A systematic review and dose-response meta-analysis of sleep duration and the occurrence of cognitive disorders.

PURPOSE: Previous researches have shown inconsistent findings on the relationship between sleep duration and cognitive disorders (including cognitive decline, dementia, Alzheimer's disease, and cognitive impairment). Furthermore, the possible dose-response association of sleep duration and cognitive disorders is still not clear.

METHODS: Two databases (PubMed and Embase) were searched for relevant studies from inception to May 4, 2016. We used the generic inverse-variance method to combine the outcomes with a random-effects model for the association between sleep duration (the shortest or longest vs. middle category) and cognitive disorders. Additionally, we used the generalized least-squares trend estimation model to estimate the potential dose-response association.

RESULTS: Finally, nine eligible cohort studies involving 22,187 participants were included in the present systematic review and meta-analysis. Compared with the middle category, both short and long sleep duration were accompanied by a significantly higher incident risk of cognitive disorders, and the relative risks (RRs) and 95% confidence intervals (CIs) were 1.34 (1.15, 1.56) and 1.21 (1.06, 1.39), respectively. Moreover, an approximately "U-shaped" dose-response relationship was observed. Stratified analyses showed that the association between short sleep duration and cognitive disorders was more robust in the subgroup of the measurement of sleep duration based on self-report.

CONCLUSIONS: The lowest incident risk of cognitive disorders was found at the sleep duration of 7-8 h per day. The present study includes a small number of studies, and the study participants mostly consist of Caucasian population. In the future, researches are also warranted to confirm the association in different groups of people.