high school etc.) and neighborhood-level characteristics such as

LMiZscore and percentage of poverty. 1 found that the capital area was predicted to

have higher incidence of low birth weight and preterm birth for U.S.-bom mothers.

The suburban areas were estimated to have lower percentage of the birth outcomes.

37

Compared to the residual map, I also found this model explained fairly well in the
suburban area and most of the central capital area.

On the contrast, foreign-bom mothers were predicted to have higher low birth
weight in the west side of Eaton County and some areas of the capital area. Also,
foreign-bom mothers were estimated to have lower preterm birth in the capital area
and the areas east and south of the capital. However, the results of foreign-bom
mothers were not explained as well as the one of U.S.-bom mothers in this model.

4.2 Adverse Birth Outcomes
4.2.1 Frequencies

Table 2 shows the mean infant birth weight of U.S.-bom mothers was slightly
higher than that of foreign-born mothers; however, the percentage of low birth weight
of U.S.-bom mothers was also higher. On the other hand, the mean gestation. of
U.S.-bom mothers was 0.2 weeks shorter and the percentage of preterm birth was
close to 2% more compared to foreign-bom mothers.

Table 3 describes the birth outcomes of the different groups among foreign-bom
mothers in the capital tri-county area in more detail. Immigrants from Asian countries
especially Eastern Asia (Korea, China, Taiwan, Japan, and Hong Kong) had the
largest portion of births, followed by women from Central/ South America and

Western Europe. I found the difference of low birth weight and preterm birth varied

38

significantly by world regions of the mother’s country of origin. For example,
mothers .from Eastern Asia and Mexico had lower percentages of low birth weight and
South Asian mothers had the highest percentage of low birth weight. As for preterm
birth, Canadian mothers had the lowest percentage while the highest percentage fell
into the Western Europe group of mothers.

4. 2. 2 Spatial Distribution

Figures 18 and 19 show the hot spots where more low birth weight and preterm
birth were observed in both U.S.-bom and foreign-bom women. I found that the
places where there were more low birth weight and preterm birth were overlapped in
the capital area, however, more cases of low birth weight were reported in the western
of the capital and spread out to one area of Eaton County while there was one hot spot
of preterm birth closed to the area of Michigan State University.

Figures 20 and 21 indicate how the percentages of low birth weight and preterm
birth were distributed in the U.S.-bom women. I found many areas were overlapped
of both low birth weight and preterm birth and the percentages were relatively low in
the north part of Clinton County. The capital area was the area with higher percentage
of both low birth weight and preterm birth as well.

Moreover, figures 22 and 23 show the results from the comparison of the

percentage of low birth weight and preterm birth of U.S.-bom women to the

39

 

percentage of state-level incidence in the year 2006. I found that two areas had the
same number of low birth weight as the state of Michigan, whereas many places in the
capital area and also the Michigan State University area had more observed cases of
low birth weight than estimated. The result of preterm birth was mainly the same;
however, less observed cases were found in the capital area and one area had the
observed numbers equal to the estimated numbers.

Figures 24 to 34 show the absolute numbers of low birth weight and preterm
birth within different foreign-bom group of mothers in the capital tri-county area.
There were more cases of low birth weight and preterm birth observed in the
southeastern and the eastern parts of the capital area among the foreign-bom mothers
from Eastern Europe while mothers from Western Europe were found more cases in
the east side of Ingham County, northwest side of the Eaton County and some in the
capital area. As for the mothers from North Africa, they were found more low birth
weight and preterm birth southern of the capital area, and more cases were found in
the capital area for the women from Sub Sahara.

Foreign-bom Mexican mothers had higher numbers of low birth weight and
preterm birth in the southeastern part of Ingham County and northern parts of Eaton
County. On the contrary, more cases could be found in the capital area and the

northern part of Clinton County for foreign-bom mothers from Canada. Higher

40

numbers of low birth weight of preterm birth of mothers from Central/South America
were in the capital area and the southwestern part of that.

South Asian mothers were more likely to have low birth weight and preterm birth
close to the Michigan State University area and the part west of the capital area,
whereas more cases were found of the foreign-bom mothers fi'om Southeast Asia and
Oceania in the west side of Baton County, the capital area, and the eastern parts of
Ingham County. Women from Eastern Asia were observed to have higher numbers of
low birth weight and preterm birth in the areas east of the Michigan State University
but the areas with the highest number were west of the campus. For the mothers from
Central Asia and Middle East, more cases were found in the eastern and western part
of the capital area.

4.3 Epidemiological Paradox

Table 4 shows the results of the logistic regressions of independent risk factors
for low birth weight in singleton live births in both U.S.-bom and foreign-bom
mothers and tables 5 and 6 show the results of U.S.-bom and foreign-bom separately.

Foreign-bom mothers were less likely to have low birth weight compared to
U.S.-bom mothers OR = 0.87 (95% CI 0.76 to 1.00) and the result was statistically
significant (p-value = 0.04). As for individual- level risk factors, 1 found maternal age

greater than 34 years was significant in both U.S.-bom OR = 1.51 (95% CI 1.39 to

41

1.65) and foreign-bom mother groups OR =1.66 (95% Cl 1.28 to 2.15), whereas some

college maternal education was less likely to result in low birth weight in only

U.S.-bom mothers OR = 0.89 (95% Cl 0.83 to 0.96).

U.S.-bom African Americans were more likely to have low birth weight OR =

1.82 (95% Cl 1.08 to 1.97) than other U.S.-bom racial groups while foreign-bom

Hawaiians and Pacific Islanders had higher risk of low birth weight OR = 1.36 (95%

CI 1.08 to 1.70) compared to other foreign-bom mother groups. Having single (one

parent) listed as marital status was highly related to low birth weight in both

U.S.-bom OR = 1.48 (95% CI 1.35 to 1.63) and foreign-bom mothers OR = 1.91

(95% Cl 1.29 to 2.81), whereas acknowledgment of paternity marital status was only

significant in U.S.-bom mothers OR = 1.25 (95% CI 1.16 to 1.36). Parity was

significant merely in U.S.-bom mothers as for the second birth OR = 0.86 (95% Cl

0.80 to 0.93); however, the odds of low birth weight (versus non low birth weight)

increased from the second birth to the fourth or more birth in both U.S.-bom and

foreign-bom women even though they were not statistically significant.

Also, whether the source of payment was Medicaid or not made a significant

difference in U.S.-bom mothers OR = 1.12 (95% Cl 1.05 to 1.20) but was not

statistically significant in foreign-bom women (p-value = 0.99). Moreover, the

initiations of prenatal care were all predictive of low birth weight in U.S.-bom

42

mothers particularly the category no prenatal care OR = 1.84 (95% Cl 1.48 to 2.29);

however, only the second trimester initiation of prenatal care was more significant to

contribute to low birth weight in foreign-bom mothers OR = 0.50 (95% CI 0.31 to

0.80). Tobacco use during pregnancy was more likely to result in low birth weight in

U.S.-bom mothers OR = 1.06 (95% CI 1.03 to 1.09) and was significant in

foreign-bom mothers as well (p-value = 0.04).

Table 7 shows the results of the logistic regressions of independent risk factors

for preterm birth in singleton live births in both U.S.-bom and foreign-bom mothers

and tables 8 and 9 indicate the results of U.S.-bom and foreign-bom separately.

Foreign-bom mothers were significantly less likely to have preterm birth

compared to U.S.-bom mothers OR = 0.81 (95% Cl 0.72 to 0.92). On the individual

level, maternal age greater than 34 years was predictive to preterm birth in both

U.S.-bom OR = 1.32 (95% Cl 1.22 to 1.43) and foreign-born women OR = 1.46 (95%

CI 1.15 to 1.87), whereas a no high school maternal education was more likely to

result in preterm birth in only U.S.-bom mothers OR = 1.12 (95% Cl 1.04 to 1.20).

Compared to other U.S.-bom racial groups of mothers, U.S.-bom African

Americans had higher risk of preterm birth OR = 1.42 (95% CI 1.32 to 1.53) while

U.S.-bom Hawaiians and Pacific Inlanders were less likely to have preterm birth OR

= 0.58 (95% Cl 0.39 to 0.87). However, no significant result that contributed to

43

preterm birth was observed in the foreign-bom racial groups. Also, marital status
made a significant difference in U.S.-bom mothers; both single and acknowledgment
of paternity were at higher risk of preterm birth, OR = 1.21 (95% Cl 1.11 to 1.32) and
OR = 1.13 (95% Cl 1.05 to 1.22) respectively. Only single marital status was
statistically significant in foreign-bom mothers (p-value = 0.02).

As for parity, the risk of preterm birth gradually increased from the second birth
to the fourth or more birth in U.S.-bom mothers, and the results were significant in the
second birth OR = 0.90 (95% Cl 0.84 to 0.96) and the fourth or more birth OR = 1.27
(95% CI 1.16 to 1.39). On the contrary, in foreign-bom mothers the third birth was
more likely to contribute to preterm birth but the result was not significant either
(p-value = 0.48). Similar to the result from low birth weight, having Medicaid as
source of payment was only predictive in U.S.-bom mothers OR = 1.09 (95% Cl 1.02
to 1.15).

The initiations of prenatal care were all significant of preterm birth in U.S.-bom
women especially the category no prenatal care OR = 2.08 (95% CI 1.70 to 2.55). In
foreign-bom mothers, having no prenatal care was more likely to result in preterm
birth OR = 1.36 but was not statistically significant (p-value = 0.53). No significant
differences contributing to preterm birth were observed in U.S.-bom and foreign-bom

mothers in terms of maternal tobacco use.

44

Table 10 shows the result of how individual- and neighborhood-level
characteristics contribute to low birth weight of both U.S.-born and foreign-bom
mothers in the capital tri-county area in Michigan by using logistic regressions, and
tables 11 and 12 represent the separate outcomes of U.S.-bom and foreign-bom
mothers.

After removing other relatively non-significant matemal-level variables, 1 found
that foreign-bom women were still less likely to have low birth weight and it was still
statistically significant (p-value = 0.04). As for neighborhood-level characteristics,
poverty level contributed to low birth weight more OR = 1.12 (95% Cl 1.03 to 1.21),
but the level of segregation was not significant. Moreover, the poverty level made a
great difference in low birth weight between U.S.-bom and foreign-bom mothers that
U.S.-bom mothers were affected more by the higher level of poverty (p-value = 0.02).

Table 13 indicates the result of the logistic regressions of individual- and
neighborhood-level characteristics for preterm birth in singleton live births in both
U.S.-bom and foreign-bom mothers, and tables 14 and 15 show the results of
U.S.-bom and foreign-born separately.

