Integrating Research into Drug Courts

United States

Summary of the Meeting

The meeting on "Integrating Research into Drug Courts" sought to explore strategies for expanding and integrating research on drug abuse treatment services into drug court settings, with particular attention to system-level factors in corrections and health system of services provided to drug court clients. The meeting addressed those issues through brief presentations and an open forum discussion.

The meeting was co-sponsored by the Department of Health and Human Services' National Institute on Drug Abuse (NIDA), the Department of Justice's Office of Justice Programs (OJP), and the National Drug Courts Institute (NDCI).

In order to set the stage for the meeting and discussion of how to improve the integration of research into the drug court system, the meeting opened with introductory comments by Marilyn Roberts of the OJP, Bennett Fletcher of NIDA, and C. West Huddleston of NDCI.

Following this, Dr. Steven Belenko of the National Center on Addiction and Substance Abuse (CASA) at Columbia University presented an overview and history of drug courts, reviewing progress and recent developments in drug court research and evaluation studies, and highlighting findings from recent research. Dr. Belenko's review included the following points:

• Most research has been poorly controlled. There are few random assignment studies; studies have no or inadequate control groups; the drug court itself is poorly characterized so that it is difficult to distinguish what factors might improve or detract from its effectiveness.
• Most studies have used a pre-post evaluation design that essentially views the drug court as a black box (that is, that doesn't examine drug court structure or process), and most have focused on treatment outcomes (especially recidivism) rather than on how the drug court interfaces with drug abuse treatment.
• Across studies, there is a lack of uniformity in outcome measurements, with the consequence that there is limited ability to arrive at consensus findings. For example, recidivism is measured in various ways (sometimes including even traffic violations, other times including only drug-crime) and at various times (for example, in the 3 months after drug court or in the 5 years after drug court).
• The research funded by NIDA under its recent RFA (Research on Drug Courts, published in 1999) is promising, but the grants funded under this program are still in progress.

Dr. Belenko identified a number of key research areas or needs. These included:

• More research on mental health, psychological, and physical health service needs.
• More standardized assessment and screening information.
• Use of better characterizations of treatment delivery models in research studies, including:
  • Description of treatment modalitis and sanctions or contingencies used by the drug court
  • Better description of intended and actual interaction between the drug court and treatment
  • More research on length of treatment needed and provided in drug courts.
• Research infrastructure needs - Management information systems that collect comprehensive and comparable data, and also the ability to merge elements of data from multiple systems across various agencies.
• Better tests of clinical decision making and service delivery models, particularly emphasizing the judicial role in these.
• Baseline data on drug use, type and frequency of criminal activity, treatment history, motivation for treatment, sociodemographic characteristics of drug offenders
• Admission process: How are offenders selected for the drug court? Who is accepted and who is screened out of the drug court? What types of treatment are available and how is the decision made to admit to types of treatment?
• How are sanctions and rewards applied by the drug court?
  • Are sanctions applied systematically and in accord with a treatment-based protocol?
  • How do sanctions and rewards affect the effectiveness of treatment?

A general discussion centered around conducting research in drug court settings followed Dr. Belenko's presentation. These touched on four areas:

1. Substantive research questions: What do we need to know about drug courts?
   Drug court role and mission - What are appropriate expectations?
   Effectiveness and efficiency of drug courts
2. Methodology: What methods are required to answer substantive research questions?
   Randomization
   Longitudinal research
    Ethical questions regarding confidentiality; control and experimental conditions
3. Implementation: How can studies be fielded?
   Mechanisms for research support
   Gaining the participation and cooperation of drug court staff, treatment providers, others
4. Dissemination: Sharing results with drug court and corrections officials, treatment providers, the public
   Using research knowledge in practice
   Informing new substantive questions

Some of the particular issues that arose in discussion include:

What Is A Drug Court?

There is a need to develop an organizational framework or typology, for drug courts. They are proliferating so rapidly that the original model is only one of many variants.

The community or environmental context is vital. Drug courts emerged as a community response to a community problem; thus some of the variety in drug courts is in response to the particular community's problems or concerns. This argues against the imposition of a single, uniform single model. It may be desirable to encourage adaptation of the drug court model to fit the community. The drug court exists in a criminal justice continuum, from diversion to incarceration, and should be understood in that context.

How do we adapt and export effective drug court models to mental health, family treatment, and other areas in which criminal justice oversight and supervision can play a role?

Research Needs

How can a research design be introduced in the implementation of the drug court?

It was suggested that research should move as quickly as possible from if the drug court works to how and when. This involves matching the drug court model to the problem, adherence to the drug court model, and training court and treatment staff, among other issues.

There is a need for theory-based research so that findings are generalizable and not limited to the specific drug court being studied. "Theory" can be based on experience (hypotheses about outcomes) as well as on behavioral or cognitive theories, organizational theory, etc.

In discussing the feasibility of comparing findings across studies, it was noted that there are problems and incompatibilities in data obtained from existing agency sources, but also in data collected in research and evaluation studies in non-uniform ways. Some of the categories of data that would benefit from standardized measurement included relapse, recidivism, follow-up time frames, definitions of "status" (e.g., "drug court failure" vs. pending case; warrant status; geographic limits that bias outcome reporting). The need for standardized data measures was mentioned several times, although it was also pointed out that local evaluations probably wouldn't benefit and therefore have little incentive to standardize data collection.

The paucity of research studies that include control or comparison conditions was discussed. It was indicated that there was a need for guidance in determining the appropriateness and adequacy of comparison and control groups.

Economics of the drug court: What is the cost-effectiveness of the drug court compared to alternatives, and what are its cost-benefits?

Transmitting Research Findings

Development of a template for drug courts so that courts know what best practices are and when they are deviating from them. May include:

• Treatment information
• Court outcomes (adjudication)
• Drug court models for particular offender profiles
• Measures of how well the drug court is working
• Utility to the community

Integrating research and practice - The need for a collaborative approach between researchers, treatment practitioners, and drug court personnel was identified. Also mentioned was the need for current information and feedback. It was suggested that longitudinal outcome studies might have limited utility in rapidly changing circumstances, since the system has often changed by the time the study has finished.

The need for training for drug court personnel and for treatment practitioners working with drug courts was mentioned. One participant suggested that most of the criminal justice "believers" in treatment - all the "easy" conversions - have been recruited to drug courts already, and many of the new drug court professionals are more punitive and less treatment-oriented. One related aspect of this is that it is becoming more difficult to recruit and train drug court judges and criminal justice staff.

In communicating research findings to the criminal justice system, the importance of professional perspective was raised. There may often be a gap between legal thinking (which focuses on legal precedents, which are case-based and give emphasis to outliers in making legal decisions) and research thinking (which is based on the statistical likelihood of a phenomenon being true on average or in aggregate, and which usually ignores or minimizes outlying behavior). The consequence is that the criminal justice system wants to protect against the "worst case" of a particular individual committing a crime, while the researcher tends to ignore non-representative "worst cases." A very good research result - an overall reduction in crime, for example - may be a poor criminal justice result, if the crime that is committed is notorious. This suggests that drug court research should include in its focus an emphasis on the relation between public safety risk level and treatment.

The meeting closed with a discussion of next steps, which include the intention of NIDA to continue to promote an active agenda of research on drug courts and other topics related to the intersection of the criminal justice system and drug abuse treatment.