PREPREGNANCY WEIGHT STATUS VALIDATION AND DIET QUALITY ASSOCIATED
WITH GESTATIONAL WEIGHT GAIN
By
Dayeon Shin

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Human Nutrition
2012

ABSTRACT
PREPREGNANCY WEIGHT STATUS VALIDATION AND DIET QUALITY ASSOCIATED
WITH GESTATIONAL WEIGHT GAIN
By
Dayeon Shin

The Institute of Medicine (IOM) of the National Academy of Sciences has established
both gestational weight gain guidelines (2009), and Dietary Reference Intakes (DRIs) for
pregnancy (2002/2005). The gestational weight gain guidelines are based on the prepregnancy
weight status and the DRIs are for each trimester. Both references, however, aim to maximize the
optimal health outcomes for both the mother and her baby. Limited data are available on the
validation of self-reported prepregnancy anthropometrics and coherence between the two
national guidelines for pregnancy. Our goal was to validate pregnant women’s self-reported
prepregnancy weight status and identify dietary determinants for achieving gestational weight
gain guidelines. The validation of self-reported prepregnancy weight status of pregnant women
was determined by drawing an inference from linear regression between self-reported and
measured weight status of age-comparable non-pregnant women. Dietary quality during
pregnancy was determined by nutrient density, nutrient adequacy, and Healthy Eating Index
(HEI) 2005 and then with the gestational weight gain, in the U.S. representative population. We
concluded that prepregnancy weight status classified by self-reported height and weight is valid
for group data, and adequate intake of selected nutrients, dark green and orange vegetables and
legumes. The adequate consumption of total grains, especially whole grains, was associated with
healthy gestational weight gain.

Dedicated to my mother for her unconditional support, encouragement and love.

iii

ACKNOWLEDGEMENTS
The thesis would not have been possible without the support and encouragement that I
have received from the guidance of my committee members, help from friends and support from
my family. I want to thank my Guidance Committee members for their valuable advice and
generosity throughout my studies: Drs. Lorraine Weatherspoon, Qing Lu and Won O. Song. I
would like to express my utmost respect and gratitude to my major advisor, Dr. Won O. Song for
her understanding, patience and brilliant insights she has shared. Her guidance, support and
encouragement have been essential to my personal and professional growth. I warmly thank to
Dr. Hwan Chung for his guidance in statistical advice. I am deeply grateful to Dr. Leonard
Bianchi, helped me develop my background in statistics and research. Lastly, and most
importantly, I would also like to thank my parents and brother for their loving support and
continual faith in me.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ........................................................................................................................ x
ABBREVIATIONS AND WORKING DEFINITIONS .............................................................. xii
INTRODUCTION .......................................................................................................................... 1
CHAPTER ONE: REVIEW OF LITERATURE
A. Importance of Nutrition before and during Pregnancy. ..................................................... 6
B. Gestational Weight Gain Guidelines ................................................................................ 18
C. Factors associated with Gestational weight gain.............................................................. 21
1. Diet Quality.............................................................................................................. 23
2. Physical Activity ...................................................................................................... 27
3. Pregnancy Outcomes ............................................................................................... 32
3.1 Infants ............................................................................................................ 32
3.2 Postpartum Weight......................................................................................... 33
D. Validity of Self-reported vs. Measured Weight and Height ............................................. 35
1. General Populations .................................................................................................. 35
2. Reproductive Aged Women ...................................................................................... 38
CHAPTER TWO: VALIDATION OF PREPREGNANCY WEIGHT STATUS ESTIMATED
FROM SELF-REPORTED DATA
A. Abstract ............................................................................................................................ 41
B. Introduction ...................................................................................................................... 42
C. Subjects and Methods....................................................................................................... 43
D. Results .............................................................................................................................. 47
E. Discussion......................................................................................................................... 50
CHAPTER THREE: HOW IS DIET QUALITY DURING PREGNANCY ASSOCIATED
WITH GESTATIONAL WEIGHT GAIN?
A. Abstract ............................................................................................................................ 60
B. Introduction ...................................................................................................................... 61
C. Subjects and Methods....................................................................................................... 62
D. Results .............................................................................................................................. 69
E. Discussion......................................................................................................................... 71
CHAPTER FOUR: CONCLUSION
A. Implications ...................................................................................................................... 85
B. Recommendations for future research.............................................................................. 87

v

APPENDICES .............................................................................................................................. 89
BIBLIOGRAPHY ....................................................................................................................... 125

vi

LIST OF TABLES

Table 1.1 Correlations and mean differences between self-reported and measured height by
sociodemographics of pregnant and non-pregnant women .......................................................... 54
Table 1.2 Correlations and mean differences between self-reported weights in non-pregnant
women by sociodemographics ...................................................................................................... 56
Table 1.3 Correlations and mean differences between self-reported and multiple imputed (MI)
weight of pregnant women by sociodemographics ....................................................................... 57
Table 1.4 Correlations and mean difference between self-reported and measured weight by
sociodemographics in pregnant women in 1st trimester ................................................................ 58
Table 1.5 Agreement on self-reported and measured BMI categories in non-pregnant women and
self-reported and multiple imputed BMI categories in pregnant women ..................................... 59
Table 2.1 Demographic distribution of pregnant women ............................................................. 74
Table 2.2 Distribution of trimester of pregnancy and gestational weight gain groups by
prepregnancy weight status ........................................................................................................... 76
Table 2.3 Prepregnancy BMI and Nutrient density per 1,000 kcal by gestational weight gain
groups ............................................................................................................................................ 77
Table 2.4 Nutritional adequacy by gestational weight gain groups .............................................. 79
Table 2.5 Healthy Eating Index 2005 score by gestational weight gain groups ........................... 82
Table 2.6 Logistic regression models for Healthy Eating Index 2005 components to inadequate
or excessive gestational weight gain in comparison with adequate gestational weight gain........ 83
Table 2.7 Logistic regression models for HEI 2005 components to inadequate or excessive
gestational weight gain in comparison with adequate gestational weight gain ............................ 84
Table A1. Correlations and mean difference between self-reported weight 1 year ago and selfreported weight by maternal age, race/ethnicity, education, country of birth, number of previous
births, and family poverty income ratio for non-pregnant women ............................................... 90
Table A2. Correlations and mean difference between self-reported and measured BMI by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
poverty income ratio for non-pregnant women ............................................................................ 92
Table A3. Pearson r, Mean, and mean difference for self-reported weight 1 year ago and
prepregnancy weight ..................................................................................................................... 94

vii

Table A4. Correlations and mean difference between self-reported and measured height, weight,
and BMI by socio-demographic characteristics ............................................................................ 95
Table A5. Sensitivity, specificity, and predictive value positive (PVP) and predictive value
negative (PVN) of self-reported BMI with measured BMI .......................................................... 97
Table A6. Distribution of pregnant women by each trimester...................................................... 98
Table A7. Correlations and mean difference between self-reported and measured height by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
poverty income ratio for pregnant women in 1st trimester ............................................................ 99
Table A8. Correlations and mean difference between self-reported and measured weight by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
poverty income ratio in pregnant women in 1st trimester ........................................................... 101
Table A9. Correlations and mean difference between self-reported and measured BMI by sociodemographic characteristics in pregnant women in 1st trimester ................................................ 103
Table A10. Correlations and mean difference between self-reported and measured height, weight,
and BMI by socio-demographic characteristics in pregnant women in 1st trimester.................. 105
Table A11. Sensitivity, specificity, and predictive value positive (PVP) and predictive value
negative (PVN) of self-reported BMI with multiple imputed BMI ............................................ 107
Table A12. Correlations and mean difference between self-reported and multiple imputation (MI)
BMI by socio-demographic characteristics of pregnant women ................................................ 108
Table A13. Mean (SEM) self-reported weight 1 year ago and weight by pregnancy status and
intentional weight lose ................................................................................................................ 110
Table A14. Mean (SEM) self-reported weight 1 year ago, weight, and measured weight by
pregnancy status and intentional weight lose.............................................................................. 111
Table A15. Comparison of mean (SEM) random value by multiple imputation (MI) methods and
actual measured BMI for randomly selected 10 pregnant women in 1st trimester ..................... 112
Table A16. Calculated pregnant women’s measured BMI by multiple imputations drawn from
non-pregnant women’s self-reported and measured BMI data ................................................... 113
Table A17. Sociodemographic characteristics of outliers above 95% confidence intervals of
height difference between self-reported height and measured height in pregnant and nonpregnant women .......................................................................................................................... 114
Table A18. Response bias of odds of under-reporting vs. accurate-reporting and over-reporting
vs. accurate group in height in pregnant women (n=279) .......................................................... 116

viii

Table A19. Response bias of odds of under-reporting vs. accurate-reporting and over-reporting
vs. accurate group in height in non-pregnant women (n=489) ................................................... 118
Table A20. Response bias of odds of under-reporting v. accurate-reporting and over-reporting vs.
accurate group in weight in non-pregnant women (n=489) ........................................................ 120
Table B1. Healthy Eating Index 2005 score by gestational weight gain groups ........................ 124
Table B2. Distribution of HEI 2005 by gestational weight gain groups .................................... 124

ix

LIST OF FIGURES

Figure 1. Overall Framework of the Research ................................................................................ 5

x

ABBREVIATIONS AND WORKING DEFINITIONS

Gestational Weight Gain. Difference between weights measured at any trimester of pregnancy
and self-reported prepregnancy weight.
Gestational Weight Gain Status. Categorized into three groups: inadequate; adequate; and
excessive.. The adequate total gestational weight gain group consisted of those pregnant women
within recommended weight gain according to prepregnancy BMI category and those less than
minimum of recommended weight gain were categorized as having inadequate gestational
weight gain . Those above the maximum recommended weight gain were categorized as having
excessive gestational weight gain when compared with the recommended gestational weight gain
by the Institute of Medicine (2009).
Healthy Eating Index Score System, 2005 (HEI, 2005). A dietary evaluation tool to quantify
the quality of diet consumed by the U.S. population. It comprises 12 components: All five major
food groups included in MyPyramid, i.e.,1) Total Fruit; 2) Total Vegetables; 3) Total Grains; 4)
Milk, which includes soy beverages; and 5) Meat and Beans, which includes meat, poultry, fish,
eggs, soybean products other than beverages, nuts, seed and legumes; additional components of
6) Whole Fruit (i.e., forms other than juice); 7) Dark Green and Orange Vegetables and 8)
Legumes; 9) Whole Grains (which mus.t include the entire grain kernel, bran, germ, and
endosperm); 10) Oils (non-hydrogenated vegetable oils and oils in fish, nuts, and seeds); 11)
Saturated Fat; 12) Sodium and Calories from Solid Fat, Alcohol, and Added Sugar (SoFAAS).
Whole Fruit was added because the 2005 Dietary Guidelines suggesting to limit juice to less than
half of total fruit intake. The Dark Green and Orange Vegetables and Legumes component was
added because those are the three subgroups of vegetables for which current intake is furthest
from recommended levels. The Whole Grains component was added because the 2005 Dietary
Guidelines specify that at least half of grain intake should be whole grains. New components
were added for Oils to reflect the recommendations for oil found in MyPyramid and for Calories
from SoFAAS.
Small-for-Gestational-Age (SGA). Defined as birth weight below the 10th percentile for
gestational age, based on a given reference population. Smaller in size than is normal for the
embryo/fetus’ gender and gestational age; occurs when an embryo/fetus undergoes intrauterine
growth restriction (IOM, 2009).
Large-for-Gestational-Age (LGA). Defined as birth weight above 90th percentile for gestational
age, based on a given reference population. Maternal diabetes is the most common cause of LGA
babies. Excessive weight gain can translate into increased fetal weight.
Low Birth Weight (LBW). Defined as infant status of weighing less than 2,500 grams.

xi

Macrosomia. Defined as infant status of weighing more than 4,000 grams.
Postpartum Weight Retention (PPWR). A woman’s weight that remains after 3 months to 3
years post-delivery minus the woman’s prepregnancy weight (Viswanathan et al., 2008).
Institute of Medicine (IOM). The Institute of Medicine (IOM) is an independent, nonprofit
organization that works outside of government to provide unbiased and authoritative evidencebased health advice to decision makers and the public. Established 1970, the IOM is the health
arm of the National Academy of Sciences (NAS), which was chartered under President Abraham
Lincoln in 1863. Nearly 150 years later, the NAS has expanded into what is collectively known
as the National Academies, which comprises the NAS, the National Academy of Engineering,
the National Research Council, and the IOM. IOM’s aim is to help those in government and the
private sector make informed health decisions by providing evidence upon which they can rely.
Each year, more than 2,000 individuals, members, and nonmembers of the IOM volunteer their
time, knowledge, and expertise to advance the nation’s health.
Self-reported BMI. BMI calculated from self-reported height and weight.
Measured BMI. BMI calculated from measured height and weight.
Prepregnancy Weight Status. Prepregnancy BMI categorized the participants as underweight
2

2

(BMI <18.5 kg/m ), normal weight (18.5≤BMI<24.9 kg/m ), overweight (25.0≤BMI<29.9
2

2

kg/m ), or obese (BMI≥30.0 kg/m ).
Diet Quality. The 2005 Dietary Guidelines for Americans emphasize the concept of nutrient
density. Nutrient dense foods provide substantial amounts of vitamins and minerals and fewer
calories. Such foods which fit the description are most often are selected in the low-energydensity diets. This approach is also consistent with the healthy eating plan, based on the Dietary
Guidelines for Americans.
The Dietary Guidelines for Americans healthy eating plan:
 Emphasizes fruits, vegetables, whole grains, and fat-free or low-fat milk and milk
products.
 Includes lean meats, poultry, fish, beans, eggs, and nuts.
 Is low in saturated fats, trans fats, cholesterol, salt (sodium), and added sugars.
Diet quality was measured by nutrient density per 1,000 kcal, and whether nutrient adequacy
meets the RDA and HEI 2005 recommendations.

xii

INTRODUCTION

There has been a great interest in public health issues associated with poor birth outcomes,
LBW, macrosomia, SGA, or LGA in relation to maternal weight gain and quality of diet
consumed before and during pregnancy. Healthy People 2020 underscores the importance of the
well-being of mothers, infants, and children, which is a major public health goal for the U.S. The
IOM of the National Academy of Sciences has established GWG guidelines (2009), and the
Dietary Reference Intakes (DRIs) for pregnancy to prevent adverse pregnancy complications and
birth outcomes. The gestational weight gain guidelines during pregnancy specify total weight
gain during the entire pregnancy and the rates of weight gain during second and third trimesters
of pregnancy for respective prepregnancy weight status (IOM, 2009). The dietary reference
intakes (DRIs) are quantitative estimates of recommended nutrient intakes to be used for
planning and assessing diets for apparently healthy people. The planning goal for individuals is
to achieve recommended and adequate nutrient intake using food-based guides such as the
Dietary Guidelines for Americans (National Academy of Science, 2003).
Determining whether gestational weight gain guidelines are followed is evaluated by selfreported prepregnancy weight status. In both clinical and public health settings, pregnant
women’s weight gain and diet quality are assessed utilizing self-reported data prepregnancy
weight and dietary intake. Valid categorization of prepregnancy weight status is thus a
prerequisite to adequately applying the IOM’s gestational weight gain guidelines. Validity of
self-reported height and weight, and BMI calculated from the self-reported data in pregnant
women and non-pregnant women, have been the focus of several investigations (Hedderson et al.
2012; Park et al. 2009; Craig & Adams, 2009; Cedergren, 2006; Olson et al. 2003). In nonpregnant women, self-reported data are compared with measured data for validation. The major
1

challenges in these research investigations with pregnant women is the inherent inability of
measuring prepregnancy weight during pregnancy, due to the unavailability of weight data right
before conception and the high prevalence of unplanned pregnancies reported in the U.S. (Finer
et al., 2011). The lack of gold standards for measuring prepregnancy weight that could be used in
validating the self-reported data has been an issue. The gold standard references used in the
previous studies have been varied and somewhat problematic. In validating self-reported
prepregnancy weight status, measured weight in the first trimester (Park et al., 2009; Olson et al.,
2003), measured weight during pregnancy (Craig & Adams, 2009), maternal weight recorded at
the first visit (Cedergren et al., 2006) and weight measured within the previous 12 months of
pregnancy (Hedderson et al., 2010) have been used as references in the previous studies.
Both adequate gestational weight gain and good nutrition are important modifiable
determinants during pregnancy. Maternal nutrition is a key factor which influences gestational
weight gain (Deierlein et al., 2008; Chapin et al., 2004; Olafsdottir et al., 2006; Tssiga et al.,
2010), adverse pregnancy and complicated birth outcomes, and hence, the health of infants and
women during pregnancy and postpartum (Anderson, 2001),
Previous research has reported inadequate dietary intake of folate and iron during the first
trimester of pregnancy (Rifas-Shiman et al., 2009) and inadequate intake of calcium, vitamin D,
iron, and folate during the second and third trimesters (Pick et al., 2005). Inadequate intake of
vitamin D, folate, calcium, iron and phosphorous throughout the entire pregnancy (Fowles, 2002)
in the U.S., was also reported. None of these studies incorporated vitamin and mineral
supplements in the assessment of nutritional quality. Due to the different socioeconomic statuses
and trimesters of pregnancy of the participants, the results had inconsistencies. Limited reports
are available on the overall diet quality of pregnant women (Pick et al., 2005; Tsigga et al.,

2

2010). Pick et al. (2005) reported that the Healthy Eating Index (HEI) 2005 provided a useful
tool for comparing diets with the Food Guide Pyramid, but the HEI did not account for the need
of micronutrients (Pick et al., 2005). Tsigga et al. (2010) assessed HEI 2005 by prepregnancy
weight status in Greek pregnant women. No studies assessed diet quality as a determinant of
gestational weight gain.
We found that there are a limited number of studies investigating the validation of
prepregnancy weight status. An important and practical means to help individual pregnant
woman to achieve the suggested gestational weight gain guidelines is to establish the degree of
validity of prepregnancy weight status calculated from self-reported height and weight and then
monitor the rates of weight gain throughout the different stages of pregnancy. Evaluation of the
reliability and validity of self-reported height and weight is important because these data are used
widely for population-based research, surveillance, and programmatic decision making. We also
find no studies that investigated diet quality by using HEI 2005 or the indices as a determinant of
optimal gestational weight gain.
The present study aimed to address the validity of prepregnancy weight status classified
based on the height and weight self-reported during pregnancy and to determine how dietary
intake during pregnancy is associated with gestational weight gain in a cross-sectional study with
a nationally representative population in the U.S.
Specific Aims
Aim 1: To assess the validity of self-reported prepregnancy weight status in U.S. representative
pregnant women

3

Hypotheses
H01) Self-reported prepregnancy weight status is valid as determined by Kappa statistics
(>0.8) and Pearson’s correlations (p<0.05).
H02) The validity of self-reported prepregnancy weight status varies by
sociodemographic subgroups.
Aim 2: To determine if diet quality during pregnancy is associated with gestational weight gain.
Hypothesis
H03) Adequate gestational weight gain during different trimesters, is predicted by diet
quality indices (nutrient density, nutrient adequacy, and HEI 2005 scores), when
prepregnancy weight status and physical activity are controlled.

4

Study 2

GWG

Maternal
Determinants
Age

Underweight

Race/Ethnicity

Normal Weight

Country of Birth
Education Level
Family PIR
Previous
Pregnancies #

H2
0

Overweight
Obese

H03

Diet
Quality
Healthy Eating
Index 2005
Nutrient
Adequacy
Nutrient Intake

Study 1
VALIDATION

Figure 1. Overall Framework of the Research

1

H01

1

For interpretation of the references to color in this and all other figures, the reader is referred to
the electronic version of this thesis.
5

Chapter One

REVIEW OF LITERATURE
A. Importance of Nutrition before and during Pregnancy.
Nutrition during pregnancy plays a key role in the well-being of the mother and the
newborn infant. It also has an impact on the overall health of the child, throughout their
childhood, until they reach adulthood (IOM, 2009). Diet in the first trimester of pregnancy is
important to the development of the fetus as differentiation of various organs including
rudimentary kidneys, liver, circulatory system, eyes, ears, mouth, hands, arms and
gastrointestinal tract occurs in the first trimester. Diet in the 2

nd

rd

and 3 trimesters in pregnancy

is important for overall fetal growth and brain development (Rifas-Shiman et al., 2006).
Nutrition before pregnancy or the periconceptional period, is a key component of healthy
2

pregnancy outcomes ( Widen& Siega-Riz, 2010). By achieving a normal BMI (18.5-24.9 kg/m )
prior to pregnancy and improving nutritional status prior to and during pregnancy, pregnant
women can lower their risk of pregnancy complications including gestational diabetes and
preeclampsia and decrease the risk of poor birth outcomes, including birth defects, intrauterine
growth restriction, and later chronic diseases.
In recent years, there has been an increasing focus on optimizing women’s nutrition and
lifestyle before pregnancy, as this is a key time for fetal development (Chapin et al., 2004).
Women, however, are often unaware that they are pregnant for the first few weeks of embryonic
life and miss any of the subsequent changes during the periconceptional period. Inskip et al.
(2009) examined the extent to which women planning a pregnancy comply with

6

recommendations for nutrition and lifestyle including folic acid supplements and alcohol
consumption in Southampton, United Kingdom. In total, 12,445 non-pregnant women aged 2034 were recruited for the Southampton Women’s Survey, 238 of those became pregnant within
three months of being interviewed. Women planning a pregnancy in the United Kingdom were
encouraged to take 400 µg/day of folic acid a day in supplements and to either completely avoid
alcohol, or at most consume 8-16 g of alcohol. Among those who became pregnant, 2.9% were
taking 400 µg/day or more folic acid supplements three months before pregnancy and drinking
≤32 g/week of alcohol, compared with 0.66% of those who did not become pregnant. Fowles
(2002) compared the differences between low- and middle-income pregnant women’s general
nutritional knowledge, usual dietary intake and weight gain with a convenient-sized sample of
women (n=109). Only 10% of the women could identify the correct number of servings for the
Food Pyramid categories. Pregnant women’s actual intake of folate (257 µg vs. 400 µg), calcium
(836 mg vs. 1,200 mg) and iron (17 mg vs. 30mg) did not meet nutritional needs of pregnancy.
Few pregnant women succeeded in complying with the recommendations on dietary intakes
during pregnancy and lacked the necessary nutritional knowledge for a healthy pregnancy.
Development of effective strategies designed to promote adequate nutritional intake by pregnant
women is needed.
Maternal nutrition during pregnancy may influence pregnancy complications (Oken et al.,
2007) and childhood outcomes (Tobias et al., 2005). Oken et al. (2007) examined the
associations of first-trimester intake of calcium, n-3 and n-6 fatty acids, trans-fatty acids,
magnesium, folate, and vitamins C, D, and E from both foods and supplements with
preeclampsia and gestational hypertension in 1,718 women in the prospective cohort study
Project Viva. A total of 59 women developed preeclampsia, and 119 developed gestational

7

hypertension. They found that a lower risk of preeclampsia was associated with higher intake of
the elongated n-3 fatty acids docosahexaenoic and eicosapentaenoic acids (OR=0.84, [95%
CI=0.75-1.09] per serving/day, and the ratio of docosahexanoic + eicosapentaenoic to
arachadonic acid (OR=0.82, [95% CI=0.66-1.01]. They found no association between other
nutrients and the prevalence of preeclampsia and gestational hypertension. Tobias et al. (2005)
examined the association between maternal diet at 32 weeks gestation and the bone mass of
4,451 children, at 9 years, which were measured by using total body Dual Energy X-Ray
Absorptiometry (DXA) in the Avon Longitudinal Study of Parents and Children. Maternal
magnesium intake was found to have a significant impact on a child’s total body bone mineral
2

content (BMC) (β=4.9, 7.4-23.1; g) and bone mineral density (BMD) (β=4.9, 2.5-7.3; g/cm x
3

10 ). Maternal intake of potassium was found to be significantly related to a child’s spinal BMC
2

3

(β=1.8, 0.8-2.9; g) and BMD (β=10.5, 4.9-16.0; g/cm x 10 ). A significant association was also
observed between maternal dietary folate intake and spinal BMC (β=0.55, 0.16-0.94; g). The
finding that maternal diet is significantly related to DXA measures of a child at age nine years of
age is consistent with the hypothesis that the trajectory of bone development in childhood is
programmed by early life factors.
Energy: Energy balance is a chief determinant of gestational weight gain. The strength of
the relation between energy intake and gestational weight gain is confounded by a number of
intervening factors: changes in basal metabolism, levels of physical activity, and the
composition of accumulated maternal and fetal tissue (Kaiser et al., 2008). The DRIs for
pregnant women specifies that pregnant women increase their daily energy intake by an
additional 340 kcal in the second trimester and 452 kcal in the third trimester to meet estimated

8

energy requirements (EER) (IOM, 2005). Basal metabolism increases during pregnancy due to
the metabolic contribution of the uterus and fetus and increased work of the heart and lungs. The
increase in basal metabolism is one of the major components of the increased energy
requirements during pregnancy (Hytten & Chamberlain, 1991). Variation in energy expenditure
between individuals is largely due to differences in fat free mass (FFM), which in pregnancy is
comprised of low energy-requiring expanded blood volume, high energy-requiring fetal and
uterine tissues, and moderate energy-requiring fetal and uterine tissues and skeletal muscle mass
(Hytten & Chamberlain, 1991). In late pregnancy, approximately one-half of the increment in
energy expenditure can be attributed to the fetus (Hytten & Chamberlain, 1991). The fetus uses
about 8 ml O2/kg body weight/min or 56 kcal/kg body weight/d; for a 3-kg fetus, this would be
equivalent to 168 kcal/d (Sparks et al., 1980). Fat mass, a low energy-requiring tissue,
contributes to the variation in energy expenditure, but to a much lesser extent than FFM, which
has been found to be the strongest predictor of basal energy expenditure (BEE) (Butte et al.,
1999).
Cumulative changes in BEE throughout pregnancy ranged from 29,636 to 50,300 kcal or
106 to 180 kcal/d. Mean increments in BEE over prepregnancy values were 48, 96, 263 kcal/d,
or 4, 7, and 19 percent in the first, second, and third trimesters in healthy women with positive
pregnancy outcomes (Prentice et al., 1996). The median change in total energy expenditure
(TEE) was 8 kcal per week of gestation with a range of -58 to 107 kcal/week (Prentice et al.,
1996).
The EER for energy during pregnancy is derived from the sum of the TEE of the woman
in the non-pregnant state plus a median change in TEE of 8 kcal/week plus the energy deposition
during pregnancy of 180 kcal/d. Since TEE changes little and weight gain is minor during the

9

first trimester, no increase in energy intake during the first trimester is recommended. The
following formula was used to calculate additional energy needed throughout pregnancy. (IOM,
2002/2005)
For pregnant women aged 19-50 years,
EER pregnant = EER non-pregnant + additional energy expended during pregnancy + energy
deposition
st

1 trimester = adult EER + 0 + 0
nd

2

trimester = adult EER + 160 kcal (8 kcal/week × 20 week) + 180 kcal

rd

3 trimester = adult EER + 272 kcal (8 kcal/week × 34 week) + 180 kcal
The additional energy required was based on trimester of pregnancy, not by prepregnancy
weight status. No information was available for the percentage contribution of protein,
carbohydrate, and fat to sum up total caloric intake during pregnancy. Further, specific food
guides to achieve additional energy kcal during each trimester of pregnancy, in regards to
prepregnancy weight status, should be fully investigated and developed.
Protein. The average requirement for the protein needed by pregnant women is 71g/day
based on calculations of the amount needed for initial deposition of pregnancy-related tissue and
the amount needed to maintain new tissue (IOM, 2005).
Olsen et al. (2007) examined whether milk consumption during pregnancy was
associated with infant size at birth in the Danish National Birth Cohort during 1996-2002. The
study concept was developed by the hypothesis that cow milk is an effective vehicle for the
delivery of many nutrients essential for fetal development and therefore of potential importance
for fetal growth. Birth weight was related to intake of protein, but not to fat, derived from milk.
10

