Childhood trauma (CT) can occur when a child witnesses or experiences overwhelming adverse events in childhood (1, 2). These can happen in relationships that involve abuse, assault, neglect, violence, exploitation, or bullying. Chronic CT causes self-defeating cognition and behaviors that hinder the healthy development of schemata (3). Moreover, previous studies have found that exposure to CT has long-term consequences (4), including depressive disorders, anxiety disorders, posttraumatic stress disorder (PTSD), self-injurious behaviors, and suicidal behavior (5–8). Three varieties of CT are frequently acknowledged: emotional abuse (EA), physical abuse (PA), and sexual abuse (SA) (7, 9–12). Globally, the prevalence of child SA varies from 2 to 62%, while the annual prevalence of PA ranges from 4 to 16%, and around 10% of children experience EA or neglected abuse (13).

Several systematic reviews and meta-analyses have found that those exposed to any form of CT are much more likely to report chronic mental health conditions (5–8). For example, McKay and colleagues reported that those exposed to multiple forms of CT have more than three times the odds of developing a mental disorder (6). Another meta-analysis reported that childhood adversities significantly increased psychosis across all research designs, with an overall effect of increasing 2.78 times (8). In addition, CT may increase the risk of depression, anxiety, PTSD, and so on. However, although many studies investigate the associations between CT and psychopathological symptoms, such as PTSD, depression, and anxiety, the interaction studies between PTSD, depression, and anxiety among individuals exposed to different types of CT are yet to be comprehensively explored.

Network analysis is one approach to describe the dynamics between symptoms from a psychopathology perspective (14). Compared with traditional latent variable models, network analysis could capture and express the correlation and structure of the observed variables (15, 16). According to the network perspective, an episode of disorder occurs whenever the necessary amount of symptoms get active for a while (17). Recovery from an illness happens when symptoms deactivate, the association between them dissolves, or both. Hence, network analysis helps guide therapeutic intervention or treatment for psychological diseases (14). In network analysis, specific symptoms (defined as nodes) are observed variables that may have solid causal links with one another. Moreover, a bridge symptom is a bridge between two other symptoms in the network (15). Similarly, specific associations (defined as edges) between two symptoms may be substantially more robust and influential than other edges (15). In a network structure, a vital edge may be thought of as a strong association between two symptoms that are likely to occur or non-occur at the same time (18). To date, the network analysis contributed to advance psychopathology research (17, 19–24). For instance, Mullarkey et al. reported that self-hatred, loneliness, and sadness are the most central symptoms of adolescent depression (23), suggesting that these symptoms (and relationships between symptoms) should be prioritized in theoretical frameworks for teenage depression and may also serve as significant therapeutic targets for adolescent depression therapies. In addition, Network analysis gives hints to the causal mechanism behind PTSD in those who have experienced a severe natural catastrophe (19) and the association between PTSD symptoms in the acute phase after trauma and the chronic phase (20). Moreover, Bryant et al. used network analysis to simulate the development of PTSD symptoms over time and to pinpoint the most crucial target symptoms for early intervention (20). McNally et al. employed it to illustrate how major depressive illness and obsessive-compulsive disorder are symptomatically related (22).

Therefore, the current study analyzed psychopathological symptoms using network analysis, which aimed to go beyond the level of symptom severity and to understand which symptoms might be particularly central to the experience of CT. We used network approaches to identify the strong association between symptoms among individuals exposed to different types of CT. Furthermore, significant differences in network structures among various CT groups could be found, which could provide advice on intervention or treatment to decrease the risk of psychosis for specific CT groups. The objectives of this study were to examine the associations between psychopathological symptoms, including PTSD, depression, and anxiety, across three subgroups (individuals exposed to EA, PA, and SA). Additionally, to identify central symptoms in the network structure within the three CT groups. Finally, to compare the three groups to find the significant differences in network interactions.

This is a large-scale network analysis to investigate the interaction of PTSD, depression, and anxiety symptoms among college students exposed to three types of CT (EA, PA, and SA). The highest prevalence was SA group, with 3.62% among participants. The results showed that central symptoms within the three groups are uncontrollable worry, sad mood, irritability, and fatigue, which indicates these core symptoms play essential roles in maintaining the whole psychological symptoms network. Results also revealed significant differences across three groups: the SA group is more likely to be associated with PTSD symptoms, the EA group is more likely to be associated with suicide-related symptoms, and the PA group is more likely to be associated with anxiety symptoms.

The current study presented several remarkable advantages. Although research has successfully established a connection between CT and psychopathological symptoms (5–7), the potential associations that may account for the connection are still not fully understood. This study revealed associations between psychological symptoms among college students exposed to different types of CT. Central symptoms of the network structures, which integrated depression, anxiety, and PTSD symptoms, were almost the same among the three subgroups. Our result showed that uncontrollable worry, sad mood, irritability, and fatigue might trigger other symptoms in psychological symptoms dynamics when college students are exposed to EA, PA, or SA. This result would be helpful to therapeutic intervention or future treatment for psychological diseases.