Unlike the result of low birth weight from table 10, poverty level did not
crucially predict preterm birth; however, it was significant that medium-level

immigrant density was less likely to result in preterm birth OR = 0.86 (95% Cl 0.77 to

45

0.96). Comparing the results separately from two groups of mothers, 1 found that
medium-level immigrant density was more predictive to contribute to less preterm
births in U.S.-bom OR = 0.88 (95% Cl 0.78 to 0.99) (p-value = 0.03) than that in
foreign-bom mothers (p-value = 0.06).

Tables 16 and 17 present how neighborhood-level characteristics affect the birth
outcomes of low birth weight and preterm birth among different foreign-bom mother
groups. For the foreign-bom mothers from Eastern Asia, even though 67.4% of low
birth weight mothers lived in the highly-segregated neighborhoods, they had the
lowest percentage (4.2%) of low birth weight, whereas the mothers from South Asia
and Central/South America, who lived in the highly-segregated neighborhoods as well,
had the highest (10.2%) and second highest percentage (8.4%) of low birth weight.
Foreign-bom mothers from Western Europe shared the similar characteristic of
U.S.-bom mothers in that they lived in the neighborhoods where the level of
segregation was lower; however, they still had fairly high percentage of low birth
weight (8.2%).

Foreign-bom mothers from North Africa and Sub Sahara had higher percentage
of low birth weight and lived in the neighborhoods with the highest (44.0%) and the
second highest (41.9%) poverty level compared to other foreign-bom groups in the

capital tri-county area in Michigan. Similarly, mothers from Canada, Eastern Europe,

46

and Mexico lived in a lower level of poverty neighborhoods and had relatively less

percentage of low birth weight. Again, having similar characteristic of the U.S.-bom

mothers, foreign-bom mothers from Western Europe, despite the fact that 84.4% of

the low birth weight women lived in low poverty neighborhood, they had higher

percentage of low birth weight.

As for the level of immigrant density, the result is varied among groups again.

62.8% of the low birth weight foreign-bom mothers from Eastern Asia lived in the

high immigrant density neighborhoods and had the lowest percentage of low birth

weight while 55.6% of low birth weight mothers from North Africa lived in a high

immigrant density neighborhoods but the percentage of low birth weight was fairly

high (7.9%). On the other hand, 83.3% of the low birth weight foreign-bom mothers

from Canada lived in the low immigrant density neighborhoods and had relatively

good birth outcome of low birth weight while the mothers from Western Europe, who

lived in the lower immigrant density neighborhoods as well, had a higher percentage

of low birth weight (8.2%).

The result of how the neighborhood-level characteristics affect the birth outcome

of preterm birth within different foreign-bom groups and in the U.S.-bom mothers is

similar to the result of low birth weight. Despite the fact that 67.2% of the preterm

birth foreign-bom mothers from Eastern Asia lived in the highly-segregated

47

neighborhoods; the percentage of preterm birth was lower (6.5%). On the contrary,
foreign-bom mothers from Canada had the lowest percentage of preterm birth (5.7%)
but 84.6% of the preterm birth women lived in the neighborhoods where the level of
segregation was lower.

The mothers from Canada, Eastern Europe, and Mexico had relatively low
percentages of preterm birth and lived in the lower poverty level neighborhoods,
while 38.8% of the preterm birth mothers from Central/South America lived in the
higher poverty level neighborhoods and had the second highest percentage of preterm
birth (9.0%). On the other hand, 50% of the preterm birth mothers from North Africa
lived in the high poverty level neighborhoods, the percentage of preterm birth was
relatively low (7.0%).

As for the level of immigrant density, 84.6% of the preterm birth Canadian
mothers lived in the lower immigrant density neighborhoods and had the best birth
outcome of preterm birth, whereas the mothers from Eastern Asia, who had the
second lowest percentage of preterm birth, were more likely to live in the higher
immigrant density neighborhoods. The foreign-bom mothers from North Africa
shared the similar characteristic as mothers from Eastern Asia; the percentage of
preterm birth was 7.0% while 50% of the preterm birth North African mothers lived in

the higher immigrant density neighborhoods.

48

5.0. Discussion

Previous research has shown that the incidence of low birth weight is increasing
among U.S.-bom mothers, while the rate for immigrant mothers is still relatively low
(Gould et al., 2003; Madan et al., 2006; El Reda et al., 2007). This phenomenon of
immigrant mothers having improved birth outcomes compared to U.S.-bom mothers
is referred to in the literature as the epidemiological paradox (Gould et al., 2003). And
the goals of my thesis are to compare the incidence of low birth weight and preterm
birth in U.S.-bom and foreign-bom mothers to determine if an epidemiological
paradox exists in the capital tri-county area in Michigan and to explore both
individual- and neighborhood- level risk factors that may contribute to these
differences.

The first hypothesis that foreign-bom mothers will have a lower incidence of low
birth weight births and preterm births than US born mothers was supported. Even
though the mean infant weight of foreign-bom mothers was 44.6 g less compared to
U.S.-bom mothers, foreign-bom mothers had lower rate of low birth weight. The
difference of the rates of preterm birth between U.S.-bom and foreign-bom mothers
was more crucial. In summary, the result of the current analysis demonstrated that
migrant selectivity and the protective factors that have given the foreign-bom a health

advantage in national or state studies have similar influence among the foreign-bom

49

 

bl

mothers in the capital tri-county area in Michigan.

The test of my second hypothesis that the incidences of low birth weight and
preterm birth will differ from different groups of foreign-bom mothers was supported
as well. The result was in agreement to the findings from Madan et a1 (2006) that
foreign-bom Mexican mothers had better birth outcomes while foreign-born
Asian-Indian had a paradoxically higher incidence of low birth weight and preterm
birth. Nevertheless, compared to the result from Urquia et a1 (2009) that migrants
from Eastern Europe and Central Asia had better birth outcomes than migrants from
others regions of the world, I found that foreign-bom mothers from Eastern Asia,
Mexico, and Canada had lower percentage of low birth weight and mothers from
Canada, Eastern Asia, and North Africa had lower percentage of preterm birth.

The third and last hypothesis that individual-level risk factors for low birth
weight and preterm birth are more important for foreign-bom mothers while
neighborhood-level risk factors have stronger affect on U.S.-bom mothers were
observed by Urquia et a1 (2009) as well. Their study pointed out that the risk of low
birth weight varied considerably according to the regions of origin of the immigrant
mothers; the country of origin appears to be a much more important factor in low
birth weight among children of recent immigrants than the neighborhood in which

they currently live.

50

From the logistic regressions, I found that living in a high poverty level
neighborhood was more likely to result in low birth weight in U.S.-bom mothers
while the result was not significant in foreign-bom mothers. Similarly, living in a
medium immigrant density neighborhood affected U.S.-bom mothers more compared
to foreign-bom mothers. And the reason was mainly because U.S.-bom mothers were
exposed to the environment more than foreign-bom mothers, who were not yet
influenced by their surrounding neighborhoods.

Moreover, Auger et a1 (2009) had the observation that high immigrant density
was protective against preterm birth for Canadian-bom mothers, but unfavorably
associated with preterm birth for foreign-bom mothers. 1 had the different findings
that even though relatively high percentage of low birth weight mother from the
foreign-bom groups of Eastern Asia and North Africa lived in the high immigrant
density neighborhoods, their birth outcomes of preterm birth were better, whereas
high immigrant density was not protective against preterm birth for U.S.-bom
mothers.

This study is subject to limitations. First, the total number of foreign-bom
mothers from 1995 to 2007 was still relatively small. Second, 1 was not able to
account for alcohol and illicit use during pregnancy, and these risk factors were

proven to contribute to significant differences of adverse birth outcomes. Also, I could

51

only obtain the data from the US. Census Bureau of the year 1999, such as immigrant

density and median household income, and this may mediate the results of the adverse

birth outcomes. Last, there was no information about mother’s diet or social support-

extended family network systems, which may explain the epidemiological paradox.

52

6.0. Conclusion

This thesis research demonstrates the association between both individual- and

neighborhood-level risk factors and adverse birth outcome for U.S.-bom and

foreign-bom mothers in the capital tri-county area in Michigan. The mechanism of .

epidemiological paradox, referring to immigrant mothers having improved birth

outcomes compared to U.S.-bom mothers, that was found in previous studies was also

observed in this thesis research. The country of origins of the foreign-bom mothers

made big differences in the birth outcomes of low birth weight and preterm birth too.

Moreover, while neighborhood-level risk factors such as level of poverty and level of

immigrant density had great influences for U.S.-born mother, foreign-bom mothers

were affected more by the individual-level risk factors such as maternal education and

marital status.

Future research can be conducted in other regions where there are greater

numbers of immigrants to see if the epidemiological paradox persists. Also, more

individual- and neighborhood— level risk factors can be included in the future study,

such as diets, family support, level of stress of mothers during pregnancy, and type of

housing. Interviewing mothers, if possible, can help better understand how personal

emotion such as stress may result in adverse birth outcomes. Moreover, instead of

having all different foreign-bom mothers as study groups, focusing on certain groups

53

of immigrant may be a good way as well. For example, the research can focus on the

adverse birth outcomes of Chinese and Mexican or Indian and Korean.

Finally, this research on the epidemiological paradox is important for

understanding ethnic differences in birth outcomes by origin of birth. Future research

should also investigate the second-generation of immigrant women in the United

States to see if their birth advantage changes and drifts more toward that of the

U.S.-bom mothers and infants. If it does then health care and public health programs

and policy must be adjusted to improve the health of mothers and children born in the

United States and to maintain the health immigrant mothers and children coming to

the United States.

54

APPENDIX A

55

Table 1 Characteristics of U.S.-bom and foreign-bom mothers and infants in the capital

tri-county area in Michigan, 1995-2007.

