The Role of Co-Occurring Disorders in Outcomes in Mental Health Courts
NRI—August 30, 2011
Well, good afternoon everyone. It’s 2:00 and I’d like to go ahead and get started. ‘Cause I know everyone’s time is precious. So welcome to the NASMHPD Research Institute’s webinar series. My name is Michael Lane. I’m the Events Manager here at NRI. We’re pleased that you all could join us for today’s 60 minute webinar. This webinar series is partially supported by SAMSHA. And we would like to thank them for their continued guidance throughout our summer webinar series. Before we get started, I’d like to go over a few housekeeping issues. If, everyone should be seeing Dr. Callahan’s presentation in the middle of their screen or sort of a left-hand side on right-hand side of your screen is, at bottom right-hand side is the closed captioning that we offer. If you do not want the closed captioning, you can minimize that and close that. In sort of the middle area where it says Q&A, that’s the Question and Answer section. You can type a question in there at any time during the presentation. However, we will not field those questions until Dr. Callahan has completed her Power Point presentation. Also, we will make the presentation available via PDF download at the conclusion of today’s talk as well as other reference materials that Dr. Callahan has provided for us all.
So why don’t we go ahead and get started now. I am pleased to welcome Dr. Lisa Callahan, a Senior Research associate, Project Director of the MacArthur Mental Health Court Study at Policy Research Associates. Dr. Callahan directs in the MacArthur Mental Health Court evaluation cost studies and NIJ Juvenile Mental Health Court Study.
Dr. Callahan first started working in at PRA when it was founded in 1988 to 1990. Prior to that she worked in the Research Bureau for the New York State Office of Mental Health. Dr. Callahan is a Professor Emeritus from the Sage Colleges where for 18 years she taught courses in Criminal Justice and Mental Health and developed the MA Forensic Mental Health Program. And today’s presentation is titled, Co-Occurring Disorders In Mental Health Courts. The MacArthur Mental Health Court studies are multi-site perspective research product that examines the outcome of MAC participants. In addition, a number of objective factors were going to be discussed today were measured, Â including official arrests, jail prison States for 3 years, pre and post-enrollment. Clinical outcomes, to the diagnoses, including Co-Occurring Disorder and Mental Health Court outcomes, such as compliance with conditions and graduation versus termination.
The results of this study will be discussed. And they show that mental health court participants are actively taking illegal drugs and during mental health court program participation and/or have a history of drug arrests have less success in mental health courts.
At this time, I’d like to welcome Dr. Callahan. Dr. Callahan, the phone is all yours.
LISA CALLAHAN, Ph.D.:
Okay. Thank you very much, Michael and thank you everyone for taking time to participate in today’s webinar. What I’d like to do is when I first begin, I’m going to give you a little bit of an overview about the study designs to setup really defining the things that I want to talk about. Because I think it has, it gives a good picture of where we started and how we’ve come to really some of the conclusions that we’ve drawn at this point that as a Mental Health Court study is coming to a conclusion. This is part of a larger set of studies that the MacArthur Foundation has funded on mandated treatment. And I’m really happy that you’re here today so I can tell you about our study. Many of you are familiar with mental health courts. This is really to start, so we’re all operatingÂ with the same frame of mind. The purpose of mental health board is really two-fold. One is to reduce recidivism in the criminal justice system and the other is to enhance treatment, not only access but engagement and community treatment. The mental health court movement has been growing quite rapidly since the middle or the end of the 1990’s Â into 2010, 2011.
And has really followed to some degree the pathway of the drug court development. There have been a number of studies on mental health courts. Maybe some ofÂ you in the audience have been those who have conducted the studies. Most studies of the single site studies, some have had a comparison group, some have not. And there are a few of the findings that I’m going to talk about today really underscore how important it is to have a comparison group if resources allowed. There have been a few multisite studies as well. But generally those have not focused as much on outcome as they have on process. The Â study that I’m basing my presentation on today remedies some of these problems, but certainly is not without its own limitations. But the MacArthur study is a foresight multisite study. We do have a comparison sample and it was a prospective study and a longitudinal study. And that what I mean is we enrolled people entering the mental health courts as they were entering the mental health court. It was not a retrospective. We didn’t capture subjects who had already been in the mental health court. Because there were measures that we wanted to take at baseline as they were coming into the court.
This is a depiction of what our study design is and what are samples are. We haveÂ Mental Health Board enrollees and the comparison group. We refer to these alternately as the jail or the TAU treatment as usual groups. Â And the Mental Health Board is designated by MHC. I generally refer to them as the TAU’s. The four sites that we have, and I’ll give a little bit more information about these in a minute. Are San Francisco, San Jose, which is Santa Clara County, Indianapolis, Indiana, which is very tiny andÂ Minneapolis, Minnesota, which is Hennepin County. And this gives you an idea of the size of our samples. We over sampled the treatment as usual groups because we thought we would have more difficulty finding them for follow-up because they weren’t under the supervision of the court. This is the, these are the criteria we used to select the courts. There certainly are many courts we’re studying throughout the United States. But in order to be in the study we wanted to make sure they were large enough. That we had something to study. That they accepted those misdemeanors and felonies. Because as many of you know, the early first generation court accepted for the most part only misdemeanor defendants and second generation began to accept felonies.
