Abstract
In the wake of several highly publicized adolescent suicides attributed to bullying victimization, national attention has been brought to bear on the profound public health problem of bullying. This article reviews the extant literature on the associations between bullying perpetration, victimization, and thoughts of or attempts at suicide and proposes five potential mediators, namely depression, anxiety, low self-esteem, loneliness, and hopelessness, that may explain this relationship. Numerous studies have found empirical support for the interrelations between internalizing behaviors and both bullying perpetration and victimization and suicide. We find that further longitudinal research needs to be conducted to more conclusively determine the role and causal ordering these various psychosocial factors may play in bullying perpetration, victimization, and suicide. Although the research literature implies causal directions among all these potential mediators, untangling the unique influence of bullying perpetration, victimization, and bully victimization on suicide and its mechanisms of action has major research and practice implications.
Key Points of the Research Review
Children and adolescents who are involved in bullying are at a heightened risk of suicide.
Depression is the most strongly associated psychosocial correlate of bullying involvement, which is also associated with suicide.
Bullying victims are also at an increased risk of anxiety, low self-esteem, loneliness, and hopelessness, which have been identified as vulnerability factors for suicide.
The causal directions among the proposed mediators between bullying involvement and suicide have been implied in prior research, but future longitudinal studies are still needed to disentangle the potential bidirectional relationship between bullying involvement and psychosocial factors.
Exploring the unique influence of these potential mechanisms on bullying perpetration, victimization, and bully victimization on suicide has important research and practice implications.
Introduction
Children and adolescents who are involved in bullying as bullies, victims, or bully victims (i.e., involved in both bullying perpetration and victimization) are significantly more likely to contemplate or attempt suicide, compared to youth who do not assume one of these three bullying roles (Klomek, Sourander, & Gould, 2010). A nationally representative study of students in Grades 6–10 in the United States (N = 15,686) found 29.9% reported some type of involvement in moderate to frequent bullying as a bully only (13.0%), victim only (10.6%), or bully victim (6.3%; Nansel et al., 2001). The Centers for Disease Control and Prevention (2010) have identified suicide as the third leading cause of death in the United States among adolescents aged 15–19, with 14.5% and 6.9% of 9th- and 12th-grade students, respectively, reporting contemplating suicide and at least one attempt within the previous year (Cash & Bridge, 2009). Although these rates call for effective prevention and intervention efforts, little research has been conducted to date on the potential mechanisms underlying the associations among these phenomena.
Bullying is defined as a subtype of aggression that involves repeated exposure to negative actions with the intent to cause harm or discomfort and an imbalance of power (e.g., physical, social, cognitive) between the bully and victim (Olweus, 1990, 1993). This construct has proved challenging to measure, with few studies simultaneously assessing its systematic, repetitive, and power dynamics (see Olweus, 2013). In addition to these measurement challenges, bullying is further categorized into two major subtypes: direct (physical and verbal) and indirect (relational and cyberbullying). Furthermore, bullying is not simply a binary construct with youth assuming a bully-only or victim-only role. Some youth are referred to as “bully victims” because they are both bullies and victims. Role type is an important consideration for future research and practice as bully victims consistently report the poorest mental health and academic outcomes, compared to their bully-only and victim-only counterparts (Nansel et al., 2001; Stein, Dukes, & Warren, 2007).
Although all three of these populations are at higher risk for suicide compared to their uninvolved peers (Klomek et al., 2010), it is important to understand that not all of these youth will become suicidal. Rather, they may possess different pathways that increase their risk. Although research findings underscore a major need to understand the association between bullying and suicide, very few studies have investigated the pathways between bullying perpetration, victimization, and likelihood of suicidal thoughts or attempts. A prior review conducted by Hong, Espelage, Grogan-Kaylor, and Allen-Meares (2012), which investigated potential mediators linking the experiences of child maltreatment at home to later bullying perpetration or victimization outside the home, has major implications for the child welfare system and school-related practice. Given that child maltreatment appears to be significantly associated with bullying (Hong, Espelage, Grogan-Kaylor, & Allen-Meares, 2012), recognizing the mechanisms that interconnect maltreatment, bullying perpetration, victimization, and suicide is an important first step toward preventing both bullying and suicide, as well as developing interventions. The extant literature points toward the complex causal ordering of these events with (a) some youth experiencing child maltreatment and its related psychosocial consequences (e.g., depression, anxiety, emotional dysregulation) prior to the onset of bullying perpetration, victimization, and suicidal thoughts or behaviors and (b) other youth experiencing psychosocial consequences solely as a result of their involvement as a bully, victim, or both.
