[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Evaluation of the Early Development of 6-Month-Old Babies in the Case of Maternal Postpartum Depression with or Without Bipolar Disorder
Previous Article in Journal
Efficacy of Health Literacy Interventions for Caregivers of Individuals with Neurodevelopmental and Chronic Conditions: A Rapid Review
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Children
Volume 12
Issue 1
10.3390/children12010010
children-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Association Between Adolescent Violence Exposure and the Risk of Suicide: A 15-Year Study in Taiwan

by
Chieh Sung
1,2,
Chi-Hsiang Chung
3,4,
Chien-An Sun
5,6,
Chang-Huei Tsao
1,7,
Daphne Yih Ng
8,
Tsu-Hsuan Weng
1,9,
Li-Yun Fann
10,
Fu-Huang Lin
3 and
Wu-Chien Chien
1,3,*
1
Department of Medical Research, Tri-Service General Hospital, National Defense Medical Center, Taipei 11490, Taiwan
2
Department of Senior Citizen Care and Welfare, Deh Yu College of Nursing and Health, Keelung 203, Taiwan
3
School of Public Health, National Defense Medical Center, Taipei 11490, Taiwan
4
Taiwanese Injury Prevention and Safety Promotion Association, Taipei 11490, Taiwan
5
Department of Public Health, College of Medicine, Fu-Jen Catholic University, New Taipei City 24205, Taiwan
6
Big Data Research Center, College of Medicine, Fu-Jen Catholic University, New Taipei City 24205, Taiwan
7
Department of Microbiology & Immunology, National Defense Medical Center, Taipei 11490, Taiwan
8
Department of Family Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei 11490, Taiwan
9
Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei 11490, Taiwan
10
Department of Nursing, Taipei City Hospital, Taipei 10684, Taiwan
*
Author to whom correspondence should be addressed.
Children 2025, 12(1), 10; https://doi.org/10.3390/children12010010 (registering DOI)
Submission received: 27 November 2024 / Revised: 16 December 2024 / Accepted: 23 December 2024 / Published: 24 December 2024
(This article belongs to the Section Global Pediatric Health)
Browse Figures
Figure 1
<p>Flowchart of the study.</p> "> Figure 2
<p>Kaplan–Meier for cumulative risk of suicide aged 10–18, stratified by violence with log-rank test.</p> "> Figure 3
<p>Joint effect for factors of suicide stratified by violence and mental disorders using Cox regression.</p> ">
Versions Notes

Abstract

:
Background/Objectives: According to the 2023 Ministry of Health and Welfare statistics, the suicide rate among adolescents aged 15 to 24 has steadily increased since 2018, from 3.7 to 5.5 per 100,000 populations, reaching a recent high. Although previous studies have pointed out that the future risk of suicide of those who had suffered from abuse was higher than that of the general population, researchers seldom focused on adolescent groups. Therefore, the aim of this study was to explore the risk of suicide after youth violence and the impact of subsequent comorbid mental illness and suicide risk. Methods: This retrospective matched cohort study analyzed data from the NHIRD, covering the period from 2000 to 2015. A total of 976 cases aged 10–18 who had experienced violence were included in this study. Controlled grouping was conducted by 1:10 matching based on gender, age, and the time of medical treatment, and a control group who had not experienced violence was selected for comparison. We used the Cox proportional hazards model to analyze the risk of suicide among adolescents after exposure to violence. Results: The suicide rate among adolescents who have experienced violence was significantly higher than that of the control group after 15 years of follow-up (1.0% vs. 0.5%). The prevalence of mental illness or disorders in adolescents exposed to violence was significantly higher than in the control group (45.2% vs. 40.1%). Among adolescents who had experienced violence, the methods of suicide included poisoning (solid and liquid) (53.6% vs. 43.2%), hanging (1.2% vs. 0.6%), firearms (2.4% vs. 0%), and cutting instruments (27.4% vs. 22.8%), all of which were significantly higher than in the control group. After adjusting for gender, age, residential area, and mental health comorbidities, the risk of suicide in those who had experienced violence was 1.475 times that of the control group (95% CI = 1.125–1.933; p = 0.005). Conclusions: In this study, female, younger age, and comorbid mental disorders were identified as risk factors for suicide among the adolescent victims of violence. Exposure to youth violence was associated with an increased prevalence of emotional disorders, including depression and social isolation, which subsequently elevated the suicide risk. These findings underscore the urgent need for governmental attention to the mental health of adolescent victims of violence. Implementing targeted psychological support and intervention programs could play a crucial role in mitigating the risk of suicide among this vulnerable population.

1. Introduction

Adolescent violence remains a pervasive global concern, defined by the physical maltreatment of children by parents or adult household members through acts such as hitting, pushing, choking, shaking, throwing, biting, and burning. These harmful behaviors can lead to bruises or more severe physical injuries. The World Health Organization [1] reports that approximately one-quarter of all adults worldwide have experienced violence during childhood. In the United States, the Department of Health and Human Services estimated that annually between 700,000 and 1.25 million children are subjected to violence or neglect [2]. Similarly, Taiwan’s Ministry of Health and Welfare reported that from 2004 to 2018, between 4000 and 19,000 children experienced violence or neglect each year [3].
Preventing youth suicide is an urgent public health imperative. Suicide is the second leading cause of death among individuals aged 15 to 24, with reported cases of suicidal thoughts and behaviors increasing over the past decade [4,5]. Evidence from self-reported data and clinical assessments indicates that maltreated youth are significantly more likely to contemplate and attempt suicide [6,7]. A recent meta-analysis revealed that young people who have experienced any form of child violence or neglect are 2.91 times more likely to attempt suicide and 2.36 times more likely to experience suicidal ideation compared to their non-maltreated peers [6]. The high prevalence of child maltreatment in the United States amplifies concerns about its impact on youth suicide rates. In 2019, over 3.4 million children were involved in child maltreatment investigations in the U.S. [8]. Furthermore, estimates suggest that 37.4% of U.S. youth will be involved in such investigations by the age of 18 [9]. Globally, the World Health Organization (2020) [10] estimates that approximately one billion children aged two to seventeen experience violence, including child maltreatment, each year. These alarming statistics underscore the critical need for effective suicide prevention strategies targeting maltreated youth worldwide.
However, the relationship between adolescent violence and suicide risk has not been thoroughly investigated. We hypothesized that adolescents who have experienced violence are at a higher risk of future suicide. Therefore, we utilized the National Health Insurance Research Database (NHIRD) to examine whether adolescents exposed to violence were at an increased risk of suicide between 2000 and 2015 in Taiwan.

