Download files
Abstract
Drug use is a highly prevalent and modifiable risk factor for recidivism. Drug use and recidivism are complex phenomena, as reflected in the diverse drug-crime relationships reported in the research literature. These relationships have not been studied in detail among community-supervised youths, who comprise the majority of supervised young offenders and differ in important ways from detainees and adults. This thesis examined drug-crime relationships among 800 Australians in community-based juvenile justice supervision. Extensive baseline health surveys were linked with lifetime court data, giving four years' prospective observation of recidivism.
Two cross-sectional studies showed that the sample was characterised by family disruption, Conduct Disorder (58%), distress (59%), cognitive deficits (67%), delinquent peers (77%), school failure (79%) and weekly drug use (69%). Psychosocial dysfunction was concentrated among frequent binge drinkers (10%), daily cannabis users (35%) and weekly users of amphetamines (9%) or opioids (5%). Frequent binge drinking was linked with five-fold higher odds of self-harm but did not affect recidivism.
Three prospective studies modelled predictors of the prevalence (79%), timing (median: eight months after baseline), frequency (mean: six convictions) and severity of recidivism (58% committed less serious offences). Frequent drug users had poorer recidivism outcomes but drug use did not predict recidivism for most drug users. However, daily cannabis use predicted more rapid violence and weekly opioid use predicted a four-fold increase in theft outcomes. Static (unmodifiable) risk factors including prior detention were stronger predictors of recidivism than dynamic (modifiable) risk factors. With the exception of opioid use, risk factors varied substantially by gender, ethnicity and age.
The original contribution of this thesis was to show that young offenders' drug-crime relationships are characterised by their variation. Drug use per se did not affect recidivism; rather, certain patterns of frequent use increased the risk of some offences. Research, policy and prevention efforts must disaggregate drug use and recidivism measures when modelling their relationships, assessing recidivism risk, and triaging offenders into education and treatment. These findings provide a clear imperative to reduce progression to frequent drug use, and to provide demographically-tailored responses that address the health and criminogenic implications of frequent drug use.