Stillbirth Working Group of Council Meeting Summary – November 22, 2022 | NICHD - Eunice Kennedy Shriver National Institute of Child Health and Human Development

On November 22, 2022, the NICHD Stillbirth Working Group of Council (the WG) convened its second meeting. The meeting, which was held virtually, was open to the public; the video recording is available online.

WG Members Present:

Co-chairs

Lucky Jain, M.D., Emory University, co-chair
Uma M. Reddy, M.D., M.P.H., Columbia University Irving Medical Center, co-chair and representative of the American College of Obstetricians and Gynecologists

Members

CAPT Amanda Cohn, M.D., Centers for Disease Control and Prevention (CDC)
Deborah Conway, M.D., University of Texas Health Science Center at San Antonio
Carla DeSisto, Ph.D., M.P.H., CDC
Ada Dieke, Dr.P.H., M.P.H., CDC
Donald Dudley, M.D., University of Virginia
Andrew Fullerton, M.P.P., March of Dimes
Isabelle Horon, Dr.P.H., CDC
Denise Jamieson, M.D., M.P.H., Emory University
Stephanie Leonard, Ph.D., Stanford University
Monica Longo, M.D., Ph.D., Maternal-Fetal Medicine Units Network, Pregnancy and Perinatology Branch, NICHD
Jenna Nobles, Ph.D., University of Wisconsin–Madison
Tina Pattara-Lau, M.D., Indian Health Service
Sonja Rasmussen, M.D., M.S., Johns Hopkins School of Medicine
Jennita Reefhuis, Ph.D., CDC
Robert M. Silver, M.D., University of Utah Health Sciences Center and representative of the Society for Maternal-Fetal Medicine
Maeve Wallace, Ph.D., M.P.H., Tulane University School of Public Health and Tropical Medicine
Monica H. Wojcik, M.D., Boston Children’s Hospital

WG Members Absent:

RADM Wanda D. Barfield, M.D., M.P.H., CDC
Catherine Vladutiu, Ph.D., M.P.H., Health Resources and Services Administration

Others Present:

Diana W. Bianchi, M.D., NICHD
Michelle Debbink, M.D., Ph.D., University of Utah
Carol Hogue, Ph.D., M.P.H., Emory University
Donna Hoyert, Ph.D., CDC
Pauline Mendola, Ph.D., University of Buffalo
Kimberly Noble Piper, Iowa Department of Public Health
Jill Wieber Lens, J.D., University of Arkansas School of Law
Other NICHD staff members

Welcome and Introductions

Diana W. Bianchi, M.D., Director, NICHD
Lucky Jain, M.D., Chair, Department of Pediatrics, Emory University School of Medicine
Uma M. Reddy, M.D., M.P.H., Vice Chair of Research and Professor of Obstetrics and Gynecology, Department of Obstetrics and Gynecology, Columbia University Irving Medical Center

Dr. Bianchi welcomed attendees and thanked those at the first session, especially those who had experienced a stillbirth, for their engagement. She expressed the WG’s commitment to joining people with this lived experience in developing a research agenda to prevent such experiences in the future.

Dr. Jain said that given the large number of stillbirths that happen—far exceeding the number of sudden infant death syndrome (SIDS) deaths—it is important to continue focusing on finding its causes and on prevention. On behalf of the co-chairs, Dr. Jain welcomed attendees to the first of two long sessions featuring expert input on stillbirth in the areas of the WG’s mandate. The day’s session would include speakers on current barriers to data collection and communities at higher risk of stillbirth.

Dr. Reddy expressed excitement about the panels of speakers and anticipated a robust discussion of the important issues they covered.

Session I: Current Barriers to Collecting Data on Stillbirths

Data Challenges in Reporting Fetal Deaths

Donald Dudley, M.D., Division Director of Maternal-Fetal Medicine, Department of Obstetrics & Gynecology, University of Virginia

Dr. Dudley shared his thoughts and insights after looking in detail at the standard fetal death certificate form. It requests a considerable amount of detail. Based on a series of queries to colleagues and acquaintances at five hospitals in Virginia, he learned that each facility has developed its own system for completing the forms, including who fills out the form, who signs off, whether it is filled out on paper or electronically, and whether there is specific guidance for completing it. In residency programs, first- and second-year residents are often the ones overseeing deliveries, so in cases of stillbirth, the attending physician may not ever complete the form.

The Virginia Department of Health does have a 25-page user manual (PDF 2.84 MB) as guidance for medical professionals filling out the form, but private physicians are unlikely to take the time to read it. In addition, Virginia has a 3-hour online video for birth registrars (MP4; streaming). One challenge presented by the death registration system in Virginia is that all pregnancy losses should be entered in its electronic death registration system, regardless of gestational age. However, outside of labor and delivery staff, few people know about fetal death certificate forms. Emergency room doctors, for example, likely are caregivers in cases of pregnancy loss but may not know about registering these deaths. As a result, the data will be incomplete.

Dr. Dudley introduced the work of the Stillbirth Collaborative Research Network (SCRN) to improve identification of cause of death. It can be a challenge to assign cause of death; this determination typically requires discussion among the doctors involved. The SCRN convened a workshop composed of worldwide experts to discuss how best to classify causes of fetal death in its own cohort (PMID: 19888051). They found that more than 35 classification systems had been developed, but none was universally applicable. SCRN researchers ultimately devised the Initial Causes of Fetal Death (INCODE) evaluation instrument in 2010 (PMID: 20664383). The researchers also created an eight-page form for identifying possible and probable causes. It organizes causes of death into seven categories: maternal medical conditions; obstetric complications; maternal or fetal hematologic conditions; fetal genetic, structural, and karyotypic abnormalities; placental infection, fetal infection, or both; placental pathologic findings; and other causes.

INCODE accounts for most of the conditions listed on the fetal death certificate form. In SCRN, an investigator reviewed each of the approximately 500 stillbirths in the cohort and discussed how to properly identify possible and probable causes of death with a colleague. This was a laborious task and, even so, cause of death could not be identified in 20% to 25% of cases. By comparison, typically a junior resident or obstetrician is the one filling out a fetal death certificate form with more than 40 values. The data will not be as thorough or complete as one would hope. Not only can it be a challenge to understand what happened; in addition, the deadline in Virginia for completing the form is 7 days, before autopsy results are returned and maybe before the placental pathology is available. It is also important to keep in mind that the fetal death certificate will only document gestational age at the time of delivery, which may differ considerably from when the baby actually died. Dr. Dudley acknowledged Dr. Conway’s work to develop methods for identifying a fetus’s actual date of death.

