Intra-Familial Adverse Childhood Experiences and Suicidal Behaviors among Tunisian Youth: The Mediating Effects of Impulsivity and Resilience

Abstract

Given that resilience is the ability to cope with adversities, and impulsivity is characterized with rapid reactions without foresight, we aimed to explore the pathways between intra-familial childhood adversities, resilience, impulsivity and suicidal behaviors among adolescents in Mahdia and Gafsa cities (Tunisia). We conducted a cross-sectional study in secondary schools from January to February 2020. Exposure to intra-familial early life experiences was evaluated by the Adverse Childhood Experiences (ACEs)-International Questionnaire. Resilience and impulsivity were assessed via the Adolescent Psychological Resilience and the Barratt Impulsivity Scale. A total of 3170 students were recruited with a response rate of 74%. About 81.1% of them reported being emotionally abused. The resilience mean score was 86.10 ± 9.85. Impulsivity was screened among 42.2% of them. About 38% of students presented suicidal thoughts, 16.4% had a suicide plan, and 10.8% have made a suicide attempt. We found that exposure to intra-familial ACEs predicts suicidal behaviors through impulsiveness (% mediated = 18% for emotional violence) and interpersonal resilience (% mediated = 24% for emotional violence; % mediated = 20.4% for physical violence). Our results emphasize the need to prevent ACEs, and to seek out a way to promote ACE protective factors among adolescents such as resilience.

1. Introduction

Adolescence is a precarious period where developmental and behavioral transitions occur over time. The recent surge in adolescent mental health problems, especially the increasing rate of suicidal behaviors among them [1, 2], suggests that this change may be more difficult in the twenty-first century than it was ever before. In fact, suicidal behaviors among youth have become a major public health problem that is very delicate to handle. It is currently the second leading cause of death among adolescents in the United States of America (USA) [3] and it has increased by 10% annually from 2014 to 2017 for adolescents and young adults [4]. In Tunisia, suicidal behaviors have also become a scourge among our youth. According to a recent Tunisian study, 26.9% of adolescents had suicidal ideations and 7.3% of them have attempted suicide [5]. In light of this, it is important to determine risk factors of suicidal behaviors in adolescents to evaluate where and how we could intervene to reduce these percentages.

According to literature [6, 7], one of the major determining factors of suicide is exposure to Adverse Childhood Experiences (ACEs). Indeed, early life adversities, categorized according to the World Health Organization (WHO) as intra-familial and social ACEs, including emotional abuse, physical neglect, and social violence, have been proved to deteriorate mental health and be responsible for an array of health problems and developmental issues [8, 9, 10]. In the USA [11], a nationwide study concluded that 57.7% of individuals experienced at least one ACE. In Tunisia, the problem seems to be an even more widespread issue, according to recent research, a staggering 89.4% of adults [12] and 97.5% of adolescents had experienced at least one ACE [13].

While both ACE categories are detrimental, intra-familial ACEs (IF-ACEs) have been linked to a high risk of suicidal behaviors, with an established dose-response relationship. In fact, a study done in the USA in 2018 [14] exhibited that having one intra-familial ACE increases the risk of attempted suicide 2- to 5-fold. The odds ratio (OR) of suicide attempt with seven or more ACEs was 31.1 (prevalence of attempts at suicide without ACEs being 1.1%). In Canada, another paper focusing on the effects of intra-familial ACEs showed that the risk of suicidal behaviors was three times higher (OR = 3.29; 99.9% confidence interval (CI) 2.33–4.64) for experiencing childhood physical abuse, four times higher (OR = 4.42; 99.9% CI 3.14–6.23) for those with a history of childhood sexual abuse, and two times higher (OR = 2.52; 99.9% CI 1.69–3.76) for those experiencing parental domestic abuse [15].

