This cross-sectional study investigates the association of adverse childhood experiences with nonsuicidal self injury and suicidality among Chinese adolescents.
Key Points
Question
Are adverse childhood experiences (ACEs) associated with nonsuicidal self-injury (NSSI) or suicidality among Chinese adolescents, and does a supportive school environment modify such associations?
Findings
In this cross-sectional study involving 95 549 adolescents in China, exposure to ACEs was associated with an increased risk of NSSI and suicidality, with particularly high increases in risk for threat-related ACEs. Associations of ACEs with NSSI and suicidality were modified by supportive school environments.
Meaning
These findings suggest the potential benefits of enhancing school environments in preventing NSSI and suicidality among individuals who have experienced ACEs.
Abstract
Importance
Studies investigating the role of supportive school environments in associations between adverse childhood experiences (ACEs) and nonsuicidal self-injury (NSSI) and suicidality among adolescents are lacking.
Objective
To assess associations of ACEs with NSSI and suicidality among adolescents and examine the modifying role of supportive school environments in such associations.
Design, Setting, and Participants
Using data from the 2021 School-Based Chinese Adolescents Health Survey, this cross-sectional study includes Chinese adolescents in grades 7 through 12 from 326 schools across 8 provinces in China. Statistical analysis was performed from March to October 2024.
Exposure
ACEs, including individual indicators and cumulative numbers by total and categorized as threat-related and deprivation-related ACEs, were assessed.
Main Outcome and Measures
NSSI, suicidal ideation, and suicide attempts were measured using validated questionnaires. Weighted Poisson regression models were used. The modifying interaction of supportive school environments was assessed using multiplicative interactions.
Results
Among 95 549 adolescents included in this study (mean [SD] age, 14.9 [1.8] years; 47 617 males [49.8%]), 45 236 individuals (47.3%) had experienced at least 1 ACE. Compared with adolescents with no ACEs, those who experienced 4 or more ACEs had an increased risk of NSSI (prevalence ratio [PR], 1.31; 95% CI, 1.30-1.33) and suicidality (ideation: PR, 1.41; 95% CI, 1.39-1.42; attempts: PR, 1.25; 95% CI, 1.24-1.27) after adjusting for covariates, including supportive school environments. A dose-response association was observed between the number of ACEs and the risk of NSSI and suicidality (eg, NSSI: PR, 1.06; 95% CI, 1.05-1.06 for exposure to 1 vs 0 ACEs; PR, 1.10; 95% CI, 1.10-1.11 for exposure to 2 vs 0 ACEs). The highest PRs were found for associations of threat-related ACEs (eg, physical abuse, emotional abuse, and bullying) with NSSI and suicidality (eg, exposure to ≥2 vs 0 threat-related ACEs: PR, 1.28; 95% CI, 1.27-1.29 for NSSI; PR, 1.33; 95% CI 1.32-1.34 for ideation; PR, 1.18; 95% CI, 1.17-1.19 for attempts). All individual ACE indicators were positively associated with NSSI and suicidality, with particularly high PRs for emotional abuse (eg, PR, 1.26; 95% CI, 1.24-1.27 for NSSI). Additionally, supportive school environments significantly modified associations of ACEs with NSSI and suicidality (eg, interaction ratio for NSSI, 0.81; 95% CI, 0.76-0.88).
Conclusion and relevance
In this study, exposure to ACEs was associated with an increased risk of NSSI and suicidality, with particularly high increases for threat-related ACEs, independent of the interaction of supportive school environments, but supportive school environments could modify such detrimental outcomes. These findings underscore the importance of enhancing school environments to prevent NSSI and suicidality among adolescents who have experienced ACEs.
Introduction
Nonsuicidal self-injury (NSSI) refers to the direct and deliberate destruction of body tissue without suicidal intent, which is distinct from suicidality (eg, suicidal ideation [SI] and suicide attempt [SA]) in both purpose and frequency. NSSI, SI, and SA are persistent public health issues among adolescents worldwide,1,2 with prevalence possibly increasing during the COVID-19 pandemic. The impact of NSSI and suicidality on adolescents can be profound, often manifesting as various subclinical psychosocial difficulties that are associated with large increases in odds of future suicidal behaviors.3 Therefore, identifying modifiable factors associated with NSSI and suicidality is crucial for planning public health interventions and promoting adolescent psychosocial health.
Although previous research has identified several risk factors associated with NSSI and suicidality, such as interpersonal relationships, academic stress, and mental illness,4 there is also evidence suggesting an association between adverse childhood experiences (ACEs) and NSSI and suicidality.5,6 ACEs encompass a broad spectrum of potentially stressful encounters during childhood,7 with estimates indicating that 27% to 70% of Chinese adolescents are exposed to at least 1 ACE.8,9 However, most previous studies have been conducted on adults using retrospectively collected information on ACEs, primarily focusing on NSSI or suicidality in adulthood.10,11 In contrast to the relatively well-studied ACEs among Western populations, only a few investigations, with limited sample sizes or restricted geographic areas, have explored the association of ACEs with NSSI and suicidality among Chinese adolescents.12,13 Moreover, aside from long-term outcomes, the role of ACEs in the sensitive neurological and brain structural development stage of adolescence14,15 has been less studied.
Additionally, previous research often focused on the conventional set of ACEs from the Centers for Disease Control and Prevention (CDC)-Kaiser Permanente ACE study,16,17 which cover abuse, neglect, and household dysfunction. However, recent studies have highlighted that conventional ACEs may not adequately reflect perceived childhood adversity. An expanded set of ACEs, including school bullying and community violence, was outlined in the ACE International Questionnaire (ACE-IQ) developed by the World Health Organization18 and has also been reported to be prevalent and associated with NSSI and suicidality.19,20 Nevertheless, previous studies typically conceptualized ACEs using the cumulative risk model, which suggested that all types of ACEs were associated with NSSI and suicidality through the same underlying mechanism.5,6,12 An alternative approach, the dimensional model of adversity and psychopathology, was further proposed to differentiate threat-related and deprivation-related ACEs because these 2 dimensions were considered to be associated with emotion regulation and cognitive control in adolescence.21
During the past decade, emerging studies have striven to identify potential effect modifiers that may interact in associations between ACEs and adverse outcomes.22,23 These may help develop tailored interventions to mitigate the negative outcomes associated with ACEs among adolescents. School is the place where adolescents spend most of their time, and it plays a critical role in promoting student health and development. Current evidence suggests that a positive and supportive school environment can promote adolescent mental health24,25 and is negatively associated with NSSI.26 Moreover, a cross-sectional study in the US27 suggested that supportive school environments may buffer the association between ACEs and suicidality. Nevertheless, knowledge regarding the modifying role of supportive school environments in the associations of expanded ACEs with NSSI and suicidality is limited.