In summary, the fetal death certificate form is cumbersome to complete and the lack of standardization in filling it out is notable. Identifying cause of death can be difficult even for experts. One way to ameliorate data collection is to link the form to electronic medical records (EMRs). It would cut out the middlemen and would make it possible to update fetal death records as new medical information becomes available until a doctor’s final sign-off. Records updating is ripe for electronic mechanisms and doing so would make the data more reliable.

Dr. Jain announced that a public listening session will be held on January 5, and on November 10, NIH published a request for information (RFI) to gather comments from the public, including people with experience of stillbirth (NOT-HD-22-054). Responses to the RFI are due January 6. He encouraged everyone to participate in giving NICHD feedback. Input from the listening session and the RFI will be used to develop a strong stillbirth research strategy.

Fetal Death Surveillance and Research

Jennita Reefhuis, Ph.D., Chief, Birth Defects Monitoring and Research Branch, Division of Birth Defects and Infant Disorders (DBDID), National Center on Birth Defects and Developmental Disabilities (NCBDDD), CDC

Dr. Reefhuis gave an overview of the work of the DBDID, one of several groups at CDC examining fetal deaths, in response to questions raised at the WG’s first meeting. She noted that 15% of the fetal deaths with a known cause in 2020 could be attributed to birth defects, which explains why this division is so heavily invested in stillbirth research. Even birth defects that do not affect major systems, such as cleft palate, are associated with higher rates of fetal death.

The surveillance that serves as the foundation of the division’s work relies on fetal death certificates, which are used nationwide. This is a passive mode of data collection; there is no follow-up to confirm the data. Birth defects data collection varies by state and, in some locations, includes active review of the records to find cases. This approach can be time- and resource-intensive, but results in the most complete data. These methods are population-based, not clinic-based, which is necessary for understanding how the entire population is affected. This approach is essential for understanding health inequities.

CDC supports birth defects surveillance and research nationwide. Ten states have surveillance programs and 7 participate in the Birth Defects Study To Evaluate Pregnancy exposureS (BD-STEPS). In 2005, with CDC’s support, Iowa and metropolitan Atlanta expanded their surveillance work to include fetal deaths. Researchers learned that in some cases, fetal death records were filed for pregnancy losses at less than 20 weeks or in duplicate, among other reasons, contributing to inaccuracy in the overall numbers of stillbirths reported. Study findings underscored the importance of fetal death records in properly ascertaining cases. For example, in Atlanta, about 13% of cases that had reports were not included in the final numbers, typically because the gestational age was reported as less than 20 weeks.

BD-STEPS, which aimed to identify risk factors for birth defects and fetal deaths, launched in 2013. Initially, investigators focused only on live births with a birth defect, but this was expanded in 2015 to include stillbirths with outward defects. In Arkansas and Massachusetts, active fetal death surveillance was conducted, including reviewing medical records and classifying causes of death. Investigators did a case-control study of nearly 900 mothers who participated in interviews to discuss illnesses or chronic conditions, medications, stress, sleep position, and other exposures they may have experienced in pregnancy, as well as the fetal movements they remembered. Currently, researchers are cleaning the data for analysis.

Dr. Reefhuis reviewed recent findings from research conducted or supported by CDC, based on National Vital Statistics Data; data from the National Birth Defects Prevention Study (NBDPS), which preceded BD-STEPS and covered 10 states; and the metropolitan Atlanta fetal death surveillance data. In 2020, the NCHS collaborative published data from 2015–2017 showing that 1 in 166 pregnancies ended in a fetal death; no cause was identified in almost one-third of these cases. Fetal death rates are also characterized by greater ethnic and racial disparities than birth defects.

Two recent papers based on NBDPS data reported fetal death risks for specific birth defects (PMID: 31809437) and examined whether an association exists between fetal death in pregnancies with birth defects and the use of medicines to control high blood pressure (PMID: 35238183). BD-STEPS data could be used to further explore the associations discovered. An analysis of fetal death data collected between 2009 and 2015 in metropolitan Atlanta was recently published, with the finding that combining fetal death certificates and active surveillance resulted in more complete case ascertainment (PMID: 35364680).

DBDID next plans to analyze BD-STEPS data; WG discussion will help prioritize areas for further examination, such as causes of death. In addition, CDC will be releasing a funding opportunity announcement for the next phase of BD-STEPS and surveillance that links pregnancy and infant data, including in cases of fetal death.

Iowa Stillbirth Surveillance Project (ISSP)

Kimberly Noble Piper, R.N., Executive Officer, Center for Congenital and Inherited Disorders; and State Genetics Coordinator, Iowa Department of Public Health

Ms. Noble Piper described Iowa’s experience with stillbirth surveillance in the field. In 2004, Iowa legislation to improve stillbirth data collection, championed by a legislator who had lost a daughter to stillbirth, was signed into law. The intent of the law was to have the Iowa Department of Health study and develop prevention strategies to reduce stillbirths and other congenital or inherited disorders that cause newborn death and disability.

With $26,000 in funding, the state convened a working group of experts who designed a stillbirth evaluation protocol, form, and manual, based on the Wisconsin Stillbirth Service Program (WiSSP) form . The group also engaged state public health staff to roll out the protocol. The form expands on the fetal death certificate form to include additional demographic detail; maternal obstetric and medical history; a narrative section for the mother to describe her experience; gross fetal, placental, and umbilical cord exams; and details of the pathology or postmortem evaluation. The manual recommends ways to follow up with family members and culturally specific grief support information. Hospital staff fill out the form. Because stillbirths do not occur often in some facilities, staff must relearn how best to support family each time. This document is intended to guide them through this process.

Iowa’s perinatal care team, which makes regular visits to all birthing hospitals in the state, disseminated the tool. The team consists of a neonatal specialist, maternal–fetal medicine specialist, obstetric nurse, neonatal intensive care unit nurse, and dietitian. They visit birthing hospitals to review records and provide education and training regularly. In addition, the Center for Congenital and Inherited Disorders’ executive officer sent the protocol to all birthing hospitals in the state and gave in-person training at large facilities. Some medical staff pushed back on filling out the evaluation form, which is voluntary, because of the lack of reimbursement. This shows how important it is to raise awareness among those who are responsible for filling out these forms that their work is an important link in improving data and ultimately reducing stillbirth.

