Surks MI and Hollowell JG (2007) Age-specific distribution of serum thyrotropin and antithyroid antibodies in the U.S. population: implications for the prevalence of subclinical hypothyroidism. J Clin Endocrinol Metab 92: 4575–4582

Surks and Hollowell analyzed whether the increase of subclinical hypothyroidism with advancing age is caused by an increased incidence of thyroid disease in older people or by an age-related shift in the TSH-distribution curve towards higher values, the latter implying a higher reference range in older people. The upper reference limit for TSH concentrations is currently 4.5 mIU/l.

Using National Health and Nutrition Examination Survey data, age-specific distribution of serum TSH and frequency-distribution curves for TSH within specific age groups were analyzed in individuals (age ≥12 years; n = 16,533) with no known thyroid abnormalities and not taking thyroid-related medications.

The percentage of individuals with TSH concentrations of 0.4–2.5 mIU/l progressively decreased with age, whereas the percentage of measurements >4.5 mIU/l progressively increased. Frequency-distribution curves showed a shift in the peak relative frequency towards higher TSH concentrations with advancing age. A progressive increase in the mean, median and 97.5 centile for TSH concentration occurred with age, and the 97.5 centile was about 3.6 mIU/l in people aged 20–39 years and 7.5 mIU/l in those over 70 years. Findings were qualitatively independent of the presence of antithyroid antibodies.

The prevalence of subclinical hypothyroidism might, therefore, currently be overestimated because a substantial proportion of individuals with TSH concentrations >4.5 mIU/l have values below their age-specific 97.5 centile. The increasing 97.5 centiles for TSH that occur with aging seem to represent, at least in part, changes in the age-specific population distribution of TSH.