In this nationwide, large-scale cross-sectional study, we aimed to comprehensively estimate associations of ACEs (individual indicator and cumulative numbers by total and categorized as threat-related and deprivation-related ACEs) with NSSI and suicidality among Chinese adolescents. The modifying interaction of supportive school environments in investigated associations was also evaluated.
Methods
Study Design and Participants
This cross-sectional study used data from the 2021 School-Based Chinese Adolescents Health Survey (SCAHS), an ongoing survey of health-related behaviors among Chinese adolescents in grades 7 through 12.28 The 2021 SCAHS used a multistage, stratified cluster random sampling method; detailed data-collection procedures have been described previously.29 A total of 98 192 students from 326 high schools across 8 provinces of China participated in this survey, yielding an overall response rate of 97.3%. Self-reported questionnaires were administered in classrooms.30 For this analysis, we excluded 2643 participants with missing data on ACEs, NSSI, and suicidality, resulting in a final sample of 95 549 participants (age range, 11-21 years). The detailed process of participant selection is listed in eMethods and eFigure 1 in Supplement 1. Participation in the SCAHS was voluntary, with written informed consent obtained from all participants and 1 of their legal guardians. Ethical approval was granted by the institutional review board of Sun Yat-sen University School of Public Health. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Adverse Childhood Experiences
Based on the CDC-Kaiser Permanente ACE study16,17 and the ACE-IQ,18 we collected data on 15 conventional and expanded forms of ACEs through self-reported questionnaires. These ACEs were further categorized into threat-related and deprivation-related dimensions. Details are presented in the eMethods in Supplement 1. Each ACE indicator was dichotomized and summed to generate a cumulative ACE score ranging from 0 to 15. Participants were further categorized into 5 groups based on cumulative ACE score: 0, 1, 2, 3, and 4 or higher.29 Additionally, cumulative scores for threat-related and deprivation-related ACEs were generated, and participants were grouped into 3 categories based on these scores: 0, 1, and 2 or higher for both threat-related and deprivation-related ACEs.31
Nonsuicidal Self-Injury
NSSI was assessed using the Chinese version of the Functional Assessment of Self-Mutilation.32 Participants were asked, “In the past 12 months, have you ever harmed yourself deliberately without intending to take your life?” We provided 8 different forms of NSSI, including head banging, hair pulling, pinching, hitting, burning, biting, cutting, and scratching. For participants who reported engaging in NSSI, the frequency of NSSI was assessed and a total NSSI frequency was calculated. In this study, NSSI was dichotomized based on total frequency, with participants reporting 3 or more incidents categorized as “yes” and those with fewer than 3 incidents categorized as “no.”12,33,34,35
Suicidal Ideation and Suicide Attempt
SI was defined as responding, “1 or more times,” to the question, “During the past 12 months, how many times did you seriously consider attempting suicide?” SA was assessed by the question “During the past 12 months, how many times did you attempt suicide?” and asked that the student rate on a scale of zero or once or more.36
Supportive School Environment
The supportive school environment was evaluated using 3 questions: (1) “How would you describe your relationship with your classmates?”, (2) “How would you describe your relationship with the teachers in your class?”, and (3) “At school, how many of your close friends frequently make mistakes?”37 Responses to each question were dichotomized, with 0 indicating a poor situation and 1 indicating a good situation. A total score ranging from 0 to 3 was then calculated, with higher scores suggesting more supportive school environments.
Covariates
Covariates assessed included age, sex (male and female), current drinking status (yes and no), current smoking status (yes and no), household socioeconomic status (>mean, mean, and <mean), academic pressure (none, moderate, and severe), and depressive symptoms (measured using the 20-item Center for Epidemiologic Studies Depression Scale, as detailed in the eMethods in Supplement 1). To minimize bias in estimating associations of ACEs and supportive school environments with NSSI and suicidality, a directed acyclic graph was used to select the appropriate set of covariates for adjustment (eFigure 2 in Supplement 1).38
Statistical Analysis
Data are presented as means and SDs for continuous variables and frequencies with percentages for categorical variables. Characteristics were compared across NSSI, SI, and SA using analysis of variance for continuous variables and χ2 tests for categorical variables, as appropriate. Prevalence estimates and corresponding 95% CIs for NSSI, SI, and SA were calculated and weighted for the complex sampling design and nonresponse.39 Weighted Poisson regression models that accounted for the stratified cluster survey design were performed to test associations of ACEs and supportive school environments with NSSI, SI, and SA, and prevalence ratios (PRs) with 95% CIs were estimated (details are presented in the eMethods in Supplement 1). ACE exposure was treated as a continuous variable, binary variable, and categorical variable, and individual ACEs were assessed. Moreover, multiplicative interaction analyses were conducted to assess the modifying interaction of supportive school environments with associations between ACEs and NSSI, SI, and SA. An interaction product term between ACEs and supportive school environments was included in models. Multiplicative interaction was measured by the interaction ratio, calculated as PRinteraction term/(PRACEs × PRsupportive school environment).40,41 A positive multiplicative interaction was indicated by an interaction ratio greater than 1, a negative interaction by a ratio less than 1, and no interaction by a ratio equal to 1.42 To explore potential variations across subgroups, we conducted subgroup analyses stratified by sex (males and females).
We conducted additional sensitivity analyses to validate the robustness of our main findings. First, we used an alternative cutoff for NSSI, categorizing individuals who reported 5 or more ACE incidents as “yes” and those with fewer than 5 ACE incidents as “no.”43 Second, we additionally adjusted for depressive symptoms, which were not initially included in the main analysis owing to their potential role as a mediator in associations of ACEs with NSSI and suicidality.44 All statistical analyses were performed using R statistical software version 4.2.2 (R Foundation for Statistical Computing), with a two-sided P-value < .05 considered statistically significant. Analyses were performed from March to October 2024.
Results
Among 95 549 participants (47 617 male [49.8%]; mean [SD] age, 14.9 [1.8] years), 45 236 participants (47.3%) had been exposed to at least 1 of 15 ACEs. Table 1 presents characteristics of the study population. The mean (SD) score of supportive school environments was 1.7 (0.9). As shown in eTable 1 in Supplement 1, and the prevalence was 15.0% (95% CI, 14.8%-15.3%) for NSSI, 18.7% (95% CI, 18.4%-19.0%) for SI, and 5.6% (95% CI, 5.5%-5.8%) for SA.
