Drug markets and COVID-19: A spatiotemporal study of drug offence detection rates in Brisbane, Australia.

In many parts of the world, the social mobility restrictions and stay-at-home orders introduced during the early months of the COVID-19 pandemic have been associated with significant reductions in crime. However, contrary to this general finding, illicit drug offence detections increased significantly. In this study, we explore the geographical distribution of the increase in Queensland, Australia, using spatiotemporal generalised additive model (GAM) to identify locations in the Local Government Area (LGA) of Brisbane where drug offence detection rates were unusually high during the three months of the COVID-19 lockdown (April-June 2020). Contrary to expectation, we find that the increase in drug offence detection rates appears to have been modest in most places, but widespread and diffuse throughout the city. We conclude that drug offence detections are most likely to have increased incidentally, probably as a consequence of general street policing initiatives which saw an increase in the visibility and vulnerability of drug user communities. We do, however, identify several locations in Brisbane where the drug offence detection rate exceeded the prediction by a considerable margin (in one case, more than double the upper limit of the prediction). We argue that in these locations the increase was likely the result of some spatial displacement of inner-city drug markets coupled with a series of targeted policing activities. Further research is needed to clarify the true mechanism of change in these locations.

How risky are heroin markets? A multi-site study of self-reported risk perceptions among police detainees in Australia.

BACKGROUND: There has long been an international interest in the eradication of drug markets-particularly heroin markets-given their documented connection to the deterioration of local community and social amenity. Recent interest in focused drug law enforcement strategies has reinvigorated debate about the potential for deterrence; however, there exists no large-scale effort to document the risk perceptions of those who transact in heroin markets. METHODS: We use data from 2,257 police detainees interviewed as part of the Australian Institute of Criminology's (AIC) Drug Use Monitoring in Australia (DUMA) program. We employ a multilevel generalised ordinal logistic regression model to explore the geographical and temporal heterogeneity of risk, controlling for individual demographic covariates. RESULTS: With one exception, we find a surprising degree of homogeneity at the high end of the risk perception scale, with between 30 and 35 percent of respondents rating their local heroin market as very risky. At the low end, there was greater geographical and temporal variability with between 15 and 30 percent of respondents rating their market as not at all risky. One location stands out as anomalous, being perceived as considerably more risky. Of the demographic and drug-use covariates, only age and gender were statistically significant. Neither the length nor frequency of association with the heroin market predicted lower levels of risk perception. CONCLUSION: A little over half of all heroin market participants consider transacting in their local market to be somewhat or very risky-the others consider it not at all or only a little risky. With few exceptions, this broad pattern appears consistent over time and between geographical locations, although an individual's personal perception is not tied to their degree of exposure to the heroin market. The policy implications of these findings are discussed.