TSH reference intervals in adolescents in the iodine deficiency region. Comparison of direct and indirect approaches for estimating reference limits

Background. Due to the difficulties of recruiting reference groups for direct determination of reference intervals, methods for reference intervals calculating using laboratory database data have been proposed in the literature. At the moment, the properties and limitations of existing methods are not sufficiently described; for example, the issue of the influence of data from patients with established pathology, the results of which are within the RI limits, remains unresolved, including for skewed distributions of analytes. The aim of the study was to evaluate the possibilities of indirect methods for determining RI in adolescents for TSH in a population including the results from individuals with established thyroid pathology. Materials and methods. The RI limits were calculated with indirect method using data from the entire database (all, n=1179, boys, n=542, girls, n=637), most of whose TSH pathological values were within the RI limits: TMC, RefineR, Kosmic, Bhattacharaya Results. The calculated limits by RefineR and TMC of the database TSH were verified. Results of calculating the simulated data were successful to varying degrees with different percentages and the distance of pathological data from the middle of the distribution of conditionally healthy individuals. Conclusions. Results with conditionally pathological values that are within the RI limits can significantly affect the quality of RI limits calculations with indirect methods. The best way to calculate RI for TSH should be considered the TMC method as the one that best meets the criteria when calculating data from the database and when conducting simulation.