Effect of sleep, sleep apnoea and insomnia symptoms on depressive symptoms in combined harmonised community-based cohorts: a cross-sectional analysis

Sleep and sleep disturbances, such as sleep apnoea and insomnia, are linked to mental health, particularly, depressive symptoms. It remains unclear however, if sleep fragmentation or hypoxia is primarily linked to depressive symptoms and how insomnia symptoms affect this relationship in samples free of clinical referral bias. We analysed cross-sectional data from two harmonised, population-based cohorts: the Sleep and HEalth in women (SHE) study and the Men Androgen Inflammation Lifestyle Environment and Stress (MAILES) study (n total  = 1238; 68% males, 57 ± 12 years, BMI 28.0 ± 4.2 kg/m 2). Objective measures of OSA (apnoea-hypopnoea index; AHI, oxygen desaturation index; ODI) and sleep fragmentation variables (sleep efficiency; SE% and wake after sleep onset; WASO) were measured, along with subjective data on insomnia symptoms and depressive symptoms. Logistic regression models were used to assess associations while adjusting for age, sex, BMI, alcohol and physical activity. Pathway analyses further explored potential mediating relationships. Depressive symptoms were present in 17% of the sample. We did not find a link between intermittent hypoxia variables (AHI, ODI) and depressive symptoms. In contrast, reduced SE% (OR 0.84; 0.72-0.98) and increased WASO (1.26; 1.06-1.51) were significantly related to depressive symptoms. When including adjustment for insomnia in the model, these associations were no longer significant. While sleep fragmentation was associated with depressive symptoms, insomnia symptoms appear to be a key factor. Assessing insomnia in depression and vice versa and addressing these conditions, may be useful for reducing symptom burden.