3

Epidemiology of Clinical Risks in Pregnancy and Childbirth

Risk is defined as “the chance of danger, loss, injury or other adverse consequence.” Generally, risk is thought of as the potential for or probability of harm. When health care providers use the term “high-risk pregnancy,” therefore, they are typically describing a situation in which the pregnant woman, fetus, or both have an increased likelihood or odds of a pregnancy complication, adverse event, or poor outcomes occurring during or after the pregnancy or birth as compared with an uncomplicated or “low-risk” pregnancy. It is important to note that an exact definition of a “high-risk pregnancy” is not available, because the term lacks conceptual precision in maternity care. Pregnancy is never without some risk; however, most studies use the absence of identified risk factors for poor outcomes as the comparator. Also worth noting is that risk in pregnancy and risk in labor are separate concepts. A person can have pregnancy risk factors such as obesity or hypertension but still have an uncomplicated labor and birth, and vice versa. Finally, all risk factors are not equally significant. Age, for example, is an independent risk factor but confers different risks from those of preeclampsia (both discussed in the text below). Nonetheless, both are similarly labeled as representing “high-risk pregnancies.”

The increased likelihood of an adverse event conferred by risk may be attributable to structural or environmental exposures; inherited or congenital conditions; chronic or acquired health problems, such as diabetes or high blood pressure; infections; complications from a previous pregnancy; risk behaviors; or other issues that might unexpectedly arise during the course of pregnancy. Moreover, these risk factors may interact and intersect. For example, maternal age is associated with diabetes and hypertension, as

well as poor outcomes such as stillbirth independently. In short, risk during pregnancy and labor can arise from many sources, both clinical and societal.

In epidemiological terms, risk takes into account not only the probability of harm but also the impact or consequence of the adverse event. In the setting of pregnancy, this becomes relevant because catastrophic losses—those resulting in the death of a pregnant woman or newborn—are infrequent events. However, the severity of these outcomes, the fear of litigation and liability on the part of providers, and the lifelong implications of loss for families have centered the practice of “high-risk” obstetrics on the prevention and mitigation of these and other types of rare but severe events.

Fortunately, the majority of pregnancies that occur in the United States are not high-risk pregnancies. The rates of diabetes, hypertension, obesity, and advanced maternal age among women of reproductive age, however, are on the rise. And while these pregnancies may not ultimately end in adverse events, they warrant additional surveillance and monitoring for disease progression and fetal compromise in women with medical, social, or obstetrical histories that confer increased risk of an adverse pregnancy outcome.

In this context, appropriate risk assessment by qualified providers to match pregnant people with the most appropriate setting and provider for their care during pregnancy and childbirth is critical. This includes consideration of the risks inherent in different settings, including iatrogenic injuries in hospitals, as well as the risk of potentially avoidable interventions. Cesarean rates in particular are known to vary widely and arbitrarily among hospitals, suggesting that hospital choice may be among the biggest independent risk factors for undergoing major surgery. The immediate risks of cesareans include three-fold higher odds of surgical complications, such as infection and hemorrhage, as well as risks in future pregnancies caused by uterine scarring, including uterine rupture and placenta accreta. In addition, as more women desire and choose birth settings other than hospitals, understanding, screening, and monitoring of the medical, obstetrical, and psychosocial risk factors that affect the care needs of women are increasingly important. At the same time, unforeseen emergencies related to either the birth process or an unrecognized condition may require immediate skilled intervention on behalf of the pregnant woman, fetus, or newborn, including cesarean birth or neonatal resuscitation. After birth, for example, the newborn may encounter difficulties adapting to the neonatal environment, suffer consequences of blood flow or oxygen deprivation during the birth process, or experience a problem based on a congenital anomaly. In light of such events, the ability to access higher-level care without delay is critical for the safety of the woman, fetus, and newborn. Moreover, because risk is not static and can change rapidly during pregnancy and the intra-

partum period, risk assessment must be continuous. Ensuring that women are effectively matched to risk-appropriate care contributes to quality and safety throughout the maternity care system.

In this chapter, we consider medical and obstetrical factors that can increase a woman’s risk of an adverse pregnancy outcome, including both maternal and fetal characteristics, and the ways in which those risk factors affect the decision making of pregnant people and their providers. Common clinical risk factors in pregnancy and childbirth and their clinical implications are described in Table 3-1 and discussed in detail in the following sections. Although this chapter focuses on individual risk factors, it is important to note that many of these factors are the result of structural conditions and societal trends. Demographic trends, such as the increasing age at first pregnancy and increased use of fertility treatments, can contribute to higher pregnancy risk profiles. The social determinants of health (discussed in greater detail in Chapters 1 and 4) also contribute to such preexisting health conditions as obesity, type 2 diabetes, and hypertension among women of reproductive age. These and other system-level risk factors are discussed in the following chapter.

MEDICAL RISK FACTORS

Medical risk factors—for example, such chronic conditions as diabetes and hypertension—are an important consideration in risk assessment for maternity care. Women with preexisting chronic conditions (such as hypertension or obesity) or conditions that develop during pregnancy (such as gestational diabetes) require more intensive care relative to women without these conditions. Demographic shifts, such as people having children later in life, and a number of growing public health challenges, such as increased opioid use, have changed the risk profile of childbearing women on a population level, increasing the proportion of people entering pregnancy with chronic conditions, including substance abuse.¹ This section examines some of the medical risk factors that are present during pregnancy.

