Bullying and Its Relation to Suicidal Ideation and Behaviors: A Meta-Analysis

Bullying is a significant public health problem associated with numerous short- and long-term psychological effects on adolescents. Of particular concern is the association between bullying involvement and suicidal ideation and behaviors. Cross-sectional studies have found that youth involved in bullying are at 1-10 times greater risk of reporting suicidal ideation and behaviors compared to uninvolved youth (Kim & Leventhal, 2008). Research to date, however, has been limited in a number of respects. For instance, there is mixed evidence about how the specific nature of bullying involvement (e.g., bully, victim, bully/victim) relates to the risk for suicidal ideation (Herba et al. 1998; Rigby & Slee, 1999). Similarly, it is unclear whether there are sex differences in the association between bullying and suicidal ideation and behaviors (e.g., Klomek et al., 2009; Laukkanen et al., 2005). Finally, while review articles on this topic exist (e.g., Kim & Leventhal, 2008), they have included studies that measure forms of peer victimization that are not necessarily bullying (e.g., studies using YRBS violence items prior to the inclusion of a bullying-specific item in 2010). The goal of this meta-analysis was to examine the association between bullying and suicidal ideation and suicidal behaviors using a strict definition of bullying, and to assess moderators of this association (e.g., bullying subtype, sex).

First, a systematic search of the literature was performed using search terms in three broad categories: bullying, suicide, and adolescents, with variations of each term included. A total of 350 articles were retrieved, and then were reviewed based on our study’s inclusion and exclusion criteria. Primary inclusion criteria were: (1) the study assessed both bullying and suicidality at a single time point; (2) participants reported on experiences occurring before age 18; and (3) peer-to-peer bullying was examined. Articles not written in English and those that did not provide adequate statistical information for the meta-analysis were excluded. This resulted in a final sample of 29 studies. Each article was coded by two individuals, and coders met to reach consensus and document final codes. Articles were coded on a wide range of indicators, including sample characteristics, measurement of bullying and suicide, and statistics for the association between bullying and suicidal ideation/behaviors.

To date, initial analyses on a sample of 26 studies have been completed using the Comprehensive Meta-Analysis statistical software. A preliminary review of data indicates that of the included studies, seven used U.S. samples and the remaining 22 used international samples. The number of participants per study ranged widely, with a low of 143 (Viljoen et al., 2005) and a high of 31,294 (Fleming & Jacobsen, 2009). The initial effect size estimate of the link between bullying victimization and suicidality is 1.74, and the effect size estimate for bullying perpetration and suicidality is 1.59. This meta-analysis will yield valuable information that can be translated into improving prevention programs and providing mental health services for youth in need.