2. Materials and Methods

2.1. Data Sources

In this study, we analyzed data from the Taiwan National Health Insurance Research Database (NHIRD) to explore the link between adolescent exposure to violence and the risk of suicide over a 15-year span. Information on adolescent violence incidents was obtained from outpatient and inpatient records in the Taiwan Longitudinal Health Insurance Database for the study duration (2000–2015). The benefits, drawbacks, and specifics of the NHIRD have been discussed in previous studies [11].
The National Health Insurance Program (NHIP) in Taiwan was launched in 1995. As of June 2009, it had contracts with 97% of the nation’s healthcare providers, covering around 23 million beneficiaries—more than 99% of Taiwan’s population [12]. The NHIRD utilizes the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes to record diagnoses [13].
All diagnoses of adolescent violence were made by pediatricians or emergency medicine physicians based on clinical findings. Furthermore, licensed medical record technicians reviewed and verified the diagnostic codes before the claim for reimbursement to the hospital was approved. According to Taiwan’s Protection of Children and Youths Welfare and Rights Act (2003) [14], clinicians who detect signs or symptoms of child and adolescent violence are mandated to report their findings to the authorized municipal or county agencies within 24 h. Clinicians must exercise meticulous care when diagnosing adolescent violence with the corresponding ICD-9-CM codes to avoid legal repercussions [15]. The diagnoses of psychiatric disorders were made by board-certified psychiatrists, according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (both original and revised editions) [16].
Additionally, certified medical record technicians review and confirm diagnostic codes before hospital reimbursement claims are processed [14]. The NHIP administration also performs random audits on outpatient records and periodically reviews inpatient claims to maintain diagnostic accuracy [16]. As a result, the data from the NHIRD are regarded as trustworthy. Thus, we used the NHIRD data to examine the relationship between adolescent violence and the occurrence of suicide.

2.2. Study Design and Sample

This study utilized a retrospective matched cohort design. Adolescents diagnosed with violence, between 1 January 2000, and 31 December 2015, were included in the adolescent violence cohort (n = 976). In addition, 29,511 controls with no history of adolescent violence during the study period were matched for age, gender, and index year at a ratio of 1:10 to the adolescent violence cohort. Participants over the age of 18 or under 10 years, and those with prior records of adolescent violence or suicide before the index date, were excluded (Figure 1).

2.3. Major Outcome Measure

This study aimed to evaluate the association between adolescent violence and the risk of suicide. All participants were followed from 1 January 2000, until the occurrence of suicide, withdrawal from the NHIP, or 31 December 2015.
In this study, the study population includes individuals who have experienced violence (ICD-9-CM codes 955.5 and E967) and suicide events (ICD-9-CM codes E950–E958).
Category 995.5 includes the following: child abuse, unspecified (995.50); child emotional/psychological abuse (995.51); child neglect (995.52); child sexual abuse (995.53); child physical abuse (995.54); shaken infant syndrome (995.55); and other child abuse and neglect (995.59). Category E967 includes perpetrators of child and adult abuse as follows: by father, stepfather, or boyfriend (E967.0); by other specified person (E967.1); by mother, stepmother, or girlfriend (E967.2); by spouse or partner (E967.3); by child (E967.4); by sibling (E967.5); by grandparent (E967.6); by other relative (E967.7); by non-related caregiver (E967.8); and by an unspecified person (E967.9). The date of the first diagnosed adolescent violence case was treated as the index date.
Suicide methods are categorized as follows: solid or liquid substances (E950), domestic gasses (E951), other gasses and vapors (E952), hanging (E953), drowning (E954), firearms (E955), cutting and piercing (E956), jumping (E957), and other unspecified methods (E958). All diagnoses were determined by certified clinicians based on clinical judgment. Additionally, the comorbidities of interest in this population are mental disorders, as identified by ICD-9-CM codes 290–319. All mental disorders diagnoses were made by certified psychiatrists and in accordance with the DSM-V criteria. All detailed ICD codes are provided in Supplemental Table S1.

2.4. Variables

Covariates incorporated into the statistical analyses encompassed a comprehensive range of demographic, socioeconomic, and clinical factors. Demographic variables included gender and age, while geographic variables covered the region of residence classified such as northern, central, southern, or eastern Taiwan and the urbanization level of the residential area, stratified into four tiers based on population density and development indicators. The type of healthcare facility accessed by participants was considered, categorized as medical centers, regional hospitals, or local hospitals. Socioeconomic status was evaluated by determining whether participants belonged to low-income households. Clinical covariates included the presence of catastrophic illnesses and mental disorders, identified through the relevant diagnostic codes. Additionally, the season during which data were collected, spring, summer, autumn, or winter, was included to control for potential seasonal variations affecting health outcomes. The urbanization level was defined according to the population size and the various indicators of development. Urbanization level 1 was defined as areas with a population of over 1,250,000 inhabitants with a specific designation of political, economic, cultural, and metropolitan development. Urbanization level 2 was defined as areas with a population between 500,000 and 1,249,999 inhabitants, playing an important role in the political system, economy, and culture. Urbanization levels 3 and 4 were defined as areas with populations of 149,999–499,999 and less than 149,999 inhabitants, respectively [17]. Comorbidities included mental disorders. The Charlson comorbidity index (CCI) is one of the most widely used comorbidity indexes [18,19], which consists of 22 conditions [20]. The score was calculated based on the presence of the relevant comorbidities (according to the ICD-9-CM codes) [21], with a score of zero indicating the absence of comorbidities and higher scores indicating a higher comorbidity burden [22].

2.5. Statistical Analysis

All statistical analyses were conducted using SPSS software version 29 (SPSS Inc., Chicago, IL, USA). Chi-square (χ2) and Mann–Whitney U test were employed to assess the distributions of categorical and continuous variables, respectively. Fisher’s exact test was used to assess differences between the two cohorts with respect to categorical variables. The associations between time-to-event outcomes and clinical characteristics were examined using the Kaplan–Meier method and multivariate Cox regression analysis with stepwise selection; the results are reported as hazard ratios (HR) and 95% confidence intervals (CI). Adjustments were made for age, gender, and covariates for inclusion in the multivariate model. Bonferroni correction for multiple comparisons was performed. A two-tailed Bonferroni-corrected p value < 0.05 was regarded as statistically significant.
The factors influencing the different suicide subgroups were analyzed using Cox proportional hazards regression models. A Bonferroni correction was applied to adjust for multiple comparisons. The Cox model estimates hazard ratios for each factor while controlling for potential confounders. The Bonferroni correction adjusts p-values by dividing the significance level by the number of comparisons to control for type I error. The results are presented as follows for each subgroup: solid or liquid substances, gasses in domestic use, other gasses and vapors, hanging, drowning, firearms, cutting and piercing, jumping, and others.

