A method of assessing serum triiodothyronine concentrations that is independent of subject's age and variations in concentrations of binding proteins in serum.

Pairs of values for serum triiodothyronine and serum triiodothyronine binding capacity were represented as maps of points for euthyroid populations (including subjects who were pregnant or were receiving estrogens) in the age groups 18-50, 51-64, 65-79, and greater than 80 years (504 subjects) and thyrotoxic populations in the same age groups (207 subjects). For each age group the two populations were effectively separated bythe technique. Statistical analysis by the method of principal components justified the combination of all the age-related maps into a single map, allowing the assessment of serum triiodothyronine regardless of age or concentration of binding proteins in serum. The overlap of the combined populations was less than 1%. The method proposed is superior to a calculation of free triiodothyronine index in several respects.