Executive summary: Workshop on preterm birth, January 28–29, 2025, co‐sponsored by the American College of Obstetricians and Gynecologists and the Eunice Kennedy Shriver National Institute of Child Health and Human Development

Abstract

The Society for Maternal‐Fetal Medicine collaborated with the American College of Obstetricians and Gynecologists and the Eunice Kennedy Shriver National Institute of Child Health and Human Development at the National Institutes of Health to co‐sponsor the annual SMFM President's Workshop on preterm birth. The workshop focused on spontaneous preterm birth and sought to explore clinical evidence, preclinical and translational evidence, environmental and social factors, and opportunities to collaborate with stakeholders on future research.

1. INTRODUCTION

Globally, the preterm birth rate has stagnated at just under 10% of all live births for at least a decade [1]. In 2023, preterm birth was the leading cause of neonatal mortality across the globe and accounted for 18% of all deaths among children under 5 years old. In the United States, the overall preterm birth rate was 10.4% in 2023, and eight states had rates ≥11.5% [2, 3]. Marked disparities in preterm birth persist, especially for Black birthing people—a result of anti‐Black racism, not Black race [2, 4]. Despite recognition, research, and guidelines, efforts to move the needle on preterm birth have had limited impact on rates of prematurity. To understand the barriers to progress and identify tangible areas where future efforts should be focused, the Society for Maternal‐Fetal Medicine (SMFM) collaborated with the American College of Obstetricians and Gynecologists and the Eunice Kennedy Shriver National Institute of Child Health and Human Development at the National Institutes of Health to co‐sponsor the annual SMFM President's Workshop on preterm birth. Workshop leaders, speakers, and invited guests represented obstetrics and gynecology; maternal‐fetal medicine; clinical, preclinical, and translational research; pharmacology; immunology; biomedical engineering; clinical informatics; and pharmaceutical research and development. The workshop also included experts in global health, health policy, and advocacy, as well as professional organizations representing various types of providers in obstetric care. The workshop focused on spontaneous preterm birth and sought to explore the following:

• Clinical evidence: landmark clinical trials, historic approaches, and new directions

• Preclinical and translational evidence: promising new strategies and informative basic and translational research

• Environmental and social factors: environmental, social, and racism‐related stress and social determinants of health (SDOH)

• Opportunities to collaborate with stakeholders on future research

2. BACKGROUND

• In 2020, there were an estimated 13.4 million preterm births worldwide, half of which occurred in just eight countries: Pakistan, Nigeria, China, Ethiopia, Bangladesh, the Democratic Republic of the Congo, and the United States [1]. Mid‐trimester cervical length measurement and obstetric history are the best available predictors of preterm birth [5]. In the United States, the preterm birth rate was essentially unchanged between 2022 and 2023 [2].

• Landmark clinical trials in the history of preterm birth prevention have evaluated transvaginal ultrasound assessment of the cervix [6], cervical cerclage for individuals with a prior mid‐trimester preterm birth [7, 8] or cervical shortening [9], and vaginal progesterone supplementation for individuals with a previous preterm birth [10] or short cervix [11]. Additional interventions that have been studied with varying results include low‐dose aspirin [12, 13], omega‐3 supplementation [14, 15], screening for infections such as bacterial vaginosis [16], and avoidance of medical procedures that may be associated with cervical insufficiency [17, 18, 19]. Preventative strategies that are no longer recommended based on the results of clinical trials include pessary [20] and 17 alpha‐hydroxyprogesterone caproate [21].

• Guidelines differ in their recommendations for preterm birth prediction and prevention. Some guidelines, such as those published by the French College of Gynaecologists and Obstetricians [22] and the International Society of Ultrasound in Obstetrics and Gynecology [23], offer a comprehensive approach to the topic. Other guidelines largely focus on individual predictors and interventions [24, 25, 26, 27]. Specific recommendations vary across guidelines. In general, guidelines discourage activity restriction, pessary, and bacterial vaginosis screening. For the most part, guidelines support low‐dose aspirin for preeclampsia prevention, but specifics vary [28, 29, 30, 31]. Most guidelines address cervical length screening in low‐ and high‐risk populations and progesterone therapy for at‐risk populations, although the risk definitions and the strength of recommendations vary.

• Intraamniotic infection has been causally linked to spontaneous preterm birth [32, 33]. Preclinical models and translational studies of intraamniotic inflammation/infection (resulting from ascending cervicovaginal microbiota or their products [34] or, less commonly, from transplacental transfer [35]) have revealed underlying mechanisms associated with preterm labor onset. Such underlying immune mechanisms may be promising therapeutic targets [36, 37, 38, 39, 40, 41, 42].