We needed to make sure there was a large enough county jail. That we had a large enough base to draw our treatment as usual sample from. And the court had to be sustainable. Some courts have not enjoyed much legislative or financial support and so they’ve gone out of business over the budgetary crises. So we wanted to make sure that there were sustainable courts. And as I describe, these are the four courts we ended up selecting. And in each case we have similar treatment as usual, and mental health court samples. But I want to just say here, is that we did not match person to person with the treatment as usual to the mental health court subjects. Instead we matched on characteristics. We wanted, first we took all the mental health court people who were becoming enrolled, who were willing to participate, obviously. And then we matched on diagnosis, age, gender and offense, their target offense. It was not a perfect match and Â certainly within and between jurisdictions, what is a misdemeanor and what is a felony varies widely but we did a pretty good job of matching. And we’re able then in most cases pool our data. And I can explain later if you want to know why we have the good Â argument for pulling the data rather than looking at it site by site. This might seem like a, for those of you who work in mental health courts, you probably can answer this question right off the top of your head.
But again, by pulling the data we get an idea of how mental health courts actually work. I should say here first that we, we believed we knewÂ how mental health courts work and then actually started studying these four courts and found that some of the beliefs that we had, were in fact mistaken. There’s a lot on this slide. But let me explain to you what’s here. We were interested in the time from arrest to mental health court enrollment. It was our belief that people who were eligible for the mental health court would become involved and enrolled in mental health court more quickly than if, then the comparison treatment as usual group would, would terminate their stay in jail and their disposition. We found that not to be the case. As you can see here across the top we simplified the stages from arrest to referral to the mental health court decision to the entry. And we found that in fact from arrest to entry is longer in terms of median days than it is for the treatment as usual group from their arrest to their disposition. Â One difference, of course, is that upon entry into mental health court, the mental health court participants are not in jail. Many times that’s a condition of entry, is that there has to be placement for them both in terms of treatment and residents.
But for the treatment as usual, they may in fact have remained in jail because their disposition might have been a conviction and a sentence. But we were not, this notÂ confirm our hypothesis that the mental health court enrollment was relatively quickly. And I would suggest this might be one area that mental health courts want to do some research into their own timeframe because many times, people who are eligible for mental health court, once they’re referred and a decision is made can be quite quick. But the resources in community aren’t there to provide for their release and connection into treatment. So in terms of expediting that process, that’s one area where mental healthÂ courts can perhaps speed it up a little bit. We looked at, again we sort ofâ€” we thought was a simple question. We assumed that for the most part, people either graduated or were terminated from mental health court. We were surprised to find that a large percentage, especially in our two California sites, San Francisco and San Jose, we were surprised to find a large percentage of mental health court participants who at 12 months were still under mental health supervision. That wasn’t the case in Minnesota or Indiana. Quite the opposite in Indiana, as you can see.
But we were surprised that the number and proportion of mental health court participants who still after 12 months were under supervision. Â I think there were some characteristics of these two courts that helped to— that we can explain those findings. But it was a surprise. Â Especially when you contrast it with the other two sites. And especially when you look at San Jose. Part of the explanation we believe lies in the fact that San Jose has a large number of people with co-occurring disorder in their mental health courts. Another process question in terms of how mental health courts work is depicted here. We collected data on those who are, just those who are in mental health courts. The TAU’s are not here. What percentage of their hearings before the mental health court judge occurred whether they were in custody or not. And how that was related to their outcome. This no surprise. People who graduated had the smallest percentages of hearings that were unattended. Â The smallest percentage that were attended in custody and the largest percent that were attended and out of custody. That doesn’t surprise us, but we were impressed with the proportion of people who were still in and terminated, who had either in custody hearings or they didn’t attend at all.
And one note I’d like to make here again for mental health courts who are thinking of what kind of data to collect for their own evaluation, this is something that was relatively easy to collect. And I would suggest can be very insightful in terms of the kinds of Â supervision and the kinds of resources that mental health courts find themselves faced, giving up. Because custody is obviously, expensive. Â We asked all of the mental health courts officials, the directors, they clearly have different names in each of the mental health courts. But we asked them to evaluate each of the mental health court clients at 12 months. Everyone who was in our study on a number of issues. But one had to do with compliance. And we had 100 percent response rate from all the courts and with all of the clients. So this, we’re very comfortable with this. And this was just their impression. There really was no wayâ€” w e didn’t ask them to go back and count or verify what percentage of meetings, they kept appointments, they kept, etc. But this was their impression. And we’ll find out later that there’s some, just a few seeds here that we’ve Â picked up on later in our analysis looking at people with co-occurring disorder. So people whoâ€” and I do want to draw attention here to the second bullet, blue bullet point. People who are charged with more serious offenses were viewed as being more compliant. And I think we can all speculate that. That might be in part because they had more to lose if they violate their terms of the mental health court. They may be facing State prison time or at least a lengthy jail time.
