An Effective Model of Prison-Based Treatment for Drug-Involved Offenders
Inciardi, James A, Martin, Steven S, Butzin, Clifford A, Hooper, Robert M, Harrison, Lana D, Journal of Drug Issues
A multistage therapeutic community treatment system has been instituted in the Delaware correctional system, and its effectiveness has captured the attention of the National Institutes of Health, the Department of Justice, members of Congress, and the White House. Treatment occurs in a three-stage system, with each phase corresponding to the client's changing correctional status-incarceration, work release, and parole. In this paper, 18 month follow-up data are analyzed for those who received treatment in: (1) a prison-based therapeutic community only, (2) a work release therapeutic community followed by aftercare, and (3) the prison-based therapeutic community followed by the work release therapeutic community and aftercare. These groups are compared with a no-treatment group. Those receiving treatment in the two-stage (work release and aftercare) and three-stage (prison, work release, and aftercare) models had significantly lower rates of drug relapse and criminal recidivism, even when adjusted for other risk factors. The results support the effectiveness of a multistage therapeutic community model for drug-involved offenders, and the importance of a work release transitional therapeutic community as a component of this model.
Drug Use and Crime
The linkages between drug abuse and crime have been well documented. Over the past 2 decades, both the National Institute on Drug Abuse and the National Institute of Justice have funded a number of major studies to generate better understandings of the drugs and crime connection and the linkages between the two phenomena The research has yielded a number of interesting findings. For example, in extensive follow-up studies of addict careers in Baltimore, the researchers found high rates of criminality among heroin users during those periods when they were active users and markedly lower rates during times of nonuse (Ball et al. 1983). Studies conducted in New York City concerning the economics of the drug-crime relationship documented the wide range of heroin usage rates among addicts and the clear correlation between the amount of drugs used and amount of crime committed (Johnson et al. 1985). Miami-based research, furthermore, demonstrated that the amount of crime committed by drug users was far greater than anyone had previously imagined, that drug-related crime could at times be exceedingly violent, and that the criminality of street drug users was far beyond the control of law enforcement (Inciardi 1979, 1992; Inciardi and Pottieger 1986, 1991, 1994). Other research has reported similar conclusions (Speckart and Anglin 1986; Anglin and Hser 1987). Together, the overall findings suggest that, although the use of heroin, cocaine, and other illegal drugs does not necessarily initiate criminal careers, drug use does intensify and perpetuate criminal activity. That is, street drugs seem to lock users into patterns of criminality that are more acute, dynamic, unremitting, and enduring than those of other offenders.
The presence of substance abusers in criminal justice settings has also been well documented. A concomitant of drug-related criminality and the war on drugs of the 1980s and early 1990s has been the increased numbers of drug-involved offenders coming to the attention of the criminal justice system (Inciardi 1993; Inciardi et al. 1996). In fact, it has been reported that perhaps two-thirds of those entering state and federal penitentiaries have histories of substance abuse (Chaiken 1989; Chavaria 1992; Leukefeld and Tims 1992). This suggests that criminal justice settings offer excellent opportunities for assessing the treatment needs of drug-involved offenders, and for providing treatment services in an efficient and clinically sound manner (Reno 1993; Hawk 1993).
Drug Treatment for Offenders
Many clinicians and practitioners have felt that the therapeutic community (TC) is one of the most viable forms of treatment for drug-involved offenders, particularly those whose criminality results in incarceration (Leukefeld and Tims 1988, 1992; Tims et al. 1994). Moreover, the TC is a total treatment environment isolated from the rest of the prison population, away from the drugs, violence, and other aspects of prison life that militate against rehabilitation. The primary clinical staff are typically former substance abusers who themselves were rehabilitated in TCs. The treatment perspective is that drug abuse is a disorder of the whole person; that the problem is the person and not the drug, that addiction is a symptom and not the essence of the disorder, and that the primary goal is to change the negative patterns of behavior, thinking, and feeling that predispose drug use (De Leon 1994).
Research on community-based residential TCs have found them most effective for those who exhibit low levels of social deviance and who remain in treatment the longest (Simpson et al. 1979; McLellan and Alterman 1991; Yablonsky 1989; Condelli and Hubbard 1994). In fact, there is consensus across studies and modalities that the longer a client stays in treatment, the better the outcome in terms of declines in drug use and criminal behavior (De Leon 1984; Anglin and Hser 1990).