Hypertensive Diseases

Hypertension during pregnancy can take several forms. Women may enter pregnancy with hypertension (chronic hypertension) or develop it during pregnancy (gestational hypertension). In addition, pregnant women can develop preeclampsia or eclampsia, conditions in which women develop high blood pressure and signs of damage to another organ system, most

___________________

¹ The number of opioid-related births in hospitals has tripled since 2005 (Admon et al., 2019).

TABLE 3-1 Clinical Risk Factors in Pregnancy and Childbirth and Clinical Implications

commonly the liver or kidneys.² In general, chronic and gestational hypertension without severe features can be managed pharmaceutically during pregnancy, while the only known treatment for preeclampsia is giving birth.³

Hypertensive disorders affect 10 percent of all pregnant women in the United States (Leeman et al., 2016) and were the cause of 6.8 percent of maternal deaths between 2011 and 2015 (Centers for Disease Control and Prevention, 2019a). About 7.7 percent of reproductive-age women in the United States have chronic hypertension (Bateman et al., 2012), which affects 2 percent of all hospital births, while gestational hypertension, preeclampsia, and eclampsia affect 9 percent of hospital births and chronic hypertension 2 percent (Centers for Disease Control and Prevention, 2019c). The prevalence of hypertensive disorders of pregnancy, including preeclampsia, has increased substantially in recent decades, from 528.9 per 10,000 deliveries in 1993 to 912.4 per 10,000 deliveries in 2014 (Centers for Disease Control and Prevention, 2019c). Both chronic and gestational hypertension can lead to such complications as preeclampsia and eclampsia, which can be life-threatening.

Preeclampsia occurs in 5 to 8 percent of all pregnant women (National Institutes of Health, 2019). All pregnant women are at risk of preeclampsia, but some women are at higher risk (refer to Table 3-1). Black women, women of lower socioeconomic status, women of advanced maternal age, and women with obesity are at greater risk of preeclampsia (U.S. Preventive Services Task Force, 2017). Although Black and White women experience preeclampsia at similar rates, Black women die of preeclampsia-related causes at three times the rate of non-Hispanic White women, which may be attributable to inequities in access to prenatal care (U.S. Preventive Services Task Force, 2017), as well as to unequal treatment within the health care system and structural racism (discussed in greater detail in the section “Race, Racism, and Risk” in Chapter 4).

Maternal Age

A woman’s age when she enters pregnancy can contribute to her risk profile in birth. Advanced maternal age, defined as pregnancy at

___________________

² In severe cases, preeclampsia can damage the mother’s organs and restrict oxygen and blood flow to the fetus. If eclampsia develops, women may experience seizure or stroke (National Institutes of Health, 2019). Women with preeclampsia may need close monitoring, specialized drugs, or treatments to prevent further complications or support fetal maturity (American College of Obstetricians and Gynecologists, 2018e).

³ In cases of severe, acute-onset hypertension in pregnancy or the postpartum period, immediate treatment to reduce the risk of maternal stroke is needed (American College of Obstetricians and Gynecologists, 2017a). Moreover, since preeclampsia usually resolves after delivery, induction of labor may be medically indicated. (Refer to Table 3-1.)

age 35 and above, is associated with greater risk of maternal mortality, preeclampsia, poor fetal growth, fetal distress, and stillbirth compared with mothers ages 25–29 (Society for Maternal and Fetal Medicine, 2014; Cavazos-Rehg et al., 2015). Likewise, teenage pregnancy is associated with a greater likelihood of endometritis, postpartum hemorrhage (ages 15–19), and mild preeclampsia and an overall likelihood of having any complication during labor and delivery for those ages 11–14 (Cavazos-Rehg et al., 2015). Pregnancy both during the teenage years and later in reproductive life is associated with higher rates of preterm birth compared with pregnancy among women in their 20s (Ferré et al., 2016). The elevated risk of preeclampsia among women with advanced and early maternal age and the higher rate of maternal mortality among women ages 35 and above frequently necessitate more intensive care during pregnancy and childbirth. Nationwide, about 5 percent of births occurred to mothers less than 20 years old in 2017, while almost 18 percent occurred to mothers ages 35 and older.

In the United States, an increasing number of births occur to women ages 35 and older. Women in this age group account for 9.1 percent of all first births in the United States, and rates of first births to these women increased by 23 percent between 2000 and 2014 (Mathews and Hamilton, 2016). Asian, Hispanic, and Native Hawaiian and Other Pacific Islander women have the highest birth rates at ages 35 and older compared with American Indian/Alaska Native, Black, and White women (Martin et al., 2018a).

Rates of first birth in the teenage years (ages 15–19) decreased by 42 percent between 2000 and 2014 (Mathews and Hamilton, 2016). Yet while teenage pregnancy rates have declined for almost all racial groups, the rates among American Indian/Alaska Native, Latinx, and non-Hispanic Black youth are substantially higher than those among their White peers (Centers for Disease Control and Prevention, 2019c).⁴ Teen pregnancy is highest in rural counties, followed by medium and small urban counties, and the rate is lowest for those residing in large urban counties (Hamilton et al., 2016).

Weight Status

Rates of overweight (body mass index [BMI] between 25.0 and 29.9) and obesity (BMI of 30.0 or higher) in the United States have been in-

___________________

⁴ For example, the birth rate among American Indian and Alaska Native youth ages 15–19 was 32.9 per 1,000, compared with 13.2 per 1,000 births among White youth (Centers for Disease Control and Prevention, 2019c).

creasing for several decades (Hales et al., 2017).⁵ Entering pregnancy with overweight or obesity may necessitate more intensive care during pregnancy and birth.⁶ Prepregnancy overweight or obesity increases the likelihood of developing gestational diabetes or a hypertensive disorder of pregnancy compared with women who enter pregnancy at a lower BMI (Institute of Medicine and National Research Council, 2009; Kim et al., 2010). These antepartum complications increase the risk of indicated preterm and cesarean birth, but women with higher prepregnancy BMI are also at greater risk of miscarriage, stillbirth, shoulder dystocia, and spontaneous preterm birth compared with normal-weight women (Declercq, et al., 2016; Catalano and Shankar, 2017; Schummers et al., 2015). The relationship between prematurity and obesity is not well understood, although maternal inflammation is hypothesized to play a role (Catalano and Shankar, 2017).

