Association between thyroid function and objective and subjective sleep quality in older men: The osteoporotic fractures in men (MrOS) study

Haruko Akatsu, Susan K. Ewing, Marcia L. Stefanick, Howard A. Fink, Katie L. Stone, Elizabeth Barrett-Connor, Reena Mehra, Sonia Ancoli-Israel, Susan Redline, Andrew R. Hoffman

Research output: Contribution to journalArticlepeer-review

11 Scopus citations


Objective: To determine the association between thyroid hormone levels and sleep quality in community-dwelling men.Methods: Among 5,994 men aged ≥65 years in the Osteoporotic Fractures in Men (MrOS) study, 682 had baseline thyroid function data, normal free thyroxine (FT4) (0.70 FT 4 1.85 ng/dL), actigraphy measurements, and were not using thyroid-related medications. Three categories of thyroid function were defined: subclinical hyperthyroid (thyroid-stimulating hormone [TSH] <0.55 mIU/L), euthyroid (TSH, 0.55 to 4.78 mIU/L), and subclinical hypothyroid (TSH >4.78 mIU/L). Objective (total hours of nighttime sleep [TST], sleep efficiency [SE], wake after sleep onset [WASO], sleep latency [SL], number of long wake episodes [LWEP]) and subjective (TST, Pittsburgh Sleep Quality Index score, Epworth Sleepiness Scale score) sleep quality parameters were measured. The association between TSH and sleep quality was examined using linear regression (continuous sleep outcomes) and log-binomial regression (categorical sleep outcomes).Results: Among the 682 men examined, 15 had subclinical hyperthyroidism and 38 had subclinical hypothyroidism. There was no difference in sleep quality between subclinical hypothyroid and euthyroid men. Compared to euthyroid men, subclinical hyperthyroid men had lower mean actigraphy TST (adjusted mean difference [95% confidence interval (CI)], -27.4 [-63.7 to 8.9] minutes), lower mean SE (-4.5% [-10.3% to 1.3%]), and higher mean WASO (13.5 [-8.0 to 35.0] minutes]), whereas 41% had increased risk of actigraphy-measured TST <6 hours (relative risk [RR], 1.41; 95% CI, 0.83 to 2.39), and 83% had increased risk of SL ≥60 minutes (RR, 1.83; 95% CI, 0.65 to 5.14) (all P>.05).Conclusion: Neither subclinical hypothyroidism nor hyperthyroidism is significantly associated with decreased sleep quality.

Original languageEnglish (US)
Pages (from-to)576-586
Number of pages11
JournalEndocrine Practice
Issue number6
StatePublished - Jun 1 2014


Dive into the research topics of 'Association between thyroid function and objective and subjective sleep quality in older men: The osteoporotic fractures in men (MrOS) study'. Together they form a unique fingerprint.

Cite this