April 25, 2018
Pregnant woman with a doctor.
Expert Q&A

Why Is Maternal Mortality So High in the U.S.?

by Jeanine Barone  

Felicia Lester, MD, is the medical director of gynecologic services at the University of California, San Francisco. She has an active obstetrics and family planning practice and works with women who have any of a variety of gynecologic conditions. We spoke with her recently about why the U.S. ranks at the bottom among developed countries in rates of women dying during pregnancy, childbirth, or the postpartum period.

How common is it for a woman to die in pregnancy in the U.S., and how does that compare to other developed countries?

It’s still rare for a woman to die in pregnancy, but it is much higher in the U.S. than it is in other developed countries. And the rate—which we call the maternal mortality ratio, or MMR—is increasing in the U.S., whereas it’s declining in most other countries. It’s currently between 17 and 28 per 100,000 live births (depending on source and calculation used), which is more than double the rate 30 years ago. The MMR includes death of a woman related to or aggravated by pregnancy while pregnant or within 42 days of the pregnancy ending, whether it’s an abortion, miscarriage, or a delivery. Some data also includes late maternal deaths, within one year of delivery if the cause was pregnancy-related. Our rates in the U.S. are still much lower than in many resource-limited settings—for example, in Uganda, where I’ve worked, the ratio is 343 per 100,000. But the U.S. rate is about triple the rate in the U.K. and Canada, and six times higher than the rate in Scandinavian countries. And it results in about 700 to 900 women dying every year in the U.S. from pregnancy-related causes, as well as about 60,000 to 65,000 of what we call “near misses,” which means women almost dying.

Why is the rate so much higher in the U.S. than elsewhere?

I think there are multiple causes. But we need to think of the system as a whole. And what we realize about the U.S. compared to other countries is that we have a real lack of integration of care between maternity care and primary care. A lot of women lack access to primary care to prevent chronic medical conditions that can contribute to maternal deaths. Obesity, diabetes, and hypertension, for example, all are conditions that can make pregnancy more of a danger for women. They can be managed, controlled, and even prevented in the primary care setting. But our healthcare system is so fractured that a lot of women lack access to basic primary health care. This is probably one of the major differences between the U.S. and other developed countries, where there are concerted efforts to address the problem of maternal mortality from the perspective of providing universal access to health care that can help prevent and treat chronic conditions. In the U.S., women who lack health insurance are four times more likely to die of a pregnancy-related complication compared to their insured counterparts. Part of the impetus behind the Affordable Care Act was to improve access to primary care for more people, but the latest trends that we’re seeing in the U.S.—that is, moving farther away from universal accessto health care rather than towards it—are extremely worrisome for worsening outcomes for moms and, frankly, for babies.

We also don’t have national protocols for conditions that are most likely to kill women in the peripartum period (the period preceding and following childbirth), such as postpartum hemorrhage. This is very different from, for example, the U.K., where there is a specific, publicly available protocol used throughout the country for this complication. And the quality of care a woman receives in the U.S. can vary a lot depending on the clinic or hospital where she happens to go. So that probably contributes as well.

Another factor is that we don’t have great database and review systems for looking at statistics and also drilling down on individual cases, referred to as morbidity and mortality reviews (M & M’s). Some hospitals and clinics have them, but many don’t. The reporting and tracking systems we have are not very well supported and are not universally implemented, though there are national efforts to work on this—to make sure there are accurate statistics and that there are always M & M’s, so we are constantly learning and improving from past near-misses or deaths.

Socioeconomic and racial factors also play a role. National statistics show that there are huge differences in the MMR by socioeconomic status and location. Poor women and rural women are more likely to die in pregnancy, for example. But there are also racial differences. Black women are far more likely to die in pregnancy than white women, regardless of socioeconomic status. Looking at and understanding this is really important as well.

Finally, the rising rate of cesarean sections (C-sections) may be a contributing factor. Repeated cesarean sections in particular can cause bleeding complications that can be deadly.

This is the 21st century. How could this be happening?

Aside from the rising cesarean section rate, I think it’s because women are getting pregnant when they have other chronic conditions that may or may not be recognized, and we haven’t been able to effectively provide access to primary care to be able to control those conditions and effectively plan pregnancy for when they are under control. Things like diabetes, obesity, and hypertension are all rising in our country, and therefore, many of the complications these conditions cause in pregnancy and childbirth are also rising. And I think our health care system is arguably more fractured than it used to be—meaning people don’t have one place or provider to go for all of their care and often seek care only for problems, not for preventive purposes. There are efforts underway to make this less so, but access to primary care is a big problem in our country. If you have or are at risk for a condition such as obesity or hypertension, and you have good primary health care, hopefully those conditions would be prevented or improving or at least not getting exacerbated as we get older. And a pregnancy would hopefully be planned, especially in women with those conditions, so that the medical condition could be optimized before the pregnancy occurred. We’re not necessarily optimizing the timing of pregnancies or optimizing access to primary care to have time to get these conditions under control or improve.