Data from the forms are sent to the Iowa Registry for Congenital and Inherited Disorders (IRCID) as a case-finding tool. The fetal death certificate is still the legal record in Iowa; this process is ancillary. About half of the facilities submit the voluntary form, and they arrive in various stages of completion. However, the information provided is still valuable. In response to state needs, the form has been modified to add questions about blood pressure measurements, sleeping position, whether the mother got information about fetal movement, and COVID status.

Ms. Noble Piper illustrated her presentation with the story of a mother who called the hospital help line when she felt her baby’s movements stop and was told to call back in the morning, by which time her baby had died. When this kind of experience is recorded, the statewide perinatal care team can provide education to those facilities based on the information received in the forms.

The initial $26,000 to start the program came from the state’s tobacco settlement funds. Iowa also received funding through a cooperative agreement with CDC to develop its surveillance system: ISSP ran from 2005 to 2015. ISSP’s aims were to evaluate the feasibility of expanding birth defects surveillance to include records of fetal deaths, monitor and report on the occurrence of fetal deaths, serve as a registry of fetal deaths, and be an educational resource for stillbirth prevention programs. Because the intent was to evaluate the feasibility of expanding the state’s birth defects registry to include stillbirth, it was time-limited and not intended for ongoing surveillance.

IRCID conducts active surveillance. A data manager looks for cases, even from obituaries initially, to identify potential cases for abstraction rather than relying on hospitals to report them. Between 2005 and 2013, about 1,600 cases were identified. In 2011, Iowa formed a consortium to share data with Hawaii, Colorado, and New York, which enhanced the racial and ethnic diversity of the sample and allowed Iowa to further evaluate the feasibility of its expanded surveillance.

Results from Iowa’s investigations showed that about one-quarter of the fetal death certificates examined did not meet state reporting criteria for fetal death, mostly due to inaccurate information (PMID: 21960514). Their analysis showed that in Iowa between 2005 and 2013, vital records identified 87% of reportable stillbirths; an additional 13% of reports were from program reports and medical records. About 80% of fetal death certificates that were issued were for reportable stillbirths; at the same time, about 20% of reportable stillbirths were not issued a certificate.

Staff faced other barriers to getting comprehensive data, including lack of autopsy results; lack of genetic testing; lack of photos of the fetus; poor-quality data (e.g., delivery weight, gestational age); lack of information on date of fetal death; variable information on prenatal history without going to the providers, especially for deaths that occurred in the emergency room; lack of information about maternal risk factors; and lack of training funeral home directors in completing the certificate. The main challenge was the limited availability of reliable data. Fetal death certificate forms were often incomplete or filled out incorrectly or for cases that did not meet requirements. Iowa’s experiences illustrate the need for data to track the current reality (where we are) and chart a path to the future (where we are headed).

Counting Stillbirth

Jill Wieber Lens, J.D., Robert A. Leflar Professor of Law, Associate Dean for Research & Faculty Development, University of Arkansas School of Law, Robert A. Leflar Law Center

Ms. Lens introduced herself as not only a law professor but a stillbirth mom, having delivered Caleb at 37 weeks about 5-1/2 years ago. This experience inspired the future direction of her professional work, the focus of which is how the law affects stillbirth, stillbirth rates, and data collection.

There are significant barriers to collecting data in the practice of conducting autopsies and in recording fetal deaths. In the United States, recent data show that fetal autopsies were conducted in only about 20% of fetal death cases overall, with disparities among racial and ethnic groups and by education and socioeconomic status (PMID: 36201780). There are many reasons for why there are not more fetal autopsies. Parents, who are shocked and distressed after a stillbirth may have difficulty deciding or agreeing to an autopsy. Doctors who are usually in the position of delivering happy news may find it challenging to talk about it, too. Even if they do, the language doctors use may be discouraging. For example, instead of saying that an autopsy will disfigure the baby, they could say that an autopsy gives doctors the best chance of finding a cause, even though it ultimately may not be successful. Doctors may also be worried about liability—for example, if an autopsy reveals that medical staff did something wrong. Overconfident doctors may downplay the value of an autopsy because they are so sure they know the cause. This happened in Ms. Lens’ case: Her doctor said a placental abruption was the cause and gave the impression that doing an autopsy would be beside the point. She later felt guilty that she did not do everything possible to find out why her son died. It would have been difficult to agree to an autopsy that day but being told that it would contribute to research might also have helped. In addition, cost is a barrier to conducting autopsies. Private insurance may or may not cover an autopsy and figuring out coverage in time to make a same-day decision is a challenge. Medicaid, which covers half of U.S. pregnancies and two-thirds of pregnancies in Arkansas, does not cover autopsy. Stillbirth is surprisingly expensive, so autopsy can end up as a low priority. Finally, autopsies can be a barrier to data collection due to the poor quality of those that are completed; the number of trained perinatal pathologists in the United States is low.

Fetal death certificates, which often have incomplete or inaccurate information, are another barrier to collecting stillbirth data. There is limited awareness of stillbirth and how important these forms are for conducting stillbirth research. The United States has gathered data on stillbirth since 1905. As with all vital statistics, data collection is state-based. Congress has no authority to mandate vital statistics. In 1905, stillbirths were widely recorded with both a birth certificate and a death certificate; in fact, for decades, birth certificates had a checkbox to answer the question of whether the baby was stillborn. At the federal level, the 1992 Model Vital Statistics Law aims to provide uniformity for states to follow. The standard for filing a fetal death certificate is for fetuses weighing more than 350 grams or, if the weight is unknown, of at least 20 weeks of gestational age. However, not all states follow the current definition of stillbirth; gestational age at the time of fetal death and fetal weight cutoffs differ by state. The 2020 National Vital Statistics Report notes that South Dakota only reported deaths of fetuses weighing more than 500 grams; however, the state law was changed in 2007 to the 20-week standard. New Mexico followed the 500-gram standard until 2014. Until 2012, Tennessee used a reporting standard of 500 grams or 22 weeks. When changes are made to differ from the model law, the reasoning may be unclear. These standards may be relatively close, but this variance typically is not seen in other contexts. In 1992, the Model State Vital Statistics Act and Regulations were revised to define the standard as 350 grams or weight unknown and 20 weeks. Before that, it stated the reverse, and it is unclear why it was changed or why the model law does not use the medical definition of stillbirth.

Ms. Lens showed the fetal death certificate from Caleb’s stillbirth. It recorded placental abruption as a contributing cause, which an expert review later discounted. It also listed gestational diabetes as a contributing cause. She noted how distressing this was, given that when she was diagnosed, she was told about the increased risk of a preterm delivery and how to manage her condition, but not about the possibility of stillbirth. The diagnosis was treated as no big deal. There is a disconnect between the conversations pregnant people have with their doctors and the information recorded on the forms. There appears to be a consistent theme: There is a lack of awareness about stillbirth until it happens to you. In health law, informed consent laws aim to reduce the knowledge disparity between the patient and the doctor, but this disparity remains stark with stillbirth. A baby’s death seems to come up only in the context of scaring people into doing certain procedures. The state of medical care in the United States leaves many to believe that stillbirth is not a current concern.