 

U.S.-boru mothers

Foreign-born mothers

 

 

Individual-level (N = 67,515) (N = 5,628)
characteristic (%) (%)

Maternal age

< 20 years 10.7 4.2

20 — 34 years 77.7 80.0

>34 years 11.6 15.8
Maternal education

No high school 20.7 14.9

High school diploma 29.3 22.9

Some college 49.0 59.6
Marital status

Single (one parent) 13.3 5.1

Married (two parents) 64.3 86.8

Acknowledgment of paternity 22.4 8.1
Parity

0 40.2 43.0

1 33.6 34.1

2 16.6 12.6

2 3 9.1 9.9
Source of payment

Private insurance 63 .0 57.9

Medicaid 35.6 40.9

Self pay and other 0.6 0.6
Initiation of prenatal care

None 0.9 0.7

First trimester 87.3 86.5

Second trimester 7.4 7.9

Third trimester 1.4 1.5
Kessner index"

Adequate 82.9 82.5

Intermediate 9.7 1 0.3

Inadequate 6.4 6.3

Continue

56

Table l (cont’d)
Substance use
Tobacco use 11.4 1.8

Neighborhood-level

characteristics

 

 

 

Level of segregation

Low (LMiZcore < 1.96) 89.6 64.6

High (LMiZcore _>_ 1.96) 10.4 35.4
Level of poverty

Low (< 30%) 89.7 74.5

High (2 30%) 10.3 25.5
Level of immigrant density

Low (0% - 5%) 72.8 38.5

Medium (5% - 10%) 18.6 24.4

High (10% - 62%) 8.7 37.0

 

* Kessner index is a classification of prenatal care developed by the Institute of Medicine in 1973 that
adjusts the timing and quantity of prenatal care for the length of gestation to determine levels of
adequate, inadequate, and intermediate prenatal care.

Source: Women Health Dictionaries http://womenhealth.medical-dictionaries.org/

57

Table 2 Birth outcomes of U.S.-born and foreign-born mothers and infants in the capital
tri-county area in Michigan, 1995-2007.

 

 

Birth U.S.-born mothers Foreign-born mothers
outcomes (N = 67,515) (N = 5,628)
Mean infant birth weight (g) 3340.8 3296.2
(SD = 629.6) (SD = 606.9)
% Low birth weight“ 7.6 6.6
Mean gestation (weeks) 38.5 38.7
(SD=2.1) (SD=2.1)
% Prematurity“ 9.9 7.8

 

* Low birth weight = infant birth < 2,500 g

** Prematurity = gestations < 37 weeks

58

Table 3 Birth outcomes within different foreign-bom groups by country/region of origin in

the capital tri-county area in Michigan, 1995-2007.

 

 

Births Mean birth Low birth Mean Prematurity“
Country/region No. (%) weight (g) weight* gestations (%)
of origin (%) (weeks)

Eastern Europe 263 (4.7) 3493.9 5.3 38.8 7.2
(SD = 615.2) (SD = 1.9)

Western Europe 551 (9.8) 3355.6 8.2 38.6 9.4
(SD = 619.6) (SD = 1.8)

North Africa 114 (2.0) 3319.6 7.9 38.9 7.0
(SD = 560.4) (SD = 1.6)

Sub Sahara 407 (7.2) 3307.0 7.6 38.6 8.1
(SD = 643.4) (SD = 2.3)

Canada 229 (4.1) 3438.3 5.2 38.7 5.7
(SD = 621.5) (SD = 2.2)

Mexico 528 (9.4) 3306.3 4.5 38.6 7.6
(SD = 627.8) (SD = 2.2)

Central/South 569 (10.1) 3296.7 6.5 38.5 9.0

America (SD = 662.6) (SD = 2.6)

Eastern Asia 1029 (18.3) 3325.2 4.2 38.8 6.5
(SD = 574.4) (SD = 1.9)

Central Asia/ 522 (9.3) 3261.6 8.4 38.6 7.7

Middle East (SD = 661.0) (SD = 2.2)

South Asia 461 (8.2) 3163.6 10.2 38.6 8.2
(SD = 561.6) (SD = 1.8)

Southeast Asia/ 904 (16.1) 3213.7 7.1 38.7 8.4

Oceania (SD = 521.8) (SD = 1.8)

 

* Low birth weight = infant birth < 2,500 g

** Prematurity = gestations < 37 weeks

59

Table 4 Binary logistic regression analyses of low birth weight in the capital tri-county area
in Michigan, 1995-2007.

 

 

 

11 Standard Odds ratio p-value
error (95% CI)
Intercept -2.72 0.04 0.07 <0.01
Maternal age
< 20 years -0.05 0.05 0.95 (0.86, 1.05) 0.35
20 — 34 years (ref.)
> 34 years 0.43 0.04 1.53 (1.41, 1.66) <0.01
Maternal education
No high school 0.08 0.04 1.08 (1.00, 1.17) 0.06
High school diploma (ref.)
Some college -0.11 0.04 0.89 (0.83, 0.96) <0.01
Country of origin
U.S.-born (ref.) .
Foreign-born -0.14 0.07 0.87 (0.76, 1.00) 0.04
Maternal race
Caucasian (ref.)
African American 0.58 0.04 1.79 (1.66, 1.94) <0.01
American Indian -0.13 0.25 0.88 (0.54, 1.42) 0.60
Asian -0.10 0.20 0.91 (0.62, 1.33) 0.62
Hawaiian and Pacific 0.23 0.10 1.26 ( 1.04, 1.51) 0.02
Islander
Marital status
Single (one parent) 0.41 0.05 1.51 (1.38, 1.66) <0.01
Married (two parents)
(ref.)
Acknowledgment of 0.23 0.04 1.25 (1.16, 1.36) <0.01
paternity
Parity
0 (ref.)
1 -0.15 0.04 0.87 (0.81, 0.93) <0.01
2 -0.02 0.04 0.98 (0.90, 1.07) 0.66
2 3 0.10 0.05 1.11 (1.00, 1.22) 0.05
Continue

60

Table 4 (Cont’d)

 

Source of payment
Private insurance (ref.)
Medicaid
Self pay and other

Initiation of prenatal care
None
First trimester (ref.)
Second trimester
Third trimester

Maternal tobacco use
Non-smoker (ref.)

Smoker

0.1 1
-0.06

0.58

-0.28

-0.67

0.06

0.03
0.19

0.11

0.06

0.15

0.01

1.11 (1.05, 1.19)
0.95 (0.65, 1.33)

1.78 (1.43, 2.21)
0.76 (0.68, 0.85)

0.51 (0.39, 0.69)

1.06 (1.03, 1.09)

<0.01
0.78

<0.01

<0.01

<0.01

0.18

 

*Adjusted for all factors listed.

61

Table 5 Binary logistic regression analyses of low birth weight of U.S.-born mother group in
the capital tri-county area in Michigan, 1995-2007.

 

 

 

[1 Standard Odds ratio p-value
error (95% CI)
Intercept -2.72 0.04 0.07 <0.01
Maternal age
< 20 years -0.06 0.05 0.94 (0.85, 1.04) 0.24
20 — 34 years (ref.)
> 34 years 0.41 0.05 1.51 (1.39, 1.65) <0.01
Maternal education
No high school 0.10 0.04 1.11 (1.02, 1.20) 0.01
High school diploma (ref.)
Some college -0.11 0.04 0.89 (0.83, 0.96) <0.01
Maternal race
Caucasian (ref.)
African American 0.60 0.04 1.82 (1.08, 1.97) <0.01
American Indian -0.06 0.25 0.94 (0.58, 1.53) 0.81
Asian 0.03 0.37 1.03 (0.50, 2.12) 0.94
Hawaiian and Pacific 022 0.20 0.80 (0.54, 1.19) 0.28
Islander
Marital status
Single (one parent) 0.39 0.05 1.48 (1.35, 1.63) <0.01
Man'ied (two parents) (ref.)
Acknowledgment of 0.23 0.04 1.25 (1.16, 1.36) <0.01
paternity
Parity
0 (ref.)
1 -0.15 0.04 0.86 (0.80, 0.93) <0.01
2 -0.01 0.04 1.00 (0.91, 1.08) 0.81
2 3 0.12 0.05 1.13 (1.02, 1.25) 0.02
Source of payment
Private insurance (ref.)
Medicaid 0.12 0.03 1.12 (1.05, 1.20) <0.01
Self pay and other -0.02 0.20 0.98 (0.67, 1.43) 0.90
Continue

62

Table 5 (Cont’d)

 

Initiation of prenatal care

None 0.61
First trimester (ref.)

Second trimester 025
Third trimester -0.73

Maternal tobacco use
Non-smoker (ref.)
Smoker 0.05

0.11

0.06
0.12

0.02

1.84 (1.48, 2.29)

0.78 (0.69, 0.87)
0.48 (0.35, 0.66)

1.06 (0.13, 1.09)

<0.01

<0.01
<0.0l

<0.01

 

*Adjusted for all factors listed.

63

Table 6 Binary logistic regression analyses of low birth weight of foreign-bom mother group

in the tri-county area in Michigan, 1995-2007.

 

 

 

[1 Standard Odds ratio p—value
error (95% Cl)
Intercept -2.67 0.15 0.07 <0.01
Maternal age
< 20 years 0.05 0.26 1.05 (0.62, 1.76) 0.85
20 - 34 years (ref.)
> 34 years 0.51 0.13 1.66 (1.28, 2.15) <0.01
Maternal education
No high school -0.23 0.15 0.80 (0.59, 1.08) 0.14
High school diploma (ref.)
Some college -0.14 0.13 0.87 (0.67, 1.12) 0.28
Maternal race
Caucasian (ref.)
African American 0.28 0.16 1.33 (0.96, 1.83) 0.08
American Indian" -18.47 7840.11 0.00 (0.00, 0.00) 0.99
Asian -0.31 0.24 0.74 (0.46, 1.17) 0.20
Hawaiian and Pacific 0.31 0.12 1.36 (1.08, 1.70) <0.01
Islander
Marital status
Single (one parent) 0.64 0.20 1.91 (1.29, 2.81) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.03 0.20 1.03 (0.70, 1.53) 0.88
paternity
Parity
0 (ref.)
1 -0.14 0.12 0.87 (0.69, 1.11) 0.26
2 -0.15 0.17 0.86 (0.62, 1.20) 0.38
2 3 -0.07 0.19 0.93 (0.64, 1.35) 0.71
Source of payment
Private insurance (ref.)
Medicaid 0.00 0.11 1.00 (0.81, 1.24) 0.99
Self pay and other -0.88 1.02 0.41 (0.06, 3.07) 0.39
Continue

64

Table 6 (Cont’d)

 

Initiation of prenatal care
None
First trimester (ref.)
Second trimester
Third trimester
Maternal tobacco use
Non-smoker (ref.)

Smoker

0.22

-0.70
-0.18

0.10

0.54

0.24
0.43

0.05

1.24 (0.43, 3.57)

0.50 (0.31, 0.80)
0.83 (0.36, 1.93)

1.10(1.01,1.21)

0.69

<0.01
0.67

0.04

 

*Adjusted for all factors listed.

*"' The total number of foreign-bom American Indian was too small (N = 26), and thus the result was

not significant in this model.