But one of the other points and guess I would ask you to make a note of this. Is when we look at the distribution of crimes or the offenses, people with co-occurring disorder are in mental health courts for less serious offenses. Â And so people with the more serious charges on the one hand may have more to lose that they violate. But on the other hand are less likely to be the individuals with co-occurring disorder. Â We found pretty much throughout our analysis and this is true no matter what we’re looking at. Is that individual characteristics such as age and gender and education have no impact on outcome. No matter how we cut it, no matter what we control for. Those factors do not seem to be connected to outcome. Â One of the hallmarks of all the treatment courts and the mental health courts are no exception. Is that there are a number of incentives and sanctions that are official and some informal in terms of what people may use for improving. Â Or they may find themselves getting in trouble for. I’m not going to go over all of these. But I think one, what I would like to draw your attention to, and this bears out I think in all of the literature, all of the studies that have been done, is the judge is very important.
Not only to the establishment and continuation of a mental health court. But a judge really matters to the individuals who are in mental health courts. And certainly anecdotally I observed a lot of mental health courts proceedings, sometimes even sitting up on the dais with the judge which is a very interesting perceptive to have. And the often times very thankful words that the participant has to say to the judge is really quite moving. And I think are the self-report data from the interviews bears this out. They also believe that the judge is a be all, is a knowing person. That they’re going to find out if they violate their conditions. Even though judges find this rather amusing, ‘cause they
don’t find out unless they’re told. But they also think that if they do violate their conditions for the most part they will be sanctioned. On the right panel, I want to draw your attention to the middle sanction here. 23 percent of the people in interviews said that they have been sanctioned by being, required to go to jail as a sanction. We all know that going back to jail is probably the last resort and certainly the most serious of the sanctions that can occur to someone short of revocation.
And it’s one that about a quarter of our sample report is that they had experienced at the time of the 6-month interview. Again, this is self-reported. Everyone in this factor bears out quite significantly with our, when we look at this by co-occurring disorder or not.
Just quickly on this slide, we asked, as I mentioned, we had interviews with all of the participants. And we asked them a number of questions about their sanctions and incentives. And then we also, and on the next slide we’ll show that, we looked atâ€” we created some variables to look at it objectively. Not to check whether they were telling the truth or not. But just to examine it a little bit differently. And the yellow check mark just depicts that each of these findings in the interviews was born out by our analysis of the objective data. So people who went to jail reported that they went to jail, we’re more likely to, they reported they had more senses of mental illness in 6 months. The were homeless at baseline. More arrests since age 15. More incarcerations and last bullet reported using more drugs in the 30 days prior to the interview. In studying, given how important sanctions appear to be in both the drug courts and mental health courts, it’s rather remarkable that it’s not defined very well.
So in order to find out if people are sent back to jail or not, we had to do some digging and some creation, I guess really to extrapolate from a lot of that we had to determine if someone was in jail for sanction or not. Mental health courts do not report why someone has returned to jail. And you may say, well we do that. If you do, I think that’sâ€” you should be applauded for that. Most don’t. And again, this is something that could be helpful in terms of looking at the use of jail as a sanction is that’s of interest to anyone. And jails don’t report why someone is being booked. They just take everybody whose brought toÂ them for whatever reason. Consequently, we don’t really know what a sanction is. And so we certainly don’t know what a jail sanction is. So what we did is we merged a number of objective data that we had. Michael referred in the introduction that we had formal arrests. And those were FBI arrest data. We have lifetime arrests for everyone in our study. We have all the jail data for everyone in our study during the duration of our study. And we also have the mental health court hearing data. So we created something called probable jail sanctions.
And as a prior slide noticed, it was pretty consistent with what the self-report was. So based on our objective data, if you will, and if you saw me you’d stand pretty close around that, individuals will more jail sanctions have lower compliance with all of our measures of compliance. They have more jail visits in 18 months prior to their entry to the mental health courts. And jail visits mean the number of times they’re admitted in and not taking into court the number of days. They have more arrests since age 15. They’re more likely to have a drug offense as their target arrest and less likely to have a violent offense. And have been using more likely been using drugs in the last 30 days. So that’s pretty consistent with what the self-report data shows. As with others, individuals with more jail sanctions were less likely to report having received treatment at baseline. But there was no difference at 6 months. They were less likely toâ€” I find this really an interesting finding. They were less likely to think they’d be sanctioned if in fact, they violated their conditions. So there was a little bit of lack of insight here in terms of what the consequences were. And they were more likely to be older and more likely to have been homeless at baseline.
So the sanctions, I think you can get sort of a picture in your mind of the individuals now who are more likely to be sent back to jail. As consistent on the interviews and on the objective records of who they are. In terms of the outcome, if you recall in the first slide I had the purpose of mental health courts is to reduce recidivism and do enhance treatment. We looked at both of those. And one of the question is whether or not participation of mental health court did in fact increase treatment when compared to the regular criminal court individuals or the treatment as usual. And one side bar I would add here is treatment data is really hard to get. It’s hard to get and it’s hard to know what you’re getting. It does not include, for example, most of the data I am going to present. Let me put this way. The data that I’m going to present is reimbursable treatment. Â In other words, there’s a financial trail that this treatment was provided. Whether or not the person was engaged while they’re in the treatment, we can’t measure that. But they showed up and there was a billing for it. This raises, of course, all sorts of concerns about peer programs and safe where are peer programs are not reimbursable. Those wouldn’t be here. And jurisdictions that have very strict disability requirements. Those treatments wouldn’t show up here.