Fetal death certificates are filled out before the autopsy or placental examination is completed. On other death certificates, there is a question asking specifically whether the autopsy findings were available. They also have language clarifying that amendments to the certificate should be reported immediately, should findings that would change the cause of death determination become available; this is not true for fetal death certificates. There is a process for adding medical information learned after the fetal death certificate was filled out, but the doctor who originally provided the information must be the one to file such changes. There are many reasons why this might not happen, including that it is no longer the doctor’s priority, the patient does not see that doctor anymore, or the doctor did not want to conduct an autopsy or otherwise does not see the value of amending the record.

Following the start of the COVID-19 pandemic, CDC quickly realized the need for data about the virus’s possible effects in pregnancy. CDC began collecting data in April 2020, but only on live birth certificates. CDC issued suggestions to states to add whether there was a COVID infection and in which trimester it occurred. On the fetal death certificate, there was a blank line for filling in type of maternal infection, but it was removed in 2014, so CDC is not collecting this information. That makes it difficult to get information on COVID and stillbirth or conduct research to understand whether is increases risk. Hospital codes are the main source of this information. First, hospitals recorded whether there was a COVID infection at the time of delivery. Data collection about COVID and stillbirth illustrates how stillbirth is still an afterthought.

There is a blueprint for how to increase the number of autopsies conducted and thus improve data collection. Many states have declared SIDS a public health emergency and pay for autopsies after suspected SIDS deaths. In addition, Arkansas’s Reproductive Health Monitoring System, created in 1985, is a good system for gathering data about stillbirth.

Other suggestions were to focus on research as prevention and to avoid blaming people who have lost a child. Medical advancements in pregnancy care are amazing, but they create unrealistic expectations about what medicine can do. As a result, stillbirths are erased. The abortion debate contributes to this erasure because neither side of the debate acknowledges pregnancy loss. As a result, pregnancy loss is seen as an aberration and leads us to blame the pregnant person when a baby is stillborn. There are cases of women being arrested for causing their child’s stillbirth; these are almost always women of color, poor White women, substance users, or other marginalized women. While the medical community continues to discuss how difficult it is to determine a cause of stillbirth, juries are determining beyond a reasonable doubt that women should be sent to jail for causing the death of their child. It is important to keep these women’s stories in mind. After a cord accident that left her daughter stillborn, 16-year-old Rennie Gibbs was indicted for killing her child when the state medical examiner found cocaine in the fetus’s blood. Such cases are likely to increase. Pregnancy loss is almost a crime. Most states preclude convicting these women, but they may still be arrested, with all the stress that entails.

The cases of Chelsea Becker and Adora Perez illustrate how the quality of medical care can decrease after a stillbirth, when a woman is suspected of being involved in the death. In California, Ms. Becker’s baby was left on a table. After Ms. Perez fell asleep, her baby was taken away. These situations add to the trauma of being investigated by police. These women often do not have easy access to mental healthcare either. It is especially important to keep these stories in mind when designing research that is about preventing, not blaming.

Reflecting on the adage that we count what matters, Ms. Lens reflected on the state of stillbirth data collection as a sign that stillbirth does not matter. Anything less than the best job of counting stillbirths means that ending stillbirths does not matter.

Dr. Bianchi thanked all the speakers. She also responded to posts on social media to explain how the voices of people with lived experience of stillbirth were being included in WG discussions. In addition to inviting speakers like Ms. Lens, the WG created a listening session on January 5, dedicated to hearing from people who have experienced stillbirth. The November and December WG sessions are focused on hearing from experts on the four topics that make up the WG’s congressional mandate, including, where possible, people with lived experience. The WG is listening and invites comments in writing and in the additional listening session.

Discussion

Dr. Longo noted that there were several questions in the chat about data collection from placental and fetal autopsies and explained that speakers at the December meeting would also address these components.

Dr. Jain asked Ms. Lens to reflect on her experience and explain what she might have done differently, given the benefit of hindsight. She said that the question of whether to consent to an autopsy is still challenging; she did not know whether that decision would change. She said she was fortunate to get involved with the stillbirth community. Three years after her stillbirth, she consulted with Harvey Kliman, M.D., Ph.D., at Yale University, who said she had perhaps experienced a placental abnormality rather than an abruption. She wished she had had that consultation earlier. She also emphasized that finding out why our children die should not depend on whether the family can afford it.

Dr. Jain also reflected on barriers to standardizing data collection. Healthcare laws are governed at the state level, and the Supreme Court is pushing more authority and responsibility to states. How can standardization of the basic elements of data collection be improved? Ms. Lens said that the answer is money. The Model State Vital Statistics Act and Regulations are one way to give out federal guidance, but even they do not use the 20-week definition of stillbirth. Congress cannot mandate data collection, but it can give states money to encourage them to do better.

Dr. Jain noted how much information in ISSP is collected manually, which is cumbersome, given how much information is transferred from one form to another. This also could lead to poor case ascertainment. He asked how it could be done better. Ms. Piper Nobles said that plenty of resources would help. With a fetal death evaluation, there are two separate processes: the voluntary forms, which typically the nursing staff fill out and submit to the executive officer, who forwards it to the registry; and the official active surveillance process, based on the fetal death certificate and medical records. That is an electronic process.

Given the limitations of fetal death certificates, Dr. Reddy asked WG members to comment on the possibilities of combining different data sources on U.S.-based pregnancies, including interviews and samples, to do individual-level meta-analysis.Dr. Reefhuis said that pulling together different data to answer research questions is a good step forward. There are few samples available, though, and those are not representative of the U.S. population. Increasing the number of autopsies would help to have a more representative sample. Additional sites are doing interviews with parents.

Dr. Dieke said that the Utah Study of Associated Risk of Stillbirth complements active surveillance in the state. Interviews with parents help in examining social determinants of health (SDOH). In Iowa, how soon after the stillbirth were parental interviews conducted? Ms. Piper Noble said that the family interview on the fetal death evaluation form in Iowa is beneficial for all the information parents provide to illuminate their perception. Nursing staff or sometimes providers fill out the forms and have a closer relationship with the families. Usually, they conduct the interview within 1–2 days of the stillbirth.

