Internalizing and Externalizing Symptoms Moderate the Relationship Between Emotion Dysregulation and Suicide Ideation in Adolescents

Abstract

The relationship between emotion dysregulation and suicide ideation may depend on the level of internalizing and externalizing symptoms. It was expected that both internalizing and externalizing symptoms would moderate the relationship between emotion dysregulation and suicide ideation, such that greater symptoms would strengthen the relationship between emotion dysregulation and suicide ideation. Adolescent participants (n = 559, Mage = 15.40, 85.0% white, 57.2% female) completed self-report measures that assessed emotion dysregulation, internalizing and externalizing symptoms, and recent suicide ideation. Both internalizing and externalizing symptoms moderated the relationship between emotion dysregulation and suicide ideation; this relationship strengthened as internalizing and externalizing symptoms increased. The results of this study indicate that internalizing and externalizing symptoms may both affect the connection between emotion dysregulation and suicide ideation. Future research should focus on targeting both internalizing and externalizing symptoms in treatment to help reduce emotion dysregulation and suicide ideation in adolescents.

Introduction

Suicide remains an urgent worldwide health concern, particularly for adolescents. In the United States, suicide is the 2nd leading cause of death among youth, and rates have increased 33% from 1997 to 2017 [1]. Suicide ideation is also widely prevalent [2, 3] and a significant predictor of suicide behaviors in adolescents [4]. Between 22% and 38% of adolescents report having experienced suicide ideation in their lifetime [5], with between 15% and 19% reporting suicidal thoughts in the past year [6]. Additionally, over one-third of adolescents who report suicide ideation transition to a suicide plan and/or attempt, and 60% of those who transition from ideation to attempt do so within one year of first experiencing suicide ideation [3, 7]. Therefore, it is critical to understand aspects related to suicide ideation in adolescents in order to improve suicide prevention for this specific age group. Emotion dysregulation, or an inability to manage the intensity and duration of negative emotions [8], has consistently been tied to increased suicide ideation in adolescents [9, 10]. However, the way in which emotion dysregulation interacts with other risk factors and associates with greater suicide ideation is less well understood, especially in this age group.

One potential factor that may impact the relationship between emotion regulation and suicide ideation is the presence of internalizing and externalizing symptoms. Internalizing symptoms typically refer to inwardly directed problems such as “disordered mood, withdrawal, anxiety or depression” and are often conceptualized through anxiety and depression symptoms, and externalizing have been classified as delinquent and aggressive behaviors, such as disobedience, lying, fighting, stealing, destruction and cruelty [11]. Both internalizing and externalizing symptoms have been shown to be predictive of suicidality in adolescents, including suicide ideation [12, 13], and are also tied to emotion dysregulation in adolescents [14, 15]. It is possible that adolescents who have difficulties regulating their emotions are more likely to develop internalizing or externalizing symptoms in response to distressing or stressful life events when healthy coping mechanisms are not available, which in turn could increase their emotion dysregulation and subsequently increase their risk for suicide ideation [16]. This study aimed to clarify the relationships between emotion dysregulation, internalizing and externalizing symptoms, and suicide ideation in a community adolescent sample.

Emotion Dysregulation and Suicide Ideation

Lack of access to emotion regulation strategies and difficulty identifying and accepting feelings have been linked to increased suicide ideation in adolescents [9, 10], while aspects such as acceptance of emotions, emotional self-efficacy, and regulation of impulsivity have been linked to decreased ideation in adolescents [10, 17]. Additionally, having fewer adaptive responses and less emotion regulation skills have been found to predict suicide ideation, plans, and attempts among adolescents and children [18, 19]. Thus, clarifying what other factors may affect how emotion dysregulation is associated with suicide risk can provide key insight into understanding and preventing suicide ideation in adolescents, which is particularly important when considering that suicide ideation is strongly tied to suicide behaviors [7].