2.6. Ethics Approvals

This research was carried out in compliance with the World Medical Association’s Code of Ethics (Declaration of Helsinki). This study was approved by the Institutional Review Board of Tri-Service General Hospital at the National Defense Medical Center in Taipei, Taiwan, and the requirement of individual consent was waived because all identifying data were encrypted (TSGHIRB No. B202405024).The NHIRD is a publicly available database that contains depersonalized patient information to ensure patient anonymity.

3. Results

3.1. Baseline Characteristics

The demographic and clinical characteristics of both groups are summarized in Table 1. We identified 909 adolescents with a documented history of violence and selected 9090 matched controls without such a history. The mean age of adolescents in the violence cohort was 14.18 ± 4.72 years, with a higher proportion of females than males. Overall, the sample was predominantly female (97.47%), with only 2.53% of participants identifying as male. Furthermore, the distribution of gender remained virtually identical across both the violence and non-violence cohorts, each consisting of 2.53% males and 97.47% females. Notably, significant differences between the violence and control cohorts were observed in terms of geographical location and urbanization levels.

3.2. Characteristics of the Study Population at Endpoint

By the end of the study period, 92 out of the 909 adolescents who had experienced violence (10.12%) died by suicide, compared to 198 out of the 9090 individuals in the control group (2.18%), demonstrating a statistically significant difference (p < 0.001; Table 2). Significant disparities were observed between the violence-exposed and control groups when comparing geographical location and urbanization level. In contrast, there were no significant differences between the two groups regarding gender, age, low-income household status, presence of catastrophic illnesses, mental disorders, CCI scores, season, or the level of care. The comprehensive data are presented in Table 2.

3.3. Risk of Suicide According to Adolescent Violence Exposure

The Kaplan–Meier survival analysis revealed that adolescents with a history of violence had a significantly higher cumulative incidence of suicide over the 15-year follow-up period compared to the matched control group (log-rank test, p < 0.001; Figure 2). The results from Tables S2 and S3 show no significant difference in follow-up years between adolescents with and without a history of violence (p > 0.05). However, in terms of years to suicide, there was a significant difference between the two groups, with a p-value of 0.031, suggesting that exposure to violence may lead to an earlier occurrence of suicide. These findings indicate that while exposure to violence does not affect the length of follow-up, it significantly impacts the age at which suicide occurs. This underscores the importance of early intervention for adolescents who have experienced violence to reduce suicide risk. The relationship between gender, violence, and psychological behavior is complex and warrants further research.

3.4. Factors of Suicide Using Cox Regression

Table 3 presents the results of the Cox proportional hazards regression analysis examining factors associated with suicide risk among adolescents who have experienced violence. The unadjusted hazard ratio (HR) for suicide in the violence cohort was 1.787 (95% CI: 1.246–2.033; p < 0.001), indicating a significantly elevated risk compared to the control group. After adjusting for multiple covariates, including gender, age group, low-income household status, presence of catastrophic illness, mental disorders, CCI score, season, geographic location, urbanization level, and the level of care, the association remained significant. The adjusted hazard ratio (aHR) was 1.592 (95% CI: 1.137–1.993; p < 0.001), suggesting that adolescent violence is independently associated with an increased risk of suicide. Several covariates were also significantly correlated with suicide risk. The crude HR for females was 2.098 (95% CI: 1.358–2.886, p < 0.001), indicating that prior to adjusting for other factors, the risk of suicide among females was approximately twice that of males. After adjusting for potential confounders, the aHR remained statistically significant at 1.523 (95% CI: 1.072–1.831, p = 0.012), suggesting that females continue to exhibit a significantly higher risk of suicide compared to males, even after accounting for other influencing variables. Additionally, the presence of mental disorders and higher levels of care were associated with an increased incidence of suicide (p < 0.05).

3.5. Factors of Suicide Stratified by Variables Listed in the Table Using Cox Regression and Bonferroni Correction for Multiple Comparisons

The patients were stratified by the variables presented in Table 3, and the adjusted hazard ratios of the different subgroups were calculated (Table 4). Over the course of the study, adolescents who had experienced violence exhibited 92 suicide events over 7074.36 person-years (PYs) of observation, resulting in an incidence rate of 1300.47 per 100,000 PYs. In contrast, the control group encountered 198 suicide events over 69,894.12 PYs, corresponding to an incidence rate of 283.29 per 100,000 PYs. After applying the Bonferroni correction for multiple comparisons, the risk of suicide was significantly higher among adolescents with a history of violence compared to those without such a history. The aHR was 1.592 (95% CI: 1.137–1.993; p < 0.001), indicating that adolescents affected by violence had nearly a 1.5 times increased risk of suicide. When stratified by gender, male exposure to violence showed a significantly elevated suicide risk (aHR = 1.500, 95% CI: 1.071–1.872, p = 0.015). Among females, the risk was even higher (aHR = 1.596, 95% CI: 1.401–1.998, p < 0.001). These findings underscore the heightened impact of violence on suicide risk, particularly among females, highlighting the need for targeted interventions for this high-risk group. Notably, the presence of mental disorders and higher levels of care were significantly associated with an increased incidence of suicide, the aHR was 2.369 (95% CI: 2.369–2.972; p < 0.001).

3.6. Factors of Suicide Subgroups Using Cox Regression and Bonferroni Correction for Multiple Comparisons

Adolescents exposed to violence exhibited a significantly elevated risk of suicide across various methods. Specifically, the adjusted hazard ratios (AHRs) for different suicide methods were as follows: ingestion of solids or liquids (AHR = 1.607), exposure to other gasses and vapors (AHR = 1.714), hanging (AHR = 2.058), cutting and piercing (AHR = 1.656), and jumping (AHR = 1.523) (Table 5).

3.7. Factors of Suicide Stratified by Violence and Mental Disorders Using Cox Regression

The analysis of suicide risk factors using the Cox regression model indicates a significant impact of both violence and mental disorders on suicide risk. In the reference group, which included individuals without a history of violence or mental disorders, the aHR was set at 1.000. In comparison, individuals without a history of violence but with mental disorders exhibited a significantly elevated risk of suicide, with an aHR of 1.465 (95% CI: 1.172–1.779, p < 0.001). Among those who experienced violence but did not have mental disorders, the suicide risk further increased, with an aHR of 1.756 (95% CI: 1.340–2.075, p < 0.001). Furthermore, individuals who experienced both violence and had mental disorders showed the highest risk of suicide, with an aHR of 3.586 (95% CI: 2.781–4.986, p < 0.001). Additionally, the interaction between violence and mental disorders was significant, as indicated by the p for interaction value (p < 0.001), highlighting a notable synergistic effect that further elevated the risk of suicide (Table 6, Figure 3).

