Harmonization protocols for thyroid stimulating hormone (TSH) immunoassays: different approaches based on the consensus mean value

The lack of interchangeable laboratory results and consensus in current practices has underpinned greater attention to standardization and harmonization projects. In the area of method standardization and harmonization, there is considerable debate about how best to achieve comparability of measurement for immunoassays, and in particular heterogeneous proteins. The term standardization should be used only when comparable results among measurement procedures are based on calibration traceability to the International System of Units (SI unit) using a reference measurement procedure (RMP). Recently, it has been promoted the harmonization of methods for many immunoassays, and in particular for thyreotropin (TSH), as accepted RMPs are not available. In a recent paper published in this journal, a group of well-recognized authors used a complex statistical approach in order to reduce variability between the results observed with the 14 TSH immunoassay methods tested in their study. Here we provide data demonstrating that data from an external quality assessment (EQA) study allow similar results to those obtained using the reported statistical approach.