HOW MOTHER-FATHER RELATIONSHIP RELATES TO FETAL GROWTH RESTRICTION (FGR) IN BLACK INFANTS By Rosemary Iganya Adaji A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Epidemiology – Doctor of Philosophy 2023 ABSTRACT Background and objectives: Black infants in the United States experience disproportionately higher rates of fetal growth restriction (FGR) at birth (i.e., reduced birth weight relative to gestational age) than White infants, and infants of other racial/ethnic groups. FGR at birth is a leading cause of infant morbidity and mortality. The racial disparity in FGR is not fully explained by maternal morbidity (e.g., hypertension), socio-economic factors, or adequacy of prenatal care. The direct effect of paternal factors on FGR generally has been explored, but little attention has been given to how these factors intersect with maternal factors to influence FGR. Studies linking mother-father relationships to FGR have broadly focused on low birthweight (LBW; <2,500 grams). Findings have shown that pregnant women whose partners are involved in their lives are less likely to have a LBW baby. However, LBW is heterogenous and can include both FGR and preterm birth (PTB); the pathways to both outcomes also differ and should be assessed distinctly. The mother-father relationship context in Black families has not been reported in the literature on FGR. Evidence supporting the hypothesized pathways by which the mother-father relationship may impact FGR has, however, been reported. A growing number of publications report that depression and depressive symptoms increase risk of FGR with several papers including Black women or exclusive to Black women. Further, while such work is not extensive, reports do link the mother-father relationship, measured in divergent ways, to depression and depressive symptoms in pregnancy among all women as well as in Black women. Another hypothesized pathway is via marital/cohabitation status. Some studies report higher risks of a LBW infant among unmarried or non-cohabiting women, but studies focusing on FGR either exclusively in Black births or in a cohort of Black births are rare. Thus, the objectives of this dissertation are to 1) determine how measures of mother-father relationship (e.g., conflict, support, involvement) relate to FGR in Black infants, 2) determine if the associations between mother-father relationship and FGR are mediated by maternal depressive symptoms, and 3) determine if marital/cohabitation status moderates the association between mother-father relationship and FGR. Methods: Data from the Biosocial Impacts on Black Births (BIBB) prospective cohort study were used. Participants completed questionnaires at 19-29 weeks’ gestation including six measures of woman’s relationship with the FOB: 1) contact with FOB, 2) FOB involvement, 3) overall relationship, 4) change in relationship prior to pregnancy to during pregnancy, 5) social support from FOB, and 6) conflict with FOB. Latent class analysis was used to identify and classify the relationship construct. To address aim 1, we fit a multivariate logistic regression model using the relationship construct created to assess its association with FGR. For aim 2, we conducted mediation analysis, based on the potential outcomes’ framework, and for aim 3, we used the same multivariate regression model in aim 1, and included an interaction term to test for moderation by marital/cohabitation status. Results: we found that 1) the mother-father relationship construct can be classified as good, conflictual and no relationship, and women in the conflictual and no relationship groups have higher odds of having a growth restricted infant compared to those in good relationship, 2) there was no evidence of mediation by maternal depressive symptoms, and 3) marital/cohabitation status did not moderate the association between the mother-father relationship construct and FGR. Conclusion: The findings reported provide evidence suggesting that the mother-father relationship is an important correlate of fetal growth restriction in Black families. More work is needed to explore pathways by which this association occurs. This dissertation is dedicated to my parents (in loving memory). To my mother who worked very hard to ensure that I got a good college education. And to my dad, who constantly prayed for me at every challenging point in this journey. I love you both. iv ACKNOWLEDGMENTS I would like to acknowledge and thank my advisor, Dr. Misra, whose mentorship, and support has made this dissertation possible. Dr. Misra took a chance on me to join the BIBB team, even when she was not privy to my background or competencies. However, I am grateful for that chance as it has afforded me several opportunities to learn, practice good research, and set me on a very promising career path. I hope that I can carry forward her work and team work ethic as I progress in my career. Further, I would like to thank each member of the BIBB team for providing data support and helping me navigate the data easily. To Dr. Pathak, I am grateful for all the personal, academic, and emotional support she gave me throughout my graduate work. She ais an integral part of this success, and I will forever remain grateful. To Dr. Todem, I am thankful for his constant support and teaching, and for continuously seeing and encouraging the potential in me even when I do not see it for myself. He was also an integral part of my journey every step of the way. I hope that I can embody his selflessness as a model for my professional career going forward, and I hope that to make him proud in all that I do. I am also thankful to Dr. Barondess’ efforts in how much he fights for and ensures students’ success. Lastly to my family and friends who have unconditionally supported me throughout my journey, thank you for being my army and ensuring that I do not fight alone. v TABLE OF CONTENTS LIST OF SYMBOLS ....................................................................................................................viii LIST OF ABBREVIATIONS .........................................................................................................ix CHAPTER 1: INTRODUCTION ....................................................................................................1 1.1 Overview ................................................................................................................................1 1.2 Significance............................................................................................................................2 1.3 Dissertation Organization .....................................................................................................3 CHAPTER 2: BACKGROUND LITERATURE ............................................................................4 2.1 Overview of Fetal Growth Restriction ...................................................................................4 2.2 Etiology and Risk Factors for Fetal Growth Restriction .......................................................5 2.3 Race as a Risk Factor for FGR ..............................................................................................6 2.4 Mother-Father of the Baby Relationship and FGR ..............................................................10 2.5 Knowledge Gaps ..................................................................................................................13 2.6 Significance of this Dissertation Research ..........................................................................14 CHAPTER 3: METHODS.............................................................................................................19 3.1 Study Population...................................................................................................................19 3.2 The BIBB Study Design ......................................................................................................19 3.3 Procedures and Data Management .......................................................................................20 3.4 Conceptual Model ................................................................................................................22 3.5 General Analytical Strategy .................................................................................................24 CHAPTER 4: MOTHER-FATHER RELATIONSHIP AND FETAL GROWTH RESTRICTION IN BLACK BIRTHS: A LATENT CLASS ANALYSIS (MANUSCRIPT 1) .............................27 4.1 Introduction...........................................................................................................................27 4.2 Methods ................................................................................................................................29 4.3 Results ..................................................................................................................................33 4.4 Discussion ............................................................................................................................36 4.5 Conclusion ...........................................................................................................................39 CHAPTER 5: MOTHER-FATHER RELATIONSHIP AND FETAL GROWTH RESTRICTION: A LATENT CLASS AND MEDIATION ANALYSIS (MANUSCRIPT 2) ....47 5.1 Introduction...........................................................................................................................47 5.2 Methods ................................................................................................................................49 5.3 Results ..................................................................................................................................54 5.4 Discussion ............................................................................................................................56 5.5 Conclusion ...........................................................................................................................59 CHAPTER 6: MOTHER-FATHER RELATIONSHIP AND FETAL GROWTH RESTRICTION IN BLACK BIRTHS: A LATENT CLASS AND MODERATION