4.1. Key Points

• Included study exposures broadly covered social-structural determinants of health; however, the identified determinants included still represent only a subset of potential social-structural determinants of interest and did not address intersections with biologic/medical risk factors.
• Limited depth and quality of available research within each risk factor domain, including racism and other forms of discrimination, impeded our ability to understand pathways connecting social determinants of health, medical comorbidities, and maternal health outcomes.
• We found an unexpectedly large volume of research on violence and trauma relative to other potential social determinants of health.
• Among outcome domains, depression/other mental health outcomes were a large proportion of the health outcomes captured.
• We found one study investigating patterns of intersecting social-structural determinants of health that is an exemplar of new approaches to risk factor research.
• Only one study reported the excess risk attributable to a specific exposure; this study reported that income inequality was associated with a 14 percent increase in excess risk of death for Black women relative to white women in Virginia; prolonged 5-year income inequality was associated with a 20 percent increase.

This chapter addresses Key Question 1 and includes studies that examined social-structural determinants of health for pregnant people during the prenatal period. We categorized studies according to 10 major exposure domains. We assigned categories based on subjective reading of the studies because this literature, especially studies examining interactions between factors, is interconnected. Therefore, we sought to present findings with a clear narrative flow. Because so many studies used correlational designs, and due to high risk of bias for those using quasi-experimental techniques, we approached this chapter from the perspective of supporting future researchers in generating hypotheses for risk factors to test potential interventions. Therefore, we report the direction of the adjusted association between risk factors and outcomes of interest. We report numbers if a study attempted to explain the results by how much a specific risk factor contributes to differences or disparities.

Overall, we identified 65 unique studies that addressed Key Question 1, which focuses on (1) the combination of risk indicators that predict poor postpartum health outcomes for pregnant people upon their potential entry into prenatal care, and (2) the extent to which the combination of predictors vary by race/ethnicity. Figure 4.1 uses colored paths to show connections between social-structural determinants of health and outcome domains. Ten exposure domains mapped to nine outcome domains. The most commonly examined outcome domain was depression, with other major domains of interest being maternal mortality, severe maternal morbidity, and hypertensive disorders. Weathering, physiological changes and premature aging caused by extended exposure to stressful experiences, was a smaller outcome domain.⁵⁷ The figure displays the complexity of the connections. In the following sections, we present brief summaries and detailed graphs of the ten identified major factors grouped into nine subsections. Detailed evidence tables can be found in Appendix C.

Figure 4.1

Reported exposures and outcomes for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the colors from left to right from the SDOH (more...)

4.2. Identity and Discrimination

We identified and categorized 27 unique studies that examined factors related to identity and discrimination. One unique study deserves special attention for investigating patterns of intersecting social-structural determinants of health.⁵⁸ Using a sample of 5,763 people with singleton pregnancies who participated in the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b), the study conducted a latent class analysis to identify six subgroups of people, or phenotypes, based on their interrelated social determinants of health. These six subgroups were then used to predict maternal health. Two subgroups in particular predicted worse scores of a composite postpartum maternal morbidity measure: young people living close to the federal poverty level with lower levels of educational attainment (subgroup 6) and people with limited English language proficiency who have lived in the United State for the shortest time (subgroup 2). The study sample was restricted to people without significant comorbidities prior to pregnancy.

The remaining 26 unique studies examined factors related to identity and discrimination, including nativity (place of birth), and acculturation.^59–79 Studies sourced data from clinic records or community-based research projects and programs,^{60–64, 71, 72, 75, 79–82} national data sets,^{59, 65, 77} multi-state records,^{83, 84} state records,^{66, 67, 69, 73, 76, 78, 85} city records,⁶⁸ and secondary randomized controlled trial data.⁷⁴