Although several surveys have studied the link between ACEs and suicidal behaviors, few have explored the mediating factors of this pathway especially among adolescents. In addition, these studies are rarely done outside of the Western world [16]. Impulsivity is one of the most frequently cited risk factors for engaging in maladaptive behaviors, such as self-harm [17]. In fact, a study conducted in the USA indicated that the odds ratio of suicidal behaviors in people suffering from impulsive-aggressive personalities was 30.3 [18]. Another South Korean study comparing planned and unplanned suicides revealed that 48% of them were impulsive in nature, and 21.1% of the impulsive suicide attempts were done using a lethal method [19]. Thus, impulsivity could be a potential mediator in the link between ACEs and suicidal behaviors mainly among adolescents. In China, research highlighted the importance of impulsivity as a mediator in that relationship in adolescents by showing a significant indirect effect on suicidal ideations [20]. A 2018 paper from Finland studying different mediators of this relationship in adolescents divulged that impulsivity was one of the rare significant personality traits [21]. Regarding protective mediators, there has been recently a surge in research focusing on the role of resiliency in maintaining and regaining mental health, despite experiencing adversity [22, 23, 24]. In 2019, a paper published in the USA showed that building resilience reduces the incidence of common risk factors for suicide, lowers suicidality, and betters the lives of people suffering from psychiatric disorders compared to less resilient people [24]. On the other hand, a meta analysis survey carried out in Taiwan focusing on youths proved that levels of resilience were significantly lower among adolescents with a higher prevalence of ACEs [25]. Thus, Resilience could also be an important mediator in the connection between ACEs and suicidal behaviors. Indeed, a survey carried out in China proved that resilience plays an important role as a mediator between ACEs and suicidal behaviors in youths [26]. Another study in Slovakia [27] underlined the role of resilience as a mediator between ACEs and emotional and behavioral problems. Unfortunately, few researches have explored the role of resilience in this pathway worldwide. To our knowledge, no research exploring either of these two pathways has been done in Tunisia so far.

As a matter of fact, while some studies focusing on ACEs have been done in oriental countries like the Kingdom of Saudi Arabia (KSA) [8, 28, 29] and some research in east and south African countries [30, 31, 32], there is a lack of research not only in Tunisia but also in the Middle East and North African region regarding the mediators between ACEs and suicide. While juvenile suicide rates are always fluctuating, recent studies suggest that it has been on the rise in Tunisia [33]. In actuality, a 12-year (2005–2016) study emphasizes that there have been two spikes of youth suicide attempt in both 2014 (17.7%) and 2016 (12.1%) [34]. In light of this, there is an urgent need to understand this pathway in order to implement preventive and protective measures to alleviate the suicidal behaviors among youth.

Given these data and the gap in the literature in this subject, the current study aimed to explore the link between intra-familial ACEs and suicidal behaviors mediated by both resilience and impulsivity among schooled youth in Mahdia and Gafsa cities, Tunisia.

2. Methods

2.1 Study population and sampling

A cross-sectional study was performed among youths enrolled in high schools of the delegations of Mahdia and Gafsa, Tunisia, during the period from January to February 2020.

Based on the cluster sampling, we randomly picked one class from every grade of each high school. All students who accepted to partake in the study were included.

According to a recent Tunisian study about suicide among adolescents [5], the prevalence of suicidal behaviors was 26.9%. Based on that and with a 0.05 probability of type I error (α) and an accuracy of 3%, we calculated the minimal sample size to be 840 students.

2.2 Data collection and study instruments

The survey was performed with a self-administered and anonymous questionnaire. Trained medical experts were present in the classrooms to explain the study and provide assistance answering the questions. Any questionnaires with missing or incomplete answers were eliminated from the study.

The tool consisted mainly of four parts:

2.2.1 Sociodemographic characteristics of the students

We collected information about gender, age, and educational characteristics of the students.

2.2.2 Measurement of childhood adversities

We used the Adverse childhood experiences-International Questionnaire (ACE-IQ) that was developed by the WHO.

The ACE-IQ is used to compute adversities encountered in the first 18 years of life [35]. It was translated and validated in Arabic by Saudi Arabia [28]. Some words were changed and added to fit the Tunisian culture and dialect during previous Tunisian ACE studies [13, 36].

The questionnaire is divided by two main categories that are subdivided into nine sections:

• Intra-familial ACEs: composed of physical, sexual, and emotional abuse; household dysfunction; emotional and physical neglect.

• Extra-familial ACEs: composed of exposure to war and collective violence, community violence, and peer violence/bullying.

In this study, we only focused on assessing intra-familial ACEs (IF-ACEs).

2.2.3 Assessment of suicidal behaviors among adolescents

We evaluated suicidal behaviors by asking simple yes and no questions about three different levels of suicide risk: having suicidal ideation, making a suicide plan, and having attempted suicide in the past.

2.2.4 Evaluation of impulsivity, resilience, anxiety, and depression

2.2.4.1 Impulsivity

Impulsivity is, according to the “Diagnostic and Statistical Manual of Mental Disorders (DSM-5),” defined as “actions without foresight that are poorly conceived, prematurely expressed, unnecessarily risky and inappropriate to the situation.” It is rarely associated with desirable outcomes. Impulsivity is also characterized by rapid, unpredictable, and spontaneous reactions to stimuli without much regard for consequences [37].

