Abstract
The purpose of this article is to review current empirical research on the effectiveness of drug treatment programs, particularly those for prisoners, parolees, and probationers. The authors reviewed empirical research published after the year 2000 that they classified as Level 3 or higher on the Maryland Scale. Participants in cognitive-behavioral therapy (CBT), therapeutic communities, and drug courts had lower rates of drug use and crime than comparable individuals who did not receive treatment. Several different types of pharmacological treatments were associated with a reduced frequency of drug use. Those who received contingency management tended to use drugs less frequently, particularly if they also received cognitive-behavioral therapy. Finally, researchers reported that drug use and crime were lower among individuals whose treatment was followed by an aftercare program. Effective treatment programs tend to (a) focus on high-risk offenders, (b) provide strong inducements to receive treatment, (c) include several different types of interventions simultaneously, (d) provide intensive treatment, and (e) include an aftercare component.
One of the major social problems in the United States is the prevalence of substance abuse. Eight percent of Americans aged 12 and older used an illicit drug during the past month—9% of youths aged 12 to 17 and 20% of those aged 18 to 25 (Substance Abuse and Mental Health Services Administration, 2009). Among prison inmates in the United States, 73% used drugs regularly prior to their incarceration (Petersilia, 2005). At the time inmates committed their latest offense, 50% were under the influence of alcohol or drugs (Karberg & James, 2005). From 1975 to 2000, there was a 400% increase in the U.S. incarceration rate, and this was due primarily to a rapid growth in incarceration for drug offenses (Blumstein & Beck, 2005).
About 95% of all inmates were released to reintegrate into communities (Petersilia, 2005). During 2008, more than 713,000 inmates were released from prison or an average of almost 2,000 per day (Sabol & Couture, 2008). This is more than 4 times the number released 25 years ago. Large numbers of those released from prison were rearrested and returned to prison, often because of their inability to refrain from substance abuse (Blumstein & Beck, 2005).
In the past decade, there have been a number of assessments of “what works” in existing crime prevention, correctional, and reentry programs (MacKenzie, 2000; Seiter & Kadela, 2003; Sherman, Farrington, Welsh, & MacKenzie, 2002; Wormwith et al., 2007). However, there has not been a recent review of the effectiveness of drug treatment programs for offenders. The purpose of this article is to review and synthesize the current empirical research on the effectiveness of drug treatment programs, particularly those used to treat offenders.
Selection Bias
It has been observed in numerous research studies that those who receive drug treatment tend to have lower rates of drug use than individuals who do not receive treatment. However, in many of the studies, selection bias may account for the differences. For example, researchers often compare those who completed a drug treatment program with a control group. It is difficult to know whether the lower drug use of the treatment group is due to the treatment or to pretreatment characteristics such as motivation, previous criminal history, or the extent of drug dependence. Researchers have confirmed that those who drop out of treatment programs tend to have more serious criminal histories and fewer ties to society than those who complete programs (Huebner & Cobbina, 2007). In this review, we pay particular attention to how researchers controlled for selection bias.
Method
In this article we review peer-reviewed journal empirical research published since 2000 that we classified as Level 3 or higher on the Maryland Scale (Sherman et al., 2002). We searched social and behavioral science databases using the terms “drug abuse,” “drug use,” “substance use,” or “drug addiction” paired with “treatment,” “treatment outcomes,” or “treatment effectiveness.” In addition, we examined the bibliographies of the articles we identified. Finally, for the years 2005 to 2009, we examined the table of contents of journals that published articles on drug use, abuse, dependence, and treatment.
Effectiveness of Drug Treatment Programs
The consensus from previous evaluations is that drug treatment programs can be modestly effective for some people (Adrian, 2001; Dutra et al., 2008; Hepburn, 2005; Hubbard, Simpson, & Woody, 2009). We turn now to a review of the research on several commonly used types of treatment programs administered to criminal offenders.