ANALYSIS (MANUSCRIPT 3) ....................................................................................................................................................67 vi 6.1 Introduction...........................................................................................................................67 6.2 Methods ................................................................................................................................69 6.3 Results ..................................................................................................................................74 6.4 Discussion ............................................................................................................................76 6.5 Conclusion ...........................................................................................................................78 CHAPTER 7: DISCUSSION.........................................................................................................86 7.1 Summary of Findings............................................................................................................86 7.2 Review of Limitations ..........................................................................................................87 7.3 Directions for Future Research .............................................................................................89 REFERENCES ..............................................................................................................................91 APPENDIX A: BERKMAN’S CONCEPTUAL MODEL OF SOCIAL NETWORKS AND HEALTH .....................................................................................................................................105 APPENDIX B: SPEARMAN CORRELATION COEFFICIENT OF RELATIONSHIP INDICATORS .............................................................................................................................106 vii LIST OF SYMBOLS ≥ Less than or equal to = Equal to < Less than viii LIST OF ABBREVIATIONS adjOR Adjusted Odds Ratio AGA Appropriate for Gestational Age BIBB Biosocial Impact on Black Births CES-D Center for Epidemiologic Studies Depression CI Confidence Interval FGR Fetal Growth Restriction FOB Father of the Baby GA Gestational Age LCA Latent Class Analysis LBW Low Birth Weight LGA Large for Gestational Age LMP Last Menstrual Period PTB Preterm Birth SAS Statistical Analytical Software SES Socio-Economic Status SGA Small for Gestational Age ix CHAPTER 1: INTRODUCTION 1.1 Overview There is evidence linking social relationships to disease risk, morbidity, mortality, and health in general (Cassel, 1976; Cobb, 1976; House et al., 1988). Berkman’s conceptual model of how social networks impact health describes upstream and downstream factors that influence this association (Berkman et al., 2000) (Appendix A, Figure 1.1.1). The upstream factors focus on the broader contexts of social structural conditions (e.g., racism, socioeconomic factors) while the downstream factors focus on the psychosocial mechanisms that enhance susceptibility to disease, mortality, and/or morbidity. Social relationships are one of the psychosocial mechanisms, and the pathways by which it can influence health are multifaceted. While the association between social relationships and health has been found for a range of health outcomes, perinatal research literature remains sparse on the topic area. This dissertation investigates how social relationships relate to adverse birth outcomes in an understudied population at increased risk: pregnant Black women. The specific aims of this dissertation will examine the pathways by which mother-father relationships relate to Fetal Growth Restriction (FGR) in Black births: Specific Aim 1: Determine how measures of mother-father relationship (e.g., conflict, support, involvement) relate to FGR in Black infants. Hypothesis: A negative mother-father relationship is associated with increased risk of FGR. Specific Aim 2: Determine if the associations between mother-father relationship and FGR are mediated by maternal depressive symptoms. Hypothesis: Maternal depressive symptoms mediate the association between negative mother-father relationships and risk of FGR. 1 Specific Aim 3: Determine if marital/cohabitation status moderates the association between mother-father relationship and FGR. Hypothesis: Marital/cohabitation status moderates the association between mother-father relationships and risk of FGR. In a broader context, this work is aimed at contributing to the body of scientific knowledge focused on assessing and addressing the persistent racial disparity in adverse birth outcomes experienced by Black women. 1.2 Significance 1.2.1 Scientific Significance This work is important because: 1) It gives context to quality of relationship. Studies have mostly focused on one or two measures of the mother-father relationship construct, as they relate to birth outcomes. This work will assess several measures of the mother-father relationship construct to provide a more comprehensive context. 2) It examines pathways to FGR relevant to Black families. The persistence of Black-White disparities in birth outcomes despite research and intervention efforts calls for new approaches. The complexities associated with Black family structures and intra-personal relationships are unique, with intricate processes families (Moynihan, 1965; Billingsley, 1992; Furstenberg Jr., 1988). Our focus on births to Black families offers the opportunity to assess within-group differences which could be useful for developing individual risk profiles for this subpopulation. 3) It examines indirect paternal effects on FGR. Direct paternal effects (i.e, age, phenotype, genetics) have been the focus of many studies linking the mother- father interplay to adverse birth outcomes like preterm birth (PTB) and low birthweight (LBW). The focus on the mother-father relationship dynamic provides other indirect ways by which fathers play a role in birth outcomes. 2 1.2.2 Public Health Significance Black infants experience the highest rates of fetal growth restriction (FGR) in the U.S compared to other racial/ethnic groups (Buck Louis et al., 2015; Frisbie et al., 1997; Zhang & Bowes, 1995). Also, compared with foreign-born African infants in the US, US-born Black infants have a higher FGR risk (M. S. Kramer et al., 2006). The focus of many public health interventions has been on upstream factors that result from social disadvantage, such as access to prenatal care. This work will provide insights for public health professionals on downstream pathways by which social disadvantage influences birth outcomes, thus leading to more targeted interventions. 1.3 Dissertation Organization This dissertation is organized into seven chapters. Chapter 1 above gives an overview of the dissertation objective, specific aims, and scientific and public health significance. Chapter 2 discusses relevant literature on FGR, the mother-father relationship, and the interplay between the two. Chapter 3 describes broadly the research approach and conceptual model guiding the dissertation. Further details will be outlined in Chapters 4 through 6, which address each specific aim, presenting three publishable manuscripts. Finally, in chapter 7, tying it all together, I will provide discussion of findings and the implications of this dissertation. 3 CHAPTER 2: BACKGROUND LITERATURE 2.1 Overview of Fetal Growth Restriction (FGR) Fetal growth restriction is a condition in which a fetus does not reach their predetermined intrauterine growth potential (Nardozza et al., 2017). It is also called intrauterine growth restriction (IUGR), and in some cases, the word ‘retardation’ is used in place of ‘restriction’. The clinical standard for assessing fetal size is a Doppler ultrasound in utero, while assessment of gestational age is based on menstrual history and ultrasound biometry (crown rump length). Diagnosis of FGR requires accurate estimation of gestational age. Hence, fetal growth rates are measured by relating the size of the infant to their expected gestational age, and not size by alone. When growth restriction is identified before 32 weeks, and in the absence of congenital anomalies, it is considered early-onset FGR, while growth restriction identified after 32 weeks is considered late onset FGR (Gordijn et al., 2016). The most widely used clinical and epidemiological measure for FGR is estimated fetal weight less than 10 th percentile for gestational age of a reference population, or small for gestational age (SGA) (“ACOG Practice Bulletin No. 204,” 2019). Information on birthweight and gestational age could also be obtained at birth, to determine SGA. Other less used measures found in epidemiology literature include birthweight adjusted for gestational age (GA), and birthweight ratio which is the ratio of the infants observed birthweight relative to GA, compared to the sex-specific mean birthweight relative to GA of a reference population (Frisbie et al., 1997). The adjustment for sex differences is based on evidence that male infants tend to be bigger than female infants and yet female infants survive equally well at the same birthweight (Wald et al., 1986). FGR is one of the most common pregnancy complications, occurs in about 5 to 10 percent of all pregnancies and is a leading cause of neonatal morbidity and mortality. (Frøen et al., 2004; Mook-Kanamori et al., 2010). 