We assessed impulsivity using the “Barratt Impulsivity Scale” (BIS-11) [38]. It showed high convergent validity and is commonly used in both research and clinical settings and has been validated in Arabic [39]. The BIS-11 is a 30-item questionnaire that measures three broad facets of impulsivity:

• The planning factor.

• The motor impulsivity.

• The cognitive instability.

Responses were classified through a Likert-type scale ranging from “1 = Rarely/Never” to “4 = Almost/Always.” The higher the score is, the more impulsive a person is. A student scoring 72 points or higher was considered to be highly impulsive.

Despite having several subscales for evaluation, we decided to focus on the total impulsivity score to assess our subjects in this study.

2.2.4.2 Resilience

The American Psychological Society (APS) defines resilience as “The Process of adapting well in the face of adversity or significant sources of stress such as relationship problems, serious health problems or workplace and financial stressors” [23].

We used “The adolescent psychological Resilience Scale” in its Arabic validated version [40].

It’s a 29-item questionnaire that measures six dimensions of resilience that can be split into two major categories:

• Intra-personal factors: Empathy, sense of struggle, and adjustment.

• Inter-personal factors: family support, school support, and confidant-friend support.

Each question has four levels of answers from “not exactly suitable for me” scored “1,” to “exactly suitable for me” scored “4.”

Each facet of resilience was scored individually with a number of questions and the total of the scores is the total resilience score of the individual.

The higher the student ranked in each category, the stronger their resilience was in that particular facet. The same is applied for the total resilience score.

2.2.4.3 Anxiety and depression

We screened for anxiety and depression using the Arabic version of the “Hospital Anxiety and Depression Scale” (HADS).

2.3 Statistical analysis

Data entry and analyses were conducted using IBM SPSS Statistics; version 25.

Quantitative variables were represented by means and standard deviations (SDs) and qualitative ones were represented by absolute and relative frequencies. Student’s tests and chi square were used to compare means and percentages, respectively.

We assessed each category of IF-ACE by gender. The number of experiences was summed up for each respondent (IF-ACE score range, 0–6) then categorized into 0, 1–2, 3, and ≥ 4.

To evaluate the impact of emotional violence, we summed up both emotional abuse and emotional neglect as a new entity.

We gave a value of “0” for the students who suffered from neither emotional neglect nor abuse. We coded a value of “1” for those presenting either or both emotional abuse and neglect. We did the same for physical violence.

Suicidal behaviors were coded “0” for people who did not have any suicidal ideations and 1 for those who had at least one of the three levels of suicide risk (suicidal ideation, making suicide plan or suicide attempts).Missing data were excluded from analyses.

We used binary logistic regression analysis to estimate the likelihood of having suicidal behaviors by the number of IF-ACE exposures, then adjusted to gender and to common mental disorders (anxiety and depression).

A p-value less than 0.05 was considered statistically significant.

Mediation analysis (Figure 1):

Figure 1.

The theoretical relationship between intra-familial ACEs (emotional violence and physical violence) and suicidal behaviors with impulsivity and resilience as mediators.

To evaluate the indirect effect of an independent variable on a dependent variable through a mediator, we used the mediation analysis [41]. Both resilience and impulsivity, as continuous variables, were explored as potential mediators in this analysis. Spearman’s correlations were used to assess the zero-order relationships among IF-ACEs (specifically exposure to emotional violence and physical violence), impulsivity, resilience, and suicidal behaviors. To determine the presence of a significant mediation (or indirect effect) of impulsivity or resilience in the relationship between IF-ACEs and suicidal behaviors, we performed mediation modeling. Resilience and impulsivity were considered as potential mediating variables when their inclusion into the model resulted in a partial or total diminution of the relationship between suicidal behaviors as the dependent variable and IF-ACEs as the independent variable. Mediation analyses were conducted using SPSS version 25 and the PROCESS macro developed by Andrew F. Hayes [42]. We used the Sobel test to verify and assess the indirect effect [43]. Results were adjusted to common mental health disorders and gender.

Some criteria need to be met to allow us to apply the mediation model:

First, there must be a significant relationship between the independent variable (IF-ACEs) and the dependent variable (suicidal behaviors) (pathway c). Second, the variable of mediation (resilience and impulsivity) must be significantly associated with suicidal behaviors (pathway b). Finally, the relationships between IF-ACEs and both impulsivity and resilience must be significant (pathway a).