For the drug-involved criminal justice client as well, it appears that those who remain in some type of treatment do better than those who either drop-out, are involuntarily discharged, or who do not participate in treatment at all. Recent studies of New York's Stayn Out and Oregon's Cornerstone prison-based TC programs have demonstrated that the longer clients remain in treatment, the less likely they are to be rearrested or return to prison (Wexler et al. 1988, 1990; Field 1992). These studies, however, provide only limited evidence that prison-based TC clients are less likely to subsequently use drugs, get arrested, or return to prison.
Other factors that have been found to be predictors of relapse or recidivism include certain demographic characteristics (age, gender, and ethnicity), life course characteristics (drug use history, delinquency, and criminal history), and treatment characteristics (Wexler et al. 1988; Hall et al. 1990; Gossop et al. 1990). However, many of these studies, which have not multivariately evaluated treatment outcome, may be confounding the effects of individual characteristics with treatment effects.
In short, research on the effectiveness of treatment alternatives for criminal justice clients has lagged behind the implementation of new programs. Most studies conducted on the effectiveness of treatment for drug-involved offenders have focused on the number who successfully complete treatment, and typically only in-prison treatment. Consequently, most assessments of program effectiveness have been more process than outcome oriented and have not incorporated multiple outcome criteria. Outcome research, where it has been attempted, has involved short follow-up time frames and has included only limited use of comparison groups, standardized measurement instruments, multivariate models, and appropriate control variables. Longitudinal outcome studies of recidivism or relapse are uncommon (Forcier 1991; Rouse 1991).
All of these issues suggest that an appropriate evaluation of the effectiveness of treatment for drug-involved offenders should longitudinally examine outcomes with a large enough sample to allow multivariate analyses. The paper reports such an analysis of the application of a TC approach to the treatment of drug-involved offenders.
The Multistage Therapeutic Community Continuum
Based on a wide body of literature in the fields of both treatment and corrections combined with clinical and research experiences with correctional systems and populations (Ball and Ross 1991; Brown 1979; Chaiken 1989; Chavaria 1992; Field 1992; Forcier 1991; Gossop et al. 1990; Hall et al. 1990; Hubbard et al. 1989; Simpson and Sells 1982; Wexler et al. 1988,1990), it would appear that the most effective strategy would involve three stages of TC intervention (Inciardi et al.1994; Martin et al. 1995). Each stage in this continuum is an adaptation to the client's changing correctional status: incarceration , work release -- parole or other form of community supervision. The approach recognizes that "the connection between rehabilitation efforts in prison and the process of integration into society after release is probably one of the most feeble links in the criminal justice system" (Wexler and Williams 1986).
The primary stage of treatment should consist of a prison-based therapeutic community. Segregated from the negativity of the prison culture, recovery from drug abuse and the development of pro-social values in the prison TC would involve essentially the same mechanisms seen in community-based TCs (see De Leon 1994; Martin et al. 1995). Therapy in this stage should be an on-going and evolving process over 12 months, with the potential for the resident to remain slightly longer, if needed. Moreover, it is important that TC treatment for inmates begin while they are still in the institution. In a prison situation, time is one of the few resources that most inmates have in abundance. The competing demands of family, work, and the neighborhood peer group are absent. Thus, there is the time and opportunity for focused and comprehensive treatment, perhaps for the first time in a drug offender's career. In addition, there are other new opportunities presented: to interact with recovering addict role models, to acquire pro-social values and a positive work ethic, and to initiate a process of understanding the addiction cycle.
The secondary stage should be a transitional therapeutic community in a work release setting. Since the 1970s, work release has become a widespread correctional practice for felony offenders. It is a form of partial incarceration whereby inmates who are approaching their release dates are permitted to work for pay in the free community, but must spend their nonworking hours either in the institution or, more commonly, in a community-based work release facility. Although graduated release of this sort carries the potential for easing an inmate's process of community reintegration, there is a negative side, especially for those whose drug involvement served as the key to the penitentiary gate in the first place. Inmates are exposed to groups and behaviors that can easily lead them back to substance abuse, criminal activities, and reincarceration. Because work release populations mirror the institutional populations from which they came, there are still the negative values of the prison culture, but now, in addition, street drugs and street norms abound.
As such, the transitional work release TC should be similar to that of the traditional TC. There should be the family setting removed from as many of the external negative influences of the street and inmate cultures as is possible. However, the clinical regimen in the work release TC must be modified to address the correctional mandate of work release.
In the tertiary stage (aftercare), clients will have completed work release and will be living in the free community under the supervision of parole or some other supervisory program. Treatment intervention in this stage should involve out-patient counseling and group therapy. Clients should be encouraged to return to the work release TC for refresher and reinforcement sessions, to attend weekly groups, to call on their counselors on a regular basis, and to spend 1 day each month at the facility.