4. Discussion

This study examined the relationship between adolescent violence and suicidal behaviors, assessing whether all forms of adolescent violence were associated with an increased risk of suicidality in both univariate logistic regression models and multivariable logistic regression models that controlled for covariates. In our study, we found that the suicide rate of adolescents who have experienced violence was significantly higher than that of the control group (10.12% vs. 2.18%; p < 0.001), with an aHR of 1.592 after 15 years of observation. The result is consistent with Zygo et al. [23], which demonstrated that psychological, physical violence, and family violence were all risk factors not only for suicide ideation but also for suicide attempt and even suicide death. The findings of this study align with those of numerous other investigations [24,25,26]. However, an alternative study suggests that emotional violence is the most significant predictor of suicide attempts, with physical violence following closely behind [27]. Several theories may account for the relationship between adolescent violence and suicidality. The Schematic Appraisals Model for Suicide posits that negative childhood experiences can foster a growing sense of self-defeat, ultimately leading individuals to perceive suicidality as an escape mechanism [28,29]. In a study from America [23], the peak age of suicide in rural or urban area was 16 to 18 years old, which was consistent with our research. Both the studies reminded us to closely monitor and observe the mental and psychologic status of the youths in this period.
Thus, concerns and caution should rise keenly for the adolescents in some situations, such as unusual wounds or bruises, rapid mood change, weird behavior, abnormal vaginal bleeding, and relationship with schoolmates. The methods and tools for adolescent suicide may be different in distinct countries or cultures. In our study, the most common methods were the consumption of liquid or solid components in both the group with/without being the victims of violence, the method of suicide is self-hanging, followed by firearms, self-burning, and self-poisoning. Thus, it is important for different countries to formulate policies exclusively, such as gun control, chemicals restriction for less accessibility, or the prohibition of entrance to the top floor in tall buildings [24,25]. In addition, with the development of technology, electronic bullying is worth being noticed. According to the study by [30] with data collected by the Centers for Disease Control and Prevention in America, not only physical bullying but also electronic bullying increased the risk for suicide ideation and attempt, pointing out the importance of education of internet politeness and online social contact for youth. The risk factors for suicide of adolescence and youth included history of violence, female, young age (from 10 to 24 years old), psychological disorders, depression, and substance use in Taiwan, which were similar to the results in other regions around the world [25,30,31,32,33,34,35,36], indicating a somewhat biophysical mechanism that crosses ethnics. However, in a study in America [30], Asian adolescents are more prone to having suicide ideation and attempts than other ethnics, making it important for Asian educators and family members to pay more attention to their adolescents, including Taiwan.
In terms of gender differences, the statistical results of this study suggest that female adolescents experience higher rates of violence and suicide-related behaviors compared to their male counterparts. Previous research has shown that girls are more likely to encounter general violence than boys, particularly sexual harassment, which is often more difficult to identify in clinical settings compared to physical violence [31]. Additionally, studies by Canetto and Sakinofsky [32] and Conner and Duberstein [33] have indicated that female adolescents are at greater risk of experiencing suicidal ideation and attempts when compared to male peers. This could help explain the higher suicide rates observed among females relative to males. The gender disparity observed in this study may also be influenced by the broader age range of the adolescent population under investigation. However, our findings contrast with those from a hospital-based study conducted in the United States, which found that physical violence in children was more frequently identified in boys than in girls within the medical system [34]. Furthermore, research by Farrell, Petros, and Hawkins [35], using data from the Centers for Disease Control and Prevention’s Compressed Mortality Files, revealed a male-dominant cohort in childhood violence. Similarly, Ashraf, Kahn, and Hussain [34] reported that male adolescents reported significantly higher levels of violence than their female counterparts in community settings. In contrast, studies by Kolev, Petrova, and Vassilev [37] and Roh, Park, and Kim [38] found that suicide rates among males were higher than those observed in females. These findings highlight the multifaceted and complex role gender plays in suicidal behavior and experiences of violence. These differences in research outcomes may be potentially influenced by societal perceptions of physical discipline or the way physicians interpret male children’s behavior. This, in turn, may impact diagnostic accuracy, suggesting the need for further research to explore the underlying causes and the potential for targeted interventions.
This study has several limitations. First, the potential for selection bias cannot be ruled out. The incidence of adolescent violence may be significantly underestimated; as medical attention is typically sought only for adolescents who experience more severe forms of maltreatment. This issue is further compounded by cases where the abuser is a parent or caregiver, reducing the likelihood of the adolescent seeking medical evaluation. Additionally, the diagnoses of adolescent violence were based solely on clinical judgment and experience, which lack standardized criteria and may result in inaccuracies or underdiagnoses. The reliance on medical evaluations also poses a limitation, as prior studies (e.g., adverse childhood experience data) suggest that the prevalence of interfamilial physical violence is much higher than what is reflected in medical reports. Consequently, some individuals in the control group may have also been exposed to violence, potentially biasing the analysis.
Second, there may be gender-based biases in identifying violence-related injuries. Violence against male adolescents is more likely to be overlooked or misinterpreted as normal behavior. In some cases, physical discipline towards male adolescents may be perceived as acceptable rather than abusive, leading to the underreporting of such incidents in the NHIRD.
Third, this study was limited by the data available in the health insurance database. We were unable to consider other important factors that may influence the risk of adolescent violence, such as the parent–child relationship, marital status, educational background, or religious beliefs. Additionally, the NHIRD did not provide clinical biochemistry data, the Glasgow Coma Scale [39], or abbreviated injury severity scores [40], which could have provided additional insights.
Fourth, privacy concerns and data protection protocols prevented direct interaction with patients, limiting access to additional information such as their mental status. Furthermore, as this study relied exclusively on inpatient data, cases involving minor injuries, outpatient care, or emergency visits were not captured. As a result, the study’s findings may be biased toward more severe injury cases.
Lastly, in compliance with the regulations outlined by the Health and Welfare Data Center (HWDC) under the Ministry of Health and Welfare (MOHW) of Taiwan, it is not feasible to acquire new analytical results in the short term. Even with the potential availability of additional data, it is unlikely to significantly impact the trajectory of our findings. The primary objective of this study is to examine trends in suicide among victims of adolescent violence in Taiwan over an extended period. Notably, the dataset employed encompasses a sample of 2 million individuals, representing a subset of Taiwan’s 23 million population, which is sufficiently large to offer a reliable reflection of broader societal trends.
Violence is caused by the presence of multiple risk factors and a combination of very few protective factors. Violence can be prevented by reducing risk factors and strengthening protective factors. Conducting this requires comprehensive policies that form part of a so-called “integrated approach” to violence prevention, in other words, an overall strategy that depends on the cooperation of many different sectors.