When pathway c is reduced significantly (partial mediation) or is no longer significant (full mediation) by including the mediator into the assessment of pathway c (pathway c’), the mediation is considered to be significant.

2.4 Ethical considerations

The Ethics Committee at the University Hospital of Mahdia (Tunisia) was charged with evaluating the study and has approved the study protocol (Approval number: P01 M.P.C-2020). The study was also approved by the Tunisian Ministry of Education. We provided them both with a copy of the questionnaire used and a detailed plan of the study subject.

We also requested authorizations from the headmasters, principals, and teachers of the participating secondary schools.

We explained to the adolescents and their parents the use of test results for research before data collection started. They were free to refuse participation.

Trained doctors were present in each classroom to explain the aim of the study and to guarantee the anonymity and confidentiality for students.

At the end of data collection, we provided the students with the address and the telephone number of a psychiatrist and we expressed our will to help any person who requests medical care.

3. Results

3.1 General characteristics of the study sample

We initially included a total of 3170 adolescents attending schools in Mahdia and Gafsa cities. The response rate was 74.2%, resulting in a sample size of 2354 students who completed and returned the questionnaire.

The average age was 17.3 ± 1.5 years. Females represented 65.8% (n = 1534) of the sample. Most of the students (90.3% (n = 2126)) were living with both parents.

3.2 Distribution of intra-familial ACEs by gender

Table 1 summarizes the distribution of IF-ACEs and its subclasses by gender.

Intra-familial ACE categories, n (%)	Total (n = 2354)	Male (n = 799)	Female (n = 1534)	p-value
Total IF-ACEs	2104 (95.2)	726 (96.84)	1378 (94.6)	0.07
Emotional abuse	1876 (81.3)	636 (80.9)	1240 (81.5)	0.7
Household dysfunction	1762 (78)	582(75.9)	1180 (79.1)	0.08
Physical abuse	1252 (54.3)	462 (58.9)	790 (51.9)	0.001
Emotional neglect	952 (41.3)	329 (41.8)	623 (41)	0.7
Physical neglect	469 (24.4)	173 (26.9)	293 (23.1)	0.06
Sexual abuse	296 (12.9)	105 (13.5)	191 (12.7)	0.6

Table 1.

Distribution of reported IF-ACEs among students by gender.

Emotional abuse was the most commonly reported IF-ACE (81.3%), followed by household dysfunction (78%) and physical abuse (54.3%). Gender comparison did not reveal any statistically significant difference in overall exposure to intra-familial violence, except for physical abuse where there was a higher prevalence of exposure among male students (58.9% vs. 51.9%, p = 0.001) (Table 1).

3.3 Mental health status among schooled youths by gender

The distribution of mental health problems by gender is shown in Table 2.

Mental health status, n (%)	Total (n = 2354)	Male (n = 799)	Female (n = 1534)	p-value
Suicidal behaviors	896 (38.6)	210 (26.4)	686 (45)	<0.001
Suicidal ideations	884 (38.1)	205 (25.8)	679 (44.4)	<0.001
Suicide planning	378 (16.3)	77(9.7)	301 (19.7)	<0.001
Suicide attempt	246 (10.6)	50 (6.3)	196 (12.8)	<0.001
Anxiety and depressive disorders	1574 (68.5)	475 (60.6)	1099 (72.6)	<0.001
Anxiety	1492 (64.9)	419 (53.4)	1073 (70.8)	<0.001
Depression	988 (43)	325 (41.4)	663 (43.8)	0.2
Impulsive behaviors	974 (42.4)	273 (34.8)	701(46.3)	<0.001

Table 2.

Self-reported mental health status among adolescents by gender.

We observed that 38.6% of our sample presented suicidal behaviors.

Gender comparison showed a higher prevalence in females for suicidal ideations (44.4% vs. 25.8%, p < 0.001), suicide planning (19.7% vs. 9.7%, p < 0.001), and suicide attempts (12.8% vs. 6.3%, p < 0.001).

The sample was heavily screened for mental disorders, with 64.9% of youth showing signs of anxiety and 43% for depression (Table 2). We noticed a higher prevalence of anxiety with females in gender comparison (70.8% vs. 53.4%, p < 0.001) but no difference in gender for depression.

The mean for total impulsivity score was 69.8 ± 10.09. Table 2 shows that 42.4% of the total sample scored for high impulsivity.