Obesity affects more than one-third of U.S. women aged 20–39 (Hales et al., 2017). Black and American Indian/Alaska Native women experience obesity and overweight at higher rates (66.7% and 73.6%, respectively) compared with non-Hispanic White and Asian women. In addition, the prevalence of obesity is estimated to be higher among women of lower socioeconomic status and women in rural areas (McLaren, 2007; Lundeen et al., 2018). For example, rates of prepregnancy obesity among non-Hispanic White, college-educated, and married women are half those of non-Hispanic Black, unmarried women with less than a high school degree (14% and 28%, respectively) (Aizer and Currie, 2014). Rates of prepregnancy obesity in the United States are highest among women of Samoan, American Indian/Alaska Native, Black, and Native Hawaiian ancestry (Singh and Dibari, 2019).

In addition to its clinical implications, having overweight and obesity may make women vulnerable to experiencing weight stigma—the societal devaluation of people with overweight or obesity—in daily life and in the health care system (Andreyeva et al., 2008; Phelan et al., 2015; Pont et al., 2017). In health care, weight stigma can manifest in negative provider attitudes or ambivalence toward patients with obesity (Phelan et al., 2015; Puhl

___________________

⁵ Obesity is further broken down into three categories: class 1 obesity (BMI between 30.0 and 34.9); class 2 (BMI between 35.0 and 39.9); and class 3, or extreme obesity (BMI of 40.0 and above). Each class of obesity is associated with a higher risk of type 2 diabetes, hypertension, and cardiovascular disease (National Heart, Lung, and Blood Institute, n.d.).

⁶ Like obesity and overweight, entering pregnancy with underweight can contribute to adverse pregnancy and birth outcomes, although it affects far fewer women. Entering pregnancy underweight—at a BMI of 18.5 or lower—increases the risk of preterm birth and low birthweight compared with normal-weight women (Han et al., 2011). About 4 percent of women enter pregnancy with underweight (Deputy et al., 2018). Underweight women are recommended to gain more weight during pregnancy to support fetal growth (Institute of Medicine and National Research Council, 2009).

and Latner, 2008; Puhl and Brownell, 2001), and studies from Australia and the United Kingdom document weight bias among maternity care providers (Mulherin et al., 2013; Furber and McGowan, 2011). For example, in an Australian study of 627 women, women with higher prepregnancy BMI reported poorer perceived quality of treatment during pregnancy and after birth relative to normal-weight women (Mulherin et al., 2013).

Diabetes

Nationwide, about 1 percent of women enter pregnancy with preexisting diabetes⁷ (Centers for Disease Control and Prevention, 2018), and between 6 and 9 percent of women develop gestational diabetes (glucose intolerance that develops during pregnancy) over the course of their pregnancy (American College of Obstetricians and Gynecologists, 2018c; DeSisto et al., 2014). While type 1 diabetes is hypothesized to be caused by genetic or environmental factors, type 2 diabetes is associated with obesity and overweight, physical inactivity, older age, high blood pressure, family history of diabetes, and history of polycystic ovarian syndrome (National Institute of Diabetes and Digestive and Kidney Diseases, 2016, 2017). Similarly, gestational diabetes is associated with overweight and obesity and previous pregnancies complicated by gestational diabetes.

Rates of prepregnancy diabetes are highest among American Indian/Alaska Native (2.1%) and Native Hawaiian and Pacific Islander (1.8%) women, followed by African American and Hispanic (1.2% and 1.0%, respectively), Asian (0.9%), and White (0.7%) women. Rates of gestational diabetes are higher among older women compared with younger women; women with obesity and overweight compared with normal-weight women; and Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, and Hispanic women compared with non-Hispanic White women.

Diabetes during pregnancy that is not well controlled is associated with a greater risk of several adverse maternal and neonatal outcomes, including the risk to both birthing women and infants of developing type 2 diabetes later in life (Centers for Disease Control and Prevention, 2018; American College of Obstetricians and Gynecologists, 2018c). Specifically, gestational diabetes increases the risks for preeclampsia, cesarean birth, fetal macrosomia (fetal weight of 9 or more pounds, which can make delivery difficult), neonatal hypoglycemia (low blood sugar immediately after birth), and birth trauma (American College of Obstetricians and Gynecologists, 2018c). Women with pregnancies complicated by diabetes may require additional resources for safe care of the women and their neonates. For

___________________

⁷ This includes type 1 and type 2 diabetes.

example, the American College of Obstetricians and Gynecologists (ACOG) recommends increased monitoring for women with gestational diabetes (2018d), and macrosomia may necessitate cesarean birth.