Cognitive-Behavioral Therapy
Cognitive-behavioral therapy (CBT) assumes that clients have maladaptive thinking patterns that need to be changed. CBT focuses on restructuring attitudes and thoughts and on developing interpersonal skills (Milkman & Wanberg, 2007). Clients are actively involved in activities and usually have homework assignments to reinforce and develop skills (Lowenkamp, Hubbard, Makarios, & Latessa, 2009).
We located seven empirical evaluations of CBT drug programs since the year 2000 that were at least Level 3 on the Maryland Scale. Two of the studies were evaluations of intensive prison residential programs that lasted at least 6 months. The first examined 760 persons from 20 prisons who received CBT. During 6 months following release, those who received CBT were less likely to use drugs or be rearrested than comparison group participants (Pelissier et al., 2001).
The second program was the Forever Free program, a CBT substance abuse treatment program for women prisoners. One year after release, the 101 women who received treatment had significantly less drug use and fewer arrests than the 79 women in the comparison group (Hall, Prendergast, Wellisch, Patten, & Cao, 2004).
We located five studies that evaluated CBT as a short-term treatment method for substance dependence. These were programs that lasted from 8 to 16 weeks, with tracking from 6 months to a year following treatment completion. All reported that CBT programs were effective in helping participants reduce drug use during follow-up periods of 6 months to a year (Budney, Moore, Rocha, & Higgins, 2006; Carroll et al., 2006; Easton et al., 2007; Kadden, Litt, Kebela-Cormier, & Petry, 2007; Rawson et al., 2006). In addition, two recent meta-analyses confirmed that CBT is an effective treatment for substance dependence (Dutra et al., 2008; Magill & Ray, 2009).
Contingency Management
A commonly used treatment for drug dependence is contingency management (CM) in which various types of rewards are given to reinforce positive behavior. A reward is given if a client receives a negative urine sample, attends treatment, or fulfills other obligations. A reward can be almost anything, but common rewards are vouchers, prizes, and medication.
We identified 12 recent studies at Level 3 or higher on the Maryland Scale that evaluated the effectiveness of drug treatments using CM. All 12 reported that clients who received vouchers or other rewards were more likely to be abstinent than clients who did not receive vouchers. CM was effective in treating a variety of drugs, including marijuana (Budney et al., 2006; Carroll et al., 2006; Kadden et al., 2007), methamphetamines (Rawson et al., 2006; roll et al., 2006), cocaine (Budney et al., 2006; Epstein et al., 2009; Groβ, Marsch, Badger, & Bickel, 2006; Higgins et al., 2006; Olmstead & Petry, 2009; Petry & Martin, 2002; Petry et al., 2005; Prendergast, Podus, Finney, Greenwell, & Roll, 2006), and opiates (Epstein et al., 2009; Groβ et al., 2006; Higgins et al., 2006; Olmstead & Petry, 2009; Petry & Martin, 2002; Petry et al., 2005; Prendergast et al., 2006). However, a limitation of CM is that when the rewards are discontinued the rates of abstinence tend to decrease (Hall, Prendergast, Roll, & Wards, 2009; Prendergast et al., 2006).
When researchers compared the effectiveness of CBT and CM, the two methods were similar in overall effectiveness (Rawson et al., 2006). Together CM and CBT were more effective than either was alone (Budney et al., 2006; Carroll et al., 2006; Dutra et al., 2008; Kadden et al., 2007).
Therapeutic Communities
A therapeutic community is a highly structured residence program where clients are organized into groups and leaders are chosen from within the group. The purpose is to give governance and accountability to the clients themselves. Because of the responsibility of the clients for each other, peer pressure within the group helps constrain individuals and encourage compliance with rules (De Leon, 2000). Individuals in therapeutic communities receive a variety of treatment modalities, including cognitive therapy, individual counseling, group counseling, and 12-step programs.