This three stage model has been operating in the Delaware correctional system since the early 1990s. It is built around three therapeutic communities: the KEY, a prison-based TC for men; WCI Village, a prison-based TC for women; and CREST Outreach Center, a residential work release center for both men and women (Lockwood and Inciardi 1993; Hooper et al. 1993; Martin et al. 1995). Preliminary data based on a 6-month follow-up of 457 clients suggested the efficacy of the three stage model in reducing relapse and recidivism (Martin et al. 1995). This report presents data on 448 clients who have been reinterviewed at 18 months. These 18month follow-up respondents are classified into four groups:
(a) those who were placed in the conventional work release setting and received neither prison-based nor community-based TC treatment (COMPARISON),
(b) those who received their primary treatment at The KEY but no secondary or tertiary treatment (KEY),
(c) those who received their primary and secondary treatment at CREST followed by aftercare (CREST), and
(d) those who received their primary treatment at The KEY (or WCI) and their secondary treatment at CREST followed by aftercare (KEY-CREST).
The basic hypothesis is that drug-involved offenders receiving primary treatment in a prison-based TC followed by secondary (transitional) treatment in a work release TC followed by aftercare will have more successful outcomes than those who have fewer stages of therapeutic community treatment.
Sample and Measures
The data for the current study originated with two research demonstration projects funded by the National Institute on Drug Abuse and awarded to the University of Delaware, James A. Inciardi, Principal Investigator. Between Summer 1990 and Spring 1994, a total of 1,002 prison inmates eligible for parole or work release were interviewed just prior to leaving prison. Respondents from both these studies are still being followed under a new NIDA continuation grant awarded to the Principal Investigator.
This report examines a subset of these respondents (N=448) who were followed up in the field some 6 and 18 months after their release from the institution. Treatment dropouts were also followed. Lengthy interview data were collected on drug use and sexual activities, criminal history, drug-abuse treatment history, psychosocial and mental health status, and sociodemographics. HIV and drug testing was also done on specimens collected at each contact. Participants were paid $50 at each of the testing intervals, $25 for completing the questionnaire and $25 for giving a blood and urine sample. Respondents' answers are confidential and protected by a Federal Grant of Confidentiality issued to the Principal Investigator of these studies. Participation in the research project is voluntary, and over 95% of all eligible subjects have agreed to participate in the study at baseline. Subsequent interviews are occurring at 42 and 54 months after release from prison.
All baseline measures reported are self-report items from these interviews. Dichotomous baseline measures used in the analyses are gender and African-American status. A measure of frequency of drug use was derived from a series of baseline questions asking the frequency of use of injecting or noninjecting cocaine, heroin and speed, crack, PCP, hallucinogens, and other opiates in the 6 months prior to prison. The maximum reported use of any of these substances was recorded on a scale of 0 (no use) to 6 (use more than once a day). Ordinal or continuous baseline measures used are age of first arrest, number of previous arrests, number of previous incarcerations, and age at baseline interview.
There are several major differences in the composition of the four groups being compared. First, assignment to the COMPARISON or CREST groups was determined by the investigators by random number. Second, the COMPARISON and CREST groups include men and women, whereas the KEY group does not-until very recently, there was no women's in-prison TC in Delaware. Third, because the determination of group membership was made at the time of the baseline interview, those included in the KEY or KEY-CREST groups were only those still in the KEY program at the time of their release, graduates. The CREST and COMPARISON groups included all those so assigned at the time of their release, regardless of actual attendance. Fourth, the KEY-only group included those clients who graduated before the CREST program was established. Once the CREST program was established, virtually all KEY graduates were assigned to it. Hence the KEY samples are nonrandom, and the KEY-only group serves as an historical comparison for the KEY-CREST group.
Another concern about making comparisons with a no treatment group is that such a distinction is far from possible in field research. Many of the so-called no treatment COMPARISON group did get some treatment help. Fully 56% of the no treatment group indeed reported some treatment during the initial 180 days after prison release (while the CREST group was at the TC). Alternately, some CREST clients did not remain in the program more than a few weeks. The mean number of days in treatment was, of course, much higher for the CREST assignees (116 days), but it was not zero for the comparison group (24 days with some treatment exposure). So random assignment still resulted in a less than perfect distinction between treatment and no treatment, though it is possible to state that the COMPARISON group did not get TC treatment.
Some baseline characteristics of the four sample groups are presented in table 1. As would be expected from the assignment process, the CREST and COMPARISON groups are similar on most baseline variables, but differ from the KEY groups on several dimensions. The two KEY groups are older and contain more AfricanAmericans and fewer whites and Hispanics. The KEY groups have, on average, longer criminal histories. All of the KEY-only and virtually all of the KEY-CREST respondents are male and have had previous drug-abuse treatment by virtue of their participation in The KEY. About a fifth of the respondents in the other two groups report no drug treatment.
