Real-World Data Improves Early Detection

A recent study showed using real-world blood pressure (BP) measurements to identify pediatric patients at risk of hypertension-related diseases in adulthood could potentially optimize patient care and help save on costs.

Mark Mitsnefes, MD, MS, director of the Clinical and Translational Research Center at Cincinnati Children’s, was one of the authors of the multi-center study published in the December 2023 issue of eBioMedicine.

“We know that accurate diagnosis of hypertension in childhood is important, since, if untreated it can lead to cardiovascular and kidney problems in adulthood, like myocardial infarction, stroke and kidney failure. But the way we currently identify hypertension in pediatric patients is based on normative data compiled from research and relatively small studies largely conducted decades ago,” Mitsnefes says.

To see if normative values could be created from real-world data, Mitsnefes and a team of researchers from eight large health systems gathered the electronic health records (EHR) of more than 8 million children—derived from the health systems and aggregated by PEDSnet, a multi-institutional pediatric health system network—and filtered them to narrow the sample size.

The study’s participating health systems included:

• Cincinnati Children’s
• Children’s Hospital Colorado
• Children’s Hospital of Philadelphia
• Lurie Children’s Hospital of Chicago
• Nationwide Children’s Hospital
• Nemours Children’s Health
• Seattle Children’s Hospital
• Stanford Medicine Children’s Health

Children between the ages of 3 and 17 years with a body mass index (BMI) lower than the 85th percentile were included. Around 1 million BP measurements from nearly 300,000 children were analyzed. This is a significantly larger sample size than the current database of normative BP values, which was compiled from roughly 60,000 to 70,000 measurements taken from around 50,000 pediatric patients between the 1970s and 1990s.

The Impact of Real-World Data

The EHR database is larger and more diverse than the current one, including patient records from across the country, according to Mitsnefes. It includes both auscultation and oscillometric BP measurements, aligning with modern practices and the guidance to average two BP measurements.

The EHR database also addresses real-life measurement circumstances. Mitsnefes notes that existing normative values are from controlled research studies.

“When we do research studies, we’re particular about everything, from choosing the right equipment and blood pressure cuff, how long we wait before taking BP measurement, making sure the patient is sitting in right position, etc.,” he says. “But in the real world, for example, the child may be talking or not sitting still, and those things can affect blood pressure."

“Limitations of the historical data are why we decided to ask whether we could create normative values from real-world blood pressure measurements taken from pediatric patients during regular visits to their provider,” says Mitsnefes.

The Clinical Significance of Real-World BP Measurements

“In addition to concluding that real-world BP measurements could be used effectively to create normative values in children, the study highlighted other important implications, including higher cutoff points for normative values using EHR.

We found that in our study, generated cutoff points to define hypertension were a few points (2-4 mmHg) higher than current normative values, especially in children between the ages of 3 and 7,” Mitsnefes says. “That means we’re probably overestimating the real prevalence of high blood pressure in that age group right now.”

Adjusting the range could lead to fewer follow-up appointments, specialist referrals and treatments,” Mitsnefes says. “This, coupled with the ability to update the EHR database regularly, could optimize patient care and reduce costs.” Also, the researchers provided statistical data, from codes to full tables, so others could replicate the study at any time—or compare the two data sets to determine which best meets their specific needs. Still, the study has limitations.

“We don’t know how many of the blood pressure measurements were done by auscultation versus with a digital device,” Mitsnefes says. “But we know that this real-world data represents how we practice, and we see important implications for applying it to clinical care.”

This study was funded by the Preserving Kidney Function in Children with Chronic Kidney Disease (PRESERVE) grant.

(Published October 2024)

Read More from Nephrology and Hypertension