Internalizing and Externalizing Symptoms and Suicide Ideation

In addition to emotion dysregulation, suicide ideation is tied to internalizing symptoms such as depression, anxiety, and rumination [20, 21], and externalizing symptoms such as hyperactivity, conduct problems, and substance abuse [22-24]. However, some studies have found internalizing symptoms to be more strongly associated with suicide ideation than externalizing symptoms (e.g. [25, 26]), indicating symptoms may differentially relate to ideation. One possible explanation for differential relationships with suicide ideation is the interactive effects of emotion dysregulation. For example, some studies suggest that externalizing symptoms are predictive of suicide ideation largely because the associated behaviors associated (e.g. substance use) exacerbate the negative emotions associated with internalizing symptoms [27]. However, previous studies were limited in that they focused only on clinical samples [20, 23, 25], did not assess recent (past month) suicide ideation [9, 22, 27], did not examine both internalizing and externalizing symptoms in the same sample [21], and/or used limited conceptions of internalizing and externalizing symptoms [24, 26).

Internalizing and Externalizing Symptoms in the Emotion Dysregulation and Suicide Ideation Relationship

As emotion dysregulation is tied to higher use of negative coping mechanisms, such as the mechanisms often present in internalizing and externalizing disorders [14, 15], it is possible that internalizing and externalizing symptoms may strengthen the emotion dysregulation-suicide ideation relationship. Specifically, internalizing symptoms often involve elements of poor emotion regulation, and some adolescents also respond to poor emotion regulation with internalizing symptoms, suggesting there may be an interactive effect that associates with more severe suicide ideation. For example, depressive symptoms often involve feelings of non-acceptance or unawareness of emotions [28, 29], both of which are aspects of emotion dysregulation [30]. Experiencing depressive symptoms in response to emotion dysregulation may increase the severity, amount, and duration of negative emotions, consequently leading to an increased inability to regulate negative emotions and higher risk of suicide ideation [31]. However, internalizing symptoms such as isolation or withdrawal may also serve as a response to emotion dysregulation, which also increases the risk of suicide ideation [32]. In adolescents, the presence of internalizing symptoms has been shown to increase along with emotion dysregulation; a 2017 meta-analysis demonstrated maladaptive emotion regulation responses (such as avoidance, rumination, and suppression) were significantly and positively related to depression and anxiety symptoms [33].

In comparison to internalizing symptoms, externalizing symptoms are conceptualized as separate, behavioral responses to the negative feelings resulting from emotion dysregulation [34, 35]. While some adolescents respond to emotion dysregulation with internalizing symptoms, others may be more likely to respond with externalizing symptoms. For instance, adolescents who are experiencing deficits in emotional awareness, emotional clarity and impulse control are more likely to report externalizing behaviors such as stealing, physical fighting [36], and risky sexual behaviors [37]. A 2016 meta-analysis found that adolescents’ emotion regulation issues, including disengagement coping and difficulties with problem-solving, increased as their levels of externalizing symptoms increased [38]. Additionally, in samples of school-aged children, lower emotion regulation skills in aspects of shifting and refocusing were associated with higher externalizing behaviors [39]. As both higher levels of emotion dysregulation and externalizing symptoms are associated with increased suicide risk (e.g. [10, 22, 40]), these findings suggest that use of externalizing behaviors in place of healthy coping strategies may also strengthen the relationship between emotion dysregulation and suicide ideation. However, a few of these studies were limited in that they did not use multiple items to assess recent suicide ideation among community adolescents [21, 23, 32], and no studies assessed recent suicide ideation, internalizing and externalizing symptoms, and emotion dysregulation among the same community adolescent sample.

Objective

This study sought to fill these gaps and clarify past findings by examining the moderating roles of both internalizing and externalizing symptoms in the relationship between emotion dysregulation and recent suicide ideation among a community sample of adolescents, while using a validated measure of recent suicide ideation. Particularly as adolescents are at heightened risk for suicide [5], it is imperative that research focus on this specific population to address their unique risk factors. Additionally, as suicide rates among adolescents have continued to increase over the years [1], there are undoubtedly gaps in clinical knowledge for how best to treat this population, and in understanding the underlying psychological mechanisms that increase risk among adolescents. It is possible that understanding the role of internalizing and externalizing symptoms in the emotion dysregulation-suicide ideation association could enhance clinical knowledge and intervention for adolescents. Internalizing symptoms (depression and anxiety symptoms) and externalizing symptoms (hyperactivity symptoms and conduct problems) were examined as moderators in the relationship between current emotion dysregulation and past-month suicide ideation severity in a sample of high school students. It was predicted that as internalizing and externalizing symptoms increased, the relationship between emotion dysregulation and suicide ideation would strengthen.