Thirteen studies examined acculturation or nativity within the context of racial/ethnic disparities in maternal morbidity and mortality. Four studies focused on acculturation and postpartum depression risk for women of Arab descent,⁶⁰ Mexican descent,⁸⁶ Hispanic subgroups,⁶¹ and Latinas.⁷⁰ Additional studies that focused on acculturation examined the association between acculturation and psychosocial stress among Latinas,⁶³ pregnancy-induced hypertension in low- and high-acculturated Hispanic women relative to pregnancy-induced hypertension in non-Hispanic white women,⁷⁹ preeclampsia by nativity (U.S.-born versus foreign born women) and duration of United States residence among non-Hispanic Black women,⁶², gestational diabetes by nativity and duration of United States residence among Hispanic and non-Hispanic Black and White women,⁸¹ trends in and causes of pregnancy-related mortality by race, ethnicity, and nativity;⁶⁵and acculturation and preeclampsia or eclampsia across racial or ethnic groups.⁷¹ One study examined the role of race, ethnicity, and immigration status on prenatal perceived stress and depressive symptomatology trends across pregnancy in a low-income, minority population of pregnant women in an urban city.⁸² Two studies sought to assess maternal health disparities in maternal morbidity during labor and delivery among Mexican-born and U.S.-born white, non-Latina women⁶⁶ and differentials in maternal morbidity and mortality between foreign and U.S.-born women.^{66, 67}

Seven studies examined how racial discrimination and bias impacts maternal health.^{64, 74, 76, 78, 80, 83, 84} One study examined associations of Epstein-Barr virus reactivation with racial discrimination in African-American and non-Hispanic white women during pregnancy,⁶⁴ while others aimed to investigate associations between reported racial discrimination⁷⁶ or racial bias exposure⁷⁸ and postpartum depression.^{83, 84} Two studies examined patterns of change over time to assess the effects of discrimination on pregnancy and postpartum mental health outcomes.^{74, 80}

The remainder of included studies investigated and compared across the following focal areas: Black-white disparity in mortality in women with severe maternal morbidity;⁸⁷ severe maternal morbidity risk factors and effect of race/ethnicity on severe maternal morbidity risk;⁶⁸ disparities in the prevalence of excess heart age (a measure for excess cardiovascular disease risk) among women with a recent live birth;⁶⁹ associations between race/ethnicity and antenatal and postpartum depressive symptoms;⁷² and risk of prenatal depression across racial/ethnic groups.⁷⁷ An additional five studies examined race/ethnicity factors as an exposure of interest.^88–91 Although studies used “race/ethnicity” as an exposure, neither race nor ethnicity are the exposure itself but, rather, each is a marker of multiple potential exposures. However, rarely did studies thoroughly explore specific aspects of the interaction between race/ethnicity and maternal health outcomes in the United States to identify the true association. Figure 4.2 displays the studies’ connections between race/ethnicity factors and eight outcome domains.

Figure 4.2

Reported outcomes for studies with identity and discrimination factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the (more...)

Among acculturation studies, one study found no effect of acculturation on postpartum depression in Arab women.⁶⁰ One study of Hispanic mothers identified single marital status as a risk factor for postpartum depression, and specifically those of Puerto Rican descent who also had a cesarean birth had higher rates of postpartum depression.⁶¹ One study found that discrimination predicted depressive symptoms one month postpartum in Latina women.⁷⁰ Another study noted that among Latina women, higher bicultural acculturation was associated with lower overall pregnancy stress compared with lower acculturation.⁶³ One study of women of Mexican descent found that sociocultural stressors decrease in the postpartum period for those who identify with Mexican culture and that increases in Mexican orientation are associated with less postpartum depressive symptoms.⁸⁶ One study found higher prevalence of preeclampsia and other cardiovascular disease among U.S.-born non-Hispanic Black women versus foreign-born non-Hispanic Black women, though the differences appeared to converge over longer periods of U.S. residence.⁶² One study found an association between more acculturation and hypertension disorders, but no association for self-reported race or ethnicity,⁷¹ while another study found higher substance abuse and interpersonal violence as well as pregnancy-induced hypertension in non-Hispanic white women compared with Hispanic women.⁷⁹ One study found foreign-born non-Hispanic Black women with a duration of U.S. residence of less than 10 years had higher odds of having gestational diabetes compared with their U.S.-born counterparts, whereas foreign-born Hispanic women with a duration of U.S. residence of less than 10 years had lower odds of having gestational diabetes.⁸¹ One study found that Mexican-born women compared with white non-Latina women were less likely to have one or more maternal morbidities, but more likely to have complications related to the quality of intrapartum care they received.⁶⁶ One study found that Mexican-born women compared with Mexican-American women had lower odds of overall maternal morbidity, but higher odds of postpartum hemorrhage, lacerations, and major puerperal infections, indicating suboptimal intrapartum obstetric care.⁶⁷ One study found Black women had higher perceived stress and higher odds of probable depression compared with U.S.-born Hispanic women.⁸²

