For Sleep plays a critical role in health and well-being of a human. It is reported that various populations are experiencing sleep disturbance frequently and suffering from its consequences. In Chinese college students, the prevalence of sleep disturbances and insomnia are reported to be 25.7% (1) and 18.5% (2), respectively.

Depression in youth has been shown to be associated with distinct sleep dimensions, such as timing, duration, and quality (3). Shochat et al. reported that those students with fewest depressive symptoms had moderate sleep time, shorter sleep onset latencies, and fewer arousals (4). In addition, depression can exacerbate the severity of insomnia in college students (5). Depression has been shown to be the most important risk factor for insomnia (6, 7). However, it is unclear if a depressive mood can lead to insomnia directly or indirectly through some mediating factors.

There may be some mediating factors between depressive symptoms and insomnia (8), such as lying awake at night (9), cognitive inflexibility (10), worry (11), social support (12), and internet addiction (13). Rumination is considered a possible mediator in relation between the depression and insomnia. It was well-known that individuals who have chronic insomnia worry about a range of topics while in bed, including “What about tomorrow's work?” (14). Rumination tends to engage in preservative and non-constructive thoughts and negative reflection on the problems and feelings in the past or present. Although previous studies have focused almost exclusively on the role of anxiety in the development of insomnia, it appears that rumination is more critical for eliciting sleep difficulties (15). Rumination about adverse events and self-reflection have been shown to heighten the levels of physiologic arousal (16), lengthen the sleep latency (17), and decrease the sleep efficiency (17). A clinical study showed that a high level of depressive symptoms significantly predicted a higher level of rumination (18). However, only a few studies have explored the mediating effect of rumination on the association between depressive symptoms and insomnia (19, 20). It has been shown that ruminative thoughts mediated the relationship between depressive symptoms and insomnia in the population of healthy young adults (20) and adolescents (19).

It has been reported that the sleep quality of students from high altitude areas is worse than that of students from plain areas, and the sleep quality of the ethnic minority students is worse than that of the Han students (21, 22). Qinghai province is located in the Qinghai-Tibet Plateau, with an average altitude of more than 3,000 meters. Many ethnic minority groups live in Qinghai province. The population of the Tibetan, Hui, Tu, Mongolian and Salar ethic groups accounts for 47.7% of the total population in Qinghai province (23). There is a lack of research on the relationship between insomnia and depressive symptoms in the population in this multi-ethnic area in Qinghai-Tibet Plateau. In this study we investigated the association between insomnia and depressive symptoms, and determined the mediating effect of rumination on the association between depressive symptoms and insomnia in college students in Qinghai province.

Our study showed that rumination mediated in the association between depressive symptoms and insomnia in college students in Qinghai province, a multi-ethnic and high-altitude area. However, the mediating impact is small.

In our study, 38.6% of college students reported that they suffered from insomnia in the past 2 weeks. Luo et al. demonstrated that the prevalence of insomnia among Chinese teenager in Guangzhou (average altitude: 21 meters) was 28.9% (35), which was lower than what we found. One reason may explain the difference is that nearly half of the participants in our study were ethnic minorities and lived in area with an average altitude of more than 3,000 meters. It has been shown that people who live at high altitudes have poorer sleep quality, which may be due to hypoxia-induced arousal and hypoxia-induced periodic breathing (35–37). Yip and Cheon reported that the prevalence of insomnia was high in ethnic minorities, which may be related to acculturation (38).

The results of the structural equation model display that depressive symptoms directly affect insomnia, and the direct effect accounted for 92.4% of the total effect. The results are consistent with previous studies that linking depressive symptoms to insomnia (8, 39). The youth with depressive symptoms demonstrate increased activity in extended medial network regions (40). Studies have shown that people with depressive symptoms have increased cortisol levels (41), hypothalamic pituitary adrenal (HPA) axis dysregulation (41), changes in inflammatory cytokines (42), and changes in sleep architecture (43). These changes can lead to insomnia. Beck's cognitive model of depression describes how people's thoughts and perceptions influence their emotional, behavioral, and physiological reactions. The elements of this model include biased attention, biased processing, biased thoughts and rumination, biased memory, and dysfunctional attitudes and schemas (44). These psychological processes may extend to the pre-sleep period, resulting in unpleasant intrusive thoughts, dysfunctional beliefs and attitudes about sleep, selective attention and monitoring to sleep-related threat, and misperception of sleep deficit. They may cause difficulty in initiating sleep and delayed sleep phase (45). Depressive symptoms are characterized by social withdrawal (46), poor interpersonal skills (47) and difficulty in coping with peer/family stressors (48). Such vigilance following social stressors may induce wakefulness before asleep.

