Regarding psychological health, the COVID-19 pandemic, considered a public health crisis, has led to the highest levels of stress and anxiety among people (1, 2). During COVID-19, new measures such as quarantine affected individuals’ daily routines and overall well-being, potentially causing increased feelings of loneliness, anxiety, and depression, as well as higher rates of insomnia, substance abuse, and self-harm or suicidal tendencies (3). As an unprecedented global crisis, it has profoundly affected people’s daily lives worldwide, not only through the threat of COVID-19 death but also through a significant rise in mental health problems observed in China and globally, as documented in various studies (4–6). Similarly, this public health crisis has inevitably had a serious impact on individuals’ psychological health, resulting in stress symptoms, fear of death, and depression (7–11), among medical staff (12), and university students (13, 14). During the pandemic, loneliness was widespread among isolated people (15) and particularly common among college students (16, 17). Research indicates that chronic or severe loneliness can easily lead to psychological and behavioral problems or worsen mental health (18–20).

Loneliness is correlated to a range of negative psychological and physical issues (21). Among loneliness-related factors, depression is the most closely influential (22). So far, an inconsistent conclusion has been drawn on how loneliness leads to depression. Some studies have indicated that causality between the two might be mutual (23), but several studies revealed loneliness was a significant outcome of depressive symptoms (24). However, this research preferred the speculation that loneliness is a substantial precursor to depression. Longitudinal research showed that loneliness would predict depression (25), both at discrete points and over time (26, 27). In addition, loneliness has not only been shown to be correlated to depression (25) but also predicted increased depression over time (22). New research has uncovered that loneliness exerts a significant influence on depressive symptoms to a moderate degree (24).

Numerous investigations have examined the link between loneliness and depressive symptoms in university students (28). Experience of loneliness seemed to depend on abnormal strategies for coping with emotional problems in the younger (29), which made them more susceptible to depressive disorders (30). For instance, ruminative thought (31) and coping mechanisms (22) served as mediators in the link. Recently, the correlation between loneliness and depression has also been supported (32–34). Research has verified that college students are particularly susceptible to feelings of loneliness and implies a growing sense of isolation among university students owing to COVID-19 (17, 35, 36). A growing sense of loneliness among university students as a result of the pandemic was closely tied to depression (37, 38). Therefore, it is important to emphasize exploring loneliness, which might predict depression, and the relationship between death anxiety and depressive symptoms in Chinese college students throughout COVID-19.

Death anxiety, defined as a stable tendency to feel negative emotions from existential death concerns (39), is universal and becomes more prominent in death-related contexts (2, 40). Additionally, it impacts daily life even unconsciously (41). It emerges in childhood (42), peaks in early adulthood—with females showing higher levels than males (43)—and is more intense in younger people than the elderly, peaking in the 20s due to unfamiliarity and uncertainty about death (44–46). According to terror management theory (TMT), depressed individuals cannot buffer death anxiety (47). Loneliness may increase youth vulnerability and social detachment, fostering death anxiety (48). College students exhibit moderate death anxiety (49), and during this public health crisis, death fear rose (2), with young people’s death anxiety harming mental health and worsening existing struggles (50). Studies confirm loneliness correlates positively with death anxiety (51, 52) and predicts it in college students (53), while death anxiety links positively to depression and exacerbates it (54, 55). Pandemic quarantine, social isolation, loneliness, and death-related stimuli harmed mental health, triggering depression and death anxiety (18, 19). Thus, this study examines whether death anxiety mediates loneliness and depression in Chinese college students.

Negative affect involves a common factor in emotional distress (56), which intensifies cognitive processes, such as reflecting on adverse feelings, which often leads to the development and escalation of depression (57). Studies indicated significant positive correlations among Chinese college students’ feelings of loneliness, negative affect, and addiction to mobile phones, with negative affect fully mediating the link between loneliness and mobile phone addiction (58). In addition, individuals with high negative emotional experiences tend to be prone to pain and depression, so negative affect is considered to be linked with depression (59, 60). Vulnerability factors for psychological issues during outbreak scenarios, as previously recognized, encompass negative affect, anxiety, aversion to unforeseen events, and perceived susceptibility to illnesses (61), for example, COVID-19 (62, 63). During this public health crisis, studies indicated that negative affect was negatively associated with depression (60, 64, 65). Therefore, this research will verify whether negative affect is another mediating factor of loneliness-induced depression in university students.