Milk consumption was inversely associated with the risk of SGA birth and directly with both
LGA birth and mean birth weight. Women who consumed more than 6 glasses of milk daily had
a 49% lower risk of SGA and 108g increase in birth weight compared with those consuming no
milk.
Essential fatty acids. Lucas et al. (2004) provided evidence that higher intake of longchain omega-3 fatty acids during pregnancy results in an increased duration of gestation with
improvement in fetal growth. Fatty acids and contaminant (polychlorinated biphenyls and
mercury) concentrations were measured in the cord plasma of Nunavik newborns (n = 454) and
compared with those of a group of newborns (n = 29) from southern Qué
bec. Gestational age and
birth weight were higher by 5.4 day [95% CI=0.7–10.1] and 77 g [95% CI=−64- 217) for babies
born to mothers in the third tertile of n-3 highly unsaturated fatty acid consumption, as compared
with those in the first tertile. In the seafood eating population of Nunavik, an increase in the
proportion of n-3 highly unsaturated fatty acids, measured in umbilical cord plasma
phospholipids, was associated with a significantly longer gestational age. However, Oken et al.
(2004) and Guldner et al. (2007) reported inverse associations between maternal fish intake
during pregnancy and birth size. Oken et al (2004) assessed the association between combined
maternal intake of EPA and DHA in early pregnancy and length of gestation or fetal growth in
2,109 of well-nourished, well-educated, ethnically diverse U.S. women. They found that the
combined daily intake of EPA and DHA ranged from 0.02 to 0.35g. Daily intakes of long chain
n-3 fatty acids were not associated with the length of gestation or preterm birth risk, but were
associated with reduced fetal growth. Guldner et al. (2007) assessed the association between
maternal intake of seafood in the first trimester and preterm birth, low birth weight, SGA birth in
2,389 French women with singleton pregnancies and with low baseline rates of adverse birth

11

outcomes. They found that an increased intake of large crustaceans decreased fetal growth. This
may possibly be due to the high level of contaminants found in large crustaceans in France.
Therefore both of these studies explained that the association was possibly due to contaminants
found in certain types of seafood. Of the three studies, only one study showed positive
association between long-chain omega-3 fatty acids during pregnancy and fetal growth, while the
other two studies showed no association between maternal fish intake and birth size due to
contaminants found in certain types of seafood. Because of the different geographical locations
and accessibility to seafood or other types of food containing fatty acids, the fatty acids baseline
would produce mixed results in showing positive or no association between fatty acids intakes
and birth outcomes.
In a prospective study with 13,475 women, who reported a singleton pregnancy between
1991 and 2001, in the Nurses’ Health Study II, Bowers et al. (2012) determined whether the total
amount and the type and source of dietary fat consumed during prepregnancy are related to the
risk of gestational diabetes. Individuals in the highest quintile of animal fat intake had 88%
increased risk of gestational diabetes after adjustment for non-dietary risk factors such as age,
parity, current smoking, BMI, physical activity, family history of diabetes, alcohol, race and total
calories, and vegetable fat intake (RR: 1.88, 95% CI=1.36-2.60). Cholesterol intake was also
positively associated with the risk for gestational diabetes mellitus after adjustment for dietary
and non-dietary risk factors, including specific fatty acids (RR:1.45; 95% CI: 1.11, 1.89). They
found higher prepregnancy intake of animal fat and cholesterol was associated with gestational
diabetes mellitus risk.
Saldana et al. (2004), examined the relation between macronutrient intake, including total
energy from protein, carbohydrate, fat percentage in the second trimester of pregnancy and

12

impaired glucose tolerance and gestational diabetes mellitus in 1,698 women in the Pregnancy,
Infection, and Nutrition Study. The Carpenter & Coustan cutoffs of 95 ml/dL for fasting, 180
mg/dL for 1-h, 155 mg/dL for 2-h and 135 mg/dL for 3-h values were used to define impaired
glucose tolerance or gestational diabetes. One abnormal value from 3-h, 100-g oral glucosetolerance test indicated impaired glucose tolerances, and two abnormal values indicated
gestational diabetes. Dietary intake during the second trimester was assessed with a foodfrequency questionnaire. They found that total energy (kcal) did not differed by glucose status.
Energy intake during the second trimester of pregnancy did not differ by glucose status: i.e.,
normal, impaired glucose tolerance or gestational diabetes. Intake of carbohydrate and fat
percentage energy significantly differed. Carbohydrate percentage of energy was significantly
higher for women with normal glucose tolerance compared to those with impaired glucose
tolerance or gestational diabetes (53 vs. 50, 51%, respectively). The percentage of energy from
fat was significantly lower in women with normal glucose tolerance compared to impaired
glucose tolerance or gestational diabetes (33 vs. 35, 35%, respectively). This study indicated that
composition of diet, carbohydrate and fat percentage were important in predicting glucose
intolerance status, rather than total caloric intake. Despite different time points of study
investigations, prepregnancy (Bowers et al., 2012), and 2

nd

trimester (Saldana et al., 2004), both

studies showed that dietary fat intakes increased the risk of gestational diabetes. However,
prepregnancy weight status and vitamins or mineral supplement information should be adjusted
in both studies.
Micronutrients. Table 1 shows the percentage of the increase in recommended dietary
intakes for pregnant adults over non-pregnant women. The high percentage of the increase over

13

that recommended for non-pregnant, non-lactating women indicates the importance of the
nutrients particularly during pregnancy.

Table 1. Dietary Reference Intakes (DRIs): Recommended Dietary Allowances and Adequate
1
Intakes, Micronutrients and Vitamins
(Source: Food and Nutrition Board, Institute of Medicine, National Academies)
Non-pregnant,
Pregnant women
% increase over
rd
non-lactating women
non-pregnant, non(3 trimester)
lactating women
for pregnant
1
women
Females
Females
PG
PG
%
19-30 y
31-50 y
19-30 y
31-50 y
Calcium (mg/d)
1,000
1,000
1,000
1,000
0
Phosphorous (mg)
700
700
700
700
0
Magnesium (mg)
310
320
350
360
13
Iron (mg)
18
18
27
27
50
Zinc (mg)
8
8
11
11
38
Iodine (mcg)
150
150
220
220
47
Selenium (mcg)
55
55
60
60
9
Copper (mg)
900
900
1,000
1,000
11
Vitamin A (mcg RE)
700
700
770
770
10
Vitamin D (mcg)
15
15
15
15
0
Vitamin E (mg a-TE)
15
15
15
15
0
Vitamin K (mcg)
90
90
90
90
0
Vitamin C (mg)
75
75
85
85
13
Thiamine (mg)
1.1
1.1
1.4
1.4
27
Riboflavin (mg)
1.1
1.1
1.4
1.4
27
Niacin (mg NE)
14
14
18
18
29
Vitamin B-6 (mg)
1.3
1.3
1.9
1.9
46
Folate (mcg)
400
400
600
600
50
Vitamin B-12 (mcg)
2.4
2.4
2.8
2.8
17
1

Calculations based on non- and pregnant females 19-30 y

IRON. Cogswell et al. (2003) evaluated the associations between maternal anemia and
adverse birth outcomes. They found that if a daily iron supplement, containing 30mg Fe as
14

ferrous sulfate or placebo, was given from enrollment to 28 weeks of gestation, to initially ironreplete, non-anemic pregnant women, it would reduce the prevalence of anemia at 28 weeks and
increase birth weight among 513 low-income pregnant women enrolled in the study before 20
weeks of gestation. Total of 275 women with a hemoglobin concentration ≥ 110 g/L and a
ferritin concentration ≥ 20 µg/L, were randomly placed in either the iron supplement group
(n=146) or the placebo group (n=129). Compared with a placebo, iron supplementation from
enrollment to 28 weeks of gestation, did not significantly affect the overall prevalence of anemia
or the incidence of preterm births, but led to a significantly higher mean (±
SD) birth weight (206
±565 g; p=0.010), a significantly lower incidence of low-birth-weight infants (4% vs. 17%;
p=0.003), and a significantly lower incidence of preterm low-birth-weight infants (3% vs. 10%;
p=0.017). However, diet intakes containing iron or other supplements taking during pregnancy,
was not investigated. Further, they did not report either the lengths of iron supplementation
during the study period, or the bioavailability of capsules, both of which could influence the birth
outcome.
Alwan et al. (2011), conducted a study on a cohort of 1,274 pregnant women, to assess
the association between dietary iron intake and birth outcomes in United Kingdom. Total intake
of iron from both diet and supplement were investigated using 24 hour dietary recalls in the first,
second and third trimesters. Total iron intake from food and supplements averaged 16.5 mg/day
in the first trimester. Mean birth weight was 3,439 g with 4.4 % of babies weighing <2,500g.
Total iron intake from foods and supplements during the first trimester was associated with birth
weight centile (unadjusted change = 5.2 centile points/10mg increase in iron, 95% CI=2.2-8.2,
p=0.001). Both studies showed a positive relationship between iron intake from foods and birth
outcome. However, none of these studies measured or controlled for baseline iron storage which

15

might impact birth outcomes. The difference in the subjects targeted in the two studies, only
non-anemic pregnant women (Cogswell et al., 2003) in one study and both anemic and nonanemic pregnant women (Alwan et al., 2011) in the other, may have been the cause of the varied
results.
FOLATE. Folate is critical to fetal development, and a cofactor for many essential
cellular reactions, including DNA synthesis. Mandatory folic acid fortification of enriched
cereal-grain products in the U.S. and Canada, implemented in 1998 (Food and Drug
Administration, 1996; Health Canada, 1997), has substantially improved the folate status of
reproductive age women and resulted in a significant decline in neural tube defects (NTDs)
(CDC, 2000).
Siega-Riz et al. (2004), conducted a prospective cohort study of 2,314 lower- to middleincome U.S. pregnant women receiving prenatal care during 24 to 29 weeks’ gestation (n=2,026)
period. They assessed the relationship between maternal folate status during second trimester of
pregnancy and the prevalence of preterm birth, defined as a delivery before 37 completed weeks
of gestation. Information on supplement use for the preconception time period, on the day of the
interview and during the past 24 hours was included. Mean daily dietary folate intake of
participants was 463 µg (±248). The relative risk is the ratio of the proportions of cases having a
positive outcome in both experiment and control groups. They reported that folate intake from
diet ≤ 500 µg/day was associated with increased risk of preterm birth with a relative risk, 1.8 [95%
CI=1.4-2.6] controlling total energy intake, with an unadjusted relative risk for preterm birth, 1.5
[95% CI=1.2-2.0]. Serum folate < 16.3 ng/mL was associated with a relative risk 1.8 [95%
CI=1.3-2.5] of preterm delivery controlling for prenatal use.

16

Bukowski et al. (2009), examined the association between duration of preconceptional
folate supplementation and the risk of spontaneous preterm birth. They conducted a multicenter
prospective cohort multiethnic, mixed socioeconomic status sample of 34,480 U.S. pregnant
women, with singleton pregnancies of <14 weeks’ gestation at enrollment. The duration of
preconceptional folate supplementation was reported and categorized into three groups: ≥ 1 y, <
1 y, and no preconceptional supplementation. They found that preconceptional folate
supplementation for ≥1 year significantly reduced the risk of preterm birth by 69% during weeks
20-28 and by 47% during weeks of 27-32 but did not reduce preterm birth risk after 32 weeks.
However, they did not report the amount of folate supplementation that the women had.
Christian et al. (2003), studied 4,926 pregnant women in Nepal, to assess the effect of
daily supplements of folic acid, folic acid-iron, folic acid-iron-zinc, or multiple micronutrients
along with vitamin A on birth size and risk of low birth weight. Total of 426 communities were
randomized to five regimens of daily supplementation: folic acid, folic acid-iron, folic acid-ironzinc, or multiple micronutrients, along with vitamin A, with vitamin A alone as the control group.
Birth weight, length, and head and chest circumference were assessed within 72 hours of birth.
Supplementation with maternal folic acid alone had no effect on birth size. Folic acid-iron
increased mean birth weight by 37g. Folic acid-iron-zinc had no effect on birth size compared
with controls. Multiple micronutrient supplementation increased birth weight by 64g.
Watanabe et al. (2008), evaluated the levels of dietary folate intake in first, second and
third trimester of pregnancy in 197 Japanese women and subsequent birth weight of their babies.
They found dietary folate intake was not a significant predictor of birth weight. This finding
might be due to the small sample size (n=197). The daily intake of folate increased from 248.5 ±
113.1 µg/d in the first trimester to 275.4±
100.2 µg/d in the third trimester. In summary, of the

17

four studies on folate nutritional status and birth outcomes, two of the studies showed significant
benefits of folate supplementation in reducing the risk of preterm births, while the other two
showed no effect of folate supplementation or dietary intake on preterm birth or birth weight.
Differences in these results might be due to the period of supplementation before or during
pregnancy, different baseline daily folate intakes or serum folate levels, race/ethnicity, study
designs and nutritional knowledge.
B. Gestational Weight Gain Guidelines
Purpose and History. The IOM guidelines for gestational weight gain aimed to improve
maternal, fetal and child health, which are key public health goals (IOM, 2009). The first IOM’s
gestational weight gain guidelines published in 1970 (IOM, 1970) has been revised regularly
based upon new scientific evidence. The 1970, Maternal Nutrition and the Course of Pregnancy
(IOM, National Research Council (NRC), Nutrition during pregnancy) had a single target
recommended average gain of 10.9 kg (24 pounds) within a range of 9.1-11.3 kg (20-25 pounds)
(NRC, 1970) for all pregnancies. This target was based on the amount of weight that healthy
women gain when meeting the physiologic needs of pregnancy: the products of conception,
expansion of plasma volume, red cell mass, and maternal fat stores (IOM, 2009). The first report
was developed from concerns about high neonatal infant mortality rates in the U.S. compared to
other developed countries. In 1983, the American Academy of Pediatrics (AAP) and the
American College of Obstetricians and Gynecologists (ACOG) published the first edition of the
AAP/ACOG Guidelines for Perinatal care, a manual outlining the two organizations’ joint
recommendations for the care of pregnant women and newborns. The guidelines increased the
previously recommended the gestational weight gain to 22-27 pounds. In 1990, the IOM
published the second gestational weight gain recommendations based on prepregnancy BMI:
18

12.5-18.0 kg for underweight women (prepregnancy BMI<19.8); 11.5-16.0 kg for normal
women (prepregnancy BMI 19.8-26.0); 7-11.5 kg for overweight women (prepregnancy BMI
26.0-29.0); <6 kg for obese women (prepregnancy BMI>29.0). The 1990 IOM report, Nutrition
During Pregnancy, offered specific recommendations for weight gain during pregnancy
stratified by prepregnancy BMI, such as population subgroups, including adolescents, members
of racial and ethnic groups, women of short stature, and women carrying twins; and detailed
historic trends in weight gain recommendations and guidelines. In the years since the release of
the weight gain recommendations in the IOM (1990) report, changes in the demographic and
epidemiologic profile of the U.S. population have occurred. There is a greater diversity among
childbearing age women in the U.S. and an increase in prepregnancy BMI and excessive
gestational weight gain in all populations groups (Yeh et al., 2005; Kim et al., 2007). The IOM
(1990) pregnancy weight guidelines were developed to reduce low birth weight infants.
Although adverse health outcomes for excess weight gain were considered in the IOM (1990)
weight gain guidelines, the recommendations were derived largely from data collected in the
1980 National Natality Survey (Retrieved from
http://www.cdc.gov/nchs/about/major/nmihs/abnmihs.htm [accessed July 3rd, 2011]) and
focused on reducing low birth weight.

19

Table 2. Institute of Medicine’s recommended total weight gain ranges for women with singleton
pregnancies (IOM, 1990) and Recommendations for total and rate of weight gain during
pregnancy, by prepregnancy BMI (IOM, 2009)
Prepregnancy BMI
Underweight
Normal Weight
Overweight
Obese (includes all
classes)

2

BMI (kg/m )
(WHO)
<19.8
19.8-26.0
26.0-29.0
>29.0

2

Total Weight Gain Range
(kg)

Total Weight Gain
Range (lbs)

12.5-18.0
11.5-16.0
7-11.5
<6

27.6-39.7
25.4-35.3
15.4-25.4
<13.2

Rates of Weight Gain
nd
rd
2 and 3 Trimester
(Mean Range in
lbs/wk)
1 (1-1.3)
1 (0.8-1)
0.6 (0.5-0.7)
0.5 (0.4-0.6)

Prepregnancy BMI

BMI (kg/m )
(WHO)

Total Weight Gain Range
(lbs)

Underweight
Normal Weight
Overweight
Obese (includes all
classes)

<18.5
18.5-24.9
25.0-29.9
≥30.0

28-40
25-35
15-25
11-20

IMPLICATIONS OF THE GESTATIONALWEIGHT GAIN GUIDELINES. To
achieve the IOM’s gestational weight gain guidelines, reproductive aged women and their health
care providers should be aware of the existence, purposes and importance of the guidelines.
Government agencies, organizations that provide health care to pregnant women or those who
are planning to become pregnant, private voluntary organizations, and medical societies that
have adopted these guidelines as their standard of care could all provide this education (IOM,
2009). In addition to being made aware of their weight gain as pregnancy progresses through the
use of weight-gain charts, women should be provided with advice about both diet and physical
activity (ACOG, 2002).

20

C. Factors associated with Gestational weight gain
OBESITY IN CHILD-BEARING AGE WOMEN. Gestational weight gain is generally
inversely associated with maternal prepregnancy BMI. For the over 2.3 million deliveries in
Germany for 1995-2000, relatively short and heavy women had lower gestational weight gain
than tall and thin women (Voigt et al, 2007). Chu et al. (2009) reported similar findings from the
PRAMS data of 52,988 women who delivered full-term singleton infants in 2004-2005.
Confounding variables were maternal age and gestational age of the infant (full term, late
preterm). In a multivariable regression model, gestational weight gain was most closely related
to maternal prepregnancy obesity, followed by higher parity, an African American or Hispanic
racial identity, and higher maternal age (Chu et al., 2009).
DISTRIBUTION OF WOMEN BY GESTATIONAL WEIGHT GAIN. Lederman et
al. (2002), conducted a study with Black and Hispanic first time and full term mothers (n=47) to
determine weight gain during pregnancy and weight changes postpartum. Overall, 64% of the
women gained weight excessively during pregnancy, as determined from their self-reported
prepregnancy weight status against the IOM recommendations (1990). Only 25% had the
recommended total gestational weight gain as, and 8.5% gained less than recommended. All
women who were obese or overweight prepregnancy (n=10), gained excessively during
pregnancy. Mean (±
SD) weight gain was highest in overweight (46.3 ±14.4 lbs) and obese (41.
0 ±16.8 lbs), greatly exceeding the recommended gain (7-11.5 lbs, and < 6 lbs, respectively).
COMPLICATIONS ASSOCIATED WITH PREPREGNANCY OBESITY. Obesity
is an epidemic for both men and women in all developed countries in the world. In the U.S., the
prevalence of obesity has increased in the last 25 years resulting in more than one-third of U.S.
adults being obese in 2005-2006 (CDC/NCHS). According to the 2005-2006 PRAMS, 23% of

21

2

U.S. women began pregnancy as overweight (BMI 25-29.9 kg/m ), and 18.7% as obese (BMI ≥
2

30 kg/m ) (Chu et al., 2009). In the past 30 years, the prevalence of obesity among American
women of childbearing age has increased (Kim, 2007). According to the data from the Pregnancy
Risk Assessment Monitoring System (PRAMS) 1993-2003 of 73,115 women, in nine different
2

states, one-fifth of American women in these states are obese (BMI > 29 kg/m ) at the start of
pregnancy, a figure that has risen 70 percent in the past decade (Kim et al., 2007).
Baeten et al. (2001), examined the associations between prepregnancy weight status and
the risk of pregnancy complications and negative birth outcomes among nulliparous women in
population-based cohort study with 96,801 Washington State birth certificates from 1992 to 1996.
Compared to prepregnancy underweight women (BMI <20.0), both prepregnancy overweight
(BMI 20.0-24.9) and obese women (BMI 25.0-29.9) had a significantly increased risk for
gestational diabetes, preeclampsia, cesarean delivery, and delivery of a macrosomic infant.
Frederick et al. (2006) reported that every unit increase in prepregnancy BMI resulted in an 8%
increased risk of preeclampsia. Maternal prepregnancy obesity during pregnancy is associated
with many pregnancy complications, such as pregnancy related gestational diabetes (Baeten et
al., 2001), preeclampsia (Baeten et al., 2001; Frederick et al., 2006), cesarean delivery (Baeten et
al., 2001), and delivery of a macrosomic infant (Baeten et al., 2001). Maternal prepregnancy
obesity yields adverse pregnancy outcomes. Thus, maintaining normal weight before pregnancy
is important for reducing adverse pregnancy complications and birth outcomes.

22

1. Diet Quality
In a cohort study of 224 pregnant women who came for their first prenatal visit and
delivered in a university hospital in Boston, maternal energy-adjusted intake of macronutrients
was associated with either maternal weight gain or birth size (Lagiou et al., 2004). The pregnant
women’s dietary intake during the second trimester was ascertained at the 27th week of
pregnancy through a food frequency questionnaire. Neither energy intake nor any of the
macronutrients was associated with birth size. In contrast, gestational weight gain by the end of
the second trimester of pregnancy was associated with energy intake (+0.9 kg/standard
deviation(s.d.) of intake) as well as energy adjusted intake of protein (+ 3.1 kg/s.d. of intake),
lipids of animal origin (+2.6 kg/s.d. of intake) and carbohydrates (-5.2 kg/s.d. of intake).
Deierlein et al. (2008), examined the effects of glycemic load and energy density on
gestational weight gain. Data were taken from 1,231 women pregnancies who participated in the
Pregnancy, Infection, and Nutrition Cohort Study and had singleton pregnancies. Dietary
information was collected during weeks 26-29 week of pregnancy, using a semi-quantified food
frequency questionnaire. Linear regression models were used to estimate the associations
between quartiles of glycemic load and energy density with total gestational weight gain and
weight gain ratio (actual weight gain/recommended weight gain). Dietary energy density (kcal/g
food consumed) was positively associated with daily energy intake, gestational weight gain, and
weight gain ratio (observed weight gain/ expected weight gain); however, glycemic load was not
associated with gestational weight gain.
Olson et al. (2003), conducted a prospective cohort study that followed 622 women from
early pregnancy until 2 y postpartum, to examine the association between potentially modifiable
psychosocial and behavioral factors and gestational weight gain. The changes in the amount of

23

food eaten throughout each trimester of pregnancy were compared with the amount eaten before
pregnancy. Women were asked, “How has the amount of food you eat out now changed
compared with times when you were not pregnant?” Other aspects of food intake were assessed
using additional behavioral questions, particularly the number of snacks consumed per day and
the frequency of regular meals and breakfast. Consuming much more or much less food during
pregnancy than prior to pregnancy were associated with greater (3.67 lbs; p<0.001) and less (3.15 lbs; p<0.05) gestational weight gain, respectively, compared with maintaining similar levels
of food intake. Women who consumed more than 3 servings of fruits and vegetables daily
throughout each trimester of pregnancy, gained significantly less weight (-1.81 lbs; p<0.05) than
those who consumed <3 servings. Women who ate “much more” during rather than before their
pregnancy had an adjusted odds ratio of 2.35 for excessive gestational weight gain.
Olafsdottir et al. (2006), identified dietary risk factors related to excessive gestational
weight gain in 495 pregnant Icelandic women. Dietary intake was estimated with a semiquantitative food frequency questionnaire addressing food intake together with lifestyle factors
for the previous 3 months. In the study, 26% gained suboptimal and 34% gained excessive
weight during pregnancy. Women in late pregnancy who had at least optimal weight gain, were
eating more (OR=3.32, 1.81-6.09) and drinking more milk (OR=3.10, 1.57-6.13), compared with
women who had suboptimal weight gain. Eating more sweets early in pregnancy increased the
risk of gaining excessive weight (OR=2.52, 1.10-5.77). There was no significant difference in
maternal macro nutrient intake (g/day) between the weight gain groups in early pregnancy, but in
late pregnancy there were significant differences across all weight gain groups and overweight
women in particular. Overweight women showed significantly higher total energy intake with
increasing weight gain. They ate more fat and carbohydrates, but there was not a significant

24

difference in protein intake in absolute amounts. Compared with women gaining suboptimal
weight, the diets of overweight women gaining excessive weight had a higher energy percentage
from fat (34.1 ±6.2 vs. 27.5 ±5.4) and lower energy percentage from carbohydrates (51.5 ±6.2
vs. 57.1 ±5.1). In logistic regression models for factors related to at least optimal and excessive
gestational weight gain, compared with women with prepregnancy BMIs between 20-24.9,
2

women with a prepregnancy BMI between 25-29.9 kg/m were more than 10 times more likely
2

to gain at least optimal weight (95% CI=4.29-24.23 kg/m ); obese women, prepregnancy BMI ≥
2

30kg/m were 3 times more likely to do so (95% CI=1.17-3.58). There was a strong relationship
between higher gestational weight gain and eating more in late pregnancy (OR=2.04, 1.17-3.58)
and drinking more milk late in pregnancy (OR=1.82, 1.08-3.06).
Laraia et al. (2004) reported that women living more than 4 miles from a supermarket
had a higher risk (adjusted OR = 2.16; 95% CI: 1.2-4.0) of falling into the lowest tertile of Diet
Quality Index for Pregnancy (DQI-P) compared to women living less than 2 miles away, after
controlling for individual characteristics such as age, race, income, education and marital status.
Food frequency questionnaires were used to construct DQI-P and included: servings of grains,
vegetables, fruits, folate, iron and calcium intake, % of calories from fat, and meal pattern score.
The DQI-P was based on eight dietary characteristics—percentage of recommended servings per
day of grains, vegetables, and fruits, percentage RDA of folate and iron, AI of calcium,
percentage of calories from fat, and a meal pattern score (Bodnar & Siega-Riz, 2002). The first
three components reflect the dietary adequacy of grain, vegetable and fruit intakes based upon
the Dietary Guidelines for Americans (USDA, 2001) and the Food Guide Pyramid (Shaw et al.,
2001). The next three components of the DQI-P reflect intake of nutrients particularly important

25

for pregnancy: folate, iron and calcium. These nutrients represent dietary intake exclusive of
vitamin/mineral supplements. The seventh component was the percentage of energy from fat in
the diet based on recommendations from the Dietary Guidelines for Americans (USDA, 2001).
The final component relates to meal/snack patterns. The Institute of Medicine recommends that
women should follow a meal pattern of three meals and at least two snacks during the gestation
period (IOM, 1992). Each of the eight characteristics is scored between 0 and 10 for adequacy in
meeting the recommendation for an overall score of 80.
Tssiga et al (2010) found that the Healthy Eating Index (HEI) score for one hundred
Greek pregnant women during pregnancy was significantly higher in participants with either an
underweight or normal prepregnancy BMI. Higher protein intake (g/kg body weight) was shown
in underweight and normal prepregnancy BMI groups compared with the obese prepregnancy
BMI group. The highest trans-fat intake was by overweight women. HEI was negatively
correlated with prepregnancy BMI (r=-0.298, p ≤ 0.003). In summary, dietary energy density
(Deierlein et al., 2008), reduced servings of fruit and vegetables (Olson et al., 2003), drinking
more milk, eating more sweets, higher energy percentage from fat, lower energy percentage from
carbohydrate (Olafsdottir et al.., 2006), distance from a supermarket, and higher prepregnancy
BMI were associated gestational weight gain.
Rifas-Shiman et al. (2006) examined the relationships between maternal characteristics,
prepregnancy BMI, age, parity, race/ethnicity, education and diet quality during the first
trimester of pregnancy (n=1,777). Middle-income women attending childbirth education classes
(n=82) and low-income women attending a free prenatal clinic (n=27) were recruited for
assessment of 2-day dietary recalls. These dietary records were analyzed with the Food Processor
computer program and compared with the RDA of each nutrient including calories, protein,

26

carbohydrate, percentage calories from fat, vitamin A, vitamin D, folate, calcium, iron,
phosphorous and sodium. They developed the Alternate Healthy Eating Index, slightly modified
for pregnancy, to measure diet quality on a 90-point scale with each of the following nine
components contributing 10 possible points: vegetables; fruit; ratio of white to red meat; fiber;
trans fat; ratio of polyunsaturated to saturated fatty acids; and folate, calcium and iron from foods
(Rifas-Shiman et al., 2006). The lowest scores were: ratio of polyunsaturated to saturated fatty
acids (5.5 ±1.9), calcium (8.4 ±1.7), folate (6.0 ± 1.8) and iron (6.0 ±1.6). Inadequate intake of
micronutrients, particularly iron, calcium, zinc, and folate has been the greatest concern for
pregnant women.
2. Physical Activity
Excessive gestational weight gain is likely to be a consequence of a persistently positive
energy balance and might therefore be influenced by physical activity and diet modification
(Streuling et al., 2010). Asbee et al. (2009) conducted a randomized controlled trial assigned to
100 women to receive either an organized, consistent program of intensive dietary and lifestyle
counseling (n=57) or routine prenatal care (n=43). The primary aim of the study was to
determine the proportion of patients whose gestational weight gain was within the IOM
recommendations. Intervention groups were instructed to engage in moderate-intensity exercise
at least three times per week and preferably five times per week. The control group received only
routine prenatal care. The participants in the intensive counseling group gained significantly less
weight (28.7 ±12.5 lb vs. 35.6 ±15.5 lb, p=0.01) than those in the routine care group. In the
routine care group, 48.8% were adherent to the IOM guidelines, and in the intensive counseling
group, 61.4% of patients adhered to the IOM guidelines.

