The following presents the results of an evaluation of the Maine Drug Court program with a special examination of the treatment offered in the drug court program. The evaluation strategy was designed to monitor the progress of drug court offenders as they progress through the various treatment stages included in the DSAT model of treatment used by the drug courts in Maine. The study team had hoped to obtain a comparison group of similar offenders that were placed on probation during the same period of time but the University of Southern Maine (USM) and the Maine Department of Corrections could not reach agreement during the two years of this study regarding access to this comparison group.
Data collection for the study took place from mid-July 2003 through November 2004. The research strategy involved collecting information on drug court offenders from administrative and case management records as well as through a series of personal interviews with drug court offenders. Interviews were conducted at a baseline period (entrance into the study) and then again at three and six month follow-up periods after the baseline interview. Of particular interest in the study was whether offenders made any clinical progress during the period of time that the offender was involved in DSAT and the impact of drug court processes (e.g. treatment, drug testing, sanctions and status hearings) on offender outcomes.
Several data sources were merged together to conduct this study. A unique research identifier was developed to allow for the linkage of data sources.
Administrative Data. The data were collected for all drug court offenders as part of a routine admissions process into the drug court program. Basic demographic information was gathered on the offenders’ education, employment, family and living situation including drug use, treatment history, and criminal history.
Case Management Data. The final status at the end of the study period and documentation of DSAT stage progression were also included. The data also included rewards and sanctions imposed on drug court offenders during involvement in the drug court.
Study Instrumentation/Personal Interviews. Personal surveys were conducted with those offenders that agreed to participate in the study (see below for a discussion of this). The surveys administered included a criminal justice supervision survey, a legal pressure survey, and addiction severity index (ASI), and outpatient treatment surveys. These surveys belong to a battery of well-established and validated instruments that are used to capture key dimensions of the offender’s progress in the drug treatment program (and other correctional interventions). They were selected for this evaluation because of their capacity to measure offender progress in treatment during the different stages considering some of the psycho-social status of the offender such as the readiness to change, social supports, self-efficacy, hostility, conformity, therapeutic engagement, personal progress, counselor rapport, and so on. (Refer to Appendix C for a copy of the instruments). The surveys were completed by the offenders and took about an hour to complete.
Each offender had to volunteer to participate in this study. Volunteers were solicited across the six county courts in the Maine drug court program during the recruitment period of July 2003 through May of 2004. Offenders were given the opportunity to participate through in-person requests made by independent researchers. Gift certificates to local grocery stores were offered as an incentive to participate.
During this recruitment period, offenders had a choice as to whether to participate in the study. A total of 99 offenders (34% of the 290 active clients) agreed to participate in the study. The primary reason that offenders indicated that they were not interested in participating in the study was that the offender had a busy schedule that did not allow them to easily juggle transportation, jobs, childcare and other responsibilities. As an alternative approach to recruitment, researchers approached offenders on treatment check-in day with case managers. On treatment check-in days, offenders are often required to spend considerable time meeting with or waiting for case managers. This approach proved to be a more effective venue for recruiting offenders to participate in the study. Grocery store gift certificates were again offered as an incentive. Researchers continued to recruit offenders for the study regardless of their phase in the program from July of 2003 until May 2004.
Of the 99 offenders that agreed to participate in the study and completed a baseline survey, 58 completed the three-month follow-up and 21 also completed the 6-month follow-up. Additional efforts to gather follow-up surveys from offenders that had been incarcerated included letters sent to the facility requesting continued participation in the study and several interviews occurred in prison.
Table 9 includes a breakdown of the sample by drug court of assignment. Most of the offenders are from Cumberland (26%) followed by Washington (23%), Penobscot (20%) Androscoggin (15%), York (11%) and Oxford (4%).
Table 9: Distribution of the Sample in the Study Across Drug Court
Characteristics of Study Population The offenders in the drug courts in Maine tend to be those with prior history of involvement in the treatment and the criminal justice system. More than three-quarters (78%) reported prior substance abuse treatment. The majority (49%) reported opiates as the substance of choice, followed by alcohol (27%), marijuana (11%), cocaine (10%) and other drugs (4%). Many of the offenders reported (74%) use of more than one substance (see Appendix A).
Offenders also started using substances at relatively young ages which according to the research literature increases their odds of having further problems with alcohol and drugs during their lives. The average age of first use was about 13. Those in the sample were also likely to report prior involvement with the criminal justice system. The average number of prior arrests was 7, which indicates that this population was also heavily involved in the criminal justice system. About 90% of the sample reported prior drug arrests. Approximately one-third of the sample (31%) has co-occurring disorders with a diagnosis of mental illness along with substance abuse.
Although a large number of the offenders reported having children (65%), there did not appear to be other personal characteristics which are consistent with stability. The majority did not have a partner at time of drug court admission (64%) and 44% were unemployed. Forty-seven percent (47%) had less than a 12th grade education. The overwhelming majority of offenders in the sample were male (67%) and white (93%).