Looking at racial discrimination and bias, one study found African American women compared with white women showed higher levels of Epstein-Barr antibody titers (a measure of stress-induced immune dysregulation), especially among those that reported higher levels of racial discrimination.⁶⁴ One study of Black and Latina women found that changes in reported discrimination predicted symptoms of depression and anxiety at future time points, with discrimination more strongly predicting anxiety when food insecurity was also present.⁷⁴ One study of multiple racial groups found that participants who experienced racial discrimination had higher odds of experiencing a depressed mood,⁷⁶ another study found that experiences of emotional upset due to racism are associated with an increased prevalence of postpartum depression symptoms,⁸⁴ while a third found non-Hispanic Black women had higher odds of experiencing racial bias but found no relationship between racial bias exposure and postpartum depression after adjusting for confounders.⁷⁸ One study found being upset from racial discrimination increased the odds of postpartum depression with the highest odds reported by women of color with some college education followed by women of color with less than a high school education.⁸⁵ One study found a variety of associations between cultural identity and discrimination and maternal health (both positive and negative aspects) in people of American and Latino cultural orientations.⁸⁰

Several studies reported nuanced findings for mortality outcomes. One study found that income inequality at 1 and prolonged inequality at 5 years was associated with a 14 percent and 20 percent increase, respectively, in excess risk of death for Black women relative to white women.⁵⁹ One study found that all racial/ethnic/nativity groups (with the exception of foreign-born white women) had higher risk of pregnancy-related mortality than U.S.-born white women.⁶⁵ One study found the greatest odds of racial disparity in mortality between Black and white women in groups with the lowest risk of pregnancy-related death, but noted no significant disparity in women at the highest risk of pregnancy-related death.⁸⁷ One study found an association between race/ethnicity and severe maternal morbidity, with greater risk among low-income Latinas and Asian-Pacific Islanders compared with white non-Latinas.⁶⁸ The same study found that living in the poorest neighborhoods increased risk of severe maternal morbidity for Black non-Latinas and Latinas.⁶⁸ Another study found a higher proportion of African American and Hispanic women compared with white women died within seven days of giving birth.⁷³

For other outcomes, one study found non-Hispanic Black women had higher prevalence of excess heart age compared with non-Hispanic white women and Hispanic women; excess heart age prevalence was highest among women who lacked health insurance.⁶⁹ One study found that depression was not associated with race for Black and Hispanic mothers after accounting for financial hardship and other comorbidities.⁷² One study found that African American, Hispanic, and women of other races had lower odds of depression before delivery than white women, with interaction effects observed between race/ethnicity and insurance status.⁷⁷

4.3. Socioeconomic Factors

We identified and categorized nine unique studies that examined socioeconomic factors as the primary social-structural determinants of interest.^{89, 92–99} The studies collected data from a multi-center cesarean registry,⁹⁴ single hospital,⁹⁸ hospital systems,^{89, 92, 99} states,⁹⁶ smaller research programs,^{93, 97} and secondary data from a randomized controlled trial.⁹⁵ Four studies focused on income and its association with cardiac events,⁹⁸ chronic placental inflammation,⁹³ depressive symptomology,⁹⁵ and severe maternal morbidity.⁸⁹ Two studies aimed to assess the impact of education levels on maternal morbidity.^{94, 97} One study sought to identify sociodemographic risk factors association to postpartum emergency department visits.⁹² One study explored how maternal sociodemographic, clinical, and care utilization characteristics are related to hypertension disorders diagnoses in the South.⁹⁶ One study sought to evaluate associations between food security and women diagnosed with gestational diabetes mellitus.⁹⁹ Eleven additional studies examined socioeconomic status as an exposure of interest.^{59, 69, 70, 72, 74, 83, 91, 100–102} Figure 4.3 shows the studies’ connection between socioeconomic factors and nine outcome domains.