Dr. Dudley commented that asking women to consent to an autopsy is a process, not an event. Sometimes, it requires several visits to discuss. In the SCRN, investigators obtained consent from 60% to 70% of the patients who were enrolled. In addition, research shows that the prevailing assumption that women who suffer stillbirth do not want to talk about it is wrong. Research has found that they desperately want to talk, to process the difficult emotions they are facing. Talking with providers who care is critically important.

Dr. Dudley asked how many birthing facilities are in Iowa. Ms. Piper Noble said there are 60, with average delivery volumes of 36,000 births each year. Dr. Dudley expanded on this: In Iowa, a lot of facilities may see only 300–400 women per year and experience only one stillbirth, adding another challenge to data collection. Virginia has 55 delivering facilities over a vast geographic region, whereas there are 240 delivery facilities in Texas and 250–260 in California. Trying to replicate and scale up Iowa’s program elsewhere will be challenging. Ms. Piper Noble said that the evaluation form is voluntary, and not all facilities submit it. Iowa has neither a carrot nor a stick to encourage participation, other than emphasizing the value of the information to inform stillbirth prevention efforts. Scaling up would mean losing relationships with individual facilities but otherwise would not require significant additional effort. It might be tricky for facilities to see value in participating.

Dr. Silver asked how best to frame the way forward in terms of research. The morning presentations highlighted the poor quality of fetal death certificate data. The forms contain inaccurate data because tissue evaluation is done after they are already filled out. Lack of resources also contributes. Accessing EMRs could help improve data quality, but even if the data are accurate, it is challenging to determine the cause. It requires discussion among many experts and an interview with the family, and even so, the determination will be somewhat subjective. Interviews are essential, for understanding SDOH factors that contributed, but they are expensive. In addition, tissue evaluation may be inadequate. Better evaluation of stillbirths and a more comprehensive evaluation of each case are necessary, as is tracking data. It is expensive to do this in every case. To justify the expense for the healthcare system, state health departments, and CDC, research must determine which approaches are best for families and most cost-effective. Research could include studies on decision aids for families to increase the use of autopsies, the most effective ways to counsel families, and how to work with states to do implementation science (e.g., the best stillbirth evaluation, how to get the most accurate data into the fetal death certificate, how to conduct audits with families to capture SDOH). With research findings showing the best way to improve stillbirth evaluation and to improve the accuracy of fetal death certificates, it will be easier to push legislation through to fund state health departments to do those things.

Dr. Longo forwarded a question about how the infrastructure for ISSP was created. What were its timeline and budget? Ms. Piper Noble explained that the fetal death evaluation form had $26,000 to start. This paid for convening the working group and drafting the original evaluation protocol and form. The ongoing work is part of the executive officer’s activities; there is no specific allocation for this work. With more resources, the state could do more, for example, to get more facilities to see the importance of submitting the form and to disseminate information learned from the forms. Iowa has a cooperative agreement with CDC, which provides about $300,000 per year.

Dr. Jain noted that it is important to clarify NIH’s task: research. NIH investigates the why. The what and how elements are the domain of HHS, public health authorities, and the Centers for Medicare & Medicaid Services. He emphasized that it will be important for the WG to frame its priorities as research priorities. Ultimately, though, there is a bigger component to stillbirth prevention and management that needs to be managed in the public health domain.

Dr. Bianchi concurred that issues surrounding stillbirth are not exclusively research problems. For example, problems with Medicaid coverage of autopsies are not in NIH’s purview. The focus of the report needs to be on establishing a research agenda for prevention. Yet as part of that, the WG can mention the need for involvement from other federal agencies, such as how CDC and NIH can collaborate moving forward. There are parallels here with government-wide efforts to reduce maternal mortality. NIH does research and collaborates with other agencies on how to reduce mortality. The focus of the WG is on research directly linked to prevention.

Dr. Hogue noted that the report could include recommendations to other agencies as appropriate.

Session II: Communities at Higher Risk of Stillbirth

Dr. Longo introduced the afternoon session.

Tell Us More: National Vital Statistics System (NVSS), Fetal Death Data

Donna Hoyert, Ph.D., Mortality Statistics Branch, Division of Vital Statistics, NCHS, CDC

Dr. Hoyert presented statistics on racial and ethnic disparities in stillbirth from CDC’s NVSS, a decentralized system of data collection governed by both state and federal mandates. Of note, international standards are embedded in the International Classification of Diseases (ICD), which influence what data are collected and how, how they are processed, and how causes of death are classified. The ICD standards are the same as those used for the other 3.4 million deaths in United States each year.

NVSS has kept fetal death information since 1918. A fetal death report is completed when a pregnancy ends with a fetal death. All conditions that could be considered a cause of death are assigned an ICD-10 code, which has about 8,000 categories and several levels of detail. The form has a standard tabulation list of most frequently reported causes.

Nearly 90% of fetal deaths are due to one of five causes, including unspecified cause. From 2018 to 2020, the data show that unspecified cause was stated in 31% of cases; stillbirth was attributed to placental, cord, and membrane complications in 25% of cases; maternal complications in 13% of cases; congenital malformations in 11% of cases; and maternal conditions unrelated to pregnancy in 11% of cases. Data on types of investigation of a fetal death show that placental exams are relatively common—conducted in nearly 70% of cases—whereas autopsies are performed about 20% of the time. These figures are similar across racial and ethnic groups. Trend data since 1942 for fetal mortality rates by race and Hispanic origin show significant decreases across all groups over time yet persistent disparities in rates between Black and White communities. Rates between Hispanic and non-Hispanic communities are similar.

Fetal mortality rates differ by other characteristics, including geography. For example, states with the highest fetal mortality rates have a higher percentage of unspecified reports than those with the lowest rates (38% versus 24%). Dr. Hoyert also shared data on racial and ethnic disparities in fetal mortality rates by the age of the mother; whether a single infant, twins, or triplets (or higher) were delivered; gestational age; birthweight; and tobacco use, which has the biggest racial difference.

Racial Disparities in Stillbirth

Carol Hogue, Ph.D., M.P.H., Professor Emerita of Epidemiology and Jules and Uldeen Terry Professor Emerita of Maternal and Child Health, Emory University

Dr. Hogue spoke specifically about racial disparities affecting African American women, both due to the higher stillbirth rates in that community and because more is known about the disparities they experience than those that other groups experience. She clarified that race is not a risk factor but rather a risk indicator. People in the African American community are twice as likely as those in the White community to experience stillbirth, due to experiencing risk factors more frequently. Known risk factors for excess stillbirth can be grouped into three large categories: social systems, genetic or medical factors, and individual or interpersonal factors.