• Spontaneous preterm birth also may be related to etiologies other than inflammation and infection. Cervical stiffness, cervical stromal composition, cervical mucus properties, and cervical tissue interaction with the cervicovaginal microbiome are among the factors that have been postulated to contribute to subtypes of spontaneous preterm birth [43, 44, 45, 46, 47, 48]. Biomedical engineering approaches can inform novel models, tests, and interventions for preterm birth related to cervical insufficiency [49, 50]. Engineering approaches can also be used to study cervical tissue remodeling [45, 46, 47] and cervical mucus properties [43, 48]. For example, digital twins, which are virtual models of human disease, can inform the mechanics of cerclage.

• Genetic variation also influences the risk of preterm birth. Candidate gene studies have shown inconsistent results, while evaluations of epigenetics show promise [51]. In addition to maternal genotype, fetal genotype, the environment, and gene–drug interactions (e.g., pharmacogenomics) are important to consider [52].

• Chemicals with the potential to cause reproductive harm are ubiquitous and inadequately regulated, but exposure is challenging to study and not equally distributed [53, 54, 55]. Environmental exposures such as extreme heat [56], air pollution [57], and wildfire smoke [58] contribute to adverse pregnancy outcomes and, like chemical exposure, these outcomes could be modified with policy change or behavioral interventions.

• While the effect of stress on preterm birth is biologically plausible [59], the existing literature is mixed because of the difficulty of distinguishing between stressor and stress response (e.g., anxiety) and between state versus trait anxiety. Studies among non‐Hispanic Black people have highlighted stress and its association with SDOH as a significant factor in the experience of preterm birth [60, 61].

3. KEY FINDINGS AND PRELIMINARY RECOMMENDATIONS

Following background presentations, workshop participants gathered into three breakout groups to identify priorities and the resources and actions needed to achieve them. Discussions of each group are summarized here.

3.1. Breakout group 1: Clinical evidence for current and emerging therapies

The group identified and ranked eight priority areas for preterm birth research. Some priorities require more interventional research, while others would benefit from implementation research.

Priority: Focus on teaching accurate, reproducible acquisition of correct cervical length images and measurements. Improve cervical assessment and treatment by consistently defining “short” cervix and advancing best practices for measurement through formal training. The use of artificial intelligence (AI) for assessing the cervix could mitigate bias.

Priority: Expand research on cerclage in those without a history of spontaneous preterm birth, use of cerclage beyond 24 weeks of gestation, and indications for cerclage in multiple gestations. Moreover, the utility of amniocentesis prior to cerclage placement and additional perioperative interventions (e.g., antibiotics or tocolytics) with cerclage placement needs further investigation.

Priority: Assess the benefits of patient‐level interventions, such as stress reduction through yoga or anxiety mitigation techniques, to reduce preterm birth in at‐risk populations. Evaluate clinical care models and social interventions, such as risk stratification tools and bundled care, that address SDOH.

Priority: Settle outstanding questions about the dose, timing, efficacy, and type of progestogen and the subpopulations most likely to benefit from it. Similarly, determine the appropriate dose and efficacy of aspirin to prevent preterm birth and the subpopulations most likely to benefit (potentially using existing data from aspirin trials).

Priority: Evaluate other potential therapeutic targets such as antibiotics, anti‐inflammatory agents, antiplatelet therapies, nutritional supplements, and novel therapeutic targets. Note that this approach requires developing screening strategies to stratify high‐risk groups that would benefit from such targeted therapies.

Priority: Leverage personalized or precision medicine, machine learning, and big data to stratify those at risk for preterm birth. Large data sets must be scrutinized for validity, reliability, and reproducibility of data collected to prevent untrustworthy findings.

Priority: Conduct more implementation research studies. Consider novel approaches to research, such as step‐wise trials or hybrid trial designs. Standardize data collection tools, outcome selection, and protocols for clinical trials, so that findings may be combined or compared across trials. Quantify and compare outcomes to determine population‐level effects and benefits following the implementation of new guidelines regarding preterm birth screening or treatment guidelines.

Breakout group 1 noted that existing trials use varying outcomes and composite measures, although the Core Outcomes in Women's Health (CROWN) Initiative was proposed to encourage standard reporting in all relevant trials [62]. In addition, better community engagement (including via social media) can help combat misinformation, educate the public on pregnancy complications such as prematurity, and promote trust in clinical research. However, funding for such engagement is lacking. Funding opportunities for efforts to reduce preterm birth should allow for and incorporate costs related to community engagement.

