Background: There is a lack of standardization of reference intervals (RIs) for ferritin across laboratories, particularly for postmenopausal women. Depending on the RI used, there can be more than a 4-fold difference in the upper limit of normal between laboratories, resulting in potential misinterpretation.
Methods: This retrospective study used a large dataset of blood test results from 25,425 healthy participants aged 18 to 97 over a 7-year period. Exclusion criteria were used to screen out individuals with conditions known to affect iron metabolism or raise ferritin as part of the acute phase response. Distributions were assessed using density and Q-Q plots, and age-banded cut-offs were determined. The non-parametric method was used to establish RIs for sex and age bands.
Results: For females, 4 age bands were established (18-39, 40-49, 50-59 and 60+). For males, 2 bands were identified (18-39 and 40+). Performance against a validation dataset, followed by an expansive validation against an inclusive dataset, demonstrated the robustness of the derived RIs.
Conclusion: This study addresses the inconsistency in serum ferritin RIs by presenting intervals based on demographic parameters. This approach can potentially enhance the accuracy of interpreting serum ferritin levels, assisting clinicians in identifying patients requiring further evaluation.
Keywords: Roche Cobas analyser; Serum ferritin; reference intervals.