Written by Kim Krisberg

In 2016, newly published data showed Texas’ maternal mortality numbers had abruptly doubled in just a few years. The number was far, far above the national average, shocking health care providers and advocates. It turns out the data were wrong — significantly wrong. The silver lining: The bad data elevated a troubling health gap.

The corrected data, published in May, showed Texas’ 2012 maternal mortality rate was at 14.6 per 100,000 live births, less than half of what researchers had initially estimated in 2016.  That put Texas’ rate more in line with national numbers. But it still didn’t make the study good news. The state’s maternal mortality rate was still inching upward, as is the nation’s, and the U.S. is home to some of the worst maternal mortality rates among high-income nations. In addition, even the corrected Texas data showed significant disparities, with black women experiencing maternal mortality rates nearly twice that of the overall rate.

“For a long time, we’ve known that maternal mortality is high for a developed country like the United States — it’s higher than it should be,” said George Saade, M.D., director of obstetrics and maternal-fetal medicine at The University of Texas Medical Branch in Galveston and a member of the Texas Perinatal Advisory Committee. “Those studies got people’s attention and showed that this is a problem, not only in Texas, but nationally.”

While the data slip-up — much of it the result of incorrect death certificate data — gave stakeholders a scare, and its retraction a momentary sigh of relief, it also galvanized new action on reducing pregnancy-related death and morbidity. Just this June, for example, the Texas Department of State Health Services officially rolled out TexasAIM, a statewide initiative to help hospitals and clinics improve maternal care and reduce preventable maternal mortality and morbidity. As of July, 75 percent of Texas’ birthing hospitals, representing about 80 percent of the state’s births, had committed to the voluntary initiative and to adopting sets of evidence-based approaches, or “bundles,” for improving maternal care. TexasAIM is a partnership with the Alliance for Innovation on Maternal Health, a quality improvement initiative operating in more than a dozen states.

The Texas Hospital Association is a TexasAIM partner and will be providing coaching, support, education and training to participating hospitals and collecting data on implementation and outcomes improvement.

TexasAIM’s first goal is implementing an obstetric hemorrhage bundle — one of 10 recommended AIM bundles — which includes a set of protocols designed to detect and intervene on hemorrhage, a leading cause of maternal deaths, as early as possible. TexasAIM’s quality improvement goals align closely with the state’s new Maternal Levels of Care Designation rule, which went into effect in March and makes designation a requirement for Medicaid reimbursement beginning in fall 2020. Texas is the first state to legislate such designations.

“It really was a perfect storm of data, legislation and research that all came together to push this work forward,” said Jeremy Triplett, Title V maternal and child health director at TDSHS.

The June launch of TexasAIM certainly marks a new chapter in statewide efforts to reduce maternal deaths, but inside Texas hospitals, work to improve outcomes for women and new mothers is well underway.

Hospitals Adopt Early Warning Systems, Bundles of Care

In Houston, Ben Taub Hospital first implemented the AIM hemorrhage bundle in 2016, followed by a bundle for severe hypertension in pregnancy in 2017. In July, the hospital started implementing its third AIM bundle, this one designed to reduce racial and ethnic disparities in maternal outcomes. The hospital also rolled out a maternal early warning system back in 2014. Carey Eppes, M.D., master of public health, chief of obstetrics at Ben Taub, said the key to such improvement efforts isn’t just the intervention itself, but the behind-the-scenes work to put it into action.

“We did a lot of simulations, we continuously did audits on what was happening and provided feedback,” said Eppes, also an assistant professor in maternal-fetal medicine at Baylor College of Medicine. “It’s not just taking what’s written down and trying to make it happen — it’s how you do it and how you adapt to the problems and challenges that arise.”

For example, the hospital’s maternal early warning system is designed to amplify vital signs that often predict adverse outcomes and then trigger an immediate response. On its face, the system doesn’t seem too different from what a hospital already does — monitor a patient and notify colleagues when troubling symptoms occur. But there is often a lag between collecting a vital sign and reporting it. The early warning system, on the other hand, standardizes which maternal symptoms should trigger immediate attention. In fact, to ensure as rapid a response as possible, Eppes said staff are encouraged to trigger the system with a face-to-face notification, rather than electronically. There’s even a script staff can use for such interactions.

The work has paid off: Eppes reported that the time between detecting an abnormal vital sign and normalizing it went from more than 400 minutes to less than 60.

“It’s been the most transformative and effective quality intervention we’ve ever done in our hospital in labor and delivery,” she said.

One of the first hospitals to enroll in TexasAIM, Ben Taub also has a head start implementing the bundles, starting in 2016 with hemorrhages. One of the biggest hurdles of that bundle, Eppes said, was switching from a reliance on estimated blood loss during delivery to quantitative blood loss. After the switch, the hospital actually recorded an increase in postpartum hemorrhage rates — “and that’s really hard to share when people are working so hard,” Eppes said — but it meant the bundle was working. The numbers were up because staff were getting better at recognizing the problem. In addition, better detection meant fewer maternal patients had to be transferred to intensive care and fewer patients required a massive blood transfusion.