Systemic factors include racism, and its effects on the environments where African American women live can increase their risk of obesity, a stillbirth risk factor. Systemic ways to address these issues include research to evaluate enumeration efforts. An example of a systemic intervention that helped reduce disparities is an NIH-funded study of school-based health centers opened in schools with a high percentage of students from homes with limited financial resources, showing an increase in the use of preventive health services by students receiving Medicaid.

Recent studies have also demonstrated the institutional racism African American women face in healthcare. A California study of women who had experienced a stillbirth between 2007 and 2011 found that 10.6% of African American mothers experienced severe maternal morbidity, compared with 5.6% of White or Hispanic mothers. Progress toward equity can be measured by reductions in factors such as maternal obesity, severe morbidity, and disparities in stillbirth.

After birth and death records were linked, starting in the 1980s, studies found that the association between mother’s education and infant mortality differed by race. Well-educated African American women were more likely than well-educated White women to deliver periviable infants at extremely low gestational ages. Later research has revealed the stress of racism and lower-quality maternal care as factors.

Dr. Hogue noted that while the fetal mortality rate is similar across races for very preterm births, African American women are three to four times as likely to deliver babies at those low gestational ages. Factors such as unnecessary cesarean deliveries can be tracked through vital records surveillance systems for live births, but this information is not available for stillbirth. Vital records tracking could be expanded to include additional measures and would be cost-effective.

Regarding the quality of fetal death certificates, for a long time, the NCHS has been unable to support better data collection in states due to a lack of resources. Federal financial support for states’ vital records collection could raise the quality of the data.

As more infants born between 20 and 25 weeks survive, those who are born alive but die soon after may not be recorded at all, because it is unclear who should record them. Situations like this argue for having the same information collected for all deliveries at 20 weeks of gestation or later.

The SCRN has shown the value of case-control studies for understanding the contribution of genetic conditions and their underlying causes to stillbirth. Medical conditions that are known risk factors for stillbirth and are more common among African American women include lupus, small-for-gestational-age fetus, twins, abruption, and preterm delivery. The disparity for stillbirth risk for infants born to African American women is starkest at very preterm ages.

Preterm delivery and stillbirth have a high correlation, especially before 24 weeks. The Pregnancy Risk Assessment Monitoring System (PRAMS) can be used for research into individual and interpersonal risk factors contributing to preterm birth and stillbirth and could be used to expand surveillance of known risk factors of stillbirth. For example, previous findings from PRAMS data indicate that chronic stress and racism may contribute to accelerated preterm birth risk among minoritized women. Measures of racial discrimination, such as the Jackson, Hogue, Phillips measure of gendered racism, could also be useful.

Data available in PRAMS include the source of funding for prenatal care and information about specific advice women received from their providers regarding marijuana or tobacco use. Analyses of such data include findings that women with limited financial resources have better pregnancy outcomes on Medicaid than with private insurance. Regarding fetal monitoring, PRAMS could add a question asking whether the patient’s provider advised them to monitor kick counts. A Utah study found an association between very preterm birth and stress, such as the loss of a job or other financial stress. A 2022 study analyzing PRAMS data by Erbetta et al. showed that racial discrimination was associated with 24% increased risk of gestational diabetes (PMID: 35928172).

More research into the elimination of discrimination may improve care for African Americans. One element of this will be emotional care, as African Americans face an increased risk of depression associated with racism and higher loads of stressful life events. Ongoing NIH studies include research into telehealth and mindfulness for depression prevention among high-risk pregnant women, but none are focused on African American women. Research into care bundles is ongoing in the United Kingdom and Australia; testing an adaptation of the content of such bundles may be a research option to consider. For example, researchers could investigate a PRAMS sample of women who experienced perinatal death and evaluate the receipt of each item, comparing those results with those for women not experiencing a fetal death. Similar studies were conducted to evaluate breastfeeding practices among women who had delivered at hospitals with and without baby-friendly breastfeeding practices.

Kaitlyn Stanhope, Ph.D., M.P.H, with the Department of Gynecology and Obstetrics at Emory University School of Medicine, has advanced research on the treatment of mothers who have experienced a stillbirth by following those who were interviewed as part of SCRN research. Two-thirds of women interviewed 6–36 months after a stillbirth said that loss was their most stressful life event. Rates were about 24% lower among women who had reported maltreatment in childhood.

More recent findings focused on coping strategies show that one in six women reported that they did not receive emotional support in the 2 months after discharge. Those who did derived support from a wide range of sources. It will be important to take this diversity into account when developing treatments.

Developing clinical education is another priority. The risk of stillbirth in a second pregnancy after the first ended in stillbirth is double the risk of stillbirth in the first pregnancy, and for African American women, the risk is even higher. One-fifth of the risk of recurrence can be attributed to a short interval between a pregnancy loss and subsequent conception. But a 2010 survey of obstetricians showed that two-thirds endorsed waiting times of less than 6 months.

Regarding screening and treatment for depression, mindfulness-based therapies are one focus of ongoing research that could be investigated for women who have experienced stillbirth.

Hearing Mothers: Using Epidemiologic and Qualitative Data Together to Explore Racial and Ethnic Inequities in Stillbirth and Pregnancy Loss in Utah

Michelle Debbink, M.D., Ph.D., Assistant Professor, Obstetrics/Gynecology, Maternal–Fetal Medicine, University of Utah

Dr. Debbink introduced a mixed-methods study focused on racial and ethnic inequities in marginalized communities in Utah. How can providers and public health experts work with women at risk of stillbirth to reduce that risk and understand what puts people at risk of risk?

Dr. Debbink introduced models of the structural causes of health inequity, including a life course model developed for maternal mortality by the World Health Organization (WHO). In 2011, Silver and Hogue published a framework for inequities in stillbirth (PMID: 21798402).

Data from Utah vital records collected between 2013 and 2019 show that people in the Native Hawaiian and Pacific Islander (NHPI) community experienced 150% to twice the stillbirth rate of other racial and ethnic groups over that period. Stillbirth rates in the Black community are also about 50% higher, although at times twice as high as those in the white community or Utah’s total stillbirth rate. The large fluctuation in rates for the American Indian/Alaska Native (AI/AN) community is due in part to low numbers overall and to various difficulties with obtaining clean data. A total of 355,310 live and stillbirths were recorded. The overall stillbirth rate in Utah was 5.8 per 1,000 live births. The rates by race and ethnic group are as follows:

NHPI: 11.6 (odds ratio [OR] 2.0)
Black or African American: 9.8 (OR 1.7)
Hispanic or Latina: 5.9 (OR 1.0)
AI/AN: 5.2 (OR 0.9)

The NHPI community is a growing population in the United States, and Utah has the highest per capita NHPI population outside Hawaii. This community also has some of the highest rates of severe maternal morbidity and SIDS in Utah. Structural factors are contributing to their risk of experience adverse pregnancy outcomes.