4 2.2 Etiology and Risk Factors for Fetal Growth Restriction The is no exact cause of FGR, however, there are three broad established risk factors that clinicians have regarded as attributable causes: fetal, placental, and maternal factors (Gardosi et al., 1992). While these are well known determinants of FGR, a substantial number of other factors remain unknown. Fetal factors: (i) Chromosomal abnormalities: specifically, trisomy, 13, 18 and 21suggested to contribute to 5 to 20 percent of FGR cases, especially early-onset FGR (Lin & Santolaya-Forgas, 1998). (ii) Genetic syndromes: genetic mutations have been shown to induce pre- and post-natal growth restriction (Abuzzahab et al., 2003; Meler et al., 2020; Netchine et al., 2011; Wit & Walenkamp, 2013) (iii) Intrauterine infections: these include but are not restricted to chromosomal ruptures, cytolysis, vascular endothelial lesions, and viruses that cause placentitis. Intrauterine infections are shown to be present in 5 to 10 percent of FGR cases (Neerhof, 1995). Placental factors: (i) Placental insufficiency: this is the most common risk factor for FGR. The placenta helps to transfer nutrients and gases from the mother to the fetus during pregnancy. It also helps to protect the fetus from pathogens. When the placenta fails to execute its functions (reduced uteroplacental perfusion), it can result in FGR. About 25 to 35 percent of FGR cases have been attributed to reduced uteroplacental perfusion associated with maternal vascular disease (Nardozza et al., 2017). 5 Maternal factors: (i) Morbidity: maternal morbidities such as congenital heart disease diabetes, renal disease, and hypertensive disorders especially pre-eclampsia, have been shown to increase the risk of FGR (Galan et al., 2001). (ii) Nutritional disorders: maternal chronic malnutrition before pregnancy has been suggested to result in about 40% of low birthweight infants which could increase the risk of having both a preterm and FGR infant (Schulz, 2010). (iii) Unhealthy behaviors: these include substance use including cigarette smoking and alcohol intake. Many studies have found a correlation between cigarette smoking during pregnancy and an increased risk of FGR (Cnattingius, 2004; Figueras et al., 2008; Janisse et al., 2014). (iv) Multiple gestation pregnancy: this is a situation where a woman is pregnant with more than one baby at a time (e.g., twins). About 15 to 30 percent of multiple gestation pregnancies, especially those of monochromatic twins result in FGR infants. The growth rates start out normal as those of singleton pregnancies up until the 20 th to 30th week, then begin to decline by 15 to 20 percent (Blickstein, 2004). Other underdiscussed factors are race related (e.g., maternal race) and include the intersection between social and environmental factors that induce stress exposures in pregnant Black women, leading to FGR. These race-related factors and the pathways by which they lead to FGR are the focus of this dissertation. 2.3 Race as a Risk Factor for FGR Racial differences in growth restricted births have been reported in the United States, with Black infants having disproportionately higher rates compared to White infants, and infants of other racial/ethnic groups (Buck Louis et al., 2015; Frisbie et al., 1997; M. S. Kramer et al., 2006; 6 Zhang & Bowes, 1995). A few studies have suggested developing a customized fetal growth standard specific to this population as they assume that Black infants are intrinsically (destined to be) smaller (Tarca et al., 2018). However, some observations about Black-White differences in birthweight relative to gestational age may challenge such approach. For example, patterns of racial differences in birthweight have varied over time, suggesting that Black infants are not intrinsically smaller to a fixed degree compared to other groups. (Kline et al., 1989). Existing strategies to bridge the racial disparity gap in adverse birth outcomes, including fetal growth restriction, have broadly focused on improvement in access to and utilization of prenatal care resources (World Health Organization, 2016). However, the persistence of disparity in birth outcomes indicates a need to shift the focus to other sources of inequities that go deeper than prenatal care, such as social disadvantage and structural racism. Researchers have hypothesized that racism, and social inequity are fundamental contributors to racial disparities in health outcomes (James, 2003; Link & Phelan, 1995). This notion is based on the evidence of a persistent relationship between socio-economic status (SES), social factors, and health outcomes; despite advancements in the study and elimination of risk factors related to or causing disease, disparities persist. Thus, they note that SES should be considered a fundamental cause of disease because it influences multiple risk factors, disease outcomes, and involves access to resources (e.g., money) that can be used to avoid risk or minimize adverse health outcomes once they occur (Link & Phelan, 1995). Socio-economic factors (e.g., income and education) are indicators of socio-economic advantage. While they may not have direct effects on fetal growth rate, they could be upstream antecedents of downstream factors that affect individual-level exposures and behaviors that lead to FGR (M. S. Kramer et al., 2000). The risk of FGR is well established to be highest among the 7 socially disadvantaged (Dominguez, 2011; Gavin et al., 2012; M. S. Kramer et al., 2000). Consequently, the literature on determinants of FGR disparities largely focuses on known, modifiable socio-economic risk factors (e.g., prenatal care access) that if controlled would eliminate the disparities. Yet despite addressing such factors, the disparities persist (Burris & Hacker, 2017; Goldfarb et al., 2018). A possible explanation is in the conceptual ambiguity of using complex constructs like SES as a composite variable to represent the experience of a certain race and/or ethnicity. The race effect is often assumed to equate the SES effect (i.e., controlling for SES eliminates the race effect) (M. S. Kramer et al., 2000; Meghani & Chittams, 2015). Further, the SES construct is operationalized simplistically in many studies, by including only a few measures such as income and education measured at one timepoint and adjusting for their effects. However, such an approach is insufficient as it does not provide information on relevant aspects of SES. For example, level of educational attainment does not give information about quality of education (i.e individual experience, prestigious vs. non-prestigious), place of education (i.e neighborhood characteristics) or trend effects (i.e., variation in meaning of education across cohorts) (Braveman et al., 2005). Further, evidence shows that different measures for socio- economic status operating at different levels (e.g individual, neighborhood) could affect health at different times in the life course, via different pathways, and could also have intergenerational effects (Adler et al., 1993; Marmot et al., 1997; Roux et al., 2001; Smith et al., 1998; Taylor et al., 1997). Such evidence warrants further evaluation of relevant aspects of SES and other potentially plausible race-specific indirect pathways (downstream factors) by which social disadvantage increases the risk of FGR within the Black pregnant population. To understand how downstream factors operate to influence FGR, we refer to the social-ecological model which explains the interplay of personal and environmental factors’ impact on health outcomes (Bronfenbrenner, 8 1986; McLeroy et al., 1988). According to this model, socio-economic factors lead to unhealthy behaviors, exposure to stress and psychological reactions that each may separately or together increase the risk of FGR. In this dissertation, the focus is mainly on maternal exposure to stress and psychologic reactions that can influence growth restricted births in Black women. Black women are more likely to be exposed to chronic stress (e.g., residential segregation), and exhibit negative emotional stress responses (e.g., depression) than White women (Collins & David, 2009; Dole et al., 2003; Hedegaard et al., 1993; Witt et al., 2015). These stressors have been reported to increase risk of adverse birth outcomes (Dole et al., 2004a; Dominguez et al., 2008; Laraia et al., 2006; Zuberi et al., 2016), For example, Black women are more likely to live in disadvantaged neighborhoods (i.e., high crime, disorder), and experiences racial discrimination compared to White women (Dominguez et al., 2008; Messer et al., 2006). In a case-control study of Black women in the United States (U.S), women who delivered preterm had more than twice the odds of high exposure to interpersonal racial discrimination (Rankin et al., 2011). These stress exposures have been noted to affect biological systems and processes throughout the life span, thus influencing health outcomes including birth outcomes. However, psychosocial resources such as social support have been found by some researchers to buffer the effects of stress, including studies of Black women (Giurgescu et al., 2006; Glazier et al., 2004; Nylen et al., 2013). Specifically, support from the father of the baby has been shown to be protective (Caldwell et al., 2018; Edwards et al., 2012; Feldman et al., 2000; Giurgescu et al., 2018). A few studies have also found that father’s involvement during pregnancy increased the woman’s likelihood of receiving early prenatal care (Martin et al., 2007). The level of support from the father of the baby is reported to be dependent on the type and quality of relationship he has with the pregnant mother of his child (Cowan & Cowan, 2000). The mother- 9 father relationship could be a missing element in understanding the factors that produce FGR and in potentially buffering impacts of other risk factors. 