Gender analyses proved that there was a statistically significant higher prevalence for females to be impulsive than males (46.3 vs. 34.8%, p < 0.000).

Table 3 explores the levels of total resilience and its subclasses according to gender.

Characteristics	Male (n = 771)	Female (n = 1511)	p-value
Total resilience score (min = 50, max = 109)	85.38 ± 10.05	86.50 ± 9.74	0.01
Inter-personal resilience	52.84 ± 8.09	53.95 ± 8.19	0.002
Family support	22.35 ± 4.21	22.77 ± 4.44	0.03
Confidant/friend support	15.60 ± 3.91	15.61 ± 4.15	0.96
School support	14.89 ± 3.64	15.56 ± 3.60	<0.0001
Intra-personal resilience	32.52 ± 3.97	32.55 ± 3.86	0.86
Adjustment	10.38 ± 2.38	10.16 ± 2.33	0.24
Sense of struggle	13.08 ± 2.1	13.05 ± 2.13	0.72
Empathy	9.16 ± 2.23	9.32 ± 2.17	0.86

Table 3.

Distribution of resilience scores among schooled youth by gender.

We found that the mean resilience score was 86.10 ± 9.85.

Gender comparison showed a higher total resilience among females than males (86.5 ± 9.74 vs. 85.38 ± 10.05, p = 0.01).

There was also a higher prevalence among females for interpersonal resilience (53.95 ± 8.19 vs. 52.84 ± 8.09, p = 0.002) with notably a higher score for family support (22.77 ± 4.44 vs. 22.35 ± 4.21, p = 0.03) and school support (15.56 ± 3.60 vs. 14.89 ± 3.64, p < 0 .001), but no gender difference in confidant/friend support (Table 3).

Gender comparison found no difference in intra-personal resilience in our sample.

3.4 Association between intra-familial ACEs and suicidal behaviors among adolescents: binary regression analysis

Table 4 supports the theory that the risk of suicidal behaviors increases with the number of intra-familial ACEs.

	Crude OR (CI 95%)	Adjusted for gender, anxiety, and depression
Suicidal behaviors
0 IF-ACE	—	—
1–2 IF-ACEs	3.15 (1.5O–6.63)*	3.21(1.51–6.85)*
3 IF-ACEs	8.08 (3.87–16.89)**	7.5 (3.54–15.88)**
≥ 4 IF-ACEs	15.96 (7.65–33.28)**	14.64 (6.92–30.97)**

Table 4.

Crude and adjusted odds ratios (95% confidence intervals) for the dose-response link between the number of IF-ACEs and suicidal behaviors among adolescents.

*

p < 0.01.

**

p < 0.001.

When the gender and common mental issues (anxiety and depression) were taken into account in the adjusted model (right column), we found a gradual increase in the odds of having suicidal behaviors in case of exposure to one to two IF-ACEs (ORa = 3.21, CI = 1.5–6.63), if the adolescent experienced three IF-ACEs (ORa = 7.5, CI = 3.54–15.88), and when they experience at least four IF-ACEs (ORa = 14.64, CI = 6.92–30.97) (Table 4).

3.5 Impulsivity and resilience as mechanisms linking IF-ACEs with suicidal behaviors among Tunisian adolescents: Mediation analysis

For suicidal behaviors, we correlated the scores by making a four-level scale depending on the suicidal risk level. First level being a student who has no suicidal ideations, second level is a student presenting only suicidal ideations, third level is for students having previously thought of a suicide plan, and the fourth level is for students having attempted suicide. Emotional violence was also taken as a continuous variable ranging from zero to two depending on whether the students suffered from emotional abuse or emotional neglect (1) or both (2). Physical violence was treated the same way using physical abuse and physical neglect.

Variables we included in the mediation analysis were significantly correlated (Table 5).

	(1)	(2)	(3)
*Emotional violence	0.14***	0.23***	0.23***
Impulsivity	—	−0.30***	0.31***
Inter-personal resilience	−0.30***	—	−0 .29***
Suicidal behaviors	0.31***	−0.29***	—
*Physical violence	0.22***	−0.26***	0.25***

Table 5.

Zero-order relationships between emotional violence, physical violence, impulsivity, inter-personal resilience, and suicidal behaviors among adolescents.

***

: p < 0.001.

Emotional violence was significantly associated with impulsivity and inter-personal resilience (pathway a). Impulsivity and inter-personal resilience (pathway b) and emotional violence (pathway c) were independently associated with suicidal behaviors (p < 0.001).

