1.1. Background

Despite spending more on maternity care than any other country, the United States has the highest rate of maternal mortality among high-income counties.¹ Further, although maternal mortality—a key indicator of health and well-being of a country—is declining globally, the United States is one of only two nations seeing a rise since 2000 in deaths from complications related to pregnancy and childbirth.² Maternal mortality, as defined by the World Health Organization, refers to the death of an individual while pregnant or within the first 6 weeks after pregnancy ends from any cause related to or aggravated by pregnancy or childbirth.³ Maternal deaths occurring after the first six weeks but prior to the first year postpartum are considered late maternal deaths.⁴ However, maternal death represents the “tip of the iceberg” as an indicator of maternal health and maternal care quality, because far more pregnant and birthing people experience life-threatening complications of pregnancy and childbirth that can undermine their well-being and functional ability.⁵ Maternal morbidity, as defined by the U.S. Centers for Disease Control and Prevention, refers to any short- or long-term health problem resulting from pregnancy and childbirth.⁶

Each year an estimated 700 pregnant and birthing people die in the United States due to pregnancy-related complications, with nearly three quarters of maternal deaths occurring either on the day of delivery or during the postpartum period.^{7, 8} Risk of maternal morbidity and mortality is unevenly distributed in the United States, with Black and Indigenous women three to four times as affected as their white counterparts,^{3, 9} and disparities in mortality worsened during the pandemic.¹⁰ The determinants of maternal morbidity and mortality and associated racial/ethnic and social inequities are complex, multi-factorial, and less well understood. Still, experts agree that many maternal deaths are preventable.^11–14 Further, trends in maternal morbidity and mortality in the United States reflect increases in rates of cesarean birth,^{15, 16} preexisting chronic medical conditions,^{17, 18} and advanced maternal age.^{16, 19} These individual-level factors do affect—but do not completely explain—the rise in maternal morbidity and mortality in the United States since 2000.²⁰

Efforts to explain the adverse maternal health outcomes have fallen short for reasons ranging from scope of the problem to methodology. For instance, maternal and infant health research has focused largely on infant outcomes, such as low birth weight and premature birth.²¹ This narrow scope is compounded by methodological limitations that restrict the breadth of maternal outcomes studied, the window of study before, during, and after pregnancy, and levels of influence of health risk factors captured in measured exposures.

Likely drivers of population-wide increases in maternal morbidity and mortality include determinants operating on multiple levels of influence—individual, interpersonal, community, and societal.^22–25 Beyond the individual level, social determinants of health represent the conditions or circumstances in which people are born, grow, live, work and age –e.g., access to healthcare, socioeconomic status, education, neighborhood and physical environment, employment, and social support networks.^26–28 Figure 1.1 provides one such conceptual model. Social determinants lie “downstream” from structural determinants of health—the structural forces that shape how social determinants are experienced by people in their neighborhoods and communities and the ways that resources and quality are distributed across individuals and communities.²⁸ Together, these social-structural determinants of health work to shape and promote maternal health for people across the different levels of influence,²⁸ such as variation in access to midwife-attended births,²⁹ linguistically and culturally appropriate care,^30–32 and geographic/local access to and use of maternity units.³³ For example, a pregnant person’s likelihood of being screened for medical risk factors such as high blood pressure, preeclampsia, cardiovascular disease, diabetes, and substance misuse is affected by social-structural factors such as systemic racism,^{34, 35} lack of stable housing³⁶ lack of food,^{37, 38} and incarceration.^{39, 40} Even properly identified medical risk factors of postpartum health may not be adequately addressed due to systemic biases (racial, ethnic, and other prejudices) during referral processes^{39, 40} or follow-up appointments (e.g., failed shared decision-making reduces treatment adherence).⁴¹

The concept of social-structural determinants of health is broadly accepted across the public health and healthcare communities, as is the concept of “social needs,” which focuses on the individual or family and includes real-time gaps that affect health, well-being, and safety. Unfortunately, the two terms are often used interchangeably, which can create confusion. Imprecise use of the terminology can overstate the reach of an intervention. Some efforts that claim to address social-structural determinants of health are not actually addressing a community’s underlying social and economic conditions, but rather aiming to mitigate the current social needs of individuals. For example, providing fresh produce to people struggling to afford food mitigates an immediate individual need, but does not address the underlying systemic issues that cause food insecurity. Addressing the social needs of an individual and the social determinants of a community require different study design approaches, unique partnerships, and innovations.

