Few things plague pediatric laboratorians like reference intervals. That’s not to say that this challenge is that much easier in adult settings. After all, one still needs to come up with ranges for new assays, find normal patients (at the least to verify those new ranges), and figure out the source of ranges that have been used as long as anyone can remember.

But pediatrics has two additional significant challenges. First, more partitions need to be addressed because analytes can change dramatically during childhood. Second, there are fewer studies on normal children than adults, largely due to the difficulty in acquiring samples from healthy children.

Despite the challenges, laboratorians still must strive to generate accurate reference intervals—correct laboratory test interpretation hinges upon it. Without appropriate, pediatric-specific intervals, clinicians may misinterpret results, leading to additional testing and possibly delayed or inaccurate diagnoses and treatments.

Current Best Practices

There are a few tricks of the trade for producing reference intervals other than using hundreds of normal patients. The Clinical and Laboratory Standards Institute (CLSI) guideline EP28-A3c outlines the criteria for defining reference intervals and describes three methods for working them up. First is to establish ranges de novo, which requires 120 healthy donors per partition. This is a particularly daunting task in pediatrics. However, the CLSI guideline provides more reasonable alternatives: transference and verification (Table 1).

Sources of Pediatric Reference Intervals

Transference adapts another lab’s interval through method comparison. This requires 40 samples spanning the analytical measurement range. Therefore, transference does not require healthy donors exclusively, allowing a laboratory to use residual clinical samples. The laboratory performs a statistical comparison of the two methods using linear regression to establish a translating equation between the methods. The slope is a multiplier and the intercept is added to the upper and lower limits of the reference interval from the reference laboratory.

Alternatively, a lab can attempt to verify any interval used by another lab or found in the literature. This requires that the lab test samples from only 20 healthy donors, and if 18 or more of the 20 samples fall within the proposed reference interval, it can be considered verified for use. However, this method requires that already good reference intervals are available to verify (1).

A Winding Road on Research

An exciting development of the last several decades is the number of large-scale efforts to establish high-quality reference intervals for children. Multiple organizations across the world have undertaken projects using different techniques to develop pediatric reference intervals. According to the International Federation of Clinical Chemistry and Laboratory Medicine, pediatric reference intervals initiative projects have taken place in Australia, the United Kingdom, Germany, South Korea, Canada, Denmark, the United States, and Scandinavia.

The Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) is one of the largest pediatric reference interval projects. CALIPER has collected thousands of samples from healthy children in Canada across several ethnic groups and published dozens of studies. These studies began with establishing intervals for pediatric populations on Abbott Architect analyzers, but CALIPER has extended its analyses to other platforms via transference and verification studies. Recently, the group has gone back to establish ranges for other platforms with fresh samples, and has extended its studies to more complex assays, from immunoassay to mass spectrometry. CALIPER researchers have also investigated biological variability of many assays using multiple samples from the same individuals. All the published reference intervals are available on the CALIPER website or smartphone app (2).

Previously, AACC collaborated with the now discontinued National Children’s Study, which was intended to be a nationwide longitudinal study of children from birth to age 21 years. This study, which ended in 2014, collected a range of data, including biological and environmental samples. The investigators obtained samples on filter paper cards rather than collect liquid blood specimens. Through collaboration with AACC, the study shared 310 of these samples from newborns in the first week of life from across the country with researchers. Dried blood spots are well suited to liquid chromatography tandem mass spectrometry methods, and the working group analyzed the samples to generate reference intervals for steroid hormones and amino acids (3). Unfortunately, with the dissolution of the National Children’s Study, there is no opportunity for additional studies.

Gaps and Goals That Remain

Despite the plethora of studies, reference interval gaps remain. Notably, analyte classes such as hematology, coagulation, and blood gases do not have well-established intervals. Reference intervals for children younger than 1 month old are sparse and often generated from numerically limited patient datasets. And reference intervals for premature infants have virtually no studies in the literature.