Because they’re not Medicaid eligible or SSCI eligible. So treatment data is really hard to get. We have all of the reimbursed treatment for everyone in our sample. But I want to get that limitation because we know there’s a lot more treatment going on that isn’t going to show up here. We found though that people who are in mental health court when compared to the jail sample or the TAU sample are moreÂ likely to get treatment. But they’re also more likely to get the treatment both before and after and entry into mental health court. And these are all significant findings. I didn’t include anything here or anything in the presentation that’s not specifically significant unless I specifically note that. And they also, and this is part of that how do mental health courts work question.
Mental health court participants are likely, or they get their treatments faster when they’re released from jail. They undo the TAU. And that’s often like and almost immediately they’re wrapped into treatment as opposed to the jail sample. They also are more the mental health court people are more likely to access therapeutic and intensive services, but they’re less likely after entering into mental health courts to use crises. And some of you I hope are seeing dollars and cents, because crises services are among the most expensive that are provided.
Lastly, what we found is, that the people who are in that still in category, not the people who were terminated, not the people who graduated but the people who were still in are very high users of the system. They’re very high users and they have very complicated issues that involve homelessness and Â co-occurring disorder and lifetime of being in jail, etc. Â Â So we’ll come back to that. But they’re very high users of the system. And again, you should probably be seeing some dollar signs on that one. Okay, the second. So mental health courts are to enhance treatment. And they seem to do that. They do get more people into treatment. And they get them in it more quickly. We also, and of course, the sales criterion, if you will, for many people who want mental health courts, is if they improve public safety outcomes. So the question is, are thereâ€” do they have better public safety incomes than the treatment usual people who just go through the regular court systems?Â We looked at re-arrests and we looked at re-incarcerations. Â And this is a simple question. Whose more likely to have an arrest post 18 months refers to the 18 months after entry into mental health court or in the case the TAU’s after their entry into the study. You can see here that treatment as usual people are less likely to be arrested and the time after their enrollment, then the treatment as usual are.
This is the type of crime that they’re likely to be rearrested for. For those who are rearrested in that 18 month period you can here I highlighted it. Mental health court people, the number one crime that they’re rearrested for is a drug offense. And there’s a pretty even distribution with the treatment as usual, in terms of property drug and other minor crimes. Just as an aside here, we were not able to, we chose not to look at misdemeanors versus felonies. Because that’s a relatively arbitrary and legal distinction that’s made for plea bargaining purposes and what might be a 4th degree felony one month, the next month is a 1st degree misdemeanor and that in fact, it happened in some of our states. So we instead are looking at it by what happened. And whether for example a victim was involved. Â We moved toward looking at something we call annualized number of arrests. Because our gut told us when we started looking at arrests that it wasn’t the full story. And so we created a variable called annualized arrests. And you can see in the box down below here, it takes into account the time the person was in the community. In other words, if they’re in jail the whole time, they don’t really have a risk of being rearrested.
So we created a new variable and we controlled for the number of days that they were in the community and whether or not they were rearrested. And what you’ll see is, if you look across the top, the mental health court people, there was a 38 percent reduction in the rate of arrests. In the 18 months before their entry into mental health court and the 18 months after their entry into mental health court. The TAU’s also had a decline. Again, just sort of surprised us. We didn’t really expect this, but they had a lesser decline. But the change for the mental health court people from before mental health court entry to after mental health court entry was a statistically significant decline. It was not as statistically significant decline for the TAU’s but it’s pretty close. But the reduction was greater for the mental health court group than it was the treatment as usual group. So this, the box that I’ve that’s at the top of the slide here taking into count, etc. this was one of those multi-varied analyses to make sure that we’re not just looking at things way too simply. So we wanted to know among the people who is the mental health court subject alone, not the TAU’s. Among the mental health court subjects who is most likely to get arrested in the 18 months after they are admitted to mental health court?
Now keep in mind that entry into mental health court almost always means released from jail, if they aren’t already in the community. They’re not sitting in jail during that time. So as you can see in this box here, we took into account, a lot of variables. We took into account individual characteristics. We took political characteristics. We took abuse history, etc. We took into account a lot. The increment genetic factors as well. All that seems to be, or not seems, all that is significantly related to likelihood of being rearrested among mental health court people is whether or not they’ve used drugs in the past 30 days and the greater number of arrests in the period leading up to mental health court. And the number of days of incarceration. In other words, the three factors that come out are criminogenic factors. They’re not clinical. They’re not individuals. They are criminogenic factors. We also looked at incarceration dates. And I want toâ€” a little bit. This is a good example of why it’s important to have a comparison group. Is we looked at the, just mental health court. If you didn’t have a comparison group and you say that in the 18 months before mental health court compared to the 18 months after mental health court Â there was actually an increase in the number of jail dates. That’d be a hard sell to the people who fund the mental health court.
