The 10,000 Women Changing the Future of Pregnancy

Improving maternal health requires scientific leaps. These women are helping make it happen.

by Molly Dickens, PhD, Head of Content and Community

We want to raise a glass to the 10,000 Bloomlife moms who helped chart a course to revolutionize maternal health.

These women have created something monumental: The largest dataset on physiological changes during pregnancy. A dataset that is helping to unlock the mysteries of the pregnant body and dramatically impact the problematic chasms in women’s health research and maternal health care.

Why this dataset is monumental

Pregnancy is a notoriously tricky subject for researchers. The female body, and women’s health in general, is a neglected field of study. But when the female body becomes pregnant, the research pool narrows further and the knowledge gap widens. Research on maternal health, women’s health during pregnancy, childbirth and postpartum, suffers from a perfect, scientific-progress-squashing storm: funding issues, politics, lack of technological innovation, the general difficulties when it comes to studying a human subject who is carrying another developing human subject, and the long-standing belief that pregnancy requires special protections for research participation.

As a result, at the most basic, biological level, we still have an incomplete understanding about the normal physiological changes that accompany pregnancy. And without knowing what is normal, we cannot identify the abnormal that can lead to pregnancy complications. This knowledge gap is especially problematic in a time when pregnancy complications, preterm birth, and maternal mortality are on the rise.

“Preterm delivery rate has not changed substantially in the last 50 years. One of the biggest problems is that we don’t really know how to predict labor at term, let alone be able to extrapolate that to the more important time frame — preterm labor.” — Dr. John Elliott, MD, MFM

So why is our dataset monumental? For one, it represents an ability to overcome limitations that have held back scientific progress in the field of pregnancy research. It represents the power of crowdsourcing data — providing a way for women to directly impact maternal health research by contributing to a database.

And it works. From the beginning, we set out with a specific scientific purpose: defining physiological biomarkers that provide clinical value for maternal health and prenatal care. Today, we have already used the dataset to identify our first set of biomarkers — uterine muscle activity (electrohysterography, EHG) combined with maternal heart rate parameters (heart rate and heart rate variability, HRV) for labor detection.

“The gathering of data points that allow researchers to sort through these signals of the uterus, at term, gives us an idea of whether that could be extrapolated to preterm labor.” — Dr. John Elliott, MD, MFM

In less than two years, we know more about how the changes in these biomarkers signal labor than any other currently available research. Our technology will soon detect when a woman is or is not in labor and give that information right back to her and her care team.

The power of longitudinal data

This dataset is also monumental because we have done the near-impossible: gathered high quality, longitudinal, physiological data in less than two years. Our dataset includes 10,000 pregnancies, 360,000 hours of data recording, across the United States. Data donated by a diverse population of women, across the United States, without geographic or care provider restrictions. Women in rural, urban, and suburban environments with a range of pregnancy experiences (e.g. ranging from high risk to low risk). Women gathering data while comfortably at home, at random times of day and often overnight for the last weeks and months of pregnancy.

“This is observational research: you observe and connect the dots. The more observations you have, the greater the chance that you come up with the dots to connect. And, oftentimes, when you think you know what you’re looking for, you find out that it was something that you didn’t expect. Having more data points allows you to come up with the unexpected. That’s the value of a robust database.” — Dr. John Elliot, MD, MFM

A traditional clinical study to acquire a similar dataset would require a laundry list of hoops to jump through: massive amounts of funding, participant recruitment, participant retention, dedicated clinics across the country with a diverse community, and active participation from clinician staff and resources. Plus, appropriate technology able to capture the physiological data, outcome tracking, and protocol implementation strategies. The more participants, the more clinics involved, the more data points gathered from each participant, the more expensive and time consuming the study gets.

The next largest available physiological dataset (also capturing EHG), came out of a Slovenian study with data from 300 pregnancies and less than 300 hours recorded at regular prenatal exams. It took over eight years to collect this dataset.

Another key distinguishing feature of our dataset from those currently available: the time points recorded. Other available datasets contain only snapshots — recordings taken opportunistically from a woman at a doctor’s appointment at a specific time of day or in labor. Our dataset is based on longitudinal data — recordings taken from women, going about their regular activities for the last weeks/months of their pregnancy. Longitudinal data allows visibility into how a woman’s physiology both trends and changes. Longitudinal data adds a time domain, a key factor for revolutionizing medical research.

This is not to say that we don’t believe in the power of the clinical study. We absolutely do. But we want to demonstrate alternative strategies to move through the discovery stage faster.

“ [With a clinical study], you’re really limited in a number of observations that you can make. You’re limited by study design. A clinical study, with a specific study design is oftentimes an extension of the of the observational research that has been done. In this area, by Bloomlife” — Dr. John Elliot, MD, MFM

Identifying our first biomarkers using consumer-generated data has informed our clinical research. We now know how to fine-tune our algorithms, our technology, and our output to adapt our model for preterm labor. The next step? Putting these questions through the rigorous control of the clinical study.

Maternal health needs scientific leaps, not baby steps.

Maternal health and our understanding of the pregnant and postpartum body is in the dark ages relative to nearly every other medical and research field. In order to deliver the care that women deserve, we need to fill in the knowledge gaps now. We need scientific leaps, not baby steps.

At Bloomlife, with the help of over 10,000 wonderful women, we are leaping.

Every Bloomlife mom, past, present, and future, has the opportunity to join the maternal health revolution and contribute data to our research efforts. If you are expecting and interested in contributing, learn more about the Bloomlife pregnancy tracker here.

If you are a healthcare professional, interested in hearing more about our technology, research, or community partnership opportunities, find that here.