Design & Methods

Participants

Data were obtained from 695 adolescents recruited from three public high schools in the south-central region of the United States. Approximately 3,000 parent consent forms were distributed across schools, and 794 (26%) were returned with written positive consent. All students with positive parent consent were approached at school to participate in the study. Among those recruited, 695 (88%) provided written assent and participated in the study. Only participants with complete data were included in study analyses (n = 559). The mean age was 15.40 years (SD = 1.95); the sample was primarily female (57.2%) and white (85.0%), with 4.3% identifying as Black/African American, 4.8% as Hispanic/Latino(a), 3.4% as Asian, 4.8% as multi-ethnic, 0.2% as Native American, and 0.4% as other. The majority of the sample identified as heterosexual (87.8%) with 1.8% identifying as gay, lesbian or queer, 4.8% as bisexual, 2.5% as unsure and 1.4% as other. Only 0.6% reported a gender identity outside the categories of cisgender boy/girl. About 24% reported ever having experienced suicidal thoughts. No significant differences were found between those with complete and incomplete data on demographics or study variables.

Measures

Difficulties in Emotion Regulation.

To assess current levels of emotion dysregulation, the Difficulties in Emotion Regulation Scale (DERS) [30] was used. The DERS is a 36-item, self-report questionnaire designed to assess multiple aspects of emotion dysregulation. The measure yields a total score and 6 subscale scores including lack of emotional awareness (AWARENESS; α = 0.80; i.e., “I am attentive to my feelings” --reverse scored), difficulties engaging in goal directed behavior (GOALS; = 0.89; i.e. “When I’m upset, I have difficulty concentrating”), nonacceptance of emotional response (NONACCEPTANCE; α = 0.85; i.e., “When I’m upset, I feel like I am weak”), limited access to ER strategies (STRATEGIES; α = 0.88; i.e., “When I’m upset, I believe there is nothing I can do to make myself feel better”), impulse control problems (IMPULSE; α = 0.86; “When I’m upset, I have difficulty controlling my behavior”) and lack of emotional clarity (CLARITY; α = 0.84; i.e., “I am confused about how I feel”). Responses are scored on a 5-point Likert scale ranging from almost never (1) to almost always (5), and 11 of the items are reverse scored. Items are summed for each subscale and the total score (range 36–180), and higher scores are indicative of greater emotion dysregulation. This scale has been found to be psychometrically sound and a valid tool for an assessment of ER difficulties [41]. Internal consistency for the total score (α=0.95) and each subscale (αs = 0.79–0.91) in the current sample was good.

Strengths and Difficulties Questionnaire.

To assess symptoms of externalizing symptoms, the Strengths and Difficulties Questionnaire (SDQ) [42] was used. The SDQ is a 25-item self-report measure that is typically divided into five subscales: emotional problems, conduct problems, hyperactivity/inattention, peer relationship problems, and prosocial behaviors. However, research has shown that in lower risk or nonclinical samples, it may be more valid to use an alternative, three subscale division [43] including internalizing problems (emotional and peer problems; 10 items; α = 0.71; “I worry a lot”), externalizing symptoms (conduct and hyperactivity symptoms; 10 items; α = 0.81; “I get very angry and often lose my temper”), and prosocial behaviors (5 items; α = 0.70; “I try to be nice to people. I care about their feelings”). The SDQ assesses these behaviors in the last six months and is scored on a three-point scale from 0 (not true) to 2 (certainly true). This study aimed to conceptualize internalizing symptoms through depression and anxiety symptoms; thus, only the externalizing subscale from the SDQ was used. Items for externalizing symptoms were summed on a scale from 0 to 20, with higher scores indicating greater externalizing symptoms. The scale has been found to be psychometrically sound and a valid tool for assessment of behavioral symptoms [44]. Internal consistency for the total score (α=0.75) and the externalizing subscale (α=0.69) was adequate.