However, for the same period of time you see that the jail or the TAU group actually increased by over 100 percent the number of days that they were in jail. Needless to say, that’s a significant increase for the TAU’s, but it isÂ not a significant increase for the mental health court. So without the comparison group this would be a hard, it’d be hard to conclude that there was an improvement in jail days. There was an improvement in light of the fact that had they not been in mental health court it’s likely their jail days would have increased even more. So this is just a quick summary of the public safety outcomes. As Michael says, these will be posted, the Power Points will be posted. But I think if the question were simply asked, do mental health courts improve public safety outcomes, I would feel comfortable saying yes, they do. Especially in comparison to similar people who do not go through the mental health court. Okay, now we get to sort of what we’ve Â been leading up here with some of the suggestions, if you will, from the data. And I think it’s really, I’m really happy that we can talk about this. Because it, in each of the analyses that we’ve done, whether we’re looking at treatments, whether we’re looking at sanctions, whether we’re looking at public safety outcomes.
The factors that keep percolating out, if you will, that seem to have an impact on outcome, is whether or not the person is actively using substances. So what I want to avoid doing, and I hope no one hangs up right now, because I don’t want to come to the conclusion that people with co-occurring disorders should not be in mental health courts.
I don’t think that, I don’t think the data suggests that and I think would not be a prudent public policy. I think it’s an important, I think this is important knowledge to have for mental health courts to know how better to address difficult clients in their mental health courts. And perhaps allocate resources appropriately. But I don’t want anyone to go please go away thinking that, well we shouldn’t have people with co-occurring disorder. As many of us knows somewhere around 75 to 80 percent of people with mental illness in jail also have a co-occurring disorder. So there wouldn’t be a lot of clients left if we did not include co-occurring disorders in our treatment course. So that’s just a little soap box point I wanted to make. So there’s the question that is do mental health court participants with co-occurring disorder have different outcomes than those without a co-occurring disorder?
Of course the answer is yes they do. Â Â Or we wouldn’t be talking about it. It’s really important to focus on co-occurring disorder in the, not only in criminal justice and mental health, but also in mental health courts. People with co-occurring disorder, I mean research is very strong on this. People with co-occurring disorder have greater impairments in their psychosocial skills. Â They’re more likely to relapse and they’re more likely to recidivate. So on the surface they’re, they don’t make great candidates for mental health courts. But as we know, they in fact can be great candidates for mental health courts with the right kind of services. One of the first steps, of course, is screening an assessment. One of the resources that Michael and Angela have to post is a document out of the game center. Which is cited here that goes through all of the screens and assessments for a co-occurring disorder. And it’d be a great resource. It is available in PDF as part of this webinar. What was really interesting to us is that we found that and when I’ve done this presentation before, not this presentation, but talked about co-occurring disorders, is the number of mental health courts that don’t screen for co-occurring disorder. They screen for a serious and persistent mental disorders such as schizophrenia, bipolar disorder, etc.
But do not really systematically screen for co-occurring disorders despite all of the evidence that people who are good candidates for diversion programs, jail diversion programs have high rates of co-occurring disorders. And we collected data from the mental health court records, and almost without exception, there was no note of a co-occurring disorder. Which of course led us to think that they’re just not recording it. Because we knew people had co-occurring disorder. And this is before we even looked at the data. Their primary diagnosis was recorded but not their, not whether they had a co-occurring disorder. So screening and assessment is really important. And the game center which is part of SAMSHA’s and CNHS recommends these screens. I’m not a clinician and I’m not going to, I’m going to make any attempts at explaining why these are the best ones. I trust the expertise of the authors of the document here. But not only is the screen necessary, but also an assessment. These are about 15 to 20 minutes. And that’s a lot of precious time and certainly and clinicians working in the jails and working in the courts. But it can save them ample time down the road if they are able to screen in and out people with co-occurring disorder.
You may be asking if you didn’t find co-occurring disorder in the records, how are you doing a presentation of co-occurring disorder?Â We had to dig into other records. As I mentioned we had enormous amounts of records and access to records. So what we did is we if the individual had a co-occurring disorder diagnosis in the court, which not a large percent did. We counted that as a co-occurring disorder. If they had a co-occurring disorder diagnosis in their treatment records, which we had access to, we counted that a s a co-occurring disorder. We also then had a combination variable that we counted as a co-occurring disorder if they received reimbursable services for specific types of Â treatment such as detox, residential treatment for drug and/or alcohol abuse and they self-reported that they had been treated for co-occurring disorders, we also then counted that as a co-occurring disorder. So we needed to put together many resources to identify individuals in our study that had a co-occurring disorder. As with many of the other analyses we looked at, with exception of education, the variables that seem to distinguish co-occurring disorder, people with co-occurring disorder from those in our study without were criminogenic factors.
Age at first arrest, the numbers of arrests since age 15, in the direction that you would predict. It also makes sense that people who begin their substance abuse history at an earlier point in their life and start getting arrested earlier are less likely going to able to finish higher levels of education. Again, gender, race, age, and history of psychiatric hospitalization were not factors that had any, made any difference. These are the target crimes of people with co-occurring disorder versus those without. Again, not surprising that I think whatâ€” what I think is one of the most interesting contrasts here if you look at the drug offense. People with co-occurring disorder are three times more likely to be arrested for drug offense than those without. And I don’t think that’s surprising, but it’s always, as a researcher it’s always a good thing when your data confirm your hypotheses and confirm the direction of other data. So this is consistent with what we have found in other types of analyses. We also, I thought it was important to look at their primary diagnosis just to make sure that, and as you can see, there’s absolutely no distinction among those with or without co-occurring disorder and what their primary diagnosis is. It’s remarkably similar across the board.