3.2. Breakout group 2: Preclinical and translational evidence for current and emerging therapies

The group focused on the importance of more precise subphenotyping of spontaneous preterm birth, drawn from molecular and biomedical engineering techniques to drive a more mechanistic approach to diagnostics and therapeutics, rather than a one‐size‐fits‐all approach. Particular attention was paid to the immunology of preterm labor; the role of the cervicovaginal microbiome in preterm birth; and the utility of biomedical engineering approaches to stiffen or support the cervix, improve the quality of cervical mucus, and model cervical changes across normal and high‐risk pregnancy using digital twins to better understand preterm birth. Advancing knowledge will also require developing common data elements, data sharing, biobanking, integrating heterogeneous data (potentially using AI), increasing and expanding funding sources, and improving communication of results among researchers and clinicians.

Priority: Leveraging clinical trials on preterm birth, collect biological specimens that can facilitate a more mechanistic understanding of precise subphenotypes of preterm birth. Once there is a better understanding of the mechanisms, investigators should integrate more translational approaches into clinical trials, such as applying more precise subphenotypes to the development of targeted diagnostics and therapeutics. Large and diverse biospecimen banks can aid the precision phenotyping needed, which ideally would be collected in a standardized manner with minimum agreed‐upon associated metadata (e.g., common data elements).

Priority: Expand options for home‐based data collection and patient involvement in research—for example, by using wearable devices (e.g., smartwatches and fitness trackers). This approach requires at‐home application of noninvasive assays in preclinical development, partnership with industry, and analytic capacity for high‐density data. It could lead to patient‐driven pregnancy monitoring.

Priority: Use technology such as predictive modeling, organ‐on‐a‐chip, or digital twins to reduce the economic burdens of trials and move toward more personalized risk prediction. This approach requires biological and psychometric data collection on a multi‐omic scale and collaboration between multidisciplinary scientists and clinicians.

Priority: Close the gap between scientific discovery and clinical implementation. Initial steps include better understanding business and regulatory issues, connecting with venture capital partners, and educating potential funders about the complexity of preterm birth phenotypes (e.g., standardizing language).

Priority: Leverage placental pathology as a “report card” to better understand the index pregnancy and inform subsequent pregnancies. This priority will require consensus on standards for clinical pathology assessments that include all spontaneous preterm births, with a subset used to establish more advanced phenotyping through a widely distributed national specimen bank.

The breakout group recommended convening a working group, modeled on the Stillbirth Surveillance Consortium [63], to

• address common data elements and common ultrasound measurements

• reach agreement on spontaneous preterm birth subphenotyping

• discuss sample collection and the establishment of a nationally distributed pregnancy biobank

Stakeholders include maternal‐fetal medicine specialists, basic and translational scientists, placental pathologists, neonatologists, biomedical engineers, the US Food and Drug Administration, business‐oriented individuals/communicators, patients, and community representatives.

3.3. Breakout group 3: Environmental and social factors

The breakout group categorized environmental exposures as encompassing multiple domains, including

• personal factors (e.g., stress, racism, relationships)

• neighborhood composition (e.g., green space, food availability)

• traditional pollutants (e.g., air and water pollution, per‐ and polyfluoroalkyl substances [PFAS])

• community and global factors (e.g., waste management, global warming, pesticides)

Addressing these factors requires additional financial resources, support and collaboration with and among healthcare providers, leveraging of evidence to inform policy and prioritize research, and more innovative ideas.

As the group identified priorities, several themes emerged. New methodologies are needed, such as mixed‐methods research, as is an increased focus on implementation and translational science. Research should incorporate biological samples and biomarkers more frequently and utilize national and global networks to facilitate collaboration across centers. Evidence must be translated into policy.

Moreover, stakeholder and community engagement are vital. Research should strive to ensure that individuals and communities see the results of the work they have made possible. More research should focus on positive factors promoting community and provider resilience. Finally, teamwork is essential. Multicenter and international collaborations and research networks promote team science.

Priority: Advance health equity by acknowledging inequity and committing to improvement. Compile existing resources into a comprehensive and more widely accepted source. Break down the silos among patients, communities, providers, and systems to help build meaningful relationships and long‐lasting engagement. Develop more objective metrics of equity, such as the time to receive care or initiate treatment.

Priority: Address environmental exposure and promote environmental justice. Develop a global approach and data to support environmental equity. For example, create a comprehensive repository of exposure data. Identify ways to boost community resilience. Broaden public health collaborations to include, for example, city planners and architects. Conduct basic science experiments to determine the timing and threshold of exposures and outcomes. Capture individuals’ environmental exposures through wearable devices and geocoded addresses (corresponding to home and work) and integrate those data into broader studies. Engage statisticians who can integrate data using advanced methodologies, including AI and machine learning.

Priority: Implement psychosocial interventions and support. Raise awareness that more equitable health care, mental health support, and supportive workforce policies can mitigate maternal stressors. Promote partnerships, including care coordination teams, to help reduce stress. More randomized trials should be conducted to demonstrate the benefits of support and individualized care specific to preterm birth prevention. Standardized, objective measures of stress‐related biological exposures and outcomes are needed, recognizing the need for multiple, longitudinal assessments.