5. Conclusions

In this study, adolescents who experienced violence had a significantly higher risk of future suicide compared to the control group, with the suicide rate increasing by up to 1.592 times. Notably, among those with comorbid mental disorders, the suicide risk rose to 2.369 times that of the control group. Exposure to youth violence may lead to emotional disorders, including depression and social isolation, which subsequently elevate the risk of suicide. Exposure to youth violence can lead to emotional disorders, including depression and social isolation, which in turn elevate the risk of suicide. To reduce the occurrence of suicide, it is crucial to not only provide psychological support and pay closer attention to the mental health of youth violence victims but also for the government to implement more practical measures. There is an urgent need to emphasize that experiencing violence or abuse is not shameful. Victims should seek medical attention or obtain protection and support from the government or healthcare institutions as early as possible to reduce their sense of guilt and prevent the loss of critical opportunities for timely intervention, documentation, and treatment. Additionally, schools and workplaces must establish comprehensive programs for gender equality education, bullying prevention, and related awareness initiatives. Third-party reporting and support mechanisms should also be put in place to ensure that victims do not fear being unable to continue functioning within their communities, which may otherwise lead them to endure in silence and ultimately result in tragic outcomes.

Supplementary Materials

The following supporting information can be downloaded at the following website: https://www.mdpi.com/article/10.3390/children12010010/s1, Table S1. Abbreviation, ICD-9-CM, and definition; Table S2. Years of follow-up; Table S3. Years to suicide.

Author Contributions

Conceptualization, C.S., L.-Y.F. and W.-C.C.; methodology, C.-A.S., C.-H.T. and F.-H.L.; software, C.-H.C. and T.-H.W.; validation, C.-H.C., D.Y.N. and F.-H.L.; formal analysis, T.-H.W.; data curation, C.-A.S. and C.-H.T.; writing—original draft preparation, C.S. and D.Y.N.; writing—review and editing, C.S., L.-Y.F. and W.-C.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Tri-service General Hospital, grant number TSGH-B-114022, TSGH-A-114010 special plan.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Tri-Service General Hospital (TSGHIRB No.: B202405024) on 7 February 2024.

Informed Consent Statement

This study was requirement of individual consent was waived because all identifying data were encrypted. The NHIRD is a publicly available database that contains depersonalized patient information to ensure patient anonymity.

Data Availability Statement

This study uses third-party data. Taiwan launched a single-payer National Health Insurance program on 1 March 1995. The database of this program contains registration files and original claim data for reimbursement. Large, computerized databases were derived from this system by the National Health Insurance Administration. Investigators interested may submit a formal proposal to NHIRD (https://dep.mohw.gov.tw/DOS/cp-5119-59201-113.html, accessed on 10 December 2024) The authors confirm that they did not have any special access privileges.

