In a recent study from the March of Dimes Prematurity Research Center (PRC) at the University of California, San Francisco (UCSF), scientists found that spontaneous preterm birth, which occurs without warning or explanation before 37 weeks, is fundamentally different from indicated preterm birth, which occurs as a result of doctor-recommended induction for the health of mom and baby.
This finding showed that spontaneous preterm birth shares no diagnosis risk factors in common with indicated preterm birth. It underscores the puzzling nature of spontaneous early labor, suggesting this biological phenomenon is in a scientific class of its own.
It also highlights the need to separate spontaneous and indicated preterm births when studying the causes of preterm birth. This is due to the unique nature of spontaneous preterm birth, which accounts for the majority of all preterm birth.
Currently, much of the retrospective patient-driven research surrounding preterm birth happens by bundling spontaneous and indicated preterm births together, netting insights about preterm birth as a whole, not two distinct types of preterm birth. (This ‘bundling’ is a result of the way Electronic Health Record—or EHR—data is organized; preterm birth, regardless of how it comes about, is categorized through diagnostic billing codes as a general event rather than indicated or spontaneous, leaving researchers without a clear way to separate the two in their data analysis).
“In the very recent past, scientists studying preterm birth, me included, lumped all preterm births together because we had to,” said Dr. Tony Capra, a UCSF epidemiology and biostatistics professor who was the study’s senior author along with UCSF PRC lead investigator Dr. Marina Sirota, a professor in the department of pediatrics and the acting director of the Bakar Computational Health Sciences Institute at UCSF. “So, in this cohort, where the different types of preterm births were clearly marked, we wanted to see how different these two groups really are, and how much this difference matters.”
“And what we found from our well-powered comparison between two groups in the same context and the same medical center is that everything we know about preconception risk factors for preterm birth as a whole is from indicated preterm births.”
While the findings spotlight the difficulty in predicting and mitigating spontaneous preterm birth because its causes are so elusive, the study’s power rests in setting a critical foundation for future work in the field. The study documents, for the first time in a large cohort with accurate distinction between the types of preterm birth, that spontaneous preterm birth is wholly unique, and must be studied apart from indicated preterm birth if its causes are to be untangled and its trajectory to be delayed or thwarted by modern medicine.
Critically, it resets the starting point for a targeted scientific undertaking to more rigorously analyze, with the most sophisticated technology, this subset of preterm birth and the women who experience it.
“Now that we have this cohort, and others like this, which clearly mark whether a preterm birth was indicated or spontaneous, we can use more complex data mining efforts to figure out what’s going on,” said Dr. Capra, adding that laboratory data and clinical notes will all be scoured in the next round of analyses, along with medical diagnoses codes.
The study, currently available as a preprint, was jointly first-authored by UCSF postdoctoral maternal and infant health researcher Jean Costello and UCSF dentistry DDS-PhD student Hannah Takasuka. Jacquelyn Roger, Ophelia Yin, Alice Tang, and Tomiko Oskotsky also significantly contributed.
In the paper, the authors analyzed 10,642 births that occurred at UCSF Medical Center between 2001 and 2022 and the pre-conception medical histories of individuals who delivered preterm vs. at term, via EHR data. The goal was to determine whether specific diagnoses in patients’ pre-conception histories were associated with preterm birth, assessing spontaneous and indicated preterm birth independently. Of the cohort, the preterm babies were split almost equally between spontaneous and indicated preterm births: 449 spontaneous and 418 indicated.
When looking at the group of individuals who had an indicated preterm birth, the researchers identified 18 associations between health histories before conception and the preterm birth: some associations were well known risk factors, like hypertension, diabetes, and chronic kidney disease, while others were lesser known, like blood, cardiac, and gynecological and liver conditions.
Then came the exciting part. The team ran statistical models on the spontaneous preterm birth group, curious whether their medical histories would give clues about their birth outcomes. The team expected their models to find some health history associations that could explain why the women went into spontaneous preterm labor.
“We expected to find unhealthy diagnoses associated with both indicated and spontaneous preterm birth,” said Ms. Takasuka.
Only that’s not what the model found—the model’s conclusions surprised even the most seasoned academics on the team.
“There was nothing,” said Dr. Capra of associations between medical histories and spontaneous preterm birth. “Just nothing.”
“What this showed us is that spontaneous preterm birth is even more mysterious than we once thought; it is completely different from indicated preterm birth.”
Optimistically, though, Dr. Capra said that pinpointing the depth of the mystery was the first step in uncovering its secrets, and that his and Dr. Sirota’s teams were already in the process of using more complex tools to sift through this 10,000+ birth cohort with finer detail to identify patterns in the individuals who gave birth early and spontaneously.
Dr. Costello said a possible explanation for the mystery behind spontaneous preterm birth lay in heterogeneity—or the existence of many different pathways that could lead to this outcome. “And we need more sophisticated approaches to identify all these,” she said.
Despite the work ahead, Dr. Capra said the efforts promised to be fruitful and soon.
“Now that we know where to look, I would say we’re just around the corner from some meaningful insights about spontaneous preterm birth, and a longer way off from the identification of a possible biomarker that could predict a woman’s risk for spontaneous preterm birth.”
“We have a lot of work ahead of us, but we know now that we’re looking in the right place.”