Abstract

Bullying is a common aggressive behaviour in school, with a number of cross-sectional studies showing that it exhibits a high comorbidity with other problem behaviours. The present study aims to estimate the comorbidity of school bullying (perpetration and victimisation) with drug use by incorporating and meta-analysing all available evidence on the cross-sectional association between the two variables. Meta-analytic results are based on a comprehensive systematic review across 20 databases and 46 journals. A total of 61 relevant manuscripts were included in the systematic review. Following explicit methodological criteria for the inclusion/exclusion of reports, 13 of them were eligible for the meta-analysis. The association of school bullying perpetration with drug use (adjusted odds ratio OR = 2.82; 95% CI 1.97-4.02; z = 5.71; p .001) suggests a very strong relationship. For example, if a quarter of children were bullies and a quarter were drug users, this value of the OR would correspond to 40.88% of bullies being also drug users, compared with 19.71% of non-bullies. The association of school bullying victimisation with drug use (adjusted odds ratio OR = 1.79; 95% CI 138-232; z = 4.41; p .001) suggests a moderate relationship. For example, if a quarter of children were victims and a quarter were drug users, this value of the OR would correspond to 33.69% of bullied youth also being drug users, compared with 22.1% of non-bullied youth. Adjusted effect sizes are based on study findings that used statistical controls for confounding variables, thus providing the unique association of school bullying with drug use over and above other important risk factors that may explain this association. Implications for policy and intervention research arising from this review are highlighted. (C) 2015 Elsevier Ltd. All rights reserved.