Consistent with previous systematic reviews and meta-analyses, the current study also found that CT experiences increase the risk of depression and anxiety (6, 7, 39, 40). Our results align with the existence of an affective pathway to psychological symptoms (5)—CT may lead to psychosis through a path of heightened emotional distress (e.g., uncontrollable worry, sad mood, irritability, and fatigue). Previous research has proposed that traumatic events may cause structural and neurochemical abnormalities in the brain and nervous system, affecting the function of the immune-inflammatory system, the autonomic nervous system (ANS), and the hypothalamus-pituitary-adrenal axis (HPA-axis), which is involved in the stress response (41–44). This may result in an increased risk of uncontrollable worry and irritability. Moreover, according to the cognitive models of psychosis perspective, trauma may lead to diminished neuropsychological performance within domains of attention and cognitive flexibility (45). Even lead to negative beliefs about the self, world, and others (46). These beliefs may, in turn, lead to distressing interpretations of everyday events, eventually resulting in feeling down, depressed or hopeless or tired, or having little energy.

Results showed significant differences in total scores of depression, anxiety, and PTSD, as well as network characteristic among three groups (EA, PA, and SA). College students exposed to SA in childhood seem to have stronger connections between PTSD symptoms, such as sleep disturbance, difficulty concentrating, flashbacks, and emotional cue reactivity. Children can be sexually assaulted by adults and other kids who, because of their age or developmental stage, have authority over the victim or are in a position of trust or obligation for them. This result could be supported by previous studies (39, 40, 47). A study reported that psychiatric outpatients with a history of childhood SA had significantly increased rates of PTSD (40). Another study found that 86% of childhood SA survivors (n = 117) met DSM-III-R criteria for a PTSD diagnosis at some point during their lives. The duration and severity of the abuse were directly correlated with the severity and variance in PTSD symptoms (39). Individuals exposed to SA in childhood are more likely to have sudden intrusive memories of the original traumatic experience and a wave of the remembered emotional response (47). Previous studies reported that in survivors of sexual assault, there is a dysregulation of the HPA axis, which may be a primary source of the structural and functional abnormalities that contribute to PTSD symptoms (48). Through victim-blaming attitudes and the perpetuating of rape myths, sociological effects of attack promote the development of PTSD (42, 49, 50). For the network research, Yang et al. also reported a strong connection between sleep disturbance and difficulty concentrating (24), which might be explained by sleep disturbance arousing a series of adverse outcomes including lack of focusing (51, 52). The stronger associations between flashbacks and emotional cue-reactivity were also found in these studies (19, 21, 24). Understanding the associations between psychological symptoms affected by SA can potentially inform targeted efforts to prevent psychopathology.

Compared with PA and SA groups, the EA group is more likely to present stronger connections with suicide-related symptoms. A meta-analysis of longitudinal studies found that EA (OR = 3.98) was associated with suicide attempts (53). Another population-based review also reported that childhood EA was significantly associated with suicide-related behaviors (54). According to cognitive theories of psychosis, EA may result in negative thoughts about oneself, the outside environment, and other people. These assumptions might result in upsetting interpretations of commonplace occurrences (46). The shared social and biological surroundings between the child and the perpetrator are possible causal processes driving the EA and suicide-related symptoms (55).

Furthermore, any acquired brain damage resulting from EA in childhood might contribute to suicide-related behaviors (56). Shared biology might play a role in the link. For example, frightening, controlling, or isolating a possibly hereditary feature linked to suicide-related behaviors may be passed on biologically from parents to children, raising the risk of suicide-related behaviors in the offspring (57). Our results were in line with these results, which support that the EA group is more likely to present symptoms associated with suicide-related symptoms. Furthermore, a systematic review and meta-analysis from 124 studies suggested that EA individuals had a higher risk of developing depressive disorders (7). Therefore, it might be found that the EA group is more likely to present depressive symptoms like anhedonia, sad mood, and restlessness.

The PA group has shown stronger connections between anxiety symptoms (i.e., trouble relaxing, restlessness, and irritability). This result indicated that these associations were more critical in the affective symptoms network structure among college students exposed to PA in childhood. A systematic review reported that exposure to PA was associated with anxiety symptoms (OR = 1.7) (58). Moreover, individuals with a history of PA recognize anger affect more quickly and accurately than those without PA history (59). Perpetrators (i.e., parents or other family members) may legitimize PA in childhood as disciplinary in a social setting. The child may absorb the viewpoint and model it with anxiety-related symptoms as a sort of self-punishment (60).

Although this is a large-scale study containing the whole colleges in Jilin province, China, several limitations should be considered. First, the CTQ-SF is a self-report, retrospective measure of CT and may thus be prone to bias (e.g., social desirability, memory bias, and demand characteristics). Second, due to the cross-sectional design, the causation between psychosis symptoms cannot be confirmed. Third, some participants who suffered from depressive or anxiety disorders may be included in this study, which may bias the findings to an uncertain extent. Finally, we only compared a single form of CT group without consideration of overlap with other types, and further studies should be explored among overlapping forms of CT groups.

In conclusion, this network analysis revealed that uncontrollable worry, sad mood, irritability, and fatigue were the most core symptoms of psychotic symptoms among college students exposed to childhood EA, PA, and SA. Moreover, there are significant differences between the three groups. The SA group is more likely to be associated with PTSD symptoms, the EA group is more likely to be associated with suicide-related symptoms, and the PA group is more likely to be associated with anxiety symptoms. Therefore, it is recommended that schools and local communities should support early intervention to improve psychological wellbeing.

The raw data supporting the conclusions of this article will be made available by the corresponding authors, under reasonable request.

The study involving human participants was reviewed and approved by the Ethics Committee at Jilin University. The participants provided their electronic informed consent to participate in this study.

Study design: YW, SX, and YJ. Study implementation and data collection: XS, XW, and HL. Data analysis: YJ and ZH. Manuscript: YJ, YW, SX, and JL. All authors read and approved the final manuscript.