Multi-Site Evaluation Demonstrates Effectiveness of Adult Drug Courts

Article excerpt

Specialized adult drug courts have proliferated during the last two decades, operating in most medium and large criminal courts, nationwide. Drug courts combine community-based substance abuse treatment and ongoing court oversight as an alternative to either incarceration or traditional probation. Participants undergo frequent drug tests, appearances before the drug court judge, and meetings with court-affiliated case managers. Supervision is most intensive at program entry, perhaps involving weekly or biweekly appearances before the judge, but becomes less frequent in response to early progress. Drawing on classic behavioral modification techniques, the judge applies a system of graduated sanctions and incentives, such as community service, more frequent court appearances, or several days in jaii for noncompliance; or verbal praise, journals, or gift certificates for progress. On average, it takes about 15 months to graduate from a drug court, at which point participants receive some legal benefit, usually a case dismissal or charge reduction. Those who fail the program are routinely sentenced to jail or prison.

As of the end of 2009, more than 1,300 adult drug courts had opened nationwide. This is in addition to more than 1,000 other drug courts that serve juveniles, family law respondents, or formerly incarcerated persons on parole or probation.1 The consensus reflected in three recent reviews of more than 60 recidivism studies is that adult drug courts reduce recidivism by an average of 8 to 13 percentage points.2 Since drug courts reduce recidivism, it also might be inferred that they succeed in rehabilitating offenders from their underlying drug problems. However, little prior research directly examines effects on drug use or, for that matter, effects on other problems, ranging from unemployment to family dysfunction to co-occurring mental health disorders. In addition, the research field is only just beginning to identify the specific policies and practices that are most responsible for producing positive outcomes.

To fill these gaps in our knowledge, the National Institute of Justice funded a five-year study, known as NIJ's Multi-Site Adult Drug Court Evaluation (MADCE). Implemented by the Urban Institute, theCenter for Court Innovation, and RTI, the study included 1,156 participants from 23 drug courts and 625 drug-involved offenders from six comparison jurisdictions that lacked adult drug courts, had a very narrowly targeted program, or had more drug-involved offenders than drug court capacity. All 29 sites were located in one of eight states: Florida, Georgia, Illinois, Pennsylvania, New York, North and South Carolina, and Washington. The sites were not randomly selected, but do comprise an intentional mix of urban, suburban, and rural locations; seven of the 23 drug courts and two of the six comparison jurisdictions were located in major cities (Atlanta, Chicago, Philadelphia, and Seattle/ Tacoma); the other 20 sites were located in suburbs, small cities, and rural areas. Thus, the sample varied demographically, as well as in drug use patterns.

Study data were collected through in-person offender interviews at baseline (entry into drug court or the equivalent for comparison subjects], as well as six and 18 months post-baseline; an oral fluids drug test at 18-months; and administrative arrest and sentencing records up to 24 months after baseline. Survey attrition was remarkably low, as 85 percent of the original sample was interviewed at the six-month follow up and 83 percent at the 18-month follow up. Survey data were applied in the cost-benefit analysis to measure resources used by participants (e.g., drug treatment, other services, drug tests, court appearances, days incarcerated, etc.) and estimates from national sources were used to estimate unit costs of each activity. Other aspects of study methodology, including statistical strategies to ensure comparability between the drug court and comparison samples, are described in the full technical report. …