Building on Hong et al.’s (2012) review, this article aims to explore the latter phenomenon and examine a number of potential mediating factors that may explicate the potential link among bullying perpetration, victimization, and suicide among youth. Although the causal directions among the potential mediating influences of internalizing behaviors have been implied in empirical studies, untangling the influences of bullying perpetration and victimization on suicidal thoughts and behaviors and examining the bidirectional mechanisms are imperative for future longitudinal research.
What Do We Know About Bullying Involvement and Suicide?
Research findings consistently indicate that victims and bully victims manifest the highest levels of suicidal thoughts and behaviors (Hepburn, Azrael, Molnar, & Miller, 2012; Hidaka et al., 2008; Kaminski & Fang, 2009; Kiriakidis, 2008; Klomek, Marrocco, Kleinman, Schonfeld, & Gould, 2007a; Liang, Flisher, & Lombard, 2007; Skapinakis et al., 2011; Viljoen, O’Neill, & Sidhu, 2005). Kaminski and Fang (2009) found from a nationally representative sample of adolescents (Grades 7–12) that victims were significantly more likely to display suicidal ideation than uninvolved peers. Controlling for differences in sociodemographic factors (i.e., age, gender, race/ethnicity) and depressive symptoms, the researchers found that youth who were frequently victimized by their peers were 1.67–3.83 times more likely to exhibit suicidal behavior. Another study, which included a sample of 152 adolescent offenders detained in Scotland, also reported that youth who were bullied while in custody were 9.22 times more likely to attempt suicide, compared to uninvolved youth (Kiriakidis, 2008).
Emerging retrospective research findings also suggest that having been bullied during childhood is linked to suicidal behavior in adulthood (Meltzer, Vostanis, Ford, Bebbington, & Dennis, 2011; Roeger, Allison, Korossy-Horwood, Eckert, & Goldney, 2010; Staubli & Killias, 2011). For instance, Meltzer, Vostanis, Ford, Bebbington, and Dennis’s (2011) research study, which examines the association between bullying victimization during childhood and suicide attempts over the life course from a random sample of 7,461 respondents in United Kingdom, reports that adults who were bullied during childhood were more than twice as likely as uninvolved adults to attempt suicide later in life.
Bullies themselves are also significantly at higher risk of suicidal behavior, compared to uninvolved youth (Hinduja & Patchin, 2010; Kim, Koh, & Leventhal, 2005; Kim, Leventhal, Koh, & Boyce, 2009; Klomek et al., 2008, 2010; Luukkonen, Rasanen, Hakko, Riala, & STUDY-70 Workgroup, 2009; Roland, 2002). Interestingly, preliminary findings suggest gender may play an important role in influencing the relationship between bullying perpetration and suicide. Klomek et al. (2009) found, from a sample of 5,302 Finnish children, that male bullies were more likely to exhibit suicidal behavior compared to female bullies and victims. However, Evans, Marte, Betts, and Silliman (2001) reported contradictory findings, with high levels of suicide risk among bullies and victims of both genders.
A recent review also revealed that, when examining the relationship between bullying perpetration, victimization, and suicide, most researchers did not control for other suicide risk factors, such as depression (Kim & Leventhal, 2008). While some studies have examined internalizing behaviors such as depressive symptoms (Kaltiala-Heino, Rimpela, Marttunen, Rimpela, & Rantanen, 1999; Kaminski & Fang, 2009; Kim et al., 2005; Park, Schepp, Jang, & Koo, 2006; Roland, 2002), only a handful of studies have examined other types of behaviors, such as anxiety (Kim et al., 2005) and low self-esteem (Park et al., 2006). Although these researchers have identified an association between bullying victimization and suicidal behavior, they have treated these internalizing behaviors as covariates. For instance, Kaltiala-Heino, Rimpela, Marttunen, Rimpela, and Rantanen (1999) found from a sample of 410 Finnish early adolescents (aged 14–16) that depression was likely to occur among bullying victims and perpetrators. When depression was controlled for, suicidal ideation occurred most frequently among youth who were identified as bullies. Analyzing a sample of 1,718 South Korean middle school students, bully victims, Kim, Koh, and Leventhal (2005) also reported that bully victims reported more suicidal/self-injurious behaviors and suicidal ideation within the past 6 months, compared to students with no bullying involvement. Interestingly, depression/anxiety was included in their regression model as a control variable rather than as a mechanism linking bullying victimization and suicidal behavior. It is particularly important that researchers examine these internalizing behaviors as mechanisms linking bullying perpetration, victimization, bully victimization, and suicidal behavior because youth who are involved in bullying are likely to experience these internalizing behaviors, which increase their risk for suicidal ideation and behavior.