65

Table 7 Binary logistic regression analyses of preterm birth in the capital tri-county area in
Michigan, 1995-2007.

 

 

 

[1 Standard Odds ratio p—value
error (95% CI)
Intercept -2.27 0.04 0.10 <0.01
Maternal age
< 20 years -0.08 0.05 0.93 (0.84, 1.02) 0.11
20 — 34 years (ref.)
> 34 years 0.29 0.04 1.34 (1.24, 1.44) <0.01
Maternal education
No high school 0.08 0.04 1.09 (1.01, 1.17) 0.02
High school diploma (ref.)
Some college -0.06 0.03 0.94 (0.89, 1.01) 0.07
Country of origin
U.S.-bom (ref.)
Foreign-bom -0.02 0.06 0.81 (0.72, 0.92) <0.01
Maternal race
Caucasian (ref.)
African American 0.33 0.04 1.40 (1.30, 1.50) <0.01
American Indian -0.19 0.23 0.83 (0.54, 1.29) 0.41
Asian 0.13 0.16 1.14 (0.84, 1.55) 0.40
Hawaiian and Pacific -0.11 0.09 0.89 (0.75, 1.07) 0.23
Islander
Marital status
Single (one parent) 0.21 0.04 1.23 (1.13, 1.34) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.12 0.04 1.13 (1.05, 1.21) <0.01
paternity
Parity
0 (ref.) .
1 -0.1 l 0.03 0.89 (0.84, 0.95) <0.01
2 0.02 0.04 1.02 (0.95, 1.10) 0.58
2 3 0.23 0.05 1.26 (1.15, 1.37) <0.01
Continue

66

Table 7 (Cont’d)

 

Source of payment
Private insurance (ref.)
Medicaid
Self pay and other

Initiation of prenatal care
None
First trimester (ref.)
Second trimester
Third trimester

Maternal tobacco use
Non-smoker (ref.)

Smoker

0.09
-0.26

0.70

-0.27

-0.60

0.02

0.03
0.19

0.10

0.05

0.14

0.01

1.09 (1.03, 1.15)
0.77 (0.53, 1.11)

2.01 (1.64, 2.45)
0.76 (0.69, 0.84)

0.55 (0.42, 0.72)

1.02 (1.00, 1.05)

<0.01
0.16

<0.01

<0.01

<0.0l

0.10

 

*Adjusted for all factors listed.

67

Table 8 Binary logistic regression analyses of preterm birth of U.S.-bom mother group in the
capital tri-county area in Michigan, 1995-2007.

 

 

 

11 Standard Odds ratio p-value
error (95% CI)
Intercept -2.28 0.04 0.10 <0.01
Maternal age
< 20 years -0.10 0.05 0.09 (0.83, 1.01) 0.06
20 — 34 years (ref.)
> 34 years -0.28 0.04 1.32 (1.22, 1.43) <0.01
Maternal education
No high school 0.11 0.04 1.12 (1.04, 1.20) <0.01
High school diploma (ref.)
Some college -0.05 0.03 0.95 (0.89, 1.01) 0.11
Maternal race
Caucasian (ref.)
African American 0.35 0.04 1.42 (1.32, 1.53) <0.01
American Indian -0.12 0.23 0.90 (0.58, 1.40) 0.63
Asian 0.16 0.31 1.18 (0.64, 2.16) 0.60
Hawaiian and Pacific -0.55 0.20 0.58 (0.39, 0.87) <0.01
Islander
Marital status
Single (one parent) 0.19 0.05 1.21 (1.11, 1.32) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.12 0.04 1.13 (1.05, 1.22) <0.01
paternity
Parity
0 (ref.)
1 -0.11 0.03 0.90 (0.84, 0.96) <0.01
2 0.02 0.04 1.02 (0.94, 1.10) 0.62
2 3 0.24 0.05 1.27 (1.16, 1.39) <0.01
Source of payment
Private insurance (ref.)
Medicaid 0.08 0.03 1.09 (1.02, 1.15) <0.01
Self pay and other -0.32 0.20 0.73 (0.50, 1.07) 0.11
Continue

68

Table 8 (Cont’d)

 

Initiation of prenatal care
None
First trimester (ref.)
Second trimester
Third trimester
Maternal tobacco use
Non-smoker (ref.)

Smoker

0.73

-0.27
-0.65

0.02

0.10

0.06
0.14

0.02

2.08 (1.70, 2.55)

0.77 (0.69, 0.85)
0.52 (0.40, 0.69)

1.02 (1.00, 1.05)

<00]

<00]
<00]

0.10

 

*Adjusted for all factors listed.

69

Table 9 Binary logistic regression analyses of preterm birth of foreign-bom mother group in

the capital tri-county area in Michigan, 1995-2007.

 

 

 

11 Standard Odds ratio p-value
error (95% CI)
Intercept -2.34 0.14 0.10 <0.01
Maternal age
< 20 years 0.08 0.25 1.09 (0.67, 1.77) 0.74
20 - 34 years (ref.)
> 34 years 0.38 0.12 1.46 (1.15, 1.87) <0.01
Maternal education
No high school ~0.21 0.14 0.81 (0.62, 1.07) 0.14
High school diploma (ref.)
Some college -0.16 0.12 0.85 (0.67, 1.08) 0.19
Maternal race
Caucasian (ref.)
African American 0.05 0.16 1.05 (0.77, 1.43) 0.75
American Indian” -18.83 8168.45 0.00 (0.00, 0.00) 0.99
Asian 0.05 0.19 1.05 (0.73, 1.52) 0.78
Hawaiian and Pacific 0.04 0.11 1.04 (0.84, 1.29) 0.72
Islander
Marital status
Single (one parent) 0.45 0.19 1.56 (1.07, 2.29) 0.02
Married (two parents) (ref.)
Acknowledgment of -0.00 0.18 1.00 (0.70, 1.43) 0.99
paternity
Parity
0 (ref.)
1 -0.19 0.11 0.82 (0.67, 1.03) 0.09
2 0.12 0.15 1.11 (0.83, 1.49) 0.48
2 3 0.07 0.17 1.08 (0.77, 1.51) 0.67
Source of payment
Private insurance (ref.)
Medicaid 0.07 0.10 1.07 (0.87, 1.31) 0.52
Self pay and other 0.23 0.62 1.26 (0.38, 4.21) 0.71
Continue

70

Table 9 (Cont’d)

 

Initiation of prenatal care

None 0.31 0.49 1.36 (0.52, 3.56)
First trimester (ref.)

Second trimester -0.35 0.20 0.71 (0.48, 1.05)
Third trimester -0.26 0.43 0.77 (0.33, 1.80)

Maternal tobacco use
Non-smoker (ref.)
Smoker 0.04 0.05 1.05 (0.95, 1.16)

0.53

0.08
0.55

0.39

 

*Adjusted for all factors listed.

** The total number of foreign-bom American Indian was too small (N = 26), and thus the result was

not significant in this model.

71

Table 10 Binary logistic regression analyses of low birth weight of both individual-level and
neighborhood-level characteristics in the capital tri-county area in Michigan, 1995-2007.

 

 

 

11 Standard Odds ratio p—value
error (95% CI)
Intercept -2.72 0.03 0.07 <0.01
Maternal age
< 20 years -0.06 0.05 0.95 (0.86, 1.04) 0.24
20 — 34 years (ref.)
> 34 years 0.45 0.04 1.56 (1.44, 1.70) <0.01
Maternal education
No high school 0.09 0.04 1.09 (1.01, 1.18) 0.03
High school diploma (ref.)
Some college 013 0.04 0.88 (0.82, 0.94) <0.01
Country of origin
U.S.-bom (ref.)
Foreign-bom -0.14 0.07 0.87 (0.76, 0.99) 0.04
Maternal race
Caucasian (ref.)
African American 0.58 0.04 1.78 (1.65, 1.92) <0.01
American Indian -0.12 0.25 0.88 (0.55, 1.43) 0.61
Asian -0.11 0.20 0.90 (0.61, 1.32) 0.58
Hawaiian and Pacific 0.23 0.10 1.26( 1.05, 1.52) 0.02
Islander
Marital status
Single (one parent) 0.46 0.05 1.58 (1.45, 1.73) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.26 0.04 1.30 (1.21, 1.40) <0.01
paternity
Initiation of prenatal care
None 0.62 0.11 1.86 (1.50, 2.30) <0.01
First trimester (ref.)
Second trimester -0.26 0.06 0.78 (0.69, 0.87) <0.01
Third trimester -0.65 0.15 0.52 (0.39, 0.70) <0.01
Continue

72

Table 10 (Cont’d)

 

Level of segregation

Low (ref.)

High -0.06 0.05
Level of poverty

Low (ref.)

High 0.1 1 0.04

0.95 (0.86, 1.05)

1.12(l.03,1.21)

0.28

<0.01

 

*Adjusted for all factors listed.

73

Table 11 Binary logistic regression analyses of low birth weight of both individual-level and
neighborhood-level characteristics of U.S.-bom mother group in the capital tri-county area in
Michigan, 1995-2007.

 

 

 

[1 Standard Odds ratio p—value
error (95% Cr)
Intercept -2.72 0.03 0.07 <0.01
Maternal age
< 20 years -0.07 0.05 0.93 (0.84, 1.02) 0.13
20 — 34 years (ref.)
> 34 years 0.44 0.04 1.55 (1.43, 1.69) <0.01
Maternal education
No high school 0.11 0.04 1.12 (1.03, 1.21) <0.01
High school diploma (ref.)
Some college -0.14 0.04 0.87 (0.81, 0.94) <0.01
Maternal race
Caucasian (ref.)
African American 0.60 0.04 1.81 (1.65, 1.92) <0.01
American Indian -0.05 0.25 0.95 (0.55, 1.43) 0.85
Asian 0.02 0.37 1.02 (0.61, 1.32) 0.95
Hawaiian and Pacific 022 0.20 0.81 ( 1.05, 1.52) 0.28
Islander
Marital status
Single (one parent) 0.44 0.05 1.56 (1.42, 1.71) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.27 0.04 1.31 (1.21, 1.41) <0.01
paternity
Initiation of prenatal care
None 0.66 0.11 1.93 (1.56, 2.40) <0.01
First trimester (ref.)
Second trimester -0.23 0.06 0.80 (0.71, 0.89) <0.01
Third trimester -0.72 0.16 0.49 (0.36, 0.67) <0.01
Level of segregation
Low (ref.)
High -0.06 0.06 0.94 (0.84, 1.05) 0.26
Continue

74

Table 11 (Cont’d)

 

Level of poverty
Low (ref.)
High 0.10 0.05

111002121)

0.02

 

*Adjusted for all factors listed.

