Truveta, a real-world data analysis company, has announced the launch of a new mother-child EHR data set.
The data set includes just over 1 million mother-child pairs collected from 30 health systems. Going back between five and 10 years, this makes it the largest mother-child EHR data set available today, according to the company. Truveta hopes researchers can find insights into the continuum of care, from pre-pregnancy through childbirth.
“I have been hearing about mother-child data for the last two years,” Michael Simonov, M.D., Truveta’s head of product management, told Fierce Healthcare. “There’s always been strong interest.”
Perinatal women and their children have historically been excluded from clinical trials and data sources, Truveta executives argue. That hinders efforts to understand the full effects of medications, vaccines and diseases on outcomes. It is also difficult to study maternal-fetal health using claims data only, executives say.
Truveta’s EHR data include clinician notes and images and are linked with claims, social determinants of health and mortality data.
“It’s over 400 variables that look at a lot of different SDOH attributes,” Simonov said. That applies to all Truveta data.
The data for mothers include all medications and immunizations taken during pregnancy as well as pregnancy outcomes, diagnoses, procedures and more. For children, data are available on the type of delivery, complications and outcomes, demographics, diagnoses, procedures and immunizations up to age 5.
A potentially powerful use case might be to better understand the various exposures that mothers have and how they affect birth or the child's outcomes, Simonov said. Another might be to study high-risk women where babies are not having good outcomes, like a prolonged NICU stay. And, it could be helpful to look for mothers that need to be seen by a maternal-fetal medicine specialist but are not.
Other uses of the data set include understanding the connection between maternal health and neonatal outcomes; monitoring postmarket safety data for chronic disease medications or vaccines; studying the use of certain medications during pregnancy and their effects on children; or exploring the correlation of pediatric conditions with birth mother demographics, medical history and more.
Earlier this year, Truveta was awarded a multimillion-dollar contract from the Centers for Disease Control and Prevention to assist with the agency’s research into maternal health, pediatric care and respiratory viruses. Other potential users of the latest data set could include health system researchers or life sciences companies, Simonov said.
The company argues that healthcare research is plagued by fragmented, heterogenous data and privacy concerns. The benefit of Truveta, it claims, is its ability to connect de-identified EHR data from millions of mothers to their children, leveraging its algorithms.
Earlier this week, Truveta also announced the availability of new concepts extracted from clinical notes, including medication details, family history and complex concepts. These are typically in unstructured text in clinician notes, which might otherwise be missing from medical claims, but Truveta extracts them with AI. In addition, a new Truveta imaging viewer enables researchers to view thousands of medical images at a time.
Truveta’s overarching data set, Truveta Data, is on more than 100 million de-identified patients across the U.S. from more than 800 hospitals and 20,000 clinics. The data are HIPAA-compliant.