Despite this, relatively few scholars have considered internalizing behaviors as potential mediating factors explaining this important relationship. Perhaps this is because the mediating influences of internalizing behaviors may increase the risk of bullying perpetration or victimization and may also have an independent relationship to suicidal behavior. However, it is also plausible that youth involved in bullying, as victims, perpetrators, or both, are at an increased risk of internalizing behaviors, which can trigger their tendency to engage in suicidal thoughts and behaviors. Thus, investigating potential mediating influences of bullying perpetration, victimization, and bully victimization, and suicidal thoughts and behaviors is warranted.
Potential Mediating Factors
A mediator is a third explanatory variable that statistically explains some amount of the observed relationship between the predictor variable and the outcome variable. For instance, bullying victimization (predictor variable) might predict hopelessness (mediator), which in turn might be associated with suicidal behavior (outcome variable; Baron & Kenny, 1986). Baron and Kenny (1986) argue that a mediator effect is often tested when there appears to be a significant direct effect between the predictor variable and outcome variable. It is important to investigate the mediating influences because there might be an important relationship masked by the mediator. In the following section, we investigate five potential mediators that can possibly shed light on the associations between bullying perpetration, victimization, bully victimization, and suicide, namely (1) depression, (2) anxiety, (3) low self-esteem, (4) loneliness, and (5) hopelessness (Figure 1).
Potential mediating factors of the association between bullying perpetration, victimization, and bully victimization, and suicidal thoughts and behaviors. The figure highlights the bidirectional relationship between bullying involvement (i.e., perpetration, victimization, or both) and the five psychosocial factors. The psychosocial factors may function as a consequence of and a risk factor for bullying involvement. This article primarily focuses on the potential mediating role of these five psychosocial factors connecting bullying involvement to suicidal thoughts and behaviors.
Depression
Depression represents the first potential mediating factor, which can explain the association between bullying perpetration, victimization, and suicidal behavior. Depression is the most commonly identified and strongly associated mental health outcome of bullying victimization (Espelage & Swearer, 2003; Hawker & Boulton, 2000). Victims (Bond, Carlin, Thomas, Rubin, & Patton, 2001; Fekkes, Pijpers, & Verloove-Vanhorick, 2004; Fleming & Jacobsen, 2009; Juvonen, Graham, & Schuster, 2003; Klomek et al., 2008; Owusu, Hart, Oliver, & Kang, 2011; Sourander et al., 2009; van der Wal, de Wit, & Hirasing, 2003), bullies (Ivarsson, Broberg, Arvidsson, & Gillberg, 2005; Klomek, Marrocco, Kleinman, Schonfeld, & Gould, 2007b; Klomek et al., 2008), and bully victims (Austin & Joseph, 1996; Kaltiala-Heino et al., 1999; Swearer, Grills, Haye, & Cary, 2004) are at heightened risk for depressive symptoms. However, significant differences exist in psychological adjustments between these bullying roles (Juvonen et al., 2003; Stein et al., 2007). Juvonen, Graham, and Schuster (2003) found from a sample of 1,985 sixth graders that victims reported the highest levels of depressive symptoms, anxiety, and loneliness, while bullies reported the lowest. In addition to role type, the type of bullying victimization (e.g., direct vs. indirect) exerts a significant influence on the development of depression. For example, Van der Wal, de Wit, and Hirasing (2003) found that direct bullying victimization (e.g., being hit, shoved, called names) was associated with depression and suicidal ideation for both boys and girls. Indirect bullying victimization (e.g., being socially excluded, target of rumors), however, had a stronger effect on depression than direct forms.