No correlations were found for sexual abuse (as independent variable) and intra-personal resilience (as a mediator).

We found that impulsivity mediated exposure to emotional violence and suicidal behaviors among adolescents (p < 0.001, mediation = 18.8%).

It also mediated exposure to physical violence and suicidal behaviors among our sample (p < 0.001, mediation = 17.4%) (Table 6).

	Coefficients*				Sobel test		% Mediated†
Mediator: Impulsivity	a	b	c	c’	SE	P
Emotional violence	2.74	0.02	0.40	0.34	0.01	p < 0.001	18.8
Physical violence	3.21	0.02	0.41	0.33	0.01	p < 0.001	17.4

Table 6.

Adjusted mediation model of IF-ACE effect on suicidal behaviors with impulsivity as a mediator among adolescents (N = 2248).

^*

Model adjusted to gender and common mental disorders (anxiety and depression).

^†

% Mediated = c – c’/c.

There was a major significant mediation between childhood emotional violence and suicidal behaviors through inter-personal resilience among youth (p < 0.001, mediation = 24%).

We also found a significant correlation between early life exposure to physical violence and suicidal behaviors among adolescents (p < 0.001, mediation = 20.4%) (Table 7).

	Coefficients*				Sobel test		% Mediated†
Mediator: Interpersonal resilience	a	b	c	c’	SE	P
Emotional violence	−3.59	−0.03	0.40	0.30	0.01	p < 0.001	24
Physical violence	−3.48	−0.02	0.41	0.33	0.01	p < 0.001	20.4

Table 7.

Adjusted mediation model of the relationship between IF-ACE types on suicidal behaviors via inter-personal resilience as a mediator among adolescents (N = 2228).

^*

Model adjusted to gender and common mental disorders (anxiety and depression).

^†

%Mediated = c – c’/c.

4. Discussion

There have not been many studies done so far that aimed to explore the relationships between exposure to early life adversities and suicidal behaviors through impulsivity or resilience. This survey goal was to examine the mediating roles of impulsive behaviors and resilience in the relationship between IF-ACEs and suicidal behaviors. Our study supports the idea that IF-ACEs predict suicidal behaviors through impulsiveness and interpersonal resilience.

Regarding exposure to intra-familial violence, our study showed an alarmingly high prevalence of IF-ACEs (95.2%). Published studies done on adolescents in developed countries showed that they only had a prevalence of 38.5% in the USA [44] and 44.3% in Canada [45] for overall ACEs. This difference is probably the result of socioeconomic and cultural differences. We tried comparing our results with those of other north African studies tackling exposure to ACEs among adolescents but we could not find any, regardless, our results were still more compatible with papers from eastern countries on adults, notably in Tunisia (99%) [46] and Saudi Arabia (82%) [8]. We found that the most frequently reported IF-ACE among adolescents in our sample was emotional abuse (81.3%), closely followed by household dysfunction (78%). These results are also incompatible with results from developed countries where the most prevalent IF-ACE was household dysfunction. In the United Kingdom, the most frequently reported IF-ACEs in a cohort study done on adolescents were first parental separation (33.8%) followed by violence between parents (25.3%), both being forms of household dysfunction. Emotional abuse and neglect only appear in third place with the same percentage of comparatively low prevalence (23.9%) [47]. In Germany, a similar study has found that the prevalence of emotional violence was also low (12.5%), while the most frequently reported IF-ACE was also household dysfunction, the prevalence was also quite low by comparison (19.4%). Some differences may be explained by differences in parental styles since corporal punishment is still regarded by most families as a normal disciplinary route for children in Tunisia. Another factor to take into consideration is the drawbacks of the 2011 revolution that not only added its fair share of social ACEs by itself, but also added further economical and social pressure on parents who may have caused further instability in the household [48, 49, 50, 51].

Focusing on eastern studies, we found that while studies done on adults do present higher percentages of reported IF-ACEs than those of Western countries, the prevalence still remains lower than our findings. In the KSA, the most commonly reported IF-ACE among adults was domestic violence against a household member (57%) followed by emotional abuse (52%) [52]. The high prevalence of emotional abuse in our sample may be explained by the locality of the study since it focused on only two governorates, a nationwide study is necessary to understand if this is a local issue or a national one. We also need to raise awareness of this issue and explain to children that such behaviors are abusive in nature, since with such a high percentage, it is worrisome that ACEs may be considered as acceptable reproducible behaviors in the community. Household dysfunction was the most frequent IF-ACE reported by adults which is probably caused by the fact that we targeted different age groups, adults tend to be more open to talk about growing up in a dysfunctional home, while it is still remains a delicate subject to address for Tunisian adolescents still living in said homes. Nonetheless, these results highlight the importance of frequent screening for ACEs in schools in Tunisia.