Figure 4.3

Reported outcomes for studies with socioeconomic factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the colors from (more...)

Eight of the studies with socioeconomic status as the major exposure found mixed or no association with maternal outcomes.^{89, 92–98} One study of women undergoing cesarean birth found women with elementary or high school education had higher odds of maternal morbidity compared with women with a college degree.⁹⁴ Another larger study of a hospital system found no association with neighborhood crime and severe maternal morbidity.⁸⁹ One study found lack of insurance prior to pregnancy was associated with pregnancy-related hypertensive disorders,⁹⁶ while a smaller study of 200 women found no association between public insurance and emergency department visits after delivery.⁹² One study of 152 women reported lower income was associated with chronic placental inflammation.⁹³ However, another smaller study of 379 high-risk pregnant women found no association between preeclampsia and food insecurity, housing quality, income, or education.⁹⁷ One study of 1,044 Black, urban, high-risk pregnant women found no association between depressive symptoms and socioeconomic status (approached as education, employment status, and public assistance).⁹⁵ While no significant difference in cardiac events rates was found between patients from low versus high income neighborhoods in another study, patients from lower income neighborhoods had higher antepartum hospitalization rates.⁹⁸ In contrast, one study found more patients in food insecure households were diagnosed with gestational diabetes compared with food secure households.⁹⁹

Studies that examined socioeconomic status, but not as their primary focus, reported mixed findings. Women with public insurance experiencing intimate partner violence had higher odds of hypertension and substance abuse,¹⁰⁰ while lack of insurance was associated with excess heart age.⁶⁹ However, education was not associated with depressive or anxiety symptoms^{72, 74} or pregnancy-related death.¹⁰¹ This study also found no association between income and pregnancy-related death.¹⁰¹ One study found no association between poverty and postpartum depression,⁷⁰ but other studies did report such associations.^{72, 102}

4.4. Violence, Trauma, and Psychosocial Stress Factors

We identified and categorized three unique studies that examined psychosocial stress factors.^103–105 The studies collected data from states^{103, 104} and study surveys.¹⁰⁵ Two studies focused on the impacts of stressful life events on maternal morbidity.^{103, 104} One sought to evaluate links between preconception life stressors (partner, traumatic, financial, and emotional) and prevalence of hypertensive disorders of pregnancy.¹⁰³ The other aimed to identify groups of women with similar patterns of antenatal stressful life experiences, and to examine their sociodemographic correlates.¹⁰⁴ The third study in the psychosocial category focused on the experience of weight stigma and subsequent depressive symptoms and reported stress.¹⁰⁵ An additional seven studies examined psychosocial stress factors as an exposure of interest.^{64, 74, 75, 95, 106–108}

We identified and categorized four unique studies examining trauma factors.^{88, 109–111} The studies sourced data from clinic-based studies or programs. These studies varied in focus, examining associations between the following: post-traumatic stress disorder and dissociation at the time of the traumatic experience;¹⁰⁹ antenatal depressive symptoms and anticipated negative police youth encounters;¹¹¹ lifetime trauma exposure and perinatal health outcomes in low-income African American women;⁸⁸ and military sexual trauma and risk of depression and suicidal ideation during and after pregnancy.¹¹⁰ An additional two studies examined trauma as an exposure of interest.^{103, 107}

We identified and categorized 12 unique studies examining violence factors.^{100, 107, 108, 112–119} Studies collected data via national data sets,¹²⁰ state health records,^{113, 114} clinic medical records,^{100, 107, 112, 117, 119} community program survey,^{108, 115, 118} and secondary program data.¹¹⁶ All included violence studies examined intimate violence and domestic violence impact on maternal health outcomes. One study also focused on violence against women and experiences of childhood violence in addition to intimate partner violence.¹¹³ An additional five studies examined violence factors as an exposure of interest.^{70, 78, 79, 95, 106} Figure 4.4 displays the studies’ connections between violence, trauma, and psychosocial stress factors and eight outcome domains.

