Bullying is a major public health concern amongst young people due to its well-evidenced associations with mental health difficulties (Wolke et al., 2013), so much so that some have called it one of the main preventable psychiatric risk factors (Arango et al., 2021). Bullying involves psychological, physical, or social aggression towards people in the same age group, which is repetitive and based on a power-imbalance (Olweus, 1993). This power-imbalance often originates with the perpetrators marginalising the victims based on a characteristic that makes them different, such as health status, disability, or an LGBTQ+ identity (Pinquart et al, 2017; Sentenac et al., 2012).
Whilst bullying and its impact on mental health are well-researched, the risk of bullying involvement in populations with mental health or neurodevelopmental conditions has been less studied, and questions continue regarding what the prevalence of different types of bullying involvement (perpetration, victimisation, and perpetration-victimisation) are, as well as what impact bullying has on mental health in these groups.
Abregú-Crespo and colleagues (2024) addressed these questions through a systematic review and meta-analysis of the global literature. The results broaden our understanding of what makes children and adolescents vulnerable to bullying involvement and its consequences.
Bullying perpetrators often marginalise victims based on a characteristic that makes them different, including mental health and neurodevelopmental conditions.
Methods
Six databases were searched with no language restrictions, and results were screened by a team of five researchers, who also extracted data and categorised the studies by disorder (DSM-5 compatible categories), and type of bullying involvement (perpetration, victimisation or perpetration-victimisation; traditional or cyber).
Pooled event rates were used for the meta-analysis of prevalence in the population(s) of interest. Pooled odds ratios were used for the meta-analysis of bullying involvement of young people with mental health or neurodevelopmental conditions relative to controls. Pooled hedge’s g effect sizes were used for meta-analysing the association between bullying involvement and mental health measures.
Results
The systematic review identified 212 relevant articles, which were included in the subsequent meta-analyses.
Meta-analysis of prevalence (153 tudies)
Most of the included studies were conducted in North America (48%; n = 73), all of them included bullying victimisation and just over half included bullying perpetration (54%; n = 83). The prevalence of traditional bullying in children and adolescents with mental health and/or neurodevelopmental conditions was found to be 42.2% for victimisation, 24.4% for perpetration, and 14.0% for perpetration-victimisation.
Case-control meta-analysis (135 studies)
Young people with mental health and/or neurodevelopmental conditions had 2.85 times greater odds of being victims, 2.42 times greater odds of being perpetrators, and 3.66 times greater odds of being perpetrator-victims. Sub-group analyses by disorder were not possible for all disorder groups, but some key findings included that greatest odds of victimisation were found amongst those with anxiety or affective disorders, and greatest odds of perpetration were found amongst those with conduct disorder.
Young people with mental health and neurodevelopmental conditions had increased odds of all types of bullying involvement. For traditional bullying, this group’s prevalence for victimisation was 42.2%, 24.4% for perpetration, and 14% for perpetration-victimisation. For cyber-bullying, prevalence for victimisation was 21.8%, 19.6% for perpetration, and 20.7% for perpetration-victimisation.
Meta-analysis of the association between bullying involvement and mental health (25 studies)
- Traditional bullying victimisation was associated with higher scores in measures of internalising, externalising, general psychopathology, suicidality, and poorer functionality in young people with these conditions, as well as with autism-associated symptoms (Hedge’s g between 0.325 and 0.742).
- Cyber-bullying victimisation was associated with internalising and externalising (Hedge’s g = 0.466, and 0.402, respectively).
- Traditional bullying perpetration was associated with internalising (Hedge’s g = 0.437).
- Cyber-bullying perpetration was associated with externalising (Hedge’s g = 0.487).
- Traditional bullying perpetration-victimisation was associated with internalising and externalising (Hedge’s g = 0.859, and 0.245, respectively).
Meta-regressions of potential moderators
The investigation of potential moderators of the significant meta-analysed effects revealed a multitude of potential moderation mechanisms. Those worth highlighting are that young people with mental health and neurodevelopmental conditions had greater odds of bullying victimisation in North American studies than those conducted in Europe, as well as in the more recently published studies; samples of children and of adolescents did not differ in odds of victimisation; finally, self-reported measures were associated with lower odds of victimisation and perpetration only in samples with neurodevelopmental conditions, but not in samples with mental health conditions.
Children and adolescents with mental health and neurodevelopmental conditions are at an increased risk of bullying involvement. Those with anxiety or affective disorders are more likely to be victims, whereas those with conduct disorders are more likely to be perpetrators.
Conclusions
This systematic review and meta-analysis identified and analysed over 200 studies, predominantly from North America and Europe, which overall highlighted that young people with neurodevelopmental or psychiatric conditions have a greater likelihood of bullying involvement. This effect is observed across different conditions, and types of bullying involvement (i.e., perpetration, victimisation, and combination of both). Furthermore, within this population, bullying involvement appears to be associated with worse mental health outcomes on measures of internalising, externalising, general psychopathology, suicidality, and functionality, as well as autism-associated difficulties.
Higher odds of bullying involvement can have a negative impact on mental health outcomes of young people with mental health and neurodevelopmental conditions.
Strengths and limitations
One notable strength of this paper is the adoption of a rigorous approach to the study selection and categorisation. Another strength is the acknowledgement of bidirectional effects between bullying involvement and mental health, evidenced in prior literature (e.g., Brunstein Klomek et al., 2019). In the approach taken by this paper, the focus on a population with mental health and neurodevelopmental conditions helps to partly mitigate the ‘chicken or the egg’ dilemma by studying the effect of bullying on mental health outcomes specifically in a population with mental health or neurodevelopmental conditions. This helps to spotlight a directional association between bullying and mental health.