Substance Use

Substance use during pregnancy is associated with several adverse outcomes, such as premature birth, low birthweight, neonatal abstinence disorder,⁸ fetal alcohol syndrome and fetal alcohol spectrum disorder, miscarriage, stillbirth, and placental abruption (Centers for Disease Control and Prevention, 2019d; Popova et al., 2017; Forray and Foster, 2016; National Institute on Drug Abuse, 2018). Nicotine is the most commonly used substance in pregnancy, followed by alcohol and marijuana (Forray, 2016).⁹ Moreover, opioid use and opioid use disorder (OUD) have increased among pregnant women in recent years. It has been estimated that the number of women with OUD at the time of labor and birth quadrupled between 1999 and 2014, with geographic variation: the lowest rates of OUD were found in Washington, DC (0.7 cases per 1,000 hospital births) and the highest in Vermont (48.6 cases per 1,000 hospital births) (Haight et al., 2018). Opioid use is most common among older pregnant women (over the age of 30) and those who are covered by Medicaid. Moreover, non-Hispanic White women have the highest rate of opioid use, followed by Hispanic and non-Hispanic Black women and those who identify as all other races.

In addition to its deleterious effects, substance use is of particular concern in maternity care because of its frequent co-occurrence with other risk factors. Substance use is often comorbid with other psychiatric illnesses (Swendsen et al., 2010; Forray, 2016). In addition, pregnant women with substance use disorders are more likely to be exposed to other risk factors, such as inadequate prenatal care, chronic medical problems, poor nutrition, and intimate partner violence (Forray, 2016).

Depression

Depression is a common but serious mood disorder that affects 10.1 percent of reproductive-age women in the United States (Brody et

___________________

⁸ Infants who are exposed to opioids during their mother’s pregnancy are commonly born with neonatal abstinence syndrome (NAS). In addition to the withdrawal symptoms they experience shortly after birth, children with NAS have disturbances in their gastrointestinal system, autonomic nervous system, and central nervous system (American College of Obstetricians and Gynecologists, 2017b).

⁹ The evidence regarding low-to-moderate use of alcohol during pregnancy is mixed, showing either inconclusive results or no increased risk for adverse pregnancy outcomes (Forray, 2016). Therefore, heavy alcohol use is of greatest concern in risk assessment of pregnant women.

al., 2018). Perinatal depression has been associated with increased risk of several adverse maternal and neonatal outcomes, including preeclampsia, gestational diabetes (Kozhimannil et al., 2009), hypertension (Kurki et al., 2000), preterm birth, and low birthweight (Grote et al., 2010). Depression is estimated to affect 12 percent of women during pregnancy (Bennett et al., 2004). Prenatal depression is more common among Black, Hispanic, and non-Hispanic White reproductive-age women compared with Asian women in the United States (Brody et al., 2018).

Women may enter pregnancy with depression or develop depression over the course of pregnancy and the postpartum period. Risk factors for developing depression during pregnancy include a history of depression and discontinuation of antidepressant medications during pregnancy (Becker et al., 2016). In addition, hormone changes during pregnancy are thought to increase vulnerability to the onset or return of depression (Bennett et al., 2004). Moreover, depressive symptoms during pregnancy are a strong predictor of postpartum depression, which affects 10 to 15 percent of people who give birth (Ko et al., 2017; Becker et al., 2016; Pearlstein et al., 2015; Halbreich and Karkun, 2006).

OBSTETRIC RISK FACTORS

As with the medical risk factors discussed above, women enter pregnancy with obstetric histories and characteristics that can confer risk. Like medical risk factors, these obstetric factors require careful consideration during the risk-assessment process. In this section, we discuss two obstetric risk factors that need to be considered when determining appropriate birth settings for pregnant women: breech presentation and previous cesarean birth.

Breech Presentation

Breech presentation refers to situations in which the fetus presents as bottom- or feet-first rather than head-first. Breech presentation occurs in 3–4 percent of term pregnancies (Royal College of Obstetricians and Gynaecologists, 2017) and is more common among nulliparous women (Fruscalzo et al., 2014). In cases of breech presentation with a single fetus, women may be offered a procedure to reposition the fetus (called external cephalic version, or ECV). If the fetus cannot be repositioned, options for birth include planned vaginal breech birth or planned cesarean birth. Planned vaginal birth with breech presentation carries higher risk of perinatal mortality than planned cesarean birth, as well as the possibility that emergency cesarean birth will be needed (Royal College of Obstetricians and Gynaecologists, 2017). However, cesarean birth carries greater risk of maternal morbidity (discussed in the following section).

Previous Cesarean Birth

Over the past five decades, the rate of cesarean birth among U.S. women has increased from 5 percent to 32 percent. This increase has been attributed to changes in medical technology (e.g., the advent of electronic fetal monitoring), decreases in operative vaginal births and attempted breech births, and the assumption that having a prior cesarean birth would disqualify a woman from having a vaginal birth (known as a vaginal birth after cesarean, or VBAC) in the future (American College of Obstetricians and Gynecologists, 2019e).

Having a prior cesarean birth, whether elective or planned, influences a woman’s risk status in any additional pregnancies. In the case of previous cesarean birth, a woman may be faced with two options in a future pregnancy: to attempt a vaginal birth or to have another cesarean birth. Both carry risks and benefits for the woman and fetus. Benefits of VBAC include avoidance of major abdominal surgery, lower rates of morbidity (such as hemorrhage, thromboembolism, and infection), and a shorter recovery period compared with women who have an elective repeat cesarean delivery (American College of Obstetricians and Gynecologists, 2019a; National Institutes of Health Consensus Development Conference Panel, 2010). In addition, women who have one successful VBAC are more likely to be able to have a vaginal birth in the future.