Beck Anxiety Inventory.

To assess anxiety issues indicative of internalizing symptoms, the Beck Anxiety Inventory (BAI) [45] was used. The BAI is a 21-item self-report measure designed to assess anxiety levels in the past month. Items are rated on a four-point scale ranging from 0 (not at all) to 3 (severely- it bothered me a lot). Items are summed for total scores ranging from 0 to 63, with a score of 22–35 indicating moderate anxiety, and a score of 36 or higher indicating concerning levels of anxiety; 11.5% of the sample reported moderate levels of anxiety and 6.9% reported high levels of anxiety. The BAI has been found to be psychometrically sound and a valid tool for an assessment of anxiety (α=0.92) [45]. Internal consistency in the current sample was good, α=0.95.

Center for Epidemiologic Studies Depression Scale for Children.

To assess depression symptoms as part of internalizing symptoms, the Center for Epidemiologic Studies Depression Scale for Children (CES-DC) [46] was used. The CES-DC is a 20-item self-report measure designed to assess recent depression symptoms in children ages 6 through 17. Sample items include “I felt down and unhappy” and “I didn’t sleep as well as I usually sleep.” Items are rated on a 4-point scale ranging from 0 (not at all) to 3 (a lot). Items are summed for total scores ranging from 0 to 60, with a clinical cut-off score of 15; 10.8% of the sample scored above the cut-off. This measure is recommended for assessment of depressive symptoms in adolescents by the PhenX Toolkit, a collection of expert-recommended measures endorsed by several funding agencies to increase common data elements across studies. The CES-DC has been found to be psychometrically sound and a valid tool for an assessment of depression (α=0.84) [47]. Internal consistency in the current sample was good, α=0.94.

Demographics.

Participants completed a measure that collected information on gender, sexual orientation, age, and race/ethnicity.

Outcomes

Suicide Ideation Questionnaire- Junior.

To assess recent suicide ideation, the Suicide Ideation Questionnaire Inventory- Junior (SIQ-JR) [48] was used. The SIQ-JR assesses suicide ideation severity in the past month with 15 items. Sample items include “I thought it would be better if I was not alive” and “I thought that killing myself would solve my problems.” Items are rated on a seven-point scale ranging from 0 (I never had this thought) to 6 (almost every day). Items are summed for a total score ranging from 0 to 90. The clinical cut-off indicating significant severity of suicide ideation is 31; 3% of the sample scored above the cut-off. The SIQ-JR has been found to be psychometrically sound and is a valid tool for assessing suicide ideation in adolescents (α=0.94–0.97) [49]. Internal consistency in the current sample was good, α=0.94.

Procedure

The research team visited high schools on multiple occasions to collect data from students with parent consent. Eligible students were gathered in the library or a classroom in small groups of 12. After learning about the study and signing the assent form, students were either given a packet of self-report measures to complete by hand, which included measures of suicide ideation, depression, anxiety, externalizing symptoms, and emotion dysregulation, or an iPad to complete additional behavioral and self-report measures. After completing the packet or iPad measures, students switched to complete the measures in the other format. Data collection took between 30 and 60 minutes. When participants had finished both tasks, they were given a debriefing form with information on mental health resources and received $5 in compensation. Before leaving the school, the research team screened all completed measures for indications of recent suicide ideation or attempts. If participants indicated suicide behavior in the past six months and/or responded to two or more critical items on the SIQ-JR, they were referred to the school guidance or crisis counselors as either being high, medium or low risk. Information about possible referral to counselors was noted in the assent forms. Among the total sample, 17% were referred as at-risk; among them, 15% were high risk, 47% were moderate risk, and 38% were low risk. Counselors followed up with at-risk students in accordance with school protocol. This study received approval from the Institutional Review Board at the university where both authors are affiliated.

