Temporary relationship between sleep duration and depression and its impact on future risk of cardiovascular disease

Background Although sleep duration and depression were correlated, their temporal sequence and how the sequence influence on future risk of cardiovascular disease (CVD) remained undetermined. This study aimed to explore the temporal relationship between sleep duration and depression, and its association with future CVD risk. Methods: We included 10,629 middle-aged and elderly participants with repeated measurements of sleep duration and depressive symptoms (measured by Center for Epidemiological Studies-Depression scale [CESD]) at the first two visits from the China Health and Retirement Longitudinal Study. Cross-lagged analysis and mediation analysis were used to examine the temporal relationship between sleep duration and depression and its impact on future risk of CVD. Results: The adjusted cross-lagged path coefficient from baseline sleep duration to follow-up CES-D (beta(1) = -0.191; 95 % confidence interval [CI], -0.239 to -0.142) was significantly greatly than that from baseline CES-D to follow-up sleep duration (beta(2) = -0.031; 95 % CI, -0.031 to -0.024) (P-difference < 0.0001). Similarly, the path coefficient from baseline sleep duration to annual changes in CES-D was significantly greater than that from baseline CES-D to annual changes in sleep duration (beta(1) = -0.093 versus beta(2) = -0.015, P-difference < 0.0001). The path coefficient from baseline sleep duration to follow-up CES-D in CVD group was significantly greater than that in those without CVD (P-difference of beta(1) = 0.0378). Furthermore, 27.93 % of the total association of sleep duration with CVD was mediated by depression symptoms. Conclusions: The findings provide evidence that decrease in sleep duration probably precedes the increased in depressive symptoms, and depression partially mediated the pathway from sleep duration to incident CVD.