Depressive symptoms, in turn, predict likelihood of suicidal thoughts and behaviors. The most researched psychosocial predictor of suicide is depression (Serocynski, Jacquez, & Cole, 2003), which is documented as second only to previous suicide attempts as the strongest predictor (Sofront, Dalgliesh, & Kosky, 2005). For adolescents, depression significantly impairs functioning and negatively affects normal psychosocial development (Cook, Peterson, & Sheldon, 2009). Depressive symptoms can induce changes in moods, interpersonal relations, and school performance, which may increase the risk of suicidal thoughts and behaviors (Lewinsohn, Rohde, Klein, & Seeley, 1999). Although depression can also hinder social interactions, which can lead to peer rejection and increase risk for bullying involvement, depressive symptoms have consistently emerged as a salient predictor for suicidal ideations, attempts, and completed suicides among adolescents in a number of research studies (Boergers & Spirito, 1998; Colluci & Martin, 2007; Dearden, De La Cruz, Crookston, Novilla, & Clark, 2005; Fergusson, Horwood, & Beautrais, 1999; Groholt, Ekeberg, Wichstrom, & Haldorsen, 2000; Haliburn, 2000; Lee, Hong, & Espelage, 2010; Lewinsohn, Rohde, & Seeley, 1994; Reifman & Windle, 1995; Russell & Joyner, 2000; Stewart, Lam, Betwon, & Chung, 1999). For instance, according to Lee (2003), 87% of South Korean adolescents admitted to the emergency room for suicide attempts manifested psychiatric problems and 66% of those individuals were diagnosed as clinically depressed. A follow-up study conducted by Weissman et al. (1999) also reports that, among adolescents diagnosed with major depressive disorder, 7% committed suicide later. Depressed adolescents were also 5 times more likely to have attempted suicide, compared to nondepressed adolescents (Weissman et al., 1999). Because of these robust findings, more research is clearly needed untangling the interconnections between bullying, perpetration, victimization, depression, and suicide. Most recently, a cross-sectional study conducted by Espelage and Holt (2013), which investigated differences in the frequency of suicidal ideation and behaviors across bully victim subgroups (i.e., verbal bullies, bully victims, victims, physically aggressive bullies, and uninvolved students) from a sample of 661 middle school students (fifth–eighth grades), reported that bully victims reported the highest percentage of suicidal ideation (60%), followed by physically aggressive bullies (43%), verbal bullies and victims (32–38%), and uninvolved students (12%). After controlling for depressive symptoms, differences in suicidal thoughts and behaviors emerged only between uninvolved youth and the victim and bully victim groups.
Anxiety
The second potential mediating factor that can explicate the bullying perpetration–victimization–suicide link is anxiety. Bullying victims are at an increased risk of anxiety during childhood (Aoyama, Saxon, & Fearon, 2011; Bond et al., 2001; Chin, 2011; Juvonen et al., 2003; Juvonen & Gross, 2008; Reuter-Rice, 2006; Salmon, James, & Smith, 1998; Slee, 1994) and adulthood (Gladstone, Parker, & Malhi, 2006; McCabe, Antony, Summerfeldt, Liss, & Swinson, 2003; Roth, Coles, & Heimberg, 2002). Victims may experience elevated levels of distress and social anxiety, as they perceive their school environment as unsafe (Juvonen & Gross, 2008; Reuter-Rice, 2006). They are also likely to experience higher levels of general anxiety and a tendency to express anxious arousal externally under stressful situations (Gladstone et al., 2006). Interestingly, indirect bullying victimization (e.g., spreading rumors, social ostracism) has been found to be the strongest predictor of anxiety. Baldry’s (2004) study, for instance, found from a sample of 661 Italian youth (aged 11–15) that being a victim of indirect bullying (independent of direct bullying) was associated with anxiety, depression, and somatic complaints.