Data management/analytic plan

Means, standard deviations, and correlations between all study variables are shown in Table 1. To test hypotheses that internalizing symptoms of depression and anxiety and externalizing symptoms would moderate the relationship between emotion dysregulation and suicide ideation, three moderation analyses were run using the PROCESS Macro for SPSS [50]. In all analyses, emotion dysregulation was entered as the predictor (DERS total score) while suicide ideation severity (SIQ-JR total score) was entered as the outcome variable. Separate analyses were run for each moderator variable (CESD-CD total score for depression; BAI total score for anxiety; externalizing subscale from SDQ). Simple slopes analyses were conducted for significant interactions, using +/− 1 SD from the mean for high and low levels of the moderator. To control for Type 1 error, the p-value was adjusted for significance to p < .01.

Table 1.

Means, standard deviations, and correlations between all study variables

Variable	M	SD	Range	1	2	3	4	5
1. Depression Symptoms	16.61	13.30	0–59	-
2. Anxiety Symptoms	11.89	12.79	0–58	0.75**	-
3. Externalizing Symptoms	7.07	2.48	0–14	0.42**	0.44**	-
4. Emotion Dysregulation	84.65	19.96	36–124	0.60**	0.59**	0.42**	-
5. Suicide Ideation	5.97	9.81	0–75	0.63**	0.57**	0.32**	0.45**	-

Note. Suicide Ideation is the total score from the Suicide Ideation Questionnaire-Junior; Emotion Dysregulation is the total score from the Difficulties in Emotion Regulation Scale; Depression Symptoms are the total score from the Center for Epidemiological Studies Depression Scale for Children; Anxiety Symptoms are the total score from the Beck Anxiety Inventory; Externalizing Symptoms is subscale from the Strengths and Difficulties Questionnaire

^*p < .05

^**p < .01

Results

Moderation Results for Internalizing Symptoms

The overall model examining depression symptoms as the moderator was significant, and accounted for 40.7% the variance, F(3, 574) = 131.16, p < .0001. Neither depression symptoms nor emotion dysregulation were significant predictors of suicide ideation. However, the interaction was significant (B = 0.003, t = 2.96, p = .003). Simple slope analysis of the interaction showed that the association between emotion dysregulation and suicide ideation was significant only when depression symptoms were high (coeff = 0.08, t = 3.68, p = .0003, [95% CI: 0.04, 0.13 ]). See results in Table 2; Fig. 1.

Table 2.

Results for internalizing and externalizing symptoms as moderators between emotion dysregulation and suicide ideation

INTERNALIZING SYMPTOMS	Total R 2	F	B	t
Moderator: Depression Symptoms	0.407	131.16**
Emotion Dysregulation (ED)			0.030	− 0.541
Depression Symptoms			0.116	0.862
ED x Depression Symptoms			0.001**	2.96
Moderator: Anxiety Symptoms	0.349	98.75**
Emotion Dysregulation (ED)			0.0264	1.36
Anxiety Symptoms			0.123	− 0.217
ED x Anxiety Symptoms			0.001**	3.19
EXTERNALIZING SYMPTOMS
Moderator: Externalizing Symptoms	0.238	59.17**
Emotion Dysregulation (ED)			0.054	0.978
Internalizing Symptoms			−1.01	−1.79
ED x Internalizing Symptoms			0.018**	2.86

Note. Suicide Ideation is the total score from the Suicide Ideation Questionnaire-Junior; Emotion Dysregulation is the total score from the Difficulties in Emotion Regulation Scale; Depression Symptoms are the total score from the Center for Epidemiological Studies Depression Scale for Children; Anxiety Symptoms are the total score from the Beck Anxiety Inventory; Externalizing Symptoms is a subscale from the Strengths and Difficulties Questionnaire

^*p < .05

^**p < .01

Fig. 1.

Moderation of emotion dysregulation and suicide ideation by depression symptoms.