27

Shirazian et al. (2010) assessed the impact of lifestyle modification program on weight
gain in pregnancy and evaluated its effect on adverse pregnancy outcomes, preeclampsia,
gestational diabetes, gestational hypertension, and various postpartum complications.
Intervention group was emphasized for a healthy diet and exercise and was goaled toward
limiting pregnancy weight gain to 15 to 20 pounds for overweight prepregnancy weight status
2

patients (BMI 25-29.9 kg/m ) and no more than 15 lbs for obese prepregnancy weight status
2

patients (BMI ≥ 30 kg/m ). For physical activity intervention, written material, seminars and
counseling sessions to encourage walking as exercise, received a pedometer to monitor. The
patients in the intervention group gained significantly less weight during their pregnancies (17.76
±16.30 lbs) compared to control group (34.00 ±16.58 lbs, p=0.003).
Claesson et al. (2008) conducted a prospective case-control intervention study to
minimize obese womens’ total weight gain during pregnancy to less than 7 kg and to investigate
delivery and neonatal outcome. An intervention group received weekly aqua aerobics especially
designed for obese women once or twice a week. The intervention group gained less weight (8.7
±5.5 kg) than the control group, (11.3 ±5.8 kg; p<0.001).
Kinnunen et al. (2007) investigated whether individual counseling on diet and physical
activity during pregnancy can have positive effects on diet and leisure time physical activity and
prevent excessive gestational weight gain in 105 pregnant women. Physical activity counseling
sessions (5 times), which included discussion about needs and opportunities to increase leisure
time physical activity and designing an individual activity plan and an opportunity to join a
group exercise session, while the control group received standard maternity care including shortterm counseling on diet, physical activity and gestational weight gain. The control group had
gestational weight gain of 14.3 ±4.1 kg and the intervention group had gestational weight gain
28

of 14.6 ±5.4 kg with no differences of gestational weight gain between these two groups. Out of
four previous studies, three of these studies showed lowering effect and one study no effect of
physical activity on gestational weight gain. Either physical activity alone or diet modification
appeared to reduce gestational weight gain. While another study showed no effect, usually
combining with diet and physical counseling and supplementary weight monitoring through
pregnancy will be successful in lowering gestational weight gain.
SOCIAL ENVIRONMENT. Laraia et al. (2007) conducted a study of neighborhood
factors associated with physical activity and weight gain during pregnancy. Study participants
were 703 pregnant women in Pregnancy, Nutrition, and Infection prospective cohort study. They
found that social spaces, defined as the presence of parks, sidewalks, porches and people
including nonresidential visitors, were positively associated with the odds for adequate
gestational weight gain. Results from a neighborhood data collection inventory identified three
social constructs, physical incivilities, territoriality, and social spaces, which were hypothesized
to influence maternal health behaviors. The physical incivility scale was associated with
decreased odds (adjusted OR=0.74, 95%CI=0.57, 0.98) in participating in vigorous leisure
activity before pregnancy, after controlling for several individual confounders, and a crude
association for decreased odds of excessive gestational weight gain (OR=0.79, 95%CI=0.64,
0.98). The social spaces scale was associated with decreased odds for inadequate (adjusted
OR=0.74, 95%CI=0.56, 0.98) and excessive (adjusted OR=0.69, 95%CI=0.54, 0.98) gestational
weight gain. The social spaces scale was also associated with decreased odds of living greater
than 3 miles from a supermarket (adjusted OR=0.03, 95%CI=0.00, 0.27). Territoriality was not
associated with any pregnancy-related health behavior. Covariates for adjusted logistic models
were age, education, race/ethnicity, income, any children, and marital status, age and

29

prepregnancy BMI. These findings suggest that neighborhood environments can influence
gestational weight gain by providing access to healthy foods and opportunities for physical
activities.
BEHAVIORAL INTERVENTION. Olson et al. (2004) evaluated the efficacy of an
intervention directed at preventing excessive gestational weight gain. Healthy pregnant women
with normal and overweight prepregnancy BMI were monitored from early pregnancy until 1
year pregnancy (Olson et al., 2004). The intervention group (n=179) encouraged pregnant
women to gain weight within the range recommended by the IOM while the control group
(n=381), only received prenatal care. The intervention had two major components: a clinical
component, including guidance about and monitoring of gestational weight gain by health care
providers using new tools in the obstetric chart, and a by-mail patient education program. Overall
gestational weight gain did not differ significantly between the control and intervention groups
(14.80 ±4.68 kg vs. 14.10 ±4.51kg), respectively; p=0.09). The proportion of women gaining
more than the recommended amount during pregnancy did not differ between the control and
intervention groups (45% vs. 41%; p=0.03).
Phelan et al. (2011) reported that a low-density behavioral intervention during pregnancy,
reduced excessive gestational weight gain in normal prepregnancy weight women and prevented
postpartum weight retention in normal and overweight/obese prepregnancy women. They
examined whether a behavioral intervention that aimed to decrease high-fat foods intakes and
increase physical activity and daily self-monitoring of eating, exercise, and weight during
pregnancy could decrease the proportion of women who exceeded the 1990 IOM
recommendations for gestational weight gain and increase the proportion of women who
returned to prepregnancy weights by 6 month postpartum. This study was a randomized,

30

assessor-blind, controlled trial. Participants were pregnant (13.5 wk gestation), normal (n=201)
and overweight or obese prepregnancy weight women (n=200) whose age average was 28.8 y.
Intervention decreased the percentage of normal weight women who exceeded the IOM
recommendations (40.2% vs. 52.1%; p=0.003) and increased the percentage of normal weight
and overweight/obese women who returned to their prepregnancy weights or below by
postpartum at 6 months (30.7% vs. 18.7%; p=0.005). In conclusion, the results of this study
showed that a low-intensity behavioral intervention reduced excessive gestational weight gain in
normal weight women and increased the percentage of normal weight and overweight/obese
women who returned to their preconception weights by 6 months postpartum.
Another randomized controlled trial was done by Guelinckx et al. (2010) to examine if
lifestyle intervention based on a brochure or education can improve dietary habits, increase
physical activity, and reduce gestational weight gain in obese pregnant women. In the
randomized controlled trial, 195 white, obese prepregnancy women (29 ±4 y; BMI 33.6 ±4.22)
were randomly assigned into 3 groups: a group that received nutritional advice from a brochure,
a group that received the brochure and lifestyle education by a nutritionist, and a control group.
Nutritional habits were evaluated every trimester through 7-day food records. Energy intake did
not change during pregnancy and was comparable in all groups. Fat intake, specifically saturated
fat intake, decreased and protein intake increased from the first to the third trimester in the
passive and active groups, while an opposite change occurred in the control group. Calcium
intake and vegetable consumption increased during pregnancy in all groups. Physical activity
decreased in all groups, especially in the third trimester. No significant differences in gestational
weight gain and obstetrical or neonatal outcome could be observed between the groups. Both
lifestyle interventions improved the nutritional habits of obese women during pregnancy. Neither

31

physical activity nor gestational weight gain was affected. The effect of nutrition lifestyle
intervention on gestational weight gain was the opposite. Of three studies of behavioral
interventions to control gestational weight gain, only one study showed effect on preventing
excessive gestational weight gain through intervention. Due to different designs of intervention
strategies, IOM gestational weight gain recommendations (Olson et al., 2004), self-monitoring of
eating and exercise (Phelan et al., 2011), dietary habits, physical activity (Guelinckx et al., 2010)
on targeted pregnant women with different socioeconomic status and baseline nutritional
knowledge , no report on adherence to supplementation resulted in mixed results.
3. Pregnancy Outcomes
The report Nutrition During Pregnancy (IOM, 1990) focused on the short-term
consequences of gestational weight gain, for little is known on the association between
gestational weight gain, intrauterine growth and long-term outcomes. The associations between
gestational weight gain and birth outcomes are affected by other factors such as maternal diet
composition, and physical activity level. It is important to determine whether these relationships
are independent of prepregnancy BMI (IOM, 2009).
3.1 Infants. Cedergren et al. (2006) reported that obese women with low gestational
weight gain reduced risk for LGA births. Increased risk was found for preeclampsia and LGA
infants among average and overweight women who had excessive gestational weight gain. Rudra
et al. (2008) examined the relationship between prepregnancy BMI and gestational weight gain
to preterm delivery in 2,468 cohort participants in western Washington State. Gestational weight
gain from prepregnancy to weeks18-22 of pregnancy was inversely associated with spontaneous
preterm delivery (0.1 kg/week adjusted OR, 0.87; 95% CI, 0.77-0.99) and not strongly associated
with preterm premature rupture of membranes (adjusted OR, 1.03; 95% CI, 0.90-1.17). Oken et
32

al. (2009) investigated the rate of gestational weight gain with the lowest combined risk of 5
short-term maternal outcomes including preterm delivery (<37 weeks), SGA and LGA and
longer-term maternal outcomes including postpartum retention at 1 y and child obesity at age 3
years (BMI > 95th percentiles) in 2,012 mother-child pairs recruited in 1999-2002 into Project
Viva, a cohort study in Massachusetts. They found that gestational weight gain was directly
associated with risk of LGA among normal-weight and overweight women, whereas it was
inversely associated with risk of SGA and preterm delivery among normal weight women.
Among women who had normal prepregnancy weight, there were inverse associations between
gestational weight gain and the risk of SGA (OR=0.68, 95% CI=0.54-0.85). Among obese
women, gestational weight gain was not associated with risk of LGA, SGA, or preterm birth. In
summary, overweight or obese prepregnancy BMI women are at a higher risk of gaining
excessive gestational weight to reduce the risk of SGA, LGA, preterm delivery, postpartum
weight retention and childhood obesity.
3.2 Postpartum Weight. Postpartum weight is a woman’s weight immediately after
delivery of the fetus, placenta, and amniotic fluid (IOM, 2009). In the subsequent days to weeks,
the excess extracellular and extravascular water was gained during pregnancy is lost and plasma
volume returns to prepregnancy values. Postpartum weight retention is the amount of weight that
remains at this later time (3 mo to 3 y) minus the woman’s prepregnancy weight; it includes the
weight of any increased breast tissue being used for lactation as well as any remaining fat mass
gained during pregnancy. Postpartum weight retention after1 year (Oken et al., 2009; Vesco et
al., 2009) or 9 months (Kac et al., 2004) was used to assess postpartum weight retention.
According to IOM (2009), modifiable determinants of gestational weight gain include 1)
Societal/institutional factors: media, culture and acculturation, health services, policy; 2)

33

Environmental factors: altitude, environmental toxicants, natural and man-made disasters; 3)
Neighborhood/Community factors: access to healthy foods, opportunities for physical activity; 4)
Interpersonal/Family Determinants: family violence, marital status, partner and family support.
Non-modifiable factors include: genetic characteristics, epigenetics and developmental
programming, prepregnancy BMI, preexisting morbidities, sociodemographic factors i.e., age,
race/ethnicity, and food insecurity. Potentially modifiable factors include: 1) physiological
factors, i.e, insulin, leptin, and hormonal milieu, basal metabolic rate (BMR); 2) medical factors,
i.e., hyperemesis gravidarum, anorexia nervosa and bulimia, bariatric surgery; 3) psychological
factors, i.e., depression, stress, social support, attitude toward low weight gain; 4) behavioral
factors, i.e., dietary intake, physical activity, substance abuse, unintended pregnancy, energy
balance; 5) vulnerable populations, i.e., seasonal migrant workers, military, incarcerated women.
A cohort study of 266 Brazilian women aged 18-45 y investigated the association
between gestational weight gain, reproductive factors and postpartum weight retention at 9
months, dichotomized cut-off at 7.5 kg (Kac et al., 2004). Of the total women,19.2% women
had a retention weight ≥ 7.5 kg at 9 months The investigated factors were the mother’s age,
parity number, infant hospitalization, infant birth weight, gestational weight gain categorized in
<9.0 kg, 9.0-11.9 kg, 12.0-15.9 kg, ≥ 16.0 kg. Women whose gestational weight gain was above
16.0 kg were 7.35 times (95% CI=2.01-26.84) more likely to have postpartum weight retention ≥
7.5 kg compared to the gestational weight gain group with < 9.0 kg. Prepregnancy BMI was not
associated with postpartum weight retention at 9 months.
Oken et al. (2009) investigated gestational weight gain to postpartum weight retention at
1 year in 2,012 women in Project Viva. Postpartum weight retention was calculated as the
difference in weight reported by mothers at 1 year following delivery and self-reported

34

prepregnancy weight at study enrollment. They defined substantial postpartum weight retention
as ≥ 5 kg compared with weight retention < 5 kg. Women who had ≥ 5 kg of postpartum weight
retention was 16%. In women who were overweight before pregnancy (BMI 25.0-29.9),
gestational weight gain was associated with risk of postpartum weight retention ≥ 5 kg (OR
=1.35 [1.01-1.81]) and no association in normal and obese preprengancy women.
In both studies (Kac et al., 2004; Oken et al., 2009), gestational weight gain was
positively associated with postpartum weight retention. However, prepregnancy BMI was
associated with postpartum weight retention only in Oken et al. (2009)’s study. It may be due to
the fact that different cut-off points of postpartum weight retention and statistical approaches
were used. Four different categories of gestational weight gain was used in Kac et al. (2004),
whereas gestational weight gain was used as continuous independent variable to see the
association with the outcome variable, postpartum weight retention dichotomized at 5 kg. This
may result in different conclusions.
D. Validity of Self-reported vs. Measured Weight and Height
1. General Populations
Many investigators validated self-reports of heights and weights in reference to measured
heights and weights in various subpopulation groups (Yun et al., 2006; Gilium & Sempos., 2005;
Lee et al., 2011; Kuczmarski et al., 2001; Nyholm et al., 2007). Yun et al. (2006) compared the
2

2

prevalence of obesity (BMI ≥ 30kg/m ) and overweight (BMI ≥ 25kg/m ) in different
demographic groups in the U.S. using the Behavioral Risk Factor Surveillance System (BRFSS)
during 1999-2000 with the National Health and Nutrition Examination Survey (NHANES) 19992000 as the gold standard. BRFSS obtained self-reported height and weight data through a statewide telephone survey, whereas NHANES had actual measured heights and weights of
35

participants. They compared the rank orders of the obesity and overweight prevalence across
different demographic groups including age, gender and race from the two data sources. The
BRFSS’ self-reported data underestimated the overall prevalence of obesity and overweight by
9.5 and 5.7%, respectively, in reference to NHANES’ measured data. The NHANES-BRFSS
differences in obesity and overweight prevalence was highest among non-Hispanic black women
(16.7%), non-Hispanic white men aged 60 or older (15.4%), and non-Hispanic white women
(12.5%); the difference in overweight prevalence was highest among non-Hispanic black men
aged 39 or younger (-15.1%), non-Hispanic black women (11.6%), and non-Hispanic white
women (13.4%) aged 39 or younger, and non-Hispanic white women aged 60 or older (12.8%).
Kuczmarski et al. (2001) compared self-reported heights and weights to measured heights
and weights of adults above 20 y in the third NHANES III. Age was an important factor in
classifying weight, height, BMI and overweight form self-reports. The mean differences for
height ranged from 2.92-4.50 cm for women and from 3.06-4.29 cm for men, 70 y and older,
which differed significantly. Validity assessed by Pearson’s correlation coefficients between
measured and self-reported heights was significant (p<0.001) for all age groups within each
gender. The correlation coefficients in the older age groups (70-79 and 80+ years), were less than
those of the younger age groups, especially for women, which suggests that self-reported height
was less valid in the older subgroups. Men typically overestimated body weight, whereas women
tended to underreport body weight. The BMI values derived from measured weight and height
and self-reported weight and height were highly correlated (r=0.89 to 0.97), and all Pearson
correlation coefficients were statistically significant (p<0.001). The prevalence of overweight
was notably greater when calculated from measured values than from self-reported values for the
2 oldest age groups. Overweight prevalence based on measured values in comparison to self-

36

reported values was 6% greater for men aged 70 to 79 years and 11% greater for men aged 80
years and older. It appears that for younger adults’ prevalence estimates of overweight based on
self-reported heights and weights are similar to those based on measured values.
In a study of 557 men and 1,010 women aged 30 to 70 years (Lee et al., 2011), selfreported height was higher than measured values in both men and women (169.98 (5.53) vs.
169.57 (5.63); 157.95 (4.95) vs. 157.43 (5.05), respectively). Self-reported weight was higher
than measured weight in women (70.96 (8.96) vs. 70.84 (9.03); 57.68 (7.09) vs. 57.42 (7.28), but
was not different in men. BMI calculated from self-reported values was lower than measured
values in both men and women (24.53 (2.62) vs. 24.61 (2.70); 23.13 (2.72) vs. 23.17 (2.82). Men
aged <45 y and women aged 45-49 y showed lowest difference between self-reported BMI and
measured BMI.
Gillum & Sempos, (2005) evaluated the effect of ethnicity to bias of mean BMI
2

and to sensitivity of overweight or obesity (BMI ≥ 25 kg/m ) derived from self-reported vs.
measured weight and height using measured BMI as the gold standard NHANES III is a crosssectional survey of a large national sample, conducted from 1988-1994. Study participants were
American men and women aged 20 years and over (n=15,025). Self-reported height, weight,
cigarette smoking, health status, and socio-demographic variables were obtained from household
interview. Height and weight were measured in a mobile examination center. In women and
Mexican American men, self-reported BMI underestimated true prevalence rates of overweight
or obesity.
Nyholm et al., (2007) conducted a cross-sectional survey to validate self-reported height
and weight and developed a useful algorithm to assess the prevalence of obesity based on selfreported information from 2001 to 2003 using BMI and age. The study included 1,703 men and
37

women of aged 30 to 74 y, in the community of Vara, Sweden. Mean differences of self-reported
minus measured values for weight and height in women were +1.8 kg and -0.4 cm, respectively.
2

Being over 70 years of age and severely obese BMI ≥ 30.0 kg/m were important factors for
misreporting height, weight, and BMI.
Race/ethnicity (Yun et al., 2006; Gullium et al., 2005) and age (Lee et al., 2011;
Kuczmarski et al., 2001; Nyholm et al., 2007) were predictors of differences between selfreports of height and weight and measured height and weight. Different study population
characteristics, social desirability and time lapse between self-reports of measured values might
have influenced the study results.
2. Reproductive Aged Women
Brunner-Huber (2007) examined the validity of self-reported height and weight among a
total of 275 women of reproductive ages of 18-45, who attended a family medicine clinic. On
average, women overestimated their height by 0.1 inches and underestimated their weight by 4.6
pounds. When self-reported height and weight were used to calculate BMI, this translated into a
2

0.8 m/kg underestimation of measured BMI.. Forty percent of Hispanic women and 33% of
non-Hispanic Black women underestimated their weight by > 5 pounds. Differences between
self-reported vs. measured BMI were highest for women with aged >35 y, those with some
college or graduate level education, non-Hispanic Blacks, and divorced, separated, widowed or
single women.
Park et al. (2011) investigated the reliability and validity of mother’s self-reported weight,
height, and BMI by comparing data from Florida birth certificates, with measured values
obtained from the Women, Infants, and Children (WIC) program in the first trimester. Mean

38

(SEM) self-reported weight was 69.9 (0.13) kg from the birth certificate and measured weight in
the first trimester was 71.8 (0.13) kg in the WIC program. Greater discrepancies between selfreported and measured data were observed in women aged 30-39 years, non-Hispanic blacks,
unmarried women, women who used tobacco during pregnancy, or women with weight
measurements taken >6 weeks of gestation. Mean height differences for self-reported height
minus measured height by race/ethnicity and education were statistically significant with a range
from 0.75 to -1.24 cm. Mean BMI differences by age, education, tobacco uses during pregnancy,
and the timing of weight measurements within the first trimester were statistically significant
2

with a range from 0.90 to 1.25 kg/m . Women aged 30-39 years, women with less than a high
school degree, women with tobacco use during pregnancy, or women with weight measurements
taken >6 weeks of gestation had the greatest discrepancy in BMI.
Craig & Adams (2009) examined the strength of agreement between BMI calculated
from self-reported and measured values using Cohen’s Kappa in the study of NHANES 19992004. The study included 724 pregnant women and 5,910 non-pregnant women aged 18 or older.
The goal was to calculate the strength of agreement between self-reported and measured BMI
categories varied significantly according to age and race. Agreement was greatest among white,
non-Hispanic women, age 26-35, and decreased significantly among non-Hispanic Black,
Hispanic, and other racial minorities and women older than 66 years old. They also reported that
pregnancy significantly decreased the strength of agreement and 92.6% of pregnant women
under-reported their weight. However, because these women were pregnant, their measured
weight during pregnancy was higher than their self-reported prepregnancy weight. The authors
compared self-reported prepregnancy weight in reference to measured weight during pregnancy,
which is not appropriate because pregnant women gain weight towards the trimester of
39

pregnancy. Agreement between BMI weight status based on self-reported and measured height
and weight varied significantly by age and race, with less agreement found for older, nonHispanic Black and Hispanic women. The strength of agreement was unrelated to SES, access to
care or health status. Age and race were important determinants impacting on the accuracy of
self-reported height and weight.
Previous studies used different gold references, measured values (Brunner-Huber, 2007;
Craig & Adams, 2009) and measured values in the first trimester of WIC program (Park et al.,
2009) to validate self-reported values in reproductive aged women might result in mixed results.
Race/ethnicity was a determinant for error in self-reported height and weight according to
Bruner-Huber (2007). According to Park et al. (2009), age and education were determinants for
error in self-reported height and weight. According to Craig & Adams (2009), age and
race/ethnicity were determinants for errors in self-reported height and weight. Due to the
inconsistent findings on the effect of maternal demographics on the inaccuracy of self-reported
height and weight in reproductive aged women, understanding the characteristics of inaccurate
reporters is important given the large values of some of the standard deviations. This implies that
for some individuals with particular characteristics of age, race/ethnicity or education, the
estimates are extremely inaccurate, which may bias average results and have important
implications for health planning for reproductive aged women.

40

Chapter Two

VALIDATION OF PREPREGNANCY WEIGHT STATUS ESTIMATED
FROM SELF-REPORTED DATA
A. Abstract
Application of the IOM’s gestational weight gain guidelines during pregnancy requires
data on prepregnancy weight status. The gestational weight gain guidelines are intended to
reduce pregnancy complications, poor birth outcomes as well as excessive postpartum weight
retention. This study aimed to determine the degree of validity of self-reported prepregnancy
weight status reported by pregnant women included in the National Health and Nutrition
Examination Survey (NHANES) 2005-6. The prepregnancy weight status was determined from
the BMI, calculated from self-reported height and weight and the classified into underweight,
normal weight, overweight and obesity. The validity of the prepregnancy weight status classified
based on the BMI calculated from self-reported data was referenced by that based on BMI
calculated from measured height and weight during the first trimester and by the imputed BMI of
279 pregnant women and 489 non-pregnant age-matched child-bearing age women. After finding
the characteristics of pregnant and non-pregnant women whose height differences between selfreported vs. measured or imputed data were above 95% CI, an inference from linear regression
between self-reported and measured weight status of age-comparable non-pregnant women.
Pearson’s correlation (r) between self-reported weight and multiple imputed weight status in
pregnant women was 0.9 (p<0.0001) with mean difference (SEM), -1.5 (0.2) kg. Mean weight
differences (SEM) between self-reported prepregnancy weight and measured weight in the first
trimester was -3.07 (±0.58) with Pearson’s correlation, 0.97 (p<0.0001). Kappa value between
41

self-reported prepregnancy BMI classification and multiple imputed BMI was 0.73 ±0.0001 (p<
0.05). Self-reported prepregnancy weight and height result in valid weight status categories to
apply for the IOM gestational weight gain guidelines in the U.S. representative population.
B. Introduction
Height and weight are of interest in epidemiological studies both as primary factors and
as potential confounding variables. Height and weight are common, straightforward components
of nutritional status assessments because they are strong predictors of functional impairment,
morbidity, and mortality (Lee et al., 2011). Body Mass Index (BMI) is calculated from height
and weight and can be used to assess the association between weight status and health outcomes
in a population. In many settings, direct measurements are not feasible, so height and weight
information is alternately collected via self-report. Although previous studies have concluded
that self-reported height and weight data are reported with acceptable accuracy in general
populations (Kuczmarski et al., 2001; Lee et al., 2011; Nyholm et al., 2007; Yun et al., 2006),
the use of self-reported prepregnancy weight has remained questionable in reproductive aged
women.
Valid categorization of prepregnancy weight status is a prerequisite to adequately
applying the IOM’s gestational weight gain guidelines. Validity of self-reported height and
weight, and the BMI calculated from the self-reported data in pregnant women and non-pregnant
women have been reported by several investigators (Craig & Adams, 2009; Hedderson et al.,
2010; Olson et al., 2003; Park et al., 2011; Cedergren, 2006). Challenges in these research
investigators were the lack of gold standards that could be used in validating the self-reported
data. The gold references that have been used in previously reported studies to validate selfreported prepregnancy weight status, maternal weight recorded at the first visit (Cedergren et al.,
42

2006) and weight measured within the previous 12 months of pregnancy (Hedderson et al., 2010).
vary with measured weight in the first trimester (Olson et al., 2003; Park et al., 2011), Further
challenges include the high prevalence of unplanned pregnancies reported in the U.S. i.e., 49% in
2006 and 48% in 2001 (Finer et al., 2011).
The primary aim of this study was to determine the degree of validity of prepregnancy
weight status classifications determined from the self-reported height and weight by pregnant
women in the U.S. representative population. We also identified socio-demographic
determinants associated with accurate self-reports of height and weight in a U.S. representative
sample of pregnant and non-pregnant women included in the 2005-6 NHANES. The findings can
help pregnant women to better achieve the suggested IOM gestational weight gain guidelines and
monitor the rates of weight gain throughout the different stages of pregnancy. Further, evaluation
of the validity of self-reported prepregnancy weight on a population-based study is important,
because these data are not only widely used for the application for gestational weight gain
guidelines, but also for population-based research, surveillance, and programmatic decision
making.
C. Subjects and Methods
Dataset
NHANES is a dataset that is used to assess the health and nutritional status of a
representative sample of adults and children in the U.S. The survey conducted by the National
Center for Health Statistics (NCHS) combines structured information from interviews and
physical examinations. Data used in this study were from participants in continuous NHANES,
2005-6 which oversampled pregnant women. Beginning in 1999, NHANES has been planned
and conducted as a continuous annual survey and the survey datasets have been released for
43

public use every two-years. During the survey, selected persons were invited to take part in the
survey by being interviewed in their homes. Household interview data were collected via
Computer Assisted Personal Interviewing (CAPI) and includes demographic, socioeconomic,
dietary, and health-related questions. Upon completion of the interview, subjects were asked to
participate in a physical examination, which included body measurements for height and weight.
Clusters of households were selected, then each person in a selected household was screened for
demographic characteristics, and one or more persons per household were selected for the
sample. Appropriate sample weights were applied in all statistical analyses to produce estimates
of means and percentiles that can be generalized to the healthy U.S. population. Reproductive
health, weight history, demographics, and body measurement files were merged for the analysis
in this study.
Study Subjects
All reproductive women aged 18-41y (n=1,474) were eligible for inclusion in this study.
From this eligible sample, subjects were excluded due to the following missing responses:
pregnancy status (n=495), self-reported height (n=54), self-reported weight (n=26), self-reported
weight prior to one year (n=8), family poverty income ratio (n=52), measured height (n=1), and
measured weight (n=58). Women who reported that they were not pregnant, but had a positive
urine test in the demographics dataset were not classified as pregnant (n=12). After the ineligible
respondents were excluded, the final analytic sample consisted of 768 women: 279 pregnant
women who reported pregnant and had positive urine tests and 489 non-pregnant women who
completed both in-home interviews and a medical examination. Self-reported weight one year
prior to the examination was used as a reference to compare self-reported prepregnancy weight
to in the preliminary study.