Anxiety is also another major psychosocial correlate of suicidal thoughts and behaviors (Fergusson et al., 1999; Koutek, Kocourkova, Hladikova, & Hrdlicka, 2009; Wunderlich, Bronisch, & Wittchen, 1998). Children who display anxiety are vulnerable to suicidal behavior, and more specifically, an anxiety disorder can increase the odds of actual suicide attempts (Kwoy & Shek, 2009; Wunderlich et al., 1998). For example, Wunderlich, Bronisch, and Wittchen (1998) found from a sample of 3,021 adolescents and young adult (aged 14–24) that the highest risk for suicide attempt was anxiety disorder (odds ratio [OR] = 4.3), followed by substance disorder (OR = 2.2) and depressive disorders (OR = 2.1). Anxiety increases suicide risk (Mykletun et al., 2007) and appears to play a mediating role in suicidal thoughts and behaviors by heightening feelings of depression and hopelessness among adolescents (Thompson, Mazza, Herting, Rantell, & Eggert, 2005). However, another study found that anxiety lowers the probability of the first suicide attempt but increases the likelihood of future attempts (Brezo et al., 2008).
Although causal relationships among bullying involvement (i.e., victimization, perpetration, and bully victimization), anxiety, and suicide-related thoughts and behaviors remains elusive, it is conceivable that bullied youth are likely to experience anxiety, which can elicit suicidal thoughts and behaviors. Further research needs to test the mediating influence of anxiety on bullying victimization, perpetration, and bully victimization, and suicide.
Low Self-Esteem
Low self-esteem is a third potential mediating factor that can shed light on the link between bullying perpetration, victimization, and suicide. Self-esteem is defined as a positive or negative attitude toward the self (Rosenberg, 1965) and is an internal representation of social acceptance or rejection (Leary & Downs, 1995). These conceptualizations underscore the fact that self-esteem is a personal value, which is influenced by one’s participation in the social world, in which interpersonal conflicts can lead to bullying and victimization (Patchin & Hinduja, 2010). Research consistently supports the association between bullying victimization and low self-esteem (Aoyama et al., 2011; Egan & Perry, 1998; Schafer et al., 2004; Undheim & Sund, 2010). For instance, Undheim and Sund (2010) examined the associations between bullying victimization and perpetration and self-reported mental health problems among a representative population sample of 2,464 Norwegian adolescents (aged 12–15). They found that both victims and perpetrators showed lower scores on the global self-worth subscale, compared to youth who were not bullies or victims. Furthermore, recent findings suggest that experiences of cyberbullying, as a victim or bully, predicted low self-esteem (Patchin & Hinduja, 2010; Ubertini, 2010). O’Moore and Kirkham’s (2001) research, which examined the relationship between self-concept and bullying behavior (i.e., perpetration, victimization, and bully victimization) from a sample of 8,249 children (aged 8–18), found higher frequencies of bullying victimization and perpetration were associated with lower levels of self-esteem. Consistent with past studies, bully victims of all ages had the lowest self-esteem, in comparison to bully-only and victim-only youth.
Findings on bullying perpetration and self-esteem, however, have been less consistent. Earlier studies (Olweus, 1990, 1994; Rigby & Slee, 1991; Salmivalli, Kaukiainen, Kaistaniemi, & Lagerspetz, 1999) reported that bullies have higher self-esteem, while more recent studies (Frisen, Jonsson, & Persson, 2007; Jankauskiene, Kardelis, Sukys, & Kardeliene, 2008; Yang, Kim, Kim, Shin, & Yoon, 2006) found the opposite. To complicate the picture, a review by Baumeister, Campbell, Krueger, and Vohs (2003) reveals that bullying perpetration is associated with high self-esteem. In sum, the relationship to self-esteem is weaker among bully-only persons than it is among those who are victim only or both bully and victim (Patchin & Hinduja, 2010).
Low self-esteem can generate a sense of worthlessness, which can negatively affect interpersonal relationships (Baldwin & Hoffmann, 2002) and put youth at a heightened risk of suicide (Beautrais, Joyce, & Mulder, 1999; Groholt et al., 2000; Lewinsohn et al., 1994; Martin, Richardson, Bergen, Roeger, & Allison, 2005; Wichstrom, 2000; Wilburn & Smith, 2005). For instance, McGee and Williams (2000) reported from a large sample of 9- to 13-year-old children in New Zealand that low self-esteem at age 9 is correlated with suicidal ideation at age 15. Chatard, Selimbegovic, and Konan (2009) also investigated whether national differences in self-esteem across 55 nations are reflected in suicide rates, using the International Sexuality Description Project data. Their findings suggest that suicide is common in nations with relatively low levels of self-esteem, regardless of age. These studies indicate low self-esteem is a major predictor of suicidal thoughts or attempts. Collectively these findings suggest bullying involvement has a negative impact on self-esteem, particularly for victims and bully victims; in return, low self-esteem is implicated in higher rates of suicide.