Acknowledgments

The authors would like to express their gratitude and appreciation to the HWDC, MOHW, Taiwan, for providing access to the NHIRD.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. World Health Organization. Child Maltreatment. Available online: https://www.who.int/news-room/fact-sheets/detail/child-maltreatment (accessed on 22 November 2024).
  2. U.S. Department of Health and Human Services; Substance Abuse and Mental Health Services Administration; Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health 2019 (NSDUH–2019). Available online: https://www.samhsa.gov/data/ (accessed on 22 November 2024).
  3. Taiwan Data from Ministry of Health and Welfare. Health Statistics. Available online: https://www.mohw.gov.tw/np-126-2.html (accessed on 22 November 2024).
  4. Curtin, S.C.; Heron, M. Death Rates Due to Suicide and Homicide Among Persons Aged 10–24: United States, 2000–2017. NCHS Data Brief. 2019; pp. 1–8. Available online: https://www.cdc.gov/nchs/products/databriefs/db352.htm (accessed on 22 November 2024).
  5. Kann, L.; McManus, T.; Harris, W.A.; Shanklin, S.L.; Flint, K.H.; Queen, B.; Lowry, R.; Chyen, D.; Whittle, L.; Thornton, J.; et al. Youth Risk Behavior Surveillance—United States, 2017. MMWR Surveill. Summ. 2018, 67, 1–114. [Google Scholar] [CrossRef] [PubMed]
  6. Angelakis, I.; Austin, J.L.; Gooding, P. Association of Childhood Maltreatment With Suicide Behaviors Among Young People: A Systematic Review and Meta-analysis. JAMA Netw. Open 2020, 3, e2012563. [Google Scholar] [CrossRef] [PubMed]
  7. Chung, C.H.; Lin, I.J.; Huang, Y.C.; Sun, C.A.; Chien, W.C.; Tzeng, N.S. The association between abused adults and substance abuse in Taiwan, 2000–2015. BMC Psychiatry 2023, 23, 123. [Google Scholar] [CrossRef] [PubMed]
  8. U.S. Department of Health and Human Services; Administration for Children and Families; Administration on Children, Youth, and Families & Children’s Bureau. Child Maltreatment 2019. Available online: https://www.acf.hhs.gov/cb/report/child-maltreatment-2019 (accessed on 22 November 2024).
  9. Kim, H.; Wildeman, C.; Jonson-Reid, M.; Drake, B. Lifetime Prevalence of Investigating Child Maltreatment Among US Children. Am. J. Public Health 2017, 107, 274–280. [Google Scholar] [CrossRef]
  10. World Health Organization. Global Status Report on Preventing Violence Against Children 2020; World Health Organization: Geneva, Switzerland, 2020. [Google Scholar]
  11. Hsieh, C.Y.; Su, C.C.; Shao, S.C.; Sung, S.F.; Lin, S.J.; Kao Yang, Y.H.; Lai, E.C. Taiwan’s National Health Insurance Research Database: Past and future. Clin. Epidemiol. 2019, 11, 349–358. [Google Scholar] [CrossRef]
  12. Ho Chan, W.S. Taiwan’s healthcare report 2010. EPMA J. 2010, 1, 563–585. [Google Scholar] [CrossRef]
  13. Chinese Hospital Association. ICD-9-CM English-Chinese Dictionary; Chinese Hospital Association Press: Taipei, Taiwan, 2000. [Google Scholar]
  14. Ministry of Health and Welfare. The Protection of Children and Youths Welfare and Rights Act. Available online: https://law.moj.gov.tw/ENG/LawClass/LawAll.aspx?pcode=D0050001 (accessed on 22 November 2024).
  15. Taiwan Data from Ministry of Health and Welfare. Taiwan Data from Ministry of Health and Welfare. 2019. Available online: https://www.mohw.gov.tw/lp-137-2.html (accessed on 22 November 2024).
  16. Charlson, M.E.; Szatrowski, T.P.; Peterson, J.; Gold, J. A New Method of Classifying Prognostic Comorbidity in Longitudinal Studies: Development and Validation. J. Chronic Dis. 1987, 40, 373–383. [Google Scholar] [CrossRef]
  17. De Groot, V.; Beckerman, H.; Lankhorst, G.J.; Bouter, L.M. How to Measure Comorbidity: A Critical Review of Available Methods. J. Clin. Epidemiol. 2003, 56, 221–229. [Google Scholar] [CrossRef]
  18. Charlson, M.; Pompei, P.; Ales, K.; MacKenzie, C. Validation of a Combined Comorbidity Index. J. Clin. Epidemiol. 1994, 47, 1245–1251. [Google Scholar] [CrossRef]
  19. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed.; text revision; American Psychiatric Association: Washington, DC, USA, 2000. [Google Scholar]
  20. Ministry of Justice. National Health Insurance Reimbursement Regulations. Available online: https://law.moj.gov.tw/eng/LawClass/LawAll.aspx?PCode=L0060001 (accessed on 22 November 2024).
  21. Chang, C.-Y.; Chen, W.-L.; Liou, Y.-F.; Ke, C.-C.; Lee, H.-C.; Huang, H.-L.; Ciou, L.-P.; Chou, C.-C.; Yang, M.-C.; Ho, S.-Y. Increased risk of major depression in the three years following a femoral neck fracture—A national population-based follow-up study. PLoS ONE 2014, 9, e89867. [Google Scholar] [CrossRef]
  22. McGrogan, A.; Madle, G.C.; Seaman, H.E.; De Vries, C.S. The epidemiology of Guillain-Barré syndrome worldwide: A systematic literature review. Neuroepidemiology 2009, 32, 150–163. [Google Scholar] [CrossRef] [PubMed]
  23. Needham, D.M.; Scales, D.C.; Laupacis, A.; Pronovost, P.J. A systematic review of the Charlson comorbidity index using Canadian administrative databases: A perspective on risk adjustment in critical care research. J. Crit. Care 2005, 20, 12–19. [Google Scholar] [CrossRef] [PubMed]
  24. Kiss, L.; Yun, K.; Pocock, N.; Zimmerman, C. Exploitation, Violence, and Suicide Risk Among Child and Adolescent Survivors of Human Trafficking in the Greater Mekong Subregion. JAMA Pediatr. 2015, 169, e152278. [Google Scholar] [CrossRef] [PubMed]
  25. MacIsaac, M.B.; Bugeja, L.; Weiland, T.; Dwyer, J.; Selvakumar, K.; Jelinek, G.A. Prevalence and Characteristics of Interpersonal Violence in People Dying From Suicide in Victoria, Australia. Asia Pac. J. Public Health 2018, 30, 36–44. [Google Scholar] [CrossRef]
  26. Zygo, M.; Pawłowska, B.; Potembska, E.; Dreher, P.; Kapka-Skrzypczak, L. Prevalence and selected risk factors of suicidal ideation, suicidal tendencies and suicide attempts in young people aged 13–19 years. Ann. Agric. Environ. Med. 2019, 26, 329–336. [Google Scholar] [CrossRef]
  27. Liu, R.T.; Walsh, R.F.L.; Sheehan, A.E.; Cheek, S.M.; Sanzari, C.M. Prevalence and Correlates of Suicide and Nonsuicidal Self-injury in Children: A Systematic Review and Meta-analysis. JAMA Psychiatry 2022, 79, 718–726. [Google Scholar] [CrossRef]