Table 1. Characteristics of Participants.
| Characteristic | Participants, No. (%) | |||
|---|---|---|---|---|
| Overall (N = 95 549) | NSSI (n = 14 481) | Suicidal ideation (n = 17 866) | Suicide attempt (n = 5315) | |
| Age, mean (SD), y | 14.93 (1.76) | 14.75 (1.69) | 14.77 (1.69) | 14.56 (1.63) |
| Sex | ||||
| Male | 47 617 (49.8) | 5357 (37.0) | 6253 (35.0) | 1598 (30.1) |
| Female | 47 932 (50.2) | 9123 (63.0) | 11 613 (65.0) | 3717 (69.9) |
| HSS | ||||
| >Mean | 24 864 (26.0) | 3230 (22.3) | 3862 (21.6) | 1225 (23.0) |
| Mean | 56 827 (59.5) | 8506 (58.7) | 10 598 (59.3) | 2989 (56.2) |
| <Mean | 13 584 (14.2) | 2713 (18.7) | 3367 (18.8) | 1087 (20.5) |
| Missing | 274 (0.3) | 31 (0.2) | 39 (0.2) | 14 (0.3) |
| Current smoking status | ||||
| No | 5299 (5.5) | 1266 (8.7) | 1628 (9.1) | 752 (14.1) |
| Yes | 90 250 (94.5) | 13 214 (91.3) | 16 238 (90.9) | 4563 (85.9) |
| Current drinking status | ||||
| No | 4826 (5.1) | 1195 (8.3) | 1522 (8.5) | 703 (13.2) |
| Yes | 90 723 (94.9) | 13 285 (91.7) | 16 344 (91.5) | 4612 (86.8) |
| Academic pressure | ||||
| None | 20 036 (21.0) | 1698 (11.7) | 2073 (11.6) | 644 (12.1) |
| Moderate | 43 441 (45.5) | 5421 (37.4) | 6570 (36.8) | 1848 (34.8) |
| Severe | 31 846 (33.3) | 7331 (50.6) | 9185 (51.4) | 2810 (52.9) |
| Missing | 226 (0.2) | 30 (0.2), | 38 (0.2) | 13 (0.2) |
| Supportive school environment, mean (SD) | 1.97 (0.91) | 1.64 (0.96) | 1.63 (0.96) | 1.50 (0.97) |
| ACEs, No./participant, mean (SD) | 0.95 (1.37) | 1.90 (1.93) | 1.96 (1.90) | 2.62 (2.16) |
| ACEs status | ||||
| No | 50 313 (52.7) | 4128 (28.5) | 4631 (25.9) | 906 (17.0) |
| Yes | 45 236 (47.3) | 10 352 (71.5) | 13 235 (74.1) | 4409 (83.0) |
| No. of ACEs | ||||
| 0 | 50 313 (52.7) | 4128 (28.5) | 4631 (25.9) | 906 (17.0) |
| 1 | 21 921 (22.9) | 3369 (23.3) | 4232 (23.7) | 1022 (19.2) |
| 2 | 12 304 (12.9) | 2591 (17.9) | 3366 (18.8) | 982 (18.5) |
| 3 | 5563 (5.8) | 1814 (12.5) | 2328 (13.0) | 835 (15.7) |
| ≥4 | 5448 (5.7) | 2578 (17.8) | 3309 (18.5) | 1570 (29.5) |
| No. of threat-related ACEs | ||||
| 0 | 69 953 (73.2) | 6424 (44.4) | 7755 (43.4) | 1744 (32.8) |
| 1 | 16 314 (17.1) | 3969 (27.4) | 5029 (28.1) | 1429 (26.9) |
| ≥2 | 9282 (9.7) | 4087 (28.2) | 5082 (28.4) | 2142 (40.3) |
| No. of deprivation-related ACEs | ||||
| 0 | 62 206 (65.1) | 7286 (50.3) | 8409 (47.1) | 1947 (36.6) |
| 1 | 20 628 (21.6) | 3816 (26.4) | 4928 (27.6) | 1460 (27.5) |
| ≥2 | 12 715 (13.3) | 3378 (23.3) | 4529 (25.3) | 1908 (35.9) |
Abbreviations: ACE, adverse childhood experience; HSS, household socioeconomic status; NSSI, nonsuicidal self-injury.
A higher number of ACEs was associated with increased risks of NSSI and suicidality (eg, PR per 1-unit increase in ACEs, 1.05; 95% CI, 1.05-1.06 for NSSI in the fully adjusted model 2). A dose-response association was observed between cumulative ACE exposure and NSSI and suicidality, even in the fully adjusted model 2 (eg, NSSI: PR, 1.06; 95% CI, 1.05-1.06 for exposure to 1 vs 0 ACEs; PR, 1.10; 95% CI, 1.10-1.11 for exposure to 2 vs 0 ACEs). For 4 or more ACEs compared with 0 ACEs in model 2, PRs were 1.31 (95% CI, 1.30-1.33) for NSSI, 1.41 (95% CI, 1.39-1.42) for SI, and 1.25 (95% CI, 1.24-1.27) for SA (Table 2). Furthermore, threat-related and deprivation-related ACEs were positively associated with NSSI and suicidality, even in the fully adjusted model 2 (eg, ≥2 vs 0 threat-related ACEs: PR, 1.28; 95% CI, 1.27-1.29 for NSSI; PR, 1.33; 95% CI 1.32-1.34 for SI; PR, 1.18; 95% CI, 1.17-1.19 for SA). Similar results were observed across male and female subgroups (eTables 2-3 in Supplement 1).
Table 2. Associations of Cumulative ACEs With NSSI and Suicidality.