2.4 Mother-Father of the Baby Relationship and FGR The complexities associated with Black family structures and intra-personal relationships are unique, with intricate processes, and differ from those of non-Black families (Moynihan, 1965; Billingsley, 1992; Furstenberg Jr., 1988). For example, Black families tend to rely on extended family relations for support and have female single-parent family structures (Jarrett, 1994; H. McAdoo, 2007). There is the general notion of fathers’ role as providers, and contributors to the emotional, social, and economic development of the family. However, Black families are characterized by the stereotype of the absent father. Hence the role of the father is mostly overlooked in discussions of Black maternal and perinatal health (Bright & Williams, 1996). Emerging literature suggests a need to assess paternal roles in birth outcomes outside traditional contexts of family (e.g., marital status), and take into consideration the dynamics of Black family structures (Bird et al., 2000; Bloch et al., 2010; Misra et al., 2010; Padilla & Reichman, 2001; Teitler, 2001a). For example, while Black fathers are less likely than White and Hispanic fathers to marry their child’s mother, they continue to provide support through visitation, cohabitation, caretaking, financial, and in-kind support (Coles & Green, 2010). The direct effect of paternal factors (e.g., age, phenotype, genetics) on FGR generally has been explored, but little attention has been given to indirect paternal factors in the psychosocial domain, and how they intersect with maternal factors to influence FGR. Quality and type of mother-father relationship is one such psychosocial factor. Studies linking mother-father relationships to FGR have broadly focused on low birthweight (LBW; <2,500 grams). Findings have shown that pregnant women whose partners are involved in their lives are less likely to have 10 a LBW baby (Alio et al., 2010; Bird et al., 2000; Bloch et al., 2010; Meng & Groth, 2018, Norland, 2001). However, LBW is heterogenous and can include both FGR and preterm birth (PTB); the pathways to both outcomes also differ (M. S. Kramer et al., 2000), and should be assessed distinctly. Further, the mother-father relationship context in Black families has not been reported in the literature on FGR. There is no standard definition of mother-father relationship. Perinatal literature that have assessed this construct in samples including but not exclusive to Black families have used varying measures such as relationship quality (good, fair, poor), and financial support (Bloch et al., 2010), cohabitation status (Padilla & Reichman, 2001), marital and cohabitation status (Teitler, 2001a), relationship type (marital and cohabitation status) and duration (Bird et al., 2000), and absence of fathers name on birth certificate (Alio et al., 2010). Table 2.1.1 summarizes those studies that have examined mother-father relationship or paternal involvement in relation to birth outcomes. These studies fall short by focusing only on one or two measures and not looking simultaneously at several measures of relationship. Considering multiple dimensions could offer a better contextualization of the relationship construct. Further, given the racial differences in family relations, it is important to focus specifically on Black mother-father relationship dynamics to help us understand group specific differences that contribute to disparity in birth outcomes. 2.4.1 Pathways Relating Mother-Father Relationships to FGR Interpretations of findings linking the mother-father relationship with adverse birth outcomes is impeded by limited understanding of the underlying causal mechanisms. It is not fully understood how a negative mother-father relationship could lead to growth restricted births and other adverse birth outcomes. One hypothesis is that mother-father relationship influences FGR indirectly through maternal factors (Misra et al., 2010). For example, Hoffman and Hatch 11 (Hoffman & Hatch, 1996), in a systematic review found that among pregnant women with life stress, intimate partner support improved fetal growth. Further evidence supporting hypothesized pathways via maternal depression, and depressive symptoms has also been reported, with several papers including Black women (Bonari et al., 2004; Diego et al., 2009; Grote et al., 2010; Janssen et al., 2016), or exclusive to Black women (Giurgescu et al., 2018; Nutor et al., 2018; Phillips et al., 2010). It has been shown that about 20 percent of women experience clinically relevant depressive symptoms during pregnancy (Lahti et al., 2017; Molyneaux et al., 2014). The rates are higher for pregnant women in lower SES, and racial and ethnic minority groups (Katz et al., 2018), with over 40 percent reporting elevated symptoms of depression (Orr et al., 2007). Maternal depressive symptoms have been associated with smaller fetal head circumferences and lower fetal weights, with increased risk for being low birthweight (birthweight <2500 grams) (Diego et al., 2006, 2009; Field et al., 2006; Oberlander et al., 2006). Evidence for the biological mechanisms by which depression or depressive symptoms affects fetal growth is established in the literature. Depression and chronic stress could lead to hormonal dysregulation and changes in the hypothalamic pituitary adrenal axis (HPA) function, stimulating high cortisol production and release resulting in restriction in flow of nutrients and oxygen to the fetus (O’Donnell et al., 2009; Sarkar et al., 2008; Van den Bergh et al., 2005). Another biological mechanism is via glucocorticoid hormone imbalance in the maternal immune system, that may increase susceptibility to infections resulting in poor fetal growth (Marcus & Heringhausen, 2009; Orr et al., 2007; Reynolds, 2013). However, the evidence for socio-environmental mechanisms remains unclear. In a systematic review, it was reported that a problematic/conflictual or dissatisfied/poor relationship with the partner was a risk factor for onset of prenatal anxiety and depression (Biaggi et al., 2016). Although, all but one of the studies included mostly focused on White women who 12 did not reside in the U.S. The literature on maternal depressive symptoms relating mother-father relationship to FGR in Black births remains understudied. Another hypothesized pathway is via marital and cohabitation status. Several studies report higher risks for adverse birth outcomes among pregnant women with uninvolved partners, and unmarried women including those in non- cohabiting relationships (MacDorman et al., 2002; Mathews et al., 2002; S. Ventura, 1995; S. J. Ventura et al., 2000). The attributable risks are even higher for Black women as the likelihood of being unmarried and non-cohabiting is much greater, with about 70% of Black births occurring to unmarried, non-cohabiting women, compared to 28% of births among White women (Lloyd et al., 2021; Wildsmith et. al., 2018). Further, unmarried Black women are more likely to be socially disadvantaged (M. S. Kramer et al., 2000), experience depression or depressive symptoms (Diego et al., 2009), and engage in unhealthy behaviors (e.g., cigarette smoking, alcohol) during pregnancy, compared to White women (Padilla & Reichman, 2001; Teitler, 2001). While these factors (depressive symptoms, and marital and cohabitation status) have been hypothesized or shown to influence adverse birth outcomes, the literature on FGR, however, does not clearly discuss the mechanisms by which they influence FGR (i.e, confounding, mediation, or moderation). 2.5 Knowledge Gaps The key points from the discussions in this chapter are the following: 1) Fetal growth restriction is a public health problem, 2) There are racial disparities in the occurrence and consequences of FGR, with Black births experiencing disproportionately higher rates compared to other races, 3) Intervention efforts to eliminate persistent racial disparities have mostly focused on upstream factors such as socio-economic status, downstream factors may be worth shifting the focus, 4) Fathers play a role in pregnancy outcomes, 5) Psychosocial factors, specifically mother- 13 father of the baby relationship may prove important. Subsequently, the following knowledge gaps could be deduced: 1) It is unclear what factors are contributing to persistent Black-White-Other race disparities in birth outcomes, specifically, FGR, 2) The effect of mother-father of the baby relationship has been shown to effect adverse birth outcomes (e.g., preterm birth and low birthweight), but the effects on FGR remain unclear, 4) The approach to assessing mot her-father relationship is underdeveloped, the multidimensional nature of the relationship construct suggests the need for newer approaches, 5) Maternal depressive symptoms, and marital and cohabitation status are hypothesized pathways by which mother-father relationship relates to adverse birth outcomes but our understanding of their mechanisms of action in FGR is limited, and 6) Studies exclusive to Black families and Black births remain rare. 2.6 Significance of this Dissertation Research This dissertation research aims to address the knowledge gaps described above. The general objective is to assess pathways by which mother-father relationship may relate to FGR, with an overarching hypothesis that a negative mother-father relationship is associated with FGR; this association is mediated by depressive symptoms and differs by levels of marital and cohabitation status. To restate the three specific aims from chapter one: Specific Aim 1: Determine how measures of mother-father relationship (e.g., conflict, support, involvement) relate to FGR in Black infants. Hypothesis: A negative mother-father relationship is associated with increased risk of FGR. Specific Aim 2: Determine if the associations between mother-father relationship and FGR are mediated by maternal depressive symptoms. 14 Hypothesis: Maternal depressive symptoms mediate the association between negative mother- father relationships and risk of FGR. Specific Aim 3: Determine if marital/cohabitation status moderates the association between mother-father relationship and FGR. Hypothesis: Marital/cohabitation status moderates the association between mother-father relationships and risk of FGR. Figure 2.1.1 summarizes the proposed hypothesized pathways by which mother-father relationship relates to FGR. 15 16 Table 2.1.1: Studies Reporting Associations Between Mother-Father Relationship and Adverse Birth Outcomes Author Sample Relationship Birth Statistical Findings (Year) measure outcome Analyses (Bird et al., Hispanic, Marital/cohab LBW Multivariate ➢ Relationship type and duration were not 2000) Non- itation status, logistic associated with LBW. Hispanic duration of regression ➢ LBW was almost six times as likely among Black, Non- relationship Hispanic women in nonmarital, non-cohabiting Hispanic relationships as among those who were White/Other married. Among non-Hispanic white women, LBW was less likely among those in nonmarital, non- cohabiting relationships than among those who were married (Padilla & Mexican, Cohabitation LBW Stepwise ➢ Mothers in non-cohabiting romantic Reichman, Non- status, and logistic relationships with the father of the baby had 2001) Hispanic financial regression, significantly higher offs of LBW compared to White, other support incremental those who cohabit with the father of the baby. Hispanic multivariate Receiving monetary support from the baby’s models father had a negative effect on the offs of LBW (Teitler, Hispanic, Marital and LBW among Logistic ➢ The effects of relationship status on LBW are 2001b) African cohabitation others regression in the expected direction but not significant. American, status Birth outcome was worse for women in a White romantic, non-cohabiting relationship relative to those no longer in a relationship with the father. 