“It’s a team effort,” Eppes said. “It takes a whole workforce to do this work.”

To the west at Children’s Hospital of San Antonio, also a TexasAIM participant, the goal is to eventually implement all 10 AIM bundles, said Peter Nielsen, M.D., the hospital's obstetrician/gynecologist-in-chief. Like Ben Taub, the San Antonio hospital was an early adopter of maternal care bundles, starting in late 2016 with hemorrhage. Nielsen noted that AIM bundles have roots in the California Maternal Quality Care Collaborative, which began in 2006 and led to a 55 percent decline in California’s maternal mortality between 2006 and 2013.

Each AIM bundle is a collection of the best evidence-based practices to improve maternal care. Of course, hospitals typically already employ many of the individual practices. But the power of bundles, Nielsen said, is two-fold: they formally link practices that combined have a greater impact than any one practice on its own, and they encourage teamwork and synchronization. The key to leveraging that intervention, Nielsen said, is constant team drilling.

“Imagine if you had a football team with all superstar athletes but they never practiced together, they just came out for high-stakes games,” said Nielsen, also vice chair for the Department of Obstetrics and Gynecology at Baylor College of Medicine. “Doing the drills exposes both an institution and an individual to some degree of risk…But we’re not here to embarrass anyone, we’re here to find opportunities for improvement.”

Today, maternity staff at Children’s Hospital of San Antonio, part of CHRISTUS Health system, participates in hemorrhage drills on a regular basis — Nielsen said the goal is to conduct at least two drills per month per shift. Clinicians base their work on the AIM bundles as well as an Agency for Healthcare Research and Quality team-based patient safety tool known as TeamSTEPPS, which Nielsen helped develop.

“The goal isn’t just to help our hospital and patients, but within the CHRISTUS system, we’ll be able to provide improved access to a certain level of team-based care,” Nielsen said. “A women should be able to show up to any hospital within CHRISTUS and receive the same level of hemorrhage care.”

Intervening in the Hospital and Out in the Community

At many Texas hospitals, work to improve maternal outcomes predates TexasAIM. However, the state-led initiative is harnessing a new wave of momentum to reduce mortality rates and close disparities. The initiative is the result of 2017 legislation that extended the work of the state’s Task Force on Maternal Mortality and Morbidity, first established in 2013, and directed state health officials to undertake maternal health and safety initiatives. 

So far, according to TDSHS' Triplett, 132 of the 178 hospitals have joined the initiative as TexasAIM Plus participants, which means participating in a comprehensive learning collaborative, submitting quarterly metrics and reviewing evaluative data every month. TexasAIM’s first focus is implementing the hemorrhage bundle, but it’s also piloting a bundle for obstetric care for women with opioid use disorders in nine hospitals, with a goal of implementing that bundle statewide in 2019. After hemorrhage and opioid addiction, focus will turn to the hypertension bundle.

“The goal is standardizing policies and procedures for care and having a uniform set of protocols so women receive the same quality of care throughout the state,” Triplett said.

Carla Ortique, M.D., an obstetrician/gynecologist at Texas Children’s Hospital Pavilion for Women in Houston and vice chair of the Texas Maternal Mortality and Morbidity Task Force, noted that other states that implemented AIM bundles reported a 20 percent decrease in severe morbidity. The key to the bundles’ effectiveness, she said, are the “four Rs”: ready, recognition, response and review. Her own hospital already has the hemorrhage protocol in place and implemented a maternal early warning system in June. 

“Often, the problem is a delay in recognizing the warning signs,” she said. “Timing is everything.”

Ortique said maternal mortality is also a policy issue, with access to and coverage of care both significant factors. In Texas, she noted, low-income pregnant women lose Medicaid coverage in the months following a birth, even though most maternal deaths happen between 42 days and a year after delivery. The majority of Texas births are covered by Medicaid, according to state health officials. Going forward, Ortique said the state task force hopes to dig deeper into the root causes of poor maternal health outcomes, identifying differences by social determinants and ZIP code. “We can prevent many of these deaths,” she said.

But there is a limit to what hospitals can do within their walls — outside in the community, social determinants such as access to care, income and environment play a big role in shaping maternal health risk, said Saade at UTMB. For example, he said UTMB data on adverse maternal outcomes showed access to care is a factor. In particular, he noted that while most women initiate early prenatal care, many lack access to specialized care for pre-existing conditions that impact maternal complications, such as heart problems, mental illness or high blood pressure.

Over the years, UTMB, a participant in TexasAIM, has developed an infrastructure for providing longer-term care for postpartum women at its network of clinics across southeast Texas, using strategies such as telemedicine to overcome geographic barriers. Still, frontline health providers can’t eliminate disparities and barriers to better health all by themselves, said Saade, who called for expanding coverage and investing more in women’s health research.

 “To me, the most important thing is access to quality and appropriate care as well as evidence to guide this care,” Saade said.