Death anxiety is a universal human occurrence, yet it can be more evident when caused by health-threatening events (66). Earlier research has suggested that death anxiety is increasing generally throughout COVID-19 (67). Research has indicated that health worries due to COVID-19 may elicit fear of death (68), and people would experience various negative affect such as sadness, worry, anger, loneliness, depression, and so on throughout COVID-19 (69). A meta-analysis revealed that participants reported a relatively higher fear of death, and suggested that females have greater death anxiety and psychological distress (2). Studies indicated that death anxiety was positively associated with negative affect throughout the pandemic outbreak in China (70). Death anxiety was more common in younger than older people (71, 72). Studies on college students during this public health crisis showed it correlated with neuroticism, with perceived stress fully mediating this link (73). Thus, this study will verify whether death anxiety and negative affect could serve as sequential mediators in the link between loneliness and depression for Chinese university students throughout COVID-19.

To substantiate our hypotheses, this research performed an online investigation among Chinese university students to study the dynamics between loneliness and depression and the mediators of death anxiety and negative affect. We proposed that (a) there is a positive correlation between loneliness and depression in Chinese college students, (b) death anxiety acts as a mediator in the link between loneliness and depression, (c) negative affect also mediates the link between loneliness and depression, and (d) loneliness predicted college students’ depression via the sequential mediating effect of death anxiety and negative affect. We conducted a chain mediation model to evaluate the mediators of death anxiety and negative affect between loneliness and depression in Chinese university students throughout COVID-19 (see Figure 1 ).

This study examined the relationship between loneliness and depression among Chinese college students during the COVID-19 pandemic, emphasizing the sequential mediating roles of death anxiety and negative affect. The results not only confirmed the positive connection between loneliness and depression but also revealed the complex psychological process driving this link—offering new insights into pandemic-related mental health issues in young adults and informing targeted interventions. Below, we discuss these findings from theoretical, empirical, and practical viewpoints, along with limitations and future research directions.

There is a considerable risk for students to experience a boost in loneliness that correlates with depression. Our study confirms the findings from earlier studies (82, 83), and college students exhibited elevated levels of loneliness throughout COVID-19 (38). Early adulthood is a vital phase for cognitive and personality growth, and it is also the time most susceptible to mental health issues (84). Loneliness can be one of the most unpleasant experiences for college students. Evidence from across the globe suggests that college students have witnessed the most substantial spike in psychological distress rates throughout the pandemic (85).

Modeling predictors of depression established risk factors related to death anxiety and negative affect among college students. This indicated the importance of monitoring death anxiety and negative affect in youth. A study that used online ecological recognition and machine-learning predictive models to analyze posts from 17,865 active users on Weibo indicated an escalation in negative emotions (86), which is parallel to other studies (87, 88). Death anxiety, alongside the general fear of death, is a common human experience. Yet, its visibility intensifies in situations where mortality is a focal point (66), with younger individuals experiencing higher levels than their older counterparts (44, 45). The majority of the public’s reactions, both behavioral and emotional, to the virus can be comprehended through the lens of terror management theory, which posits that death anxiety is a central force influencing much of human behavior.

The study’s findings strongly align with and extend Terror Management Theory (TMT) (66), which posits that death anxiety is a core driver of psychological distress, and that social connection serves as a key “buffer” against this anxiety. Our results indicated that loneliness was positively associated with death anxiety and that death anxiety further predicted higher negative affect and depression. This supports TMT’s core tenet: when individuals lack stable social bonds, their ability to cope with mortality threats (e.g., COVID-19’s death risk) is compromised, leading to elevated death anxiety. Critically, we extended TMT by demonstrating that death anxiety does not directly drive depression in isolation—it acts as a “precursor” to negative affect, which then amplifies depressive symptoms. This sequential pathway suggests that the psychological impact of mortality salience during COVID-19 is not immediate but unfolds through cumulative emotional distress, a nuance rarely highlighted in prior TMT-based pandemic research (2, 40).