75

Table 12 Binary logistic regression analyses of low birth weight of both individual-level and

neighborhood-level characteristics of foreign-bom mother group in the capital tri-county area

in Michigan, 1995-2007.

 

 

 

11 Standard Odds ratio p-value
error (95% Cl)
Intercept -2.76 0. 14 0.06 <0.01
Maternal age
< 20 years 0.16 0.28 1.17 (0.68, 2.02) 0.56
20 — 34 years (ref.)
> 34 years 0.34 0.13 1.71 (1.32, 2.22) <0.01
Maternal education
No high school -0.31 0.18 0.73 (0.51, 1.05) 0.09
High school diploma (ref.)
Some college -0.12 0.13 0.88 (0.68, 1.15) 0.35
Maternal race
Caucasian (ref.)
African American 0.27 0.17 1.31 (0.94, 1.83) 0.12
American Indian" -18.48 8147.76 0.00 (0.00, 0.00) 0.99
Asian -0.24 0.24 0.78 (0.49, 1.26) 0.31
Hawaiian and Pacific 0.30 0.12 1.34 ( 1.06, 1.71) 0.02
Islander
Marital status
Single (one parent) 0.68 0.21 1.98 (1.32, 2.96) <0.01
Married (two parents) (ref.)
Acknowledgment of -0.11 0.22 0.90 (0.58, 1.39) 0.62
paternity
Initiation of prenatal care
None -0.43 0.74 0.65 (0.15, 2.74) 0.56
First trimester (ref.)
Second trimester -0.62 0.25 0.54 (0.33, 0.88) 0.01
Third trimester -0.01 0.43 1.00 (0.43, 2.32) 0.99
Level of segregation
Low (ref.)
High -0.13 0.13 0.88 (0.68, 1.14) 0.32
Continue

76

Table 12 (Cont’d)

 

Level of poverty
Low (ref.)
High -0.18 0.13 1.19 (0.93, 1.54) 0.17

 

*Adjusted for all factors listed.
** The total number of foreign-born American Indian was too small (N = 26), and thus the result was

not significant in this model.

77

Table 13 Binary logistic regression analyses of preterm birth of both individual-level and
neighborhood-level characteristics in the capital tri-county area in Michigan, 1995-2007.

 

 

 

[5 Standard Odds ratio p-value
error (95% CI)
Intercept -2.26 0.03 0.1 1 <0.01
Maternal age
< 20 years -0.11 0.05 0.90 (0.02, 0.98) 0.02
20 — 34 years (ref.)
> 34 years 0.33 0.04 1.39 (1.29, 1.49) <0.01
Maternal education
No high school 0.10 0.04 1.11 (1.03, 1.19) <0.01
High school diploma (ref.)
Some college -0.08 0.03 0.92 (0.87, 0.98) <0.01
Country of origin
U.S.-bom (ref.)
Foreign-bom -0.19 0.06 0.83 (0.73, 0.94) <0.01
Maternal race
Caucasian (ref.)
African American 0.36 0.04 1.43 (1.33, 1.54) <0.01
American Indian -0.17 0.22 0.85 (0.55, 1.32) 0.46
Asian 0.12 0.16 1.12 (0.82, 1.53) 0.47
Hawaiian and Pacific -0.10 0.09 0.90 ( 0.75, 1.08) 0.28
Islander
Marital status
Single (one parent) 0.25 0.04 1.28 (1.18, 1.39) <0.01
Married (two parents) (ref.)
Acknowledgment of 0.15 0.04 1.16 (1.09, 1.24) <0.01
paternity
Initiation of prenatal care
None 0.74 0.10 2.09 (1.72, 2.55) <0.01
First trimester (ref.)
Second trimester -0.26 0.05 0.78 (0.70, 0.86) <0.01
Third trimester -0.59 0.14 0.56 (0.43, 0.72) <0.01
Continue

78

Table 13 (Cont’d)

 

Level of segregation

Low (ref.)

High 0.06
Level of poverty

Low (ref.)

High 0.07
Level of immigrant density

Low (ref.)

Medium 015

High 0.01

0.06

0.04

0.06
0.03

1.06 (0.95, 1.18)

1.08 (0.99, 1.17)

0.86 (0.77, 0.96)
1.01 (0.95, 1.08)

0.31

0.29

<0.01
0.74

 

*Adjusted for all factors listed.

79

Table I4 Binary logistic regression analyses of preterm birth of both individual-level and
neighborhood-level characteristics of U.S.-bom mother group in the capital tri-county area in
Michigan, 1995-2007.

 

 

 

0 Standard Odds ratio p—value
error (95% CI)
Intercept -2.26 0.03 0.1 0 <0.01
Maternal age
< 20 years -1.13 0.05 0.88 (0.81, 0.97) <0.01
20 — 34 years (ref.)
> 34 years 0.32 0.04 1.38 (1.28, 1.49) <0.01
Maternal education
No high school 0.13 0.04 1.14 (1.06, 1.22) <0.01
High school diploma (ref.)
Some college -0.08 0.03 0.92 (0.87, 0.99) 0.02
Maternal race
Caucasian (ref.)
African American 0.38 0.04 1.46 (1.35, 1.57) <0.01
American Indian -0.09 0.23 0.92 (0.59, 1.43) 0.70
Asian ' 0.15 0.31 1.16 (0.63, 2.13) 0.63
Hawaiian and Pacific 054 0.20 0.58 ( 0.39, 0.87) <0.01
Islander
Marital status
Single (one parent) 0.23 0.04 1.26 (1.16, 1.37) <0.01
Married (two parents)
(ref.)
Acknowledgment of 0.16 0.04 1.17 (1.09, 1.25) <0.01
paternity
Initiation of prenatal care
None 0.77 0.10 2.17 (1.77, 2.65) <0.01
First trimester (ref.)
Second trimester -0.25 0.05 0.78 (0.70, 0.87) <0.01
Third trimester -0.63 0.14 0.53 (0.40, 0.70) <0.01
Level of segregation
Low (ref.)
High 0.06 0.06 1.07 (0.95, 1.20) 0.28
Continue

80

Table 14 (Cont’d)

 

Level of poverty
Low (ref.)
High 0.06 0.05
Level of immigrant density
Low (ref.)
Medium -0. l 3 0.06
High 0.01 0.04

1.06 (0.97, 1.16)

0.88 (0.78, 0.99)
1.01 (0.95, 1.09)

0.17

0.03
0.71

 

I"Adjusted for all factors listed.

81

Table 15 Binary logistic regression analyses of preterm birth of both individual-level and

neighborhood-level characteristics of foreign-bom mother group in the capital tri-county area

in Michigan, 1995-2007.

 

 

 

13 Standard Odds ratio p-value
error (95% CI)
Intercept -2.27 0.13 0.10 <0.01
Maternal age
< 20 years 0.14 0.26 1.15 (0.69, 1.93) 0.59
20 - 34 years (ref.)
> 34 years 0.42 0.12 1.52 (1.19, 1.94) <0.01
Maternal education I
No high school -0.31 0.17 0.73 (0.53, 1.02) 0.07
High school diploma (ref.)
Some college -0.17 0.12 0.84 (0.66, 1.08) 0.17
Maternal race
Caucasian (ref.)
African American 0.07 0.16 1.07 (0.78, 1.48) 0.66
American Indian" -18.82 8539.39 0.00 (0.00, 0.00) 0.99
Asian 0.08 0.19 1.09 (0.75, 1.59) 0.66
Hawaiian and Pacific 001 0.12 0.99 ( 0.79, 1.24) 0.91
Islander
Marital status
Single (one parent) 0.47 0.20 1.61 (1.08, 2.38) 0.02
Married (two parents)
(ref.)
Acknowledgment of -0.18 0.21 0.84 (1.56, 1.26) 0.40
paternity
Initiation of prenatal care
None -0.14 0.61 0.87 (0.26, 2.87) 0.82
First trimester (ref.)
Second trimester -0.35 0.21 0.71 (0.47, 1.07) 0.10
Third trimester -0.08 0.43 0.92 (0.40, 2.16) 0.85
Level of segregation
Low (ref.)
¥ High -0.05 0.17 0.96 (0.68, 1.34) 0.80
Continue

82

Table 15 (Cont’d)

 

Level of poverty
Low (ref.)
High 0.19 0.13
Level of immigrant density
Low (ref.)
Medium -0.26 0.19
High -0.04 0.13

1.21 (0.94, 1.56) 0.15

0.77 (0.54, 1.11) 0.06
0.96 (0.74, 1.24) 0.74

 

*Adjusted for all factors listed.

" The total number of foreign-born American Indian was too small (N = 26), and thus the result'was

not significant in this model.

83

$3 - $2 323.38 531:388 .«0 ow3=ooeoa n bicep EEEEE :wE mo\oo_

- can macaw—smog EBA—328 mo owficooeoa u bicep Haw—ESE Mo .05. E338 axon - o\oo acorn—smog Eon-=w_28 mo ”$85an I bicep EEEEE mo .05. 33 1......
.x. an m .9869 .3 owficoobq n argon mo _o>o_ cw:— mo\ocm v 5.56.“ do owficooeoa .1. .338 .«o _o>o_ 264 4......

8.. N oeoomfi—ZA n casewoewom mo :32 :wE x5.— v oeoowNmiq u cotawouwom mo .96. So; I.

w 83 v 5.5 85...; .1. Ewes 5.5 33 ..

 

 

 

do ”.2 33 of 3m 2 Es GE as m Egan:
:a 3m gm :8 SK 2: mm» : .HS 2580 33.. 286538
53. a: 4.3. EN MS 3:. Zn 3.25 was... 5:8

“mam
E... 52 2m 3m was gm 2:. is 3. 23:2 32 assoc
”.8 a: as non Woe is 3m 53 a. 22... 583
we. in QR «.2 NS o. _ N 3: 9.8 R 8:65. esogaaoo
98 5.2 new a: on. he. as Ge em 882
9o 3. on» 3 92: 52 no. so 2 2.23
Na. 3: mom a; Sm ea a: 65 a 82% 8m
6.? an _.: is. 93 4.3. can as a 8:2 552
he 4.4m 3o 92 3e _.: a: 3.8 we 82...". 5883
«.3 3a E 3; new 52 36 2.3 3 683 seam

1?: 52.82 33 $5 33 $5 :3
£9 3: 3.; 3: .oz 52.5
3:53:05 «443L959 «eaotawfiwom «EH93 be cinching—:5
«HF—35:: ho 3.54 a: _0>01— ha _o>01— at:— .53

 

.8882 59:22 5 85. 5.58-3 .838 65 5 e652:
Sonnmj 05 EB Eon—SE ho 3:on Eon-:w_2£ EpcoEw wcofia 85:88:30 _o>o_-_ooo:._on:wmo: 98 Emma? 5.5 32 5953 cote—oboe 2:. m: «Bah.