17 Table 2.1.1 (cont’d) Author Sample Relationship Birth Statistical Findings (Year) measure outcome Analyses (Hohmann- White, Relationship PTB and Weighted ➢ Odds of PTB were higher for when only one Marriott, Black, quality, LBW multivariate partner or neither partner intended the 2009) Hispanic, marital status, logistic pregnancy. Asian/Pacific joint regression ➢ Risk of LBW was not associated with Islander, pregnancy pregnancy intentions, but with the father not Native intentions having discussed pregnancy with mother American Indian, and Mixed race (Alio et al., Black, Paternal LBW, very Multivariate ➢ Higher rates of LBE, VLBW, PTB, VPTB, and 2010) White, and involvement: LBW logistic SGA among father-absent births Hispanic Paternal PTB, very regression ➢ Black women with absent fathers had the women details on PTB highest rates of LBW, VLBW, PTB, VPTB, birth SGA and SGA. certificate (Bloch et al., Non- Mother-father LBW, among Multivariate ➢ The worse the quality of relationship the lower 2010) Hispanic relationship other analyses and the outcome, with dose-response associations White, quality among outcomes Poisson between the quality of relationship and Hispanic, unmarried regression birthweight among other outcomes and Non- women Hispanic Black and other pregnant women Abbreviations: LBW, low birthweight (<2500grams); VLBW, very low birthweight (<1500grams); PTB (<34 weeks gestation), preterm birth, VPTB; very preterm birth (28 to 32 weeks gestation); SGA, small for gestational age (BW <10 th percentile for GA). 18 CHAPTER 3: METHODS 3.1 Study Population This dissertation leverages a unique set of data from the Biosocial Impacts on Black Births (BIBB) study, to address the specific aims. Data from self-report questionnaires and medical record abstractions were collected for approximately 600 Black women from prenatal care sites in Michigan and Ohio. Our sample is comprised of women who had enrolled and delivered between December 18, 2017, and March 13, 2020, when COVID-19 pandemic restrictions were instituted. Structured surveys including valid and reliable instruments were used to collect data on maternal risk factors, emotional stress, social stressors, and psychosocial resources. 3.2 The BIBB Study Design The BIBB study is a prospective cohort study with a primary objective to determine how social stressors alter systemic inflammation during pregnancy and lead to preterm birth (PTB) in Black women. The study population comprises literate, self-identified Black or African American women 18-45 years old, and between 8-29 weeks of gestation from Detroit Michigan (MI) and Columbus Ohio (OH) metropolitan areas giving birth to singleton infants from three prenatal clinics: Providence, Southfield MI; St. John, Detroit, MI; The Ohio State University Wexner Medical Center, Columbus, OH. At enrollment, women provided data at up to 3 time points by completing questionnaires and providing saliva and blood specimens. A detailed protocol on the recruitment process can be found elsewhere (Vaughan et al., 2022). This dissertation will focus on questionnaire data collected from 19 through 29 weeks of pregnancy. Hence, a cross-sectional study design describes the sample for analysis. Questionnaires included information on psychosocial measures (mother-father relationship measures), emotional stress (measures of depressive symptoms) and maternal risk factors (SES; marital/cohabitation status etc). Birthweight 19 and gestational age assessments were abstracted from medical records. Data on covariates and potential confounders needed for the analyses are also available. The study sample for this dissertation will include 404 pregnant Black women. A summary of variables to be used in analyses is presented in Table 3.1.1. 3.3 Procedures and Data Management The BIBB study was approved by the institutional review boards at Wayne State University, and participating sites. A waiver was obtained by the principal investigators to access medical records of women, and to identify potential participants. Research staff approached and explained the study to eligible women and obtained informed consent from those who agreed to participate. Questionnaires were administered and a $30 gift card was provided to each woman who completed the questionnaire. Questionnaire data were entered into Qualtrics Research Suite, a web-based platform for creating surveys. Password-protected, customer-controlled survey data were captured in real-time and stored on Qualtrics’ secure and Transport Layer Security (TLS) encrypted servers. 20 Table 3.1.1: Variables and Operational measures Concepts/Variables Measures Description Maternal Risk Self-report Questionnaire Age, education, marital/cohabitation Factors status, household income, cigarette Socio-demographics, smoking unhealthy behaviors Emotional Stress Center for Epidemiological 20 items; range 0 – 60; scores >16 Depressive symptoms Studies Depression Scale represent clinically relevant depressive (CES-D) symptoms; Cronbach’s α=0.85-0.95 (ref: Mother-Father Two subscales: Relationship (1) support subscale; 5 items; range 0- Support Social Networks in Adult 25; higher score= father-mother Conflict Relations Questionnaire relationship is supportive; (2) conflict subscale; 4 items; range 0- 20; conflict in the father-mother relationship (e.g. mother/father criticizing the other); Cronbach’s α=0.71-0.94 for Black mothers and Frequency of contact fathers in pilot BIBB study. with FOB 1 item, “how often do you have contact FOB involvement with the FOB?” 1 item, “how often is the father of this FOB relationship baby involved with you before (and with MOB before and during) your pregnancy?” during pregnancy 2 items, 5-point likert scale response (1= very close, 2= somewhat close, 3= sometimes close/sometimes cold, 4= Change in somewhat cold, 5= very cold) relationship (5) Difference relationship before and during pregnancy Birth outcomes Birth weight Medical Records Birthweight (grams) Gestational age at Abstraction for mothers Gestational age at birth (days) Birth Birthweight percentiles FGR 21 3.4 Conceptual Model In figure 3.1.1, 3.1.2, and 3.1.3, we present the conceptual models guiding this dissertation. The models display the hypothesized pathways by which the mother-father relationship relates to FGR, for each specific aim. This dissertation is motivated by the ecological model (Bronfenbrenner, 1986) which states that both personal and environmental factors impact health outcomes. Concepts from the fundamental cause theory (Link & Phelan, 1995), and stress and emotions model (Folkman, 2013; Lazarus, 1999) will also be highlighted throughout this dissertation. Figure 3.1.1: Hypothesized pathway between mother-father relationship and FGR for Aim 1. In this model we assume that there is a direct association between mother-father relationship and FGR, but because this is a non-randomized study, confounding is bound to occur. We will consider multiple measures of mother-father relationship, outlined in Table 3.1.1, the maternal risk factors are confounders which need to be adjusted for, to estimate the true effect of mother-father relationship on FGR. This model represents a simplified ‘exposure-outcome’ relationship, as ideally, there are more complex hypothesized pathways by which mother-father relationship can influence FGR. 22 Figure 3.1.2: Hypothesized pathway between mother-father relationship and FGR for Aim 2. This model assumes the following 1) that there is a direct pathway between the mother- father relationship and FGR, 2) part of the effect of the mother-father relationship on FGR is due to the mediated process of depressive symptoms, and 3) these effects are subject to confounding by maternal risk factors. It is important to note that true and valid causal interpretations of mediation effects requires that all relevant confounding covariates must be measured and included in the causal mediation analysis. To this effect, there are four assumptions that need to be met to identify causal mediation effects, 1) no unmeasured ‘exposure’ to ‘outcome’ confounding, 2) no unmeasured mediator to outcome confounding, 3) no unmeasured exposure to mediator confounding and 4) no mediator to outcome confounder that is caused by exposure. However, because of the observational (non-randomized) nature of our study, some of these assumptions may be violated. The goal of specific aim 2 is not to infer true causal interpretations, but to assess a plausible pathway by which mother-father relationship can influence FGR. 23 Figure 3.1.3: Hypothesized pathway between mother-father relationship and FGR for specific Aim 3. Finally, in this model for specific aim 3, we assume that there is an association between mother-father relationship and FGR, however, the effect varies by levels of marital/cohabitation status (moderator). 