The overall model examining anxiety symptoms as the moderator was significant, and accounted for 34.9% the variance, F(3, 553) = 98.75, p < .0001. Neither anxiety symptoms nor emotion dysregulation were significant predictors of suicide ideation. However, the interaction was significant (B = 0.004, t = 3.190. p = .002). Simple slope analysis of the interaction showed that the association between emotion dysregulation and suicide ideation was significant when anxiety symptoms were moderate (coeff = 0.06, t = 2.79. p = .006 [95% CI: 0.02, 0.11]) and more so when anxiety symptoms were high (coeff = 0.13, t = 5.l7, p < .0001, [95% CI: 0.08, 0.18]; see Table 2; Fig. 2).

Fig. 2.

Moderation of emotion dysregulation and suicide ideation by anxiety symptoms.

Moderation Results for Externalizing Symptoms

The overall model examining externalizing symptoms as the moderator was also significant and accounted for 23.8% the variance, F(3, 570) = 59.17, p < .0001. Neither emotion dysregulation nor externalizing symptoms were significant predictors of suicide ideation, but the interaction was significant (B = 0.018, t = 2.86, p = .005). Simple slope analysis of the interaction showed that the association between emotion dysregulation and suicide ideation was significant when externalizing symptoms were low (coeff = 0.14, t = 5.34, p < .0001, [95% CI: 0.09, 0.20]), moderate (coeff = 0.18, t = 8.74, p < .0001, [95% CI: 0.14, 0.22]), and high (coeff = 0.23, t = 9.72, p < .0001, [95% CI: 0.19, 0.28]). Increased emotion dysregulation associated with increased suicide ideation severity at all levels of externalizing symptoms (see Table 2; Fig. 3).

Fig. 3.

Moderation of emotion dysregulation and suicide ideation by externalizing symptoms.

It is important to note that this study utilized total emotion dysregulation scores rather than individual components of emotion dysregulation. As previous research has focused more on general issues with emotion dysregulation in relation to suicide ideation, internalizing symptoms, and externalizing symptoms, examining individual subscales of emotion dysregulation was outside the scope of this research study. However, future research should include individual emotion dysregulation subscales in analyses to examine how the various dimensions of emotion dysregulation may interact with internalizing and externalizing symptoms and suicide ideation.

Discussion

In our sample of community adolescents, both internalizing and externalizing symptoms moderated the relationship between emotion dysregulation and suicide ideation. This relationship was significant when anxiety symptoms were at moderate and high levels, and at high levels of depression symptoms. In comparison, the relationship between emotion dysregulation and suicide ideation was significant at all levels of externalizing symptoms. These findings indicate that both types of symptoms are relevant to suicide ideation, which is consistent with past findings (e.g. [22, 23, 51-53]). However, the results also indicate that externalizing symptoms may be more of a risk factor for suicidal thoughts for adolescents, which is in contrast to this study’s hypotheses and is inconsistent with previous findings (e.g. [25, 26]). However, internalizing symptoms still significantly moderated the emotion dysregulation and suicide ideation relationship, which is consistent with prior research that shows suicide ideation is strongly predicted by levels of depression and anxiety, emotion-based responses to distress, and withdrawal/isolation [45, 54, 55]. Additionally, while the current results align with past research that has found both emotion dysregulation and internalizing symptoms to be related to past-year and recent suicide ideation when examined as separate variables [9, 54, 56], this study demonstrated the interactions between emotion dysregulation and internalizing/externalizing symptoms in the context of recent suicide ideation severity.

The finding that both internalizing and externalizing symptoms moderated the relationship between emotion dysregulation and suicide ideation suggests that when adolescents are unable to access healthy emotion regulation skills, they may be more likely to respond to negative emotions in maladaptive ways. These maladaptive responses seem to associate with thoughts of suicide, regardless of if those responses are directed inwardly or outwardly. These results add to the existing literature that shows higher levels of internalizing and externalizing symptoms can increase suicidality when interacting with emotion dysregulation [22, 23, 51-53]. However, past research has examined these behaviors in relation to past-year [22] or lifetime (23, 51, 53] suicide ideation; this study examined these relationships in respect to recent suicide ideation severity. Furthermore, this study expands on past research by exploring the relationships between internalizing and externalizing symptoms with emotion dysregulation and recent suicide ideation specifically. These differences fill necessary gaps in the research; it is imperative to understand the individual and unique relationships internalizing and externalizing symptoms have with emotion dysregulation and suicide ideation to tailor suicide intervention and prevention plans that can target the presenting set of symptoms.

