Longevity, demographic characteristics, and socio-economic status are linked to triiodothyronine levels in the general population

Abstract

The hypothalamic-pituitary-thyroid (HPT) axis is fundamental to human biology, exerting central control over energy expenditure and body temperature. However, the consequences of normal physiologic HPT-axis variation in populations without diagnosed thyroid disease are poorly understood. Using nationally representative data from the 2007 to 2012 National Health and Nutrition Examination Survey, we explore relationships with demographic characteristics, longevity, and socio-economic factors. We find much larger variation across age in free T3 than other HPT-axis hormones. T3 and T4 have opposite relationships to mortality: free T3 is inversely related and free T4 is positively related to the likelihood of death. Free T3 and household income are negatively related, particularly at lower incomes. Finally, free T3 among older adults is associated with labor both in terms of unemployment and hours worked. Physiologic TSH/T4 explain only 1.7% of T3 variation, and neither are appreciably correlated to socio-economic outcomes. Taken together, our data suggest an unappreciated complexity of the HPT-axis signaling cascade broadly such that TSH and T4 may not be accurate surrogates of free T3. Furthermore, we find that subclinical variation in the HPT-axis effector hormone T3 is an important and overlooked factor linking socio-economic forces, human biology, and aging.

Keywords: T3; health demographics; mortality; socioeconomic outcomes; thyroid.

Fig. 1.

Inter-relationships between TSH, free T4, and free T3 in SDs. (A) Scatter plot showing TSH and free T3 for the same individual, in terms of SDs of the adult population distribution. (B) Scatter plot showing free T4 and free T3 for the same individual, in terms of SDs of the adult population distribution. Euthyroid, hypo-, and hyper- thyroid individuals in both panels classified by TSH (hypo: TSH > 4.1 mIU/L, hyper: TSH < 0.4 mIU/L). Nonparametric LOWESS estimates shown stratifying by summer measurement (May 1 to October 31). SI Appendix, Figs. S1 and S2 display parallel results in nonstandardized units, and stratifying overt and subclinical hypothyroidism.

Fig. 2.

Thyroid hormones, age, and longevity. (A) Weighted nonparametric LOWESS estimates of the relationships between thyroid hormones and age among adults, stratified by sex. (B) Weighted survival curves since measurement date for each quartile of TSH, free T4, and free T3, among adults over age 50 (Q1: 0 to 25%, Q2: 25 to 50%, Q3: 50 to 75%, Q4: 75 to 100% within each hormone’s distribution).

Fig. 3.

Thyroid hormones, household income, and unemployment. Panels A, B, and C show relationships for TSH, free T4, and free T3, respectively. Panels A, B, and C show relationships for TSH, free T4, and free T3, respectively. Weighted nonparametric LOWESS estimates of the relationship between thyroid hormones and the natural logarithm of real (2007 dollars) household income per capita are displayed (within the central 5th to 95th percentiles of the distribution), stratified by sex and employment status at the time of measurement. Data from 2007 to 2012 NHANES.