Abstract
While the link between the use of illegal drugs and individuals' involvement in crime is well established, there have been few studies of the effects of drug treatment upon levels of acquisitive crime in the UK. This article examines the association between acquisitive crime and drug treatment using data from the longitudinal Drug Outcome Research in Scotland (DORIS) study. Follow‐up interviews were conducted with 1033 individuals who started treatment for problem drug use in, 2001/2. Respondents were interviewed on four occasions over a 33‐month period. Stepwise logistic regression models were constructed to test the independent effect of 22 co‐variables upon the commission of acquisitive crime or the likelihood of being arrested for it. The outstanding feature of the results is the strong independent effect of drug consumption and drug consumption‐related variables in accounting for acquisitive crime. In contrast, treatment‐related variables have hardly any independent associations with the outcome variables. While the results show substantial reductions in acquisitive crime following treatment, it appears that the influence of treatment is indirect and mediated by its effect on drug use. However, insofar as drug treatment reduces the need for individuals to engage in acquisitive crime by moderating their use of illegal drugs, the social and economic benefits to society from such programmes are likely to be substantial.
Acknowledgements
The Drug Outcome Research in Scotland (DORIS) study is funded by the Robertson Trust and the Scottish Executive. We wish to thank all members of the DORIS research team for their assistance in data collection and all study participants for agreeing to be interviewed. We are also grateful to Professor Neil McKeganey for comments on an earlier version of the paper. The views expressed in the paper are those of the authors and should not be attributed to the funding bodies.
Notes
1. For this type of data, the degrees of fit (r2 values) are quite satisfactory for the committed crimes models (Tables , and ) and for the prison population arrests model (). The degrees of fit for Tables and are less satisfactory, explaining only 14 and 11% of the variance, respectively.