Okay, in terms of their participating in the mental health court, some of these variables should be familiar to you by now. I talked about them before. They’re people with co-occurring disorder are less likely or perceived as being less likely to comply with Â their conditions of the court. And I have an asterisk here. We didn’t ask them, are people with co-occurring disorders less likely to be comply. This is just a statistical relationship. They’re also more likely to have their mental health hearings while they’re in custody. In other words, they’re in jail. And they’re more likely to be sent to jail as a sanction. The outcomes of persons with co-occurring disorder again are less positive, I guess you could say, than those without a co-occurring disorder. And these are all statistically significant. Almost, well 81 percent of people with co-occurring disorder are rearrested after enrollment into mental health court versus those without. On the average, they spend0 more time in jail. After they’ve been in mental health court than those without. And we know that from the prior slide they’re more likely to have their (WORD?) in jail. And more likely to be sanctioned by being sent back to jail.
And they’re more likely to be terminated from the mental health court. And I might add termination doesn’t come easily to the judges. They like to give participants, they seem to like to give participants many chances for success. Because relapse is part of the process in a mental health court, while abstinence is one of the key factors in drug courts.
Abstinence is not mentioned in the key elements or the essential elements of mental health courts. It’s really more of a recovery model. So they’re often given a chance to prove themselves and eventually, but they’re more likely to be terminated from the court. Often the bottom line has to do with cost. And we are in the process, Michael mentioned at the introduction, we also have a cost study on, going on right now looking at our mental health courts. It’s extremely complicated for the reasons that I mentioned that the way in which treatment is billed. It has an impact on what the costs appear to be. And so we find that there’s wide variation across our three sites in terms of the cost. And we know that’s in part because, not because they’re not getting treatment in other jurisdictions, it’s just not billable treatment. And so we can’t capture it.
But in terms of people with co-occurring disorder. We k now they’re high users of the treatment in criminal justices and particularly jail days. Recall that we found that people with this co-occurring disorder, or the people who are still in are likely to be people who Â haveÂ co-occurring disorder. They so they’re high users of both the treatment and justice system. They go Â in and out of jail, whether it’s for a re-arrest or whether it’s for sanction. We ended up having to drop San Jose from the cost part of the study for reasons that are multiple. And we constituted. But in theÂ three sites that is we have in the cost study, the cost for persons without co-occurring disorder went down in over half of the participants in the period of time after they entered mental health court. Boy, that was a complicated sentence. In other words, we all, we want the cost to go down. And we Â have criminal justice costs plus mental health costs, which I can have an e-mail conversation with anyone who might want to know how we calculated that. But in over half of the participants without a co-occurring disorder their cost went down after entry into mental health court. But with people with co-occurring disorders they went down only for a third of the participants. So again, we can see that there’s a cost factor here and part of it’s because of the high user factor.
The only variables that are related to costs are the pre-enrollment incarceration days. In other words, people who spent a lot of time in jail before mental health court, it costs more after the fact. And those who have a co-occurring diagnosis. And when I say only variables I would like you to trust me to know we have, we put at least 50 variables both individual, and manyâ€” and criminogenic factors, clinical factors into the model. And the Â only two that are related are the number of incarceration days before mental health court and whether or not they have a co-occurring diagnosis. That was the only factors related to cost. So conclusion. Â Conclusions from the entire, these are sort of summing up the whole 45 minutes here. Mental health courts reduce recidivism compared to the TAU sample. These are all significant findings. MentalÂ health court participants receive more treatment in the community than the TAU sample. So mental health courts do meet the goal of treatment and reduction of recidivism. Mental health court participants with co-occurring disorder generally have less favorable outcomes. And listed here, are those, well many of them having to do with process and some having to do with their additional arrests.
I wanted to, race was point again for people who are in the mental health courts that are in the audience to remind them to make sure that this becomes part of the diagnostic process. Most co-occurring disorder does not show up in the mental health records. Mental health courts should screen and assess for co-occurring disorder using validated tools, specifically the tools that are recommended by the game center and by clinicians who study these things. And rather than conclude that mental health courts are not good diversion programs, for people with co-occurring disorder, I would like to argue that the data really shows that it gives us an opportunity to deal with treatment and supervision for people who have co-occurring disorder so that they have positive outcomes. And consequently would reduce the costs for mental health courts. And that’s the end of my presentation. And here’s my contact information.
Well thank you, Dr. Callahan. At this time, I’d like toÂ solicit any questions you may have for Dr. Callahan. We have a couple right now in the queue. And why don’t we go ahead and take those at this time. One of the questions we have, is for those for whom costs did not go down, did they go up or were they comparable to what they would have been without the mental health court involvement?