3.5 General Analytical Strategy The analytic procedures used in this dissertation integrates multiple analytical techniques that test the robustness and sensitivity of our findings. Our main analytic methods are guided by the underlying hypothesis for each specific aim. These methods are further detailed in chapters 4 through 7. The methodology for obtaining the study population, independent and dependent variables, and covariates in each of these chapters will generally be similar for each specific aim. However, the models addressing each specific aim will differ. Briefly, to address specific aim 1, we will use latent class analysis (LCA) methods to derive a latent construct using the measures of mother-father relationship listed in Table 3.1.1, then assess associations with FGR. Latent class analysis is a statistical method for identifying and grouping individuals based on conditional probabilities. It belongs to a larger family of latent variable techniques called finite mixture models. The LCA model uses responses to a set of observed 24 measures or indicators to identify groups of people that have similar response patterns. There are two parameters of interest in a standard LCA model: i) Class membership probabilities (i.e., the relative size of each class), and ii) item response probabilities conditional on class membership. As an example, suppose that there are M binary observed latent class indicators, u1 , u2 , u3 …, uM, observed on n individuals, The LCA model assumes that M indicators denote measures of an unobserved latent class variable, c, that has k latent classes. The class membership probabilities in a given class k, (i.e., Pr(c = k), is denoted by πk , where K classes are mutually exclusive and exhaustive, with each individual in the population having membership in exactly one of the K latent classes (i.e., 𝛴𝜋𝑘 = 1). The model follows the conditional independence assumption for the set of M indicators such that any association observed among the indicators is entirely explained by the latent class variable, which once identified, the indicators are no longer correlated . The joint distribution of the indicators and latent class variable, c, under the conditional independence assumption is then specified as: 𝑃𝑟(𝑢1𝑖 , 𝑢 2𝑖 , … , 𝑢 𝑀𝑖 ) = ∑𝐾𝑘=1[𝜋𝑘 (∏𝑀 𝑚=1 𝑃𝑟 ( 𝑢 𝑚𝑖 |𝑐𝑖 = 𝑘) )] (EQ. 3.1) The LCA approach is considered in view of its ability to address correlation between measures, estimate combined effects of the relationship measures, and findings may be easier to interpret. Further details of the LCA procedure will be provided in chapters 4 through 6. The counterfactual framework for mediation analysis will be used to address specific aim 2. This approach allows for the quantification and estimation of total, direct and indirect (mediated) effects of the mother-father relationship on FGR. As an example, if the optimal number of classes determined by the LCA model in aim 1 is 2, we will consider a binary measure of Mother-Father relationship, ‘X’ conceptualized as ‘good’ and ‘poor’ relationship, in relation to our FGR outcome. 25 Formally, we denote Xi as the binary indicator for the relationship measure, with 1 representing good relationship, and 0, poor relationship. We also denote by Mij the observed level of depressive symptoms (mediator). Because depressive symptoms can be influenced by mother-father relationship, there exist two potential values Mij (0) and Mij (1), only one of which is observable at a time, i.e Mij (Xi) = Mij. For example, if a level of depressive symptoms is assigned to the measure of relationship, i.e Xi = 1, then we only observe Mij (0) and not Mij (1). Next, we define the infant’s potential outcome which we denote by Yij (𝑥, m) for relationship Xi = 𝑥, and depressive symptoms Mij = m. We further define the indirect effect of mother-father relationship mediated by depressive symptoms, and the direct effects as: indirect effect: 𝐼𝐸𝑖𝑗 (𝑥) = 𝑌𝑖𝑗 (𝑥 , 𝑀𝑖𝑗 (1))-𝑌𝑖𝑗 (𝑥 , 𝑀𝑖𝑗 (0)) (EQ 3.2) and direct effect: 𝐷𝐸𝑖𝑗 (𝑥) = 𝑌𝑖𝑗 (𝑥 , 𝑀𝑖𝑗 (1))-𝑌𝑖𝑗 (𝑥 , 𝑀𝑖𝑗 (0)), for 𝑥 = 0, 1, respectively. (EQ 3.3) Finally, the total effect is decomposed into the direct and indirect effect, i.e., total effect: TEij (x) = 𝐼𝐸𝑖𝑗 (𝑥) + 𝐷𝐸𝑖𝑗 (𝑥 ). (EQ 3.4) For specific aim 3, we will fit a multivariate logistic regression model using the latent construct created in aim 1. This model will include an interaction term which will assess the moderation effects of marital/cohabitation status. All models stated for each aim allow for covariate adjustment (i.e., controlling for confounders). Multiple imputation methods will be used to treat missing data and all analyses will be performed using SAS version 9.4 (Statistical Analysis Software, Cary, NC). 26 CHAPTER 4: MOTHER-FATHER RELATIONSHIP AND FETAL GROWTH RESTRICTION IN BLACK BIRTHS: A LATENT CLASS ANALYSIS (MANUSCRIPT 1) 4.1 Introduction Fetal growth restriction (FGR) is one of the most common adverse birth outcomes, and a leading cause of perinatal morbidity and mortality (Malhotra et al., 2019). FGR occurs when a fetus fails to reach their predetermined intrauterine growth potential (Nardozza et al., 2017). Infants that survive are at increased risk of having long term needs due to a range of morbid conditions including neurodevelopmental impairments, cardiovascular, and metabolic problems (Malhotra et al., 2019). Compared to other racial/ethnic groups, Black infants in the United States constantly experience disproportionately higher rates of FGR at birth (Buck Louis et al., 2015; Frisbie et al., 1997; Nasiri et al., 2020). Similarly, compared with foreign-born African infants in the US, US- born Black infants have a higher FGR risk (M. S. Kramer et al., 2006). Although maternal factors (e.g., maternal morbidity, socio-economic factors, adequacy of prenatal care) have been associated with FGR, these factors do not fully explain the FGR rates in Black infants (Burris & Hacker, 2017; Mutambudzi et al., 2017). Among plausible factors, psychosocial factors, specifically, relationship with the father of the baby have been posited to influence birth outcomes (Giurgescu & Misra, 2018; Mutambudzi et al., 2017; Uchino et al., 1996). Direct paternal effects (i.e, age, phenotype, genetics) have been the focus of many studies relating the mother-father interplay to adverse birth outcomes. Paternal support and involvement with the mother during pregnancy has also been shown to have protective effects on birth outcomes (Feldman et al., 2000; Meng & Groth, 2018; Misra et al., 2010). However, the mother-father 27 relationship context in Black families has not been reported in the literature on FGR. Beyond the sparse literature on FGR and mother-father relationship effects, the approach to analysis is underdeveloped. Several measures have been used in the perinatal research literature to define the mother-father relationship construct, including relationship quality (good, fair, poor) and financial support (Bloch et al., 2010), social support and conflict during pregnancy (Caldwell et al., 2018), and relationship status (married or not) (Padilla & Reichman, 2001). However, these studies fall short by focusing only on one or two measures and not looking simultaneously at several measures. The complexities associated with Black family structures and intra-personal relationships are unique, with intricate processes that differ from those of non-Black families (Moynihan, 1965; Billingsley, 1992; Furstenberg Jr., 1988). For example, Black families tend to rely on extended family relations for support and have female single-parent family structures (Jarrett, 1994; H. McAdoo, 2007). Further, mother-father relationships in Black pregnant partners is not often characterized by formal marital status, as seen from the high number of births to unmarried women (70 percent). While there is no standard measure of Mother-Father relationship, considering multiple dimensions in this subpopulation could offer a better contextualization of the relationship construct. The disproportionately higher rates of FGR in Black infants, and the inadequacy of maternal specific factors to explain this persistent disparity suggests a need for research on other influences on FGR in Black births. This study extends the literature on FGR in Black births in two ways: 1) we identify a construct of mother-father relationship by assessing several pregnant women’s self-reported measures of relationship 2) we assess how this mother-father relationship construct relates to FGR in black births. 28 4.2 Methods 4.2.1 Study Population This study included a subsample of 404 women pregnant, participating in the Biosocial Impacts on Black Births (BIBB) study, a prospective cohort study with enrollment and delivery between 2017-2020. Recruitment was from two sites: Detroit, Michigan (MI) and Columbus Ohio (OH) metropolitan areas, and from three prenatal clinics: Providence, Southfield MI; St. John, Detroit, MI; The Ohio State University Wexner Medical Center, Columbus, OH. Eligibility criteria included: English speaking, self-identified Black or African American women, singleton pregnancy, 18-45 years old, and between 8-29 weeks of gestation. At enrollment, women provided data at up to three time points by completing questionnaires and providing saliva and blood specimens. A detailed protocol on the recruitment process can be found elsewhere (Vaughan et al., 2022). In this study, we use a cross-sectional study design, focusing on questionnaire data collected from 19 through 29 weeks pregnancy, including measures of mother-father relationship, measures of depressive symptoms, maternal characteristics, birthweight, and gestational age estimates abstracted from medical records. 4.2.2 Measures Maternal characteristics Data on baseline maternal characteristics included age (<30years and ≥30 years), household income (<$10,000 and ≥$10,000) cigarette smoking (yes/no), and education (3.0.CO;2-Y James, S. A. (2003). Confronting the Moral Economy of US Racial/Ethnic Health Disparities. American Journal of Public Health, 93(2), 189–189. https://doi.org/10.2105/AJPH.93.2.189 Janisse, J. J., Bailey, B. A., Ager, J., & Sokol, R. J. (2014). Alcohol, Tobacco, Cocaine, and Marijuana Use: Relative Contributions to Preterm Delivery and Fetal Growth Restriction. Substance Abuse, 35(1), 60–67. https://doi.org/10.1080/08897077.2013.804483 Janssen, A. B., Capron, L. E., O’Donnell, K., Tunster, S. J., Ramchandani, P. G., Heazell, A. E. P., Glover, V., & John, R. M. (2016). Maternal prenatal depression is associated with decreased placental expression of the imprinted gene PEG3. Psychological Medicine, 46(14), 2999–3011. https://doi.org/10.1017/S0033291716001598 Jarrett, R. L. (1994). Living Poor: Family Life among Single Parent, African-American Women. Social Problems, 41(1), 30–49. https://doi.org/10.2307/3096840 Kahn, R. L., & Antonucci, T. C. (1980). Convoys over the life course: Attachment, roles, and social support. Life-Span Development and Behavior. https://scholar.google.com/scholar_lookup?title=Convoys+over+the+life+course%3A+att achment%2C+roles%2C+and+social+support&author=Kahn%2C+R.L.