  28. Hua, L.L.; Lee, J.; Rahmandar, M.H.; Sigel, E.J.; Committee On Adolescence, & Council On Injury, Violence, And Poison Prevention. Suicide and Suicide Risk in Adolescents. Pediatrics 2024, 153, e2023064800. [Google Scholar] [CrossRef]
  29. Angelakis, I.; Gillespie, E.L.; Panagioti, M. Childhood maltreatment and adult suicidality: A comprehensive systematic review with meta-analysis. Psychol. Med. 2019, 49, 1057–1078. [Google Scholar] [CrossRef]
  30. Olsson, P.; Wiktorsson, S.; Sacuiu, S.; Marlow, T.; Östling, S.; Fässberg, M.M.; Skoog, I.; Waern, M. Cognitive Function in Older Suicide Attempters and a Population-Based Comparison Group. J. Geriatr. Psychiatry Neurol. 2016, 29, 133–141. [Google Scholar] [CrossRef]
  31. Castro, Á.; Ibáñez, J.; Maté, B.; Esteban, J.; Barrada, J.R. Childhood Sexual Abuse, Sexual Behavior, and Revictimization in Adolescence and Youth: A Mini Review. Front. Psychol. 2019, 10, 2018. [Google Scholar] [CrossRef]
  32. Canetto, S.S.; Sakinofsky, I. The Gender Paradox in Suicide. Suicide Life-Threat. Behav. 1998, 28, 1–23. [Google Scholar] [CrossRef] [PubMed]
  33. Conner, A.; Azrael, D.; Miller, M. Suicide Case-Fatality Rates in the United States, 2007 to 2014: A Nationwide Population-Based Study. Ann. Intern. Med. 2019, 171, 885–895. [Google Scholar] [CrossRef] [PubMed]
  34. Allareddy, V.; Asad, R.; Lee, M.K.; Nalliah, R.P.; Rampa, S.; Speicher, D.G.; Rotta, A.T.; Allareddy, V. Hospital-Based Emergency Department Visits Attributed to Child Physical Abuse in the United States: Predictors of In-Hospital Mortality. PLoS ONE 2014, 9, e100110. [Google Scholar] [CrossRef] [PubMed]
  35. Farrell, C.A.; Fleegler, E.W.; Monuteaux, M.C.; Wilson, C.R.; Christian, C.W.; Lee, L.K. Community Poverty and Child Abuse Fatalities in the United States. Pediatrics 2017, 139, e20161616. [Google Scholar] [CrossRef]
  36. Ashraf, F.; Niazi, F.; Masood, A.; Malik, S. Gender Comparisons and Prevalence of Child Abuse and Post-Traumatic Stress Disorder Symptoms in Adolescents. JPMA J. Pak. Med. Assoc. 2019, 69, 320–324. [Google Scholar]
  37. Kõlves, K.; De Leo, D. Suicide Methods in Children and Adolescents. Eur. Child. Adolesc. Psychiatry 2017, 26, 155–164. [Google Scholar] [CrossRef]
  38. Roh, B.R.; Jung, E.H.; Hong, H.J. A Comparative Study of Suicide Rates Among 10–19-Year-Olds in 29 OECD Countries. Psychiatry Investig. 2018, 15, 376. [Google Scholar] [CrossRef]
  39. Chung, C.H.; Lai, C.H.; Chu, C.M.; Pai, L.; Kao, S.; Chien, W.C. A nationwide, population-based, long-term follow-up study of repeated self-harm in Taiwan. BMC Public Health 2012, 12, 744. [Google Scholar] [CrossRef]
  40. Farst, K.; Ambadwar, P.B.; King, A.J.; Bird, T.M.; Robbins, J.M. Trends in hospitalization rates and severity of injuries from abuse in young children, 1997–2009. Pediatrics 2013, 131, e1796–e1802. [Google Scholar] [CrossRef]
Children 12 00010 g001
Figure 1. Flowchart of the study.
Figure 1. Flowchart of the study.
Children 12 00010 g001
Children 12 00010 g002
Figure 2. Kaplan–Meier for cumulative risk of suicide aged 10–18, stratified by violence with log-rank test.
Figure 2. Kaplan–Meier for cumulative risk of suicide aged 10–18, stratified by violence with log-rank test.
Children 12 00010 g002
Children 12 00010 g003
Figure 3. Joint effect for factors of suicide stratified by violence and mental disorders using Cox regression.
Figure 3. Joint effect for factors of suicide stratified by violence and mental disorders using Cox regression.
Children 12 00010 g003
Table 1. Characteristics of study in the baseline.
Table 1. Characteristics of study in the baseline.
ViolenceTotalWithWithoutp
Variablesn%n%n%
Total9999 909 9090
gender 0.999
Male2532.53232.532302.53
Female974697.4788697.47886097.47
Age (years)14.28 ± 4.8514.18 ± 4.7214.29 ± 4.860.314
Low-income household 0.948
Without740574.0667474.15673174.05
With259425.9423525.85235925.95
Catastrophic illness 0.704
Without887188.7280388.34806888.76
With112811.2810611.66102211.24
Mental disorders 0.999
Without946094.6186094.61860094.61
With5395.39495.394905.39
CCI0.74 ± 0.680.72 ± 0.650.74 ± 0.680.107
season 0.999
Spring (Mar–May)239823.9821823.98218023.98
Summer (Jun–Aug)267326.7324326.73243026.73
Autumn (Sep–Nov)255225.5223225.52232025.52
Winter (Dec–Feb)237623.7621623.76216023.76
Location <0.001
Northern Taiwan350035.0037541.25312534.38
Middle Taiwan254625.4623425.74231225.43
Southern Taiwan235823.5825127.61210723.18
Eastern Taiwan107210.72495.39102311.25
Outlets islands5235.2300.005235.75
Urbanization level <0.001
1 (The highest)292229.2229832.78262428.87
2342134.2134638.06307533.83
3163816.3811212.32152616.79
4 (The lowest)201820.1815316.83186520.52
Level of care 0.701
Hospital center525252.5348953.80476352.40
Regional hospital333133.3129732.67303433.38
Local hospital141614.1612313.53129314.22
p: Chi-square/Fisher’s exact test on category variables and U-test on continuous variables.
Table 2. Characteristics of study at the endpoint.
Table 2. Characteristics of study at the endpoint.
ViolenceTotalWithWithoutp
Variablesn%n%n%
Total9999 909 9090
Suicide <0.001
Without970997.1081789.88889297.82
With2902.909210.121982.18
Gender 0.999
Male2532.53232.532302.53
Female974697.4788697.47886097.47
Age (yrs)21.99 ± 9.9621.95 ± 9.8622.00 ± 9.970.763
Low-income household 0.948
Without740174.0267273.93672974.03
With259825.9823726.07236125.97
Catastrophic illness 0.683
Without886388.6480288.23806188.68
With113611.3610711.77102911.32
Mental disorders 0.707
Without945494.5585794.28859794.58
With5455.45525.724935.42
CCI0.75 ± 0.690.74 ± 0.670.75 ± 0.690.310
Season 0.873
Spring 241524.1522624.86218924.08
Summer267326.7324727.17242626.69
Autumn254725.4723025.30231725.49
Winter236423.6420622.66215823.74
Location <0.001
Northern Taiwan341134.1137240.92303933.43
Middle Taiwan255025.5024627.06230425.35
Southern Taiwan234123.4122324.53211823.30
Eastern Taiwan116311.63596.49110412.15
Outlets islands5345.3490.995255.78
Urbanization level <0.001
1 (The highest)293429.3429932.89263528.99
2337433.7434237.62303233.36
3169516.9511712.87157817.36
4 (The lowest)199619.9615116.61184520.30
Level of care 0.700
Hospital center523552.3648353.14475252.28
Regional hospital332433.2429132.01303333.37
Local hospital144014.4013514.85130514.36
p: Chi-square/Fisher exact test’s on category variables and U-test on continuous variables.