Both planned labor after cesarean and repeat cesarean delivery are associated with increased risks. Planned labor after a cesarean birth is associated with greater risk of maternal infection, surgical injury, and uterine rupture (American College of Obstetricians and Gynecologists, 2019e). However, most maternal morbidity related to labor occurs when surgical birth becomes necessary, rather than when vaginal birth is successful (American College of Obstetricians and Gynecologists, 2019e). Women with multiple surgical births are at greater risk for complications associated with repeat abdominal surgeries (such as bowel and bladder injuries) and for issues of placental position and growth¹⁰ in subsequent pregnancies (National Institutes of Health Consensus Development Conference Panel, 2010). In addition, laboring after prior cesarean birth carries some risks for the fetus. Rates of perinatal mortality and hypoxic ischemic encephalopathy associated with labor after prior cesarean birth are higher than those for repeat cesarean birth without labor (National Institutes of Health Consensus Development Conference Panel, 2010).

Given the available evidence on risk and benefit, the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and ACOG recommend that VBAC birth be offered to women

___________________

¹⁰ For example, placenta previa, placenta accreta, increta, and percreta.

who meet certain conditions (primarily, one previous cesarean birth with a low-transverse incision, which carries the lowest risk of uterine rupture). However, other risk factors, such as maternal age, weight status, chronic health conditions, and obstetrical history must also be considered (Wu et al., 2019; American College of Obstetricians and Gynecologists, 2019e). In general, ACOG recommends that planned labor after previous cesarean delivery be attempted at facilities capable of performing emergency deliveries (American College of Obstetricians and Gynecologists, 2019e). Moreover, NICHD calls for the use of a shared decision-making process between women and their providers when planned labor and elective repeat cesarean birth are medically equivalent options (National Institutes of Health Consensus Development Conference Panel, 2010).

MEDICAL AND OBSTETRIC RISK FACTORS BY BIRTH SETTING IN THE UNITED STATES

In general, planned home and birth center births are much less likely to be affected by complications than are hospital births because of the risk selection process conducted by providers in those settings. Women with complicated pregnancies, whether due to medical risk factors or previous obstetric outcomes, are more likely to give birth in a hospital. This distribution of risk is reflected in the birth certificate data on risk factor by place of birth (see Table 3-2). Women with planned home and birth center births in 2017 were much less likely to have medical risk factors, such as prepregnancy or gestational diabetes, hypertensive disorders, or obesity, than women who gave birth in hospitals (refer to Table 3-2). Births to adolescents also occurred at a greater rate in hospitals, while a greater proportion of planned home and birth center births than hospital births were to mothers ages 35 and older (23.6 and 18.1%, respectively, compared with 17.5%). However, VBAC occurred more frequently in home and birth center settings (2.0% of hospital births versus 3.4% of out-of-hospital births). This difference was driven by both planned and unplanned home births, of which VBACs made up about 4 percent (refer to Table 3-2).

In light of the various medical, obstetrical, and social risk factors that can affect a woman and fetus during birth, the risk selection process employed by maternity care providers is critical for promoting patient safety. An analysis of birth certificate data by Grünebaum and colleagues (2015a) found that more than 30 percent of midwife-attended planned home births that occurred between 2010 and 2012 were to women that had at least one perinatal risk factor (breech presentation, prior cesarean birth, more than 41 weeks gestation, or twin gestation). Risk assessment and selection is an important process that requires monitoring and evaluation to support patient safety and promote favorable outcomes at the systems level.

TABLE 3-2 Percentage of Births with Selected Risk Factors by Place of Birth, United States, 2017

NOTES: Column percentage computed per 100 women in specified group. Not-stated responses (<4% for all variables) were dropped before percentages were computed.

^aIncludes prepregnancy and gestational diabetes.

^bIncludes prepregnancy hypertension, gestational hypertension, and eclampsia.

SOURCE: Based on birth certificate data.

Despite the fact that hospitals are at present the safest place for women in some high-risk situations to obtain desired care options for vaginal birth (Bovbjerg et al., 2017), many women cannot find hospitals and physicians offering such care, such as VBACs, which may in part explain the higher percentage of women having VBACs in home and birth center settings. Other maternity care services that are often not available in hospital settings include external cephalic version, vaginal breech birth, and planned vaginal twin birth. Further, some women face challenges in finding hospitals that support intermittent auscultation, nonpharmacologic measures for labor comfort and progress, freedom to drink fluids and eat solids, freedom of movement in labor, and freedom of choice of birth positions, as well as the related essential care option of the choice between midwifery- or medical-led care (Bovbjerg et al., 2017).

CHOICE, RISK, AND DECISION MAKING

In the face of a maternal health crisis in the United States, including maternal mortality and severe maternal morbidity, the nation needs to take seriously the reality that birth, a natural process that in a majority of cases occurs without complication, also can result in devastating outcomes for women, their infants, and their families. Importantly, disparities in these outcomes disproportionately affect the most vulnerable populations. Women, however, may conceive of risk differently; may understand risk differently or tolerate risk differently; or may simply have competing values (e.g., control, respect, faith) that they prioritize over and above medical risks.

Given the prevalence of medical and obstetric risks in the U.S. population, risk assessment and risk selection in birth settings are critical to decision making and choice among birth settings. It is clear that some women desire birth setting options other than hospitals, as evidenced by the increased number of women choosing home and birth center births in recent years (MacDorman and Declercq, 2019). Moreover, among participants in the population-based Listening to Mothers in California Survey who had given birth in hospitals in 2016, a majority expressed an interest in midwifery care and doula support, 40 percent in birth center care, and 22 percent in home birth. Box 3-1 further details the literature regarding pregnant people’s preferences for birth settings and birth experiences and the cultural, social, and religious factors that influence these preferences.

