Background: Patients with depression often suffer from circadian rhythm disruptions, such as insomnia and a lower amplitude of daily rhythms of locomotor activity. However, whether circadian disorders precede depression and act as a predictor for subsequent depression have not been unanimously approved. In light of increasing evidence of an association between disrupted circadian rhythms and affective symptoms, a meta-analysis was conducted to examine the effect of circadian disruptions on incident depression. Methods: A systematic search was performed in PubMed, Embase, Web of Science, PsycINFO, PsycArticles, CINAHL database. The search results were restricted to be published in English up to April 10, 2020.Reference lists from included studies and relevant reviews were also screened. We included longitudinal cohort studies that examined the predictive role of all various types of circadian disruption in incident depression. Two reviewers independently extracted information on study design, participants’ characteristics and diagnostic criteria for circadian disruption and depression. Study quality was assessed using the Newcastle-Ottawa Scale. Random-effect models were applied to all meta-analyses given potential heterogeneity. For insomnia, subgroup analyses were performed based on subject’s characteristics, insomnia symptom domains and numbers and the investigated methods. Pool-effect estimates of relative risks to develop incident depression in groups with circadian disruption compared to groups without. This study is registered with PROSPERO, CRD42020179135. Findings: A total of 49 studies were eligible for this meta-analysis. Of 49 eligible studies, included types of circadian disruption were sleep disturbances (n = 45), chronotype (n = 2), cortisol awakening response (n = 1) and rest - activity rhythm (n = 1). Insomnia, hypersomnia, excessive daytime sleepiness (EDS), short sleep duration, obstructive sleep apnea (OSA), restless legs syndrome (RLS), an evening preference, heightened levels of cortisol in awakening response and low robustness of rest - activity rhythm all acted as significant predictors of occurrence of depression. Among insomnia subtypes, difficulty initiating sleep (DIS) and difficulty maintaining sleep (DMS) predicted future depression. Of note, persistent insomnia (OR: 5.02) contributed to more than two - fold risk of incident depression compared to insomnia (OR: 2.34). Moreover, insomnia symptom numbers were associated with incident depression in a gradient trend. Interpretation: Different types of circadian disorders act as risk factors of onset of depression, which clinicians should be informed of. Future research should be aimed to determine the casual relationship between other types of circadian disorders and onset of depression, and moreover to further elucidate the underlying mechanisms. Funding Statement: This work was supported in part by the National Natural Science Foundation of China (no. 81871071 and 81171251) and Beijing Municipal Natural Science Foundation (no. 7162101). Declaration of Interests: All of the authors declare no conflict of interest.