84

fame - £3. macaw—smog Eonéwfish 3 uwficooba n .3553 EEwEE. :3: 2x6—

- 3m 20:232. Eonéwéo. 3 ow8=35a n .9353 EEEEE 3 65. 6:32: ”fem - 9.6 22338 Eon-:w_o..£ .3 owficoobg n .3653 EEEEE .3 _o>o_ 3o: 1......
.x. an N 55>?“ .3 owflcooba n 55>?— .3 .05. :wE ”foam v .338 .3 owficoobq H 5.955. .3 $5. Be.— 2....

co; m 038N§§ u couawohwom .3 .05. am:— moa; v 20852.. n coumwoaom 3 BR: 304 1.

w 83 v 5.... 5...... u :32: 5.5 so. ..

 

 

 

m... 5.8 2: ..~. 3.. 3 .55 a... 558 58.5.:
:2 3m 1% 08 SK n... on. a»... 2 55.80 Em... 3.85:8
5.5. «.2 MGM 3.. 5.... 3m 5.: a... an «.2 58m

“mam
3.. ow.“ 3... 5.8 3... 3m 3... .5. 3. 2.8.2 Rm... .8500
Zn 5.2 5.8 3.... 3... NS 3m .3. S «.2 523.”.
mm .2... ... . m .5. $5 5.2 2... 8.... .m 8.35... 538.5550
2. 3m 35 ms 08 3. 2.. Gt 2. 8.52

o... 3.. 3... o... 5.8. 5.2 3... .5. m. 526
5.9. 3. «.9 .5. in NE «.3 2.... mm «555 55
9% 5.2 5.2 can 9% 2m 35 85 w 8.5... 582

5.5 5.2 5.3. 5.2 3... Na. .2... :3. mm 2.23. 5383

mm 5.. 5.5. on 5.5 ...N 3: as. a. 2.25m. 533m.

Imma— a....52 33 5w... 33 a... :3
3.. 3.. 55 £3 .oz 5&8
33.37.56 1. .3295.— ...aotawouwom «239$ 3 5323.35.50
2.25.5.5... .35 5 .26.. .c .33 5...: E:

 

. .hoomfiaa— £3282 E g .958...“ .938 2: E mafia...
Econmd 2: EB E0508 .3 anew Eon-:w_2£ 320E“. wcofim 85:32.35 _o>o_-coo€onsw_o= 98 5.3 5.88.. 503.3 3322.8 2:. 2 035,—.

85

APPENDIX B

86

433:an 5 «2a b.5095 1338 2.: we QaE 02.20QO m 9...»:—

 

bcsou :SEC B
5:300 Exam _ fi

@500 Ens»:— D
528.: 5 sea 3:33.518 2: ____

.hooméma— .SwwEomE E 8.3 353.3 .333 05 E boon .«o ems—.0209 05.3 30% 2250: RF N 9...»:—

.8_ Es 5. .8 .8 .85 a“ .2. 228.5 €98 2: 3 “£2 .8...

    

Son dam and A I

500 .3m an." - m5.— I
.BQ .Bm em; . me; I
.50 .3m 3; - no.7 I
Son— dum no.7 . em._- I
509 .3m cog- - ”ON. I
Son— .Em «WN- v I
encumNmuv

1.8...—

Hoamfiaa— .5%2034 E 8.5 b.5035 =3an 05 5 so: b05956 mo ems—523 2: mo 30% 28:0: 2: n PEEK

.oou 55 8.35 2 35 0.2.3:: $.33 05 8 Sumo.— cut bun—Penn.

    

an .3 ”3 A I

3a .Em and - 03 I

Son .3m 2m.— - no; I

son .3 3.. - 3..- I
s8 .30, 8..- - 8..- _

an .3 3;. - ”2.

sun .3 and. v

2836

.25. reaches—

 

89

.Somfima— .caw222 E no.3 mac—50;: =3an 2: E 2:85 20:852— 562: we :ousnmafiu 33% 2:. v Bani

.82 a Sci“ 83 53:22 a. 3% 28° 2:82 2288.. saga 2:.

    

ohmda - ommdh I
mmmdb - mmmmmo I
mefio - $93 I
Roe“: - mowém
Nafivm . NhNdN
KNdN - cmmd

 

93.33 .32 a. 2:8... 22.8.5: 5:52

90

.82 go 3222 s «an $38.2. 3&8 as 5 £83. “swag 3:0 8:335 swam 2: m 25E

     

2.8 . 8.2 I
8.2 - ad I

oo.m - ocd

 

3L 8.: we E22— .5235...—

91

88-3m— dmm232 E «Ba 3:28.15 358 2: E 233030 Econmd mo 30% 28:0: 2: w 2:“:—

92

 

393 an? I
son .3 ”3-2: a fi
.593 3.. - 3.. m _ ,
589m “3 -37 .‘ I _ ,
.593 37-8.7 . ‘ u .A .
58.9w 8.7-»3- HH— , ,
sooémwwwv I _ 7
BoomNfiv ,
83$qu 2:835 EB-.m.: ‘ 1 f

62532 53:32 5 8; 558+: 558 of E mfiauozno E3-.=w_28 mo 30% 28%.. 2F b BauE

 

son gm ”2A I w
39.3 ”3-2: I ,
58.38.73; a .m
.59.ch 3.73..- \
son .3 m3- - om. _- ,
39.3 8.7%.”- a

30.3 wwwv I , ; ‘ J

98550 ‘ W
Swim u 2v 233030 =._cn-=w_o..oh . i ,l

93

88-32 .3222 5 83 5.50.5 .358 2: a 285:2 58:? Easemdmo 38m 28%; 2F a 2...»:—

      

94

.>on .em an.” A
55 .3m “2 - e3
.59 9m 03 - 8;

son .3 n2 . 8.7 _ 1 1.,

son .3 n3- . 22. I m

son .3 3;. - ”3. I _

.59 .3 ”2- v I ,

I
I
I

2835 , r E
fissuzvastoiw 53:? 53%: _l y. a t ‘

.hoomfiaa— .§m_:o_2 E 83 953+: 3:98 2: :_ 28:03 58E< EBA—EBB mo 30% 29.50: 2: a 9...»:—

    

.>8 .3 ”3 A I
in .Em an.” - 3.. I
son .3 02 - 3; MI

son .3 m3 - m3- 1
son .5 n3. - 2:. m.
son .3 3.7 - ”3- I

son .3 ”3- v I

9.335
93 u ZV 2.3.505 =35 Eonéflflek

95

$93-32 5&322 E 83 b58045 =3an 2: E 9:23. Eastmd mo 30% 28:2. 2:. 3 BauE

son Em ”2A I
58.3 «2-03 I
5938.78; m M .
55.3 8.73..- ,1 ,, .. L ..
song 3.7-8..- ._: . . . N .
58.337”? E _ , . .

33% ”36 I .

 

2835 .
8.: "2:55 58$: .- r

96

Noon-mg. .fiwanz E 83 bczoo-mb .8an o... E m§_m< Eon-$.28 mo 30% 2850: of. = 9...»:—

 

.>3. .3 ”3 A I , . . . H
in .Em and - 3.. E ,. . . .-
53. .3 3.. - 3.. . m .

son .3 3.. - 2. .-
53. .3 8..- - 3..- , ,
sun .3 3..- - .2. E _ .
son. .3 ”2. v I - - _

«LoomNTr-v
:5. u 75 2:33. Eon-$.93.— . .- - _

97

SEN-mg. flaming—.4 5 no... fiasco-5 .838 o... E 2590-5 Eon-.mj no 30% 28:2. 2.... an 2:»...—

 

5o .3 33 A I A .
5.5.3 ”3-3.. I _ . - . f
55.5 3.. -3. was“. W
.50 .3 3.. - Q..- .

.>u...u.m we..-- W3..- --- . . ,
$5.2m 3..--..2- m ., .. . .
Son. .Ew wmfi- v I , 1 - .- , -iL. . l.

283.0 _
$750.33.: Egg-m... - . ,.>- z. .i.

98

Room-moa— .=mmEo_2 E no; .9550... 3:98 2: E 2.5.5: Eon-=92... mo 30% 28:0: 2:- 2 2:»:—

99

 

.50 .3 .2 A
.5o .3 ”3 - 3.. ,. M... _ . .
.5o .3 3.. - 3. , , . _

.5n. .2m 3.. - 8..- .. . - . i , -

.5o .3 3..- - 3..- H , _ .

.5a .3 3..- - ”on. a, .u,

.59 .3 ”3- v n

 

283.0 . .
53.. u z. 5.5%.: Eon-$.28. .. ,

38.33.
.5922: =. 83 9.58.... .598 2.. a. eases 58-9.9 .0 .825. 3. 2a 2.9.53 5.... so. 85...»... .8 8.25.55... 3.3% 2: z 2......—

.m..oz.. .5 .8533...
.8893». 5382.. 523955.39 as. 0.89.. £859.29 .o. 3:32..

 

3. -3: I

o... -2 I

2.2 I

3-3 I

3.2. I

35.3....

3.55 Bea-.9... 5 2.9.5.5 at... :3

.59 ,3 an“ A I

.59 .3 an.” - 3.. I
.59 fim 3.. . a... a
.59 .39 3.. - 8... D
.59 69 no..- - 3..- a
.59 .39 3..- . an”- I
.59 .39 .3- v I

Sam—Em

100

SST-32
.§w_:o_2 E a2“ .9597... 3an 05 S. 2052.: EB-.m.D «o 3:28.. m: 98 5.3 5.82.. “.30an mo cause-8% 33% 2:- m— 9...»:—

 

.2.ozn. as. among.
6833.. .32.. .o.=§.:...§.m<.. 5... :8... 653.2. 8.. 8.3.65

3. - 3.. I

3.. - n... I

n... - ad I

ed - 3 I

We - o4. I

33...»...

