Investigating heterogeneity in violent offending liability among injection drug users from a developmental perspective. (September 2016)

• Record Type: Journal Article
• Authors: Torok, Michelle; Darke, Shane; Shand, Fiona; Kaye, Sharlene
• Is Part Of: Addictive behaviors. Volume 60(2016)
• Page Start: 165
• Page End: 170
• Publication Date: 2016-09
• Subjects: Violent offending -- Injecting drug use -- Developmental; Substance abuse -- Periodicals; Alcoholism -- Periodicals; Drug addiction -- Periodicals; Nicotine addiction -- Periodicals; Smoking -- Periodicals; Gambling -- Psychological aspects -- Periodicals; Electronic journals; 362.29
• DOI: 10.1016/j.addbeh.2016.04.013
• Languages: English
• ISSNs: 0306-4603

Abstract

Abstract: Introduction: Violence is a major burden of harm among injecting drug users (IDU), however, the liability to violent offending is not well understood. The current study aimed to better understand differences in the liability to violence by determining whether IDU could be disaggregated into distinct violent offending classes, and determining the correlates of class membership. Methods: A total of 300 IDU from Sydney, Australia were administered a structured interview examining the prevalence and severity of drug use and violent offending histories, as well as early life risk factors (maltreatment, childhood mental disorder, trait personality). Results: IDU were disaggregated into four distinct latent classes, comprising a non-violent class (24%), an adolescent-onset persistent class (33%), an adult-onset transient class (24%) and an early-onset, chronic class (19%). Pairwise and group comparisons of classes on predispositional and substance use risks showed that the EARLY class had the poorest psychosocial risk profile, while the NON class had the most favourable. Multinomial logistic regression revealed that higher trait impulsivity and aggression scores, having a history of conduct disorder, frequent childhood abuse, and more problematic alcohol use, were independently associated with more temporally stable and severe violent offending. The model explained 67% of variance in class membership (χ 2 = 207.7, df = 51, p < 0.001). Conclusions: IDU can be meaningfully disaggregated into distinct violent offending classes using developmental criteria. The age of onset of violence was indicative of class membership insomuch as that the extent of early life risk exposure was differentially associated with greater long-term liability to violence and drug use. Highlights: This review examined the effectiveness of awareness campaigns in preventing suicide. Awareness campaigns were not effective when implemented as a standalone strategy. Suicide literacy increased post-intervention but effects were not maintained. Impact and repeat exposure underpinned campaign efficacy, but difficult to sustain. Increased use of social media to deliver campaigns could better reach those at risk.