The results suggest that adolescents who experience internalizing symptoms may need different intervention and prevention strategies than adolescents who experience externalizing symptoms. It may be important to prioritize the treatment of emotion dysregulation and concurrent externalizing behaviors. Additionally, it is essential we understand these relationships in respect to recent suicide ideation, so that intervention effects can target symptoms indicative of acute risk. Understanding these relationships in the context of recent ideation may allow clinicians to more accurately identify proximal risk factors for current or near-future suicide ideation and behaviors. Furthermore, it is significant that the current results are focused specifically on the adolescent population. As adolescence is a time of increased risk for suicide ideation and behaviors, understanding the underlying mechanisms for this heightened risk will allow clinicians to tailor prevention and intervention efforts to the adolescent population.

Prior research suggests that internalizing symptoms are more strongly tied to both emotion dysregulation [57] and suicide ideation [54] than externalizing symptoms. Specifically, depression and anxiety symptoms have been found to be stronger predictors of emotion dysregulation [57] and suicide ideation [54] than conduct and attention issues. Based on existing research, we expected internalizing symptoms to have more of an effect on the emotion regulation-suicide ideation relationship, but our findings suggest the opposite. Emotion dysregulation significantly associated with suicide ideation severity at all levels of externalizing symptoms, while anxiety and depression symptoms moderated the relationship at moderate and high or high levels only, respectively. It is possible that adolescents who experience externalizing symptoms are more impulsive and reactive than adolescents with internalizing symptoms [39, 58]. Because depression and anxiety symptoms involve aspects of withdrawal and isolation, and people with depression and anxiety symptoms often report feeling numb [59], individuals who respond to distress with externalizing symptoms may be more reactive to negative emotions.

Additionally, it is also possible that extrinsic consequences of externalizing symptoms are more severe than extrinsic consequences of internalizing behaviors, leading to a stronger desire to escape through suicide ideation. For example, while withdrawing from friends and family at a minimal level may not significantly impact the social standing or support levels of an adolescent, engaging in any type of externalizing behavior even at a minimal level, such as abusing a substance or acting out in class, may result in direct consequences (e.g. suspension, hostile reactions from parents or peers) [60]. These consequences, in turn, may drive adolescents to want to escape their current situations, prompting them to consider suicide [61, 62]. Along those lines, adolescents who are unable to access emotion regulation skills and use externalizing behaviors as a coping response to negative emotions, but then have that coping response removed or blocked through punishment or intervention, may feel hopeless and come to view suicide as their only option [63, 64]. Further, as externalizing symptoms are more likely to be dangerous, public and draw attention [11], engaging in even low levels of externalizing symptoms may indicate an underlying disregard for personal well-being, and/or signify a cry for help. In comparison, it may be that low levels of anxiety and depression symptoms are more common in adolescence and do not necessarily indicate a desire to hurt oneself. It is when emotion dysregulation becomes more consistent, and these feelings become more frequent or severe (indicating more moderate or high levels of anxiety or depression symptoms), that adolescents may start to think about suicide [65]. Thus, clinical programs may want to consider implementing differently tailored prevention and intervention plans for adolescents, depending on the type and severity of internalizing or externalizing symptoms.

Interestingly, this study found that emotion dysregulation predicted suicide ideation at moderate and high levels of anxiety, but only at high levels of depression. This is unexpected considering that anxiety and depression were highly correlated in this study (r = .75), have been significantly linked in adolescents in previous research (e.g., [66, 67]), and are both are classified as internalizing symptoms [68]. This finding may be because of the heightened arousal and panic associated with anxiety [45], and if these symptoms are experienced in response to a stressful event, it may be more likely to worsen rumination, catastrophic thinking, and feelings of hopelessness [44, 64], and consequently increase suicide ideation [69]. In comparison, if adolescents experience moderate depression symptoms in place of healthy coping skills, the moderate symptoms may work to numb or detach the adolescent [59], preventing them from spiraling into suicide ideation. It may be that it is only when adolescents are experiencing severe depression symptoms in addition to poor emotion regulation skills that they feel completely hopeless, withdrawn, and upset, increasing the likelihood of using suicide thoughts as a means of coping [70]. Also, suicide ideation outcomes may differ depending on which specific emotion regulation skills are failing. Future studies should examine the specific emotion regulation subscales in relation to anxiety and depression symptoms and suicide ideation in order to clarify their relationships and strengthen clinical prevention and intervention plans.