| ACEs | PR (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| NSSI | Suicidal ideation | Suicide attempt | |||||
| Model 1a | Model 2b | Model 1a | Model 2b | Model 1a | Model 2b | ||
| Cumulative ACEs, per 1-unit increase | 1.31 (1.30-1.32) | 1.05 (1.05-1.06) | 1.32 (1.31-1.33) | 1.07 (1.07-1.07) | 1.46 (1.44-1.48) | 1.04 (1.04-1.04) | |
| Any ACE, No | |||||||
| 0 | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | |
| ≥1 | 2.56 (2.46-2.66) | 1.11 (1.11-1.12) | 2.87 (2.77-2.98) | 1.15 (1.15-1.16) | 4.81 (4.43-5.23) | 1.07 (1.06-1.07) | |
| No. of ACEs | |||||||
| 0 | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | |
| 1 | 1.86 (1.77-1.95) | 1.06 (1.05-1.06) | 2.05 (1.96-2.14) | 1.08 (1.08-1.09) | 2.64 (2.39-2.93) | 1.03 (1.02-1.03) | |
| 2 | 2.47 (2.34-2.60) | 1.10 (1.10-1.11) | 2.83 (2.70-2.96) | 1.15 (1.14-1.16) | 4.34 (3.92-4.81) | 1.06 (1.05-1.06) | |
| 3 | 3.63 (3.43-3.84) | 1.20 (1.18-1.21) | 4.06 (3.87-4.26) | 1.26 (1.25-1.28) | 7.41 (6.66-8.24) | 1.12 (1.11-1.13) | |
| ≥4 | 4.83 (4.59-5.07) | 1.31 (1.30-1.33) | 5.42 (5.19-5.66) | 1.41 (1.39-1.42) | 13.07 (11.88-14.38) | 1.25 (1.24-1.27) | |
| Cumulative threat-related ACEs, per 1-unit increase | 1.51 (1.49-1.53) | 1.09 (1.09-1.10) | 1.50 (1.48-1.51) | 1.11 (1.11-1.11) | 1.70 (1.67-1.73) | 1.07 (1.06-1.07) | |
| No. of threat-related ACEs | |||||||
| 0 | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 (Reference | 1 [Reference] | 1 [Reference] | |
| 1 | 2.48 (2.37-2.58) | 1.12 (1.11-1.13) | 2.53 (2.44-2.62) | 1.15 (1.14-1.16) | 3.25 (3.00-3.52) | 1.05 (1.05-1.06) | |
| ≥2 | 4.07 (3.91-4.24) | 1.28 (1.27-1.29) | 4.06 (3.93-4.21) | 1.33 (1.32-1.34) | 7.46 (6.92-8.04) | 1.18 (1.17-1.19) | |
| Cumulative deprivation-related ACEs, per 1-unit increase | 1.34 (1.32-1.36) | 1.04 (1.04-1.05) | 1.42 (1.41-1.44) | 1.07 (1.07-1.07) | 1.71 (1.67-1.76) | 1.04 (1.04-1.05) | |
| No. of deprivation-related ACEs | |||||||
| 0 | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | |
| 1 | 1.50 (1.44-1.56) | 1.05 (1.04-1.05) | 1.66 (1.60-1.72) | 1.07 (1.07-1.08) | 2.10 (1.94-2.27) | 1.03 (1.03-1.04) | |
| ≥2 | 2.01 (1.92-2.09) | 1.10 (1.09-1.11) | 2.35 (2.26-2.43) | 1.16 (1.15-1.17) | 4.03 (3.74-4.34) | 1.10 (1.09-1.11) | |
Abbreviations: ACE, adverse childhood experience; NSSI, nonsuicidal self-injury; PR, prevalence ratio.
Model 1 was adjusted for age, sex, current drinking status, current smoking status, household socioeconomic status, and academic pressure.
Model 2 was adjusted for the variables in model 1, plus supportive school environments.
Most ACE subtypes were positively associated with an increased risk of NSSI and suicidality, even in the fully adjusted model 2. Specifically, the associations with the highest PRs were found for emotional abuse (eg, PR, 1.26; 95% CI, 1.24-1.27 for NSSI) (Table 3). Similar findings were observed in male and female subgroups (eTables 4-5 in Supplement 1).
Table 3. Associations of Each ACE Subtype With NSSI and Suicidality.
| ACE subtype | PR (95% CI) | |||||
|---|---|---|---|---|---|---|
| NSSI | Suicidal ideation | Suicide attempt | ||||
| Model 1a | Model 2b | Model 1a | Model 2b | Model 1a | Model 2b | |
| Threat-related ACEs | ||||||
| Physical abuse | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.72 (1.65-1.78) | 1.25 (1.23-1.27) | 2.81 (2.69-2.93) | 1.30 (1.28-1.32) | 4.40 (4.04-4.79) | 1.21 (1.19-1.23) |
| Sex abuse | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.59 (2.44-2.75) | 1.21 (1.19-1.23) | 2.31 (2.19-2.44) | 1.21 (1.19-1.23) | 3.52 (3.19-3.87) | 1.15 (1.13-1.17) |
| Emotional abuse | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.86 (2.74-2.97) | 1.26 (1.24-1.27) | 2.98 (2.89-3.08) | 1.33 (1.32-1.34) | 4.99 (4.67-5.33) | 1.22 (1.21-1.23) |
| Household substance abuse | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.37 (2.17-2.59) | 1.19 (1.16-1.22) | 2.36 (2.19-2.53) | 1.23 (1.21-1.26) | 3.12 (2.69-3.62) | 1.14 (1.11-1.17) |
| Witness of community violence | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.18 (2.08-2.28) | 1.15 (1.13-1.16) | 1.97 (1.89-2.05) | 1.15 (1.13-1.16) | 2.50 (2.31-2.72) | 1.08 (1.07-1.09) |
| Household domestic violence | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.12 (2.03-2.21) | 1.14 (1.13-1.15) | 2.26 (2.18-2.34) | 1.19 (1.18-1.2) | 2.89 (2.69-3.10) | 1.09 (1.09-1.10) |
| Sex discrimination | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.22 (2.11-2.33) | 1.18 (1.17-1.20) | 2.11 (2.02-2.20) | 1.20 (1.19-1.22) | 2.67 (2.44-2.91) | 1.11 (1.10-1.13) |
| Bullying | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.51 (2.42-2.60) | 1.16 (1.15-1.17) | 2.37 (2.29-2.44) | 1.18 (1.17-1.19) | 3.26 (3.06-3.49) | 1.10 (1.09-1.10) |
| Deprivation-related ACEs | ||||||
| Physical neglect | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.47 (1.42-1.53) | 1.05 (1.04-1.06) | 1.65 (1.60-1.71) | 1.08 (1.08-1.09) | 2.42 (2.27-2.58) | 1.06 (1.05-1.06) |
| Emotional neglect | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.80 (1.73-1.88) | 1.09 (1.08-1.10) | 2.05 (1.98-2.11) | 1.14 (1.14-1.15) | 3.18 (2.98-3.39) | 1.10 (1.09-1.10) |
| Parental divorce | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.35 (1.29-1.41) | 1.04 (1.04-1.05) | 1.52 (1.47-1.58) | 1.08 (1.07-1.09) | 1.77 (1.64-1.90) | 1.04 (1.03-1.05) |
| Household criminality | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.46 (1.31-1.63) | 1.06 (1.04-1.08) | 1.8 (1.66-1.95) | 1.13 (1.11-1.15) | 2.06 (1.76-2.42) | 1.06 (1.04-1.08) |
| Household mental illness | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 2.42 (2.24-2.63) | 1.20 (1.17-1.23) | 2.18 (2.03-2.34) | 1.20 (1.18-1.23) | 3.19 (2.79-3.64) | 1.14 (1.11-1.17) |
| Family financial problems | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.68 (1.58-1.79) | 1.09 (1.07-1.10) | 1.62 (1.54-1.71) | 1.10 (1.08-1.11) | 1.74 (1.56-1.94) | 1.04 (1.03-1.05) |
| Parental death | ||||||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 1.16 (1.05-1.29) | 1.02 (1.01-1.04) | 1.16 (1.06-1.27) | 1.03 (1.01-1.05) | 1.17 (0.98-1.40) | 1.01 (1.00-1.02) |
Abbreviations: ACE, adverse childhood experience; NSSI, nonsuicidal self-injury; PR, prevalence ratio.
