Abstract
Testosterone is one possible biomarker for depression risk among men and women. Both high and low levels of testosterone have been associated with depression, at least among men. Testosterone may be associated only with specific symptoms of depression, which might help to explain inconsistencies in previous results. We examined the cross-sectional associations between total testosterone and the specific symptoms of depression using pooled data across three cycles of NHANES (2011-2012, 2013-2014, and 2015-2016). The sample included 4253 men and 5102 women. Testosterone was modelled as 1) a dichotomous (low testosterone cut-off <300 ng/dL for men and 15 ng/dL for women) and 2) a continuous variable using cubic splines. In men, very low testosterone was weakly associated with problems with appetite, whereas very high testosterone was associated with sleep problems and weakly associated with tiredness. There were no consistent symptom-specific associations among women. These findings provide only suggestive evidence for symptom-specific associations between testosterone and depression, mainly related to somatic complaints. Further data are needed to assess the reliability of these associations.