Objective: Poor sleep quality and mental health problems are common in college students. The objective of this study is to examine whether sleep quality predicts the risk of future mental health problems, and vice versa.

Methods: The sleep quality and mental health status of 686 male college students were estimated, and 582 of them were followed up a year later. Subjective sleep quality and mental health problems were measured with the Pittsburgh Sleep Quality Index (PSQI) and the Depression Anxiety Stress Scale-21 (DASS-21), respectively.

Results: Either at baseline or during follow-up, the PSQI global score was positively associated with scores for depression, anxiety, and stress on the DASS-21 (p's < 0.001). Longitudinal analyses revealed that DASS-21 total score increased in line with increased of PSQI global score during the year (p < .001). More importantly, the cross-lagged analysis showed that (i) PSQI global score at baseline was positively related to depression (β = 0.261), anxiety (β = 0.321), and stress (β = 0.311) scores a year later (p's < 0.001) and (ii) depression (β = 0.259), stress (β = 0.245) and anxiety (β = 0.292) scores at baseline were related to PSQI global score a year later (p's < 0.001). Finally, we further found that among those without mental health problems at baseline, poorer baseline sleep quality predicted a higher risk of anxiety symptoms a year later (RR 3.07, 95% CI 1.36-6.97, p = .007).

Conclusions: These data may suggest a bidirectionally relationship between sleep quality and mental health problems.

Keywords: College students; DASS-21; Longitudinal observational study; Mental health problems; PSQI; Sleep quality.

Copyright © 2020 Elsevier Inc. All rights reserved.