It is broadly accepted that women with decisional capacity have the right to make informed decisions about their care, including crucial, highly determinative, and interrelated decisions about choice of care provider and choice of place of birth (American College of Nurse-Midwives, 2016; American College of Obstetricians and Gynecologists, 2016a). Informed choice, however, requires a set of real options, accurate information about

the risks and benefits of those options, appropriate and ongoing medical/obstetrical risk assessment, respect for women’s informed decisions, and recognition that those choices may change over the course of care. Indeed, true choice requires both that obstetricians inform patients of the availability of alternative care settings and midwifery providers, and that midwives inform patients about the limits of their scope of practice and when medical and/or hospital-based care would be more appropriate.

The discussion in this section considers the concept of risk—how it is assessed (by physicians and midwives versus by pregnant women) and how that risk assessment can and should factor into a provider’s recommendation for a given birth setting. We consider how the skills of shared decision making might facilitate provider–patient communication regarding risks, benefits, and alternatives, as well as elicit values and help women negotiate competing priorities to make the choice that best aligns with their risk profile and values. Recognizing that not all women are candidates for birth center and home birth based on medical and obstetric risks and that women in hospital settings may decline some interventions, providers will inevitably find themselves in a position in which a patient declines or refuses medically recommended care. Therefore, we discuss the provider’s professional and ethical obligation to ensure that a refusal is an informed one and consider best options for respecting patient autonomy while supporting patient safety.

In all of these areas, of course, there is the risk of decisions being made because of unacknowledged normative assumptions. Using end-of-life decision making as an analogy, for example, a normative assumption might be that life must be preserved at all costs, which could lead to choosing medical interventions at the end of life, regardless of their impact on quality of life. A competing normative assumption might be that quality of life is the most important thing, which could lead to foregoing lifesaving interventions (e.g., chemotherapy) that have uncomfortable side effects. These types of normative assumptions might be the basis for decisions made by policy makers, payers, administrators, or providers, but they might not align with a particular person’s or family’s values and preferences. In the case of maternity care, for example, a provider might advise a nulliparous woman with a breech fetus to schedule a cesarean in order to minimize risks to the baby, but the woman might prefer to attempt a vaginal birth because of concerns about operative risks and recovery. Thus, it is important to be aware of how normative assumptions may influence decision making, and to be cognizant of when and how different assumptions are in conflict.

When one considers normative assumptions through the lens of population and public health, tensions often emerge between individual rights or preferences and population-based efforts that seek to maximize health and safety. At the bedside, maternity care providers can prioritize patients’ indi-

vidual preferences in light of their individual risk profile. At the population level, policy makers are tasked with developing strategies for minimizing pregnancy-related morbidity and mortality for women and infants. As a result, they may operate under the normative assumption that their role is to drive down perinatal mortality at all cost without recognizing that doing so may cause maternal mortality or morbidity or other neonatal morbidity to rise, or that many potentially avoidable cesareans may lead to life-threatening conditions in future pregnancies. This is particularly the case in settings, such as the United States, where interventions for “safe maternity” are tertiary in nature, relying on obstetric intervention and surgical “rescue” rather than preventive and safety net strategies designed to ensure that all women have an equitable prospect of entering pregnancy in good physical and mental health and with adequate support. This trade-off is not intended to pit women and their babies against one another, as their interests are, in fact, almost always aligned. However, it does raise important considerations for policy makers regarding the normative assumption that “perfect” is possible or that all risk of adverse perinatal outcomes can ever be perfectly known and mitigated. Two normative questions thus arise: What risk of maternal morbidity and mortality is U.S. society willing to accept in efforts to reduce perinatal mortality and morbidity? Conversely, what risk of perinatal morbidity and mortality is U.S. society willing to accept to prevent maternal mortality and reduce morbidity? The answers to these questions are fraught with practical, political, and ideological implications (Cahill, 2001). However, they are central to any discussion of birth setting, and the collective decisions made with regard to these questions communicate, implicitly and explicitly, the nation’s norms, values, and biases.

Risk Assessment, Informed Choice, and Shared Decision Making

In the late 1990s, Charles and colleagues (1997) developed the framework of shared decision making (SDM), defining it as the bidirectional flow of information between patient and provider resulting in deliberation and negotiation between the two parties, after which patient and provider jointly decide on a treatment strategy. SDM is distinguished from informed decision making, which involves a one-way communication (provider to patient) of medical information, with patients being left to deliberate and decide on their own. Informed decision making is considered more of a “menu of options” approach, in contrast to the more deliberative and negotiated partnership conceptualized in SDM (Charles et al., 1999). SDM has since been designated the optimal model for treatment decision making to promote patient-centered care, particularly when the treatment decision is preference-sensitive (Institute of Medicine, 2013).

Preference-sensitive decision making has been defined as “medical care for which the clinical evidence does not clearly support one treatment option such that the appropriate course of treatment depends on the values of the patient or the preferences of the patient…regarding the benefits, harms and scientific evidence for each treatment option” (Centers for Medicare & Medicaid Services, 2016). Preference-sensitive care does not mean simply that patients may have preferences about their care, as this can be assumed for almost every treatment decision, but rather that patients’ preferences, values, and goals determine which of a number of equally medically indicated treatment alternatives is most suitable and effective for each patient.

Reproductive health in general, and obstetrical care in particular, is replete with preference-sensitive decision making because there is often insufficient or poor-quality evidence to inform treatment decision making, given that it is often infeasible and/or unethical to perform randomized trials of interventions manipulating birth experiences. (See Chapter 5 for further discussion of the strengths and limitations of methodologies used in birth settings research.) Furthermore, pregnant women have historically been excluded from research studies and discoveries, leaving practitioners with little information to guide prescribing practices and clinical management (McCormack and Best, 2014). Accordingly, there is frequently a degree of uncertainty surrounding obstetrical management decision making. However, practitioners may not be skilled or well practiced in navigating or disclosing this uncertainty, and may be biased in their assessments of risks and benefits associated with medical therapies.