LISA CALLAHAN, Ph.D.:
I’m going to, well for the most part they went up. And I would like to I would like to add a contingency there. We are still analyzing the cost data, but that phase of the study is not over. But there clearly are layers of people for whom costs either went down, they stayed the same or they went up. For people who went up there, that they’re who you would predict in terms of being the high users. People with lots of criminal justice backgrounds and lots of treatment demands. And a variety of treatment demands, not just for co-occurring disorder but for other really behavioral health problems that encompass homelessness and the like. We are in the process of finishing up that analysis. And hopes to be able to be definitive on that. And I’m not trying to be evasive. But that’s really the best answer I can give you. We really have three layers as you mentioned. And you know, stay the same or go up. They’re clearly for people for whom they stay the same. But they’re also people for whom they go up.
Okay, great. Here’s another question:Â Did any of your data come from Ohio Mental Health Courts?
LISA CALLAHAN, Ph.D.:
They did not. Our, all of our data are from the four sites that were listed, Santa Clara, San Francisco, Minneapolis and Indianapolis. I know Ohio has a lot of mental health courts. They probably have among the largest number of treatment courts proportionately in the whole country.
Okay, great. Here’s another question:Â How does tracking COD fit in with extra confidentiality requirements of 42 CFR for substance use?
LISA CALLAHAN, Ph.D.:
You’ll have to tell me what 42 CRF is.
Okay. Well, why wereâ€” let’s see. Here we go. WeÂ have another question here. It’s been very difficult to find co-occurring treatment programs that deviate from highly confrontational models and truly accommodate serious and persistent mental illness. Are MAC and Las Vegas does diagnose for co-occurring disorders?Â That was a comment.
LISA CALLAHAN, Ph.D.:
And one of the resources that I sent along to be attached to this is a document put out byÂ Fred OsherÂ a psychiatrist wrote about integrated treatment. And that maybe will speak to that question. And it’s a two or three page document that speaks specifically to the type of integrated treatments that are important.
I think what people should be seeing on their screen now is Angela is going to load up the resources that Dr. Callahan’s been referring to. As well as the presentation of it for PDF. So there should be probably about five things will be available for download. And I know we had one question from Tricia. Who were looking for what the 42 CFR for substance abuse, what that is so that we could effectively answer your question. It’s the federal law for protecting release of substance abuse information that limits what can be released to whom.
LISA CALLAHAN, Ph.D.:
I really feel that they’ll find in terms of what impact that might have on a person’s legal state is I feel solely outside of my element being able to answer that. The diagnostic process is a clinical process to the best of my knowledge that’s done. And I realize that in a forensic setting and a jail setting or a court setting there are other limitations on that.
I don’t know the answer to that question. And I’d be wading into an area that I’m not Â capable if I tried to answer to that specific piece of it. Maybe there’s a lawyer in the group or a clinician who would be able to answer that.
Okay, Dr. Callahan. The other question we have is what behavioral health diagnoses was the highest percentage of mental health court participants?
LISA CALLAHAN, Ph.D.:
Oh, well I know, I can tell you which article up here has that. It is the Steadman et al, which is the Archive Journal Psychiatry article we lay out the diagnosis. It varies some by site. If my recollection is, both schizophrenia and bipolar disorder or depressive disorders come in pretty close, they come in fairly high together. We chose people into the study who had a serious mental illness intentionally and in many cases the mental health courts limited to people with serious disorders anyway. Not with less, you know, Axis I type diagnoses. But I would, if the person can look at the Steadman et al article and I know there’s a neat table there that compares TAU’s and their mental health court and also by site.
Everyone should see in the center of their screen a file transfer. And if you were to click on either one of these five file names, and you click on it and you can download and it would download directly to your computer. And as Dr. Callahan said, there’s four resources and then her presentation is the bottom one there, NRI, mental health court COD, PDF, that is the Power Point presentation.
LISA CALLAHAN, Ph.D.:
Can I warn people that the first one is quite long. That’s screening and assessment is quite large. I just wanted to warn them ahead of time. It’s like 2 meg’s I think.
Yeah, I think it’s the size right up there. And so is your presentation it’s pretty large as well. But we have a couple of more questions coming in at this time. So we have one from Kathy here that says, thank you for such a great presentation. I am interested in hearing more about your rational for pooling the individual samples. Perhaps this is not of interest to others. So a e-mail response would be fine.
LISA CALLAHAN, Ph.D.:
I’d be happy to explain to either to the group or individually. It’s, I kind of dropped a little Â hint in the beginning and never followed up on it. The main reason and I could go into the statistical reasons for it off, you know, the regular e-mail conversation is. If we were to do it every time, we controlled for site in those cases and the Steadman et al article everything is controlled by site. So we do take into account site differences in all of our analyses. But as part of the purpose of doing a multisite study is to not do four case studies. Which if everything is done by site, we might as well as have just done four Â single-site studies. And s we work with our statistician who helped and worked with us in terms of creating theâ€” all of the right formulas for waiting and controlling for site. And in each case that was introduced into the model to make sure we weren’t just displaying, you know, the oddities rather than the normal distributions. But she’s, feel free to e-mail me if she wants more information on that. I mean it’s something we struggled with, by the way. I’ll just add it, it is something we struggle with on a regular basis. In our own minds we go back and forth and remember why we pool the data.
Okay we have some more questions coming in. Were the mental health courts studied using pre or post adjudication models?
LISA CALLAHAN, Ph.D.:
Okay. Let’s see and another one here. What benefit is there to tracking what kind of sanctions are used in mental health courts?