We observed that 38.6% of our sample presented suicidal behaviors which is a little higher than another local 2019-study (26.9%) [53]. When comparing our suicidal ideation rate to those given in previous studies, our results came higher than estimates observed in developed countries [21, 54, 55, 56, 57, 58] and also in developing countries [21, 49, 56, 57]. The variability in prevalence of suicidal thoughts may be the result of different methodological approaches (sample selection, study site, perception of suicide). In terms of suicidal planning, our results showed higher prevalence than a Chinese study (16.3 and 12%, respectively) [56]. Alarmingly, our results showed that 10.6% of our adolescents had a suicide attempt, which is higher than previous international data that ranged between 3.3 and 7.96% [49, 51, 53, 58, 59]. Gender comparison showed a higher prevalence in females for all suicidal behavior types. These outcomes resonate with Canadian [60] and Chinese [56] findings revealing that girls experienced a greater and more rapidly increasing rate of suicidal ideation and three times more likely to attempt suicide than boys. In light of these alarming findings, it is recommended to implement accessible mental health support with gender-specific interventions.

The prevalence of signs of depression in the current sample is relatively high (43%). The finding prevalence is higher than described in high-income countries [28, 57, 61] and also in middle- and low-income countries [21, 49, 57]. However, a previous local study showed that 71.8% of the adolescents had depressive manifestations of varying intensity [53]. The difference in the occurrence of depressive symptoms could potentially stem from varying measurement scales. Signs of anxiety were the most screened mental disorders, with 64.9% of youngsters showing them. This rate is too high when comparing to an Arabic finding (17%) and to two Tunisian outcomes (42.3 and 55.5%) [13, 53]. We noticed a higher frequency of anxiety with females in gender comparison which is comparable to local Tunisian articles [13, 46].

We found that impulsivity (42.2%) was mostly prevalent among teenagers. Almost, the same rate (43.5%) was found in a 2019-study among Tunisian students [13]. Gender analyses proved that there was a statistically significant higher prevalence for females to be impulsive. When analyzing gender and cultural factors, there was no distinct indication of elevated or diminished impulsivity rates in Arabic cohorts when contrasted with Western samples [13]. In order to enhance the strength and credibility of the findings, it is necessary to incorporate substantial and representative samples locally and regionally.

Our study has unveiled noteworthy gender-based disparities in adolescent resilience and its association with adverse childhood experiences (ACEs). These outcomes are consistent with those of prior research underlining the necessity of incorporating gender considerations for a comprehensive understanding of resilience outcomes [10, 54, 59, 62]. Importantly, our study revealed that females exhibited higher scores in interpersonal resilience, particularly within the domains of family and school support. These results echo the findings of studies emphasizing the significance of family and peer relationships in fortifying adolescent resilience [63, 64, 65, 66]. The greater overall resilience observed in females compared to that observed in males suggests gender-specific coping strategies and patterns of social support. This finding contrasts with an Italian study pointing to higher resilience scores among male students exposed to a traumatic event (earthquake) [67]. Additionally, it counters the outcomes of an American community survey centered on demographic factors predictive of enhancing resilience [68]. Nevertheless, among a cohort of Chinese adolescents, resilience was identified as a significant moderating influence on the connection between emotional abuse and depressive symptoms for both genders. Notably, the moderating effect of resilience was more pronounced among females than males [69].

Our investigation also brought to light a concerning link between exposure to IF-ACEs and the escalation of suicidal behaviors among adolescents. This finding concurs with an expanding body of literature highlighting the detrimental effect of ACEs on mental health outcomes. In a recent study conducted in the USA, involving individuals from sexual and gender minority groups, findings underscored that adults having higher ACE scores registered greater odds of mental distress [70]. The graded dose-response relationship highlights the cumulative influence of these encounters on the susceptibility to mental health disorders, hazardous conduct, and ultimately, suicidal behaviors. These findings are corroborated by recent investigations carried out in South Korea, the United States of America, and Ireland [71, 72, 73]. These observations bear crucial implications for clinicians, educators, and policymakers. Acknowledging the gender-specific dynamics of resilience and the pernicious effects of ACEs is pivotal. Tailoring interventions that foster coping strategies and social support systems, particularly within familial and educational contexts, could potentially bolster adolescent resilience and mitigate the fallout of adverse experiences. Further longitudinal studies are warranted to delineate the intricate interplay between gender, resilience, and mental health outcomes over time.

