What works and what does not.

• Position: Drug courts - Panel Discussion

FRIDAY, MARCH 1, 2002

MORNING SESSION

PANELISTS

Michael Rempel Center for Court Innovation

Adele V. Harrell The Urban Institute

Jeff Fagan Columbia University School of Law

Barbara Babb University of Baltimore School of Law

Michael Rempel Center for Court Innovation

I want to talk about treatment modality and participant progress in recovery. Most of my examples are going to be regarding drug courts, with just one other kind of court, the domestic violence court. Of cases handled in the Bronx Misdemeanor Domestic Violent Court, seventy-five percent are mandated to a batterer intervention program.

There are a number of other types of programs--substance abuse treatment, alcohol treatment, combined programs--that are available. For instance, one type of combined program, batterer intervention and alcohol treatment, is a function of the existence of the domestic violence court in the Bronx. Recently that program was determined to be poorly run, so that's no longer used anymore. So variables that may not directly relate to treatment, but relate more to the relationships between the court and available treatment providers may affect the modality used.

There are actually four New York City drug courts. This just gives the distribution of first treatment modality for those four courts. Immediately you see that Brooklyn and Manhattan are fairly similar. Both start out by sending about half of their participants to a long-term residential program.

Queens sends far fewer to a residential program, and when you look at other data about Queens, you see that they have a primarily young population, a population whose primary drug tends to be marijuana. So the clinical need is less, so they send fewer people to a residential program.

But then you get to the Bronx. Only four percent go to residential programs, eighty-five percent go to intensive outpatient. Well, that is not a function of clinical need. That is a function of the particular relationship between the court and treatment providers in the Bronx. In the Bronx, instead of referring out to a large number of treatment providers, they have a core of about ten to fifteen providers, all of which run intensive outpatient programs, some of which run other modalities.

But the dominant modality of treatment providers in the Bronx is intensive outpatient. Bronx courts tend to work very closely with these providers, so that the structural relationship between the court and those providers leads to an especially high degree of outpatient use.

I will just preview the results. If you were to look at retention rates across these courts, you could not necessarily conclude that one court is doing better or worse than the other based on modality they are using. But you would conclude that there are different types of relationships going on that are affecting the kinds of modalities being used.

Just to give a little bit of a overview of the other theme of the day that I want to talk to you about, I want to switch to the page that is titled "Graduate Compliance in the Brooklyn Treatment Court."

This is the other kind of necessity under treatment, that you have your modality, relationships between court and providers, and then in turn you have the participant response to that and their progress, not their outcomes but their progress, during the treatment and recovery process.

In this chart you see just the basic distribution of results for successful participants at the Brooklyn Treatment Court, and despite being; successful--everyone represented in this chart is a graduate--only twenty-eight percent had zero positive drug tests during their participation; only thirty-eight percent had no unexcused treatment absences; and only fifteen percent avoided going out on a warrant.

Immediately you see that the recovery process tends to involve extensive relapses. And for any type of problem-solving court, that in turn raises the question of second chances: How many second chances are appropriate for the population you're working with? Are there types of third or fourth chances that are appropriate?

For instance, in this context in the Brooklyn Treatment Court, if there were no second chances, there would not be a particularly good graduation rate.

This sort of details the process a little bit more. This shows again just for graduates at the Brooklyn Treatment Court what their non-compliance rates are across five periods of their treatment: the first thirty days of treatment; between thirty and ninety days of treatment; after ninety days of treatment but prior to completing phase one of that program, which requires 120 days--and also, to complete phase one, that 120 days must be consecutive, drug-free, and sanctionless time--and then phase two and phase three.

Immediately you see, after just the first thirty days of treatment and a positive drug test, for an example, a tremendous improvement. After averaging a positive drug test on thirty-eight percent of tests in those first thirty days, immediately after participants are past that point, you get tremendous improvement in level of compliance.

There is one final point I want to make about treatment and recovery process. That was just for graduates. We did an analysis where we looked at graduates and dropouts, and the question is, what is a warning sign that someone is about to drop out?

We found that positive drug tests were completely not significant as a warning sign that someone was going to drop out. Continued use was not important. What was important was attendance, commitment to the recovery process, that the participants who did not go out on warrants, who did not miss treatment appointments, tended ultimately to succeed, even if they had multiple positive drug tests, whereas the participants that maybe improved in their drug use but didn't have the commitment to the recovery process tended ultimately to drop out. The participants with absences tended to drop out. But positive drug tests were not at all an indicator that someone was headed ultimately to fail the program.

That's a very brief snapshot of two process issues that, by and large, have not been analyzed extensively but that give you a flavor for where I think the next five years of research is going to be in problem-solving courts, less focus on do they work--they are here, we all know that--and more on how the treatment and recovery process goes, what modalities are used, and what is the course of recovery.

Adele V. Harrell The Urban Institute

I think the defining feature of problem-solving courts is the difference in the way they use legal coercion. Traditionally, courts have concentrated on using their authority to appropriately punish behaviors in the hopes that somehow that would deter future non-compliance. Problem-solving courts are really turning their attention directly on the subject of the offender's behavior, with less concentration on the appropriate penalty for the immediate instant offense.

I think that change means that problem-solving courts have had to rethink our normal approach to research.

Deterrence theory is what most criminologists immediately turn to when they think about using legal pressure, and this sort of posits that people choose to comply or violate laws based on their perceptions of the costs and benefits associated with the behavior. There are three main components: the severity of the penalty they face, the certainty with which they will face it, and how quickly they will face it.

There are two types of deterrents, specific and general. Specific means that if you give the penalty to the person, they won't do it again. General means that the imposition of penalties will discourage others. Drug courts actually use this when they have people come to review hearings and watch other people being penalized. That's their effort at general deterrence.

What does the research say about it? It says certainty is the key; severity of punishment is important but not always significant, and it may depend on the certainty of punishment and on the salience of the penalty to the individual; and severity has rarely been found to have an effect on offender behavior.

I think what problem-solving courts need to do is switch and begin to think about what psychologists or behavior management theory has to say about the use of sanctions.

Important insights from that literature are that there are actually four kinds of sanctions. There are two kinds of positive sanctions and two kinds of negative sanctions.

There are rewards, which in drug courts are in the form of tokens and recognition, judicial praise. There is positive reinforcement. This means dropping something onerous that is going to happen to the person as a reward good behavior, such as requiring less appearances in court or fewer drug tests.

There are also punishments, which are in the form of graduated sanctions, which are immediate responses or penalties to negative behavior. And there is negative reinforcement, and that is the imposition of the penalty for failing to show the good behavior; that is, imposing the alternative sentence.

I think it is important to begin to think clearly about the different types of sanctions, and document what we are doing in these problem-solving courts that works.

The other insights from that literature are that the salience of the sanction varies by individuals. For example, one of the things you see in domestic violence research is that arrest is a deterrent for some people, those who are employed, and not for others.

The other insight from that literature is that contingency management is very important.

The principles of punishment are:

A punishment must be severe enough to motivate efforts to avoid it.

Punishment should be consistent, and should consistently follow unwanted behaviors.

Punishments and negative reinforcement should be contingent upon offenders' behavior in a predictable and controllable way.

There are several implications for the way...