Background: The rise in serum thyroid stimulating hormone (TSH) in healthy older adults is well established. However, age-related reference intervals are not widely used. We aimed to establish all-adult and age-related reference intervals for TSH, free thyroxine (FT4) and free triiodothyronine (FT3) and determine their impact on primary care thyroid testing results. Methods: We measured TSH, FT4 and FT3 by Roche Cobas assays in a reference cohort of 1364 anti-thyroid peroxidase antibody negative adults with no self-reported medical conditions or medications. Reference intervals were generated for all-adult, 18-60 and over 60-years. Reference intervals were applied to 21,286 primary care tests with no known thyroid disease to assess effect on test interpretation. Results: In the reference cohort, 23.2% were over 60 years compared to 50.4% of those undergoing thyroid testing in primary care. With an all-adult reference interval, 8.2% of over 60s had an elevated TSH, 7.0% were classified as subclinical hypothyroid and 0.8% as overt hypothyroid. With age-related reference intervals, this fell to 4.4% with an elevated TSH, 3.7% subclinical hypothyroid and 0.4% overt hypothyroid. Minimal changes were seen in the 18-60 years group. Conclusions: An all-adult reference interval derived from healthy and therefore younger individuals is less appropriate for the older subset of the population being tested. Application of an age-related reference interval for over 60s would reduce the proportion of patients with abnormal thyroid test results. In turn, this would decrease potentially unnecessary cascade and repeat testing as well as regular follow-up in primary care.
Allen et al. (2026) studied this question.