Thus, for a maternity care provider, determining the optimal approach to counseling first requires determining whether the medical and obstetric risk and benefit assessment for a patient results in a clear recommendation for hospital, birth center, or home birth. If risks are comparable in all settings, the “right choice” of birth setting depends entirely on what is “right” for that woman. Similarly, if risks are not equivalent across settings, the pregnant woman must weigh this trade-off. She must assess, through the lens of her personal values, preferences, and lived experiences, the probability and severity of potential adverse outcomes, and make the choice she deems safest for her and her child. The same criteria apply within hospital settings when women need to make informed choices about interventions.

In the absence of a medical recommendation for in-hospital care and/or a provider of high-risk maternity care, decision making with respect to birth setting and maternity care provider requires explicitly eliciting a woman’s values, preferences, fears, and concerns regarding her hoped-for birth experience (e.g., family involvement, support persons, pain management, mobility). This includes presenting a full array of options together with an unbiased explanation of the maternal and neonatal risks—both absolute and relative—and benefits associated with each option. These

options need to be presented in appropriate language, which considers not only a language other than English if needed, but also health literacy, such as vocabulary, culturally appropriate terminology, and terminology consistent with levels of education and familiarity with the physiology of pregnancy and birth (National Academies of Sciences, Engineering, and Medicine, 2018).

Risk Assessment and Informed Refusal

As pregnancy progresses, assessments need to be ongoing for maternal or fetal risk factors that would place a woman at increased risk of requiring medical therapies and interventions accessible to her or her newborn only in the inpatient setting, and perhaps only at a higher level of hospital care. These risk factors include her medical history (e.g., cardiovascular disease, autoimmune disorders, chronic renal disease), obstetric history (e.g., prior cesarean, shortened cervix), and psychosocial background (e.g., substance use disorder, current or prior trauma, intimate partner violence, homelessness). For example, professionally and ethically speaking, “to provide safe care, midwives need to be able to tell parents that they can no longer participate in their birth because of changes in risk status” (Jankowski and Burcher, 2015). Out-of-hospital providers are encouraged to practice “preventive ethics” by making the parameters of their care explicit at their first visit with the pregnant woman, as well as being transparent about liability coverage and the potential for redress (McCullough and Chervenak, 1994). Similarly, it is incumbent upon in-hospital maternity care providers to be transparent and forthcoming about the harms associated with hospital-based care—specifically, the use of interventions to induce or augment labor, which can introduce their own side effects and risks for maternal morbidity.

Women who “risk out” of or are deemed poor candidates for home or birth center care still have the right to refuse recommended care, and may do so for any number of reasons, including inability to access the type of care they desire, such as VBAC, in a hospital setting. Informed refusal also takes place within hospitals with regard to specific interventions or types of care (Declercq et al., 2007). Maternity care providers have a responsibility to ensure that these are informed refusals, offering resources and information to support informed choice and mitigate bias and misinformation where possible. Nevertheless, “pregnancy is not an exception to the principle that a decisionally capable patient has the right to refuse treatment, even treatment needed to maintain life. Therefore, a decisionally capable pregnant woman’s decision to refuse recommended medical or surgical interventions should be respected” (American College of Obstetricians and Gynecologists, 2016a, p. 1). Indeed, a woman’s informed refusal provides

the challenge and opportunity for in-hospital and birth center and home birth providers to work collaboratively in a fully integrated system to try to work with the woman to facilitate a safe set of alternatives in support of her well-being.

In the face of such refusal to accept a recommendation for a hospital birth, home birth and birth center providers may find themselves in a quandary, wondering whether they do more good or harm by not providing the woman with care that is outside of their scope of practice. Fears of patient coercion and abandonment may lead these providers to accept patients despite or precisely because of the increased risk, particularly when a woman has either refused or been denied in-hospital vaginal birth for a given indication, such as planned VBAC, vaginal twin birth, and vaginal breech birth.

In their review of a case of home birth with anticipated congenital anomalies, Jankowski and Burcher (2015, p. 31) provide the following guidance for out-of-hospital maternity providers: “Careproviders are obligated to define the boundaries of practice for patients, but careproviders cannot be compelled by patients’ assertion of their positive right for care that is beyond the careproviders’ skill set. To do so, in violation of professional standards, out of a fear that patients will fare even worse if their requests are refused, is a misapplication of the principle of beneficence.” The authors remind care providers that speculative fears “do not outweigh a careprovider’s professional obligation to recognize her own limitations and act accordingly” (p. 32), reasoning that “if a patient’s autonomy could override physicians’ and midwives’ responsibility to remain within their respective scopes of practice, then a patient’s request to her obstetrician to provide a home cesarean section has no grounds for denial” (p. 34). They stress that “careproviders cannot be held hostage by parents’ poor choices” (p. 33), concluding that in doing so, a birth center or home birth careprovider “threatens birth options for other women by opening herself, and her profession, up to criticism” and jeopardizing her own “professional status and the perception of her profession in the broader healthcare community” (p. 34). Similar cautions apply in the realm of hospital care, wherein a woman’s refusal of interventions must be respected, yet providing care beyond one’s scope of practice or skillset places patients and the profession at large at risk.