44

Data analyses
Statistical software. Data preparation was performed using SAS version 9.2 (SAS Institute, Cary,
NC). Because NHANES 2005-6 was conducted in a stratified, multi-stage design probability
design, SAS-callable SUDAAN was used. Appropriate sample weights were applied in all
statistical analyses to produce estimates of means and percentiles that can be generalized to the
healthy adult U.S. population.
Statistical methods.
Validity of self-reported height Self-reported height and weight (proxy measure) was compared
to measured height (gold standard) in both pregnant and non-pregnant women to verify accuracy.
To determine the association between self-reported vs. measured height in pregnant and nonpregnant women separately, Pearson’s correlation was used. To determine if mean self-reported
height was different from mean measured height, general linear regression was used. Mean
height differences between self-reported vs. measured height by maternal demographic
characteristics were examined via the general linear model.
Validity of self-reported weight: Pearson’s correlations were conducted between a) self-reported
prepregnancy weight and imputed weight and b) self-reported prepregnancy weight and
measured weight in the first trimester, when GWG is not expected. Then, to determine if the
amount of difference between a) self-reported prepregnancy weight and imputed weight and b)
self-reported prepregnancy weight and measured weight in the first trimester was related to
maternal demographic characteristics such as age, race/ethnicity, country of birth, family poverty
income ratio, and education levels, general linear regressions were conducted.
Pearson’s correlation was used to find a correlation between self-reported weight and measured
weight. BMI calculated from self-reported height and weight was compared to BMI computed

45

from measured height and weight in non-pregnant women. To determine if the difference
between self-reported weight and measured weight differed by maternal demographic
characteristics such as age, race/ethnicity, country of birth, family poverty income ratio, and
education levels, general linear regressions were conducted.
Validation of self-reported prepregnancy weight status:
The multiple imputations method provides a useful strategy for dealing with data sets with
missing values, in this instance, the measured prepregnancy women’s height and weight. Instead
of filling in a single value for each missing value, the multiple imputation procedure replaces
each missing value with a set of plausible values that represent the uncertainty about the right
value to impute (Rubin, 1987). Multiple imputed inferences were drawn from available selfreported and measured data (gold reference)data of age-matched non-pregnant women for
which , to calculate BMI, in order to validate prepregnancy weight status categories. Kappa is
the probability of agreement adjusted for the probability of agreement at random (Cohen, 1960).
If Kappa equals one, the proxy and the gold standard are in perfect agreement. The strengths of
agreement is interpreted base on Kappa: slight (0.00-0.20), fair (0.21-0.40), moderate (0.41-0.60),
substantial (0.61-0.80) and almost perfect (0.81-1.00) (Landis, 1977).
Self-reported prepregnancy weight status, stratified by cut points of <18.5, 18.5-24.9, 25.0-29.9,
2

≥30 kg/m (IOM, 2009) were compared to imputed weight status divided into same cut points by
Kappa statistics. Self-reported prepregnancy weight status, stratified by cut points of <18.5, 18.52

24.9, 25.0-29.9, ≥30 kg/m (IOM, 2009) were compared to imputed weight status divided into
same cut points by Kappa statistics.

46

D. Results
A total of 279 pregnant and 489 non-pregnant women were included in the study. The
mean age (SEM) was 27.1 ±0.4 years for pregnant women, 32.3 ±0.4 years for non-pregnant
women. The majority of pregnant women were less than 29 years of age (71.3%), non-Hispanic
White (49.5%), and born in the U.S. (74.9%).Education levels and family poverty income ratio
levels were well distributed. The majority of non-pregnant women were either less than 29 years
of age (57.5%), non-Hispanic White (35.6%), or born in the U.S. (77.7%).Education and family
poverty income ratio levels were well distributed.
The mean self-reported height (SEM) of pregnant women was 164.3 (±0.9) cm, and the
mean measured height was 163.7 (±
0.9) cm. The correlation coefficient (r) between the selfreported and measured height of pregnant women was 0.9 (p<0.0001). The correlation
coefficient (r) between self-reported height and measured height was significant in each category
of maternal sociodemographic characteristics, ranged from 0.7 (p=0.0002) in women with
education levels between 0 to 8 years, to 0.9 (p<0.0001) for women of other races, aged between
18-29 and 30-41, born in U.S., with some college education or above, and PIR ≥ 2. Self-reported
height differed significantly from measured height (p<0.0001). Height differences did not
significantly differ across the categories of maternal sociodemographic characteristics (Table 1).
The mean self-reported height (SEM) of non-pregnant women was 163.3 (±0.6) cm and
the mean measured height was 162.8 (±0.5) cm. The mean difference between self-reported
height and measured height was 0.5 (±0.1) cm. The correlation coefficient (r) between the selfreported height and measured height in non-pregnant women was 0.9 (p<0.0001). The
correlation coefficient (r) between self-reported height and measured height was significant in
each category of maternal sociodemographic characteristics, ranged from 0.8 (p<0.0001) in

47

Mexican American or other Hispanic women and women born in Mexico or other countries to
0.9 (p<0.0001) in women in all education levels and family PIR above 1 (Table 1).
The mean self-reported weight (SD) of non-pregnant women was 73.4 (1.2) kg, and the
mean measured weight was 75.0 (1.3) kg. On average, non-pregnant women underestimated their
weight by 1.6 (0.2) kg. The correlation coefficient (r) between self-reported weight and measured
weight was 0.9 (p<0.0001). The correlation coefficient (r) between self-reported weight and
measured weight was significant in each category of maternal sociodemographic characteristics,
ranged from 0.95 (p<0.0001) in women born in Mexico or other countries, to 0.99 (p<0.0001) in
women with education levels between 19 and 21 years. Weight difference between self-reported
weight and measured weight was not significantly different among the categories of maternal
sociodemographic characteristics (Table 2).
A total of 2,790 (original number of pregnant women 279 (* number of imputations=10)
women were included in the study. Mean self-reported weight (SEM) was 70.1 (2.3) kg and
multiple imputed weight was 71.7 (2.4) kg. Overall mean weight difference for self-reported data
versus measured data was -1.5 (0.2) kg, indicating that self-reported weight was lower by an
average of 1.5 kg. Mean weight differences by age, race/ethnicity, country of birth, education,
and family poverty income ratio ranged from -2.26 to 0.36 kg. The mean weight differences for
pregnant women were as follows: -1.6 kg for women 18-29 years , -1.7 kg for non-Hispanic
Blacks women -1.7 kg, -1.6 kg for women born in U.S., The greatest discrepancy was among
women with education levels between15-18 years (-2.1 kg), or with a family PIR<1 (-2.2 kg)
(Table 3).
A total of 47 women were in their first trimester of pregnancy. Self-reported weight
(SEM) was 78.0 (4.9) kg, and measured weight (SEM) was 81.1 (5.1) kg. Overall, mean weight

48

differences was -3.0 (0.5) kg. The greatest discrepancy between self-reported weight and
measured weight in the first trimester was shown in non-Hispanic Black with mean weight
differences, -15.4 (0.6) kg and family PIR<1 with mean weight differences, -7.2 (0.3) kg (Table
4).
Agreement between categories of weight status calculated from self-reported height and
weight and BMI calculated from measured height and weight in non-pregnant women was
significantly high by Kappa estimates (0.84). BMI categories in non-pregnant women
determined by self-reported height and weight, were equivalent to BMI categories based on
measured height and weight, for 83.6% (409/489) of women (Table 5).
Agreement between BMI categories determined by self-reported height and imputed
weight (the gold standard) and BMI categories calculated from self-reported height and
prepregnancy weight (the proxy) in pregnant women was significantly high by Kappa estimates
(0.73). BMI categories in pregnant women determined by self-reported height and weight were
equivalent to BMI categories determined by multiple imputations, for 72.8% (2031/2790) of
women when BMI category was based on multiple imputations as the gold reference (Table 5).
The mean BMI calculated from self-reported height and self-reported weight of non2

pregnant women, 27.50 (0.41) kg/m was lower than the mean BMI estimated from measured
2

height and measured weight, 28.27 (0.47) kg/m . Overall, non-pregnant women under-reported
2

their BMI by 0.77 (0.09) kg/m less than one unit of BMI. The correlation coefficient (r)
between BMI calculated from self-reported height and self-reported weight and BMI calculated
from measured height and measured weight was significant for each category of maternal
sociodemographic characteristics, ranging from 0.94 to 0.98 (p<0.0001) in women with

49

education levels between 9 to 12 years and family PIR 1 to 1.999 in non-Hispanic White women,
education levels 15 to 18 years, 19 to 21 years, and family PIR > 5 with overall correlation
coefficient (r), 0.97 (p<0.0001).
E. Discussion
Optimally, BMI calculations are based on height and weight measured through physical
examination (e.g. balance beam scales and calibrated stadiometers by trained personnel).
Validation of proxy measure is assessed by the level of agreement with the gold standard. In this
study, we used multiple imputations as the gold standard to validate self-reported prepregnancy
weight status because measured prepregnancy weight value is not available. In this study, we
used imputed value as a reference instead of using weight measured in the first prenatal visit
(Olson et al., 2003; Park et al., 2011). However, gestational weight gain may occur during the
first trimester and the first prenatal visit can occur at any stage of pregnancy, between the first to
the third trimester. Imputation, the practice of ‘filling in’ missing data, measured prepregnancy
weight with plausible values, an approach to analyze incomplete data fitted to our study aims to
validate self-reported prepregnancy weight. This increased the accuracy of validation of pregnant
women’s self-reported prepregnancy weight, when comparing measured weight during
pregnancy.
The BMI calculated from self-reported height and weight of non-pregnant women was
2

0.77 (±
0.09) kg/m underestimated compared to the BMI computed from measured height and
weight. Less than one unit of BMI difference may not be significant in clinical terms. This was
confirmed by Brunner-Huber (2007), reproductive aged women under-reported their self2

reported BMI by 0.8 mg/kg in reference to measured BMI. Agreement in weight status
classification (underweight, normal, overweight and obese) by BMI between those calculated
50

from self-reported data vs. measured data was high (kappa = 0.84) in non-pregnant women. Our
result for non-pregnant women was similar to the previous findings of Craig and Adams (2009)
which determined a Kappa value 0.705 for t BMI data determined using self-reported and
measured data were found in non-pregnant women with BMI categorical agreements between
self-reported and measured data were found in 60% (Craig and Adams, 2009) and 76% (Park et
al., 2011) of pregnant women. Similar to the agreement among these pregnant women, the BMI
categorical agreement between prepregnancy weight status categories based on self-reported data
based upon weight status categories based on measured height and imputed weight was 72.8%.
While the first trimester measured weights are not the same as prepregnancy weights,
obtaining actual prepregnancy weight measurements immediately before pregnancy is difficult,
especially because approximately half of all live births in U.S. are unplanned and a majority of
women do not seek care immediately before pregnancy. Average weight gain during the first
trimester was approximately 1 kg among healthy women without eating restrictions (IOM, 2009).
Therefore, in the present study, a mean weight difference of 3.07 kg between self-reported and
measured weight in the first trimester might be partially caused by the actual physiologic weight
change during the first trimester. In ideal situations, child-bearing age women should have
routine visits to physicians’ office where height and weights are measured and the prenatal visit
immediately before pregnancy can yield useful data for calculation and classification of accurate
prepregnancy weight status.
Overall, pregnant women and non-pregnant women who had education levels between 08 years had greatest discrepancy between self-reported and measured height. The greatest
discrepancy between self-reported height and measured height was found for women with
Hispanic ethnicity and education level less than high school (Park et al., 2011). Because there

51

are no readily available actual measured data for prepregnancy weight, actual weight
measurements within the first trimester were considered as reference (the standard) and were
compared with self-reported prepregnancy weight. Pregnant women aged 18-29, non-Hispanic
Black, family PIR <1 had greatest discrepancy between self-reported weight and measured
weight in the first trimester. In the same study, it was found pregnant women aged 30-39 years
and non-Hispanic Black women had the greatest weight discrepancy between self-reported
weight on the birth certificate and measured weight in the first trimester WIC program.
Furthermore, 33% of non-Hispanic Black women underestimated their self-reported weight by >
5 pounds in reference to measured weight (Brunner-Huber, 2007). It was confirmed by previous
studies that non-Hispanic Black women had the greatest differences between self-reported and
measured weight in the first trimester. Differences in groups found with the greatest difference
between self-reported data vs. measured data among age groups, 18-29 years in our study, 30-39
years (Park et al., 2011), >75 years (Craig & Adams, 2009) might be due to the different
distribution of age groups in each study.
The present study provides important information to researchers, epidemiologists, and
policy analysts who use self-reported data for maternal and infant health research and
surveillance purposes. Errors in the measurement of prepregnancy height, weight, and BMI used
for research and surveillance as exposures, outcomes, or confounders, can have significant
effects on the findings.
In summary, overall maternal self-reported prepregnancy mean height, weight, and BMI
were considered valid as determined by the reference data generated by multiple imputations.
Agreement between self-reported data and measured data was high on the basis of Kappa
estimates, Pearson’s correlation and multiple imputations, which indicate self-reported

52

prepregnancy height, weight, and BMI are generally reliable. When recall or reporting bias by
maternal characteristics is considered, self-reported height, weight, and BMI of reproductive
aged non-pregnant women, are generally reliable and valid, for population-based research and
surveillance purposes.

53

Table 1.1 Correlations and mean differences between self-reported and measured height by
sociodemographics of pregnant and non-pregnant women
n
Self-reported Measured
Pearson r
Differences
p-value
Pregnant
Women
a
All
279
164.3 (0.9)
163.7 (0.9)
0.9***
0.6 (0.2)
***
Age
18-29

199

163.1 (1.1)

162.7 (1.1)

0.9***

0.7 (0.2)

80

166.3 (1.5)

165.7 (1.4)

0.9***

Race/Ethnicity
MA or O H

87

160.4 (0.9)

159.4 (0.9)

0.8***

1.0 (0.4)

NHW
NHB
Other race

138
38
16

165.9 (0.8)
162.1 (1.2)
162.9 (5.1)

165.5 (0.7)
161.6 (1.1)
160.9 (5.7)

0.9***
0.8***
0.9***

0.4 (0.2)
0.4 (0.3)
1.9 (0.6)

209

165.0 (0.8)

164.5 (0.7)

0.9***

0.4 (0.1)

70

160.6 (0.9)

159.3 (0.9)

0.8***

1.3 (0.3)

Education
0-8

20

159.2 (0.5)

157.7 (0.6)

0.7**

1.5 (0.9)

9-12
13-14
15-18
19-21

53
61
81
64

159.4 (1.6)
161.8 (1.2)
165.0 (1.3)
167.6 (1.2)

158.4 (2.0)
161.7 (1.4)
164.6 (1.0)
166.8 (1.1)

0.8***
0.9***
0.9***
0.9***

0.9 (0.8)
0.1 (0.3)
0.4 (0.4)
0.8 (0.3)

72

161.4 (1.3)

161.1 (1.1)

0.8***

0.3 (0.5)

60
39
60
48

162.2 (1.4)
166.8 (1.1)
164.5 (1.1)
166.4 (0.5)

161.7 (1.5)
165.7 (1.2)
163.5 (1.0)
166.0 (0.7)

0.8***
0.9***
0.9***
0.9***

0.4 (0.3)
1.1 (0.1)
0.9 (0.1)
0.4 (0.3)

b

0.6 (0.3)

30-41

Country of Birth
United States
M or other

Family PIR
PIR <1
1=<PIR<1.9
2=<PIR<2.9
3=<PIR<4.9
5=<PIR

54

0.9

0.6

0.1

0.8

0.4

b

b

b

b

Table 1.1 (con’t)
n

Self-reported

Measured

Pearson r

Differences

p-value

Non-Pregnant
Women
All

489

163.3 (0.6)

162.8 (0.5)

0.9***

0.5 (0.1)

***

Age
18-29

208

163.0 (0.7)

162.6 (0.6)

0.9***

0.4 (0.1)

0.6

30-41

281

163.5 (0.7)

162.9 (0.6)

0.9***

0.5 (0.2)

Race/Ethnicity
MA or O H

146

159.1 (0.8)

157.9 (0.7)

0.8***

1.1 (0.3)

NHW
NHB
Other race

174
140
29

164.8 (0.6)
169.9 (0.5)
159.7 (1.5)

164.3 (0.4)
163.5 (0.5)
159.4 (1.3)

0.9***
0.9***
0.9***

0.4 (0.2)
0.4 (0.1)
0.3 (0.5)

380

164.1 (0.6)

163.8 (0.5)

0.9***

0.3 (0.1)

109

159.7 (1.1)

158.3 (1.0)

0.8***

1.3 (0.3)

Education
0-8

25

157.4 (1.3)

155.8 (0.9)

0.9***

1.6 (0.5)

9-12
13-14
15-18
19-21

86
132
173
73

162.3 (0.9)
162.5 (1.2)
164.5 (0.7)
164.0 (0.8)

161.5 (0.8)
162.0 (1.0)
164.0 (0.6)
163.8 (0.7)

0.9***
0.9***
0.9***
0.9***

0.8 (0.3)
0.5 (0.2)
0.5 (0.1)
0.1 (0.2)

131

162.4 (0.8)

161.4 (0.8)

0.8***

1.0 (0.3)

Country of Birth
United States
M or other

Family PIR
PIR <1

0.2

0.03

0.03

0.02

1≤PIR<1.9
119
162.8 (0.8)
162.3 (0.8)
0.9***
0.5 (0.2)
2≤PIR<2.9
71
161.5 (1.2)
161.2 (1.1)
0.9***
0.2 (0.1)
3≤PIR<4.9
109
165.0 (1.0)
164.6 (0.9)
0.9***
0.4 (0.3)
5≤PIR
59
163.8 (0.7)
163.4 (0.7)
0.9***
0.3 (0.2)
Mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics
*** p<0.0001
*** p<0.05
MA or O H = Mexican American or other Hispanics
NHW = Non-Hispanic White
NHB = Non-Hispanic Black
55

a

b

b

b

b

b

Table 1.2 Correlations and mean differences between self-reported weights in non-pregnant
women by sociodemographics
Non-Pregnant
n
SelfMeasured
Pearson r
Difference
p-value
Women
reported
All

489

73.4 (1.2)

75.0 (1.3)

0.9***

-1.6 (0.3)

***

Age
18-29

208

70.2 (1.5)

71.8 (1.7)

0.9***

-1.5 (0.3)

0.8

30-41

281

74.8 (1.4)

76.4 (1.5)

0.9***

-1.6 (0.2)

Race/Ethnicity
MA or O H

146

71.0 (1.3)

72.5 (1.4)

0.9***

-1.5 (0.3)

NHW
NHB
Other race

174
140
29

73.1 (1.8)
81.9 (2.0)
63.8 (2.5)

74.8 (2.0)
83.3 (2.5)
65.5 (2.6)

0.9***
0.9***
0.9***

-1.6 (0.3)
-1.4 (0.8)
-1.6 (0.3)

380

75.1 (1.2)

76.7 (1.4)

0.9***

-1.6 (0.3)

109

65.5 (1.1)

66.9 (1.2)

0.9***

-1.3 (0.3)

Education
0-8

25

66.4 (2.2)

67.4 (2.6)

0.9***

-1.0 (0.4)

9-12
13-14
15-18
19-21

86
132
173
73

76.6 (2.0)
74.2 (2.3)
76.3 (2.0)
66.8 (2.6)

78.1 (1.7)
76.5 (2.5)
77.9 (2.1)
67.5 (2.7)

0.9***
0.9***
0.9***
0.9***

-1.4 (1.0)
-2.3 (0.4)
-1.6 (0.3)
-0.7 (0.3)

131

74.8 (1.9)

76.4 (2.2)

0.9***

-1.6 (0.5)

a

Country of Birth
United States
M or other

Family PIR
PIR <1

119
78.5 (2.2)
80.3 (2.3)
0.9***
-1.8 (0.7)
1≤PIR<1.9
71
72.6 (2.7)
74.3 (3.1)
0.9***
-1.7 (0.7)
2≤PIR<2.9
109
73.2 (1.9)
75.0 (2.0)
0.9***
-1.8 (0.2)
3≤PIR<4.9
59
66.9 (2.3)
67.7 (2.4)
0.9***
-0.7 (0.2)
5≤PIR
Mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics
*** p<0.0001; ** p<0.05
56

0.9

0.6

0.2

0.4

b

b

b

b

b

Table 1.3 Correlations and mean differences between self-reported and multiple imputed (MI)
weight of pregnant women by sociodemographics
n
SelfMultiple
Pearson r
Mean
p-value
reported
Imputed
Differences
a

All

2790

70.1 (2.3)

71.7 (2.4)

0.9***

-1.5 (0.2)

***

Age
18-29

1990

67.9 (2.3)

69.5 (2.4)

0.9***

-1.6 (0.3)

0.7

30-41

800

74.7 (3.9)

76.2 (4.0)

0.9***

-1.4 (0.2)

Race/Ethnicity
MA or O H

870

67.4 (1.1)

68.6 (1.4)

0.9***

-1.2 (0.5)

NHW
NHB
Other race

1380
380
160

69.8 (2.0)
75.9 (7.0)
70.6 (15.1)

71.5 (2.1)
77.7 (7.1)
71.8 (15.7)

0.9***
0.9***
0.9***

-1.7 (0.2)
-1.7 (0.4)
-1.2 (0.5)

2090

71.5 (2.6)

73.1 (2.6)

0.9***

-1.6 (0.2)

700

63.7 (2.6)

65.1 (3.0)

0.9***

-1.3 (0.6)

Education
0-8

200

57.1 (3.1)

56.7 (3.1)

0.8***

0.3 (0.6)

9-12
13-14
15-18
19-21

530
610
810
640

65.9 (3.6)
70.7 (2.2)
73.7 (4.2)
70.2 (3.9)

67.7 (3.8)
72.6 (2.4)
75.9 (4.0)
71.3 (4.0)

0.9***
0.9***
0.9***
0.9***

-1.8 (0.6)
-1.8 (0.4)
-2.1 (0.3)
-1.1 (0.4)

720

71.8 (5.0)

74.1 (5.1)

0.9***

-2.2 (0.4)

Country of Birth
United States
M or other

Family PIR
PIR <1

600
66.8 (2.0)
68.9 (2.3)
0.9***
-2.1 (0.4)
1=<PIR<1.9
390
74.6 (3.8)
76.2 (4.1)
0.9***
-1.6 (0.3)
2=<PIR<2.9
600
73.9 (3.4)
74.8 (3.5)
0.9***
-0.8 (0.3)
3=<PIR<4.9
480
63.5 (2.4)
64.6 (2.6)
0.8***
-1.1 (0.4)
5=<PIR
Mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics
*** p<0.0001; ** p<0.05
57

b

b

0.8

b

0.7

b

0.8

0.03

b

Table 1.4 Correlations and mean difference between self-reported and measured weight by
st
sociodemographics in pregnant women in 1 trimester
Weight (kg)
n
Self-reported
Measured
Pearson
Mean
pWt
Wt
r
differences value
All
47
78.0 (4.9)
81.1 (5.1)
0.9***
-3.0 (0.5) ***a
Age
18-29

37

75.5 (5.0)

79.4 (5.6)

0.9***

-3.8 (0.6)

30-41
p-value
Race/Ethnicity
MA or O H

10

86.1 (9.8)

86.6 (10.0)

0.9***

12

62.4 (.)

63.5 (.)

0.9***

-1.0 (.)

NHW
NHB
Other race
p-value
Country of Birth
United States

23
8
4

80.8 (4.6)
106.8 (5.4)
53.6 (.)

83.4 (4.5)
122.2 (6.1)
52.7 (.)

0.9***
0.9***
0.9***

-2.5 (0.5)
-15.4 (0.6)
0.9 (.)

39

81.8 (4.2)

85.4 (4.1)

0.9***

-3.5 (0.5)

M or other
p-value
Education
0-8

8

56.3 (.)

56.6 (.)

0.9***

-0.3 (.)

2

72.5 (.)

75.8 (.)

9-12
13-14
15-18
19-21
p-value
Family PIR
PIR <1

12
13
11
9

72.0 (.)
68.1 (0.8)
79.8 (4.1)
85.4 (3.9)

76.9 (.)
74.5 (1.6)
83.1 (4.8)
85.5 (3.7)

0.9***
0.9***
0.9***
0.9***

-4.9 (.)
-6.3 (0.8)
-3.2 (0.9)
-0.1 (0.2)

16

81.0 (2.7)

88.3 (3.1)

0.9***

-7.2 (0.3)

11
6
5
9

62.5 (6.5)
95.2 (.)
83.3 (.)
64.4 (0.6)

64.5 (6.3)
94.5 (.)
90.6 (.)
66.1 (0.6)

0.9***
0.9***
0.9***
0.9***

-2.0 (0.2)
0.6 (.)
-7.2 (.)
-1.1 (0.1)

b

-0.4 (0.3)

1=<PIR<1.9
2=<PIR<2.9
3=<PIR<4.9
5=<PIR
p-value

.

-3.2 (.)

0.1

0.7

0.1

0.2

0.6

Mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics
*** p<0.0001
** p<0.05
58

b

b

b

b

Table 1.5 Agreement on self-reported and measured BMI categories in non-pregnant women and
self-reported and multiple imputed BMI categories in pregnant women
Self-reported
Measured BMI
Total
Kappa
BMI
categories
Underweight Normal
Overweight
Obese
Underweight
13 (n)
4
0
0
17
0.84
809302(wt n)
100229
0
0
909532
Normal

4
78155

143
7562348

28
1214899

1
20984

176
8876385

Overweight

0
0

12
667548

95
4269967

27
1253761

134
6191276

Obese

0
0

0
0

4
296148

158
6634891

162
6931038

159
127
8330125
5781013
Imputed BMI

186
7909636

Total
Self-reported
BMI
category
Underweight

Normal

17
887457

Underweight Normal
Overweight
Obese
56 (n)
34
0
409355 (wt
309585
0
n)

489
22910007
Total
Kappa

0
0

90
718941

127
1794646

1090
13750007

297
3835733

6
107388

1520
19490007

Overweight

0
0

96
718483

452
4079752

172
1712985

720
6511220

Obese

0
0

0
0

27
693724

433
7867886

460
8561611

183
2204001

1220
14780007

776
8609210

611
9688259

2790
35280007

Total

59

0.73

Chapter Three

HOW IS DIET QUALITY DURING PREGNANCY ASSOCIATED WITH
GESTATIONAL WEIGHT GAIN?

