Study 1
Jones (2013) conducted a nonblinded randomized controlled trial to examine the impact of the intensive judicial supervision (IJS) program on drug use, sanctions, program progression, and program termination of drug court clients in the Parramatta Drug Court in Sydney, Australia. All drug court program participants received identical drug treatment; the differences were in their respective judicial supervision requirements and the length of time they were scheduled to remain in the first phase of the program. Participants were either randomly assigned to the IJS group or the supervision as usual (SAU) control group. Those in the SAU group were required to appear in court once per week during Phase 1 (for a minimum of 3 months), once every 2 weeks during Phase 2 (for a minimum of 3 months), and once every 4 weeks thereafter. Those in the IJS group were required to appear in court twice per week (instead of once per week) during Phase 1 for a minimum of 4 months (instead of 3 months). During Phase 2, supervision was gradually reduced to weekly report-back court visits, and then to visits every 2 weeks by Phase 3, if participants showed sustained abstinence.
Of the first 160 drug court program clients who had their cases held for initial sentencing on or after March 1, 2010 to November 8, 2011, 24 were removed from the randomized controlled trial due to ineligibility for the trial or to receiving treatment in residential rehabilitation facilities. The names of the remaining 136 participants were forwarded to the New South Wales (NSW) Bureau of Crime Statistics and Research, and each participant was randomized within blocks to ensure a balanced accrual in the IJS program group and control group. All court officers were blinded to the randomization.
There were 66 drug court clients randomly assigned to the IJS program (treatment group) and 70 drug court clients assigned to the SAU condition (control group). The average age of IJS program participants was 32.2, 84.9 percent were male, and 13.6 percent were from indigenous Australian groups. In terms of seriousness of offense, 37.9 percent of IJS program participants had committed a break, enter, and steal offense; 33.3 percent a theft offense; 12.1 percent a driving offense; and 16.7 percent another offense. Fifteen percent of IJS participants were being treated for use of alcohol, 35.4 percent for amphetamines, 29.2 percent for benzodiazepines, 52.3 percent for cannabis, 15.4 percent for cocaine, and 70.8 percent for heroin. The average age of SAU participants was 32.5, 82.9 percent were male, and 8.6 percent were from indigenous Australian groups. In terms of seriousness of offense, 25.7 percent of SAU participants had committed a break, enter, and steal offense; 40 percent a theft offense; 15.7 percent a driving offense; and 18.6 percent another offense. Seven percent of SAU participants were being treated for use of alcohol, 35.8 percent for amphetamines, 23.2 percent for benzodiazepines, 46.4 percent for cannabis, 8.7 percent for cocaine, and 75.4 percent for heroin. There were no statistically significant differences between IJS and SAU participants on any of these characteristics.
Data on participants? demographic characteristics, urinalysis outcomes, sanctioning, and time in the program was collected from the NSW Drug Court database. Data on initial and final sentences was obtained from manual spreadsheets maintained by the Parramatta Drug Court. Data on prior criminal offenses was collected from a reoffending database maintained by the NSW Bureau of Crime Statistics and Research.
There were four outcome measures for this trial: 1) drug tests, 2) sanctions, 3) progression to Phase 2, and 4) termination. A positive drug test was defined as any test in which the participant tested positive to any drug, admitted any drug use to the court, failed to attend a scheduled drug test, or failed to provide a urine sample at a testing episode. Two separate sanctioning measures were constructed to determine how many sanctions were accrued and served. The first measure counted the number of sanctions (days in custody) accrued per ?free week? on the programs. Free weeks were counted as those weeks when one or more judicial status hearings were listed, while weeks when no judicial status hearings were listed were recorded as missing. The second measure counted the number of sanction days issued to serve during each free week on the programs. Progression to Phase 2 was measured as a categorical outcome: yes or no. Termination was measured as a categorical outcome: not terminated to prison or terminated to prison.
Multilevel logistic regression model analyses were conducted to test whether there was any difference between groups in their odds of returning positive urinalysis tests. A chi-squared test was employed to test whether there was any difference between groups in their likelihood of progressing to Phase 2 of the program. The Kaplan-Meier estimator was used to assess whether there was any difference between groups in the number of days it took participants to be terminated and returned to prison. Cox proportional hazards regressions models were .used to assess whether there was any interaction between supervision, risk, and time on the program. The study authors did not conduct subgroup analyses