&publication_yea r=1980 Katz, J., Crean, H. F., Cerulli, C., & Poleshuck, E. L. (2018). Material Hardship and Mental Health Symptoms Among a Predominantly Low Income Sample of Pregnant Women Seeking Prenatal Care. Maternal and Child Health Journal, 22(9), 1360–1367. https://doi.org/10.1007/s10995-018-2518-x Kline, J., Stein, Z., & Susser, M. (1989). Conception to Birth: Epidemiology of Prenatal Development. Oxford University Press. Kramer, M. S., Ananth, C. V., Platt, R. W., & Joseph, K. (2006). US Black vs White disparities in foetal growth: Physiological or pathological? International Journal of Epidemiology, 35(5), 1187–1195. https://doi.org/10.1093/ije/dyl125 Kramer, M. s., Séguin, L., Lydon, J., & Goulet, L. (2000). Socio-economic disparities in pregnancy outcome: Why do the poor fare so poorly? Paediatric and Perinatal Epidemiology, 14(3), 194–210. https://doi.org/10.1046/j.1365-3016.2000.00266.x Lahti, M., Savolainen, K., Tuovinen, S., Pesonen, A.-K., Lahti, J., Heinonen, K., Hämäläinen, E., Laivuori, H., Villa, P. M., Reynolds, R. M., Kajantie, E., & Räikkönen, K. (2017). Maternal Depressive Symptoms During and After Pregnancy and Psychiatric Problems in Children. Journal of the American Academy of Child and Adolescent Psychiatry, 56(1), 30-39.e7. https://doi.org/10.1016/j.jaac.2016.10.007 97 Lanza, S. T., Bray, B. C., & Collins, L. M. (2012). An Introduction to Latent Class and Latent Transition Analysis. In I. Weiner (Ed.), Handbook of Psychology, Second Edition (p. hop202024). John Wiley & Sons, Inc. https://doi.org/10.1002/9781118133880.hop202024 Lanza, S. T., Tan, X., & Bray, B. C. (2013). Latent Class Analysis With Distal Outcomes: A Flexible Model-Based Approach. Structural Equation Modeling: A Multidisciplinary Journal, 20(1), 1–26. https://doi.org/10.1080/10705511.2013.742377 Laraia, B. A., Messer, L., Kaufman, J. S., Dole, N., Caughy, M., O’Campo, P., & Savitz, D. A. (2006). Direct observation of neighborhood attributes in an urban area of the US south: Characterizing the social context of pregnancy. International Journal of Health Geographics, 5(1), 11. https://doi.org/10.1186/1476-072X-5-11 Lausman, A., Kingdom, J., Gagnon, R., Basso, M., Bos, H., Crane, J., Davies, G., Delisle, M.-F., Hudon, L., Menticoglou, S., Mundle, W., Ouellet, A., Pressey, T., Pylypjuk, C., Roggensack, A., & Sanderson, F. (2013). Intrauterine Growth Restriction: Screening, Diagnosis, and Management. Journal of Obstetrics and Gynaecology Canada, 35(8), 741–748. https://doi.org/10.1016/S1701-2163(15)30865-3 Lazarus, R. S. (1999). Stress and emotion: A new synthesis (pp. xiv, 342). Springer Publishing Co. Lin, C.-C., & Santolaya-Forgas, J. (1998). Current concepts of fetal growth restriction: Part I. Causes, classification, and pathophysiology. Obstetrics & Gynecology, 92(6), 1044– 1055. https://doi.org/10.1016/S0029-7844(98)00328-7 Link, B. G., & Phelan, J. (1995). Social Conditions As Fundamental Causes of Disease. Journal of Health and Social Behavior, 80–94. https://doi.org/10.2307/2626958 Lloyd, C. M., Alvira-Hammond, M., Carlson, J., & Logan, D. (n.d.). Family, Economic, and Geographic Characteristics of Black Families with Children. 10. Loggins Clay, S., & Andrade, F. C. D. (2015). Role of stress in low birthweight disparities between black and white women: A population-based study. Journal of Paediatrics and Child Health, 51(4), 443–449. https://doi.org/10.1111/jpc.12735 MacDorman, M. F., Minino, A. M., Strobino, D. M., & Guyer, B. (2002). Annual summary of vital statistics—2001. Pediatrics, 110(6), 1037–1052. https://doi.org/10.1542/peds.110.6.1037 Malakouti, S. K., Pachana, N. A., Naji, B., Kahani, S., & Saeedkhani, M. (2015). Reliability, validity and factor structure of the CES-D in Iranian elderly. Asian Journal of Psychiatry, 18, 86–90. https://doi.org/10.1016/j.ajp.2015.08.007 Malhotra, A., Allison, B. J., Castillo-Melendez, M., Jenkin, G., Polglase, G. R., & Miller, S. L. (2019). Neonatal Morbidities of Fetal Growth Restriction: Pathophysiology and Impact. 98 Frontiers in Endocrinology, 10. https://www.frontiersin.org/articles/10.3389/fendo.2019.00055 Manderbacka, K., Merilainen, J., Hemminki, E., Rahkonen, O., & Teperi, J. (1992). Marital Status as a Predictor of Perinatal Outcome in Finland. Journal of Marriage and the Family, 54(3), 508. https://doi.org/10.2307/353237 Marcus, S. M., & Heringhausen, J. E. (2009). Depression in childbearing women: When depression complicates pregnancy. Primary Care, 36(1), 151–165, ix. https://doi.org/10.1016/j.pop.2008.10.011 Marmot, M., Ryff, C. D., Bumpass, L. L., Shipley, M., & Marks, N. F. (1997). Social inequalities in health: Next questions and converging evidence. Social Science & Medicine, 44(6), 901–910. https://doi.org/10.1016/S0277-9536(96)00194-3 Martin, L. T., McNamara, M. J., Milot, A. S., Halle, T., & Hair, E. C. (2007). The effects of father involvement during pregnancy on receipt of prenatal care and maternal smoking. Maternal and Child Health Journal, 11(6), 595–602. https://doi.org/10.1007/s10995-007- 0209-0 Mathews, T. J., MacDorman, M. F., & Menacker, F. (2002). Infant Mortality Statistics from the 1999 Period: Linked Birth/Infant Death Data Set: (558952006-001) [Data set]. American Psychological Association. https://doi.org/10.1037/e558952006-001 Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12, 23–44. https://doi.org/10.1037/1082-989X.12.1.23 McAdoo, H. (2007). Black Families (4th ed.). https://doi.org/10.4135/9781452226026 McAdoo, J. L. (1993). The Roles of African American Fathers: An Ecological Perspective. Families in Society, 74(1), 28–35. https://doi.org/10.1177/104438949307400103 McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An Ecological Perspective on Health Promotion Programs. Health Education Quarterly, 15(4), 351–377. https://doi.org/10.1177/109019818801500401 Meghani, S. H., & Chittams, J. (2015). Controlling for Socioeconomic Status in Pain Disparities Research: All-Else-Equal Analysis When “All Else” Is Not Equal. Pain Medicine, 16(12), 2222–2225. https://doi.org/10.1111/pme.12829 Meler, E., Sisterna, S., & Borrell, A. (2020). Genetic syndromes associated with isolated fetal growth restriction. Prenatal Diagnosis, 40(4), 432–446. https://doi.org/10.1002/pd.5635 Meng, Y., & Groth, S. W. (2018). Fathers Count: The Impact of Paternal Risk Factors on Birth Outcomes. Maternal and Child Health Journal, 22(3), 401–408. https://doi.org/10.1007/s10995-017-2407-8 99 Messer, L. C., Kaufman, J. S., Dole, N., Herring, A., & Laraia, B. A. (2006). Violent crime exposure classification and adverse birth outcomes: A geographically-defined cohort study. International Journal of Health Geographics, 5(1), 22. https://doi.org/10.1186/1476-072X-5-22 Misra, D. P., Caldwell, C., Young, A. A., & Abelson, S. (2010). Do fathers matter? Paternal contributions to birth outcomes and racial disparities. American Journal of Obstetrics & Gynecology, 202(2), 99–100. https://doi.org/10.1016/j.ajog.2009.11.031 Molyneaux, E., Poston, L., Ashurst-Williams, S., & Howard, L. M. (2014). Obesity and mental disorders during pregnancy and postpartum: A systematic review and meta-analysis. Obstetrics and Gynecology, 123(4), 857–867. https://doi.org/10.1097/AOG.0000000000000170 Mook-Kanamori, D. O., Steegers, E. A. P., Eilers, P. H., Raat, H., Hofman, A., & Jaddoe, V. W. V. (2010). Risk Factors and Outcomes Associated With First-Trimester Fetal Growth Restriction. JAMA, 303(6), 527–534. https://doi.org/10.1001/jama.2010.78 Mutambudzi, M., Meyer, J. D., Reisine, S., & Warren, N. (2017). A review of recent literature on materialist and psychosocial models for racial and ethnic disparities in birth outcomes in the US, 2000–2014. Ethnicity & Health, 22(3), 311–332. https://doi.org/10.1080/13557858.2016.1247150 Muthén, B., & Muthén, L. K. (2000). Integrating Person-Centered and Variable-Centered Analyses: Growth Mixture Modeling With Latent Trajectory Classes. Alcoholism: Clinical and Experimental Research, 24(6), 882–891. https://doi.org/10.1111/j.1530- 0277.2000.tb02070.x Nardozza, L. M. M., Caetano, A. C. R., Zamarian, A. C. P., Mazzola, J. B., Silva, C. P., Marçal, V. M. G., Lobo, T. F., Peixoto, A. B., & Araujo Júnior, E. (2017). Fetal growth restriction: Current knowledge. Archives of Gynecology and Obstetrics, 295(5), 1061– 1077. https://doi.org/10.1007/s00404-017-4341-9 Nasiri, K., Moodie, E. E. M., & Abenhaim, H. A. (2020). To What Extent Is the Association Between Race/Ethnicity and Fetal Growth Restriction Explained by Adequacy of Prenatal Care? A Mediation Analysis of a Retrospectively Selected Cohort. American Journal of Epidemiology, 189(11), 1360–1368. https://doi.org/10.1093/aje/kwaa054 Neerhof, M. G. (1995). Causes Of Intrauterine Growth Restriction. Clinics in Perinatology, 22(2), 375–385. https://doi.org/10.1016/S0095-5108(18)30289-6 Netchine, I., Azzi, S., Le Bouc, Y., & Savage, M. O. (2011). IGF1 molecular anomalies demonstrate its critical role in fetal, postnatal growth and brain development. Best Practice & Research Clinical Endocrinology & Metabolism, 25(1), 181–190. https://doi.org/10.1016/j.beem.2010.08.005 100 Nordentoft, M., Lou, H. C., Hansen, D., Nim, J., Pryds, O., Rubin, P., & Hemmingsen, R. (1996). Intrauterine growth retardation and premature delivery: The influence of maternal smoking and psychosocial factors. American Journal of Public Health, 86(3), 347–354. Nutor, J. J., Slaughter-Acey, J. C., Giurgescu, C., & Misra, D. (2018). Symptoms of Depression and Preterm Birth among Black Women. MCN. The American Journal of Maternal Child Nursing, 43(5), 252–258. https://doi.org/10.1097/NMC.0000000000000464 Nylen, K. J., O’Hara, M. W., & Engeldinger, J. (2013). Perceived