Figure 4.4

Reported outcomes for studies with violence, trauma, and psychosocial stress factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by (more...)

Among the studies focused on psychosocial stress, the two that examined life stress events found a positive correlation with maternal morbidity. All four types of life stress (partner, traumatic, financial, and emotional) were linked with increased prevalence of pregnancy-related hypertensive disorders, with the strongest association observed for financial stress.¹⁰³ The highest prevalence of severe pregnancy-associated nausea/vomiting and postpartum depression occurred in those experiencing multiple types of stress.¹⁰⁴ The other included study suggested that experiencing weight stigma may contribute to unfavorable physical and mental health outcomes for pregnant and postpartum women.¹⁰⁵

All studies focused on factors related to trauma found that experiences of trauma were associated with negative maternal health outcomes. Race-related anticipatory stress around potential negative youth-police experiences was associated with antenatal depressive symptoms for expectant African American mothers.¹¹¹ Women with maltreatment history and post-traumatic stress disorder were at higher risk of experiencing re-traumatization or becoming overwhelmed and experiencing dissociation while giving birth.¹⁰⁹ Military sexual trauma was associated with higher pre- and postnatal symptoms of depression and suicidal ideation.¹¹⁰ Lifetime trauma exposure was significantly associated with depressive symptoms, anxiety, and generalized stress. This study found 87 percent of the women reported at least one traumatic life event.⁸⁸

Among the studies focused on the domain of violence, all studies on intimate partner violence found the violence to have a negative effect on maternal health outcomes.^{100, 107, 108, 112–120} In a predominately Latina sample, one in five mothers exposed to intimate partner violence showed depressive symptoms when screened during the perinatal period.¹¹² Recent intimate partner violence exposure was found to be a prenatal predictor of postpartum depression^{118, 119} and post-traumatic stress among Latinas.^{117, 118} A history of experiencing violence, including intimate partner violence increased medical and obstetrical complications, including perinatal and postpartum depressive symptoms across populations.^{100, 108, 113, 115, 116} One study also noted that the association between intimate partner violence and postpartum depression persisted regardless of socioeconomic status.¹¹⁵ One study sourced from the National Inpatient Sample found when compared to those with no exposure, individuals of all ethnic groups exposed to violence had increased risk of all adverse maternal/fetal outcomes.¹²⁰ Another study reported domestic violence contributed to death in 14 percent of cases of maternal mortality in their sample, with 65 percent of cases considered preventable.¹¹⁴

4.5. Structural/Institutional Factors

We identified and categorized 17 unique studies that examined structural/institutional factors as the primary social-structural determinants of interest.^{102, 121–133} The studies sourced data from city public health records,¹²¹ state/county records,^{102, 122, 124, 128, 132, 134, 135} multi-state data,^136–138 national data,¹²³ and records from a single hospital.^{91, 125–127, 130, 139} Structural/ institutional factors explored in this literature set include the relationship between urban food deserts and obstetric outcomes,¹²¹ associations between residing in a maternity care desert and risk of death during pregnancy,¹²⁹ associations between family leave after childbirth and maternal outcomes for mental and physical health,¹²³ impacts of pregnancy-related public health programs,¹²² municipal expenditures,¹³⁷ incarceration exposure during pregnancy,^{102, 136} neighborhood indices on maternal health outcomes^{125–128, 130} and severe maternal morbidity,⁹¹ and area-level population distributions of race and income to predict death during pregnancy or up to 1 year postpartum.¹²⁴ Another study used Black-White racial segregation along with pandemic timing to examine racial disparities,¹³⁵ while another aimed to identify the geographic distribution of and disparities in cardiomyopathy outcomes.¹³⁹ One study sought to identify the effects of the Affordable Care Act’s non-Medicaid provisions on insurance coverage and postpartum depressive symptoms.¹³⁸ An additional three studies examined structural/institutional factors as an exposure of interest.^{89, 97, 140} Figure 4.5 displays the studies’ connections between structural/institutional factors and outcomes.