Model 1 was adjusted for age, sex, current drinking status, current smoking status, household socioeconomic status, and academic pressure.
Model 2 was adjusted for the variables in model 1, plus supportive school environments.
A supportive school environment was negatively associated with NSSI and suicidality, even in the fully adjusted model 2. For example, a 1-unit increase in the supportive school environment score was associated with a lower prevalence of NSSI (PR, 0.75; 95% CI, 0.73-0.76) (Table 4). Similar results were observed across male and female subgroups (eTables 6-7 in Supplement 1).
Table 4. Associations of Supportive School Environment With NSSI and Suicidality.
| Supportive school environment outcomea | PR per 1-unit increase (95% CI) | |
|---|---|---|
| Model 1b | Model 2c | |
| NSSI | 0.70 (0.68-0.71) | 0.75 (0.73-0.76) |
| Suicidal ideation | 0.69 (0.68-0.70) | 0.74 (0.73-0.75) |
| Suicide attempt | 0.60 (0.59-0.62) | 0.66 (0.64-0.68) |
Abbreviations: NSSI, nonsuicidal self-injury; PR, prevalence ratio.
The supportive school environment is a continuous variable, which was evaluated using 3 questions. The total score ranges from 0 to 3, with higher scores suggesting more supportive school environments.
Model 1 was unadjusted.
Model 2 was adjusted for age, sex, current drinking status, current smoking status, household socioeconomic status, and academic pressure.
Moreover, we observed multiplicative interactions between ACEs and supportive school environments in the association with NSSI and suicidality, even in the fully adjusted model 2 (eg, interaction ratio for NSSI, 0.81, 95% CI, 0.76-0.88) (Table 5). Similar results were observed across male and female subgroups (eTables 8-9 in Supplement 1).
Table 5. Interactions Between ACEs and Supportive School Environment.
| Outcome | Model 1a | Model 2b |
|---|---|---|
| NSSI | ||
| PR per 1-unit increase (95% CI) | ||
| ACE (≥1 vs 0) | 2.10 (1.93-2.28) | 1.89 (1.74-2.05) |
| Supportive school environmentc | 0.71 (0.69-0.74) | 0.74 (0.71-0.76) |
| ACE × supportive school environment | 1.12 (1.07-1.17) | 1.13 (1.09-1.18) |
| Interaction ratio (95% C)d | 0.75 (0.73-0.80) | 0.81 (0.76-0.88) |
| Suicidal ideation | ||
| PR per 1-unit increase (95% CI) | ||
| ACE (≥1 vs 0) | 2.32 (2.15-2.51) | 2.07 (1.92-2.23) |
| Supportive school environmentc | 0.71 (0.69-0.74) | 0.73 (0.71-0.76) |
| ACE × supportive school environment | 1.13 (1.09-1.17) | 1.15 (1.11-1.19) |
| Interaction ratio (95% CI)d | 0.67 (0.62-0.73) | 0.78 (0.70-0.81) |
| Suicide attempt | ||
| PR per 1-unit increase (95% C) | ||
| ACE (≥1 vs 0) | 4.63 (3.89-5.52) | 3.85 (3.24-4.57) |
| Supportive school environmentc | 0.70 (0.64-0.75) | 0.70 (0.65-0.76) |
| ACE × supportive school environment | 1.02 (0.94-1.12) | 1.06 (0.98-1.16) |
| Interaction ratio (95% CI)d | 0.31 (0.27-0.37) | 0.39 (0.33-0.47) |
Abbreviations: ACE, adverse childhood experience; NSSI, nonsuicidal self-injury; PR, prevalence ratio.
Model 1 was unadjusted.
Model 2 was adjusted for age, sex, current drinking status, current smoking status, household socioeconomic status, and academic pressure.
The supportive school environment is a continuous variable, which was evaluated using 3 questions. The total score ranges from 0 to 3, with higher scores suggesting more supportive school environments.
Calculated as PRACE × supportive school environment/(PRACEs × PRsupportive school environment).
As shown in eTables 10 to 12 in Supplement 1, results remained consistent with main findings when using an NSSI threshold of 5 or more times. Additionally, after further adjusting for depressive symptoms, findings were generally similar to main results (eTables 13-16 in Supplement 1).
Discussion
This cross-sectional study found that exposure to ACEs, particularly threat-related ACEs, was independently associated with NSSI, SI, and SA among Chinese adolescents. Associations exhibited a dose-response pattern. Moreover, our study found that a supportive school environment was negatively associated with NSSI and suicidality, and it modified associations of ACEs with NSSI, SI, and SA.
In line with previous studies,33,45 we found that cumulative ACEs and each subtype of ACE were positively associated with NSSI, SI, and SA among adolescents. Similarly, a cross-sectional study conducted in the US27 found that compared with youths with no ACE exposure, middle school youths with 3 or more ACEs had a prevalence of suicidal behaviors that was 5.50 to 8.03 times higher, while high school youths with 3 or more ACEs had a prevalence that was 3.63 to 4.35 times higher. Although associations of ACEs with NSSI and suicidality have been consistently reported across various cultural contexts,13,46,47 limited research has explored the independent role of threat-related and deprivation-related ACEs in these outcomes in general adolescent populations. A previous cross-sectional study in a province of China13 reported that threat-related ACEs had associations with the greatest increases in odds of psychosocial difficulties. Another longitudinal study48 found a more pronounced increase in depression trajectories among individuals with threat-related ACEs than among individuals with other types of ACEs. Our study extends these findings by highlighting the specific association of threat-related ACEs, particularly emotional abuse, with NSSI, SI, and SA among adolescents. Although these findings should be validated through future cohort studies and randomized clinical trials, they highlight the potential for early identification and targeted interventions for adolescents exposed to threat-related ACEs, which may help reduce the risk of NSSI and suicidality.