Additionally, the mediating role of negative affect aligns with emotion processing theories (57), which argue that unregulated negative emotions amplify cognitive biases (e.g., rumination about loneliness or death) and exacerbate depression. Our data showed that negative affect had the strongest correlation with depression among all mediators, and that the indirect effect of loneliness on depression via negative affect alone was far larger than the effect via death anxiety alone. This highlights that negative affect is not just a “byproduct” of loneliness or death anxiety but a central “amplifier” of depressive symptoms—consistent with studies showing that COVID-19-related negative emotions (e.g., sadness, worry) mediate mental health struggles in young adults (65, 69). By integrating TMT and emotion processing frameworks, our study offers a more holistic theoretical model for explaining how pandemic-related social isolation (loneliness) translates to depression.

Our findings are consistent with a growing body of literature on college students’ mental health during COVID-19, while also uncovering unique nuances. For instance, we confirmed prior observations that pandemic-induced loneliness correlates with higher depression in college students (37, 38, 82). Notably, our study found a strong direct correlation, which is larger than the effect sizes reported in non-pandemic studies (22, 25). This suggests that the context of COVID-19—with its mandatory home isolation, disrupted campus life, and uncertainty—may have intensified the loneliness-depression link, as social connection was actively restricted. Özgüç et al.’s (2) meta-analysis found that females report higher death anxiety during COVID-19, and our study partially supports this: gender was positively correlated with death anxiety, indicating slightly higher levels among females. However, the effect size was small, which may reflect cultural differences. Chinese college students (especially females) often receive strong family and community support during crises (5, 13), which could mitigate gender-based disparities in death anxiety—a pattern not observed in Western samples (50).

Prior research has examined either death anxiety (52, 53) or negative affect (58, 60) as single mediators in the link between loneliness and depression. Our research advances this by showing that these two factors act sequentially: loneliness first increases death anxiety (likely due to COVID-19’s mortality threat), which then heightens negative affect (as death-related worries trigger emotional distress), and finally leads to depression. The serial indirect effect may seem small, but it is theoretically critical—it reveals that ignoring either mediator would oversimplify the psychological chain linking loneliness to depression during public health crises.

While the study provides valuable insights, it has several limitations that should be addressed in future research. First, the investigation’s main goal was to examine the elements that shape the link between loneliness and depression throughout COVID-19. The sample was limited to College students from three universities in Guangdong Province, China, limiting generalizability to students in other regions or other countries. Additionally, convenience sampling (via Wenjuanxing) may have introduced selection bias. Future studies should use stratified random sampling to include students from diverse backgrounds and regions. Second, the cross-sectional nature of the data prevents us from establishing causality. For example, while we hypothesize that loneliness predicts death anxiety and negative affect, it is also possible that depression increases loneliness (24) or that negative affect exacerbates loneliness (58). Longitudinal studies tracking students’ mental health across different phases of the pandemic would help clarify the direction of these relationships. Third, all variables were assessed via self-reported scales, which may be influenced by social desirability bias. Future research could complement self-reports with objective measures to improve data accuracy. Fourth, we did not assess factors that may moderate the observed pathways, such as social support or resilience. For example, high social support may weaken the link between loneliness and death anxiety, while resilience may buffer the effect of negative affect on depression. Including these moderators would provide a more comprehensive understanding of pandemic-related mental health. Lastly, the tested model may not provide the best fit to the data. Alternative model structures may be more parsimonious: For example, a “parallel multiple mediation model” or a “moderated mediation model” might yield better fit indices. However, our choice of sequential model was prioritized to test the theoretically derived “chain reaction” of psychological distress during COVID-19, rather than purely pursuing statistical fit. To address this limitation in future research, we propose two key adjustments: Incorporate additional variables or test alternative model structures.

In summary, this study demonstrates that loneliness is significantly positively associated with depression among Chinese college students during COVID-19, and this association is partially explained by the sequential mediating role of death anxiety and negative affect. The findings extend theoretical frameworks (TMT and emotion processing) to pandemic contexts, align with and refine prior empirical research, and offer actionable strategies for universities and policymakers to support students’ mental health. While limitations exist, the study highlights the importance of addressing not just direct mental health symptoms (e.g., depression) but also the underlying social and emotional factors (e.g., loneliness, death anxiety) that drive them—especially during public health crises. By targeting these interconnected factors, interventions can more effectively mitigate the long-term mental health impacts of pandemics on young adults.