LISA CALLAHAN, Ph.D.:
Well I think, I mean what I would say, there’s really, aside from just it being an interesting question in and of itself. I think that it’s, sanctions are resources. And sanctions, for example, in order toâ€” we found that a number of people were given jail days not asâ€” they weren’t standing before the judge and the judge saying, you haven’t been taking your medication or showing up to meetings. I’m ordering you to jail for three days. That wasn’t typically what happened. What would happen is someone would not show up for a couple of meetings and a bench warrant would be issued. And the police would either directly go find them. Or if they happen to find them would take them to jail for a few days. So I think that there is the use of sanctions like that, is worth examining. Because I think that it has, it’s clearly has cost implications that get spread across primarily the criminal justice system. And it also interrupts treatment. And very few jurisdictions is the community treatment then seamlessly provided in the jail. There are a few who do that. But the treatment interrupt, their medications interrupt it. And it is worse I think examining whether that‘s an appropriate use of jail for both treatment reasons and just cost reasons.
Other sanctions don’t have the same kind of impact, I mean in terms of cost impact being scolded by the judge, they don’t like it. But it doesn’t cost anything financially.
Okay. Thanks. We have a couple more questions here. And one of them is from Paul. He asks:Â one of the issues we have in our mental health courts is so many with Axis I as primary also have secondary Axis II personalities. Could you discuss this?
LISA CALLAHAN, Ph.D.:
We did not, much like co-occurring disorder, we found that variable documentation of any Axis II diagnosis. I think that it’s just extrapolating from the jail population generally there’s such a high incidence of it. We didn’t, we did not look at either whether it be psychophacy as a measure orient, i.e. social personality disorder. And it was we did attempt toâ€” we recorded it as it was there. But it was not there very often and it certainly was underreported. Clinically, I’m sure it has profound implications. We were not able to capture that in our study.
Okay. Here’s another question. This is question is from Allison. She asks, you mentioned that you didn’t distinguish between felony and misdemeanor arrests, but might felony convictions have a big negative impact on employability which in turn will have a negative impact on co-occurring disorders and criminal justice outcomes?
LISA CALLAHAN, Ph.D.:
What I meant, when weâ€” our analysis is based upon what they were charged with. You know, what their target offense was. We attempted initially to look at misdemeanor versus felony. Because it seemed, it was their literature and it seemed to make sense for reasons that Allison mentions. It proved out to us to be, for purposes of analysis, and certainly there’s practical implications that are not incidental. It was really, it was not analytically making any sense. Because for example, just in Minnesota, I’ll just give you an example. One crime that many were charged with was a type of assault that went from being a 5th degree felony to a 1st degree misdemeanor over the course of our study. And nothing changed about the type of crime what happened, or even the outcome. And so we found that it was really a false dichotomy for our purposes for analysis. You are absolutely right in that it has practical, I mean very important implications in terms of whether it could carry a state prison sentence or a jail sentence. I’m not sure that that necessarily matters so much clinically. I mean it I think it matters criminal justice, in terms of criminal justice outcomes. But in terms of what they‘re called or what the potential sanction is.
But it really, we found and just a false dichotomy for analytical purposes. And also it was an area where there was a lot of negotiation between your public defender and the prosecutor. It didn’t change the nature of what happened, it just changed the nature of what it was called. And I’m sure that was an inadequate response. I think she’s absolutely right. It has practical implications. And I don’t think it was something that because of our unitâ€” because we how we were doing the analysis and questions we were asking, it didn’t make any sense to us.
Okay. Let’s see it is right at 3:00. We probably have time for a couple of more questions, if there are any. We’ve had some individuals have trouble accessing the Power Point download. And we will have this available on our website if individuals are still having trouble being able to access the file transfer box or you could e-mail myself or Dr. Callahan’s e-mail is there, too. And she could send you the resources directly. But for those that are having trouble we will have that posted on our website.
LISA CALLAHAN, Ph.D.:
And I’m also happy to answer, ‘cause I think I’m going to lose my conference room here pretty soon. I’m happy to answer questions, you know, if people want to e-mail me the information that is there is accurate. And I’d be happy, if something may to come to mind later as people are trying to digest the materials.
Well, I just want to thank you from NRI’s perspective, Dr. Callahan for putting this together for us and being part of our webinar series with SAMSHA. I mean this has been very useful. And I appreciate your time. And I look forward to hopefully having you come back and present, you know, follow-up results. You know, and I think it would be a great opportunity to share maybe, you know, six months to a year from now what PRA is doing. So we would be interested in having. I just want to say thank you so much for putting this together.
LISA CALLAHAN, Ph.D.:
Thank you. And thanks for inviting me. I enjoyed talking to others. It was good to think about it in terms of co-occurring disorders specifically, too.
Well, if we don’t have anymore questions at this time. And we have a couple more comments. Thanks again from Virginia. And thanks very much, very interesting. We can, we’ll end the presentation. And if there’s anymore questions, you can contact myself, the Events Manager at NRI. My contact information can be found on our website. Or you can contact Dr. Callahan. But again, thank you all for joining us for today’s webinar and have a wonderful rest of your day.
LISA CALLAHAN, Ph.D.:
Thank you, everybody. Goodbye Michael.