Furthermore, our study found that certain types of ACEs under an expanded definition (eg, bullying) were associated with greater increases in risk of NSSI and suicidality than those observed with some conventional ACEs (eg, physical neglect). These findings suggest that the associated outcomes and significance of certain additional ACEs are substantial and warrant attention. A related study31 highlighted that if only conventional ACEs were considered, ACE exposure would have been underreported in 14.2% of included participants. Furthermore, the conventional ACE framework was originally developed based on a predominantly White, educated sample.17 Given that the perception and types of adversities may vary across different populations, there remains a need to develop culturally relevant ACE measurements. However, the exact underlying mechanisms in the association of ACEs with NSSI and suicidality remain unclear. One possible explanation is that early life adversity, especially including threat-related ACEs, may disrupt the development of brain regions involved in coping, self-regulation, and emotional management. Such disruption can result in emotional dysregulation, leading to maladaptive behaviors, including NSSI and suicidality, as a means of managing negative emotions.49
Previous research has established associations between social support and adolescent NSSI and suicidality.33,43,50 Our study extended this understanding by highlighting the association of support from school with risk of NSSI and suicidality among adolescents. Consistent with previous studies,51,52 our findings suggest that supportive school environments may be associated with reduced risk of adolescents engaging in NSSI and suicidality. Moreover, our study found that supportive school environments mitigated associations of ACEs with NSSI and suicidality, particularly among individuals with the highest level of ACE exposure. Interaction analysis suggest that supportive school environments could buffer the association of ACEs with these outcomes. These findings highlight the crucial role of the school environment in promoting adolescent mental health, which aligns with existing evidence.25,53 For instance, a 2021 cross-sectional study in the US27 found that school connectedness, an indicator of supportive school environments, moderated the association between ACEs and suicidal behaviors (eg, suicidal planning and suicide attempts). Similarly, a quasiexperimental study in the US54 revealed that individuals attending high-performing schools had lower rates of risky health behaviors, particularly substance use. Notably, another cross-sectional study26 suggested that students who perceived greater support from teachers were less likely to engage in NSSI. A supportive school environment is characterized by strong teacher-student relationships and a sense of belonging to the school. Theoretical frameworks suggest that in such environments, students are more likely to be engaged academically and feel a connection to the school community. This engagement is thought to foster the development of emotional and social skills and promote positive relationships, which, in turn, enhance mental health, buffer against adverse outcomes associated with ACEs, and reduce the risk of NSSI and suicidality.24,55
Strength and Limitations
The primary strength of our study is the use of a nationally representative sample of Chinese adolescents, which provides sufficient statistical power and broad generalizability, enabling a robust investigation of associations between ACEs and outcomes (ie, NSSI, SI, and SA), considering school environments. However, several limitations should be noted. First, owing to the cross-sectional nature of our study design, we can draw no causal conclusions concerning ACEs and NSSI and suicidality. Second, the reliance on self-reported data for ACEs, supportive school environments, NSSI, and suicidality may introduce the possibility of recall bias. Third, the study sample primarily consisted of school-attending adolescents, which may limit the generalizability of findings to all Chinese adolescents. Fourth, although we used a directed acyclic graph to identify a plausible set of confounders, our study may still not have accounted for all confounders, such as parenting style.
Conclusions
In this cross-sectional study of a nationally representative sample of school-based adolescents, we identified independent associations of cumulative and specific subtypes of ACEs with NSSI and suicidality among adolescents. A dose-response association was observed, with cumulative ACEs associated with increased risks of NSSI and suicidality. Notably, threat-related ACEs, particularly emotional abuse, exhibited associations with the highest increases in risk of these outcomes. Additionally, a supportive school environment was found to mitigate associations of ACEs with NSSI and suicidality. These findings underscore the association of ACEs and school environments with these outcomes and suggest that interventions targeting ACEs, along with school-level interventions, may be effective in reducing risks of NSSI and suicidality among adolescents.
eMethods. Supplemental methods
eFigure 1. Flowchart of participant selection
eFigure 2. Directed acyclic graph depicting the appropriate covariate set
eTable 1. Prevalence and 95% CIs of NSSI, suicidal ideation, and suicide attempts among study participants
eTable 2. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt in males
eTable 3. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt in females
eTable 4. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt in males
eTable 5. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt in females
eTable 6. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt in males
eTable 7. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt in females
eTable 8. Interactions between ACEs and supportive school environments in males
eTable 9. Interactions between ACEs and supportive school environments in females
eTable 10. Associations between cumulative ACEs and NSSI (defined as reporting 5 or more times)
eTable 11. Associations between each ACE subtype and NSSI (defined as reporting 5 or more times)
eTable 12. Interactions of ACEs and supportive school environments with NSSI (defined as reporting 5 or more times)
eTable 13. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 14. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 15. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 16. Interactions between ACEs and supportive school environments, further adjusted for depressive symptoms
eReferences.
Data Sharing Statement
References
- 1.Biswas T, Scott JG, Munir K, et al. Global variation in the prevalence of suicidal ideation, anxiety and their correlates among adolescents: a population based study of 82 countries. EClinicalMedicine. 2020;24:100395. doi: 10.1016/j.eclinm.2020.100395 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Moloney F, Amini J, Sinyor M, Schaffer A, Lanctôt KL, Mitchell RHB. Sex differences in the global prevalence of nonsuicidal self-injury in adolescents: a meta-analysis. JAMA Netw Open. 2024;7(6):e2415436. doi: 10.1001/jamanetworkopen.2024.15436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mars B, Heron J, Klonsky ED, et al. Predictors of future suicide attempt among adolescents with suicidal thoughts or non-suicidal self-harm: a population-based birth cohort study. Lancet Psychiatry. 2019;6(4):327-337. doi: 10.1016/S2215-0366(19)30030-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jin MK, Wang XY, Wang RX, et al. A systematic review and meta-analysis of factors related to non-suicidal self-injury among Chinese adolescents. Psychiatry Res. 2023;326:115329. doi: 10.1016/j.psychres.2023.115329 [DOI] [PubMed] [Google Scholar]