social support interacts with prenatal depression to predict birth outcomes. Journal of Behavioral Medicine, 36(4), 427–440. https://doi.org/10.1007/s10865-012-9436-y Nylund-Gibson, K., & Choi, A. Y. (2018). Ten frequently asked questions about latent class analysis. Translational Issues in Psychological Science, 4, 440–461. https://doi.org/10.1037/tps0000176 Oberlander, T. F., Warburton, W., Misri, S., Aghajanian, J., & Hertzman, C. (2006). Neonatal outcomes after prenatal exposure to selective serotonin reuptake inhibitor antidepressants and maternal depression using population-based linked health data. Archives of General Psychiatry, 63(8), 898–906. https://doi.org/10.1001/archpsyc.63.8.898 O’Donnell, K., O’Connor, T. G., & Glover, V. (2009). Prenatal stress and neurodevelopment of the child: Focus on the HPA axis and role of the placenta. Developmental Neuroscience, 31(4), 285–292. https://doi.org/10.1159/000216539 Orr, S. T., Blazer, D. G., James, S. A., & Reiter, J. P. (2007). Depressive symptoms and indicators of maternal health status during pregnancy. Journal of Women’s Health (2002), 16(4), 535–542. https://doi.org/10.1089/jwh.2006.0116 Orr, S. T., James, S. A., & Blackmore Prince, C. (2002). Maternal prenatal depressive symptoms and spontaneous preterm births among African-American women in Baltimore, Maryland. American Journal of Epidemiology, 156(9), 797–802. https://doi.org/10.1093/aje/kwf131 Oxman, T. E., Berkman, L. F., Kasl, S., Freeman, D. H., & Barrett, J. (1992). Social support and depressive symptoms in the elderly. American Journal of Epidemiology, 135(4), 356– 368. https://doi.org/10.1093/oxfordjournals.aje.a116297 Padilla, Y. C., & Reichman, N. E. (2001). Low birthweight: Do unwed fathers help? Children and Youth Services Review, 23(4), 427–452. https://doi.org/10.1016/S0190- 7409(01)00136-0 Phillips, G. S., Wise, L. A., Rich-Edwards, J. W., Stampfer, M. J., & Rosenberg, L. (2010). Prepregnancy Depressive Symptoms and Preterm Birth in the Black Women’s Health Study. Annals of Epidemiology, 20(1), 8–15. https://doi.org/10.1016/j.annepidem.2009.09.009 101 Radloff, L. S. (1977). The CES-D Scale: A Self-Report Depression Scale for Research in the General Population. Applied Psychological Measurement, 1(3), 385–401. https://doi.org/10.1177/014662167700100306 Ramaswamy, V., Desarbo, W. S., Reibstein, D. J., & Robinson, W. T. (1993). An Empirical Pooling Approach for Estimating Marketing Mix Elasticities with PIMS Data. Marketing Science, 12(1), 103–124. Rankin, K. M., David, R. J., & Collins, J. W. (2011). African American women’s exposure to interpersonal racial discrimination in public settings and preterm birth: The effect of coping behaviors. Ethnicity & Disease, 21(3), 370–376. Reynolds, R. M. (2013). Glucocorticoid excess and the developmental origins of disease: Two decades of testing the hypothesis--2012 Curt Richter Award Winner. Psychoneuroendocrinology, 38(1), 1–11. https://doi.org/10.1016/j.psyneuen.2012.08.012 Roux, A. V. D., Merkin, S. S., Arnett, D., Chambless, L., Massing, M., Nieto, F. J., Sorlie, P., Szklo, M., Tyroler, H. A., & Watson, R. L. (2001). Neighborhood of Residence and Incidence of Coronary Heart Disease. New England Journal of Medicine, 345(2), 99– 106. https://doi.org/10.1056/NEJM200107123450205 Sarkar, P., Bergman, K., O’Connor, T. G., & Glover, V. (2008). Maternal antenatal anxiety and amniotic fluid cortisol and testosterone: Possible implications for foetal programming. Journal of Neuroendocrinology, 20(4), 489–496. https://doi.org/10.1111/j.1365- 2826.2008.01659.x Schulz, L. C. (2010). The Dutch Hunger Winter and the developmental origins of health and disease. Proceedings of the National Academy of Sciences, 107(39), 16757–16758. https://doi.org/10.1073/pnas.1012911107 Schwarz, G. (1978). Estimating the Dimension of a Model. The Annals of Statistics, 6(2), 461– 464. Smith, G. D., Hart, C., Blane, D., & Hole, D. (1998). Adverse socioeconomic conditions in childhood and cause specific adult mortality: Prospective observational study. BMJ (Clinical Research Ed.), 316(7145), 1631–1635. https://doi.org/10.1136/bmj.316.7145.1631 Steen, J., Loeys, T., Moerkerke, B., & Vansteelandt, S. (2017). Flexible Mediation Analysis With Multiple Mediators. American Journal of Epidemiology, 186(2), 184–193. https://doi.org/10.1093/aje/kwx051 Talge, N. M., Mudd, L. M., Sikorskii, A., & Basso, O. (2014). United States Birth Weight Reference Corrected For Implausible Gestational Age Estimates. Pediatrics, 133(5), 844– 853. https://doi.org/10.1542/peds.2013-3285 102 Tarca, A. L., Romero, R., Gudicha, D. W., Erez, O., Hernandez-Andrade, E., Yeo, L., Bhatti, G., Pacora, P., Maymon, E., & Hassan, S. S. (2018). A new customized fetal growth standard for African American women: The PRB/NICHD Detroit Study. American Journal of Obstetrics and Gynecology, 218(2 Suppl), S679-S691.e4. https://doi.org/10.1016/j.ajog.2017.12.229 Taylor, S. E., Repetti, R. L., & Seeman, T. (1997). Health psychology: What is an unhealthy environment and how does it get under the skin? Annual Review of Psychology, 48, 411– 447. https://doi.org/10.1146/annurev.psych.48.1.411 Teitler, J. O. (2001a). Father involvement, child health and maternal health behavior. Children and Youth Services Review, 23, 403–425. https://doi.org/10.1016/S0190-7409(01)00137- 2 Teitler, J. O. (2001b). Father involvement, child health and maternal health behavior. Children and Youth Services Review, 23(4), 403–425. https://doi.org/10.1016/S0190- 7409(01)00137-2 Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The relationship between social support and physiological processes: A review with emphasis on underlying mechanisms and implications for health. Psychological Bulletin, 119(3), 488–531. https://doi.org/10.1037/0033-2909.119.3.488 Van den Bergh, B. R. H., Mulder, E. J. H., Mennes, M., & Glover, V. (2005). Antenatal maternal anxiety and stress and the neurobehavioural development of the fetus and child: Links and possible mechanisms. A review. Neuroscience and Biobehavioral Reviews, 29(2), 237–258. https://doi.org/10.1016/j.neubiorev.2004.10.007 Vaughan, S. E., Misra, D. P., Wong, A. C., Rengers, B., Jablonski, S., Price, M., & Giurgescu, C. (2022). Successful Recruitment Strategies for Engaging Pregnant African American Women in Research. Western Journal of Nursing Research, 44(1), 94–100. https://doi.org/10.1177/01939459211005808 Ventura, S. (1995). Births to unmarried mothers: United States, 1980-92. Vital and Health Statistics. Series 21, Data on Natality, Marriage, and Divorce, 21, 1–55. Ventura, S. J., Martin, J. A., Curtin, S. C., Mathews, T. J., & Park, M. M. (2000). Births: Final data for 1998. National Vital Statistics Reports: From the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System, 48(3), 1–100. Vilagut, G., Forero, C. G., Barbaglia, G., & Alonso, J. (2016). Screening for Depression in the General Population with the Center for Epidemiologic Studies Depression (CES-D): A Systematic Review with Meta-Analysis. PLOS ONE, 11(5), e0155431. https://doi.org/10.1371/journal.pone.0155431 103 Wald, N., Cuckle, H., Nanchahal, K., & Turnbull, A. C. (1986). Sex differences in fetal size early in pregnancy. British Medical Journal (Clinical Research Ed.), 292(6513), 137. Weissman, M. M., Sholomskas, D., Pottenger, M., Prusoff, B. A., & Locke, B. Z. (1977). Assessing depressive symptoms in five psychiatric populations: A validation study. American Journal of Epidemiology, 106(3), 203–214. https://doi.org/10.1093/oxfordjournals.aje.a112455 Weller, B. E., Bowen, N. K., & Faubert, S. J. (2020). Latent Class Analysis: A Guide to Best Practice. Journal of Black Psychology, 46(4), 287–311. https://doi.org/10.1177/0095798420930932 Wit, J. M., & Walenkamp, M. J. (2013). Role of insulin-like growth factors in growth, development and feeding. World Review of Nutrition and Dietetics, 106, 60–65. https://doi.org/10.1159/000342546 Witt, W. P., Park, H., Wisk, L. E., Cheng, E. R., Mandell, K., Chatterjee, D., & Zarak, D. (2015). Neighborhood Disadvantage, Preconception Stressful Life Events, and Infant Birth Weight. American Journal of Public Health, 105(5), 1044–1052. https://doi.org/10.2105/AJPH.2015.302566 Zhang, J., & Bowes, J., Watson A. (1995). Birth-weight-for-gestational-age patterns by race, sex, and parity in the United States population. Obstetrics & Gynecology, 86(2), 200–208. https://doi.org/10.1016/0029-7844(95)00142-E Zuberi, A., Duck, W., Gradeck, B., & Hopkinson, R. (2016). Neighborhoods, Race, and Health: Examining the Relationship Between Neighborhood Distress and Birth Outcomes in Pittsburgh. Journal of Urban Affairs, 38(4), 546–563. https://doi.org/10.1111/juaf.12261 104 APPENDIX A: BERKMAN’S CONCEPTUAL MODEL OF SOCIAL NETWORKS AND HEALTH Figure A1.1.1: Berkman’s Conceptual Model of Social Networks and Health. 105 APPENDIX B: SPEARMAN CORRELATION COEFFICIENT OF RELATIONSHIP INDICATORS Table A1.1.1: Spearman Correlation Coefficient of Relationship Indicators Spearman Correlation Coefficients Prob > |r| under H0: Rho=0 Number of Observations Contact with father Change in of the baby FOB involvement Overall relationship relationship Conflict with FOB Support from FOB Contact with father of the baby (FOB) Spearman corr 1 0.58754 0.60994 0.30732 0.09782 -0.5429 P-value <.0001 <.0001 <.0001 0.0612 <.0001 N 367 367 367 367 367 367 FOB involvement Spearman corr 0.58754 1 0.63491 0.5093 0.2535 -0.71974 P-value <.0001 <.0001 <.0001 <.0001 <.0001 N 367 404 404 404 404 404 Overall relationship Spearman corr 0.60994 0.63491 1 0.43045 0.26467 -0.68872 P-value <.0001 <.0001 <.0001 <.0001 <.0001 N 367 404 404 404 404 404 Change in relationship Spearman corr 0.30732 0.5093 0.43045 1 0.16064 -0.42445 P-value <.0001 <.0001 <.0001 0.0012 <.0001 N 367 404 404 404 404 404 Conflict with FOB Spearman corr 0.09782 0.2535 0.26467 0.16064 1 -0.35846 P-value 0.0612 <.0001 <.0001 0.0012 <.0001 N 367 404 404 404 404 404 Support from FOB Spearman corr -0.5429 -0.71974 -0.68872 -0.42445 -0.35846 1 P-value <.0001 <.0001 <.0001 <.0001 <.0001 N 367 404 404 404 404 404 Notes: ‘Don’t know’ responses in contact with FOB were treated as missing (9%) 106