As shown in Table 10, 45% of the offenders who participated were still active in the drug court. Thirty-seven (37) percent had graduated and 19% had been expelled. The most common reason for expulsion was for positive drug tests (9 of the 16 offenders). New criminal behavior was reported for 2 of the expelled offenders and 2 were expelled for failure to report to the drug court. No reason was reported for expulsion for four offenders
Table 10: Status at the Completion of the DSAT Study
Generalizability of the DSAT Study Participants
Since this study sample was voluntary, there was some concern about the generalizability of the results to the population of drug court offenders. Using the demographic data from the administrative dataset for all offenders admitted to the drug court from January 2002 until May of 2004, we were able to determine the extent to which this sample was representative of the larger drug court population. The administrative variables examined included: the offenders court, employment at admission, gender, race, education, family status, prior drug treatment, prior arrests and convictions and status at the end of the study period (see appendix B tables 1-10).
Using a chi-square measure of association or means test, the study subjects were found to be similar to the general population of drug court participants. That is the study subjects matched the drug court participants that did not volunteer for this study based on age, prior arrests, prior drug treatment, court, employment, race, education, and family status. Gender appears to be the one differentiating feature with more female volunteering to be in this study than present in the general drug court population; those participating in the study were less likely to have a partner at admission. These results support the generalizability of our findings to the Maine drug court population.
Offender Progress through the Maine Drug Court
Offenders admitted into the drug court are expected to move through the various treatment components based on the DSAT treatment model. The advantage of the DSAT treatment model is that it is a comprehensive treatment model that includes different levels of care based on the psychosocial functioning of the offender in the treatment process. Prior work on the DSAT model (see prior work that reviewed the DSAT model) found that the treatment curriculums were comprehensive and integrated state-of-the-art knowledge about cognitive behavioral strategies that present the offender with the skills to manage their addiction. The phases of the DSAT model are: motivational, intensive, and maintenance; most of the treatment time is spent in intensive sessions. The model allows offenders to move up and down in the treatment process according to their progress. One drawback is that many of the sites use closed sessions where offenders might have to wait before they can enter the program. According to Table 10, about 34% of the sample was recruited in the study during the motivational phase of treatment when they took the survey, 39% were in the intensive phase and 18% were in the maintenance phase. Several offenders remained in the drug court program after the maintenance phase as they waited to graduate from the drug court.
Table 11A: Treatment Phase at the Time of Entrance into the Study.
Table 11B: Length of Time in Each Phase of the DSAT Treatment Model
Table 11B presents the length of time spent in each phase during the study period for all offenders in the study. Sixty-eight offenders entered phase 1. The average number of days spent in phase 1 was approximately 87. Of the 68 offenders, 43 entered phase 2 with an average stay of 129 days. Phase 1 and 2 had a similar standard deviation of about 58 days or approximately two months. For phase three, the average stay was similar to phase 2, about 128 days but the standard deviation was much larger at 85 days or almost three months.
To examine if those who successfully completed the program differed from those who did not in terms of length of stay in each phase, we conducted an analysis of variance to test for significant differences. Table 12 indicates that there are no significant differences in length of stay by status (active, graduated and expelled) at the end of the study.
Table 12: Length of Stay in Each Phase by Status at End of Study
Sum of Squares
Sanctions are used by the drug court program to address noncompliant behavior. Table 12 is a list of violations documented by case manager for the offenders in the study. Of the 99 offenders in the study, only 58 (58%) received any sanctions for negative behavior. The negative behavior was generally for positive drug testing (32%). At least one offender had up to six violations for this issue. Program violations and missed calls were also issues for which offenders were sanctioned with about 14% of the population committing at least one of these behaviors. Nine percent were sanctioned for admitted drug use. It does not appear that sanctions are used when offenders do not obtain employment.
Drug courts used a number of sanctions to address behavior of the drug court offenders. The nature of the sanctions range from treatment oriented sanctions to more control oriented sanctions. Table 13 includes the sanctions used for negative behaviors in this study group. The most common form of sanction was incarceration with 40% of the offenders receiving that sanction, followed by increased reporting (18%) and termination from drug court (14%). Some of the available control oriented sanctions were not used at all including level demotion and house arrest. The treatment oriented sanctions used included community service with 13% of the offenders receiving this sanction at least once.
Table 14: Type of Sanctions By Specific Categories
From the sanctions options, we constructed a variable to capture the sanction orientation on a continuum ranging from treatment to control oriented for each of the offenders. The number of control sanctions was divided by the total sanctions and a four category continuum was created that ranged from no sanctions, treatment sanctions, mixed and control. Table 15 includes a frequency of this variable.
More than half of the sample experienced some form of sanctions which is expected in a drug court setting. Thirty seven (37%) percent of the sample experienced control oriented sanctions followed by mixed (9%) and treatment (8%). Table 16 shows that expelled offenders were more likely to experience control-oriented sanctions than the graduated or active groups. The types of violations by status group were investigated to determine if expelled offenders were more likely to engage in serious violations. In table 17, an analysis of variance is presented that compares the number of violations for particular behaviors across the active, graduated and expelled groups. The only significant differences in the number of violations among the three groups was for the number of positive drug tests and violations for new criminal activity. To determine which groups differ, an examination of the means of these violations across groups in Table 17 shows that expelled offenders were more likely to have violations for positive drug tests and new criminal activity.