Most lab test results come with reference intervals—the range of numerical results that define “normal” from “high” or “low.” However, defining “normal” requires scientific evidence that demonstrates what healthy levels are for a range of chemicals in the body for different populations.

Establishing normal levels is especially challenging to accomplish for children. They’re not miniature adults, and their bodies are changing rapidly, sometimes wildly within a matter of days in the case of newborns. “I don’t know which is worse: not having a range or having an inaccurate range,” said Amy Pyle-Eilola, PhD, lab director of Nationwide Children’s Hospital.

She, along with Hubert Vesper, PhD, director of clinical standardization programs at the Centers for Disease Control and Prevention (CDC) and Dennis Dietzen, PhD, lab director of St. Louis Children’s Hospital, recently published a paper in JAMA Open Network showing for the first time the breadth of significant inconsistencies with pediatric reference intervals (JAMA Pediatr 2022; doi: 10.1001/jamapediatrics.2022.0794).

They also identified some of the most pressing issues in this area, including the lack of data and how small studies can be interpreted the wrong way, especially by clinicians who find reference intervals online without any context, don’t consult a lab director, and make care decisions based on those intervals. The study also paves the way for the medical community to develop more reliable pediatric reference intervals, for the betterment of children’s medical care.

This has been an especially pressing need since the start of the COVID-19 pandemic, which also has coincided with a steep rise in pediatric cases of Multisystem Inflammatory Syndrome in Children (MIS-C).

For these conditions, “children usually get dozens if not more laboratory tests, and we’re stuck with trying to reconcile them for a condition that still is fairly nebulous,” said Alan Schroeder, MD, associate chief for research in the division of pediatric hospital medicine at Lucile Packard Children’s Hospital Stanford, who wrote the accompanying editorial in the same issue of JAMA Open Network. “Contextualizing lab results is so important, and trying to understand what’s normal for a kid really matters.”

The Challenge of Accurate Pediatric Reference Intervals

Gathering enough data about every single stage of a child’s development is difficult, especially when children change so much, and so often. And even then, relying solely on numerical age doesn’t account for how children progress and develop on different timelines. “Just because you turn seven or eight years old, doesn’t mean your IGF goes up by ten points,” Pyle-Eilola said.

In their JAMA Open Network study, Pyle-Eilola's group found that reference intervals published in scientific journals for free thyroxine, thyrotropin, ferritin, cystatin C, estradiol, and testosterone were inconsistent, especially during developmental stages where children undergo rapid biochemical changes. Some pediatric reference intervals failed to account for the free thyroxine and thyrotropin surge in the first few days of life, which could lead to these newborns being diagnosed with thyroid disease.

Part of the problem is sheer lack of data. While sick children are often subject to a range of medical tests, parents aren’t often willing to have the same tests done on healthy children, especially babies, who by nature have very small volumes of blood. Accurate ranges can’t be created based on the test results of sick children alone; healthy children need to be part of the sample set too.

“You need a lot of blood from a lot of people. In most cases, new parents aren’t about to let their child undergo phlebotomy,” Dietzen said. “If they’re sick, it’s not a great thing to do, and if they’re well, it’s a worse thing to do. By and large getting samples to do these studies is a big deal.”

The pediatric intervals that do exist also aren’t standardized across instrumentation, he added. “If somebody does a big study on thyroid stimulating hormone on a Siemens platform, and I have a Roche platform, I cannot take that data and put it onto my Roche platform because I don’t know how well it agrees,” he said. “It’s all over the place.”

The second part of the problem is that, despite these inconsistencies, existing reference intervals are still being used. That can put children at risk for inappropriate or harmful clinical care. At the very least, it means subjecting a baby or child to unnecessary visits to specialists, which most likely require their parent or parents to undergo additional travel to and from appointments, take off work, or find childcare for other children. At worst, children are given the medication or treatment for a medical condition they don’t have.

“Abnormal tests lead to more abnormal tests and interventions, and it keeps going. In some cases it can be really harmful to patients,” Schroeder said.

It’s also easy for physicians to rely on these available pediatric reference intervals, even those who typically do their due diligence. Dietzen has found that clinicians who are what he calls “critical consumers of data” can still be unaware of the problem of interpreting data across platforms. “They typically don’t understand the equipment that’s present in the lab and how those reference intervals are generated,” he said. “The problem is they don’t know what they don’t know.”

Clinical laboratorians could “probably be better about making sure that people are aware” of the problem with these intervals, Schroeder said, and help out physicians who can be bombarded, even overwhelmed with data. At the time he wrote his opinion, Schroeder had a pediatric ICU patient who had undergone 65 individual laboratory tests by 6 a.m. Clinicians get “so much information and that’s just the lab on top of all of the patient’s vital signs,” he said.

That’s why it’s critical for standardize pediatric reference intervals, added Pyle-Eilola. “The problem is that clinicians don’t know to call and ask somebody.”

How To Create Accurate Pediatric Reference Intervals

The answer starts with data—and a lot of it. “The holy grail is a continuous, smooth distribution of data from birth all the way into adulthood, and that is not where we are at today,” said Dietzen. “We end up having to make bins of data, and those bins are limited in how well they can describe the dynamic change of a pediatric population.”

In addition to raising awareness with both clinical laboratorians and clinicians about the problem with existing pediatric reference intervals, AACC and CDC have partnered on a project to create better reference intervals.

The effort includes using data from the National Health and Nutritional Examination Survey, which is conducted by CDC, to generate continuous pediatric reference intervals and provide them to laboratories, clinicians, and researchers. The project will also facilitate the adoption of these intervals by clinical laboratories through assay standardization.

“There is a need to correctly describe the biochemistry of child development, as well as to identify strategies to develop accurate and consistent pediatric reference intervals for improved pediatric care,” the study authors wrote. “Continued communication and collaboration between clinicians and their laboratory colleagues ensures appropriate clinical test interpretation and patient assessment and remains essential to effective implementation of common [pediatric reference intervals], they added.

Canada has tackled this problem for more than a decade through the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) project. Since 2009, it has recruited more than 12,300 healthy children and adolescents, established pediatric reference intervals for more than 200 laboratory biomarkers, and created an online database and mobile application for free access to the CALIPER reference intervals.

Part of the success with CALIPER, said Pyle-Eilola, was in convincing parents to allow their healthy children to be part of it, which is why AACC working with CDC is critical and why more funding is needed to ensure a U.S.-specific set of standards that can be used across the country.

“We need a very large, nationwide reference interval study,” she said. “These are really big problems that no single laboratory or organization can deal with. We need a large-scale approach, with federal funding, to make it happen.”

“We just need Congress to love this idea to get it down the road,” Dietzen added.

Jen A. Miller is a freelance journalist who lives in Audubon, New Jersey. +Twitter: @byJenAMiller