Moreover, covariates such as body mass index and ethnicity likely affect certain analytes; however, this has remained largely unexplored. Additionally, ethnic distributions are unequally sampled in all of the large-scale studies due to the study populations available, leading many studies to reflect only one ethnicity. Finally, many reference intervals were established years ago using now discontinued assays on analyzers no longer available, and without harmonization or standardization. Even new, well-done reference interval studies only apply to the analyzer on which the samples were tested, unless additional transference studies are performed. Though there are ongoing initiatives to harmonize assays across multiple platforms, until those efforts are realized, reference intervals will not be harmonized, and must be reconsidered for every analyzer.

How AACC Is Leading Progress

AACC has initiated a legislative effort to develop better pediatric reference intervals by collecting samples from healthy children. AACC has met with members of Congress and recently held a briefing for congressional staff to further describe the issues. With congressional support and funding, AACC believes the nation can make progress on pediatric reference intervals.

The association has called on Congress to increase the Centers for Disease Control's (CDC's) Environmental Health Laboratory budget by $10 million for such an effort, as the agency already has the expertise to tackle such a project. For example, CDC’s National Health and Nutrition Examination Survey (NHANES) has been assessing the health status of populations across the U.S. since the 1960s, and routinely collects biological samples for laboratory tests and extensive demographic, socioeconomic, dietary, and health information. NHANES recruits participants from a variety of geographic locations, socioeconomic statuses, and ethnic groups. Data from NHANES studies is freely available for researchers, who also may access residual biological samples. As such, NHANES is an ideal source for collecting healthy pediatric samples from a sample set that is representative of the U.S. population. And in fact, studies already have tapped into NHANES data to generate reference intervals (4-6).

Building Momentum

At the 71st AACC Annual Scientific Meeting in August, the Pediatric and Maternal Fetal Medicine Division hosted a special session about the status of pediatric reference intervals that was well attended by pediatric laboratorians from around the world, including several with involvement in other large-scale reference intervals studies. The meeting discussed the means for carrying out such a project, as well as to identify solutions to possible pitfalls. Many issues were brought to the attention of these stakeholders and will be considered as the project moves forward.

Though much progress has been made in developing pediatric reference intervals, the laboratory medicine community is eager and ready to grow and scale these efforts and make a significant difference in the health of children. The new and ongoing efforts of large-scale studies to establish population-based reference intervals in children will be of great benefit for the use and interpretation of laboratory tests.

Amy L. Pyle-Eilola, PhD, DABCC, FAACC, is the director of clinical chemistry at Nationwide Children’s Hospital and assistant professor of clinical pathology at Ohio State University Wexner Medical Center in Columbus. +Email: amy.pyle-eilola@nationwidechildrens.org


  1. CLSI. Defining, establishing, and verifying reference intervals in the clinical laboratory; approved guideline—third edition. Wayne, PA: CLSI 2010.
  2. Tahmasebi H, Higgins V, Fung AWS, et al. Pediatric reference intervals for biochemical markers: Gaps and challenges, recent national initiatives and future perspectives. EJIFCC 2017;28:43-63.
  3. Dietzen DJ, Bennett MJ, Lo SF, et al. Dried blood spot reference intervals for steroids and amino acids in a neonatal cohort of the National Children’s Study. Clin Chem 2016;62:1658-67.
  4. Wener MH, Daum PR, McQuillan GM. The influence of age, sex, and race on the upper reference limit of serum C-reactive protein concentration. J Rheumatol 2000;27:2351-9.
  5. Fulgoni VL 3rd, Agarwal S, Kellogg MD, et al. Establishing pediatric and adult RBC reference intervals with NHANES data using piecewise regression. Am J Clin Pathol 2019;151:128-42.
  6. Cheng CK, Chan J, Cembrowski GS, et al. Complete blood count reference interval diagrams derived from NHANES III: Stratification by age, sex, and race. Lab Hematol 2004;10:42-53.