AUTHOR=Marrington Rachel , Anderson Michelle , Roch Martin , Davies Sarah , Forster Rosie , Williams George , d’Oyen-Fitchett Johnathan , MacKenzie Finlay 
  
TITLE=Use of EQA for assessing variation in post-analytical interpretation of results using a clinical case of ?adrenal insufficiency taking into consideration variation in analytical performance of cortisol
  
JOURNAL=British Journal of Biomedical Science
  
VOLUME=Volume 83 - 2026
  
YEAR=2026
  
URL=https://www.frontierspartnerships.org/journals/british-journal-of-biomedical-science/articles/10.3389/bjbs.2026.16559
  
DOI=10.3389/bjbs.2026.16559
  
ISSN=2474-0896
  
ABSTRACT=IntroductionLaboratory medicine is evolving at a rapid rate, not only in terms of the repertoire of tests available and automation of the laboratory and procedures, but also in how laboratory data is used within “calculated tests,” algorithms and guidelines. The goal is diagnostic and treatment improvements for the patient; however, there are real patient safety concerns if the limitations of what is often described as the “simple blood test” are not sufficiently understood and appreciated. The aim of this study was to investigate how laboratories take into consideration variation in cortisol assays when managing potential cases of adrenal insufficiency when they had a single cortisol measurement. For English laboratories also assessed compliance to NICE Guidance NG243 — “Adrenal insufficiency: identification and management,” published in 2024.Materials and methodsThe UK NEQAS for Steroid Hormones Scheme provides an External Quality Assessment (EQA) service for Cortisol. In December 2025, three serum specimens were distributed for cortisol analysis and interpretation based on a provided scenario — each specimen taken from a 25-year-old female at 08:30, clinical details — ?Adrenal insufficiency. Participants were required to interpret results into three categories — (1) Person may have adrenal insufficiency, (2) Probability of adrenal insufficiency uncertain or (3) Adrenal insufficiency very unlikely. Data has been evaluated based on the reported cortisol result and corresponding interpretation, broken down by manufacturer and location of laboratory.ResultsCortisol results were returned by ∼370 Participants with ∼300 interpretating their result. Specimen 536A had a cortisol of 122 nmol/L (SD 9 nmol/L, %CV 7.5%), 536B — 307 nmol/L (SD 24 nmol/L, %CV 7.8%) and 536C — 260 nmol/L (SD 19 nmol/L, %CV 7.2%) based on a mass spectrometry method mean (validated by a candidate reference method). According to NG243, these cortisol results should correspond to Codes (1), (3) and (2) respectively. As expected, there was variation in the cortisol result reported and the subsequent interpretation.DiscussionThis study utilises EQA to look at the variation in cortisol measurement specifically in the context of identifying adrenal insufficiency. By linking post-analytical interpretation to the analytical measurement we have identified variation in both cortisol assays and interpretation of results.