- 5.Miu AC, Szentágotai-Tătar A, Balázsi R, Nechita D, Bunea I, Pollak SD. Emotion regulation as mediator between childhood adversity and psychopathology: a meta-analysis. Clin Psychol Rev. 2022;93:102141. doi: 10.1016/j.cpr.2022.102141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shapero BG, Steinberg L. Emotional reactivity and exposure to household stress in childhood predict psychological problems in adolescence. J Youth Adolesc. 2013;42(10):1573-1582. doi: 10.1007/s10964-013-9954-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hughes K, Bellis MA, Hardcastle KA, et al. The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. Lancet Public Health. 2017;2(8):e356-e366. doi: 10.1016/S2468-2667(17)30118-4 [DOI] [PubMed] [Google Scholar]
- 8.Chan KL. Victimization and poly-victimization among school-aged Chinese adolescents: prevalence and associations with health. Prev Med. 2013;56(3-4):207-210. doi: 10.1016/j.ypmed.2012.12.018 [DOI] [PubMed] [Google Scholar]
- 9.Fang X, Fry DA, Ji K, et al. The burden of child maltreatment in China: a systematic review. Bull World Health Organ. 2015;93(3):176-85C. doi: 10.2471/BLT.14.140970 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chen Z, Li J, Liu J, Liu X. Adverse childhood experiences, recent negative life events, and non-suicidal self-injury among Chinese college students: the protective role of self-efficacy. Child Adolesc Psychiatry Ment Health. 2022;16(1):97. doi: 10.1186/s13034-022-00535-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Laporte N, Ozolins A, Westling S, Westrin Å, Wallinius M. Adverse childhood experiences as a risk factor for non-suicidal self-injury and suicide attempts in forensic psychiatric patients. BMC Psychiatry. 2023;23(1):238. doi: 10.1186/s12888-023-04724-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Guo S, Jiao T, Ma Y, et al. Association of adverse childhood experiences with non-suicidal self-injury and suicidality: baseline survey of the Chinese Adolescent Health Growth Cohort. Future. 2023;1(3):76-92. doi: 10.3390/future1030009 [DOI] [Google Scholar]
- 13.Li Y, Shang Y, Yang Y, et al. Association of childhood adversities with psychosocial difficulties among Chinese children and adolescents. Int J Epidemiol. 2023;52(6):1887-1897. doi: 10.1093/ije/dyad117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McCrory E, De Brito SA, Viding E. The link between child abuse and psychopathology: a review of neurobiological and genetic research. J R Soc Med. 2012;105(4):151-156. doi: 10.1258/jrsm.2011.110222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McLaughlin KA, Weissman D, Bitrán D. Childhood adversity and neural development: a systematic review. Annu Rev Dev Psychol. 2019;1:277-312. doi: 10.1146/annurev-devpsych-121318-084950 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Anda RF, Croft JB, Felitti VJ, et al. Adverse childhood experiences and smoking during adolescence and adulthood. JAMA. 1999;282(17):1652-1658. doi: 10.1001/jama.282.17.1652 [DOI] [PubMed] [Google Scholar]
- 17.Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4):245-258. doi: 10.1016/S0749-3797(98)00017-8 [DOI] [PubMed] [Google Scholar]
- 18.Pace CS, Muzi S, Rogier G, Meinero LL, Marcenaro S. The Adverse Childhood Experiences-International Questionnaire (ACE-IQ) in community samples around the world: a systematic review (part I). Child Abuse Negl. 2022;129:105640. doi: 10.1016/j.chiabu.2022.105640 [DOI] [PubMed] [Google Scholar]
- 19.Cronholm PF, Forke CM, Wade R, et al. Adverse childhood experiences: expanding the concept of adversity. Am J Prev Med. 2015;49(3):354-361. doi: 10.1016/j.amepre.2015.02.001 [DOI] [PubMed] [Google Scholar]
- 20.Finkelhor D, Shattuck A, Turner H, Hamby S. Improving the adverse childhood experiences study scale. JAMA Pediatr. 2013;167(1):70-75. doi: 10.1001/jamapediatrics.2013.420 [DOI] [PubMed] [Google Scholar]
- 21.Lambert HK, King KM, Monahan KC, McLaughlin KA. Differential associations of threat and deprivation with emotion regulation and cognitive control in adolescence. Dev Psychopathol. 2017;29(3):929-940. doi: 10.1017/S0954579416000584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Longhi D, Brown M, Fromm Reed S. Community-wide resilience mitigates adverse childhood experiences on adult and youth health, school/work, and problem behaviors. Am Psychol. 2021;76(2):216-229. doi: 10.1037/amp0000773 [DOI] [PubMed] [Google Scholar]
- 23.Wang W, Liu Y, Yang Y, et al. Adverse childhood and adulthood experiences and risk of new-onset cardiovascular disease with consideration of social support: a prospective cohort study. BMC Med. 2023;21(1):297. doi: 10.1186/s12916-023-03015-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bonell CP, Fletcher A, Jamal F, et al. Theories of how the school environment impacts on student health: systematic review and synthesis. Health Place. 2013;24:242-249. doi: 10.1016/j.healthplace.2013.09.014 [DOI] [PubMed] [Google Scholar]
- 25.Leurent B, Dodd M, Allen E, Viner R, Scott S, Bonell C. Is positive school climate associated with better adolescent mental health: longitudinal study of young people in England. SSM Ment Health. 2021;1:100033. doi: 10.1016/j.ssmmh.2021.100033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Madjar N, Ben Shabat S, Elia R, et al. Non-suicidal self-injury within the school context: multilevel analysis of teachers’ support and peer climate. Eur Psychiatry. 2017;41:95-101. doi: 10.1016/j.eurpsy.2016.11.003 [DOI] [PubMed] [Google Scholar]
- 27.Lensch T, Clements-Nolle K, Oman RF, Evans WP, Lu M, Yang W. Adverse childhood experiences and suicidal behaviors among youth: the buffering influence of family communication and school connectedness. J Adolesc Health. 2021;68(5):945-952. doi: 10.1016/j.jadohealth.2020.08.024 [DOI] [PubMed] [Google Scholar]
- 28.Lai W, Li W, Guo L, et al. Association between bullying victimization, coping style, and mental health problems among Chinese adolescents. J Affect Disord. 2023;324:379-386. doi: 10.1016/j.jad.2022.12.080 [DOI] [PubMed] [Google Scholar]
- 29.Zheng X, Jiang W, Peng S, et al. Childhood adversity and adolescent smartphone use across sexual orientation and gender expression. JAMA Netw Open. 2024;7(4):e246448. doi: 10.1001/jamanetworkopen.2024.6448 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Guo L, Wang W, Gao X, Huang G, Li P, Lu C. Associations of childhood maltreatment with single and multiple suicide attempts among older Chinese adolescents. J Pediatr. 2018;196:244-250.e1. doi: 10.1016/j.jpeds.2018.01.032 [DOI] [PubMed] [Google Scholar]