Loneliness
Loneliness is the fourth potential mediating factor that may explain the link between bullying involvement and suicidal thoughts or behaviors. Considering the importance of social belonging among adolescents (Hamm & Faircloth, 2005), it is not surprising that an association between bullying victimization and loneliness has been observed (Berguno, Leroux, McAinsh, & Shaikh, 2004; Brown, Riley, Butchart, & Kann, 2008; Fleming & Jacobsen, 2009; Forero, McLellan, Rissel, & Bauman, 1999; Juvonen et al., 2003; Nansel et al., 2001; Owusu et al., 2011; Storch & Masia-Warner, 2004). Victims frequently experience boredom, inactivity, a tendency to withdraw into fantasy, and passive attitudes toward social interactions (Berguno et al., 2004). The linkage between bullying victimization and loneliness is significant for both male and female adolescents (Forero et al., 1999; Gofin & Gordon, 2002; Ireland & Power, 2004; Ireland & Qualter, 2008; Schafer et al., 2004; Storch & Masia-Warner, 2004; Tritt & Duncan, 1997). For instance, from a sample of 86 male and 120 female university students, Tritt and Duncan (1997) found that young adults who were bullies or victims during childhood reported significantly more loneliness during adulthood than did those who had no bullying involvement as children. Ireland and Power (2004) also reported from a sample of adult and adolescent male offenders that bully victims reported significantly higher scores on emotional loneliness than did those who were only bullies, only victims, or were uninvolved in bullying at all. We should note, however, that peer rejection can precede bullying, and youth may experience loneliness, which may put them at risk for bullying victimization. Moreover, victimization and loneliness may mutually reinforce one another and lead to suicidal behavior.
Loneliness has been implicated as a vulnerability factor for adolescent suicidal behavior. Theorists have long argued that loneliness, attributed to a lack of social connectedness, is predictive of suicidal behaviors, because loneliness may be an indicator that a fundamental psychological need is unmet. Joiner’s interpersonal psychological theory of suicide also posits that thwarted belongingness (i.e., frustration when a person’s inclination to connect with others is not achieved, and loneliness and social isolation occur) is a salient risk factor for suicide. Durkheim’s (1897) dysregulation of social forces theory also purports that too little integration can lead to suicide because individuals lack connection to other people and to something that transcends themselves (as cited by Van Orden et al., 2010). With regard to empirical findings, loneliness and social isolation have consistently been identified as the strongest and most reliable predictors of suicidal behavior (Borowsky, Ireland, & Resnick, 2001; Garnefski, Diekstra, & De Heus, 1992; Guertin, Lloyd-Richardson, Spirito, Donaldson, & Boergers, 2001; Jones, Schinka, van Dulmen, & Bossarte, 2011; Rich, Kirkpatrick-Smith, Bonner, & Jans, 1992; Van Orden et al., 2010; Yang & Clum, 1994). More specifically, adolescents with suicidal ideation (Roberts, Roberts, & Chen, 1998) and those with a prior history of suicide attempts (Page et al., 2006) consistently reported higher levels of loneliness than adolescents without suicidal tendencies.
Despite the need for longitudinal research to sort out the causal ordering of these psychosocial factors, there is an even more current need to empirically examine loneliness, as a potential point of intervention or harm reduction strategy for reducing suicide rates among youth involved in bullying perpetration and victimization.
Hopelessness
The fifth and final potential mediating factor that could explain the pathway from bullying involvement to suicidal thoughts or behaviors is that of hopelessness. Hopelessness is another common psychosocial result from bullying victimization (Bonnano & Hymel, 2010; Coggan, Bennett, Hooper, & Dickinson, 2003; Fleming & Jacobsen, 2009, 2010; Swearer, Song, Cary, Eagle, & Mickelson, 2001). If they are unable to change their situation (e.g., stopping their peers from bullying them,) victims may develop learned helplessness (Besag, 1989), which can subsequently lead to a sense of hopelessness (Koszeghy, 2010). One of the few bullying studies to investigate the relationship between social hopelessness and suicide, Bonnano and Hymel (2010) found hopelessness partially mediated the association between bullying victimization and suicidal ideation for students in Grades 8–10 (N = 399). Their findings suggest that one potential mechanism by which victimized students become suicidal is through the impact of victimization on social hopelessness.