4. STAKEHOLDER PANEL

Several nonprofit and industry stakeholders shared information about their organizations. Apple conducts its own health research, and Apple devices support secure health data storage. Apple offers tools such as ResearchKit to create study components (e.g., consent forms and surveys optimized for digital devices), capture sensor data, and implement studies without building a custom platform. Investigators can use Apple CareKit to customize and share treatment plans between patients and providers. Apple's investigator support program can furnish devices to patients enrolled in a study.

Ferring Pharmaceuticals, a specialty biopharmaceutical group with a spectrum of maternal and reproductive health products from contraception to birth, offers generous paid leave and wellness initiatives that can be a model for how companies treat pregnant employees. Ferring is engaged in public–private partnerships focused on equitable access to necessary pharmaceuticals to reduce maternal mortality and morbidity.

The March of Dimes has a long history of improving maternal and child health (e.g., implementing newborn screenings and promoting polio vaccines). It is a trusted source for data, and reports include detailed trends of preterm birth rates overall and among subgroups in the United States. The March of Dimes aims to reduce preventable maternal and infant mortality and morbidity across the pregnancy journey by addressing access to care, hypertensive disorders of pregnancy, and preterm birth, among other conditions. It provides millions of dollars in funding for original scientific research each year.

Organon produces drugs, devices, and diagnostics focused on women's health (including the Jada device for postpartum hemorrhage). It is seeking to collaborate to stimulate joint action and innovation. For example, it has partnered with Pregnolia (described below) on clinical development and regulatory plans.

The Pregnolia System is a simple device that safely and objectively measures cervical stiffness—a novel biomarker for cervical maturation—using an aspiration technique. Pregnolia's clinical research in Europe focuses on tools to assess the risk of spontaneous preterm birth within 7 days among symptomatic individuals.

5. OTHER DISCUSSION POINTS AND NEXT STEPS

5.1. The future of research and development investment

Workshop participants agreed on the need for alternative funding approaches given the current climate around federal funding for research. Cuts to NIH funding will necessitate rethinking the research landscape, redistributing funding to more investigators and laboratories. New approaches could involve increasing the number of multicenter studies, opening archives, making data sharing more facile, and crowd‐sourcing. Despite uncertainty about future funding and concerns about the devaluing of science, healthcare providers and researchers will continue to focus on helping patients achieve healthy pregnancies.

Data clearly show that addressing environmental and social factors improves health, although concerns persist about a waning focus on health equity [64, 65, 66]. Continuing to stratify data to show where interventions can have the most impact remains a persuasive argument for funders and legislative partners.

The lack of real‐world clinical uptake frustrates developers who have invested millions in new technology. Historically, clinicians have been skeptical of new technology; however, they may be more likely to adopt new tools and strategies quickly if these approaches have been proven effective by the academic community. The development and utilization of risk stratification tools, for example, is complicated by the inability to link the risk for preterm birth with effective, evidence‐based interventions (ranging from mitigation of exposures to treatment with medication). It also underscores the importance of conducting implementation research to illuminate how effective new products can and should be used in real‐life clinical settings.

5.2. Collaboration and commitment are key

The workshop elicited a broad consensus on the need for collaboration moving forward. Biomedical engineers are eager to tackle tough issues in reproductive health. More team science and more communication across the domains of medicine and engineering will pave the way for better technology. Collaboration with relevant influential voices in social media can help disseminate accurate, evidence‐based information around maternal‐fetal medicine. Acknowledging inequity and making a personal commitment to advocate on the patient's behalf can effectively build trust with individual patients.

6.

The use of this information is voluntary, and clinicians should be familiar with and comply with all applicable laws and regulations.

All authors and committee members have filed a disclosure of interests delineating personal, professional, business, or other relevant financial or nonfinancial interests in relation to this publication. Any substantial conflicts of interest have been addressed through a process approved by the Society for Maternal‐Fetal Medicine (SMFM) Board of Directors. SMFM has neither solicited nor accepted any commercial involvement in the specific content development of this publication.

This document has undergone an internal peer review through a multilevel committee process within SMFM. This review involves critique and feedback from the SMFM Document Review Committee and final approval by the SMFM Executive Committee. SMFM accepts sole responsibility for the document content. SMFM publications do not undergo editorial and peer review by Pregnancy. Further details regarding SMFM publications can be found at www.smfm.org/publications.

SMFM recognizes that obstetrical patients have diverse gender identities and strives to use gender‐inclusive language in all publications. SMFM uses terms such as “pregnant person” and “pregnant individual” and the singular pronoun “they.” When describing study populations used in research, SMFM uses the terminology reported by the study investigators.

All questions or comments regarding the document should be referred to the SMFM Publications Committee at pubs@smfm.org.

Reprints will not be available.