The therapeutic community is a widely used treatment modality both within and outside correctional facilities. In reviews of research, Sherman et al. (2002) and Seiter and Kadela (2003) identified three major research projects that evaluated therapeutic community programs in prison: (1) Wexler and his colleagues evaluated therapeutic communities in California (Wexler, De Leon, Thomas, Kressell, & Peters, 1999; Wexler, Falkin, & Lipton, 1990; Wexler, Melnick, Lowe, & Peters, 1999); (2) Inciardi, Martin, and their colleagues evaluated the Key-Crest therapeutic community in Delaware (Inciardi, Martin, Butzin, Hooper, & Harrison, 1997; Martin, Butzin, & Inciardi, 1995; Martin, Butzin, Saum, & Inciardi, 1999); (3) Knight and colleagues assessed a therapeutic community in Texas (Hiller, Knight, & Simpson, 1999; Knight, Simpson, & Hiller, 1999) In all three projects, individuals who were involved in prison therapeutic communities were more likely to remain drug and arrest free than comparable individuals who were not treated in a therapeutic community.
From 2000 to 2009, we discovered eight additional studies on therapeutic communities that met our criteria. Those studies replicated previous findings and extended our knowledge in several ways.
First, 5 years after completion, those who participated in the Delaware program were less likely to have used drugs or have a new arrest. Participants who dropped out of the program were more likely to remain drug free than a comparison group. Those who participated in aftercare had less recidivism and higher levels of employment than individuals who did not receive aftercare (Inciardi, Martin, & Butzin, 2004).
Second, the 5-year evaluation of the California therapeutic community produced similar results, although the treatment effect diminished for those who did not receive aftercare (Prendergast, Hall, Wexler, Melnick, & Cao, 2004). Among high-risk individuals, those who participated in the therapeutic community had lower recidivism than individuals who were not treated. However, among low-risk individuals there was no difference between the treated and untreated individuals (Wexler, Melnick, & Cao, 2004).
Third, two new evaluations of therapeutic communities were completed. Two years after release, those who participated in five prison therapeutic communities in Pennsylvania had lower rearrest and reincarceration rates than a no-treatment group (Welsh, 2007). A study of 1,193 federal prisoners revealed that prisoners in therapeutic communities had lower rates of drug relapse and recidivism than two untreated groups (Rhodes et al., 2001). The cumulative evidence is that therapeutic communities can be effective in reducing the risk of drug relapse and rearrest, particularly among high-risk individuals and when followed by aftercare programs.
Drug Courts
A drug court is a specialized program that is designed to use the power of the court to encourage individuals to receive treatment and decrease drug use. Specific procedures vary across drug courts although most combine drug treatment with judicial monitoring, drug testing, and intensive supervision (Carey & Finigan, 2006; Giacomazzi & Bell, 2007; Sanford & Arrigo, 2005; Wiseman, 2005).
Typically, individuals who participate in a drug court are required to plead guilty to their charges, but the charges are held in abeyance as long as they comply with the requirements of the court. All participants are required to take a urinalysis test regularly and attend all components of their treatment plans. If they comply with treatment requirements and successfully graduate from drug court, their charges are dropped and expunged from their record. If they do not, the judge may sentence them to jail or prison since they previously pled guilty (Cooper, 2003; Jensen & Mosher, 2005-2006; Turner et al., 2002).
A review of 27 drug court evaluations conducted from 1993 to 2002 revealed that drug courts can be effective in helping offenders reduce their criminal activity (MacKenzie, 2006). We identified 14 evaluations of drug courts from 2000 to 2009 that we judged to be at least Level 3 on the Maryland Scale. Ten were new studies that were not available when MacKenzie conducted her review.
In a series of evaluations of the Baltimore City Drug Treatment Court, participants tended to have less drug use and crime than nonparticipants (Banks & Gottfredson, 2003, 2004; Gottfredson & Exum, 2002; Gottfredson, Kearly, Najaka, & Rocha, 2005; Gottfredson, Najaka, & Kearly, 2003). Particularly important was the fact that they randomized participants into treatment and control groups and followed them for 3 years (Gottfredson et al., 2005). Evaluations of drug courts in eight other states (Florida, Idaho, Missouri, Nebraska, Nevada, Ohio, Oregon, and Pennsylvania) and Australia provided additional evidence of the effectiveness of drug courts (Brewster, 2001; Goldkamp, White, & Robinson, 2001; Listwan, Sundt, Holsinger, & Latessa, 2003; Shaffer, Hartman, & Listwan, 2009; Spohn, Piper, Martin, & Frenzel, 2001; Truitt et al., 2003; Weatherburn, Jones, Snowball, & Hua, 2008). However, in two studies in Las Vegas and California there were no differences between those who did and did not participate in a drug court (Miethe, Lu, & Reese, 2000; Wolfe, Guydish, & Termondt, 2002).