In addition, other studies have reported a connection between hopelessness and subsequent suicidal thoughts or behaviors (Beautrais et al., 1999; Beck, Brown, Berchick, Stewart, & Steer, 1990; Beck, Steer, Kovacs, & Garrison, 1985; Donaldson, Spirito, & Farnett, 2000; Morano, Cisler, & Lemerond, 1993; Negron, Piacentini, Graae, Davies, & Shaffer, 1997; Russell & Joyner, 2000). For example, Negron, Piacentini, Graae, Davies, and Shaffer’s (1997) research, which compared psychological outcomes of suicidal ideation and attempts, found that attempters reported significantly more hopelessness than nonattempters. Research on the association between hopelessness and suicidal behavior, however, yields inconsistent results (Ayyash-Abdo, 2002). Lewinsohn, Rohde, and Seeley’s (1993) study findings, which included a representative sample of 1,710 older adolescents, reveal that hopelessness was not a significant predictor of suicide attempts, when controlling for depression. In contrast, other researchers found that hopelessness was a stronger predictor than depression (Beck et al., 1990; McLaughlin, Miller, & Warwick, 1996). Additional research should explore feelings of hopelessness among youth involved in bullying and explore whether such feelings induce suicidal thoughts and behaviors.
Implications for Research, Prevention, and Intervention
Five potentially relevant mediating factors (depression, anxiety, low self-esteem, loneliness, and hopelessness) were identified in this review. The relationships proposed are complex and require further empirical investigation, which can enhance our understanding of the relationship between bullying involvement and suicidal tendencies. A rigorous investigation that employs a longitudinal research design to illuminate the causal order of events and utilizes advanced statistical analyses with a clinically and nationally representative sample can elucidate the multiple pathways linking bullying involvement and suicide (Bender, 2010). Greater understanding of these relationships is imperative, considering the profound impact bullying perpetration and victimization has on the lives of children and adolescents.
Most research on bullying and victimization conceptualizes psychiatric or psychological problems as direct consequences. Rather than assuming such a causal direction, some researchers have examined whether psychological problems predict poor peer relations including peer rejection and bullying victimization. For instance, a longitudinal study by Kochel, Ladd, and Rudolph found depressive symptoms predicted subsequent peer difficulties, including bullying victimization. No support was found for an interpersonal risk model in which peer difficulties predict depressive symptoms or for a transactional model in which depressive symptoms and peer difficulties are reciprocally related over time. The authors found that youth with depressive symptoms are more likely to become victims of bullying. Similarly, bullying victimization is associated with environmental factors such as child maltreatment, exposure to domestic violence, parental depression, and low socioeconomic status (Arseneault, Bowes, & Shakoor, 2010). Psychological distress and emotional dysregulation have been proposed as potential mediators, explaining the association between exposure to family violence (e.g., maltreatment) and increased peer rejection and bullying involvement (Hong et al., 2012). Similar to internalizing behaviors (e.g., depression, anxiety, low self-esteem, loneliness, and hopelessness) as potential mechanisms explaining the link between bullying involvement and suicidal thoughts or behaviors, mediating pathways, in general, remain largely underexplored within the bullying literature. One of the goals of this review is to draw attention to the need for more theoretical, conceptual, and empirical research to explain the complex interrelationships between early childhood risk factors (e.g., family violence), bullying perpetration and victimization, psychological and internalizing problems, and suicidal ideation and behaviors.
As a starting point, it is imperative that researchers investigate the directionality of the five mediating variables discussed in this review. It may be that youth with these internalizing problems are more vulnerable to peer rejection and subsequent bullying victimization. Longitudinal research will be able to disentangle these causal pathways and discover whether these variables persist or become worse with subsequent bullying. We view this as a critical direction for future research. It is possible that these potential mediators between bullying victimization and suicide actually precipitate the onset of being bullied. A salient research question to pursue is whether bullying victimization is a mediator between depression and other psychological problems and suicide. It may be that depressed youth who are victimized are more likely to become suicidal.
