This short has been reviewed and updated. You can find a link to the author's latest short on the subject here.
A Reference Interval Harmonization (hRI) Working Group was formed in 2016 by the Canadian Society of Clinical Chemists (CSCC) to coordinate harmonization of pediatric and adult reference intervals across Canada. As part of this mandate, the working group initiated a national survey to collect data on current reference intervals used in clinical practice by Canadian laboratories (1). Participating laboratories were asked to report reference intervals for a select list of analytes and measure these analytes in a commutable reference sample. Reported reference intervals varied substantially, even between laboratories using instruments from the same manufacturer. For example, the upper reference limit for alkaline phosphatase (ALP) for an adolescent female varied 42% across laboratories, ranging from 23% to 52% between laboratories using instruments from the same manufacturer. The inter-laboratory variation in reference intervals was greater for pediatrics compared to adults. For example, some laboratories reported an adult creatinine reference interval for 2 year old males that did not even overlap with appropriate pediatric limits.
To determine if the observed variation in reference intervals could be explained by variation in laboratory test results across laboratories, commutable reference samples were also measured. It was found that the variation in reference intervals was much greater than the variation in laboratory test results. For example, compared to the 42% variation in reported ALP upper reference limits, ALP test results only varied 7%. Additionally, if all test results for one analytical platform were biased (e.g., consistently lower than other methods), this may explain the extensive variation in reference intervals if reported reference intervals for that specific analytical platform were also biased similarly. However, there was no relationship found between the bias of commutable sample results and reference intervals, suggesting that reference interval variation is not explained by true analytical differences.
Unexplained inter-laboratory variation in reference intervals has also been observed by other countries, including Australia and New Zealand (2) and is likely an international phenomenon. The variation in reference intervals is likely due to laboratories simply using the same reference intervals they always have, even though the original source may no longer be known. Furthermore, reference intervals are often inappropriately acquired (e.g. from out-of-date textbooks, manufacturer product inserts) and thus may not be appropriate for the target population (e.g. using adult reference intervals to interpret pediatric test results). Without a scientific justification (e.g. method or population differences) for laboratories to use different reference intervals, there is no reason for reference intervals to vary. Rather, a common reference interval should be used to ensure consistent laboratory test result interpretation.
Several initiatives have begun establishing harmonized reference intervals and recommending their use in clinical practice. These include initiatives in the Nordic countries, Australia and New Zealand, United Kingdom, and Japan. The recently established CSCC hRI Working Group is also working towards establishing evidence-based harmonized reference intervals to be used in laboratories across Canada. Reducing between-laboratory variation in reference intervals will improve test result interpretation and ultimately patient care.
References
- Adeli K, Higgins V, Seccombe D, Collier CP, Balion CM, Cembrowski G, et al. National Survey of Adult and Pediatric Reference Intervals in Clinical Laboratories across Canada: A Report of the CSCC Working Group on Reference Interval Harmonization. Clin Biochem. 2017 Jun 21;
- Jones GR, Koetsier SD. RCPAQAP First Combined Measurement and Reference Interval Survey. Clin Biochem Rev. 2014 Nov;35(4):243–50.>