Background/Objectives: Reference intervals (RIs) for thyroid-stimulating hormone (TSH), free thyroxine (FT4), and free triiodothyronine (FT3) are essential for accurate interpretation of thyroid function tests. Given the pronounced age- and sex-related biological variability of these analytes, locally derived and population-specific RIs are required. Methods: This multicenter study estimated age- and sex-specific RIs for TSH, FT4, and FT3 using an indirect statistical approach applied to routine laboratory information system data from three hospitals in Ankara, Türkiye. Measurements were performed on Maglumi X8 and X6 analyzers (Snibe Diagnostic, Shenzhen, China) using chemiluminescent immunoassays. Reference intervals were calculated using the refineR algorithm implemented in the R programming language, with Box–Cox or shifted two-parameter (mod) Box–Cox transformations selected according to data distribution. Results: More than 100,000 test results were included after predefined exclusions. In adults (≥18 years), the estimated TSH reference intervals were 0.495–5.83 mIU/L in males and 0.485–6.40 mIU/L in females, with higher upper limits observed in females. Adult FT4 reference intervals were approximately 12.9–21.5 pmol/L, and FT3 intervals were 3.46–6.04 pmol/L in males and 3.46–5.76 pmol/L in females. Pediatric reference intervals for TSH and FT4 were also derived across predefined age categories, demonstrating age-dependent variation. Conclusions: This study establishes age- and sex-stratified reference intervals for TSH, FT4, and FT3 using large multicenter routine laboratory datasets, encompassing both adult and pediatric populations. The findings highlight limitations of manufacturer-provided intervals and support the application of indirect statistical approaches, such as refineR, for local reference interval determination in clinical laboratories.
Koca et al. (Thu,) studied this question.