Several studies have investigated the link between ACEs and suicidality [54, 59, 74, 75]. However, analyzing mediators in this relationship has limited consideration in the literature. Regarding impulsivity, we found that impulsivity mediated exposure to emotional violence and suicidal behaviors among adolescents (mediation = 18.8%). It also mediated exposure to physical violence and suicidal behaviors in our sample (mediation = 17.4%). Our results are congruent with a study reporting a positive major indirect effect of ACEs on suicidality through impulsivity among adolescents in Finland [21]. Similarly, a 2020 survey suggested that impulsivity mediate the childhood trauma effect on suicidal behavior in patients with major depressive disorder in Spain [76]. Specifically, the indirect effect of childhood physical abuse on suicidality via attentional impulsivity was significant among patients with bipolar disorder in Greece [77]. Several explanations were found according to previous research on the developmental pathway from ACEs to suicidality through impulsive behaviors. In fact, ACEs have been observed to increase the suicidality risk via their influence on child brain structures. One of the major biological consequences of ACEs is the hypofunction of the serotonergic system [78]. Indeed, low serotonin function has been associated with higher impulsive behaviors among youth [21].

Regarding resilience as a protective mediator, interestingly, inter-personal resilience was a major mediator between childhood emotional violence and suicidal behaviors among youth in our sample (mediation = 24%). Similarly, a significant mediation effect between physical violence and suicidal behaviors via inter-personal resilience was found (mediation = 20.4%). A number of studies support our results. In fact, psychological resilience was a mediator (mediation = 29.6%) in the association between intra-familial childhood abuse and suicidal ideation among Chinese adolescents [58]. Similar to our findings, childhood emotional abuse had the higher mediation effect on suicidal ideation when comparing to other intra-familial ACEs in this survey [58]. Another recent survey showed the moderating influence of resilience on the relation between ACEs and mental health among students in the United States-Mexico Border region [79]. A possible explanation is that ACEs are linked to lower resilience level, which is related to higher mental health problems such as depression [80, 81]. Indeed, a 2022 survey among patients with inflammatory bowel disease in Canada [82] revealed that resilience mediated the association between childhood adversities and suicide by increasing depression risk. Otherwise, individuals with greater resilience score are more likely to recover from stressful events such as intra-familial childhood adversities [83]. Our findings highlight the urgent need to implement cost-effective strategies aimed at enhancing resiliency, especially among those exposed to childhood adversities in Tunisia.

Results of our survey should be viewed in light of some limitations. First, the cross-sectional type of the study provides no insight into the temporal nature of those associations. Second, well-known risk factors for suicidal behaviors include socioeconomic status, history of mental illness, or suicidal behavior in the family, which were not assessed in the current study. Third, it is important to note that we only evaluate schooled adolescents who tend to be at a lower risk of exposure to harm compared to unschooled youths so the results may not reflect the full weight of the problem. Finally, it should also be underlined that this study targeted two Tunisian governorates which may askew the representativeness of the results.

5. Conclusion

This is one of the first studies in Tunisia that investigated the role of both impulsivity and resilience in the relationship between IF-ACEs and suicidal behaviors in adolescents. Specifically, we found that both impulsivity and interpersonal resilience contributed heavily to the relationship between physical violence and suicidal behaviors (% mediated = 17.4 and 20.4%, respectively). Furthermore, we found that both variables also mediated the path between emotional violence and suicidal behaviors (% mediated = 18.8% for impulsivity and 24% for inter-personal resilience). This not only opens the path for further studies and screening for suicidal risk factors among youth, but also allows for future interventions to be focused on promoting and strengthening inter-personal resilience of vulnerable individuals. Our results suggest that adverse childhood experiences, impulsivity, and resilience are factors that should be taken into consideration when assessing suicidality among adolescents.

Acknowledgments

We seize this opportunity to express our gratitude to team members of The Epidemiology and Preventive Medicine Department at the University Hospital Taher Sfar Mahdia who participated in data collection.

This research did not benefit from any dedicated grants from funding agencies in the public, commercial, or nonprofit sectors.

Declaration of competing interest

The authors disclose that they have no known competing financial interests or personal relationships that might be perceived to influence this work.