Structural risks can be better understood by listening to the communities most affected. Hearing the Voices of Utah Mothers is an ongoing project using a qualitative approach to understand the why and how of epidemiological data. The study uses the grounded theory approach. Researchers are conducting focus groups with NHPI and AI/AN women who have given birth in the previous 5 years. Questions were developed with local groups and cover the women’s community, factors in their environment that reduce or create barriers to support during pregnancy, and their interactions with providers and the healthcare system. Five focus groups with 32 women from the NHPI community were held in 2021; those with AI/AN women are ongoing but so far have included 12 women.

Discrimination and dismissal were big themes in the responses. NHPI women reported trauma and said they felt that providers brushed off their concerns. Stereotyping and bias were two more important themes, both within and outside the community. A lack of cultural humility created distance between the women and their White healthcare providers. The women in the focus groups felt the providers dismissed or downplayed things that could contribute to meaning-making.

Melding epidemiological and qualitative data can result in a community translational approach, spanning bench to bedside and ultimately centering communities that have been most harmed.

There are many ways to interpret the data. For those in healthcare, hearing from participants that they want providers to listen and take their pain seriously echoes what we’ve heard from African American women, which formed the basis of the CDC HEAR HER campaign. This research highlights structural influences that put NHPI women and others at risk of pregnancy loss. A lack of listening and cultural humility impedes meaning-making, which affects women’s ability to cope with feelings around loss, demonstrating where the caregiving and medical communities can improve.

Environmental Factors Associated with Stillbirth Risk

Pauline Mendola, Ph.D., Chair, Department of Epidemiology and Environmental Health, University of Buffalo

Dr. Mendola discussed two studies: the Air Quality and Reproductive Health Study (AQRHS) and the Consortium on Safe Labor (CSL), both funded through NICHD’s intramural program. AQRHS linked records from NICHD’s Consecutive Pregnancy Study and the CSL to weather and air pollution models.

The CSL collected data from electronic medical records from more than 220,000 deliveries between 2002 and 2008. The investigators linked these data to measurements of air pollution and temperature extremes from meteorology and emissions data and analyzed them by hospital referral region. For some analyses, investigators took the average of hourly environmental measurements in a region over a day, trimester, or whole pregnancy.

Investigation of temperature extremes (those at the 10th or 90th percentile for local temperatures) showed that extreme cold increased stillbirth risk 4.75 times; extreme heat was associated with 3.71 times greater risk of stillbirth (PMID: 28650842). This translated to about 1,000 additional stillbirths per year associated with both very cold and very hot conditions.

The researchers also assessed the acute effects of temperature changes. They found that each 1°C rise in temperature in warm seasons translated to an additional 4 stillbirths per 10,000 live births, an increase of about 6%. Cold weather was not associated with increased acute risk.

Data from NICHD’s Consecutive Pregnancies Study included women in Utah with at least two deliveries between 2002 and 2010, a low-risk population (PMID: 32980027). After linking to the same environmental data sources and conducting the same analyses, the researchers found that the results from the CSL and the Consecutive Pregnancies Study were similar, but in some cases, they showed higher risks for women in the Consecutive Pregnancies Study, a surprising result. Odds for stillbirth at temperature extremes in the Consecutive Pregnancies Study were up to 7 times higher than at other temperatures. The results also showed an increase in acute risk of stillbirth for each 1°C colder by about 6%.

Researchers also examined chronic and acute exposures to air pollution over the whole pregnancy and in the first trimester. Results indicated consistent risks associated with ozone exposure but no other pollutants. When examining ozone exposure as a triggering event, increased risk was highest (22%) 5 days before delivery. There were an 18% increase in risk in the first trimester and a 40% increase for whole pregnancy. The researchers concluded that about 8,000 stillbirths per year could be attributed to ozone exposure.

Mechanistically, air pollutants are associated with oxidative stress, which could be associated with stillbirth, either acutely or over time through a cascade of exposures. Temperature stress, especially at extreme heat; reduction in placental blood flow; dehydration; inflammatory response; and hypercoagulability can harm the fetus or pregnant person. Likewise, extreme cold can affect blood flow and otherwise stress the body.

Analysis of the CSL data also showed racial and ethnic disparities. Stillbirth rates were two to three times higher for Black mothers. Using two segregation indices to examine data from the U.S. Census and the American Community Survey, the researchers found that lower segregation had more benefit for Black mothers, with the most positive effects in the areas with least segregation (PMID: 29715639). Stillbirth risk was lowered as much as 80% when segregation decreased over time. No increased risks were found for White mothers. This could translate to 900 fewer stillbirths among Black mothers.

Large population studies have demonstrated high stillbirth risks associated with environmental factors. In addition, factors such as segregation and temperature and air pollution extremes are increasing. With climate change, we are also experiencing more extreme weather events, including flooding, hurricanes, and wildfires. These factors and pregnant people’s social and living environments should be considered in future research.

Discussion

Dr. Longo relayed a question from the chat for Drs. Hoyert and Hogue seeking clarification on the role of racism in stillbirth. Dr. Hogue said that the SCRN’s case-control study assessing stressful life events, researchers used a 13-item survey developed at CDC in interviews at the time of the stillbirth. A higher number of stressful life events was associated with stillbirth risk in that pregnancy. Although there was no statistical interaction between African American, Hispanic, and White women regarding the risk of stressful life events, African American women reported far more stressful life events. Racism can lead to stillbirth by being the trigger of stressful life events. One possible trigger is maternal depression. Interviewers asked about depression at a follow-up session and found it to be a risk factor for stillbirth. African American women are at greater risk for depression, due to gendered racism and stressful life events. Gendered racism reflects the stressors to which African American women are uniquely exposed. The Jackson, Hogue, Phillips gendered racism measure is a 39-item instrument measure to assess those stressors.

Dr. Dudley asked about the possible mechanisms for associating stillbirth with environmental conditions. Are there any animal models to look at air pollution and pathology leading to stillbirth? Dr. Mendola said there are, but litter size and other characteristics of the animals make it difficult to examine stillbirth and pregnancy loss, and there are similar animal model limitations in studying preterm birth. There are more data for air pollution models than for temperature models; she and her colleagues were shocked by their own results.