However, the paper is not without limitations. Most studies identified and analysed were conducted in North America and Europe, revealing greater odds of bullying victimisation of the studied population in North American studies compared to those in Europe. Discrepancies in school, social, and health systems across regions might account for this variation, likely limiting the generalisability of findings, especially to other regions of the world, such as the Global South or East Asia.
In addition, the combination of mental health with neurodevelopmental conditions within one review may limit the utility of the study and could potentially perpetuate stigma surrounding neurodiversity. Notably, the authors do not discuss the findings in the context of neurodiversity.
Another limitation is the heterogeneity among the included studies, particularly in relation to sample characteristics, measurement of bullying involvement, and outcome measures. Consequently, general conclusions should be taken with a pinch of salt.
Lastly, most of the included studies were cross-sectional, which limits causal inference, even if a degree of directionality can be inferred. The researchers also did not apply a standard quality assessment tool, making it challenging for readers to gauge the reliability of evidence, particularly concerning the meta-results, which may differ by analysis.
This study’s strength are rigour and the consideration of bidirectional effects between bullying and mental health, but future research should consider expanding beyond the Global North and acknowledging neurodiversity.
Implications for practice
Schools
Young people with neurodevelopmental or psychiatric conditions may be perceived as more vulnerable and/or different by their peers and they require more pastoral support. Amongst other implications, it is worth highlighting the need for supporting youth with conduct problems. Some argue that since conduct disorder is based on behavioural symptoms, it overlooks underlying emotional difficulties that might lead to it (Fairchild et al., 2019). Labelling youth in such a way, without consideration of internal difficulties, may in fact be maintaining the ‘conduct disorder’ instead of helping young people heal emotionally.
Clinical practice
Clinicians working with youth need to be aware that difficulties experienced by clients are complex and they must provide comprehensive support even when working with a particular primary diagnosis (e.g., depression or ADHD). Since young people with neurodevelopmental or psychiatric conditions are more likely to experience bullying; pointing to difficulties in their social lives, this can itself be a struggle that they need support with. Furthermore, bullying exposure can trigger new difficulties, not directly linked to the primary diagnosis. In summary, clinicians must be aware of the complexity of experiences and difficulties, especially interpersonal difficulties, when working with young clients, and not be blind to the difficulties that are not part of a textbook diagnosis.
Research
Future research needs to investigate more specific populations. The scope of the population in this meta-analysis is very broad, challenging the utility of the findings pertaining to all adolescents with these conditions. Future studies should also apply longitudinal methods to address directionality, as longitudinal directionality of effects can strengthen causal inference. Finally, the findings of meta-regressions lead to potential new directions for working with neurodiverse participants. Self-report was a moderator of the odds of victimisation in samples with neurodevelopmental conditions, but not in samples with mental health conditions. Young people with neurodevelopmental conditions, notably those associated with social difficulties (e.g., autism), may not easily identify bullying situations or connect questionnaire items to their own bullying experiences. As such, future research should consider whether existing self-report measures are suitable tools for measuring bullying and similar interpersonal experiences in young people with neurodevelopmental conditions associated with social difficulties. One possibility would be to develop and validate self-report measures for the specific population instead, or to use other informants (e.g., teachers), although they are likely to witness and remember only a limited number of situations, which creates its own biases.
This systematic review and meta-analysis makes a strong case that more pastoral support is needed for pupils with neurodevelopmental or mental health conditions within educational settings.
Statement of interests
No conflict of interest to disclose.
Links
Primary paper
Abregú-Crespo, R., Garriz-Luis, A., Ayora, M., Martín-Martínez, N., Cavone, V., Carrasco, M. Á., … & Díaz-Caneja, C. M. (2024). School bullying in children and adolescents with neurodevelopmental and psychiatric conditions: a systematic review and meta-analysis. The Lancet Child & Adolescent Health, 8(2), 122-134. https://doi.org/10.1016/S2352-4642(23)00289-4
Other references
Arango, C., Dragioti, E., Solmi, M., Cortese, S., Domschke, K., Murray, R. M., … & Fusar‐Poli, P. (2021). Risk and protective factors for mental disorders beyond genetics: an evidence‐based atlas. World Psychiatry, 20(3), 417-436.
Brunstein Klomek, A., Barzilay, S., Apter, A., Carli, V., Hoven, C. W., Sarchiapone, M., … & Wasserman, D. (2019). Bi‐directional longitudinal associations between different types of bullying victimization, suicide ideation/attempts, and depression among a large sample of European adolescents. Journal of Child Psychology and Psychiatry, 60(2), 209-215.
Day, J. K., Perez-Brumer, A., & Russell, S. T. (2018). Safe schools? Transgender youth’s school experiences and perceptions of school climate. Journal of Youth and Adolescence, 47, 1731-1742.
Fairchild, G., Hawes, D. J., Frick, P. J., Copeland, W. E., Odgers, C. L., Franke, B., … & De Brito, S. A. (2019). Conduct disorder. Nature Reviews Disease Primers, 5(1), 43.
Pinquart, M. (2017). Systematic review: bullying involvement of children with and without chronic physical illness and/or physical/sensory disability—a meta-analytic comparison with healthy/nondisabled peers. Journal of Pediatric Psychology, 42(3), 245-259.
Sentenac, M., Arnaud, C., Gavin, A., Molcho, M., Gabhainn, S. N., & Godeau, E. (2012). Peer victimization among school-aged children with chronic conditions. Epidemiologic Reviews, 34(1), 120-128.
Wolke, D., Copeland, W. E., Angold, A., & Costello, E. J. (2013). Impact of bullying in childhood on adult health, wealth, crime, and social outcomes. Psychological Science, 24(10), 1958-1970.