- 31.Lin L, Cao B, Chen W, Li J, Zhang Y, Guo VY. Association of adverse childhood experiences and social isolation with later-life cognitive function among adults in China. JAMA Netw Open. 2022;5(11):e2241714. doi: 10.1001/jamanetworkopen.2022.41714 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Qu D, Wang Y, Zhang Z, et al. Psychometric properties of the Chinese version of the Functional Assessment of Self-Mutilation (FASM) in Chinese clinical adolescents. Front Psychiatry. 2022;12:755857. doi: 10.3389/fpsyt.2021.755857 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Wan Y, Chen R, Ma S, et al. Associations of adverse childhood experiences and social support with self-injurious behaviour and suicidality in adolescents. Br J Psychiatry. 2019;214(3):146-152. doi: 10.1192/bjp.2018.263 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Jiao T, Guo S, Zhang Y, et al. Associations of depressive and anxiety symptoms with non-suicidal self-injury and suicidal attempt among Chinese adolescents: the mediation role of sleep quality. Front Psychiatry. 2022;13:1018525. doi: 10.3389/fpsyt.2022.1018525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lai W, Wu H, Yang L, et al. Prevalence of unhealthy behaviors and their associations with non-suicidal self-injury, suicidal ideation and suicide attempt among Chinese adolescents. Child Adolesc Psychiatry Ment Health. 2024;18(1):61. doi: 10.1186/s13034-024-00742-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Guo L, Xu Y, Deng J, et al. Association between nonmedical use of prescription drugs and suicidal behavior among adolescents. JAMA Pediatr. 2016;170(10):971-978. doi: 10.1001/jamapediatrics.2016.1802 [DOI] [PubMed] [Google Scholar]
- 37.Pulimeno M, Piscitelli P, Colazzo S, Colao A, Miani A. School as ideal setting to promote health and wellbeing among young people. Health Promot Perspect. 2020;10(4):316-324. doi: 10.34172/hpp.2020.50 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Med Res Methodol. 2008;8:70. doi: 10.1186/1471-2288-8-70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Bell BA, Onwuegbuzie AJ, Ferron JM, Jiao QG, Hibbard ST, Kromrey JD. Use of design effects and sample weights in complex health survey data: a review of published articles using data from 3 commonly used adolescent health surveys. Am J Public Health. 2012;102(7):1399-1405. doi: 10.2105/AJPH.2011.300398 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.VanderWeele TJ. The interaction continuum. Epidemiology. 2019;30(5):648-658. doi: 10.1097/EDE.0000000000001054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.VanderWeele TJ, Knol MJ. A tutorial on interaction. Epidemiol Methods. 2014;3(1):33-72. doi: 10.1515/em-2013-0005 [DOI] [Google Scholar]
- 42.Knol MJ, VanderWeele TJ. Recommendations for presenting analyses of effect modification and interaction. Int J Epidemiol. 2012;41(2):514-520. doi: 10.1093/ije/dyr218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Ma Y, Li Y, Xie X, et al. The role of depressive symptoms and social support in the association of internet addiction with non-suicidal self-injury among adolescents: a cohort study in China. BMC Psychiatry. 2023;23(1):322. doi: 10.1186/s12888-023-04754-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Morrow EL, Duff MC, Mayberry LS. Mediators, moderators, and covariates: matching analysis approach for improved precision in cognitive-communication rehabilitation research. J Speech Lang Hear Res. 2022;65(11):4159-4171. doi: 10.1044/2022_JSLHR-21-00551 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Shi X, Meng Y, Cheng B, et al. Association between traumatic events with suicidality among adolescents: a large-scale cross-sectional study of 260,423 participants. Psychiatry Res. 2024;333:115762. doi: 10.1016/j.psychres.2024.115762 [DOI] [PubMed] [Google Scholar]
- 46.Anderson KN, Swedo EA, Trinh E, et al. Adverse childhood experiences during the COVID-19 pandemic and associations with poor mental health and suicidal behaviors among high school students—Adolescent Behaviors and Experiences Survey, United States, January-June 2021. MMWR Morb Mortal Wkly Rep. 2022;71(41):1301-1305. doi: 10.15585/mmwr.mm7141a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Liu RT, Scopelliti KM, Pittman SK, Zamora AS. Childhood maltreatment and non-suicidal self-injury: a systematic review and meta-analysis. Lancet Psychiatry. 2018;5(1):51-64. doi: 10.1016/S2215-0366(17)30469-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Bevilacqua L, Kelly Y, Heilmann A, Priest N, Lacey RE. Adverse childhood experiences and trajectories of internalizing, externalizing, and prosocial behaviors from childhood to adolescence. Child Abuse Negl. 2021;112:104890. doi: 10.1016/j.chiabu.2020.104890 [DOI] [PubMed] [Google Scholar]
- 49.Wolff JC, Thompson E, Thomas SA, et al. Emotion dysregulation and non-suicidal self-injury: a systematic review and meta-analysis. Eur Psychiatry. 2019;59:25-36. doi: 10.1016/j.eurpsy.2019.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Tordoff DM, Wanta JW, Collin A, Stepney C, Inwards-Breland DJ, Ahrens K. Mental health outcomes in transgender and nonbinary youths receiving gender-affirming care. JAMA Netw Open. 2022;5(2):e220978. doi: 10.1001/jamanetworkopen.2022.0978 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Shim-Pelayo H, De Pedro KT. The role of school climate in rates of depression and suicidal ideation among school-attending foster youth in California public schools. Child Youth Serv Rev. 2018;88:149-155. doi: 10.1016/j.childyouth.2018.02.033 [DOI] [Google Scholar]
- 52.Yu C, Xie Q, Lin S, et al. Cyberbullying victimization and non-suicidal self-injurious behavior among Chinese adolescents: school engagement as a mediator and sensation seeking as a moderator. Front Psychol. 2020;11:572521. doi: 10.3389/fpsyg.2020.572521 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Hatzenbuehler ML, Birkett M, Van Wagenen A, Meyer IH. Protective school climates and reduced risk for suicide ideation in sexual minority youths. Am J Public Health. 2014;104(2):279-286. doi: 10.2105/AJPH.2013.301508 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Dudovitz RN, Chung PJ, Reber S, et al. Assessment of exposure to high-performing schools and risk of adolescent substance use: a natural experiment. JAMA Pediatr. 2018;172(12):1135-1144. doi: 10.1001/jamapediatrics.2018.3074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Markham WA, Aveyard P. A new theory of health promoting schools based on human functioning, school organisation and pedagogic practice. Soc Sci Med. 2003;56(6):1209-1220. doi: 10.1016/S0277-9536(02)00120-X [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods. Supplemental methods
eFigure 1. Flowchart of participant selection
eFigure 2. Directed acyclic graph depicting the appropriate covariate set
eTable 1. Prevalence and 95% CIs of NSSI, suicidal ideation, and suicide attempts among study participants
eTable 2. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt in males
eTable 3. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt in females
eTable 4. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt in males
eTable 5. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt in females
eTable 6. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt in males
eTable 7. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt in females
eTable 8. Interactions between ACEs and supportive school environments in males
eTable 9. Interactions between ACEs and supportive school environments in females
eTable 10. Associations between cumulative ACEs and NSSI (defined as reporting 5 or more times)
eTable 11. Associations between each ACE subtype and NSSI (defined as reporting 5 or more times)
eTable 12. Interactions of ACEs and supportive school environments with NSSI (defined as reporting 5 or more times)
eTable 13. Associations of cumulative ACEs with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 14. Associations of each ACE subtype with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 15. Associations of supportive school environments with NSSI, suicidal ideation, and suicide attempt, further adjusted for depressive symptoms
eTable 16. Interactions between ACEs and supportive school environments, further adjusted for depressive symptoms
eReferences.
Data Sharing Statement