1.3.0 53$... he 5.... .539...

son .3 ”3 A I

sun .3 m2 - 8.. I

.59 .am 8.. - 3.. HH—

.>on .3 8.. - 8... E

sun .5 3... - 8... fl.
sun .3 3..- - .3."-

.>3 .3 ”3. v "

Ens—Em

101

$8.32 59%.:
E 83 b:=o?tu .838 05 E 2050.: Eon-£220.. mo 3:23.. a. was 3E2: 5.3 32 “.8069... no .833ng 33% 2F 3 ensur—

.....oz.. 2:. .omowns...
6828... .0582. 6.5%....522. E... .89... 98$... 8.. 8.3.8....

 

3.. -3: I
3.. - m...- I

3.9.8...
.955 Eon-aye... .c 2.393 5.... :3

son. .30. an.” A I

son. .3 an.” - 8.. I
son .5 3.. - 8.. a
.25 .3 8.. - no..-
son .5 we..- - 8..-
son. .3 8..- - .3-
son. .5 .3- v

IIUE

Eng—3m

102

«bow-noo—
.53322 E 3.8 .0589... .333 o... a. 9650:. East—.328 be 3:23.. 3. E... 5.5 8.82.. 3862.. we 533.5% :5an 2:. 5 Bani

 

£32.. Ea .onomas...
.8250: 6382.. .o.§.........§..<.. ..<._ .80.... .28....2. 8. 3.3.3.

o6. - of I

o... - m... I

m... - od 3

ed - no I

m... - o... I

38:8...—

¢A=EU Eoceuie .3 5.5. .589...—

.>oo .5 an.” A I
5.5.3 ”3.8.. I
5.5.3 8.. .3. new.
58.3 8.78..- TL
son. .3 3..--8..- ]
.>on..zm8..--..n~- I

58.3 awn-v I

Bad—5m

103

$8-3... 5.0.30.2 =. 8... 5.58.... 3...... 2.. a. 32.68 E....-..m......¢. .2... Eon-dd .0 2...... 5.... 3...... no... 28...... 2: a. 9...»...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

sun .3 an. A I

3.. .3 an. - o... I
5.. .3 0.... . 8.. D
.53 gm 8.. - 8..- in.
son. .3 8..- - ea..- D
son. .3 8..- - .3- I
son .3 .2- v I

 

2836
6mm...- " z. .535 5.... 33

104

88-32 525.2 ... 8... €58... .538 2.. ... e058. 53-.....2... as. 53%: ..o 5.... 5.32.. .8 we... 28%.. 2: a. 2...»...

 

 

 

E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

son .3 an.” A I _ .

son. Sm and - 8.. I
son .3 8.. - 3.. nu _
59.3 3.. -3..- U
5.5.3.. 2.7-8..- 3
5.5.5 8..--..3- I a

Son— .me and- V I

 

 

 

 

 

 

 

 

 

28mm...
82.5 n E at... 5.32..

105

Hooa$¢2
.5m322 E 3.... .0583.» :3an 05 E 5:83 Eastmd E 3E0? 5.3 Bo. mo OMS—.028 on. we cotangflu 33% 2: cu PEEK

   

no.2 - 3.9
3nd.— - 5.2
8.3 - no.0
no.0 - 3..».
Sum - cod

DIIII

3.. 5......» ..........m.: 5 2.9.3 5.... :3

106

soofig. .525... a. .2. 55.8.... 3..... 2.. ... 5.53 52-..... ... 5.... 5.32.. .o 0.382... 2.. .o 5.5.5.... 3..... 2F a 2......—

107

   

w—AN . mod—
36— - «fig
Zhu— - we.»
5'.” - mfiv
Vué - cod .

UIIII

3.. 55.3 E35... 5 5.... 5......—

28.3... 2.25.2
2 .2. 52.3... .......o 2.. 2 5...... 23-..... 5 2...... 5.... 3...... $2.22. 822...... 2.. .o 8.3.5.... .5... 2F a. 2......

      

8.. - .2 I
on. . 8.. I
8.. I

a... - .m... I
on... - 8... I

.5503 Eon-dd E 2393 5...: .5..— 322.5%

108

.5822...
2......22 .. .2. 5.58.... 3..... 2.. 5 .28.. 53...: 2 5.... 5.22.. .o 0.252... .3255. 2.... 8.5.5.... 3..... 2F a 2......

     

a... - 2.... I

on. - 8.. I

8.. I

86-..... I

93.8.. I

22.5.5 E213. a. 5.5 5.89..— 3..—.2525

l09

58.32 .3222
E 8.3 .958...“ .338 05 5 2.95m Eofinm Eob noEoB 509:928 05 no 55 5.805 ES 2303 5.5 32 .«o whoa—:52 en 95!..—

e; I c; I

—! 1‘]

OD .. .2.

5...: 529... .e 22.5.2 2...; 5....— ..3 E 22.52

 

110

$8-32 52.3.2
E 8.5 b50045 3:93 05 5 383m E383 EB.“ 5:83 Eonéfiouom 05 mo 5.3 .5205 98 Ego? 5.5 26. mo £3834 mm 253

0-. I c-— .I.

c c

5.5— .589..— uo 9.2.552

 
 

Ill

.8828. 52222
S 35 2258.3 2523 2: E 35¢. 5.82 80¢ 5:53 502-2328 2: mo 523 2:82: 98 Emma? :23 >5. mo $3822 on Pinw—

f I f I
on

on.

5...: 2.82.— ... 22.2.2 2...»? 22.2 E: 2 23.5.2

 

112

scam-mmfl
£53222 5 83 2258.5 2338 05 E Ewfim 25 59a :oEok Eon-2928 05 mo 5.5 2:822 25 £903 :25 32 mo 23:52 pa 9...»:—

o; I
a

m.-. I
o , .
525 5280.5 we Eon—E22

2.293 5.2. 33 2 232:2

113

 

Khan-moo —
233222 5 22¢ Dance-E .838 05 E 5230 Bot =0an Eon-2222 2: mo 5.5 2:222 25 Emma? 5.22 32 .wo 232:2 an Esau-.—

o - _ I e - . I
o a A o . _._
52m 2:895 no 23222 2.20.5 2...:— tcd no 23222

 

H4

.hoomfiafl
“53:22 E 8.8 b.5973 .838 05 E cons—2 Eat 5:53 EBA—£28 05.3 5.5 E82; van 5325 5.5 >5. Mo 33:52 an 9:5:—

o-_l @4'
c5;

O .1 .V

5.5 5.32.. he E£E=z 2903 at:— 33 a. 22.332

 

llS

.88-?2 flaming—2 5
no... b.5975 .838 o... 5 812.2. fiaom >350 Eob 5:83 E09535. on. no 5.3 8.82..— E... EMF? 5.3 Bo. mo E09852 en 9...»:—

o - . II c - . I

o H u o _. I .

I — , .

5....— .Eeu... .9 23.5.: 2.393 at... :3 .c 22.53.

 

116

6.08-33
.aamEouz E 8.0 .9503... 3:98 05 E 0_m< 5.5m E9... :25? Eon...w_0.o.« 05 no 5.3 8.000.... 2... 5903 5.5 >5— uo 0.03852 3 9:53

o.-. I
o u . a
. o “r! .
at... 2.52.. .0 22.35.. 2.0.0..» at... :3 .0 23.52

6

UI

-—
.c

 

117

88-3... 50.5.2 =.
no... .0550... .8500 05 E 0.5000 33. .maofinom E8. 5:53 83-3.0.0. 0... .5 5.5 5.0.0... 9... 5303 5.5 Bo. mo 9.05.5.2 an 0.5»...—

o..- 3'
o... c-..

5.... 5.32.. .0 22......2 2.0.03 5.... E... .e 22......7.

 

118

800-80. 000.00....
a. 00... 5.500.... .8500 05 a. 0.3.. 50.00.". 50.... .5503 53-5.0.5 05 .5 5.5 5.0.0... 5... 5203 5.5 Bo. .5 0.05502 mm Bani

2-.. I
0.-. I
0.”.1.“

0. _

5...: 2.32.. .0 E00057.

 
 

119

83-30. 5550.2 5 00.0
55.00.... .5500 05 5 60m. 0523. 5.3. 5.500 50... :05...» E09055...“ 05 .5 5.5 5.0.0... .50 5303 5.5 325 0.35:2 vm 0.5..—

0. - . I

0 - . I. 0 - . E
0 . l. 0

5.5 5.0.0...— .5 0.05557. EH03 5.:— 39. .5 0.2.557.

120

 
 

REFERENCE

121

Acevedo-Garcia, D., Barbeau, E., Bishop, J.A., Pan, J. and, Emmons, K.M. 2004.
Undoing an epidemiological paradox: the tobacco industry’s targeting of US.
immigrants. American Journal of Public Health 94:2188-93.

Acevedo-Garcia, D., Soobader, M.J., and Berkman, LP. 2003. The differential effect
of foreign-bom status on low birth weight by race/ ethnicity and education.
Pediatrics 115: e20-30.

Akbar, M. Arab American Health in Michigan. Arab American BRFS conducted in
1994.

Alexander, GR. and Korenbrot, CC. 1995. The role of prenatal care in preventing
low birth weight. T he future of Children 5: 103-20.

Anselin, L. 1995. Local indicators of spatial association (LISA). Geographical
Analysis 27(2): 93-116.

Arab American Institute Foundation. Arab American State Profile: Michigan, 2003.

Auger, N., Luo, Z.C., Platt, R.W., and Daniel, M. 2008. Do mother’s education and
foreign born status interact to influence birth outcomes? Clarifying the
epidemiological paradox and the healthy migrant effect. Journal of
Epidemiological Community Health 62: 402-09.

Begman, I., Hirsch, R.P., Fria, T.J., el al. 1985. Cause of hearing loss in the high-risk
premature infant. Journal of Pediatrics 106: 95-101.

Births of low birthweight as a percent of all births by race/ ethnicity, 2006. Retrieved
October, 16 2009, from The Henry J. Kaiser Family Foundation website:
http://www.statehealthfacts.org/comparemaptable.jsp?ind=43&cat=2

Bratton, S.L., Shoultz, D.A., and Williams, MA. et al. 1996. Recurrence risk of low
birth weight deliveries among women with a prior very low birth weight delivery.
American Journal of Perinatal l3: 147-50.

Collins, J.W. Jr, and David, R.J. 1990. The differential effects of traditional risk
factors in infant birthweights among blacks and whites in Chicago. American
Journal of Public Health 80; 679-81.

122

Crump, C., Lipsky, S., and Mueller, B.A. 1999. Adverse birth outcomes among
Mexican-Americans: are U.S.-bom women at greater risk than Mexico-bom
women? Ethnic Health 4; 29-34.