Overall, the evidence indicates that if implemented well, drug courts can be effective in helping offenders reduce drug use and crime. The success of drug courts appears to be due to the combination of judicial oversight, intense supervision, drug testing, and rehabilitative services (Fischer, Geiger, & Hughes, 2007; Giacomazzi & Bell, 2007; Jensen & Mosher, 2005-2006).
Pharmacological Treatment
During the past two decades, there have been significant developments in the pharmacological treatment of drug and alcohol abuse. First, a drug has been used as a substitute for a more harmful drug. These are classified as agonists because they induce a full or partial pharmacological response. An example of this type of treatment is methadone, which is used to treat heroin addiction. Second, medications are used to counteract the effects of another drug. These are termed antagonists and may reduce cravings (O’Brien, 1997). In this section, we review evidence relevant to the effectiveness of several pharmacological treatments for drug dependence.
Topiramate
Although originally used in the treatment of epileptic seizures, topiramate has been evaluated as a possible treatment of alcohol dependence because it tends to decrease the release of dopamine in the midbrain after alcohol intake (Johnson et al., 2007). In a comparison of topiramate and a placebo, those on topiramate had fewer drinks per day, fewer heavy drinking days, and more days abstinent (Baltieri, Daro, Ribeiro, & de Andrade, 2008; Johnson et al., 2003, 2007). Those taking topiramate also showed an increase in safe drinking periods compared to those on a placebo (Ma, Ait-Daoud, & Johnson, 2006).
Buprenorphine
In January of 2003, buprenorphine became available for the treatment of opiate dependence (Colameco, Armando, & Trotz, 2005). In a 28-day outpatient treatment with either buprenorphine or clonidine, those receiving buprenorphine had significantly higher retention in treatment and higher percentages of opiate abstinence (Marsch et al., 2005). In a comparison of 126 heroin dependent patients receiving detoxification and drug counseling, those treated with buprenorphine had a longer time to first heroin use and more abstinent days than those using a placebo (Schottenfeld, Chawarski, & Mazlan, 2008). Another study examined the impact of psychotherapy and buprenorphine treatment on outpatients dependent on both cocaine and heroin. Those who received both psychotherapy and buprenorphine had lower rates of drug use (Montoya et al., 2005).
Groβ et al. evaluated three randomly assigned treatment groups. The first group received standard treatment including buprenorphine, regardless of their urinalysis results. The second group not only received standard treatment and buprenorphine but also earned vouchers for each negative urine sample on an escalating schedule. The third group received half of their prescribed buprenorphine dose if they attended the clinic and the other half if their urinalysis was negative. Those in the medication contingent (third) group had more weeks of continuous abstinence from opiates and cocaine than those in the voucher group (Groβ et al., 2006).
Naltrexone
Naltrexone is an opioid antagonist that has been used to treat individuals dependent on opioids or cocaine. The research on naltrexone has been mixed. A study of naltrexone and alcohol dependence revealed that naltrexone-treated participants tended to have lower relapse rates, consume fewer drinks, and have slower progression to drinking (Anton, Drobes, Voronin, Durazo-Avizu, & Moak, 2004; Guardia et al., 2002; Morley et al., 2006). However, other research indicates that naltrexone was not as effective as topiramate or buprenorphine in helping individuals reduce their dependence (Baltieri et al., 2008; Schottenfeld et al., 2008).
Methadone
Methadone maintenance is a specific type of pharmacological treatment in which heroin dependence is replaced with methadone dependence. The objective is to control the dosage and enable the addicts to live relatively normal lives. Although there have been a number of studies of methadone maintenance, we located only two which were Level 3 or higher on the Maryland Scale. The first randomly assigned 197 prison inmates to one of three conditions: (1) counseling only, (2) counseling plus methadone maintenance at release, and (3) counseling and methadone maintenance in prison. A follow-up 90 days after release revealed that the counseling + methadone group had more frequent attendance at drug treatment, less heroin use, and less reincarceration than the counseling-only group (Kinlock, Gordon, Schwartz, & O’Grady, 2008). The second study tested a combination of methadone maintenance and CM on heroin and cocaine use (Epstein et al., 2009). Using two different levels of methadone administration, they found that the higher dose increased heroin abstinence but not cocaine abstinence. CM tended to reduce the use of both heroin and cocaine and a higher-value incentive and a higher dose of methadone resulted in greater abstinence (Epstein et al., 2009).
In summary, the evidence indicates that drugs can be effective supplements in treating alcohol and drug dependence, particularly topiramate and buprenorphine. Pharmacological treatments appear to be particularly useful when paired with therapy or CM.
Boot Camps
Boot camps were designed as a military-type regimen to rehabilitate offenders through strict discipline and swift punishment for rule infractions. They emphasize vigorous physical activity, manual labor, and discipline. The goal is to help offenders learn respect for authority and develop skills that will enable them to desist from crime.
Research has been consistent in showing that boot camps do not reduce recidivism among juvenile or adult offenders (Wilson & MacKenzie, 2005). We identified 7 studies from 2000 to 2009 that evaluated the effectiveness of boot camps. Similar to previous research, no differences were found between boot camp participants and control groups (Bottcher & Ezell, 2005; Stinchcomb & Terry, 2001). However, in one study boot camp participants with a previous record had lower recidivism than controls (Kempinen & Kurlychek, 2003). In three other studies, boot camp arrestees were less likely to be convicted (Steiner & Giacomazzi, 2007), less likely to be recommitted (Wells, Minor, Angel, & Stearman, 2006), or if reincarcerated, spent less time in prison (Duwe & Kerschner, 2008). These findings suggest that arrestees who have been to boot camps may commit less serious offenses than arrestees who have not been to boot camps. Finally, boot camp participants who received a 90-day aftercare were significantly less likely to be rearrested 2 years after completion—22% for the aftercare group compared to 33% among those who did not receive aftercare (Kurlychek & Kempinen, 2006). In summary, the evidence indicates that boot camps might reduce recidivism among high-risk participants or if the boot camp is followed by aftercare.
Twelve-Step Programs
Twelve-step programs are among the oldest and most well-known drug and alcohol treatment programs. The first 12-step program began in the mid-1930s, and the programs have grown to become one of the most widely used drug treatment approaches (Alcholics Anonymous World Services, 1957, 1976). It is estimated that 3% of Americans will attend some type of 12-step program during their lives (Fiorentine, 1999). Examples of 12-step programs are Alcoholics Anonymous (AA), Narcotics Anonymous (NA), and Cocaine Anonymous (CA). In addition, 12-step principles have been used as part of many other treatment programs.
Twelve-step programs assume that substance dependence is a life-long disease that can be managed but not cured. The program is based on 12 steps of recovery that individuals strive to attain while regularly reporting their progress and struggles. Key components of 12-step programs are (a) recognizing that you will always be an addict, (b) weekly or biweekly meetings with a nonprofessional support group of individuals with similar problems, (c) recognizing and relying on a higher power, (d) performing service to others, and (e) group and individual counseling sessions (Foundation, 1987; Schneider, 2006).
Although 12-step programs are widely used, there has been relatively little systematic research evaluating their effectiveness. Selection bias is a major limitation, as only about half of AA participants make it past 3 months (Fiorentine, 1999). Similar to other substance abuse programs, dropouts tend to have higher rates of relapse and criminal recidivism (Schneider, 2006).
We discovered only five recent studies that met Level 3 on the Maryland Scale. Fiorentine (1999) found that those who participated in a 12-step program were more likely to be abstinent from both alcohol and other drugs. On the other hand, Zanis et al. observed that 12-step participation did not reduce recidivism. Parolees treated in a 12-step program were not less likely than untreated parolees to be reconvicted of a new offense within 24 months of release (Zanis et al., 2003). In a comparison of (a) 12-step + CM, (b) CBT, or (c) CBT + CM, 12-step participants had higher levels of marijuana use than those receiving the other treatments (Carroll et al., 2006). In a study of AA as a supplement to alcohol couples in behavioral therapy, there was no additional benefit to combining AA attendance with behavioral therapy (McCrady, Epstein, & Hirsch, 1999). Finally, a comparison of CBT, motivational enhancement therapy, and Twelve Step Facilitation (TSF) on drinks per day and the number of days abstinent over 3 years revealed no difference among the three treatments (Babor & Del Boca, 2003). Overall, the evidence suggests that 12-step programs are not as effective as other treatments in reducing drug use and recidivism.
Summary, Discussion, and Conclusion
In summary, there is evidence that drug courts, therapeutic communities, cognitive-behavioral treatment, CM, and pharmacological treatment can be effective in helping individuals decrease their drug use and desist from criminal activity. Effective treatment programs tend to (a) focus on high-risk offenders, (b) provide strong inducements to receive treatment, (c) include several different types of interventions simultaneously, (d) provide intensive treatment, and (e) include an aftercare component.
Research demonstrated that aftercare increased the impact of both therapeutic communities and boot camps. The findings on boot camps were particularly surprising given the fact that previous research had found that boot camps were not effective.
Another important issue is whether or not mandated treatment is effective. The clients of drug courts often are coerced in that they must accept treatment in the drug court or face prison time. Since retention is important for drug treatment success, strong inducements to continue treatment would appear to be important. Recent research confirms that legally mandating treatment tends to lower dropout rates and reduce illicit drug use and criminal offending (Kelly, Finney, & Moos, 2005; McSweeney, Stevens, Hunt, & Turnbill, 2007; Perron & Bright, 2008; Young & Belenko, 2002).
A final issue is whether individuals can desist from drug abuse without treatment. Although spontaneous remission is not uncommon, the evidence indicates that treatment can be an effective tool in helping many individuals reduce their drug use (Price, Risk, & Spitznagel, 2001).
Several existing theories of change are useful in interpreting and applying these results. Social learning and social control theories are consistent with much of the research we reviewed (Agnew, 2005; Gottfredson & Hirschi, 1990; Hirschi, 1969). The findings illustrate that learning and reinforcement are useful change tools along with attempts to develop appropriate bonds to individuals and groups that do not abuse drugs.
Another useful theoretical perspective is to view drug addiction as a chronic brain disease. From this perspective drug abuse may produce changes in the structure and functioning of the brain that are long lasting and difficult to modify. These changes may decrease the ability of people to control their drug use (Leshner, 1997; Powledge, 1999).
Viewing drug abuse as a brain disease has a number of implications for treatment. First, if individuals have a disease that they no longer control, then they need treatment rather than criminal sanctions. Second, the focus may need to shift from curing the disease to managing it. Chronic diseases are not cured after initial treatment but require long-term management and ongoing treatment. Third, treatments may need to include pharmacological as well as behavioral methods. The pharmacological research demonstrates the usefulness of this perspective and that drug therapies can be an effective supplement to other types of treatment.
After reviewing the empirical research, we recommend the following. First, the use of therapeutic communities should be expanded for prisoners and others in residential settings. Second, the use of drug courts should be expanded for offenders on probation and in the community. Given the costs of incarceration and recidivism, the expense of expanding therapeutic communities and drug courts would be offset by reductions in incarceration, crime, and substance dependence. Within therapeutic communities and drug courts, cognitive behavioral, CM, and pharmacological treatments should be made readily available.
Footnotes
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Department of Sociology and the College of Family, Home and Social Sciences of Brigham Young University.