A. Abstract
Two thirds of pregnant women gain more weight than what is recommended by the IOM
(Schieve, 1998). Diet quality and physical activity have been recognized as important
contributing factors in achieving optimal gestational weight gain (GWG).Both GWG and Dietary
References Intakes (DRIs) and Dietary References Intakes for pregnant women aim to reduce
adverse pregnancy complications and optimize birth outcomes, and related to dietary intake.
DRIs for pregnancy underscore dietary quality and nutritional adequacy. However, GWG is
based on the prepregnancy weight status and DRIs is by trimester. We aimed to test the
hypothesis that diet quality during pregnancy measured by daily nutrient intakes, nutrient density,
and Healthy Eating index (HEI) 2005 are good predictors for overall GWG cross-sectionally.
Data are from 238 pregnant women in NHANES 2005-6, which oversampled pregnant women.
Dietary intake was the average of two 24-hour recalls. Each diet quality indices were divided
into three GWG and compared within the groups of GWG. Logistic regression and linear
regression models were used to determine if diet quality indicators predict adequate GWG at
different trimesters of pregnancy. Adequate GWG was predicted (p<0.05) by nutrient density of
cholesterol, total choline and selenium, and daily nutrient adequacy for vitamin B2, vitamin B6
and phosphorous. However, scores for HEI 2005 did not differ by GWG. In conclusion, daily
intake of several nutrients, dark green and orange vegetables and legumes, total grains and whole
grains were modifiable factors to achieve adequate GWG.
60

B. Introduction
The GWG guidelines are intended to minimize the risk of adversary pregnancy outcomes
for both infants and mothers. Women with excessive or outside of the recommended ranges of
GWG have been documented to have increased risk for cesarean delivery, preeclampsia and
large-for-gestational age (LGA) infants (Cedergren, 2006). Previous studies have shown that
most pregnant women gain weight above the recommended gestational weight gain guidelines
(Lederman et al., 2002). The ability to properly adjust dietary patterns during pregnancy may
also be a determinant of appropriate weight gain during pregnancy (Deierlein et al., 2008),
although advice from health providers (Stotland et al., 2005), psychosocial factors (Strychar et
al., 2000), and physical activity and smoking (Olson et al., 2003) have also been found to
influence gestational weight gain.
Both adequate GWG and good nutrition are important modifiable predictors during
pregnancy. Maternal nutrition is a key factor influencing gestational weight gain (Chapin et al.,
2004; Deierlein et al., 2008; Olafsdottir et al., 2006; Tsigga et al., 2011), and hence the health of
women during pregnancy and postpartum (Anderson, 2001), adverse pregnancy and birth
outcomes. Previous research reported inadequate nutrient intakes of folate and iron during first
trimester of pregnancy, not including vitamins or supplements (Rifas-Shiman et al., 2009);
inadequate intake of calcium, vitamin D, iron, folate during second and third trimesters (Pick et
al., 2005); inadequate intake of vitamin D, folate, calcium, iron, phosphorous throughout the
pregnancy (Fowles, 2002) in pregnant women in U.S. However, none of these studies included
use of vitamin and mineral supplements as part of the nutritional assessment. Due to the different
socioeconomic status and trimester of pregnancy of the participants included in these studies, the
results were inconsistent with limited reports on overall diet quality of pregnant women (Pick et

61

al., 2005; Tsigga et al., 2011). The HEI 2005 serves as a useful tool for comparing diets with the
Food Guide Pyramid, but HEI 2005 does not discern the needs for micronutrients (Pick et al.,
2005). Diet quality has been assessed by HEI 2005 in Greeks pregnant women (Tsigga et al.,
2011). To date, no studies have reported diet quality during pregnancy as a determinant of GWG.
We aimed to test the hypothesis that diet quality during pregnancy measured by daily
nutrient intake, nutrient density, and the HEI 2005 predicts adequate gestational weight gain
cross-sectionally. Data are from 238 pregnant women in NHANES 2005-6. Dietary intake was
the average of two 24-hour recalls. Each diet quality indices were divided into three groups of
GWG and compared among three groups. Logistic regression and linear regression models were
used to determine if diet quality indicators predict adequate gestational weight gain at different
trimesters of pregnancy.
C. Subjects and Methods
Dataset
Data used in this study were from participants in continuous NHANES, 2005-06.
Beginning 1999, NHANES has been planned and conducted as a continuous annual survey and
the survey datasets have been released for public use every two-year increments. During the
survey, selected persons were invited to take part in the survey by being interviewed in their
homes. Household interview data were collected via Computer Assisted Personal Interviewing
(CAPI) and include demographic, socioeconomic, dietary, and health-related questions. Upon
completion of the interview, sample persons were asked to participate in physical examination.
The examination included body measurements for height and weight (US DHHS, 2005). Clusters
of households are selected, each person in the selected household is screened for demographic
characteristics, and one or more persons per household are selected for the sample. Appropriate
62

sample weights were applied in all statistical analyses to produce estimates of means and
percentiles that could be generalized to the healthy adult U.S. population. The sample for the
survey NHANES 2005-6 is selected to represent the U.S. population of all ages. Reproductive
health, individual foods day 1 and day 2, total nutrient intakes day 1 and day 2, weight history,
demographics, and body measurement file were merged for the analysis.
Study Subjects
All pregnant women were eligible for inclusion (n=349) in this study. From this eligible
sample, subjects were excluded were: those aged less than 18 years (n=7) and subjects who did
not respond self-reported prepregnancy height and weight, measured height and weight, country
of birth, education level, family poverty income ratio, self-reported pregnancy status, month of
pregnancy (n=69) or day 1 24-hour recall diet information (n=35). The final analytic sample in
this study consisted of 238 pregnant women, aged between 18-43 y, which completed both the
interview questionnaire and the medical examination.
Data analyses
Statistical software. Data preparation was performed using SAS version 9.1 (SAS Institute, Cary,
NC). Because NHANES 2005-6 was conducted in a stratified, multi-stage design probability
design, SAS-callable SUDAAN was performed. Appropriate sample weights were applied in all
statistical analyses to produce estimates of means and percentiles that can be generalized to the
healthy adult U.S. population.
GWG groups were categorized into three groups following the recommendation (IOM,
2009): 1) Inadequate GWG group for those who gained less than the minimum weight of the
recommendation; 2) adequate GWG group for those who gained between the minimum and
maximum weight of recommendation; 3) excess GWG group for those who gained above the

63

maximum weight of recommendation. The one-way Anova test was used to determine if three
different categories of GWG group had significant differences in nutrient intakes or density per
1,000 kcal for of protein, carbohydrate, dietary fiber, fat, niacin, total folate, iron, magnesium
and zinc. Chi-square tests were performed to evaluate if percentage of women whose daily
nutrient intake meet RDA or AI differ among GWG groups. Differences in the HEI 2005 score
among three GWG groups were tested by one-way Anova. To test if nutrient adequacy predicts
GWG categories, polytomous logistic regression was performed for three-level outcome
variables, i.e., inadequate; adequate; excessive GWG groups. In the polytomous logistic
regression, adequate GWG group was set as the reference in comparison with the combined
inadequate or excessive levels. The odds ratio was calculated by comparing inadequate or
excessive GWG groups to adequate GWG group in assessing the effect of HEI 2005 on GWG
groups in adjusted model controlling for maternal age, race/ethnicity, country of birth, family
poverty income ratio, education levels, and physical activity.
Dietary Assessment
Detailed dietary interview was conducted to obtain dietary intake information from
NHANES participants. The dietary intake data contains the types and amounts of food and
beverages consumed during the 24-hour period prior to the interview (midnight to midnight) in
order, be able to estimate intakes of energy, nutrients, and other food components from those
foods and beverages. The first dietary recall interview was collected in-person in the Mobile
Examination Center (MEC) and the second interview was collected by telephone 3 to 10 days
later, both of which dietary recalls done with a multiple-pass, automated data collection system.
To assess diet quality during pregnancy, total nutrient intakes derived from the average of two
24-hour dietary recalls was used. Total nutrient intakes files reflect only nutrients obtained from

64

foods, beverages, and water including tap and bottled water. Nutrients obtained from dietary
supplements, antacids, or medications were not included in this analysis.
Diet quality was assessed by four methods: 1) Total daily nutrient intakes from foods and
beverages; 2) Nutrient density calculated by the sum of daily intake for each nutrient per 1,000
kcal; 3) Nutritional adequacy assessed by the sum of daily intake from 24-hour recalls and
compared to the Recommended Dietary Allowances (RDA) or Adequate Intakes (AI) for each
nutrient by individual and then by aggregated subgroup of the pregnant women. RDAs or AIs
were used for protein, carbohydrate, dietary fiber, niacin, total folate, iron, magnesium, and zinc;
4) HEI 2005 on 12 components including total fruit, whole fruit, total vegetables, dark green and
orange vegetables and legumes, total grains, whole grains, milk, meat and beans, oils, saturated
fat, sodium, and calories from solid fat, alcohol, and added sugar (SoFAAS) for each individual
pregnant woman.
The HEI 2005
The HEI was developed by the US Department of Agriculture (USDA) “to provide a
single summary of diet quality based on different aspects of a healthy diet” (Kennedy et al.,
1995). This tool was developed to measure how well the diet adhered to the recommendations of
the Dietary Guidelines for Americans and to assess diet quality using a single composite number
(US Department of Agriculture & US Department of Health and Human Services, 2000). The
HEI was based on foods rather than nutrients, and was designed to measure diet quality as
compared with U.S. national guidelines (Pick et al., 2005). MyPyramid Equivalents Database for
USDA Survey Food Codes and CNPP, MyPyramid Equivalent Database for whole fruit and fruit
juice, individual food intake data in 1 day, individual nutrient intake data in 1 day were
combined to calculate HEI 2005 score for each individual. The standards were created using a

65

density approach, expressed as a percent of calories or per 1,000 calories. The four major
strengths of the HEI-2005 are (1) capturing the key recommendation regarding diet quality found
in the 2005 Dietary Guidelines for Americans, (2) assessing diets on a per 1,000 calorie basis in
order to characterize diet quality while controlling for diet quantity, (3) addressing the
consumption of energy-dense, nutrient-poor foods and ingredients, and (4) emphasizing those
aspects of the American diet that are further from current recommendations: whole fruit, dark
green vegetables, orange vegetables, legumes, whole grains, sodium, and discretionary calories
(Guenther et al., 2008).
Prepregnancy BMI and GWG groups
Prepregnancy BMI was calculated by dividing the self-reported prepregnancy weight in
pounds by the self-reported prepregnancy height in inches squared and then multiplying by 703.
GWG at month of gestation was calculated by subtracting self-reported prepregnancy weight
from measured weight at the specific month of pregnancy (1-10 months). Month of pregnancy
was obtained by the question, “what month of pregnancy are you in?” Adequacy in the rate of
GWG were assessed following IOM’s 2009 GWG guideline based on self-reported prepregnancy
BMI group (i.e., underweight, normal, overweight, obese)for specific month of pregnancy
(Table 1).

66

Table 1. Recommended GWG by prepregnancy BMI and month of pregnancy (IOM, 2009)
Month
1
2
3
4
5
6
7
8
9
10
Underweight
Min
0.4 1.2
2.2
5
9
13
17
20
24
28
BMI<18.5
Max
1.2 3.6
6.6
10
15
20
25
30
35
40
Normal
Min
0.4 1.2
2.2
5
8
11
15
18
22
25
BMI 18.5-24.9 Max
1.2 3.6
6.6
10
14
18
22
27
31
35
Overweight
Min
0.4 1.2
2.2
4
6
7
9
11
13
15
BMI 25.0-29.9 Max
1.2 3.6
6.6
9
11
14
17
20
22
25
Obese
Min
0.2 0.6
1.1
2
4
5
7
8
10
11
BMI >30
Max
0.8 2.4
4.4
6
8
11
13
15
18
20
Source: N.C. Department of Health and Human Services; Women’s and Children’s Health
Section. Prenatal Weight Gain Chart. Adapted from Institute of Medicine, 2009. Weight Gain
During Pregnancy: Reexamining the Guidelines. Washington, DC. National Academies Press;
Committee to Reexamine IOM Pregnancy Guidelines.
Prepregnancy BMI

Physical Activity Levels
An MET (metabolic equivalent) score was used as a measure for physical activity and
calculated by summing the MET score for each activity engaged in by an individual based on
responses provided as part of the Physical Activity Questionnaire. The MET score of each
activity was multiplied by the reported frequency and duration and then summed and divided by
30 before being multiplied by 7 to obtain the total MET-minutes per week. MET refers to
metabolic equivalent, and 1 MET is the rate of energy expenditure while sitting at rest. It is taken
by convention to be an oxygen uptake of 3.5 milliliters per kilogram of body weight per minute.
Physical activities frequently are classified by their intensity using the MET as a reference. For
the analysis to see adjust for physical activity level to see association with GWG, METminutes/week as continuous variable was used.

67

METs and MET-minutes
A well-known physiologic effect of physical activity is that it expends energy. A
metabolic equivalent, or MET is a unit useful for describing the energy expenditure of a specific
activity. A MET is the ratio of the rate of energy expended during an activity to the rate of
energy expended at rest. For example, 1 MET is the rate of energy expenditure while at rest. A 4
MET activity expends 4 times the energy used by the body at rest. If a person does a 4 MET
activity for 30 minutes, he or she has done 4 × 30 = 120 MET-minutes (or 2.0 MET-hours) of
physical activity. A person could also achieve 120 MET-minutes by doing an 8 MET activity for
15 minutes.

68

D. Results
Two hundred thirty eight pregnant women were included in the study. The demographics
of the groups are provided in Table 1. The pregnant women were most likely to be younger (1829 y), White, non-Hispanic, had some college or have an Associate’s degree or above, born in
the U.S. The family poverty income ratio (PIR) evenly distributed among the participants, most
had more than one previous pregnancies, and normal prepregnancy weight status (BMI 18.5-25
2

kg/m ). There was significant association between trimester of pregnancy and excessive GWG
(p=0.0011). The majority of women (83%) were in between the 2

nd

rd

and 3 trimester. The

majority of women in the first trimester gained inadequate GWG and the majority of women in
the second and third trimester gained excessive GWG. More women (75.0%) in the third
trimester were in excessive gestational weight group compared to women in their first trimester
(28.5%) or second trimester (59.6%). Prepregnancy weight status was not significantly
associated with each trimester of pregnancy. The distribution of the three GWG groups was not
significantly associated with prepregnancy weight status. Pregnant women who were in normal
(45.8%) or obese (43.8%) prepregnancy weight status were categorized as inadequate GWG. In
adequate or excessive GWG, most pregnant women had normal prepregnancy weight status
(64.6%, 54.7%, respectively) (Table 2).
Nutrient density per 1,000 kcal significantly differed by GWG for cholesterol, total
choline and selenium. Cholesterol intakes was lowest in adequate GWG compared to inadequate
or excessive GWG (p=0.032). Total choline and selenium intakes were highest compared to
inadequate or excessive GWG (p=0.023, p=0.027, respectively) (Table 3). There was significant
association between nutrient below or above RDA or AI for Vitamin B2, Vitamin B6,
Phosphorous and Selenium by GWG groups when average of day 1 and day 2 24-hour recall is
69

included (p=0.015, p=0.034, p=0.010, p<0.0001, respectively). Problematic nutrients whose
nutrient intake is less than the recommendation of RDA or AI shown in over 50% of pregnant
women were fiber, folate, iron, magnesium and potassium (Table 4).
Their Healthy Eating Index 2005 total score and each 12 components did not differ by
GWG. HEI 2005 total score was 61.7 in adequate GWG, 55.0 in inadequate GWG, and 54.0 in
excessive GWG. When inadequate and excessive GWGs were combined and compared with
adequate GWG, HEI 2005 total score and each of the 12 components did not differ by two GWG.
The HEI 2005 total score was 61.7 for women who had adequate GWG and 54.2 for women who
had inadequate or excessive GWG (Table 5). Logistic regression was performed on the effect of
serving sizes of HEI 2005 components to inadequate or excessive GWG in comparison with
adequate GWG adjusted for maternal age, race/ethnicity, country of birth, family poverty income
ratio and education levels. Pregnant women whose dark green and orange vegetables and
legumes intake was less than 0.4 cup equivalent per 1,000 kcal compared to above 0.4 cup were
4.7 times more likely to be in inadequate or excessive GWG compared to those had adequate
GWG (p=0.046). Those whose total grain intake was less than 3.0 cup equivalent per 1,000 kcal
compared to above 3.0 cup were 0.5 times less likely in inadequate or excessive GWG compared
to adequate GWG (p=0.039). Those whose whole grain intake less than 1.5 cup equivalent per
1,000 kcal compared to above 1.5 cup were 0.1 times less likely to be in the inadequate or
excessive GWG group compared to the adequate GWG group (p=0.023) (Table 6).
Logistic regression was performed to examine the association between serving sizes of
HEI 2005 components to inadequate or excessive GWG as a reference to adequate GWG
controlled for maternal age, race/ethnicity, country of birth, family poverty income ratio,
education levels, physical activity levels and prepregnancy weight status. Pregnant women

70

whose dark green and orange vegetables and legumes intake was less than 0.4 cup equivalent per
1,000 kcal compared to those whose intake was above 0.4 cup were 8.4 (1.8-38.5) times more
likely to be in inadequate or excessive GWG compared to those in adequate GWG (p=0.01).
Those whose whole grain intake less than 1.5 cup equivalent per 1,000 kcal compared to those
whose intake was above 1.5 cup were 0.1 (0.03-0.81) times less likely to be in the inadequate or
excessive GWG group compared to the adequate GWG group (p=0.03). Pregnant women whose
calories from solid fat, alcohol and added sugar (SoFAAS) above 20% compared to those whose
intake was less than 20% were 7.8 (1.6-38.7) times more likely to be in inadequate or excessive
GWG rather than adequate GWG (p=0.01) (Table 7). In both logistic models, dark green, orange
vegetables and legumes and whole grain intake were associated with GWG. However, when
controlled for physical activity and prepregnancy weight status, calories from SoFAAS was
associated with GWG in addition to dark green and orange vegetables and legumes and whole
grain.
E. Discussion
Total grain or whole grain intakes were positively related to GWG, whereas dark green
and orange vegetables and legumes intakes were inversely related to GWG. Our study showed
that the HEI score for pregnant women was 55.9, whereas the HEI scores for pregnant women
were 75.0 ±0.99 (Pick et al., 2005) and 66.9 ±7.6 (Tsigga et al., 2011). Further, HEI 2005 score
from this study was lower than the mean HEI 1995 score for women ages 19 to 50 years
(HEI=61.9) in the USDA population survey from 1994-1996, diets with an HEI score of 80 or
more were considered good; those scoring between 50 and 80 needed improvement (Basiotis et
al., 2002). Our results showed that pregnant women had mean HEI 2005 score which indicate
there needs improvement for healthy diet. Two percentages of pregnant women had HEI score
71

>80, and 65.5% of pregnant women consumed diets needing improvement, HEI score 50-80.
Only 32.5% of pregnant women had diets considered poor, HEI score<50. Pick et al. (2005)
included only 52 pregnant women in 20-38 weeks of gestation with high education and
household income levels using 4-day diet records excluding supplements and minerals. They
calculate the previous version of composed of 10 components, grains, vegetables, fruits, milk,
meat, total fat (%), saturated fat (%), cholesterol (mg), sodium (mg), and variety. HEI 2005
reflects the 2005 Dietary Guidelines for Americans in comparison to HEI 1995. Tssiga et al.
(2010) included 100 Greek pregnant women in the first trimester (n=50), in the second trimester
(n=32) and in the third trimester (n=18) with the majority women having normal prepregnancy
weight status (n=62). Differences in prepregnancy weight status of study population, geographic
location, HEI 1995 vs. HEI 2005 and socioeconomoic status in each study may influence in
inconsistencies in HEI 2005 score.
HEI 2005 was used to assess diet quality in this study. Laraia et al. (2007) used Diet
Quality Index for Pregnancy which reflected 2000 Dietary Guidelines for Americans and the
Food Guide Pyramid, nutrients particularly important for pregnancy such as folate, iron and
calcium, and meal and snack patterning from foods only. Rifas-Shiman et al. (2009) developed
alternate HEI, modified for pregnancy to measure diet quality on nine components, vegetables,
fruit, ratio of white to red meat, fiber, trans fat, ratio of polyunsaturated to saturated fatty acids,
folate, iron and calcium from foods, not including vitamins or supplements. However, folate,
iron and calcium intake may compensate from vitamins or supplements if not adequately
consumed from diets. HEI 2005 is developed to measure diet quality on foods components so
this may provide more accurate information in assessing diet compared to DQI-P (Laraia et al.,
2007) or alternate HEI (Rifas-Shiman et al., 2009)

72

In our study, more than half of pregnant women’s diet did not meet the RDA for folate,
calcium, magnesium, and iron, concern of micronutrients during pregnancy. This finding was
confirmed by previous studies (Fowles, 2002; Pick et al., 2005; Rifas-Shiman et al., 2009) which
showed that pregnant women consumed less than the recommended daily intake of
micronutrients, particularly calcium, iron and folate. This finding suggested that nutritional
supplementation during pregnancy is highly recommended. Ideally, all child-bearing age young
women consume a variety of foods for optimal nutrition during perinatal period, in preparation
for a healthy pregnancy and baby (Pick et al., 2005). The low compliance to the nutrition
recommendations during pregnancy may partly due to the fact that majority of pregnancies are
unplanned; thus women do not have adequate time to adapt to the new recommendations
promptly (Inskip et al., 2009).
Pregnant women were inadequate in folate, iron, magnesium and potassium intake
compared to the RDA. Our research indicated that dark green and orange vegetables and
legumes, total grains, and whole grains were associated with GWG. We investigated a snapshot
of the diets of women of childbearing age and pregnant women at 1 to 10 months’ gestational.
The study included U.S. representative population to find a new way to formulate the HEI 2005
to reflect dietary quality and needs in pregnant women. For future studies, further infestation on
the relationship between HEI 2005 to pregnancy complications and birth outcomes is
recommended.

73

Table 2.1 Demographic distribution of pregnant women
No.
Wt. No.
Age (y)
18-29
166 2038880
30-43
73 1065639
Race/Ethnicity
Mexican American or Other
Hispanic
White, Non-Hispanic
African American
Other race

Wt. %
65.7
34.3

71

517961

16.7

121
31
15

2024560
351357
210640

65.2
11.3
6.8

17
39
51

141254
377771
532530

4.6
12.2
17.2

71
60

884826
1168137

28.5
37.6

177
61

2583453
521065

83.2
16.8

Family Poverty Income Ratio (PIR)
PIR <1.85
1.85 ≤PIR< 4
4 ≤PIR

99
73
66

956219
1112975
1035325

30.8
35.9
33.3

Number of pregnancies before
0
1
2
3+
Missing

13
83
44
27
71

191765
1009408
665034
231634

9.1
48.1
31.7
11.1

Trimester
st
1 trimester

40

805251

25.9

trimester

98

1131585

36.4

3 trimester

100

1167683

37.6

Education Level
0-8
9-12
High School Grad/GED or
Equivalent
Some College or AA degree
College Graduate or above
Country of Birth
United States
Mexico or other

nd

2

rd

74

Table 2.1 (con’t)
No.
Prepregnancy Wt Status (BMI)
2
Underweight (<18.5 kg/m )

Wt. No.

Wt. %

9

71894

2.3

131

1702309

54.8

Overweight (25-30 kg/m )

60

551882

17.8

Obese (>30 kg/m )
Total

38

778433

25.1

238

3104519

100.0

2

Normal (18.5-25 kg/m )
2

2

75

Table 2.2 Distribution of trimester of pregnancy and gestational weight gain groups by prepregnancy weight status
st
nd
a
3rd trimester
1 trimester
2 trimester
p-value
Prepregnancy Wt
No. Wt. No. Wt. % No. Wt. No. Wt. % No. Wt. No. Wt. %
Status
Underweight
3
17438
2.2
1
9100
0.8
5
45356
3.9
0.34
2
(<18.5 kg/m )
Normal
23
310248
38.5
50
644100
56.9
58
747961
64.1
2
(18.5-25 kg/m )
Overweight
6
172937
21.5
33
224052
19.8
21
154894
13.3
2
(25-30 kg/m )
Obese
8
304628
37.8
14
254333
22.5
16
219472
18.7
2
(>30 kg/m )
Total
Prepregnancy Wt
Status
Underweight
2
(<18.5 kg/m )
Normal
2
(18.5-25 kg/m )
Overweight
2
(25-30 kg/m )
Obese
2
(>30 kg/m )
Total
a
Chi-square

40
805251
100.0
Inadequate GWG

98

No.

Wt. No.

Wt. %

No.

2

23293

3.4

2

28

311323

45.8

8

47395

13
51

1131585 100.0
Adequate GWG
Wt. No.

100

1167683
100.0
Excessive GWG

Wt. %

No.

13722

2.1

5

34879

2.0

35

417377

64.6

68

973610

54.7

7.0

12

137643

21.3

40

366844

20.6

297195

43.8

4

76986

11.9

21

404252

22.7

679206

100.0

53

645728

100.0

134

1779585

100.0

76

Wt. No.

p-value

a

Wt. %
0.17

Table 2.3 Prepregnancy BMI and Nutrient density per 1,000 kcal by gestational weight gain
groups
Inadequate
Adequate GWG
Excessive
GWG
GWG
Mean
SEM
Mean
SEM
Mean
SEM P-value a
Prepregnancy BMI
Kcal/d
Protein (gm)
Carbohydrate (gm)
Dietary fiber (gm)
Total fat (gm)
Cholesterol (mg)
Vitamin E (mg)
Added Vitamin E (mg)
Retinol (mcg)
Vitamin A, RAE (mcg)
Alpha-carotene (mcg)
Beta-carotene (mcg)
Beta-cryptoxanthin (mcg)
Lycopene (mcg)
Lutein + zeaxanthin (mcg)
Thiamin (Vitamin B1)
(mg)
Riboflavin (Vitamin B2)
(mg)
Niacin (mg)
Vitamin B6 (mg)
Total folate (mcg)
Folic acid (mcg)
Food folate (mcg)
Folate, DFE (mcg)
Total choline (mg)
Vitamin B12 (mcg)
Added Vitamin B12 (mcg)

27.7
2276
38
129
7
38
162.5
3.3
0.2
232.1
299.2
116.2
693.8
106.6
2785.7
603.4

1.5
197
0.9
3
0.4
1
19.3
0.2
0.1
29.3
37.8
26.3
124.0
19.6
550.3
174.7

24.3
2264
41
133
9
35
116.1
4.3
0.4
269.4
366.3
166.0
1040.0
83.5
4118.1
663.4

1.3
55
1
1
0.5
1
6.4
0.3
0.2
23.8
16.8
35.3
150.8
8.4
1260.0
111.1

25.4
2437
36
133
8
37
122.0
3.4
0.3
255.9
332.3
145.2
793.9
86.0
2884.4
628.1

0.7
220
1
2
0.4
0.6
7.0
0.3
0.1
13.1
12.4
27.4
122.5
12.1
346.6
89.1

0.40
0.30
0.07
0.51
0.68
0.83
*
0.68
0.81
0.50
0.62
0.72
0.93
0.55
0.76
0.95

0.8

0.0

0.9

0.1

0.9

0.0

0.08

1.0

0.1

1.2

0.1

1.2

0.0

0.14

10.9

0.6

12.7

1.0

11.5

0.4

0.73

0.9

0.1

1.2

0.1

1.0

0.1

0.57

199.0
101.3
97.6
269.9
157.4
2.5
0.3

13.3
11.0
6.6
20.3
10.4
0.1
0.1

270.6
161.6
108.9
383.7
159.3
3.2
1.0

35.5
39.1
4.9
62.8
9.0
0.5
0.3

229.1
132.8
96.2
321.9
135.8
2.7
0.7

7.8
10.8
6.0
14.7
6.3
0.1
0.1

0.53
0.42
0.54
0.49
*
0.98
0.21

77

Table 2.3 (con’t)
Inadequate
GWG
Mean
SEM
Vitamin C (mg)
Vitamin K (mcg)
Calcium (mg)
Phosphorus (mg)
Magnesium (mg)
Iron (mg)
Zinc (mg)
Cooper (mg)
Sodium (mg)
Potassium (mg)
Selenium (mcg)
Caffeine (mg)
Theobromine (mg)
Average of two 24h recalls
a
One-way Anova
*p<0.05

Adequate GWG
Mean

SEM

46.9
39.8
506.8
637.2
138.2
7.6
5.6
0.6
1611.7
1260.2
54.9
31.4
16.0

57.2
55.1
532.9
699.3
162.5
9.2
7.1
0.7
1697.9
1465.4
57.3
18.4
26.6

2.6
5.5
22.8
36.0
11.7
1.1
1.2
0.0
143.0
103.2
4.0
3.9
6.1

7.6
7.7
46.1
31.1
6.4
0.4
0.2
0.0
33.7
60.9
3.2
11.7
2.6

78

Excessive
GWG
Mean
SEM
50.8
43.8
546.1
646.3
136.8
8.3
6.0
0.6
1612.5
1239.4
50.0
19.8
20.8

5.3
4.9
26.4
23.6
5.8
0.2
0.3
0.0
50.7
42.2
2.1
2.6
2.2

a

P-value
0.23
0.96
0.54
0.90
0.40
0.54
0.94
0.38
0.76
0.39
*
0.28
0.53

Table 2.4 Nutritional adequacy by gestational weight gain groups
Inadequate GWG
Adequate GWG
No. Wt. No. Wt. % No. Wt. No. Wt. %
Protein (g/d)
RDA<71
RDA≥71
Carbohydrate
(g/d)
RDA<175
RDA≥175
Fiber (g/d)
AI<28
AI≥28
Vit A (µg/d)
RDA<770
RDA≥770
Vit B1(mg/d)
RDA<1.4
RDA≥1.4
Vit B2 (mg/d)
RDA<1.6
RDA≥1.6
Niacin(mg/d)
RDA<18
RDA≥18
Vit B6 (mg/d)
RDA<1.9
RDA≥1.9

Excessive GWG
No. Wt. No.
Wt. %

p-value

a

20
31

241839
463315

34.3
65.7

16
37

175232
501741

25.9
74.1

42
92

858771
1548392

35.7
64.3

0.72

5
46

81330
623824

11.5
88.5

4
49

8689
668284

1.3
98.7

14
120

320422
2086742

13.3
86.7

0.21

44
7

650192
54962

92.2
7.8

44
9

594248
82724

87.8
12.2

114
20

1840332
566831

76.5
23.5

0.10

37
14

476183
228971

67.5
32.5

33
20

294523
382449

43.5
56.5

81
53

1382625
1024539

57.4
42.6

0.51

17
34

257604
447550

36.5
63.5

20
33

138081
538891

20.4
79.6

40
94

763850
1643313

31.7
68.3

0.29

10
41

156915
548239

22.3
77.7

5
48

37386
639586

5.5
94.5

20
114

231086
2176077

9.6
90.4

*

11
40

171442
533712

24.3
75.7

14
39

138973
537999

20.5
79.5

38
96

587178
1819986

24.4
75.6

0.92

29
22

457737
247417

64.9
35.1

25
28

218556
458416

32.3
67.7

53
81

834123
1573040

34.7
65.3

*

79

Table 2.4 (con’t)
Inadequate GWG
No. Wt. No. Wt. %
Folate (µg/d)
RDA<600
RDA≥600
Vit B12 (µ
g/d)
RDA<2.6µg
RDA≥2.6µg
Vit C (mg/d)
RDA<85
RDA≥85
Vit K (µg/d)
AI<90
AI≥90
Calcium (mg/d)
AI<1000
AI≥1000
Phosphorous
(mg/d)
RDA<700
RDA≥700
Magnesium
(mg/d)
RDA<350
RDA≥350
Iron (mg/d)
RDA<27
RDA≥27

Adequate GWG
No. Wt. No. Wt. %

Excessive GWG
No. Wt. No.
Wt. %

p-value

a

41
10

618347
86806

87.7
12.3

38
15

367057
309915

54.2
45.8

96
38

1527768
879395

63.5
36.5

0.13

11
40

75375
629779

10.7
89.3

6
47

82350
594622

12.2
87.8

18
116

264017
2143146

11.0
89.0

0.98

22
29

346934
358219

49.2
50.8

22
31

216928
460044

32.0
68.0

50
84

1104185
1302978

45.9
54.1

0.61

35
16

561885
143269

79.7
20.3

36
17

319001
357971

47.1
52.9

84
50

1426821
980343

59.3
40.7

0.18

34
17

439366
265788

62.3
37.7

33
20

324128
352844

47.9
52.1

80
54

1523324
883839

63.3
36.7

0.54

4
47

70286
634868

10.0
90.0

1
52

1363
675609

0.2
99.8

6
128

39626
2367537

1.6
98.4

0.01

38
13

515192
189962

73.1
26.9

35
18

357777
319195

52.8
47.2

94
40

1594311
812852

66.2
33.8

0.48

43
8

653214
51940

92.6
7.4

47
6

477225
199747

70.5
29.5

114
20

1976775
430388

82.1
17.9

0.42

80

Table2.4 (con’t)
Inadequate GWG
No.

