Depressive symptom severity mediates the association between sleep disturbance and obesity in US adults: Results from the NHANES

BACKGROUND: The clustering of sleep alterations, cardiometabolic risk, and depressive symptoms suggests a convergence in their pathophysiology. We quantify the role of depressive symptoms in mediating the association between empirically derived sleep indices and body mass index (BMI). METHODS: Data were derived from 8,844 adult participants of the 2005 to 2008 US National Health and Nutrition Examination Survey. Factor analysis of the Sleep Disorders Questionnaire was conducted. Ordinary least squares path analysis quantified the effects of sleep indices on BMI directly and indirectly via depressive symptom severity (ie, Patient Health Questionnaire). RESULTS: Three sleep indices were extracted: poor sleep-related functional impairment, sleep disturbance, and daytime sleepiness. The associations between functional impairment, sleep disturbance, and daytime sleepiness and BMI were mediated by the effects of sleep on depressive symptoms (kappa(2) = 0.02) after adjustment for covariates. Daytime sleepiness was associated with BMI independent of depressive symptoms, whereas poor sleep-related functional impairment and sleep disturbance were not. CONCLUSIONS: Higher subjective ratings of sleep-related functional impairment, sleep disturbance, or daytime sleepiness indirectly increased BMI by worsening overall depressive symptom severity. A testable hypothesis is whether preemptive targeting of depressive symptoms in populations with sleep disturbances may decrease risk for obesity and other concurrent metabolic comorbidities.