We also need to recognize that the disparities in maternal mortality—with black women, regardless of socioeconomic status, dying in much higher rates than white women—means that our health care system is not meeting the needs of all pregnant women equally, which represents systemic racism.

Another factor contributing to our country’s worsening outcomes is a lack of emphasis on maternal health as compared to fetal health. It used to be that the specialists in obstetrics were specialists in maternal health, and they really understood and could provide the critical care needed for women when complications arose in pregnancy or childbirth. But now a lot of our maternal-fetal medicine programs focus on fetal diagnosis, and a lot of the providers who are being trained in this area don’t even work on a labor and delivery unit and don’t necessarily feel comfortable managing complications that may arise.

Some of the increase in risk could be due to assisted reproductive technology, which I mentioned earlier—which means that more older women who may have other medical conditions are getting pregnant. Assisted reproductive technology also leads to increased pregnancies with multiples (twins and triplets), which have a higher risk of complications in pregnancy and childbirth.

What are the most common medical reasons for a woman to die in pregnancy or childbirth?

Globally, the most likely cause is hemorrhage; it accounts for at least a quarter of the maternal deaths worldwide. But in our country—and I think this goes along with the risk factors of older moms with obesity, diabetes, and hypertension that I already mentioned—15 percent are from cardiovascular disease, 11 percent from cardiomyopathy (an enlargement of the heart muscle), nine percent from pulmonary embolism (a blood clot that travels to the lung), seven percent from hypertensive disorders—including preeclampsia, and seven percent from stroke. Eleven percent are from hemorrhage and another 15 percent are from pre-existing non-cardiovascular conditions, such as kidney disease, hepatitis, or cancer.

Any pre-existing chronic condition increases the likelihood that a woman will experience a fatal event in pregnancy or childbirth. For example, women with pre-existing hypertension are at higher risk to have pre-eclampsia or stroke. Women with pre-existing diabetes are more likely to have ischemic heart disease or renal failure; women with lupus are more likely to have kidney failure or pulmonary embolism; women who are obese are more likely to have ischemic heart disease or pulmonary embolism. Obesity is also a risk factor both for poor wound healing, which can lead to infection. The changes to the cardiovascular system that occur during pregnancy and childbirth—increased blood volume, increased cardiac output, decreased lung capacity, and after delivery, a big shift in fluid back to the heart—can cause a collapse when there’s already an unaddressed medical problem. This can occur during the pregnancy, during delivery, or postpartum.

Is maternal mortality particularly high in certain states?

There is broad variation of maternal mortality by state, from a low of about six per 100,000 (California, Massachusetts) to a high of almost 41 per 100,000 (District of Columbia). This is probably due to a combination of things. The quality of care varies so greatly from site to site and I think partly that lower quality of care can be due to lack of protocols and standardized care. You might have providers who are not in the hospital when women are delivering; the nurses call the doctor if something goes wrong, and he or she has to drive in from home. So there may be a delay in responding to abnormal vital signs, blood loss, and other emergencies. In a system where people are in-house with all the providers actually in the hospital, that might be less likely to occur. Also, not all hospitals have adopted or adhere to evidence-based practice protocols that say, for example, “if the blood pressure is greater than 140 over 90, then the provider needs to come into the hospital and help the nurses manage it,” or if it’s above a certain level, then antihypertensive drugs need to be given. There’s also the lack of standardized M & M’s I mentioned earlier, in which, when there’s a death or a near-miss, we delve down into it, do a root-cause analysis, and try to determine why that death occurred—and if there are systemic factors that need to be changed in order to prevent something like that from happening in the future. One thing we’ve learned over the years regarding complications in medicine in general is that there’s often a system-level problem. Protocols are put into place to try to prevent those types of system-level mistakes from occurring and leading to bad outcomes.

There are also policy-level things that contribute to maternal death or help prevent it, and these vary widely by state.For instance, programs that provide access to highly effective contraceptive methods and abortion can help pregnancies be planned and desired, which can decrease rates of maternal death. Some states, like California, have robust programs, while many others, like Texas (which has a very high maternal mortality ratio 31/100,000), do not.

Can you talk more about the racial difference in maternal mortality in the U.S.? Why are black women dying at such a high rate during pregnancy and childbirth compared to their white peers?