Figure 4.5

Reported outcomes for studies with structural/institutional factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the colors (more...)

Two studies found that racial residential segregation was negatively associated with maternal health, with higher rates of hypertensive disorders,¹²⁷ severe maternal morbidity, ¹³⁵ and increased risk for pregnancy-associated death¹²⁴ among those living in neighborhoods with high proportions of low-income and Black residents. Other neighborhood indices, such as crime¹²⁶ and adverse neighborhood physical environment traits, were associated with an increase in hypertensive disorders,¹²⁵ while neighborhood walkability decreased risk for pregnancy-related hypertension.¹²⁸ One study found after adjustment, there was no association between neighborhood deprivation and severe maternal morbidity. ⁹¹ One study found patients with severe peripartum cardiomyopathy were more likely to live in communities with greater social vulnerability.¹³⁹ In an urban setting, multipurpose and walkable communities were associated with a lower risk of postpartum depression.¹³⁰

Risk of death during pregnancy and up to 1 year postpartum was higher among women who lived in maternity care deserts than women in areas with greater access.¹²⁹ Study authors noted that racial inequity in risk persisted above and beyond differences in geographic access to maternity care.¹²⁹ For individuals residing in urban food deserts, mothers were more likely to be younger, obese, and of minority race/ethnicity; however, they did not experience worse perinatal health outcomes.¹²¹ For mothers who worked prior to childbirth and returned in the first year, having less than 12 weeks of maternal leave and less than 8 weeks of paid maternal leave was associated with increased depressive symptoms and reduced overall health status.¹²³ Exposure to incarceration (personally or through a romantic partner’s incarceration) was associated with postpartum depression, but not more so than other typical stressors such as financial hardship.¹⁰² Another study identified increased odds of prepregancy hypertension and pre- or postpartum depression in women with incarceration exposure.¹³⁶ One study found that increased public-health spending reduced mortality rates among Black mothers and narrowed Black-white outcome disparities.¹²² One study found reduced odds of severe maternal morbidity in areas where annual municipal expenditures (such as fire, ambulance, transportation, health, housing and libraries) were $1000 higher per capita. The same study found higher expenditures on police were associated with increased odds of severe maternal morbidity.¹³⁷ One study found the Affordable Care Act’s non-Medicaid provision expansion was associated with increased retention of postpartum insurance and reductions in postpartum depression after controlling for maternal characteristics.¹³⁸

4.6. Rural/Urban and Environment Factors

We identified three studies that examined rural/urban factors using nationwide data sources as the primary social-structural determinant of interest.^{90, 141, 142} One examined geographic variations in 3,747 pregnancy-related deaths from 2011 to 2016 among Hispanic, non-Hispanic white, American Indian or Alaska Native, and Asian or Pacific Islander women (using National Center for Health Statistics Urban-Rural Classification Scheme for Counties to determine urban-rural groupings).⁹⁰ Another study included 17,229 perinatal women from 14 states and examined differences in risk of perinatal depression between women in rural and urban areas.¹⁴¹ The third study aimed to evaluate how rural/urban status along with other risk factors alter a women’s risk of severe maternal morbidity at delivery using Kentucky state delivery hospitalization records.¹⁴²

We identified and categorized six unique studies examining environmental factors.^143–147 Studies collected data from states,^{143, 147} city programs,^{144, 146} and a single hospital.¹⁴⁵ Two studies focused on air pollution, one on associations between proximity of key land features and risk of negative health outcomes,¹⁴³ and the other on ambient air pollution as a modifiable risk factor for postpartum depression.¹⁴⁶ One study examined the association between exposure to trace minerals and heavy metals and preeclampsia.¹⁴⁴ Two studies examined associations between neighborhood greenness and hypertensive disorders¹⁴⁵ and how greenspace may positively impact pregnancy health for racially and economically minoritized populations.¹⁴⁷ One study examined special clustering of severe maternal morbidity across South Carolina and its associations with place-based social and environmental factors, primarily heat exposure.¹⁴⁰ Figure 4.6 displays the studies’ connections between rural/urban and environment factors and five outcome domains.