Table 3. Factors of suicide using Cox regression.
Table 3. Factors of suicide using Cox regression.
VariablesCrude HR95% CI95% CIpaHR95% CI95% CIp
Violence
WithoutReference Reference
With1.7871.2462.033<0.0011.5921.1371.993<0.001
Gender
MaleReference Reference
Female2.0981.3582.886<0.0011.5231.0721.8310.012
Age (yrs)0.8940.5891.1820.4860.9710.6321.2890.570
Low-income household
WithoutReference Reference
With2.3031.4933.701<0.0011.5721.0991.9770.001
Catastrophic illness
WithoutReference Reference
With1.8621.1982.596<0.0011.3031.0501.6840.024
Mental disorders
WithoutReference Reference
With2.9791.8654.228<0.0012.6661.3563.784<0.001
CCI1.2011.1861.277<0.0011.1421.0631.2250.018
Season
SpringReference Reference
Summer1.7791.2872.074<0.0011.4401.0881.7200.006
Autumn 1.9331.4832.335<0.0011.5181.1651.836<0.001
Winter1.5021.0951.9820.0031.3861.0241.6890.037
Location
Northern TaiwanReference Multicollinearity with urbanization level
Middle Taiwan0.8420.3011.3780.672Multicollinearity with urbanization level
Southern Taiwan0.9860.3891.4600.573Multicollinearity with urbanization level
Eastern Taiwan0.5970.2840.9970.048Multicollinearity with urbanization level
Outlets islands0.7320.042196.6780.925Multicollinearity with urbanization level
Urbanization level
1 (The highest)2.1351.4872.865<0.0011.7951.1942.512<0.001
21.9111.3742.784<0.0011.7831.1132.505<0.001
31.5061.0101.7010.0451.3220.8071.5710.189
4 (The lowest)Reference Reference
Level of care
Hospital center2.7852.0133.389<0.0012.1061.3702.864<0.001
Regional hospital1.8431.5622.131<0.0011.4421.1151.765<0.001
Local hospitalReference Reference
HR = hazard ratio, CI = confidence interval, and aHR = adjusted HR: Adjusted variables are listed in the table.
Table 4. Factors of suicide stratified by variables listed in the table using Cox regression and Bonferroni correction for multiple comparisons.
Table 4. Factors of suicide stratified by variables listed in the table using Cox regression and Bonferroni correction for multiple comparisons.
ViolenceWithWithout (Reference)With vs. Without (Reference)
StratifiedEventsPYsRateEventsPYsRateaHR95% CI95% CIp
Total927074.361300.4719869,894.12 283.29 1.592 1.137 1.993 <0.001
gender
Male3175.231712.0471768.33 395.85 1.500 1.071 1.872 0.015
Female896899.131290.0219168,125.79 280.36 1.596 1.401 1.998 <0.001
Low-income household
Without675230.011281.0714651,734.81 282.21 1.574 1.121 1.963 <0.001
With251844.351355.495218,159.31 286.35 1.642 1.178 2.059 <0.001
Catastrophic illness
Without796236.101266.8217561,981.84 282.34 1.556 1.113 1.949 <0.001
With13838.261550.83237912.28 290.69 1.852 1.327 2.320 <0.001
Mental disorders
Without846669.751259.4218766,103.87 282.89 1.542 1.104 1.913 <0.001
With8404.611977.21113790.25 290.22 2.369 1.688 2.972 <0.001
Season
Spring201758.821137.134516,831.25 267.36 1.475 1.053 1.849 0.022
Summer271992.331355.205318,653.31 284.13 1.656 1.180 2.072 <0.001
Autumn 261789.141453.215417,815.09 303.11 1.667 1.188 2.085 <0.001
Winter191534.071238.544616,594.47 277.20 1.543 1.104 1.940 <0.001
Urbanization level
1 (The highest)322323.021377.525920,262.43 291.18 1.641 1.177 2.059 <0.001
2352661.671314.966623,313.17 283.10 1.611 1.152 2.023 <0.001
311908.411210.913412,133.66 280.21 1.498 1.070 1.878 0.015
4 (The lowest)141181.261185.183914,184.86 274.94 1.494 1.067 1.870 0.017
Level of care
Hospital center523758.981383.3510736,538.04 292.85 1.647 1.173 2.066 <0.001
Regional hospital282253.181242.696423,231.55 275.49 1.562 1.118 1.982 <0.001
Local hospital121062.201129.732710,124.53 266.68 1.460 1.049 1.842 0.026
PYs = person-years, Rate: per 100,000 PYs, and aHR = adjusted hazard ratio, adjusted for the variables listed in Table 3. CI = confidence interval.
Table 5. Factors of suicide subgroups using Cox regression and Bonferroni correction for multiple comparisons.
Table 5. Factors of suicide subgroups using Cox regression and Bonferroni correction for multiple comparisons.
ViolenceWithWithout (Reference)With vs. Without (Reference)
SuicideEventsEventsaHR95% CI95% CIp
Overall92 198 1.592 1.137 1.993 <0.001
 Solid or liquid 38 81 1.607 1.148 2.012 <0.001
 Gasses in domestic use0 2 0.000 --0.999
 Other gasses and vapors12 24 1.714 1.223 2.145 <0.001
 Hanging3 5 2.058 1.468 2.571 <0.001
 Drowning0 2 0.000 --0.999
 Firearms1 0 ∞--0.997
 Cutting and piercing29 60 1.656 1.172 2.073 <0.001
 Jumping4 9 1.523 1.081 1.909 0.009
 Others5 15 1.142 0.823 1.424 0.157
aHR = adjusted hazard ratio, adjusted for the variables listed in Table 3. CI = confidence interval.
Table 6. Factors of suicide stratified by violence and mental disorders using Cox regression.
Table 6. Factors of suicide stratified by violence and mental disorders using Cox regression.
ViolenceMental DisordersaHR95% CI95% CIpp for Interaction
WithoutWithout1.000 <0.001
WithWithout1.465 1.172 1.779 <0.001
WithoutWith1.756 1.340 2.075 <0.001
WithWith3.586 2.781 4.986 <0.001
aHR = adjusted hazard ratio, adjusted for the variables listed in Table 3. CI = confidence interval.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Sung, C.; Chung, C.-H.; Sun, C.-A.; Tsao, C.-H.; Ng, D.Y.; Weng, T.-H.; Fann, L.-Y.; Lin, F.-H.; Chien, W.-C. Association Between Adolescent Violence Exposure and the Risk of Suicide: A 15-Year Study in Taiwan. Children 2025, 12, 10. https://doi.org/10.3390/children12010010

AMA Style

Sung C, Chung C-H, Sun C-A, Tsao C-H, Ng DY, Weng T-H, Fann L-Y, Lin F-H, Chien W-C. Association Between Adolescent Violence Exposure and the Risk of Suicide: A 15-Year Study in Taiwan. Children. 2025; 12(1):10. https://doi.org/10.3390/children12010010

Chicago/Turabian Style

Sung, Chieh, Chi-Hsiang Chung, Chien-An Sun, Chang-Huei Tsao, Daphne Yih Ng, Tsu-Hsuan Weng, Li-Yun Fann, Fu-Huang Lin, and Wu-Chien Chien. 2025. "Association Between Adolescent Violence Exposure and the Risk of Suicide: A 15-Year Study in Taiwan" Children 12, no. 1: 10. https://doi.org/10.3390/children12010010

APA Style

Sung, C., Chung, C. -H., Sun, C. -A., Tsao, C. -H., Ng, D. Y., Weng, T. -H., Fann, L. -Y., Lin, F. -H., & Chien, W. -C. (2025). Association Between Adolescent Violence Exposure and the Risk of Suicide: A 15-Year Study in Taiwan. Children, 12(1), 10. https://doi.org/10.3390/children12010010

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop