Intervention Comparisons

Journal of Substance Abuse Treatment 41 (2011) 288–293

Brief article

Preliminary evaluation of extended-release naltrexone in Michigan and Missouri drug courts

Michael W. Finigan, (Ph.D.)a,⁎, Tamara Perkins, (Ph.D.)a, Phyllis Zold-Kilbourn, (Ph.D.)b, Joseph Parks, (M.D.)c, Mark Stringer, (M.A.)c

aNorthwest Professional Consortium, Inc., Portland OR, USA bUnder contract to Northwest Professional Consortium, Inc., Portland OR, USA

cMissouri Department of Mental Health, Jefferson City, MO, USA

Received 7 October 2010; received in revised form 20 April 2011; accepted 22 April 2011

Abstract

This pilot study, a retrospective case series analysis, examined the feasibility and effectiveness of treating alcohol dependence with extended-release naltrexone (XR-NTX) in the drug court setting. In two Michigan courts and in one Missouri court, 32 clients were treated with XR-NTX and were matched with 32 clients with standard care in an open-label, voluntary recruitment design. Treatment with XR-NTX was associated with relative risk reductions (RRRs; p = ns) of 57% fewer missed drug court sessions, a 35% reduction in the monthly ratio of positive drug and alcohol tests to total tests, and 35% fewer individuals with greater than 25% overall positive alcohol or drug tests. In the principal end-point analysis of annualized number of new arrests, 26% of standard-care clients were rearrested versus 8% on XR-NTX (RRR = 69%; p b .05). Treatment with XR-NTX appeared to be feasible and was associated with a consistently large treatment effect across multiple outcomes relevant to the drug court setting. © 2011 Published by Elsevier Inc.

Keywords: Alcohol dependence; Criminal justice; Law enforcement; Naltrexone; Vivitrol

1. Introduction

Alcohol dependence is a major public health problem, with an estimated prevalence of 6.3% in adults 18–29 years old and 6.4% in adults 30–44 years old (Kessler & Wang, 2008). Alcohol has been identified as a significant risk factor for criminal behavior (Greenfield & Henneberg, 2001). Approximately 36% of all violent and nonviolent crimes (nearly 2 million per year) are committed under the influence of alcohol (Greenfield, 1998). A 10% increase in heavy drinking rates is associated with a 2% increase in arrest rates among young adults (Carpenter & Dobkin, 2010). Despite reductions in rates of drunk driving over the past 20 years, alcohol continues to be a major public health problem and is estimated to be a contributory cause in 39%

⁎ Corresponding author. Northwest Professional Consortium, Inc., 4380 SW Macadam Ave., Ste. 530, Portland, OR 97239, USA.

E-mail address: [email protected] (M.W. Finigan).

0740-5472/11/$ – see front matter © 2011 Published by Elsevier Inc. doi:10.1016/j.jsat.2011.04.003

of fatal motor vehicle accidents (National Highway Traffic Safety Administration, 2009). Approximately one third of DUI or DWI arrests are rearrests of recidivists (Nochajski & Stasiewicz, 2006).

The high proportion of offenders with alcohol and substance abuse problems and the high rate of recidivism among this offender subgroup led to the introduction of drug courts in the early 1990s (Nolan, 2001) and more recently of DUI/DWI courts. Drug court and DUI/DWI court (hereafter drug court) represents a “combined-systems” approach to offenders. The drug court team typically consists of personnel from the district attorney and public defender and probation offices, as well as treatment providers, all under the leadership of a judge. The goal is to provide integrated and seamless management of the substance- dependent offender with a single point of accountability. Studies evaluating the effectiveness of drug courts suggest that they reduce recidivism rates by at least 50% (Bhati, 2006; Roman, Townsend, & Bhati, 2003; Gottfredson, Najaka, & Kearley, 2003). This has raised the question as to

289M.W. Finigan et al. / Journal of Substance Abuse Treatment 41 (2011) 288–293

whether specific treatment interventions can be identified that might further reduce recidivism rates (Marchand, Waller, & Carey, 2006).

Extended-release naltrexone (XR-NTX) is an injectable, once-per-month formulation of the opioid antagonist nal- trexone (Vivitrol; Alkermes, Inc., Waltham, MA) that has demonstrated significant efficacy in treating alcohol depen- dence (Garbutt et al., 2005). In alcohol-dependent patients with at least 4 days of abstinence, treatment with XR-NTX is associated with a threefold increase in the rate of 6-month abstinence, a 90% reduction in the median number of drinking days per month, a 95% reduction in the number of heavy drinking days, and a greater than ninefold delay in the median time to the first heavy drinking day (N180 vs. 20 days; O'Malley et al., 2007).

A pilot program using XR-NTX in alcohol-dependent offenders in the drug court setting was funded through the Missouri Department of Mental Health's Division of Alcohol and Drug Abuse in cooperation with the Missouri Department of Corrections. The objective of the current analysis was to obtain preliminary data on the effectiveness of XR-NTX in reducing rearrest rates and maintaining abstinence and compliance in alcohol-dependent offenders in the drug court setting.

2. Materials and methods

2.1. Study plan

This was a retrospective records analysis of anonymized naturalistic data comparing rearrest rates and other outcomes at approximately 1 year among alcohol-dependent offenders who were referred for treatment with XR-NTX, open label, and a matched (post hoc) control group who received standard care. Outcome data on alcohol-dependent offenders were analyzed from three drug court sites that were early adopters of XR-NTX. Because the analysis was retrospec- tive and used only data that were available as part of the standard (nonclinical) legal and case management drug court record, no informed consent was obtained, and no protocol was submitted to local institutional review boards (IRBs). However, to ensure confidentiality and ethical use of the retrospective data, the following steps were taken: (a) We consulted the national drug court IRB that had previously approved the use of these methods to conduct retrospective analyses without the use of a local IRB; (b) in spite of the fact that these were public records, the iden- tifying information (names, etc.) were redacted once the data were gathered; (c) Michigan required that we submit a separate protocol to the State of Michigan IRB, which provided approval; (d) We put into place data-sharing agreements with all of the court (this meant that they could share information on current clients and they had release forms from the clients for this); and (5) the Michigan data collector signed a special confidentiality agreement with the

Michigan DUI courts to access their database to generate a matched comparison group. She extracted the data using a redacted Excel database.

2.2. Subjects

Study subjects were male and female adult volunteers identified from a criminal population that consisted of anyone charged with a DUI, or who was charged with another offense but had co-occurring alcohol dependence disorder. Subjects had a diagnosis of alcohol dependence, tested positive for alcohol and other drugs multiple times, and had problems with compliance with the demands of drug court and had continued to drink after all other interventions had been tried (daily AA meetings, inpatient and outpatient treatment). Subjects were excluded, from both the XR-NTX and control group, if they had a history of violence, or if they were arrested for a violent offense (e.g., assault).

Subjects in the control group consisted of defendants within the same drug court with alcohol dependence and similar offenses who were arrested in the 12 months prior to the availability of XR-NTX, and who were matched, post- hoc, on five baseline demographic variables: age, gender, race, diagnosis and criminal history. The period for the col- lection of data was short and we detected no major changes in drug court policy or procedures during the period during which individuals in the control group and the XR-NTX treatment group were arrested.

2.3. Treatment

Candidates for treatment with XR-NTX underwent a medical evaluation to determine whether any contraindica- tions to treatment with XR-NTX were present (including concomitant use of or dependence upon opioids, pending chronic pain that would be expected to require opioid analgesia, clinically significant hepatic laboratory findings, and so on). Subjects who were medically approved were scheduled to receive intramuscular injections of XR-NTX 380 mg every 4 weeks for at least 9 months. However, as is typical in naturalistic settings, actual doses varied. The experimental group received a mean of 4.33 injections with about a third receiving six or more injections. In addition, individuals in the experimental group received standard care in the community. The matched control group received only standard drug court with care in the community. In standard care, subjects were scheduled to attend group treatment sessions four times per week for the first month and two times per week thereafter; to attend individual treatment sessions once per week (at least for the first month); to attend drug court sessions once per week for the first month, once every 2 weeks for the next 3 months, and once per month thereafter; to attend 12-step self-help meetings each week; and to provide at least four random breath alcohol or urine drug screen tests per week during the first month, two per week for the next 3 months, and one per week thereafter.

290 M.W. Finigan et al. / Journal of Substance Abuse Treatment 41 (2011) 288–293

2.4. Data collection

Data collection for the Michigan courts was based on the statewide Drug Court Case Management Information System, supplemented, where needed, by a review of paper records. Data from the St. Louis site were collected based on a review of paper records.

2.5. Drug court procedures

There were two Michigan drug courts. One, first established in the fall of 1999, was an adult drug court with a large population of DUI offenders. The second was a DUI court established in the fall of 2004. Both had clients with second or third DUI offenses. The St. Louis drug court was first established in the Spring of 1997. It is an adult criminal drug court with a substantial number of DUI cases and with a target population of nonviolent felony offenders with substance abuse.

Once a defendant was arraigned and entered a voluntary plea agreeing to participate in an alcohol intervention program, the probation officer conducted a brief evaluation to determine whether a Diagnostic and Statistical Manual of Mental Dis- orders, Fourth Edition substance abuse diagnosis might be present and what the appropriate level of care was based on Patient Placement Criteria of the American Society of Addiction Medicine(Mee-Lee, Shulman,Fishman, Gastfriend, & Griffith, 2001). As part of the program, participants were required to attend review hearings, report to their drug court case manager, submit to random alcohol and drug testing, attend self-help groups, and attend substance abuse treatment. If the decision to use XR-NTX was made, the patient was referred for medical screening to determine whether there were any medical contraindications to XR-NTX treatment.

2.6. Outcomes analyzed

Data on three component outcomes and a principal end point were available for analysis. Compliance was analyzed based on the number of missed drug court sessions per month. Abstinence data were analyzed based on the number of episodes of positive alcohol and drug tests per month. The proportion of subjects with low levels of abstinence or persistent drinking or drug use, defined as greater than 25% positive tests, was also analyzed. Because no consensus criterion was available, a 25% criterion level was used based on face validity (i.e., feedback from court evaluators). Finally, for the principal end point measure, rearrest rates were analyzed in terms of number of new arrests per month for subjects in the XR-NTX treatment group compared with the matched control group (standard care). Because the mean duration of treatment was longer for subjects treated with XR-NTX (13 months) compared with the control group (11 months), the new arrest data were annualized.

Because this was a pilot analysis of retrospective data and sample sizes were small, the use of inferential statistics was

of limited value. Descriptive statistics were provided on baseline characteristics of each treatment group and on compliance, abstinence outcomes, and rearrest rates. This latter variable, baseline criminal history data, was only available from two of the three sites (Southgate and St. Louis). The relative risk reduction achieved by XR-NTX treatment was calculated by dividing the absolute risk reduction (difference in experimental vs. control event rates) by the event rate in the control group.

3. Results

A pilot sample of 32 subjects who had been treated with XR-NTX between June 2008 and December 2009 was matched with 32 control subjects. The treatment subjects were all the subjects that had been placed on XR-NTX in that period less those who were assessed for opiods only (two individuals). The control group was developed by matching the pool of drug court clients not on XR-NTX to the treatment group based on baseline diagnosis, demographic, and prior criminal history variables.

At baseline, the XR-NTX and control groups showed no significant differences on the key demographic variables of age (mean year of birth: 1976 vs. 1976), gender (24% vs. 21% female), and race (40.0% vs. 42.5% non-White, respectively). The mean number of prior convictions (available from two of the three sites, Southgate and St. Louis), however, was somewhat higher in the XR-NTX group (3.20 vs. 2.44; p = ns). Data on other variables such as marital status or em- ployment history were not recorded with consistency in the drug court database.

In addition to case matching on the baseline diagnosis and demographic and prior criminality variables, as a measure of control for bias during the study, court procedures were analyzed. In terms of the ratio of sanctions to rewards plus sanctions, the courts were found to have treated the two groups similarly: XR-NTX = 0.5 and standard care = 0.47 (p = ns).

Results with the component outcomes are shown in Fig. 1A, B, and C. The number of missed court sessions was lower for the XR-NTX group compared with the control group. The number of missed court sessions was used as a behavioral measure of compliance with the demands and expectations of the drug court. The mean number of missed court sessions per month was small for both groups, aver- aging less than 1%, but the missed-session rate was more than twice as frequent among control subjects than those treated with XR-NTX (Fig. 1A). This represents a relative risk reduction of 57% for XR-NTX treatment.

Abstinence is a critical objective of treatment courts; therefore, the component measure of drinking and/or drug use behavior was analyzed in two ways, both of which were objective (rather than via self-report). Episodic drinking or drug use was examined via the mean ratio of positive alcohol or drug tests per month. Persistent drinking or drug use was

Fig. 1. Comparative effect of XR-NTX and control treatment on compliance and efficacy outcomes.

291M.W. Finigan et al. / Journal of Substance Abuse Treatment 41 (2011) 288–293

calculated as the proportion of subjects whose alcohol or drug tests were positive greater than 25% of the time. The mean ratio of positive alcohol or drug tests per month was higher among control subjects than those treated with XR- NTX (Fig. 1B). This represents a relative risk reduction of 35% for XR-NTX treatment in the proportion of positive tests. Furthermore, control cases had a greater proportion of subjects in the high-risk group with greater than 25% positive tests than subjects treated with XR-NTX (Fig. 1C). Thus, XR-NTX treatment was associated with a 33% relative risk reduction for producing urines indicating poor consis- tency with abstinence. No evidence could be gathered about attendance at such interventions as the 12-step program, nor was there evidence in the record of naloxone challenges being performed.

On the principal outcome of interest, the mean annualized rearrest rate, control subjects were more than three times more likely to be rearrested within a year than subjects who also received treatment with XR-NTX. This represents a relative risk reduction of 69% in the annual risk of having a new arrest while in drug court.

4. Discussion

The results of this pilot study indicate that XR-NTX is a feasible and potentially effective treatment for offenders with alcohol dependence in the drug court setting. XR-NTX

was associated with increased compliance with regular court-mandated appearances and increased rates of absti- nence. Most importantly, the rearrest rate was 69% lower, a significant reduction, among subjects treated with XR-NTX compared with standard care.

These results are consistent with a case series reporting open-label XR-NTX treatment of 10 DUI court-sentenced repeat offenders with alcohol dependence that reported significant within-subject decreases from pretreatment in average drinks per day (p b .01), decreased drinks per drinking day (p = .04), and increased abstinent days (p = .02), with biomarker measures that were consistent with reduced drinking. Interestingly, the vehicles of these repeat DUI offenders had breath alcohol interlock devices installed, and XR-NTX was associated with a reduction in failures to start due to elevated breath alcohol from 3.1% to 1.29% of tests (p = ns; Lapham & McMillan, 2010).

The patients selected for treatment in this drug court study were at very high risk for relapse despite all the external supports and structures of the treatment-as-usual DUI/drug court system. XR-NTX has also demonstrated efficacy in other high-risk situations. For example, in a post hoc analysis of a double-blind, placebo-controlled trial, treatment with XR-NTX, in combination with psychosocial intervention, resulted in a reduction to zero in protocol-defined drinking days during high drinking risk holidays (e.g., New Year's eve, Labor day, Fourth of July, Super Bowl Sunday; Lapham, Forman, Alexander, Illeperuma, & Bohn, 2009). Rosenbloom

292 M.W. Finigan et al. / Journal of Substance Abuse Treatment 41 (2011) 288–293

(2009), in a commentary on this finding, noted the connec- tion between holidays and high risk for relapse, stating “The Lapham study also suggests an entirely new way to help society deal with the special dangers of alcohol-impaired driving so prevalent at holiday times.”

The findings of this study are especially promising given that the drug court setting itself, using a combination of psychosocial intervention with drug and alcohol monitoring, has been reported to reduce recidivism rates by at least 50% compared with what is observed in individuals who receive typical criminal justice penalties (Roman et al., 2003; Gottfredson et al., 2003; Galloway & Drapela, 2006; Ronan et al., 2009).

There are important limitations that are inherent in the pilot nature of these data. The results are based on a retro- spective records analysis of naturalistic data, without the benefit of prospective, random assignment, and the standard care control group was matched post hoc on a limited range of variables. Despite matching, the mean rate of prior criminal convictions was higher in the XR-NTX group compared with the control group (3.20 vs. 2.44). This suggests that the XR-NTX subjects might be a more difficult group of offenders with a higher recidivism rate. Higher rates of past criminal behavior are a known predictor of future criminality (AviBhati, Urban, Institute, July 2006). This might argue that in a larger more controlled study, greater differences in recidivism may be found and that these pilot findings may be conservative. No apparent bias was evident, however, in how the drug courts managed the two offender groups. There was a different duration of community exposure between the two groups, and a follow-up to this pilot study would need to take this variable into account. Finally, the sample size in this pilot was small, and the component outcome findings did not reach significance. Nevertheless, component behaviors of compliance, episodic drinking, and persistent drinking collectively contribute to the “bottom line” end point of rearrest, lending this variable greater statistical power. Given that the result with this variable was significant, the small sample numbers appear to have been sufficient to yield a meaningful signal of potential treatment impact with XR-NTX.

The overall pharmacoeconomic benefit of XR-NTX treatment is an important aspect of feasibility that remains to be addressed. Using data from a previous Michigan DUI court study not involving XR-NTX and from other costs studies of drug court (Carey, Finigan, Crumpton, & Waller, 2006; Carey, Waller, & Marchand, 2006), the annual reduc- tion in rearrest rate found in this study suggests that treat- ment with XR-NTX may be associated with a cost offset advantage in the range of $4,000–$12,000 per person over the 2 years following the initial arrest. (If a formal cost analysis on a larger controlled sample confirms a cost offset advantage, this would potentially have policy implications for drug courts nationwide.)

In spite of robust evidence of efficacy and preliminary evidence for a cost offset advantage, use of medication in the

treatment of alcohol dependence and abuse remains quite low (Mark et al., 2003; Mark, Kassed, Vandivort-Warren, Levit, & Kranzler, 2009; Harris, Kivlahan, Bowe, & Humphreys, 2010). The small sample size in the current project, in spite of training and policy explicitly supporting use of these medications, reflects the challenge that adoption of a medication treatment approach continues to face. In Missouri, for example, all certified substance abuse treat- ment programs that receive state and federal funds (including Medicaid) have been encouraged and lately required to include medication-assisted treatment in the array of services available to people for whom it is clinically appropriate (ADAW, 2009). Continued education and training of substance abuse treatment providers and other health care professionals are vitally important (Glantz et al., 2009).

In conclusion, alcohol dependence is a chronic disease characterized by high risk for relapse, with staggering impli- cations for public safety and criminal justice in this country (Greenfield, 1998). Despite the advance that these nationally recognized and replicated models of progressive jurispru- dence represents, the system continues to struggle with the problem of high recidivism among offenders with severe alcohol dependence (Nolan, 2001; Roman et al., 2003; Gottfredson et al., 2003; Galloway & Drapela, 2006; Ronan et al., 2009). The average offender in this study was facing his or her fourth conviction. Despite extensive monitoring, treatment, and sanction measures meted out by the treatment court model, one in every four offenders in standard treat- ment was rearrested within a year. These pilot findings, viewed in the context of prior reports (Lapham & McMillan, 2010) and prospective random controlled trial results with XR-NTX (Garbutt et al., 2005; O'Malley, Garbutt, Gastfriend, Dong, Kranzler, 2007; Lapham et al., 2009), suggest that XR-NTX may be a promising and effective tool to use in the drug court setting for the management of alcohol- dependent offenders who are at high risk for recidivism.

Acknowledgments

Funding for the purchase of Vivitrol for clients of Missouri Drug and DWI Courts was provided by the Missouri Department of Mental Health's Division of Alcohol and Drug Abuse.

The Medisorb preparation used in XR-NTX was developed with support from National Institute on Drug Abuse Grant R43DA013531 and National Institute on Alcohol Abuse and Alcoholism Grant N43AA001002.

This study was funded by Alkermes, Inc., under a contract with NPC Research. Research design, data collection, data analyses, and report writing were performed primarily by NPC Research. Dr. Finigan, Dr. Perkins, and Dr. Zold-Kilbourn were paid consultants for Alkermes, Inc., in connection with this study. Dr. Edward Schweizer of Paladin Consulting Group, a paid consultant to Alkermes, Inc., provided editorial assistance on an early draft of this article.

293M.W. Finigan et al. / Journal of Substance Abuse Treatment 41 (2011) 288–293

The study was presented in part at the annual meeting of the American Psychiatric Association in New Orleans, May 22–26, 2010.

References

ADAW. (2009). Missouri requires providers to offer medication-assisted treatment. Alcoholism and Drug Abuse Weekly, 21, 3.

Bhati, Avi. (2006). Studying the effects of incarceration on offending trajectories. Urban Institute Policy Center. Accessed online (2/10/2011) at: http://www.urban.org/publications/411427.html.

Carey, S. M., Finigan, M. W., Crumpton, D., & Waller, M. S. (2006). California drug courts: Outcomes, costs and promising practices: An overview of phase II in a statewide study. Journal of Psychoactive Drugs, 38(4), 345–356.

Carey, S. M., Waller, M. S., & Marchand, G. (2006a). Kalamazoo County Adult Drug Treatment Court Outcome and Cost Evaluation: Final Report. Portland, OR: NPC Research.

Carpenter, C., & Dobkin, C. (2010). Drinking age, alcohol consumption, and crime (working paper). AccessedAugust 31, 2010 at: http://people. ucsc.edu/~cdobkin/.

Garbutt, J. C., Kranzler, H. R., O'Malley, S. S., Gastfriend, D. R., Pettinati, H. M., Silverman, B. L., et al. (2005). Efficacy and tolerability of long- acting injectable naltrexone for alcohol dependence: A randomized controlled trial. Journal of the American Medical Association, 293, 1617–1625.

Glantz, M. D., Anthony, J. C., Berglund, P. A., Degenhardt, L., Dierker, L., Kalaydjian, A., et al. (2009). Mental disorders as risk factors for later substance dependence: Estimates of optimal prevention and treatment benefits. Psychological Medicine, 39, 1365–1377.

Gottfredson, D., Najaka, S., & Kearley, B. (2003). Effectiveness of drug treatment courts: Evidence from a randomized trial. Criminology and Public Policy, 2, 171–196.

Greenfield, L. A., & Henneberg, M. (2001). Victim and offender self-reports of alcohol involvement in crime. Alcohol Research and Health, 25, 20–31.

Greenfield, A. (1998). U.S Department of Justice, Office of Justice Programs; Bureau of Justice Statistics. An analysis of national data on the prevalence of alcohol involvement in crime, Washington D.C. Accessed July 19, 2010 at: http://bjs.ojp.usdoj.gov/index.cfm?ty=pbdetail&iid=385.

Harris, A. H., Kivlahan, D. R., Bowe, T., & Humphreys, K. N. (2010). Pharmacotherapy of alcohol use disorders in the Veterans Health Administration. Psychiatric Services, 61, 392–398.

Kessler, R. C., & Wang, P. S. (2008). The descriptive epidemiology of commonly occurring mental disorders in the United States. Annual Review of Public Health, 29, 115–129.

Lapham, S., Forman, R., Alexander, M., Illeperuma, A., & Bohn, M. J. (2009). The effects of extended-release naltrexone on holiday drinking in alcohol-dependent patients. Journal of Substance Abuse Treatment, 36, 1–6.

Lapham, S. C., & McMillan, G. P. (2010). Open-label pilot study of extended-release naltrexone to reduce drinking and driving among repeat offenders. Journal of Addiction Medicine, Epub ahead of print, doi:10.1097/ADM.0b013e3181eb3b89.

Marchand, G., Waller, M. S., & Carey, S. M. (2006). Barry County adult drug court outcome and cost evaluation: Final report. Portland, OR: NPC Research.

Mark, T. L., Kassed, C. A., Vandivort-Warren, R., Levit, K. R., & Kranzler, H. R. (2009). Alcohol and opioid dependence medications: Prescrip- tion trends, overall and by physician specialty. Drug and Alcohol Dependence, 99, 345–349.

Mark, T. L., Kranzler, H. R., Poole, V. H., Hagen, C. A., McLeod, C., & Crosse, S. (2003). Barriers to the use of medications to treat alcoholism. American Journal of Addiction, 12, 281–294.

Mee-Lee, D., Shulman, G. D., Fishman, M., Gastfriend, D. R., & Griffith, J.H. (2001). ASAM patient placement criteria for the treatment of substance-related disorders, (2nd ed.-Revised (ASAM PPC-2R)). Chevy Chase, MD: American Society of Addiction Medicine, Inc.

National Highway Traffic Safety Administration. (2009). Traffic safety facts 2008. Washington, DC: National Highway Traffic Safety Administration.

Nochajski, T. H., & Stasiewicz, P. R. (2006). Relapse to driving under the influence (DUI): A review. Clin Psych Rev, 26, 179–195.

Nolan, J. L. (2001). Reinventing justice: The American drug court movement. Princeton, NJ: Princeton University Press.

O'Malley, S., Garbutt, J. C., Gastfriend, D. R., Dong, Q., & Kranzler, H. R. (2007). Efficacy of extended-release naltrexone (XR-NTX) in alcohol- dependent patients who are abstinent before treatment. Journal of Clinical Psychopharmacology, 27, 507–512.

Roman, J., Townsend, W., & Bhati, A. S. (2003). Recidivism rates for drug court graduates: Nationally based estimates (NCJ 201229). Washing- ton, DC: National Institute of Justice.

Rosenbloom, D. L. (2009). Holidays, triggers, and willpower—Is there a role for medications? A commentary on “The effects of extended-release naltrexone on holiday drinking in alcohol-dependent patients”. Journal of Substance Abuse Treatment, 36, 7.