In our study the structural equation model results display that rumination had mediating effects on the association between depressive symptoms and insomnia. Three possible mechanisms may explain our results. Firstly, depressive symptoms are considered the failure of emotion regulation. People with depressive symptoms may elaborate negative information. The reactivation of these memories during the pre-sleep period (i.e., ruminating) may exacerbate the difficulty in falling asleep (49). Secondly, ruminative thinking is considered invasive thinking. Individuals with excess rumination tend to focus on negative emotions continuously and repeatedly. This may further increase their selective attention to adverse events (50) and stimulate more cognitive awakening, leading to the delay of sleep initiation (8). Thirdly, rumination can cause psychological arousal and autonomic excitation, such as elevated heart rate, elevated body temperature, increased basal metabolic rate and electrodermal activity (51). Such changes in the body may cause insomnia.

Our results are consistent with previous studies (19, 20, 52) showing that rumination is a mediating factor between depressive symptoms and insomnia. However, there are some differences between our results and others. Firstly, the mediating effect by rumination in our study is smaller than that in the previous studies. Previous studies have reported that rumination fully mediated the relationship between depression and insomnia (19, 20). However, we found that rumination only plays a relatively limited role on the association between depressive symptom and insomnia. Secondly, previous studies included some other mediators, such as perfectionism (19), neuroticism (52), and self-reported health (20). Incorporating other factors into the model may change the indirect effects. Thirdly, about half of the participants in our study were from multiple ethnic minority groups. It has been shown that ruminative thinking from different cultural backgrounds is bound to have differences in content, form and function (53). For example, the everyday worship and annual Ramadan of the Hui Muslim group, including meditation and social networking, may promote psycho-physical well-being (54).

The present study has several strengths. Firstly, we applied a complex structural equation model to perform path analysis and quantify these paths. Secondly, the large sample size is a highlight of our study. Thirdly, the present study is the first sleep study conducted in the population of Qinghai-Tibet Plateau. It is also one of the few large-scale surveys conducted in the minority/multi-ethnic areas. However, several limitations should be noted when interpreting our findings. Firstly, due to the cross-sectional nature of the study, we cannot determine the causality among depressive symptoms, rumination, and insomnia. A longitudinal study will be designed to define the causality of these variables. Secondly, although rumination mediated the association between depressive symptoms and insomnia, the contribution of its effect was much less than that of the direct effect. Other unmeasured variables, such as attentional biases, self-control or social support (12), may play a role on the relationships among depressive symptoms, rumination, and insomnia. Thirdly, all the questionnaires in this study were self-reported. Although the self-report questionnaires provide a rapid access to a large number of college students' mental health status data via the Internet (45), the data may be prone to social desirability bias (55).

In summary, this study shows that insomnia, depressive symptoms, and rumination are related constructs in college students in Qinghai province. It demonstrates the direct effects and the rumination-mediated indirect effects between depressive symptoms and insomnia; the direct effects seem to be dominant.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of Nanfang Hospital of Southern Medical University and the Third People's Hospital of Qinghai Province. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

SX and BZ: conceptualization. SX, ZX, and BZ: methodology. SX: writing—original draft. SX and SL: formal analysis. SL, PZ, JY, HA, HW, FZ, YX, NM, XiuZ, XM, JL, XW, XS, WL, XiaZ, WW, LW, RW, YH, LC, and BZ: investigation and resources. SL and BZ: writing—review and editing. PZ, JY, HA, HW, FZ, YX, NM, XiuZ, XM, JL, XW, XS, WL, XiaZ, WW, LW, RW, YH, LC, and SD: data curation. SL and BZ: funding acquisition. BZ: supervision and project administration. All authors have approved the final manuscript.

This work was supported by the President Foundation of Nanfang Hospital, Southern Medical University (Grant No. 2019Z014); the Scientific Research Foundation of Southern Medical University (Grant No. CX2018N018); the National Natural Science Foundation of China (Grant No. 81901348); the Chinese Sleep Research Society Hansoh Project, China (Grant No. 2019HSC03); and the National Natural Science Foundation of China (Grant No. 82071488).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.