Figure 4.6

Reported outcomes for studies with rural/urban and environment factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the (more...)

One rural/urban study reported that in small metro, micropolitan populations (10,000 to 50,000), and rural counties, pregnancy-related mortality ratios were 2 to 3 times higher among non-Hispanic, American Indian or Alaska Natives than among non-Hispanic white women in the same areas.⁹⁰ Black women had a pregnancy-related mortality ratio 3 to 4 times higher than white women in the same area regardless of urban-rural classification.⁹⁰ One study found that the percentage of delivery hospitalizations with severe maternal morbidity was higher for women with rural vs metro vs metro-adjacent residence.¹⁴² Findings from one study suggested the odds of perinatal depression were higher among rural than urban women. Non-Hispanic Black women were more likely to report perinatal depression than non-Hispanic white women in rural areas.¹⁴¹

All six studies found negative associations between environmental factors and maternal health outcomes. One study found that second trimester exposure to ambient particulate matter and living close to a major roadway were associated with higher odds of gestational diabetes mellitus, and that living more than 500 meters from a recreational facility was associated with lower odds of gestational hypertension.¹⁴³ None of these factors were associated with preeclampsia.¹⁴³ One study found an association between higher manganese concentration in maternal red blood cells and lower risk of preeclampsia, and another association between higher cadmium concentration and higher risk of preeclampsia.¹⁴⁴ One study found an association between lower levels of neighborhood tree canopy cover and higher odds of hypertensive disorders in an urban population, and this association was stronger among non-Hispanic Black women.¹⁴⁵ One study found an association between increased odds of postpartum depression for Hispanic/Latina and second trimester exposure to nitrogen dioxide and particulate matter.¹⁴⁶ One study found that those with the lowest access to publicly available and accessible greenspace had an increased risk for mental disorders, depressive disorders, and gestational diabetes.¹⁴⁷ One study identified an impact of hot ambient temperatures on maternal morbidity.¹⁴⁰

4.7. Comorbidity Factors

We identified and categorized two unique studies that examined comorbidity factors as major exposures.^{106, 148} The first study examined medical records of Black women living with HIV,¹⁰⁶ and the other recruited women who were seeking prenatal medical care from four Midwestern clinics.¹⁴⁸ An additional five studies examined comorbidities as an exposure of interest.^{95, 96, 109, 112, 114} Comorbidities in these studies included mental illness,^{109, 114} existing clinical conditions,⁹⁶ stress and anxiety,^{106, 148} and substance use.^{95, 112, 114} Figure 4.7 displays the studies’ connections between comorbidities and five outcome domains.

Figure 4.7

Reported outcomes for studies with comorbidity factors for pregnant people. Note: The figure displays colored paths connecting risk factor and outcome domains. Each outcome domain is represented by a different color and by following the colors from left (more...)

One study reported low income, exposure to intimate partner violence, and childcare burden as having a negative impact on the psychological health of Black peripartum women living with HIV.¹⁰⁶ The other study reported a positive correlation for white women between their beliefs about keeping depression a secret and experiencing depression; however, secrecy and depression were uncorrelated for Black women.¹⁴⁸

Studies that examined comorbidities as an additional factor of interest reported mixed findings. Lifetime post-traumatic stress disorder contributed, independent of other factors, to dissociation in delivery,¹⁰⁹ while substance use was independently associated with depression,¹¹² and illicit drug and alcohol use was associated with depressive symptoms in African-American expecting mothers.⁹⁵ However, substance use and mental illness did not lead to more preventable deaths.¹¹⁴ Obesity and diabetes were associated with hypertensive disorders.⁹⁶

4.8. Hospital and Healthcare Use Factors

We identified and categorized one unique study that examined healthcare use factors as a major exposure.¹⁰¹ The study sourced data from North Carolina state death certificate codes and linkage of birth and death files to determine pregnancy-related death among all cesarean births and all vaginal births for a 7-year period. Receipt of any prenatal care was associated with a decreased risk for pregnancy-related mortality. An additional study examined healthcare use factors as an exposure of interest.⁶⁸