During the 10 years she was a nurse working with pregnant women at an Alexandria, Virginia hospital, Sacha Han, B.S.N., R.N., and her colleagues were forever on alert.
“I set timers on my phone,” Han recalled. If a woman’s systolic blood pressure spiked to a dangerous 160 mmHg and stayed there after 15 minutes — a potential sign of life-threatening complications — they rushed her to the labor and delivery unit. There, colleagues rapidly implemented a series of evidenced-based protocols that Han is certain saved the lives of patients and their unborn babies.
“We acted so quickly,” said Han, who is now a special assistant in the immediate office of the director of the National Heart, Lung, and Blood Institute (NHLBI). “We often averted a crisis.”
And that was the point. The time-sensitive protocols Han’s colleagues used — called “care bundles” — were developed to improve outcomes for women who develop dangerously elevated blood pressure, or severe hypertension, during pregnancy. In the United States, about one in 12 women fall into this category. And one in 25 develop preeclampsia, a severe type of hypertension that can affect the liver, kidneys, and brain. A delayed or missed diagnosis can lead to the mother falling into a coma, having a stillborn baby, or even dying herself. It can also mean heart disease later in life for the mother and her baby. Around 50,000 U.S. women develop these and other pregnancy complications each year.
“Implementation is everything,” said Lisa Hollier, M.D., M.P.H., a professor of obstetrics and gynecology at Baylor College of Medicine and chief medical officer of Texas Children’s Health Plan, speaking at an NIH-hosted maternal health symposium. The Alliance for Innovation on Maternal Health (AIM) has made implementing the science easier by developing bundles to guide not only the treatment of severe hypertension, but other conditions, including obstetric hemorrhage and sepsis. Today, 1,500 hospitals in 38 states have used the safety bundles to deliver 2 million babies each year, about half of all newborns in the United States.
Their use is paying off. Hospitals in California and Illinois saw pregnancy complications associated with elevated blood pressure drop, respectively, by 16% and 20% in 2016 and 2017. The improvements have been so notable that in a special issue published by the Journal of Women’s Health, clinicians and researchers, including Han and Hollier, published a review about the bundles, which double as best practices for obstetric safety. The authors also identify tools and research that may help clinicians detect pregnancy complications earlier.
The care bundles use evidence-based research about pregnancy complications that most clinicians already know. What’s unique is the way they present that information: breaking down what works and why into tiny, routine steps every person in a medical unit must put into practice. Han and the researchers found that health outcomes improve as multiple units in hospitals, and multiple hospitals in states, identify pregnancy complications early and respond with timely treatment.
At the hospital Han worked at, Inova Alexandria Hospital, nurses using the hypertension bundles worked with doctors to administer antihypertensive treatment within a preferred timeframe. “That was huge,” Han explained.
Some physicians and nurses checked off “to-do” items from printed bundles they attached to clipboards. Others used reminders from tablets or cell phones. But all found that by having a list handy, the steps they needed to take became quick and automatic: monitor a patient’s blood pressure, analyze protein levels from urine samples, provide intravenous medications and magnesium sulfate within 60 minutes of blood-pressure spikes, and, if necessary, escalate care.
This universal process eliminated subjective decision making and made it less likely that unconscious bias would creep in. “There’s a human being in front of you that’s presenting with this type of blood pressure, and we need to do this about it, regardless of anything else,” Han said. “Everyone followed a prescribed pattern of actions regardless of a patient’s race or socioeconomic status, which put the patient’s health and safety first.” And those variables have historically mattered.
Government data show that compared to white women, Black women are nearly three times as likely to die while giving birth. For every 100,000 women delivering a baby, about four white women and 11 Black women die. And, according to research published in Clinical Obstetrics and Gynecology, when Black women have children at the same hospitals as white women, the rates of Black women experiencing severe complications, including death, fall by one-third.
In Illinois, these kinds of disparities were no longer apparent after 110 medical teams —responsible for delivering 95% of births in the state — adopted the AIM bundles. In fact, they reported no differences in health outcomes related to race, ethnicity, or health insurance, according to Hollier.
“That’s where we need to be in this country,” said Hollier, a former president of the American College of Obstetricians and Gynecologists. “We all need the same treatment — no matter who we are, no matter what we look like, and no matter where we go to get our treatment.”
In the Illinois hospitals, which formed the Illinois Perinatal Quality Collaborative, teams treated nearly 10,000 cases of severe hypertension. As hospitals implemented steps in the bundle, statewide improvements followed: Post-treatment debriefs among medical teams rose from 2% to 44%. Rates for treating most women with hypertension within 60 minutes rose from 14% to 65%. The number of women who had follow-up appointments 10 days after leaving the hospital increased from 53% to 75%. Preeclampsia education appointments soared from 37% to 81%.
Using digital innovations
The success of the bundles has led to increased use of other innovations. For example, researchers at the University of Pennsylvania found that when 103 new mothers in Philadelphia were given remote blood pressure monitors to use at home, 90% shared their medical updates with their team, texting their blood pressure readings twice a day for 10 days. Conversely, of 103 patients who scheduled in-person appointments for blood pressure readings, 33% of Black women and 70% of nonblack women showed up.
At Vanderbilt University Medical Center in Tennessee, researchers used CDC data to create an algorithm to identify patients at risk for severe pregnancy complications. Using the electronic health records of 45,000 women who delivered a baby at the hospital over a 13-year period, the equation predicted who would experience complications, such as an enlarged heart, hypertension, or heart failure. The algorithm also found that women diagnosed with conditions that affect their fluid levels or immune function were more likely to experience complications during delivery.
Finally, researchers in Korea used a machine-learning model to predict late-stage preeclampsia, which develops 34 weeks after pregnancy and affects 5-8% of women worldwide.
“There’s so much you can add in with machine learning and artificial intelligence,” especially when eliminating bias, Han explained. “But, there’s always going to be a portion of maternal fetal medicine that’s an art.” Personal experience and education, she added, is essential for “nebulous gray areas” that defy checklists, categories, and computer algorithms.
As clinicians, researchers, and ethicists continue to hone best practices for maternal health care, they have been turning to trends in basic science and genomics. Since 2008, the NIH has invested more than $50 million into research to better predict, manage, and treat pregnancy-related complications, including preeclampsia.
“It would be really fascinating to see if we could prevent preeclampsia,” Han said. She noted with excitement the possibilities that could come with future discoveries, such as clinical screenings that could identify genes or biomarkers associated with high-risk pregnancies. “How would that change a woman’s plan of care?” Han asked. Might she seek treatment earlier? Have a different discussion about family planning with her provider? All this and more could be on the table, Han said.
Already, researchers at Baylor College of Medicine have identified genes that activated an immune pathway associated with preeclampsia. Others have identified novel biomarkers. Both sets of findings provide clues about preeclampsia pathways, but they require further study.
“This is a very complex question about a complex, multisystem disease,” said Jasmina Varagic, M.D., Ph.D., program director of the vascular biology and hypertension branch within NHLBI. And so, while researchers work to know more, the care bundles are providing the best resources to help physicians detect and treat complications that might arise during pregnancy, according to the review by Han, Hollier, and their colleagues.
As evidence: in 2019, the Maternal and Child Health bureau, part of the Health Resources and Services Administration, provided $3 million in grants to bring the bundles to hospitals in every state — in an effort to reach every mother and baby.
“It’s really about leveling the playing field,” Han explained. “The bundles use the best available science to support every woman during one of the most momentous journeys of her life, guided by data to support the future of her and her child’s health.”
To learn about maternal health research at the NIH, visit http://www.nih.gov/women/maternalhealth.