Dr. Jain asked about the effect of structural racism on the delivery of healthcare and whether there are things providers can do to remove gaps in healthcare delivery. Some structural issues may take decades to solve, but how can the medical community start making improvements in their patients’ care immediately? Dr. Debbink said that simulation training can help providers repair cultural ruptures, be aware of implicit bias, and avoid microaggressions. They could adopt a design similar to simulations currently being planned to help providers interact in a way that centers anti-bias and avoids stigma around substance use disorder. There are opportunities for providers to practice breaking bad news and having respectful interactions with their patients. Health systems, especially those at academic institutions, can center patients’ experiences by including them on advisory boards and helping them navigate financial services, insurance, and other aspects of the patient experience to achieve outcomes they desire. Care navigators based on the community health worker model are another possible intervention. Racism defines access to people who know how to navigate the system, but healthcare systems can act now to interrupt this.

Dr. Reddy asked WG members to propose studies that could elevate the field’s knowledge and help deliver solutions. Members provided the following suggestions:

Dr. Hoyert said one priority is to improve cause-of-death data and to be able to make better use of that information to know what to do. Studies could investigate what different locations have tried to do to lower the number of forms with an unspecified cause of death and compare their results to see what works. The lessons learned could be expanded across the country.
Dr. Hogue suggested pilot studies of increased measures of racism in PRAMS, linking fetal death certificates and recruiting women in areas with high proportions of African American women and in high enough numbers to test the methodology. There is an additional possibility of looking at correlations between extreme prematurity and stillbirth.
Dr. Debbink said that in her work, NHPI mothers have asked for implementation of perinatal mental healthcare to diminish adverse outcomes. This could be done with a rollout study or stepped wedge design. In addition, there is a unique opportunity to study spikes in air pollution in Salt Lake City, where day-to-day fluctuations are unlike those elsewhere in the United States due to inversions, wildfires, and other factors. A Diné member of their focus group said that when she was pregnant, she would leave the city and go to the reservation every week so she could get to clean air.
While noting that immediate clinical questions are a higher priority for research dollars, Dr. Mendola said it is important to look at macroenvironmental stressors because of their large effect. It would be good to see the results of previous studies confirmed in more recent cohorts, because these factors are unappreciated risks. Conversations about pollution standards do not typically take pregnancy into account but should. Research should look at these effects more systematically and provide protection where possible. People living in substandard housing cannot mitigate temperature risks.
Dr. Hogue added that high-risk communities have not been involved in studies investigating the potential of mindfulness-based interventions, such as group telehealth. Data from these studies could be combined across communities and states or with clinical populations.

Dr. Jain noted that one common theme in the morning and afternoon sessions was the importance of accurate data about stillbirth and accurate assessment of cause of death. Without both, there will continue to be a struggle to frame public health policy and convince people to do right by the mothers who have been affected.

Dr. Reefhuis noted that solid case definitions are a necessary starting point so that it is clear everyone is talking about the same thing. It would allow researchers to analyze cord incidents compared with other causes, for example, and there is much to be gained by looking at subsets of stillbirth. Right now, they are all in one group. Dr. Reddy concurred that combining information would be extremely helpful; information from regional studies could be combined to answer questions until better data are available from fetal death certificates. Combining data would also help individual meta-analysis. It is likely that parents who participate would be willing to provide more information.

Dr. Dudley proposed implementation research as a priority. Research could examine ways to link EMRs to fetal death certificates so that there is less pressure on frontline staff. Doing so could also improve the number of completed certificates. Research could also help determine best practices for consenting families for autopsies; consenting has not been studied systematically, but research could help determine how to best inform families about their options sensitively and culturally appropriately.

Noting that there are at least 20,000 stillbirths annually, Dr. Jain asked whether there is an opportunity to create a national biorepository. In cases where no cause is readily apparent, would there be value to genome sequencing? Dr. Dudley said that a national biorepository would be a great resource, but given the expense and space required, it would be important to define its purpose clearly. Dr. Reddy added that a speaker at the WG’s next meeting will present on genetics. SCRN research compared microarrays and karyotypes to see whether they could increase the number of genomic causes for stillbirth, and further research with SCRN samples examined whole exomes and found other genetic changes associated with stillbirth. The researchers concluded that about 18% of stillbirths have a likely genetic cause. Knowing this can influence providers’ management of the next pregnancy and is important for families. Dr. Jain said that with the cost of whole genome sequencing now below $1,000 per test, such research is within reach. It would be beneficial to look even deeper at some cases.

Dr. Hogue asked whether it would be important to have whole genome sequencing for live births and preterm births, too, to get the most out of that data. Dr. Reddy said that it would be possible to compare data from stillbirths with large databases already available for live births.

Dr. Longo relayed a question from the chat: Could interventions for cleaner indoor air improve pregnancy outcomes? Dr. Mendola said that the data in her presentation were all about outdoor air, not temperatures in the house. Results from a colleague’s research indicate that air filtration may lower some exposures in the home. However, none of the studies are big enough to look at stillbirth. It would be interesting to compare a working filter and a sham filter.

Dr. Longo relayed another question: Why is the U.S. ranking in the average rate of stillbirth reduction so low? Dr. Reddy said one reason is racial disparities in the United States. It is important to understand why Black women have three times the risk of stillbirth that White women have. Research must get at issues such as access to care, quality of care, and empowering individuals to ask questions about their care. To reduce stillbirths, it is essential to understand why they occur. Dr. Hogue added that another contributing factor is that many countries count stillbirths at 24 or 26 weeks, while the United States uses the 20-week standard. A large proportion of stillbirths happen between 20 and 24 weeks. Dr. Reddy noted that there is still a disparity among high-income countries when looking at stillbirths above 28 weeks.

Wrap-Up

Dr. Jain emphasized that the WG wanted everyone to feel they had an opportunity to weigh in on its work. The WG’s next meeting is December 9, a virtual public listening session is scheduled for January 5, and the request for information (RFI) remains open until January 6. He invited attendees to send feedback on strategies for research to reduce stillbirth in the meeting question and answer session, the listening session, and the RFI.

Dr. Bianchi thanked the co-chairs, moderator, and speakers. She highlighted two opportunities for research: the NIH Climate Change and Health Initiative, in which NICHD participates and which would support research on temperature issues; and NICHD’s Data and Specimen Hub (DASH) resource, which includes deidentified clinical data many NIH studies. Data from the CSL are the most downloaded study data in DASH.