Limitations

Despite several strengths of the study, there are some limitations that need to be addressed. First, the study lacked diversity both in terms of race/ethnicity, sexuality, and gender identity, which limits the generalizability of study results. Particularly because LGBTQ adolescents and black, indigenous and people of color (BIPOC) are underrepresented in suicide research [71, 72] and may be at higher risk for suicidality [73, 74], it is especially important that these groups are included in research that could highlight potential suicide intervention and prevention strategies. Future studies should replicate this study with a more diverse sample. Additionally, the measures used to assess internalizing and externalizing symptoms failed to encompass all elements of related symptomatology, such as substance use, stealing, phobias, and obsessions. Future studies should include more comprehensive measures of internalizing and externalizing symptoms. Further, the study only utilized self-report measures and thus may have influenced results through responses bias and memory issues. Future studies should incorporate a mix of self-report and parent/teacher report measures, particularly to assess internalizing and externalizing symptoms. The study is also limited by its cross-sectional design; it is important that future studies use longitudinal designs to examine how increasing, decreasing or consistent levels of emotion dysregulation and internalizing/externalizing symptomatology interact to impact levels of suicide ideation and to evaluate directionality of the relationships. Similarly, this study examined the relationships between internalizing and externalizing symptoms, suicidality and emotion dysregulation in relation to recent suicide ideation severity. However, this study did not examine these relationships with past-year or lifetime suicidality, thus limiting the generalizability of findings with other studies. Lastly, future research should account for the comorbidity of internalizing and externalizing symptoms and attempt to expand on the relationships between internalizing and externalizing symptoms and emotion dysregulation with different aspects of suicidality, in order to tailor interventions for adolescents experiencing both types of symptoms, and further understand how internalizing and externalizing symptoms interact with to the development and progression of suicide.

Conclusion

The current findings suggest prevention and intervention plans should target both internalizing and externalizing symptoms, including symptoms of depression, anxiety, hyperactivity, and conduct problems, and include teaching and practice of emotion regulation strategies. Longitudinal studies and a more comprehensive assessment of internalizing and externalizing symptoms may demonstrate that having emotion regulation skills is protective against the development of maladaptive symptoms and suicide risk.

Summary

This study explored the relationships between emotion dysregulation, internalizing and externalizing symptoms, and suicide ideation in non-clinical adolescents. Adolescent participants (n = 559, Mage = 15.40, 85.0% white, 57.2% female) completed self-report measures that assessed emotion dysregulation, internalizing and externalizing symptoms, and recent suicide ideation. Both internalizing and externalizing symptoms moderated the relationship between emotion dysregulation and suicide ideation; this relationship strengthened as internalizing and externalizing symptoms increased. Externalizing symptoms were significant at low, moderate and high levels, while internalizing symptoms were significant at moderate and high levels. Because externalizing symptoms were consistently related to suicide ideation across all levels of symptoms, this study suggests externalizing symptoms may be more relevant to the emotion dysregulation and suicide ideation relationship among community adolescents. However, internalizing symptoms also significantly moderated the relationship between emotion dysregulation and suicide ideation, although anxiety and depression symptoms moderated the relationships to different extents. It may be that anxiety is more predictive of suicide ideation when experienced in conjunction with emotion dysregulation; future studies should closely examine differences between anxiety and depression in relation to emotion dysregulation and suicide. Future suicide intervention and prevention efforts should focus on both internalizing and externalizing symptoms.

Data Availability

Data was uploaded and shared through the NIMH Data Archive.