The five proposed mediating variables are conceptually and empirically linked to the possibility of suicide. For example, the extant literature robustly identifies these factors as predictors of increased likelihood in children and adolescents. The proposed variables’ relationships to bullying perpetration and victimization, however, remain unclear. Because of the relational nature of bullying, the effects of these five psychosocial factors on peer relationships require further study. These psychosocial factors may lead to increased peer rejection or social isolation, increasing the risk for bullying involvement. However, these factors may also function as mediators leading to increased suicidal thoughts or behavior after involvement in bullying as either a perpetrator or a victim.
Heightened attention to the problem of bullying and suicide in recent years has led to significant development in prevention programs. Antibullying prevention and intervention programs have been widely developed and implemented in school districts (Brown, Low, Smith, & Haggerty, 2011; Hirschstein, Van Schoiak Edstrom, Frey, Snell, & MacKenzie, 2007; Polanin, Espelage, & Pigott, 2012). Bullying and suicide prevention approaches have traditionally relied on identifying and assessing individuals most at risk for referral to treatment and other risk reduction services (see Hong & Espelage, 2012; Lubell & Vetter, 2006). Evidently, these types of approaches have generally targeted children and adolescents who have already been exposed to (e.g., peer rejection) or begun to develop (e.g., antisocial behavior) risk factors (Rigby, 2003).
We recommend that future researchers investigate moderators that potentially buffer the association between bullying perpetration and victimization and suicide across multiple contexts (e.g., increased family support, friendship quality, antibullying school-level policies). Family and parental support is a relevant protective factor that has been frequently examined in bullying (Baldry, 2004) and suicide (Morano et al., 1993) research. Family support can potentially strengthen or attenuate the link between these factors (Bonnano & Hymel, 2010; Herba et al., 2008). Victims who lack family support appear to be more vulnerable to suicidal ideation (Herba et al., 2008). However, youth who report strong family connections (Borowsky et al., 2001) and those who receive an adequate amount of social support at home (Bonnano & Hymel, 2010) may be less prone to contemplate suicide even when bullied (e.g., Borowsky et al., 2001). Furthermore, family support is shown to have an impact on the mental health of children (Colarossi & Eccles, 2003; Repetti, Taylor, & Seeman, 2002). Identifying and assessing potential moderators can translate into more effective prevention and intervention strategies.
Recent research has emphasized the importance of primary prevention that focuses on moving up the casual chain to prevent the development of certain key risk factors, thus reducing future problems such as bullying involvement and suicidal tendencies. Indeed, it is imperative that clinicians and school practitioners screen early for symptoms of depression, anxiety, low self-esteem, loneliness, and hopelessness among children, whether identified as bullies, victims, or both. It is important that clinicians and school practitioners (e.g., educational psychologists, school psychologists, counselors, and social workers) assess these symptoms as early as preschool or elementary school. Early assessment can provide a strong foundation for developing interventions that can effectively facilitate the development of social and emotional skills for middle and high school students.
And finally, we also recommend that researchers and practitioners with expertise in school violence and peer conflicts form a collaborative partnership with those working in youth suicide prevention. Such partnership can significantly enhance the development, dissemination, and implementation of programs that can address the link between bullying involvement and suicide in a more integrative, comprehensive, and effective manner.
Implications for Research, Practice, and Policy
An investigation that uses a longitudinal research design and advanced statistical analyses can enhance our understanding of the mechanisms linking bullying perpetration, victimization, and bully victimization, and suicidal thoughts and behaviors.
It is also imperative that researchers investigate the potential dual role of the five psychosocial factors—depression, anxiety, low self-esteem, loneliness, and hopelessness—as (a) mediators bridging the connection between bullying involvement and suicidality and (b) risk factors to future bullying involvement.
Future research should also examine moderators that potentially amplify the association between bullying perpetration, victimization, bully victimization, and suicidality across multiple contexts.
Practitioners should screen early for symptoms of depression, anxiety, low self-esteem, loneliness, and hopelessness among children involved in bullying.
Researchers and practitioners with expertise in bullying should collaborate with those working in youth suicide prevention, which can enhance the development of programs that can effectively address bullying perpetration, victimization, and bully victimization, and suicide.
Footnotes
Authors’ Note
Dr. Jun Sung Hong conceived the study. However, all of the authors contributed equally to the article. Dr. Hong wishes to express his deepest gratitude to Mr. Al Acker for proofreading this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.