Impaired sleep is associated with low testosterone in US adult males: results from the National Health and Nutrition Examination Survey

Premal Patel, Benjamin Shiff, Taylor P. Kohn, Ranjith Ramasamy

Testosterone deficiency has been linked to several adverse health outcomes and recent data have suggested that abnormal sleep quality may result in lower testosterone levels. We assessed the effect of self-reported sleep patterns on serum testosterone while controlling for co-morbidities, and baseline demographics. Using data collected from the 2011-2012 National Health and Nutrition Examination Survey (NHANES), we extracted serum total testosterone level, sleep duration, demographic, and co-morbidities for men aged 16 years and older. Univariate and multivariate linear regression was used to estimate the association of number of hours slept, co-morbidities, and demographics with serum testosterone. Among the 9756 individuals in the NHANES dataset, 2295 (23.5%) were males 16 years and older with a median (interquartile range) age of 46 years (29-62) who also had serum testosterone levels drawn. Median serum testosterone level was 377 ng/dL (IQR: 279-492 ng/dL). Median number of hours slept was 7 h (IQR: 6-8 h). On multivariate linear regression, we found serum testosterone decreased by 0.49 ng/dL per year of age (p = 0.04), 5.85 ng/dL per hour loss of sleep (p < 0.01) and 6.18 ng/dL per unit of body mass index (BMI) increase (p < 0.01). Among men aged 16-80 in the United States, we found increasing age, impaired sleep and elevated BMI is associated with low testosterone. It is important, therefore, that evaluation and treatment of reduced serum testosterone should also include improving sleep duration in combination with weight management.