Risk Communication

Communicating risk in a way that is appropriate for a variety of literacy levels and in culturally and linguistically concordant ways is quite difficult. The literature on health literacy has demonstrated multiple barriers to appropriate risk communication (see, e.g., National Academies of Sciences, Engineering, and Medicine, 2015). Written materials on risk are

constrained by requirements for informed consent documents to be written at vocabulary and complexity levels far beyond the average reading level of the U.S. population. In addition to vocabulary, literacy, and numeracy barriers, cultural background and lived experiences can shape how messages are heard (Nielsen-Bohlman et al., 2018). In Spanish, for example, the word “risk” translates directly as “riesgo,” but pregnant women in one study said that “peligro”—literally “danger”—better communicated what was meant by risk (Alcalay et al., 1993).

In addition, the limited time that some types of maternity care providers can spend with patients can impede implementation of the shared decision-making model discussed earlier (Luntz, 2007). The midwifery model of care, which as noted can be applied in all birth settings and implemented by all types of clinicians, including physicians, provides the time necessary for shared decision making. Nurse practitioners and labor and delivery nurses also have more time for communication and shared decision making, although this can depend on their patient load and clinic or hospital policies.

Decision aids have been found to be useful and effective adjuncts to provider counseling to help health care consumers access and understand treatment options and their risks and benefits (Stacey et al., 2017). These tools not only present information in support of informed choice, but also can include clarification of values to facilitate deliberation and negation of competing priorities. Evidence suggests that decision support tools can help increase patient knowledge and activation and facilitate shared decision making, and in some cases have been shown to result in patients opting for less interventional and costly treatment options (Alston et al., 2014). Calculators and assessment tools can even be embedded in these decision aids to help tailor decision making to personal medical or obstetric risk factors. Were decision aids available to assist in the related choices of maternity care provider and birth setting at the onset of or even prior to pregnancy, women might enter care more activated, engaged, and knowledgeable about these choices (O’Connor et al., 1999; O’Connor, 2001; Stacey et al., 2017). Practical options include making such decision aids available on the intranets of health plans and employers and on respected websites that support childbearing women.

Social media also can be used for clear communication before and during pregnancy, as well as the postpartum period (Scheufele, 1999; Scheufele and Tewksbury, 2006). Women with access to the Internet and the literacy level and language background to utilize that access can find multiple ways to learn about choices for prenatal and intrapartum care and the risks around those choices. Many women with fewer resources have limited access to the Internet, including linguistic and educational barriers to full understanding of Internet materials, but access is rapidly increasing (Kontos et al., 2014; Kim and Xie, 2017). One report notes that in 2018, 68 percent

of Americans used Facebook, and nearly three-quarters accessed YouTube (Smith and Anderson, 2018).

It is also important to note that the potential of social media to facilitate communication about risks and choices in pregnancy and childbirth is complicated by the fact that not all media sources are objective and reliable (Southwell et al., 2018). Some are driven by special interests, and some by individuals or groups with perspectives that are not supported by science and best practices. It remains for respected institutions in government (e.g., the Centers for Disease Control and Prevention, state and local health departments) and the private sector to provide sources that are linguistically, educationally, financially, and culturally accessible (Scrimshaw, 2019).

CONCLUSION

In summary, risk is the potential or probability of harm occurring. In the context of maternity care, clinicians conceive of risk as the potential for pregnancy complications, adverse events, or poor outcomes occurring during pregnancy or after delivery. Risk in this context is influenced by a host of medical and obstetrical factors, as well as systems-level determinants (discussed in the next chapter). Some population groups, particularly women from historically marginalized communities, face a disproportionate burden of pregnancy-related risk, indicating greater care needs that are appropriately provided only in certain birth settings. Although the likelihood of catastrophic losses, such as the death of a pregnant woman or newborn, is low, many pregnancies in the United States warrant additional surveillance and monitoring, and, often, access to medical resources. In addition, members of the maternity care team have a responsibility to inform women accurately and transparently about the risks and benefits of their options, and do so in a way that is culturally concordant, easily understandable, and respectful.

Risk assessment is the process of identifying and assessing sources of risk. In maternity care, the risk-assessment process can be used to match women with the settings and resources they need, focusing more resources on those who need them most and avoiding overuse of technology and intervention for those who do not need them (Institute of Medicine, 2013). In short, the risk assessment process can be used to indicate which settings are most appropriate for a pregnant woman’s care during pregnancy and childbirth. Greater understanding of essential resources for each of the various birth settings, predictors of neonatal complications to guide decisions about level of neonatal care, predictors of maternal complications to guide decisions about level of maternal care, and predictors that should prompt maternal transport between birth settings is needed to inform continuous risk assessment and to guide decisions about which level of

care a woman should receive (Institute of Medicine and National Research Council, 2013). Appropriate risk communication is also essential.

Such consideration and assessment to match women appropriately to the setting and care they need and desire, when carried out continuously and effectively, results in risk stratification across birth settings. That is, lower-risk women predominate in home and birth center settings, while higher-risk women are generally treated in hospital settings. However, in reality, women’s options will be limited by the availability of different types of birth settings and maternity care providers within or near their community, including hospital resources and within-hospital options. Availability is particularly challenging in rural areas and in some inner cities. Also, a woman’s choices are further limited by health insurance and Medicaid restrictions; economic circumstances; access to transportation; and cultural and linguistic factors, such as language barriers with providers and her perception of how she will be received and treated. In short, many nonclinical factors, such as where a woman lives, her opportunities for employment and education, her exposure to discrimination and stress, and her access to services, can influence the level of clinical risk she carries into pregnancy and childbirth. These social and environmental factors impact not only her health, but also the health and well-being of her child, both in the immediate postpartum period and for years to come. These system-level influences on access to and choice in birth settings are the focus of the next chapter.

This page intentionally left blank.