Are drug arrests a valid measure of drug use? A time series analysis

Abstract

PurposeThe purpose of this paper is to determine whether trends in arrests for heroin, amphetamine‐type substances (ATS) and cocaine can be used as indicators of trends in the use of these drugs.Design/methodology/approachThe question was addressed using ARIMA models to analyse the relationship between arrests and emergency department (ED) admissions for narcotics, amphetamine type substances (ATS) and cocaine.FindingsStrong positive correlations were found for the narcotics and cocaine series between arrests and EDs in the same month (contemporaneous correlation) and between arrests in the current month and overdoses in earlier months (lagged correlation). The contemporaneous correlation between ATS arrests and EDs was slightly less strong than the lagged correlations at two and four months. A jump in ATS EDs, was followed by a jump in arrests in the same month and then two and four months later.Practical implicationsArrests for narcotics use/possession, ATS use/possession and cocaine use/possession may in some circumstances provide useful intelligence about drug trends and/or a basis for evaluating the impact of police drug law enforcement activity on the use of narcotics, ATS and cocaine when other stronger measures of drug use are not available.Originality/valueEfforts to evaluate local drug law enforcement activity on illicit drug use have been hampered by poor measures of trends in illicit drug use at small area levels. This is the only study the authors are aware of that has examined the long‐term relationship between illicit drug arrests and emergency department admissions for illicit drug use.