Wt. No.

Zinc (mg/d)
RDA<11
26
Table 4 (Con’t)
RDA≥11
25
Copper (µg/d)
RDA<1000
13
RDA≥1000
38
Sodium (mg/d)
AI<1500
2
AI≥1500
49
Potassium
(mg/d)
AI<4700
49
AI≥4700
2
Selenium (µg/d)
RDA<60
1
RDA≥60
50
Total
51
Average of two 24h recalls
a
Chi-square
*p<0.05
**p<0.001

Adequate GWG

Excessive GWG

a

Wt. %

No.

Wt. No.

Wt. %

No.

311316

44.1

21

242192

35.8

52

1050467

43.6

393838

55.9

32

434780

64.2

82

1356696

56.4

141637
563517

20.1
79.9

12
41

76606
600366

11.3
88.7

21
113

585370
1821794

24.3
75.7

0.40

19079
686074

2.7
97.3

2
51

3000
673973

0.4
99.6

0
134

0
2407163

0
100.0

.

687092
18062

97.4
2.6

45
8

632002
44971

93.4
6.6

126
8

2025888
381276

84.2
15.8

0.07

3719
701435
705154

0.5
99.5
100.0

2
51
53

5567
671405
676972

0.8
99.2
100.0

6
128
134

301645
2105518
2407163

12.5
87.5
100.0

**

81

Wt. No.

p-value

Wt. %
0.87

Table 2.5 Healthy Eating Index 2005 score by gestational weight gain groups
Inadequate GWG
Adequate
Excessive
GWG
GWG
Parameter
Mean
SEM
Mean SEM Mean SEM
Healthy Eating Index
1

2005
TOTAL FRUIT
WHOLE FRUIT
TOTAL VEGETABLES
DARK GREEN &
ORANGE VEG &
2

LEGUMES
TOTAL GRAINS
WHOLE GRAINS
3

MILK
MEAT & BEANS
OILS4
SATURATED FAT5
SODIUM5
CALORIES FROM
SOLID FAT, ALCOHOL
& ADDED SUGAR
(SoFAAS)
Average of two 24h recalls
a
One-way Anova

P-value

a

55.0

2.6

61.7

4.2

54.0

1.5

0.63

2.4
1.9
3.0

0.6
0.5
0.2

3.7
3.3
3.4

0.3
0.5
0.2

3.1
2.5
2.7

0.4
0.3
0.3

0.46
0.63
0.64

1.0

0.3

1.5

0.5

0.9

0.2

0.83

4.6
0.9

0.1
0.2

4.2
1.5

0.1
0.5

4.4
1.4

0.1
0.2

0.72
0.32

5.2

0.7

7.1

0.8

6.8

0.4

0.29

8.3
6.0
5.8
3.9

0.3
0.5
0.4
0.3

8.0
5.8
6.3
3.8

0.4
0.2
0.7
0.5

7.6
5.3
5.5
3.9

0.5
0.4
0.4
0.4

0.54
0.37
0.67
0.82

12.0

1.0

13.1

1.8

9.7

0.7

0.12

1 Intakes between the minimum and maximum levels are scored proportionately, except for
Saturated Fat and Sodium.
2 Legumes counted as vegetables only after Meat and Beans standard is met.
3 Includes all milk products, such as fluid milk, yogurt, and cheese, and soy beverages.
4 Includes nonhydrogenated vegetable oils and oils in fish, nuts, and seeds.
5 Saturated Fat and Sodium get a score of 8 for the intake levels that reflect the 2005 Dietary
Guidelines, <10% of calories from saturated fat and 1.1 grams of sodium/1,000 kcal, respectively.

82

Table 2.6 Logistic regression models for Healthy Eating Index 2005 components to inadequate
or excessive gestational weight gain in comparison with adequate gestational weight gain
Excessive or Inadequate GWG
vs.
Adequate GWG (REF)
HEI 2005 Components
TOTAL FRUIT
WHOLE FRUIT
TOTAL VEGETABLES
DARK GREEN &
ORANGE VEG &
LEGUMES
TOTAL GRAINS
WHOLE GRAINS
MILK
MEAT & BEANS
SATURATED FAT

Effect

REF

< 0.8 cup eq
/1000kcal
< 0.4 cup eq
/1000kcal
< 1.1 cup eq
/1000kcal

Adjusted OR
(95%Cl)

a

p-value

≥ 0.8

0.81

1.56 (0.65-3.73)

0.31

≥ 0.4

4.69 (1.02-21.48)

0.04

≥ 3.0

0.46 (0.22-0.96)

0.03

≥ 1.5

0.12 (0.02-0.74)

0.02

≥ 1.3

1.02 (0.21-4.93)

0.97

≥ 2.5

0.75 (0.17-3.32)

0.70

≤ 7%

> 7% of energy

1.13 (0.39-3.27)

≥ 1.1

< 3.0cup eq
/1000kcal
< 1.5 cup eq
/1000kcal
< 1.3 cup eq
/1000kcal
< 2.5 oz eq
/1000kcal

0.92

≥ 0.4

< 0.4 cup eq
/1000kcal

1.05 (0.38-2.85)

0.58 (0.09-3.55)

0.55

CALORIES FROM
SOLID FAT, ALCOHOL
> 20% of energy
≤ 20%
4.83 (0.65-35.68)
0.12
& ADDED SUGAR
(SoFAAS)
Average of two 24h recalls
a
Adjustments were made of maternal age, race/ethnicity, country of birth, family poverty
income ratio and education levels.

83

Table 2.7 Logistic regression models for HEI 2005 components to inadequate or excessive
gestational weight gain in comparison with adequate gestational weight gain
Excessive or Inadequate GWG
vs.
Adequate GWG (REF)
HEI 2005 Component
TOTAL FRUIT
WHOLE FRUIT
TOTAL VEGETABLES
DARK GREEN &
ORANGE VEG &
LEGUMES
TOTAL GRAINS
WHOLE GRAINS
MILK
MEAT & BEANS

Effect
< 0.8 cup eq
/1000kcal
< 0.4 cup eq
/1000kcal
< 1.1 cup eq
/1000kcal
< 0.4 cup eq
/1000kcal
< 3.0cup eq
/1000kcal
< 1.5 cup eq
/1000kcal
< 1.3 cup eq
/1000kcal
< 2.5 oz eq
/1000kcal

REF
≥ 0.8
≥ 0.4
≥ 1.1

a

Adjusted OR (95% Cl)

p-value

1.37 (0.24-7.88)
0.40 (0.06-2.86)
2.26 (0.67-7.62)

≥ 0.4

0.73
0.36
0.19
0.01

8.36 (1.81-38.51)
≥ 3.0
≥ 1.5
≥ 1.3
≥ 2.5

SATURATED FAT

> 7% of energy

≤ 7%

CALORIES FROM
SOLID FAT, ALCOHOL
& ADDED SUGAR
(SoFAAS)

> 20% of
energy

≤ 20%

a

0.37 (0.13-1.07)
0.15 (0.03-0.81)
0.72 (0.15-3.45)
0.33 (0.10-1.10)
0.93 (0.14-6.16)
7.82 (1.58-38.67)

Adjustments were made of maternal age, race/ethnicity, country of birth, family poverty
income ratio, education levels, physical activity levels and prepregnancy BMI
Average of two 24h recalls

84

0.07
0.03
0.68
0.07
0.94

0.01

Chapter Four

CONCLUSION
A. Implications
Pre-pregnancy weight status must be known in order to apply the gestational weight gain
guidelines developed by the IOM (2009). Information of prepregnancy weight status is usually
obtained from pregnant women retroactively, through self-report (Cedergren, 2006). The selfreported data are often validated by measured weight during the first trimester assuming that
little change in weight would occur during the first trimester (Olson et al., 2003; Park et al.,
2009), by weight measured during the first prenatal visit, which can occur at varying trimesters
of pregnancy, or even by weight measured within the previous 12 months of pregnancy
(Hedderson et al., 2010). All of these approaches, however, deviate from the common ways to
validating self-reported data, i.e., self-reported data validated by objectively measured data.
Consequently previous research used various references to make up for the lacked gold standards
to validate the self-reported prepregnancy weight. In this study, since no measured prepregnancy
weight values were available, multiple imputed statistical approach was used to establish a gold
reference and then validate self-reported prepregnancy weight status.
A strong agreement was found between self-reported prepregnancy weight status and prepregnancy weight determined using the multiple imputed value method (Kappa value, 0.73).
However, pregnant women with a low family poverty income ratio (<1.0) had the greatest
discrepancy between self-reported pre-pregnancy weight and multiple imputed pre-pregnancy
weight, due to under-reporting. A low education level, i.e., less than 8 years, was associated with
under-reporting of self-reported height in pregnant women. This information will be useful to
85

researchers and clinicians applying gestational weight gain guidelines by using prepregnancy
weight status based on self-reported height and weight.
Maternal nutrition is one key factor which impacts gestational weight gain. In this study,
diet quality as a determinant of gestational weight gain was assessed. Each component of the
foods included in Healthy Eating Index (HEI) 2005 and total HEI 2005 scores in relation to
adequate gestational weight gain. Consumption of dark green and orange vegetables and legumes
were inversely associated with gestational weight gain whereas total grains and whole grains
were positively associated with gestational weight gain. This finding provides food guidelines
for women who are planning to become pregnant, to help them achieve optimal gestational
weight gain according to the guidelines developed by the IOM (2009). However, no significant
difference was found in the HEI 2005 total composite scores among the three gestational weight
gain groups: i.e., inadequate, adequate, and excessive. The present study included the U.S.
representative population to find a new way to formulate HEI 2005 to assess dietary quality and
needs in pregnant women.
The main strength of this research is that the average of day 1 and day 2 24-hour recalls
was used to determine the mean usual intake. To our knowledge, this is the first study to use the
multiple-imputed approach to validate self-reported pre-pregnancy weight, and to examine HEI
2005 to gestational weight gain in the representative U.S. national population. The limitations of
this study are that diet quality was examined from foods only, and micronutrients from vitamins
or supplements were not taken into account. We utilized NHANES 2005-6. However, to increase
the total number of pregnant women for this analysis, combining the dataset from NHANES
1999-2000, NHANES 2001-2002, NHANES 2003-2004 would be recommended to increase the
study’s impact. The study provided a snapshot design of gestational weight gain in months 1-10

86

of pregnancy when the survey was conducted, and this may not reflect the total gestational
weight gain.
B. Recommendations for future research
It is still unknown whether women whose dietary quality is high, are able to accurately
recall their pre-pregnancy weight status. Future research on the associations between groups of
accurate self-reporters and their diet quality assessed by HEI 2005 will be recommended.
Encouraging more prenatal visits, during which weight can be measured, can assist pregnant
women in more accurately recalling their prepregnancy weight.
This study included gestational weight gain at each month of pregnancy because total
gestational weight gain was not available. Elucidating the weight gain pattern of pregnant
women throughout pregnancy and total gestational weight gain in relation to diet quality will be
recommended. Combining foods and supplements might be useful in assessing diet quality and
discerning micronutrient needs more accurately in pregnant women. Our study was crosssectionally designed. A prospective cohort study will be useful to examine the associations and
establish a linkage between prepregnancy weight status, gestational weight gain, postpartum
weight retention, pregnancy complications, birth outcomes and childhood health status in their
later lives.
The surveillance and monitoring programs or datasets to fulfill this study’s
recommendations are the following: Pregnancy Risk Assessment Monitoring System is an ongoing surveillance project the CDC and state health departments, collects state-specific
population-based data on maternal attitudes and experiences before, during, and shortly after
pregnancy. The Pregnancy, Infection, and Nutrition study is a longitudinal, prospective
investigation of adverse birth outcomes being conducted at selected prenatal clinics in central
87

North Carolina. The National Institutes of Health (NIH) established the Women's Health
Initiative (WHI) in 1991 to address the most common causes of death, disability and impaired
quality of life in postmenopausal women. The Women’s Health Initiatives (WHI) addressed
cardiovascular disease, cancer, and osteoporosis among women aged 72 and older. This dataset
will be useful for investigating middle-aged or older women’s health status in relation to their
pregnancy complications and birth outcomes.

88

APPENDICES

89

Table A1. Correlations and mean difference between self-reported weight 1 year ago and selfreported weight by maternal age, race/ethnicity, education, country of birth, number of previous
births, and family poverty income ratio for non-pregnant women
Weight (kg)
SelfSelfDifferences
n
reported
Pearson r
reported
1 yr ago
All
489 72.73 (1.28) 73.44 (1.23) -0.71 (0.33)
0.92
(p<0.0001)
a
p-value
<0.0001
Age
18-29

208

68.20 (1.36)

70.28 (1.56)

-2.09 (0.79)

30-41

281

74.74 (1.48)

74.84 (1.45)

-0.10 (0.38)
b

p-value

0.88
(p<0.0001)
0.93
(p<0.0001)

0.05

Race/Ethnicity
MA or O H

146

70.11 (1.12)

71.00 (1.35)

-0.90 (1.39)

NHW

174

72.85 (2.00)

73.19 (1.87)

-0.33 (0.54)

NHB

140

80.13 (1.96)

81.91 (2.04)

-1.78 (0.60)

29

62.75 (2.83)

63.86 (2.50)

-1.11 (1.56)

Other race

b

p-value

0.79
(p<0.0001)
0.93
(p<0.0001)
0.93
(p<0.0001)
0.91
(p<0.0001)

0.61

Country of Birth
United States

380

74.32 (1.39)

75.10 (1.28)

-0.78 (0.39)

M or other

109

65.14 (0.95)

65.54 (1.13)

-0.39 (0.84)
b

p-value

0.92
(p<0.0001)
0.81
(p<0.0001)

0.70

Education
0-8

25

66.67 (2.64)

66.40 (2.27)

0.27 (1.11)

9-12

86

75.87 (1.89)

76.67 (1.20)

-0.80 (1.17)

13-14

132

72.41 (2.40)

74.23 (2.32)

-1.82 (1.02)

90

0.79
(p<0.0001)
0.90
(p<0.0001)
0.89
(p<0.0001)

Table A1. (con’t)
Weight (kg)
Selfreported
1 yr ago

n

Selfreported

Differences

15-18

173

75.84 (1.97)

76.33 (2.04)

-0.49 (0.57)

19-21

73

66.94 (2.58)

66.82 (2.66)

0.12 (0.64)
b

p-value

Pearson r
0.93
(p<0.0001)
0.94
(p<0.0001)

0.22

Family PIR
PIR <1

131

71.46 (1.88)

74.81 (1.98)

-3.35 (0.64)

1=<PIR<1.999

119

77.62 (2.16)

78.58 (2.22)

-0.96 (1.11)

2=<PIR<2.999

71

73.86 (3.72)

72.60 (2.75)

1.25 (1.45)

3=<PIR<4.999

109

73.27 (2.03)

73.25 (1.95)

0.01 (0.66)

59

66.35 (2.44)

66.98 (2.31)

-0.63 (0.77)

5=<PIR

b

p-value

0.91
(p<0.0001)
0.90
(p<0.0001)
0.88
(p<0.0001)
0.95
(p<0.0001)
0.95
(p<0.0001)

0.007

mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

91

Table A2. Correlations and mean difference between self-reported and measured BMI by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
poverty income ratio for non-pregnant women
2

BMI (kg/m )
Measured
[mean
(SEM)]

Differences
[mean
(SEM)]

489

Selfreported
[mean
(SEM)]
27.50 (0.41)

28.27 (0.48)

-0.77 (0.09)
a
<0.0001

0.97 (p<0.0001)

18-29

208

26.45 (0.57)

27.16 (0.62)

-0.71 (0.11)

0.95 (p<0.0001)

30-41

281

27.96 (0.49)

28.76 (0.57)

-0.80 (0.10)

0.97 (p<0.0001)

n
All
p-value

Pearson r

Age

p-value

0.38

b

Race/Ethnicity
MA or O H

146

28.05 (0.46)

29.05 (0.45)

-1.00 (0.17)

0.95 (p<0.0001)

NHW

174

26.93 (0.61)

27.65 (0.74)

-0.73 (0.16)

0.98 (p<0.0001)

NHB

140

30.49 (0.86)

31.23 (1.10)

-0.73 (0.33)

0.96 (p<0.0001)

29

25.00 (0.78)

25.74 (0.83)

-0.74 (0.10)

0.96 (p<0.0001)

Other race
p-value

0.46

b

Country of Birth
United States

380

27.85 (0.42)

28.57 (0.51)

-0.72 (0.12)

0.97(p<0.0001)

M or other

109

25.80 (0.47)

26.80 (0.43)

-0.99 (0.17)

0.94(p<0.0001)

p-value

0.32

b

Education
0-8

25

27.11 (1.17)

28.02 (1.15)

-0.91 (0.23)

0.96 (p<0.0001)

9-12

86

28.96 (0.77)

29.84 (0.73)

-0.88 (0.43)

0.94 (p<0.0001)

13-14

132

28.10 (0.71)

29.12 (0.78)

-1.02 (0.19)

0.97 (p<0.0001)

92

Table A2. (con’t)
2

BMI (kg/m )
n
15-18
19-21
p-value
Family PIR
PIR <1
1=<PIR<1.999
2=<PIR<2.999
3=<PIR<4.999
5=<PIR
p-value

173
73

Selfreported
[mean
(SEM)]
28.21 (0.74)
24.77 (0.81)

Measured
[mean
(SEM)]

Differences
[mean
(SEM)]

Pearson r

28.99 (0.79)
25.12 (0.88)

-0.79 (0.09)
-0.35 (0.13)

0.98 (p<0.0001)
0.98 (p<0.0001)

0.07
131
119
71
109
59

28.28 (0.78)
29.64 (0.74)
27.75 (1.06)
26.93 (0.70)
24.88 (0.74)

29.30 (0.83)
30.48 (0.80)
28.51 (1.24)
27.72 (0.78)
25.25 (0.79)

-1.02 (0.20)
-0.85 (0.31)
-0.76 (0.26)
-0.80 (0.12)
-0.37 (0.12)
0.12

a

b

b

0.97 (p<0.0001)
0.94 (p<0.0001)
0.97 (p<0.0001)
0.97 (p<0.0001)
0.98 (p<0.0001)

Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

93

Table A3. Pearson r, Mean, and mean difference for self-reported weight 1 year ago and
prepregnancy weight
Reliability
n
Wt at preWt at 1 yr
pregnancy
95%
Pearson r
ago
Mean
Differences
Confidence
Mean (SEM)
(SEM)
Interval
Pregnant
women
Nonpregnant
women

279

0.96
(p<0.0001)

69.49 (2.36)

70.15 (2.33)

-0.66 (0.49)

[-1.70:0.37]

489

0.92
(p<0.0001)

72.73 (1.28)

73.44 (1.23)

-0.71 (0.33)

[-1.422:0.003]

94

Table A4. Correlations and mean difference between self-reported and measured height, weight, and BMI by socio-demographic
characteristics
2
Height (cm)
Weight (kg)
BMI (kg/m )
Mean
Mean
Mean
n
Pearson r
Pearson r
Pearson r
differences
differences
differences
All
489 0.52 (0.16)
0.94 -1.60 (0.26)
0.98 -0.77 (0.09)
0.97
(p<0.0001)
(p<0.0001)
(P<0.0001)
Age
18-29

208

0.44 (0.18)

30-41

281

0.55 (0.20)

MA or O H

146

1.12 (0.36)

NHW

174

0.41 (0.21)

NHB

140

0.46 (0.14)

29

0.30 (0.55)

United States

380

0.35 (0.16)

M or other

109

1.34 (0.37)

0.92
(p<0.0001)
0.95
(p<0.0001)

-1.55 (0.33)

0.87
(p<0.0001)
0.95
(p<0.0001)
0.95
(p<0.0001)
0.92
(p<0.0001)

-1.50 (0.33)

0.95
(p<0.0001)
0.87
(p<0.0001)

-1.64 (0.32)

-1.62 (0.26)

0.96
(p<0.0001)
0.98
(p<0.0001)

-0.71 (0.11)

0.97
(p<0.0001)
0.98
(p<0.0001)
0.96
(p<0.0001)
0.97
(p<0.0001)

-1.00 (0.17)

0.98
(p<0.0001)
0.95
(p<0.0001)

-0.72 (0.12)

-0.80 (0.10)

0.95
(P<0.0001)
0.97
(P<0.0001)

Race/Ethnicity

Other race

-1.65 (0.38)
-1.45 (0.81)
-1.66 (0.38)

-0.73 (0.16)
-0.73 (0.33)
-0.74 (0.10)

0.95
(P<0.0001)
0.98
(P<0.0001)
0.96
(P<0.0001)
0.96
(P<0.0001)

Country of Birth

-1.39 (0.33)

95

-0.99 (0.17)

0.97
(P<0.0001)
0.94
(P<0.0001)

Table A4. (con’t)
n

Height (cm)
Mean
Pearson r
differences

Weight (kg)
Mean
Mean
differences differences

0.90
(p<0.0001)
0.91
(p<0.0001)
0.95
(p<0.0001)
0.95
(p<0.0001)
0.95
(p<0.0001)

-1.08 (0.42)

0.89
(p<0.0001)
0.91
(p<0.0001)
0.97
(p<0.0001)
0.96
(p<0.0001)
0.95
(p<0.0001)

-1.61 (0.56)

2

BMI (kg/m )
Mean
Pearson r
differences

Education
0-8

25

1.69 (0.53)

9-12

86

0.80 (0.36)

13-14

132

0.50 (0.27)

15-18

173

0.52 (0.16)

19-21

73

0.18 (0.23)

PIR <1

131

1.02 (0.32)

1=<PIR<1.999

119

0.58 (0.26)

2=<PIR<2.999

71

0.26 (0.19)

3=<PIR<4.999

109

0.40 (0.31)

59

0.34 (0.25)

-1.43 (1.00)
-2.35 (0.48)
-1.65 (0.30)
-0.77 (0.31)

0.96
(p<0.0001)
0.96
(p<0.0001)
0.98
(p<0.0001)
0.98
(p<0.0001)
0.99
(p<0.0001)

-0.91 (0.23)

0.97
(p<0.0001)
0.94
(p<0.0001)
0.97
(p<0.0001)
0.97
(p<0.0001)
0.98
(p<0.0001)

-1.02 (0.20)

-0.88 (0.43)
-1.02 (0.19)
-0.79 (0.09)
-0.35 (0.13)

0.96
(P<0.0001)
0.94
(P<0.0001)
0.97
(P<0.0001)
0.98
(P<0.0001)
0.98
(P<0.0001)

p-value
Family PIR

5=<PIR

-1.81 (0.72)
-1.77 (0.74)
-1.84 (0.25)
-0.73 (0.26)

96

-0.85 (0.31)
-0.76 (0.26)
-0.80 (0.12)
-0.37 (0.12)

0.97
(P<0.0001)
0.94
(P<0.0001)
0.97
(P<0.0001)
0.97
(P<0.0001)
0.98
(P<0.0001)

Table A5. Sensitivity, specificity, and predictive value positive (PVP) and predictive value
negative (PVN) of self-reported BMI with measured BMI
BMI categories
Validity
Sensitivity
Specificity
PVP
PVN
Underweight
91.19%
99.54%
88.98%
99.64%
Normal
90.78%
90.99%
85.20%
94.53%
Overweight
73.86%
88.78%
68.97%
90.96%
Obese
83.88%
98.03%
95.73%
92.02%

97

Table A6. Distribution of pregnant women by each trimester
st

1 trimester
N
Wt. N
%

47
879822
24.9%

nd

2

trimester
115
1273669
36.1%

98

rd

3 trimester
117
1374197
39.0%

Total
279
3527688
100.0%

Table A7. Correlations and mean difference between self-reported and measured height by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
st
poverty income ratio for pregnant women in 1 trimester
Height (cm)
Self-reported
Measured
Mean
n
Pearson r
Ht
Ht
differences
All
47
165.62 (2.58)
164.76
0.87 (0.44) 0.96 (p<0.0001)
(2.70)
<0.0001
p-value
Age
37
164.49 (2.75)
163.66
0.84 (0.47) 0.96 (p<0.0001)
18-29
(3.14)
10
169.32 (5.18)
168.37
0.95 (0.98) 0.99 (p<0.0001)
30-41
(4.20)
0.93
p-value
Race/Ethnicity
12
156.97 (.)
156.67 (.)
0.30 (.) 0.74 (p=0.0060)
MA or O H
NHW

23

168.48 (2.27)

NHB

8

164.92 (0.88)

Other race

4

156.42 (.)

p-value
Country of Birth
United States
M or other

167.74
(2.22)
165.10
(0.78)
153.12 (.)

0.75 (0.51)

0.98 (p<0.0001)

-0.18 (0.10)

0.97 (p<0.0001)

3.30 (.)

0.99 (p=0.0080)

0.21
39

166.94 (2.18)

8

166.37
(2.07)
155.60 (.)