Figure 1.1

Williams framework for study of racism and health, adapted for colorism and maternal health. Williams (2013) as adapted by Dr. Jaime Slaughter-Acey. Abbreviations: SES= Socioeconomic Status

Current research focuses disproportionately on risk factors at the individual levels, particularly those representative of social identity (e.g., race, education-level, gender). This unbalanced approach has obscured the role of social-structural determinants of health as the root cause of health inequities in maternal morbidity and mortality that affect marginalized or vulnerable people, including Black, Indigenous, People of Color (BIPOC) individuals.⁴³ Notably, race itself is not a risk factor; rather, racism puts BIPOC mothers at risk. Race is a social construct⁴⁴ used to categorize people within a hierarchical system of unearned advantage (privilege) and power that unequally provides access to material, cultural, and psychological resources based on presumed value judgements related to racial status. In turn, unequitable access to resources along with high exposure to cumulative stress resulting from discrimination and marginalization creates a “web of missed opportunities” (e.g., differential access to care and treatment, lack of coordinated care, missed or delayed diagnoses, and unrecognized warning signs by patient or provider related to pregnancy-related death and morbidity).⁴⁵ Such missed opportunities threaten maternal health and deepen health inequities. Additionally, BIPOC mothers and their family members experience higher rates of incarceration, illness, and death than their white counterparts. Lost or systematic removal of family members in BIPOC communities severs access to practical knowledge of pregnancy, birthing, breastfeeding, and postpartum health.⁴⁶ Not only do individuals who have lost their own mothers, either through death or disconnection, often experience profound grief during and after pregnancy. And to continue the example of racism, this information void created by maternal loss is further compounded by the lack of social and medical capital BIPOC people disproportionately experience.

Just as problematic, maternal health literature tends to assume that the role of motherhood and the experiences attached to the identity of motherhood are shared similarly by all pregnant or birthing people—regardless of socioeconomic status, geographic location, racial or ethnic background, age, or other group identities. For example, literature often reports differing rates of maternal morbidity and mortality according to social group category without acknowledging experiences of violence, trauma, or privilege.^{25, 47–49} The overuse of social group categories as proxies for experience obscures the unique concerns and priorities of vulnerable or underserved pregnant and birthing people. This practice also stands in the way of addressing the root causes that are central to creating unique and/or marginalizing experiences of motherhood associated with race, gender, class, disability, and maternal morbidity and mortality inequities. We need to better understand how unfair treatment and structural barriers associated with race, sex, gender, class, and disability impact postpartum health for birthing people.⁵⁰ Intersectionality offers a valuable framework for illuminating inequities in maternal morbidity and mortality, because it allows us to examine how people and their health are affected by two or more intersecting social forces that affect social position and access to resources (e.g., racism, classism) and shape experience (e.g., unfair treatment, discrimination).^{29–41, 51} While each potential pregnant or birthing person will confront their own unique patterns of individual risk, research that identifies themes and patterns at the population level can help highlight opportunities to deliver interventions that address the impact of these determinants of health.

Research often describes comorbid conditions such as obesity, chronic hypertension, and mental health disorders as intrinsic or independent biomedical risk factors for poor health outcomes. However, these conditions are often the physiologic consequences of transgenerational stress and protracted exposure to racial discrimination for BIPOC people.⁵² These conditions (and others including cardiovascular disease and diabetes) have short- and long-term impacts on women and birthing people’s health, and significantly contribute to the disparities in maternal health outcomes. Importantly, an interacting framework allows us to examine how people and their health are affected by multiple intersecting social forces. Such a framework also shifts away from examining these comorbidities as inherent and one-directional, opting instead to view the interactions as a multidimensional feedback loop that compounds risk.

1.2. Terminology

We acknowledge that terms labeling racial and ethnic identities are embedded in a history that includes problematic and painful political and cultural experiences of many groups. Further, terms describing racialized people and marginalized groups have shifted over the years, merging and diverging categories in ways that make simple aggregation impossible. Medical terminology has also shifted; for example, the term pregnancy-induced hypertension is no longer used. Therefore, we chose to use the same language with which study authors presented their results. We also recognize that not all people who become pregnant or give birth identify as women and have attempted to use gender-neutral language to reflect the diversity of the birthing and postpartum experience where possible. We likewise acknowledge that there are women who may object to what may be felt as an erasure of a long history of advocating for women’s rights, and that finding the language balance is fraught. When citing specific study outcomes that identify “mothers” or “women,” we have used those terms for consistency with the research. However, when discussing risk factors such as racial discrimination, we use the word “reported” rather than “perceived,” regardless of study author choices. In this way, we seek to highlight the fact that perception in this case is the act of recognizing the presence of discrimination. While academic fields use the term to denote a necessary step in an appraisal process, this usage could be misread as questioning the validity of “reported” experiences of discrimination.

1.3. Purpose and Scope of the Systematic Review

To better understand the factors underlying postpartum maternal morbidity and mortality in the United States, the Office of Disease Prevention requested this systematic review of available evidence to inform the November 29 – December 1, 2022 Pathways to Prevention workshop, “National Institutes of Health Pathways to Prevention Workshop: Identifying Risks and Interventions to Optimize Postpartum Health,” cosponsored by the National Institutes of Health’s Office of Research on Women’s Health, the National Heart, Lung, and Blood Institute, the National Institute of Minority Health and Health Disparities, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development. The Office of Disease Prevention anticipated complex patterns associated with social and structural drivers of health, including maternal health at the intersections of race and other social group memberships. Therefore, we focused mainly on research examining factors to which pregnant and birthing people have been exposed that may underlie poor postpartum health outcomes. Our scope does not include assessing the effectiveness of interventions aimed at improving maternal morbidity and mortality. Our results will inform research on approaches to address risk factors and improve health outcomes over the postpartum period.