The CDC reports the maternal mortality ratio in the U.S. for white women as 12.7 per 100,000; for black women, it’s 43.5 per 100,000. Can you believe that disparity? It’s unconscionable. And it’s not just about socioeconomic status or educational status or rural/urban dwelling status. It’s something else. Highly educated African American women of high socioeconomic status are still more likely to die in childbirth than a white woman, even one of a lower socioeconomic status. We need to think about our society in general and determine why this might be. And some people are starting to study this more, which is wonderful. We think that at least part of it is due to systemic racism in the medical system. On an individual level, we can think of unconscious bias that providers and staff may have against black women; on a broader level, this translates to systemic racism in which we consistently devalue, disrespect, and distrust black women. Women are more reluctant to seek care in a system that treats them this way, and when they do, they are less likely to get the same quality of care. We think that this, along with increased stress during pregnancy, can exacerbate conditions that lead to maternal death and cause delays in getting high-quality medical care to identify and address the problems.

In California, the Preterm Birth Initiative at UCSF is looking at racial disparities in terms of rates of preterm birth, but some of the work they are doing also applies to maternal mortality. And it’s not only in maternal health care that this disparity comes out. It comes out in all sorts of conditions: Black women are more likely to die of cervical cancer and cardiovascular disease, for example. But it’s extremely stark in maternal health.

Why have maternal mortality rates declined in California?

California has tried to be more proactive for a while now. We have a statewide initiative, the California Maternal Quality Care Collaborative (CMQCC), to look at maternal mortality and try to decrease it. They also have protocols for many of the conditions that kill women in childbirth and tools for hospitals to use to reduce maternal death at their sites. There’s been a big emphasis on hemorrhage and hypertensive disorders, and having protocols and toolkits for those conditions; they are developing others as well. We use these at my institution and they are being used at many smaller hospitals throughout California as well. One of the reasons we have high rates of hemorrhage is because of repeat C-sections. After many C-sections, the placenta is more likely to implant in the uterus in a dangerous way, called placenta accreta. The placenta implants into the scar tissue and, when the placenta comes out after the baby is born, the uterus will just continue to bleed. There is an emphasis on preventing the first cesarean section so as to prevent this condition from the outset. The CMQCC has been very successful at decreasing maternal mortality rates in California. It declined by 55 percent between 2006 and 2013, from 16.9 per 100,000 to 7.3, and is now 5.9. The emphasis on effective contraception and access to abortion has also been important in California.

What are ways to lower the rates of maternal mortality in the U.S.?

On the national level, making sure that women have access to primary health care as well as pregnancy-related care. Planned pregnancies decrease the risks to moms and babies of closely spaced pregnancies (which include increased risk of preterm birth, infant mortality, and maternal mortality), as well as of pregnancies that occur when other health conditions are not effectively controlled, like diabetes and hypertension. All of these things can help us improve pregnancy-related outcomes. Having initiatives like the CCMQC on the national level would also help. There is also the current emphasis on fetal and infant care as opposed to maternal health care that is important to look at. For example, the Title X Fund (part of the U.S. Public Health Service Act, a federal program that provides funding for family planning services) spends 6 percent of its funds on the mom vs. 78 percent on infants and children. Furthermore, Medicaid covers the mom for 60 days after delivery, but the baby is covered for a full year. If the mother had a chronic medical condition, she wouldn’t be able to get ongoing care for it. Then suppose three years later, she gets pregnant again. Not only is she older, but if she hasobesity, diabetes, or hypertension—and there are a lot of women whohave all three—and has gotten no health care between pregnancies, her medical conditions have likely worsened. This is an issue that should be addressed.

The U.S. has a joint commission that accredits health care facilities, and they have a set of five core values that they look at regarding maternity care in hospitals.Four of those are about the baby, and only one is about the mom (the C-section rate). So I think having more of an emphasis on maternal outcomes and not only the infant could help as well. We could also do better in terms of training for high-risk obstetrics care. There’s a financial incentive for recognizing fetal conditions. There is a high rate of insurance reimbursement for all the ultrasounds during pregnancy and hospitals get reimbursed well for babies in the neonatal intensive care unit. These things are not bad, but we need to think about why we are emphasizing this rather than maternal health at the same time as we’re seeing higher rates of maternal death.

Finally, we need to recognize and take responsibility for the racial disparities in maternal mortality and address them head on. This means teaching providers and staff about the dire consequences of structural racism in healthcare and changing how we provide care to make sure that all patients feel comfortable seeking care, their concerns are heard and addressed, and high-quality care is provided to everyone, regardless of race. I think we will be seeing more research into how to best addressstructural racism to narrow these disparities and reverse the trends that we see now.

This opinion does not necessarily reflect the views of the UC Berkeley School of Public Health or of the Editorial Board at BerkeleyWellness.com.

Also see The Social Dynamics of Health.