158.12 (.)

0.57 (0.45)

0.98 (p<0.0001)

2.52 (.)

0.80 (p=0.0177)

0.28

p-value
Education
0-8

2

164.54 (.)

162.08 (.)

2.46 (.)

.

9-12

12

156.35 (.)

155.84 (.)

0.51 (.)

0.85 (p=0.0005)

13-14

13

160.08 (3.19)

1.55 (0.38)

0.98 (p<0.0001)

15-18

11

166.91 (1.63)

0.82 (0.36)

0.93 (p<0.0001)

19-21

9

172.89 (0.49)

158.53
(3.53)
166.09
(1.60)
172.19
(0.11)

0.71 (0.60)

0.92 (p=0.0005)

0.90

p-value

99

Table A7. (con’t)

9-12

12

Self-reported
Ht
156.35 (.)

13-14

13

160.08 (3.19)

15-18

11

166.91 (1.63)

19-21

9

Height (cm)
Measured
Mean
Ht
differences
155.84 (.)
0.51 (.)

172.89 (0.49)

n

p-value
Family PIR
PIR <1

158.53
(3.53)
166.09
(1.60)
172.19
(0.11)

Pearson r
0.85 (p=0.0005)

1.55 (0.38)

0.98 (p<0.0001)

0.82 (0.36)

0.93 (p<0.0001)

0.71 (0.60)

0.92 (p=0.0005)

0.90
16

160.04 (0.42)

-0.97 (0.20)

0.95 (p<0.0001)

1.55 (1.16)

0.92 (p<0.0001)

174.55 (.)

161.01
(0.51)
158.33
(5.15)
172.82 (.)

1=<PIR<1.999

11

159.88 (4.15)

2=<PIR<2.999

6

1.73 (.)

0.84 (p=0.0369)

3=<PIR<4.999

5

163.84 (.)

161.99 (.)

1.85 (.)

0.99 (p=0.0004)

5=<PIR

9

169.18 (0.66)

169.98
(1.37)

-0.80 (0.71)

0.94 (p=0.0001)

0.81

p-value

mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

100

Table A8. Correlations and mean difference between self-reported and measured weight by
maternal age, race/ethnicity, education, country of birth, number of previous births, and family
st
poverty income ratio in pregnant women in 1 trimester
Weight (kg)
n
All

47

Selfreported Wt
78.03 (4.96)

Measured
Wt
81.11 (5.14)

Mean
differences
-3.07 (0.58)
<0.0001

a

Age
18-29

37

75.54 (5.05)

79.42 (5.63)

-3.88 (0.69)

30-41

10

86.19 (9.85)

86.63
(10.03)

-0.44 (0.33)

p-value
Race/Ethnicity
MA or O H

0.10

b

12

62.46 (.)

63.55 (.)

-1.09 (.)

NHW

23

80.87 (4.65)

83.45 (4.53)

-2.57 (0.58)

NHB

8

Other race

4

106.81
(5.45)
53.63 (.)

122.23 -15.43 (0.68)
(6.13)
52.73 (.)
0.90 (.)

p-value
Country of Birth
United States
M or other

0.73

b

39

81.85 (4.26)

85.40 (4.11)

-3.55 (0.52)

8

56.31 (.)

56.68 (.)

-0.38 (.)

p-value

0.16

b

Education
0-8

2

72.57 (.)

75.84 (.)

-3.26 (.)

9-12

12

72.02 (.)

76.95 (.)

-4.93 (.)

13-14

13

68.12 (0.81)

74.51 (1.60)

-6.38 (0.84)

15-18

11

79.84 (4.13)

83.11 (4.83)

-3.27 (0.95)

19-21

9

85.41 (3.94)

85.56 (3.71)

-0.15 (0.24)

p-value

0.27
101

b

Pearson r
0.97
(p<0.0001)

0.97
(p<0.0001)
0.99
(p<0.0001)

0.98
(p<0.0001)
0.98
(p<0.0001)
0.98
(p<0.0001)
0.99
(p=0.0046)

0.96
(p<0.0001)
0.98
(p<0.0001)

.
0.99
(p<0.0001)
0.97
(p<0.0001)
0.99
(p<0.0001)
0.99
(p<0.0001)

Table A8. (con’t)
Weight (kg)
n

Selfreported Wt

Measured
Wt

Mean
differences

PIR <1

16

81.01 (2.78)

88.31 (3.11)

-7.29 (0.33)

1=<PIR<1.999

11

62.59 (6.50)

64.59 (6.31)

-2.00 (0.21)

2=<PIR<2.999

6

95.20 (.)

94.52 (.)

0.68 (.)

3=<PIR<4.999

5

83.37 (.)

90.61 (.)

-7.24 (.)

5=<PIR

9

64.97 (0.61)

66.15 (0.67)

-1.17 (0.07)

Pearson r

Family PIR

p-value

0.67

b

0.98
(p<0.0001)
0.96
(p<0.0001)
0.99
(p<0.0001)
0.99
(p=0.0003)
0.99
(p<0.0001)

mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

102

Table A9. Correlations and mean difference between self-reported and measured BMI by sociost
demographic characteristics in pregnant women in 1 trimester
2

n
All

47

Self-reported
BMI
28.20 (1.09)

BMI (kg/m )
Measured
Mean
BMI
differences
29.64 (1.27)
-1.44 (0.24)

p-value

<0.0001

a

Pearson r
0.96
(p<0.0001)

Age
18-29

37

27.77 (1.28)

29.49 (1.55)

-1.72 (0.33)

30-41

10

29.60 (1.53)

30.15 (1.96)

-0.55 (0.43)
b

p-value
Race/Ethnicity
MA or O H

0.13
12

25.56 (.)

25.94 (.)

-0.37 (.)

NHW

23

28.31 (1.13)

29.62 (1.36)

-1.31 (0.32)

NHB

8

39.08 (1.64)

44.51 (1.95)

-5.43 (0.30)

Other race

4

21.82 (.)

22.35 (.)

-0.53 (.)
b

p-value
Country of Birth
United States
M or other

0.97
(p<0.0001)
0.98
(p<0.0001)

0.91
(p<0.0001)
0.96
(p<0.0001)
0.98
(p<0.0001)
0.99
(p<0.0089)

0.55
39

29.19 (1.06)

30.77 (1.24)

-1.57 (0.28)

8

22.55 (.)

23.25 (.)

-0.70 (.)
b

p-value

0.96
(p<0.0001)
0.70
(p=0.0531)

0.28

Education
0-8

2

26.90 (.)

29.30 (.)

-2.40 (.)

.

9-12

12

28.99 (.)

30.76 (.)

-1.77 (.)

13-14

13

26.57 (1.34)

29.70 (1.83)

-3.13 (0.51)

15-18

11

28.59 (1.01)

30.01 (1.20)

-1.42 (0.39)

19-21

9

28.33 (1.14)

28.70 (1.25)

-0.37 (0.11)

0.98
(p<0.0001)
0.97
(p<0.0001)
0.98
(p<0.0001)
0.98
(p<0.0001)

b

p-value

0.26

103

Table A9. (con’t)

2

BMI (kg/m )
Self-reported
BMI

n

Measured
BMI

Mean
differences

Pearson r

Family PIR
PIR <1

16

31.23 (0.94)

33.49 (1.13)

-2.26 (0.19)

1=<PIR<1.999

11

24.31 (1.36)

25.51 (0.97)

-1.20 (0.40)

2=<PIR<2.999

6

31.22 (.)

31.60 (.)

-0.38 (.)

3=<PIR<4.999

5

30.71 (.)

34.20 (.)

-3.49 (.)

5=<PIR

9

22.96 (0.09)

23.14 (0.26)

-0.19 (0.16)
b

p-value

0.98
(p<0.0001)
0.89
(p=0.0002)
0.97
(p=0.0010)
0.96
(p=0.0089)
0.99
(p<0.0001)

0.81

mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

104

Table A10. Correlations and mean difference between self-reported and measured height, weight, and BMI by socio-demographic
st
characteristics in pregnant women in 1 trimester
2
Height (cm)
Weight (kg)
BMI (kg/m )
n
All

47

Mean
differences
0.87 (0.44)

0.96
(p<0.0001)

<0.0001

p-value
Age
18-29

37

0.84 (0.47)

30-41

10

0.95 (0.98)

p-value
Race/
Ethnicity
MA or O
H

Pearson r

Mean
differences
-3.07 (0.58)

Pearson r

Mean differences

Pearson r

0.97 (p<0.0001)

-1.44 (0.24)

0.96 (p<0.0001)

-3.88 (0.69)

0.97 (p<0.0001)

-1.72 (0.33)

0.97 (p<0.0001)

-0.44 (0.33)

0.99 (p<0.0001)

-0.55 (0.43)

0.98 (p<0.0001)

<0.0001
0.96
(p<0.0001)
0.99
(p<0.0001)

0.93

0.10

12

0.30 (.)

0.74
(p=0.0060)

-1.09 (.)

0.98 (p<0.0001)

-0.37 (.)

0.91 (p<0.0001)

NHW

23

0.75 (0.51)

-2.57 (0.58)

0.98 (p<0.0001)

-1.31 (0.32)

0.96 (p<0.0001)

NHB

8

-0.18 (0.10)

-15.43 (0.68)

0.98 (p<0.0001)

-5.43 (0.30)

0.98 (p<0.0001)

Other race

4

3.30 (.)

0.98
(p<0.0001)
0.97
(p<0.0001)
0.99
(p=0.0080)

0.90 (.)

0.99 (p=0.0046)

-0.53 (.)

0.99 (p<0.0089)

0.96 (p<0.0001)

-1.57 (0.28)

0.96 (p<0.0001)

p-value
Country of
Birth
United
States

0.21

39

0.57 (0.45)

0.73

0.98
(p<0.0001)

-3.55 (0.52)

105

Table A10. (con’t)
Height (cm)
n
M or other

8

2

9-12

12

13-14

13

15-18

11

19-21

9

2.46 (.)

0.80
(p=0.0177)

Mean
differences
-0.38 (.)

BMI (kg/m )

Pearson r

Mean differences

0.98 (p<0.0001)

Pearson r

-0.70 (.)

0.70 (p=0.0531)

-2.40 (.)

.

0.16
.

-3.26 (.)

0.51 (.) 0.85(p=0.000
5)
1.55 (0.38)
0.98
(p<0.0001)
0.82 (0.36)
0.93
(p<0.0001)
0.71 (0.60)
0.92
(p=0.0005)
0.90

-4.93 (.)

0.99 (p<0.0001)

-1.77 (.)

-6.38 (0.84)

0.97 (p<0.0001)

-3.13 (0.51)

0.98
(p<0.0001)
0.97 (p<0.0001)

-3.27 (0.95)

0.99 (p<0.0001)

-1.42 (0.39)

0.98 (p<0.0001)

-0.15 (0.24)

0.99 (p<0.0001)

-0.37 (0.11)

0.98 (p<0.0001)

-7.29 (0.33)

0.98 (p<0.0001)

-2.26 (0.19)

0.98 (p<0.0001)

-2.00 (0.21)

0.96 (p<0.0001)

-1.20 (0.40)

0.89 (p=0.0002)

0.68 (.)

0.99 (p<0.0001)

-0.38 (.)

0.97 (p=0.0010)

-7.24 (.)

0.99 (p=0.0003)

-3.49 (.)

0.96 (p=0.0089)

-1.17 (0.07)

0.99 (p<0.0001)

-0.19 (0.16)

0.99 (p<0.0001)

16

-0.97 (0.20)

1=<PIR<1.9

11

1.55 (1.16)

2=<PIR<2.9

6

1.73 (.)

3=<PIR<4.9

5

1.85 (.)

5=<PIR

9

-0.80 (0.71)

p-value

Pearson r

0.28

p-value
Education
0-8

p-value
Family PIR
PIR <1

Mean
differences
2.52 (.)

2

Weight (kg)

0.81

0.95
(p<0.0001)
0.92
(p<0.0001)
0.84
(p=0.0369)
0.99
(p=0.0004)
0.94
(p=0.0001)

.

0.27

0.67
106

Table A11. Sensitivity, specificity, and predictive value positive (PVP) and predictive value
negative (PVN) of self-reported BMI with multiple imputed BMI
BMI categories
Validity
Sensitivity
Specificity
PVP
PVN
Underweight
81.43%
91.06%
56.94%
94.81%
Normal
93.04%
72.01%
70.55%
93.49%
Overweight
47.39%
90.88%
62.66%
92.34%
Obese
81.21%
97.29%
91.90%
93.19%

107

Table A12. Correlations and mean difference between self-reported and multiple imputation (MI)
BMI by socio-demographic characteristics of pregnant women
2

BMI (kg/m )

All

2790

Self-reported
BMI
25.92 (0.60)

p-value
Age
18-29

1990

25.41 (0.64)

30-41

800

26.96 (1.14)

n

p-value
Race/Ethnicity
MA or O H

26.31 (0.70)

NHW

1380

25.26 (0.53)

NHB

380

28.70 (2.30)

Other race

160

26.22 (4.04)

M or other

26.70
(0.64)

Mean
differences
-0.79 (0.07)

26.20
(0.69)
27.74
(1.24)

-0.79 (0.08)

27.06
(0.82)
26.02
(0.58)
29.53
(2.37)
27.25
(4.07)

-0.75 (0.14)

26.91
(0.78)
25.71
(1.31)

-0.77 (0.07)

22.90
(1.24)
26.77
(1.19)
27.84
(0.96)
27.94
(1.20)
25.51
(1.19)

-0.40 (0.18)

-0.78 (0.11)

Pearson r
0.97
(p<0.0001)
0.95
(p<0.0001)
0.94
(p<0.0001)

0.94
870

p-value
Country of Birth
United States

MI BMI

-0.77 (0.08)
-0.83 (0.10)
-1.03 (0.04)

0.94
(p<0.0001)
0.93
(p<0.0001)
0.98
(p<0.0001)
0.92
(p<0.0001)

0.29
2090

26.15 (0.72)

700

24.82 (1.21)

p-value
Education
0-8

0.51
200

22.51 (1.06)

9-12

530

25.80 (1.22)

13-14

610

27.12 (0.93)

15-18

810

26.99 (1.16)

19-21

640

24.87 (1.10)

p-value

0.58

-0.89 (0.17)

0.95
(p<0.0001)
0.94
(p<0.0001)

108

-0.98 (0.10)
-0.73 (0.08)
-0.95 (0.07)
-0.64 (0.13)

0.92
(p<0.0001)
0.96
(p<0.0001)
0.94
(p<0.0001)
0.95
(p<0.0001)
0.92
(p<0.0001)

Table A12. (con’t)
2

BMI (kg/m )
n

Selfreported
BMI

MI BMI

Mean
differences

Pearson r

Family PIR
PIR <1

720

27.37 (1.51)

28.33 (1.54)

-0.97 (0.10)

1=<PIR<1.999

600

25.40 (0.87)

26.33 (0.93)

-0.94 (0.11)

2=<PIR<2.999

390

26.53 (1.09)

27.46 (1.17)

-0.93 (0.12)

3=<PIR<4.999

600

27.32 (0.89)

27.95 (0.98)

-0.63 (0.12)

5=<PIR

480

22.96 (0.85)

23.49 (0.90)

-0.53 (0.10)

p-value

0.96
(p<0.0001)
0.93
(p<0.0001)
0.96
(p<0.0001)
0.93
(p<0.0001)
0.89
(p<0.0001)

0.021

mean (SEM)
a
Regression analysis was used to determine self-reported height is different from measured
height
b
General linear model was used to examine mean differences mean height difference by
maternal characteristics

109

Table A13. Mean (SEM) self-reported weight 1 year ago and weight by pregnancy status and
intentional weight lose
‘SelfTrying
Selfreported
to lose
Selfreported
Wt 1 yr
weight
reported
n
Wt 1 yr
ago’ –
Pearson r
in the
Wt
ago
‘Selfpast
reported
year
Wt’
Pregnant (missing=21)
Yes
83
73.99
77.63
-3.64
0.94
(2.77)
(2.81)
(0.59)
(p<0.0001)
62.07
62.52
-0.45
0.97
No 175
(1.33)
(1.53)
(0.38) (p<0.0001)
Yes 222
74.14
77.57
-3.43
0.94
Non-Pregnant
(1.57)
(1.49)
(0.42)
(p<0.0001)
(missing=61)
No

206

64.93
(1.50)

mean (SEM)

110

66.68
(1.81)

-1.75
(0.52)

0.95
(p<0.0001)

Table A14. Mean (SEM) self-reported weight 1 year ago, weight, and measured weight by
pregnancy status and intentional weight lose
Self-reported
Self-reported
Measured Wt
n
Wt 1 yr ago
Wt
All
489
72.73 (1.28)
73.44 (1.23)
75.03 (1.34)
Trying to lose weight in the past
222
74.14 (1.57)
77.57 (1.49)
79.07 (1.51)
year (yes)
Trying to lose weight in the past
206
64.93 (1.50)
66.68 (1.81)
68.16 (2.03)
year (no)
Missing
61
st

Pregnant women in 1 trimester
Trying to lose weight in the past
year (yes)
Trying to lose weight in the past
year (no)
Missing
mean (SEM)

47

76.74 (5.89)

78.03 (4.96)

81.11 (5.14)

16

76.09 (3.68)

81.25 (2.30)

84.73 (2.32)

29

65.49 (2.06)

65.85 (2.14)

69.29 (3.18)

2

111

Table A15. Comparison of mean (SEM) random value by multiple imputation (MI) methods and
st
actual measured BMI for randomly selected 10 pregnant women in 1 trimester
Mean SEM
MI BMI 1
30.17 1.32
Observed BMI 1
30.09 1.23
MI BMI 2
Observed BMI 2

33.72
37.35

0.98
0.99

MI BMI 3
Observed BMI 3

29.79
30.99

2.54
2.98

MI BMI 4
Observed BMI 4

32.31
32.89

1.04
0.74

MI BMI 5
Observed BMI 5

32.44
32.73

1.01
1.18

Random 6
Observed BMI 6

28.45
29.18

0.32
0.24

Random 7
Observed BMI 7

28.85
28.22

1.43
1.40

Random 8

30.67

1.04

Observed BMI 8

30.95

1.29

112

Table A16. Calculated pregnant women’s measured BMI by multiple imputations drawn from
non-pregnant women’s self-reported and measured BMI data
Self-reported BMI
Measured BMI
Mean
Mean (SD)
Pregnant women
279
25.92 (0.60)
29.41 (1.82)*
Non-pregnant women 489
27.50 (0.41)
28.27 (0.48)
*Multiple Imputed Estimates

113

Table A17. Sociodemographic characteristics of outliers above 95% confidence intervals of
height difference between self-reported height and measured height in pregnant and nonpregnant women
Pregnant women (n=46)
Non-pregnant women (n=86)
Age Groups
No. Wt. No.
%
Age Groups
No.
Wt. No.
%
18-29
18
89910
68.6
18-29
20
515615
42.2
30-41
5
41246
31.4
30-41
23
707431
57.8
Total
23 131156 100.0
Total
43 1223046
100.0
Race/ethnicity
MA or O H
NHW
NHB
Total

No. Wt. No.
18
99147
1
6119
4
25890
23 131156

Education
Levels
0-8
9-12
13-14
15-18
19-21
Total

No.

Country of
Birth
United States
M or other
Total

No.

Family poverty
income ratio
(PIR)
PIR <1
1=<PIR<1.9
2=<PIR<2.9
3=<PIR<4.9
5=<PIR
Total

No.

5
9
5
3
1
23

7
16
23

9
9
3
1
1
23

%
75.6
4.7
19.7
100.0

Race/ethnicity
MA or O H
NHW
NHB
Other race
Total

No.
26
4
11
2
43

Wt. No.
536892
270941
282907
132305
1223045

%
43.9
22.2
23.1
10.8
100.0

Wt. No.

%

No.

Wt. No.

%

27485
62589
16573
16647
7862
131156

21.0
47.7
12.6
12.7
6.0
100.0

Education
Levels
0-8
9-12
13-14
15-18
19-21
Total

213115
283719
262608
319381
144222
1223046

17.4
23.2
21.5
26.1
11.8
100.0

Wt. No.

%

No.

Wt. No.

%

23235
107921
131156

17.7
82.3
100.0

Country of
Birth
United States
M or other
Total

545781
677265
1223046

44.6
55.4
100.0

Wt. No.

%

No.

Wt. No.

%

50512
42478
24185
7862
6119
131156

38.5
32.4
18.4
6.0
4.7
100.0

Family poverty
income ratio
(PIR)
PIR <1
1=<PIR<1.9
2=<PIR<2.9
3=<PIR<4.9
5=<PIR
Total

579062
282156
69414
220442
71971
1223046

47.3
23.1
5.7
18.0
5.9
100.0

114

7
14
8
12
2
43

17
26
43

22
11
4
5
1
43

Table A17. (con’t)
Pregnant women (n=46)
Previous
No. Wt. No.
birth(s)
1
7
36506
2
7
38948
3
2
21026
4+
1
6045
Total
17
102526
Frequency Missing = 6

%
35.6
38.0
20.5
5.9
100.0

Non-pregnant women (n=86)
Previous
No.
Wt. No.
birth(s)
1
14
430402
2
13
344157
3
7
190819
4+
6
159497
Total
40 1124875
Frequency Missing = 3

115

%
38.3
30.6
17.0
14.2
100.0

Table A18. Response bias of odds of under-reporting vs. accurate-reporting and over-reporting vs. accurate group in height in
pregnant women (n=279)
Polytomous regression
Predictor
Under-reporting
Over-reporting
vs. accurate reporting
vs. accurate reporting
Unadjusted OR
pAdjusted OR
pUnadjusted OR
pAdjusted OR
(95% CI)
value (95% CI)
valu (95% CI)
value (95% CI)
e
Age
18-29 vs.
1.47 (0.77-2.82)
0.23
0.98 (0.40-2.40)
0.97 1.07 (0.30-3.85)
0.90
0.85 (0.23-3.08)
30-41
1
1
1
1
Country of Birth
Mexico or other
1.24 (0.38-4.04)
0.71
0.68 (0.21-2.20)
0.52 2.78 (0.92-8.40)
0.06
2.39 (0.78-7.30)
countries vs.US
1
1
1
1
Education Level
0-8 vs.
1.33 (0.11-16.06) 0.82
0.91 (0.06-13.50) 0.95 2.18 (0.28-16.80) 0.45
1.30 (0.18-8.98)
15-18
1
1
1
1
9-12
5.25 (0.55-49.58) 0.14
3.51 (0.29-42.0)
0.32 7.05 (1.98-25.11) 0.002 4.25 (1.1415.80)
13-14
1.68 (0.42-6.69)
0.45
1.56 (0.34-7.18)
0.56 2.75 (0.59-12.65) 0.19
2.43 (0.4213.96)
19-21
0.52 (0.10-2.76)
0.44
0.77 (0.13-4.46)
0.77 2.04 (0.51-8.16)
0.31
3.06 (0.6015.58)
Family PIR
<1 vs.
4.41 (1.60-12.13) 0.004 3.10 (0.58-16.58) 0.18 1.22 (0.45-3.32)
0.69
0.92 (0.22-3.72)
3-4.9
1
1

116

pvalue

0.81

0.12

0.78
0.03
0.31
0.17

0.90

Table A18. (con’t)
Polytomous regression
Predictor

1-1.9
2-2.9
≥5
Race/Ethnicity
Mexican
American vs.
Non-Hispanic
White
Other Hispanic

Under-reporting
vs. accurate reporting
Unadjusted OR
pUnadjusted OR
(95% CI)
value (95% CI)
4.48 (1.35-14.79) 0.01
3.65 (0.8715.26)
0.30 (0.08-1.09)
0.06
0.28 (0.090.82)
0.86 (0.10-7.03)
0.89
0.94 (0.127.43)
2.74 (1.01-7.46)
1

0.04

2.20 (0.43-11.20)

1.53 (0.465.07)
1

pvalue
0.07

Over-reporting
vs. accurate reporting
Unadjusted OR
pUnadjusted OR
(95% CI)
value (95% CI)
1.41 (0.49-4.05)
0.51
1.07 (0.30-3.78)

pvalue
0.91

0.02

0.79 (0.20-3.02)

0.73

0.68 (0.12-3.84)

0.66

0.95

0.24 (0.08-0.71)

0.009

0.22 (0.07-0.65)

0.006

0.47

3.18 (1.24-8.18)
1

0.01

1.68 (0.54-5.21)
1

0.36

0.33

1.17 (0.130.88
0.65 (0.09-4.59)
0.66
0.37 (0.03-4.10)
0.42
9.95)
Non-Hispanic
4.52 (1.29-15.79) 0.01
2.64 (0.820.10
2.80 (0.89-8.79)
0.07
2.20 (0.66-7.29)
0.19
Black
8.48)
Other Race
0.37 (0.05-2.54)
0.31
0.23 (0.020.18
2.53 (0.26-24.29)
0.41
1.52 (0.19-11.85) 0.68
1.98)
(Under-reporting group : <25% of height difference: Ht difference<-0.82; Accurate reporting group: 25-75% of height difference: 0.82≤Ht Difference<-1.98; Over-reporting: >75% of height difference: 1.98≤Ht difference)

117

Table A19. Response bias of odds of under-reporting vs. accurate-reporting and over-reporting vs. accurate group in height in nonpregnant women (n=489)
Polytomous regression
Predictor
Under-reporting
Over-reporting
vs. accurate reporting
vs. accurate reporting
Unadjusted OR
pAdjusted OR
pUnadjusted OR
pAdjusted OR
p(95% CI)
value (95% CI)
value (95% CI)
value (95% CI)
value
Age
18-29 vs.
1.18 (0.70-1.97) 0.52
1.09 (0.67-1.78)
0.71
1.19 (0.77-1.84)
0.42
1.04 (0.60-1.80) 0.87
30-41
1
1
1
1
Country of Birth
Mexico or other
1.04 (0.53-2.04) 0.90
1.06 (0.53-2.12)
0.86
2.44 (1.17-5.09)
0.01
2.11 (0.85-5.19) 0.10
countries vs.US
1
1
1
1
Education Level
0-8 vs.
0.53 (0.13-2.09) 0.37
0.34 (0.09-1.23)
0.10
2.19 (1.17-4.09)
0.01
0.64 (0.27-1.49) 0.30
15-18
1
1
1
1
9-12
1.11 (0.43-2.82) 0.82
0.89 (0.31-2.51)
0.82
1.47 (0.78-2.78)
0.22
0.94 (0.43-2.06) 0.88
13-14
1.05 (0.50-2.20) 0.89
0.91 (0.41-1.99)
0.81
0.73 (0.41-1.28)
0.27
0.57 (0.29-1.14) 0.11
19-21
0.91 (0.49-1.68) 0.77
1.11 (0.62-1.99)
0.70
0.54 (0.23-1.28)
0.16
0.66 (0.26-1.64) 0.37
Family PIR
<1 vs.
1.32 (0.71-2.46) 0.37
1.51 (0.80-2.82)
0.19
2.45 (1.14-5.27)
0.02
2.19 (0.93-5.13) 0.06
3-4.9
1
1
1
1
1-1.9
1.10 (0.48-2.51) 0.81
1.20 (0.56-2.53)
0.63
1.38 (0.67-2.85)
0.37
1.25 (0.58-2.69) 0.56
2-2.9
1.32 (0.54-3.18) 0.53
1.38 (0.57-3.34)
0.47
1.01 (0.44-2.29)
0.97
1.00 (0.46-2.20) 0.98

118

Table A19. (con’t)
Polytomous regression
Predictor

≥5
Race/Ethnicity
Mexican
American vs.
Non-Hispanic
White
Other Hispanic
Non-Hispanic
Black
Other Race

Under-reporting
vs. accurate reporting
Unadjusted OR
pUnadjusted OR
(95% CI)
value (95% CI)
0.59 (0.17-2.08) 0.42
0.54 (0.14-2.11)

pvalue
0.38

Over-reporting
vs. accurate reporting
Unadjusted OR
pUnadjusted OR
(95% CI)
value (95% CI)
0.62 (0.16-2.35)
0.48
0.59 (0.15-2.25)

pvalue
0.44

1.45 (0.61-3.46)
1

0.39

1.28 (0.44-3.74)
1

0.64

3.23 (1.50-6.95)
1

0.002

1.49 (0.66-3.37)
1

0.09

0.56 (0.13-2.34)
1.10 (0.55-2.19)

