A Robust Control Group Identification Approach for Reference Intervals of Thyroid Biomarkers: The Case of Jordan

Purpose: This study introduces a robust approach for establishing reference intervals (RIs) for thyroid biomarkers in the Jordanian population by incorporating symptoms-based criteria into the control group selection. This method enhances the traditional inclusion-exclusion criteria by adding a symptom layer. Five thyroid biomarkers were analyzed: thyroid-stimulating hormone (TSH), free triiodothyronine (FT3), free thyroxine (FT4), thyroid receptor antibodies (TRAbs), and antithyroid peroxidase antibodies (TPOAb). The study also investigated the prevalence of thyroid disorders using the established population-specific RIs. Patients and Methods: Samples from 6782 participants across different regions were collected between June 2016 and May 2017. The Kruskal-Wallis and the Mann-Whitney tests identified the best partitioning based on age-gender significant differences, and 2.5th-97.5th percentiles were used to establish the RIs. Results: The TSH, FT4, and FT3 differed significantly among males and females, while no significant differences were found among different age groups. The RIs for males were as follows: TSH, 0.50-4.18 mIU/L; FT4, 12.00-21.16 pmol/L; and FT3, 2.28-4.17 pmol/L, while the corresponding RIs for females were as follows: TSH, 0.57-4.61 mIU/L; FT4, 12.02-24.69 pmol/L; and FT3, 2.53-3.71 pmol/L. TRAbs and TPOAb RIs were 0.3-1.89 IU/L and 0.83-23.87 IU/mL, respectively. The prevalence of overt hypothyroidism was 10.63%, subclinical hypothyroidism was 3.53%, overt hyperthyroidism was 0.66%, and subclinical hyperthyroidism was 1.52%. Conclusions: This study emphasizes the role of population-specific RIs in minimizing diagnostic errors. The addition of symptom criteria improved control group identification and RI accuracy. These findings improve the diagnosis of thyroid disorders in Jordan and suggest a reproducible framework for worldwide application.