Dejin-Karlsson, E. and Ostergren, PO. 2004. Country of origin, social support and the
risk of small for gestational age birth. Scand Journal of Public Health 32:442-9.

Dejmek, J., Solansky, 1., Podrazilova, K., and Sram, R.J. 2002. The exposure of
nonsmoking and smoking mothers to environmental tobacco smoke during
different gestational phase and fetal growth. Children 19 Health [10:601-06.

Dyck, I. 1990. Context, culture, and client: Geography and the health for all strategy.
Canadian Geographer 342338-41.

Dyck, I. and Keams, R. 1995. Transforming the relations of research: Toward
culturally safe geographies of health and well being. Health and Place 12137-47.

El Reda, D.K., Grigorescu, V., Posner, S.F., and Davis-Harrier, A. 2007. Lower rates
of preterm birth in women of Arab ancestry: An epidemiologic paradox-
Michigan, 1993-2002. Maternal Child Health Journal 11:622-27.

Elliott, S. 1995. Psychosocial stress, women and heart health: A critical review of the
evidence. Social Science and Medicine 40: 105-16.

Elo, LT, and Preston, SH 1996. Educational differentials in mortality: United States,
1979-1985. Social Science and Medicine 42: 47-57.

Ferraz, M.B., Gray, RH, and Fleming, P. et a1. 1988. Interpregnancy interval and low
birth weight: Finding from a case-control study. American Journal of
Epidemiology 1988: 111 1-6.

Fisher, E., Lo Gerfo, J., and Darling, J. 1995. Prenatal care and pregnancy outcomes
during the recession. American Journal of Public Health 75: 866-69.

Frank, R. 2005. International migration and infant health in Mexico. Journal of
Immigration Health 7:1 1-22.

Gavin, N.I., Adams, E.K., Hartmann K.E., Benedict, M.B., and Chireau, M. 2004.
Racial and ethnic disparities in the use of pregnancy-related health care among
Medicaid pregnant women. Maternal and Child Health Journal 8: 1 13-26.

123

Getis, A., 0rd, J.K. 1992. The analysis of the spatial association by use of distance
statistics. Geographical Analysis 23(3): 189-207.

Glewwe, P. 1997. How does schooling of mothers improve children health? Evidence
fiom Morocco. Washington, DC: World Bank

Goldon, M. 1964 Assimilation in American Life. New York: Oxford University Press.

Gould, J.B., Madan, A., Qin, C., and Chavez, G 2003. Perinatal outcomes in two
dissimilar immigrant populations in the United States: a dual epidemiologic
paradox. Pediatrics 11; 676-82.

Hoagberg, S., Rode, P., and Skovolt, C. et a1. 1990. Barriers and motivators to
prenatal care among low-income women. Social Science and Medicine
30:487-95.

Holt, V.L., Danoff, N.L., and Mueller, B.A. et al. 1997. The association of change in
maternal marital status between births and adverse pregnancy outcomes in the
second birth. Paediatric and Perinatal Epidemiology 11:31-40.

House, J., Kessler, R., and Herzog, R. et a1. 1990. Age, socioeconomic status, and
health. Milbank Quarterly 68: 383-411.

Huynh, M., Parker, J.D., and Harper, S. et al. 2005. Contextual effect of income
inequality on birth outcomes. International Journal of Epidemiology 22:467-77.

Keams, RA. 1993 Place and health: Towards a reformed medical geography. The
Professional Geographer 45: 139-47.

Kliegman, R.M., Madura, D., and Kiwi, R. et al. 1994. Relation of maternal cocaine
use to the risks of prematurity and low birth weight. Journal of Pediatrics 124:
751-6.

Kramer, MS. 1990. Birth weight and infant mortality: Perceptions and pitfalls.
Pediatric and Perinatal Epidemiology 4:381-89.

Leung, T.N., Roach, V.J., and Lau, T.K. 1998. Incidence of preterm delivery in Hong

124

Kong Chinese. Australian and New Zealand Journal of Obstetrics and
Gynaecology 38: 138-41 .

Low birthweight Babies (percentage) — Michigan, 1995 — 2007. Retrieved April, 23
2010, from The Annie E. Cassie Foundation website: http://
http://datacenter.kidscount.org/datafbystate/Rankings.aspx?state=MI&loct=5&by
=a&order=a&ind=1 65 8&dtm=3 523&tf=6

Madan, A., Palaniappan, L., Urizar, G, Wang, Y., Fortmann, SP, and Gould, J.B.
2006. Sociocultural factors that affect pregnancy outcomes in two dissimilar
immigrant groups in the United States. The Journal of Pediatrics 148:341-46.

Markides, KS. and Coreil, J. 1986. The health of Hispanics in the southwestern
United States: an epidemiological paradox. Public Health Reports 101:253-65.

Markides, K.S., and Coreil, J.C. 1986. The health of Hispanics in the Southwestern
United States: An epidemiologic paradox. Public Health Reports 101:253-265.

McFarlane, J., Parker, B., and Soeken, K. 1996. Physical abuse, smoking, and
substance use during pregnancy: Prevalence, interrelationships, and effects on
birth weight. Journal of Obstetric, Gynecologic and Neonatal Nursing 25:313-20.

McLafferty, S., and Tempalski, B. 1995. Restructuring and women’s reproductive
health: Implications for low birthweight in New York City. Geofiom 26: 309-23.

Nothnagle, M., Marchi, K., Egerter, S., and Braveman, P. 2000. Risk factors for late
or no prenatal care following Medicaid expansions in California. Maternal and
Child Health Journal 4: 251-59.

O’regan, K., and Wiseman, M. 1990. Using birthweights to chart the spatial
distribution of urban poverty. Urban Geography 11:217-33.

Palloni, A., and Morenoff, JD. 2001. Interpreting the paradoxical in the Hispanic
paradox: demographic and epidemiologic approaches. Annals of the New York
Academy of Science 954: 140—74.

Paneth, NS. 1995. The problems of low birth weight. The Future of Children 5:
19-34.

Peak, C. and Weeks, JR. 2002. Does community context influence reproductive

125

outcomes of Mexican origin women in San Diego, California? Journal of
Immigrant Health 4: 125—36.

Pritchard, C., and Teo Mpfan, M. 1993. Preterm birth, low birthweight and the
stressful of the household role for pregnant women. Social Science and Medicain
38:89-96.

Rauh, V.A., Andrews, HE, and Garfinkel, RS. 2001. The contribution of maternal
age to racial disparities in birthweight: A multilevel perspective. American
Journal of Public Health 91: 1815-24.

Reagan, PB, and Salsberry, RJ. 2005. Race and ethnic difference in determinants of
preterm birth in the USA: Broadening the social context. Social Science and
Medicine 60:2217-28.

Reichman, NE. and Pagnini, BL. 1997. Maternal age and birth outcomes: Data from
New Jersey Family. Planning Perspectives 29: 268-72.

Roth, J., Hendrickson, J., and Schilling, M. et al. 1998. The risk of teen mothers
having low birth weight babies: Implication of recent medical research for school
health personnel. The Journal of Scholl Health 68: 271-5.

Rush, D., and Cassano, P. 1983. Relationship of cigarette smoking and social class to
birth weight and perinatal mortality among all births in Britain. Journal of
Epidemiology and Community Health 37: 249-55.

Shiono, P.H., Rauh, V.A., Park, M., Lederrnan, S.A., and Zuslar, D. 1997. Ethnic
differences in birthweight: the role of lifestyle and other factors. American
Journal of Public Health 87: 787-93.

Sidel, R. 1992. Women and children first: Towards a US. family policy. American
Journal of Public Health 82:825-29.

Silva, A.A., Barbieri, M.A., and Gomes, U.A. et a1. 1998. Trends in low birth weight
and association maternal factors in an urban area. Indian Pediatrics 76: 73-84.

Singh, GK., and Miller, B. 2001. Health, life expectancy, and mortality patterns
among immigrant populations in the United States. American Journal of Public
Health 91:392-99.

126

Singh, GK., and Siahpush, M. 2002. Ethnic immigrant differentials in health
behaviors, morbidity, and cause specific mortality in the United States: An
analysis of two national database. Human Biology 74; 83-109.

Skilnik, R. 2008. Essentials of Global Health. Sudbury: Jones and Barlett Publishers
Press

Smeriglio, V.L., and Wilcox, HQ 1999. Prenatal drug exposure and child outcome.
Past, present, and future. Child Perinatol 26: 1-16.

Starfield, B., Shapiro, S., and Weiss, J. et a1. 1991. Race, family income, and low birth
weight. American Journal of Epidemiology 134: 1167-74.

Tore, E., Kunst, A.E., Bos, V., Deerenberg, M.1., Joung, I.M.A., and Mackenbach, JP.
2006. The effect of age at immigration and generational status of the mother on

infant mortality in ethnic minority populations in The Netherlands. European
Journal of Public Health 17: 134-38.

Uretsky, MO, and Mathiesen, SO 2007. The effects of years lived in the United
States in the general health status of Califomia’s foreign-bom populations.
Journal of Immigrant Health 9: 125-36.

Urquia, M.L., Frank, J.M., Glazier, R.H., Moineddin, R., Matheson, EL, and Gagnon,
A. 2009. Neighborhood context and infant birthweight among recent immigrant
mothers: A multilevel analysis. American Journal of Public Health 99: 285-93.

Vahratian, A., Buekens, P., Delvaux, T., Boutsen, M., Wang, Y., and Kupper, LL.
2004. Birthweight differences among infants of North African immigrants and
Belgians in Belgium. European Journal of Public Health 14: 381-83.

Valero de Bembé, J., Soriano, T., Albaladejo, R., Juarranz, M., Calle, M.B., Martinez,
D., and Dominguez-Rojas, V. 2004. Risk factors for low birth weight: A review.

European Journal of Obstetrics and Gynecology and Reproductive Biology 116:
3-15.

Wilborg, K., Henriksen, TB, and Hedegaard, M. et a1. 1996. Smoking during
pregnancy and preterm birth. BJOG: An International Journal of Obstetrics and
Gynaecology 103: 800-5.

World Health Organization, Expert Committee on Maternal and Child Health. Public

127

health aspect of low birth weight. WHO Technical Report Series, No. 27. Geneva,
Switzerland: WHO, 1950.

Zeitlin, J .A., Ancel, P.Y., and Saurel-Cubizolles, M.J. et al. 2001. Are risk factors the
same for small for gestational age versus other preterm births? American Journal
of Obstetrics and Gynecology 185:208-15.

128

N

”'llllllll11111111fllilflllflllllljlllllf