0.42
0.77

0.49 (0.07-3.36)
0.93 (0.48-1.79)

0.47
0.84

0.91 (0.25-3.27)
1.31 (0.78-2.19)

0.88
0.29

0.32 (0.08-1.22)
0.99 (0.54-1.82)

0.09
0.99

1.33 (0.40-4.33)

0.63

1.33 (0.35-5.00)

0.66

0.89 (0.26-3.00)

0.85

0.61 (0.22-1.70)

0.34

119

Table A20. Response bias of odds of under-reporting v. accurate-reporting and over-reporting vs. accurate group in weight in nonpregnant women (n=489)
Polytomous regression
Predictor
Under-reporting
Over-reporting
vs. accurate reporting
vs. accurate reporting
Unadjusted OR
pAdjusted OR
pUnadjusted OR pAdjusted OR
p(95% CI)
value (95% CI)
value (95% CI)
value
(95% CI)
value
Age
18-29 vs.
0.90 (0.55-1.48)
0.68
0.79 (0.52-1.17)
0.24
1.32 (0.68-2.57) 0.40
1.28 (0.67-2.46) 0.44
30-41
1
1
1
1
Country of Birth
Mexico or other
countries vs.US

0.99 (0.46-2.13)
1

0.99

0.88 (0.36-2.12)
1

0.78

1.57 (0.96-2.55)
1

0.06

1.64 (0.95-2.83)
1

0.07

Education Level
0-8 vs.
15-18
9-12
13-14
19-21

0.39 (0.11-1.30)
1
1.28 (0.59-2.75)
1.25 (0.64-2.46)
0.50 (0.22-1.16)

0.12

0.24 (0.05-1.06)
1
1.05 (0.42-2.58)
1.14 (0.57-2.28)
0.58 (0.24-1.39)

0.06

1.07 (0.39-2.98)
1
1.80 (0.78-4.14)
1.08 (0.48-2.42)
1.25 (0.68-2.32)

0.88

0.59 (0.18-1.94)
1
1.44 (0.55-3.78)
0.92 (0.46-1.86)
1.48 (0.69-3.17)

0.38

0.52
0.50
0.10

120

0.91
0.69
0.22

0.16
0.84
0.46

0.45
0.83
0.30

Table A20. (con’t)
Polytomous regression
Under-reporting
vs. accurate reporting
Unadjusted OR
pAdjusted OR
(95% CI)
value (95% CI)
Family PIR
<1 vs.
3-4.9
1-1.9
2-2.9
≥5
Race/Ethnicity
Mexican
American vs.
Non-Hispanic
White
Other Hispanic
Non-Hispanic
Black
Other Race

0.96 (0.45-2.07)
1
0.98 (0.40-2.42)
0.53 (0.22-1.27)
0.21 (0.06-0.68)

0.93

pvalue
0.96

0.97
0.15
0.009

1.01 (0.44-2.33)
1
1.03 (0.44-2.40)
0.52 (0.21-1.27)
0.21 (0.06-0.76)

1.26 (0.70-2.26)
1

0.43

1.08 (0.52-2.22)

1.14 (0.36-3.56)
1.20 (0.61-2.37)

0.81
0.58

1.00 (0.29-3.44)

0.99

Over-reporting
vs. accurate reporting
Unadjusted OR pAdjusted OR
(95% CI)
value (95% CI)
1.39 (0.67-2.87)
1
1.20 (0.44-3.24)
0.80 (0.21-3.10)
0.69 (0.35-1.36)

0.37

0.83

0.97 (0.33-2.80)
1.03 (0.48-2.19)
1.06 (0.30-3.6 7)

121

pvalue
0.39

0.71
0.75
0.28

1.44 (0.62-3.36)
1
1.22 (0.48-3.10)
0.81 (0.21-3.00)
0.64 (0.36-1.22)

1.75 (0.98-3.13)
1

0.05

1.32 (0.57-3.03)

0.50

0.95
0.93

1.17 (0.37-3.69)
1.74 (0.83-3.65)

0.78
0.14

0.85 (0.18-3.90)
1.52 (0.68-3.39)

0.83
0.30

0.92

1.66 (0.56-4.87)

0.35

1.49 (0.51-4.35)

0.45

0.94
0.15
0.01

0.66
0.75
0.18

Table A21. Response bias of odds of under-reporting vs. accurate-reporting and over-reporting
vs. accurate group in BMI in non-pregnant women (n=489)
Predictor

Age
18-29
vs. 30-41
Country of Birth
Mexico or other
countries vs.US
Education Level
0-8
vs. 15-18
9-12
13-14
19-21
Family PIR
<1 vs.
3-4.9
1-1.9
2-2.9
≥5
Race/Ethnicity
Mexican American
vs.
Non-Hispanic White
Other Hispanic
Non-Hispanic Black
Other Race

Under-reporting
vs. accurate reporting
Unadjusted OR
p-value
(95% CI)

Over-reporting
vs. accurate reporting
Unadjusted OR
p-value
(95% CI)

0.97 (0.73-1.30)
1

0.87

1.77 (0.99-3.16)
1

0.05

1.40 (0.80-2.47)
1

0.23

1.16 (0.59-2.27)
1

0.6

0.29 (0.12-0.74)
1
1.24 (0.53-2.87)
1.25 (0.69-2.26)
0.90 (0.53-1.53)

0.009

0.26 (0.07-0.96)

0.04

0.60
0.45
0.92

1.63 (0.63-4.19)
0.90 (0.53-1.53)
0.96 (0.42-2.15)

0.30
0.72
0.15

1.39 (0.57-3.39)
1
1.52 (0.64-3.57)
0.71 (0.28-1.84)
0.24 (0.08-0.75)

0.46

1.09 (0.60-1.99)
1
1.17 (0.58-2.37)
0.81 (0.55-1.20)
0.51 (0.26-1.00)

0.76

1.28 (0.44-3.74)
1

0.64

1.49 (0.66-3.37)
1

0.33

0.49 (0.07-3.36)
0.93 (0.48-1.79)

0.47
0.84

0.32 (0.08-1.22)
0.99 (0.54-1.82)

0.09
0.99

1.33 (0.35-5.00)

0.66

0.61 (0.22-1.70)

0.34

0.33
0.49
0.01

122

0.64
0.31
0.05

Table B1. Healthy Eating Index 2005 score by gestational weight gain groups

Parameter
Healthy Eating Index
2005
Total Fruit
Whole Fruit
Total Vegetables
Dark Green & Orange
Veg & Legumes
Total Grains
Whole Grains
Milk
Meat & Beans
Oils
Saturated Fat
Sodium
Calories from Solid Fat,
Alcohol & Added Sugar
(SoFAAS)
Average of two 24h recalls
a

Inadequate or
Excessive GWG
(n=185)
Mean
SEM

Adequate GWG
(n=53)
Mean

SEM

P-value a

54.2

1.5

61.7

4.2

0.21

3.0
2.4
2.8

0.3
0.3
0.2

3.7
3.3
3.4

0.3
0.5
0.2

0.10
0.22
0.37

0.9

0.1

1.5

0.5

0.37

4.4
1.3
6.5
7.8
5.5
5.6
3.9

0.1
0.1
0.4
0.3
0.3
0.3
0.3

4.2
1.5
7.1
8.0
5.8
6.3
3.8

0.1
0.5
0.8
0.4
0.2
0.7
0.5

0.46
0.74
0.55
0.79
0.61
0.50
0.91

10.2

0.7

13.1

1.8

0.30

T-test

123

Table B2. Distribution of HEI 2005 by gestational weight gain groups
Inadequate GWG
(N=51)

Adequate GWG
(N=53)

Excessive GWG
(n=134)

HEI 2005

No.

Wt. No.

Wt. %

No.

Wt. No.

%

No.

Wt. No.

Wt. %

Total

< 50

18

321668

45.6

17

207795

30.7

42

775225

32.2

77

50-80

31

370956

52.6

34

424838

62.8

91

1627221

67.6

156

80

2

12530

1.8

2

44339

6.5

1

4718

0.2

5

Total

51

705154

100.0

53

676972

100.0

134

2407163

100.0

238

Average of two 24h recalls
a

Chi-square

124

p-value
0.05

a

BIBLIOGRAPHY

125

BIBLIOGRAPHY

Abu-Saad, K., Fraser, D. (2010). Maternal nutrition and birth outcomes. Epidemiologic Reviews,
32, 5-25.
Food and Drug Administration. (1996) Food standards: amendment of standards of identity for
enriched grain products to require addition of folic acid: Vol. 61 (pp. 8781-8807).
Allen, L. H. (2000). Anemia hemoglobin concentration and birth weight. American Journal of
Clinical Nutrition, 71(5 Suppl), 1280S-1284S.
Allen, L. H. (2005). Multiple micronutrients in pregnancy and lactation: an overview. American
Journal of Clinical Nutrition 81(5), 1206S-1212S.
Alwan, N. A., Greenwood, D. C., Simpson, N. A., McArdle, H. J., Godfrey, K. M., Cade, J. E.
(2011). Dietary iron intake during early pregnancy and birth outcomes in a cohort of
British women. Human Reproduction, 26(4), 911-919.
Anderson, A. S. (2001). Symposium on 'nutritional adaptation to pregnancy and lactation'.
Pregnancy as a time for dietary change? Proceedings of the Nutrition Society 60(4), 497504.
Asbee, S. M., Jenkins, T. R., Butler, J. R., White, J., Elliot, M., Rutledge, A. (2009). Preventing
excessive weight gain during pregnancy through dietary and lifestyle counseling: a
randomized controlled trial. American Journal of Obstetrics and Gynecology, 113, 305312.
Baeten, J. M., Bukusi, E. A., Lambe, M. (2001). Pregnancy complications and outcomes among
overweight and obese nulliparous women. American Journal of Public Health, 91(3),
436-440.
Baeten JM, B. E., Lambe M. (2001 ). Pregnancy complications and outcomes among overweight
and obese nulliparous women. American Journal of Public Health 91(3), 436-440.
Bailey, L. B. (2000). New standards for dietary folate intake in pregnant women. American
Journal of Clinical Nutrition, 71(5 Suppl), 1304S-1307S.
Barger, M. K. (2010). Maternal nutrition and perinatal outcomes Journal of Midwifery &
Women's Health, 55(6), 502-511.
Basiotis, P. P., Center for Nutrition Policy and Promotion (U.S.); et al. (2002). The healhty
eating index : 1999-2000. Washington, D.C.: Center for Nutrition Policy and Promotion,
U.S. Dept. of Agriculture.

126

Bothwell, T. H. (2000). Iron requirements in pregnancy and strategies to meet them. American
Journal of Clinical Nutrition 72(1 Suppl), 257S-264S.
Bowers, K., Tobias, D. K., Yeung, E., Hu, F. B., Zhang, C. (2012). A prospective study of
prepregnancy dietary fat intake and risk of gestational diabetes. American Journal of
Clinical Nutrition 95(2), 446-453
Bowers, K., Tobias, D. K., Yeung, E., Hu, F. B., Zhang, C. (2012). A prospective study of
prepregnancy dietary fat intake and risk of gestational diabetes. American Journal of
Clinical Nutrition, 95(2), 446-453.
Britten, P., Marcoe, K., Yamini, S., Davis, C. (2006). Development of food intake patterns for
the MyPyramid Food Guidance System. Journal of Nutrition Education and Behavior,
38(6 Suppl), S78-92.
Brunner Huber, L. R. (2007). Validity of self-reported height and weight in women of
reproductive age. Maternal Child Health Journal 11(2), 137-144.
Bukowski, R., Malone, F. D., Porter, F. T., Nyberg, D. A., Comstock, C. H., Hankins, G. D.,
Eddleman, K., Gross, S. J., Dugoff, L., Craigo, S. D., Timor-Tritsch, I. E., Carr, S. R.,
Wolfe, H. M., D'Alton, M. E. (2009). Preconceptional folate supplementation and the risk
of spontaneous preterm birth: a cohort study. PLoS Med, 5(6).
Butte, N. F., Hopkinson, J. M., Mehta, N., Moon, J. K., Smith, E. O. . (1999). Adjustments in
energy expenditure and substrate utilization during late pregnancy and lactation.
American Journal of Clinical Nutrition, 69, 299-307.
Canada, H. (1997). Regulations amending the Food and Drug Regulations Can Gaz Part I (Vol.
131, pp. 3702-3737).
Canada., H. (1997). Regulations amending the Food and Drug Regulations (1066). In C. G. P. I
(Ed.), (Vol. 131).
Cedergren, M. (2006). Effects of gestational weight gain and body mass index on obstetric
outcome in Sweden. International Journal of Gynecology & Obstetrics, 93(3), 269-274.
Chapin, R. E., Robbins, W. A., Schieve, L. A., Sweeney, A. M., Tabacova, S. A., Tomashek, K.
M. (2004). Off to a good start: the influence of pre- and periconceptional exposures,
parental fertility, and nutrition on children's health. Environmental Health Perspectives,
112(1), 69-78.
Christian, P., Khatry, S. K., Katz, J., Pradhan, E. K., LeClerq, S. C., Shrestha, S. R., Adhikari, R.
K., Sommer, A., West, K. P. . (2003). Effects of alternative maternal micronutrient
supplements on low birth weight in rural Nepal: double blind randomised community
trial. BMJ, 326(7389), 1-6.
127

Chu, S. Y., D'Angelo, D. V. (2009). Gestational weight gain among US women who deliver
twins, 2001-2006. American Journal of Obstetrics and Gynecology, 200(4), 390.e391396.
Claesson, I. M., Sydsjo, G., Brynhildsen, J., et al. (2008). Weight gain restriction for obese
pregnant women: a case-control intervention study. BJOG, 115, 44-50.
Cogswell, M. E., Parvanta, I., Ickes, L., Yip, R., Brittenham, G. M. (2003). Iron supplemnetation
during pregnancy, anemia, and birth weight: a randomized controlled trial. American
Journal of Clinical Nutrition, 78(4), 773-781.
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological
Measurement, 20(1), 37-46.
Craig, B. M., Adams, A. K. (2009). Accuracy of body mass index categories based on selfreported height and weight among women in the United States. Maternal Child Health
Journal, 13(4), 489-496.
Deierlein, A. L., Siega-Riz, A. M., Herring, A. (2008). Dietary energy density but not glycemic
load is associated with gestational weight gain. American Journal of Clinical Nutrition,
88(3), 693-699.
Finer, L. B., Zolna, M. R. (2011). Unintended pregnancy in the United States: incidence and
disparities, 2006. Contraception, 84(5), 478-485.
Fowles, E. R. (2002). Comparing pregnant women's nutritional knowledge to their actual dietary
intake. MCN: The American Journal of Maternal/Child Nursing, 27(3), 171-177.
Gillman, M. W., Rifas-Shiman, S., Berkey, C. S., Field, A. E., Colditz, G. A. (2003). Maternal
gestational diabetes, birth weight, and adolescent obesity. Pediatrics., 111(3), 221-226.
Gillum, R. F., Sempos, C. T. (2005). Ethnic variation in validity of classification of overweight
and obesity using self-reported weight and height in American women and men: the
Third National Health and Nutrition Examination Survey. Nutrition Journal, 6(4), 27-34.
Grandjean, P., Bjerve, K. S., Weihe, P., et al. (2001). Birthweight in a fishing community:
significance of essential fatty acids and marine food contaminants. International Journal
of Epidemiology, 30(6), 1272-1278.
Guenther, P. M., Reedy, J., Krebs-Smith, S. M. (2008). Development of the Healthy Eating
Index-2005. Journal of American Dietetic Association, 108(11), 1896-1901.
Guldner, L., Monfort, C., Rouget, F., et al. (2007). Maternal fish and shellfish intake and
pregnancy outcomes: a prospective cohort study in Brittany, France. Journal of
Epidemiology & Community Health 6(33).
128

Hedderson, M. M., Gunderson, E. P., Ferrara, A. (2010). Gestational weight gain and risk of
gestational diabetes mellitus. Obstetrics & Gynecology, 115(3), 597-604.
Hytten, F. E., Chamberlain, G. (1991). Clinical Physiology in Obstetrics. Oxford: Blackwell
Scientific Publications.
Inskip, H. M. (2009). Women’s compliance with nutrition and lifestyle recommendations before
pregnancy: general population cohort study. BMJ, 338, b481.
IOM. (2009). Weight gain during pregnancy: reexamining the guidelines. Washington, DC: The
National Academy Press.
IOM, National Academy of Science. (2005). Dietary Reference Intakes for Energy,
Carbohydrate, FIber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids.
Washington, DC: National Academy Press.
Kac, G., Bení M. H., Velá
cio,
squez-Melé
ndez, G., Valente, J. G., Struchiner, C. J. (2004).
Breastfeeding and postpartum weight retention in a cohort of Brazilian women. American
Journal of Clinical Nutrition, 79(3), 487-493.
Kaiser, L., Allen, L. H; American Dietetic Association. (2008). Position of the American Dietetic
Association: nutrition and lifestyle for a healthy pregnancy outcome. Journal of
American Dietetic Association, 108(3), 553-561.
Kaiser L, A. L. A. D. A. (2008). Position of the American Dietetic Association: nutrition and
lifestyle for a healthy pregnancy outcome. Journal of American Dietetic Association,
108(3), 553-561.
Kennedy, E. T., Ohls, J., Carlson, S., Fleming, K. (1995). The Healthy Eating Index: design and
applications. Journal of American Dietetic Association, 95(10), 1103-1108.
Kim, S. Y., Dietz, P. M., England, L., Morrow, B., Callaghan, W. M. (2007). Trends in
prepregnancy obesity in nine states, 1993-2003. Obesity (Silver Spring), 15(4), 986-993.
Kinnunen, T. I., Pasanen, M., Aittasalo, M., et al. (2007). Preventing excessive weight gain
during pregnancy: a controlled trial in primary health care. European Journal of Clinical
Nutrition, 61, 884-891.
Kuczmarski, M. F., Kuczmarski, R. J., Najjar, M. (2001). Effects of age on validity of selfreported height, weight, and body mass index: findings from the Third National Health
and Nutrition Examination Survey, 1988-1994. Journal of American Dietetic Association,
101(1), 28-34.
Lagiou, P., Tamimi, R. M., Mucci, L. A., Adami, H. O., Hsieh, C. C., Trichopoulos, D. (2004).
Diet during pregnancy in relation to maternal weight gain and birth size. European
129

Journal of Clinical Nutrition, 58(2), 231-237.
Landis, J. R., Koch, G. G. . (1977). The measurement of observer agreement for cateogorical
data. Biometrics, 33(1), 159-174.
Laraia, B., Messer, L., Evenson, K., Kaufman, J. S. (2007). Neighborhood factors associated
with physical activity and adequacy of weight gain during pregnancy. Journal of Urban
Health, 84(6), 793-806.
Lederman, S. A., Alfasi, G., Deckelbaum, R. J. (2002). Pregnancy-associated obesity in black
women in New York City. Maternal Child Health Journal, 6(1), 37-42.
Lee, D. H., Shin, A., Kim, J., Yoo, K. Y., Sung, J. (2011). Validity of self-reported height and
weight in a Korean population. Journal of Epidemiology, 21(1), 30-36.
Lucas, M., Dewailly, E., Muckle, G., Ayotte, P., Bruneau, S., Gingras, S., Rhainds, M., Holub, B.
J. (2004). Gestational age and birth weight in relation to n-3 fatty acids among lnuit
(Canada). Lipids, 39(7), 617-626.
Nyholm, M., Gullberg, B., Merlo, J., Lundqvist-Persson, C., Rå
stam, L., Lindblad, U. (2007).
The validity of obesity based on self-reported weight and height:
Implications for population studies. Obesity (Silver Spring), 15(1), 197-208.
Oken, E., Kleinman, K. P, Olsen, S. F., et al. (2004). Associations of seafood and elongated n-3
fatty acid intake with fetal growth and length of gestation: results from a US pregnancy
cohort. American Journal of Epidemiology, 160(8), 774-783.
Oken, E., Ning, Y., Rifas-Shiman, S. L., Rich-Edwards, J. W., Olsen, S. F., Gillman, M. W.
(2007). Diet during pregnancy and risk of preeclampsia or gestational hypertension. Ann
Epidemiology, 17(9), 663-668.
Olafsdottir, A. S., Skuladottir, G. V., Thorsdottir, I., Hauksson, A., Steingrimsdottir, L. (2006).
Maternal diet in early and late pregnancy in relation to weight gain. Int J Obes (Lond).
30(3), 492-499.
Olsen, S. F., Halldorsson, T. I., Willett, W. C. (2007). Milk consumption during pregnancy is
associated with increased infant size at birth: prospective cohort study. American Journal
of Clinical Nutrition, 86(4), 1104-1110.
Olson, C. M., Strawderman, M. S. (2003). Modifiable behavioral factors in a biopsychosocial
model predict inadequate and excessive gestational weight gain. Journal of American
Dietetic Association, 103(1), 48-54.
Organization, W. H. (2010). Global recommendations on physical activity for health. Geneva,
Switzerland.

130

Park, S., Sappenfield, W. M., Bish, C., Bensyl, D. M., Goodman, D., Menges, J. (2011).
Reliability and validity of birth certificate prepregnancy weight and height among women
enrolled in prenatal WIC program: Florida, 2005. Maternal Child Health Journal, 15(7),
851-859.
Pfeiffer, C. M., Johnson, C. L., Jain, R. B., Yetley, E. A., Picciano, M. F., Rader, J. I., Fisher, K.
D., Mulinare, J., Osterloh, J. D. (2007). Trends in blood folate and vitamin B-12
concentrations in the United States, 1988 2004. American Journal of Clinical Nutrition,
86(3), 718-727.
Pick, M. E., Edwards, M., Moreau, D., Ryan, E. A. (2005). Assessment of diet quality in
pregnant women using the Healthy Eating Index. Journal of American Dietetic
Association, 105(2), 240-246.
Prentice, A. M., Spaaij, C. J., Goldberg, G. R., Poppitt, S. D., van Raaij, J. M., Totton, M.,
Swann, D., Black, A. E. . (1996). Energy requirements of pregnant and lactating women.
European Journal of Clinical Nutrition, 50, S82-S111.
Prevention., C. f. D. C. a. (2000). Folate status in women of childbearing age—United States. In
M. M. M. W. Rep (Ed.), (Vol. 49, pp. 962-965).
Rifas-Shiman, S. L., Rich-Edwards, J. W., Willett, W. C., Kleinman, K. P., Oken, E., Gillman,
M. W. (2006). Changes in dietary intake from the first to the second trimester of
pregnancy. Paediatric and Perinatal Epidemiology, 20(1), 35-42.
Rifas-Shiman, S. L., Rich-Edwards, J. W., Kleinman, K. P., Oken, E., Gillman, M. W. (2009).
Dietary quality during pregnancy varies by maternal characteristics in Project Viva: a US
cohort. Journal of American Dietetic Association, 109(6), 1004-1011.
Rode, L., Hegaard, H. K., Kjaergaard, H., Mø
ller, L. F., Tabor, A., Ottesen, B. (2007).
Association between maternal weight gain and birth weight. Obstetrics & Gynecology,
109(6), 1309-1315.
Rogers, I., Emmett, P., Ness, A., et al. (2004). Maternal fish intake in late pregnancy and the
frequency of low birth weight and intrauterine growth retardation in a cohort of British
Infants. Journal of Epidemiology & Community Health, 58(6), 486-492.
Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Surveys. New York: John Wiley &
Sons.
Saldana, T. M., Siega-Riz, A. M., Adair, L. S. (2004). Effect of macronutrient intake on the
development of glucose intolerance during pregnancy. American Journal of Clinical
Nutrition, 79(3), 479-486.

131

Schieve, L. A., Cogswell, M. E., Scanlon, K. S. (1998). Trends in pregnancy weight gain within
and outside ranges recommended by the Institute of Medicine in a WIC population.
Maternal Child Health Journal, 2(2), 111-116.
Scholl, T. O. (2005). Iron status during pregnancy: setting the stage for mother and infant.
American Journal of Clinical Nutrition, 81(5 Suppl), 1218S-1222S.
Shapira, N. (2008). Prenatal nutrition: a critical window of opportunity for mother and child.
Womens Health, 4(6), 639-656.
Shirazian, T., Monteith, S., Friedman., F., Rebarber., A. (2010). Lifestyle modification program
decreases pregnancy weight gain in obese women. American Journal of Perinatology, 27,
411-414.
Siega-Riz, A. M., Savitz, D. A., Zeisel, S. H., Thorp, J. M., Herring, A. (2004). Second trimester
folate status and preterm birth. American Journal of Obstetrics and Gynecology, 191(6),
1851-1857.
Sparks, J. W., Girard, J. R., Battaglia, F. C. (1980). An estimate of the caloric requirements of
the human fetus. Biol Neonate Biol Neonate, 38, 113-119.
Spencer, E. A., Appleby, P. N., Davey, G. K., Key, T. J. (2002). Validity of self-reported height
and weight in 4808 EPIC-Oxford participants. Public Health Nutrition, 5(4), 561-565.
Spencer, E. A., Appleby, P. N., Davey, G. K., Key, T. J. (2002). Validity of self-reported height
and weight in 4808 EPIC-Oxford participants. Public Health Nutrition, 5(4), 561-565.
Steer, P. J. (2000). Maternal hemoglobin concentration and birth weight. American Journal of
Clinical Nutrition, 71(5 Suppl), 1285S-1287S.
Stotland, N. E., Haas, J. S., Brawarsky, P., Jackson, R. A., Fuentes-Afflick, E., Escobar, G. J.
(2005). Body mass index, provider advice, and target gestational weight gain. Obstetrics
& Gynecology, 105(3), 633-638.
Streuling, I., Beyerlein, A., von Kries, R. (2010). Can gestational weight gain be modified by
increasing physical activity and diet counseling? A meta-analysis of interventional trials.
American Journal of Clinical Nutrition, 92, 678-687.
Strychar, I. M., , Chabot, C., Champagne, F., Ghadirian, P., Leduc, L., Lemonnier, M. C.,
Raynauld, P. (2000). Psychosocial and lifestyle factors associated with insufficient and
excessive maternal weight gain during pregnancy. Journal of American Dietetic
Association, 100(3), 353-356.

132

Tobias, J. H., Steer, C. D., Emmett, P. M., Tonkin, R. J., Cooper, C., Ness, A. R.; ALSPAC
study team. (2005). Bone mass in childhood is related to maternal diet in pregnancy.
Osteoporosis International, 16(12), 1731-1741.
Tsigga, M., Filis, V., Hatzopoulou, K., Kotzamanidis, C., Grammatikopoulou, M. G. (2011).
Healthy Eating Index during pregnancy according to pre-gravid and gravid weight status.
Public Health Nutrition, 14(2), 290-296.
van Eijsden, M., Hornstra, G., van der Wal M. F. et al. (2008). Maternal n-3, n-6, and trans fatty
acid profile early in pregnancy and term birth weight: a prospective cohort study.
American Journal of Clinical Nutrition, 87(4), 887-895.
Viswanathan, M., Siega-Riz, A. M., Moos, M-K., Deierlein, A., Mumford, S., Knaack, J., Thieda,
P., Lux, L. J., Lohr, K. N. . (2008). Outcomes of Maternal Weight Gain, Evidence
Report/Technology Assessment No. 168. (Prepared by RTI International-University of
North Carolina Evidence-based Practice Center under Contract No. 290-02-0016.) In A.
P. N. 08-E009. (Ed.). Rockville, MD: Agency for Healthcare Research and Quality
Watanabe, H., Fukuoka, H., Sugiyama, T., et al. (2008). Dietary folate intake during pregnancy
and birth weight in Japan. European Journal of Nutrition, 47(6), 341-347.
Widen, E., Siega-Riz, A. M. (2010). Prenatal Nutrition: A Practical Guide for Assessment and
Counseling. Journal of Midwifery & Women's Health, 55(6), 540-549.
Widen, E., Siega-Riz, A. M. (2010). Prenatal nutrition: a practical guide for assessment and
counseling. Journal of Midwifery & Women's Health, 55(6), 540-549.
Yun, S., Zhu, B. P., Black, W., Brownson, R. C. (2006). A comparison of national estimates of
obesity prevalence from the behavioral risk factors surveillance system and the National
Health and Nutrition Examination Survey. International Journal of Obesity, 30(1), 164170.

133