There are several outcome measures. Each of the outcome measures combine information and test results collected at both the 6-month and 18-month follow-up interviews. The outcome measure of arrest status is dichotomous, coded as 1 if the respondent reported no arrests for new offenses since release from prison, 0 if otherwise. The first outcome measure of drug use was based upon a composite of questions similar to those used to indicate frequency of drug use before prison, combined with urinalysis results at both the 6-month and 18-month follow-up interviews. If the respondent reported that none of the drugs on that list had been used and tested negative, drug status was coded as 1 for "drug-free at 18 months."
Any positive response or test results to any drug on the list was coded as 0. The second outcome measure of drug use measures drug-free status in the 30 days prior to the 18-month interview, by self-report of use of any of the categories of illegal drugs and by a negative urine screen.
Two other outcome measures ascertain the maximum frequency that the respondent used drugs at any time during the follow-up period. For each different living arrangement that the respondents had during follow-up, divided into months, they were asked how often they used each of 16 different drugs or combinations of drugs, including alcohol. Responses were scored on an ordinal scale from 0 (never used) to 6 (used several times a day or more). The outcome measure used in the analyses reported in figure 4 below, maximum frequency of illegal drug use, combines answers for all illegal drugs and all time periods and reports the maximum frequency that the respondent reported use of any illegal drug. The outcome measure for figure 5, maximum frequency of intoxication, reports on a legal drug used to the point of drunkenness. Alcohol use, particularly to intoxication, is often associated with illegal drug use and with rearrest. As such, it is an important criterion for judging drug-treatment effectiveness, and one that is often overlooked in evaluation studies.
Analyses and Results
The statistical analyses reported use both logistic regression and ordinary least squares (OLS) regression, depending on the metric of the dependent variable. Logistic regression is closely related to the more familiar OLS regression, but is more appropriate for the statistical evaluation of predictors of dichotomous outcomes such as relapse to criminal behavior or drug use (Hosmer and Lemeshow 1989). The effect of group assignment is modeled by a dummy classification with the COMPARISON group, i.e., the excluded category. The other independent control variables are those shown in table 1.
For both the logistic and OLS regression analyses reported in figures 1 through 5, the results are presented as regression histograms, showing the predicted score on the dependent variable for each comparison group, controlling for the effects of the other independent baseline variables. For the logistic regression analyses in figures 1 through 3, the group scores are predicted probabilities (or percentages) for each group with the other co-variates in the regression set at their mean value. For the OLS regression results in figures 4 and 5, the group scores are means from multiple classification analysis, adjusted for the effects of the other co-variates.
As indicated in figures 1 and 2, those in the CREST and KEY-CREST groups had significantly higher arrest-free and drug-free rates (p < .01), compared to those in the KEY and COMPARISON groups. The proportions of arrest-free and drug-free reported are parameter estimates that control for demographic, criminal, and drug history variables using logistic regression techniques.
These analyses include everyone who had an opportunity to recidivate or relapse, even those who were back in prison at the 18-month interview as the result of an earlier arrest or relapse. Importantly, more than half of the CREST clients and more than three-fourths of the KEY-CREST clients had no new arrests during the 18month period after release from prison. The KEY group actually was slightly more likely than the COMPARISON group to have a new arrest. Analyses of the KEY group not shown here suggest a bi-polar pattern for the KEY group. Graduates of the in-prison TC were likely to either relapse or recidivate soon after release from prison or else to do quite well; there was not as gradual a pattern of relapse and rearrest, as seen in the other groups.
The noteworthy significance of the drug-free estimates is in the highly stringent criteria used for inclusion in this group. To be classified as drug-free, clients must have had no self-reports of any use of any illegal drug at the 6-month and 18-month interviews, and no positive urinalysis on either occasion. As such, even a single use of cocaine or an occasional marijuana cigarette in the 18-month period would have placed a client in the relapse category. Nevertheless, almost a third of the CREST clients and almost half of the KEY-CREST clients were drug free after 18 months, two and three times, respectively, the proportion of the COMPARISON group.
Relapse is a common phenomenon associated not only with the treatment of substance abuse, but also with smoking cessation, weight reduction, and other behavioral change programs. In fact, even occasional and minor relapses seem to be the norm rather than the exception. As such, being totally drug-free for the entire 18month post-release period is a particularly stringent criterion and could be considered an unrealistic expectation. Thus, several other drug-use criteria are considered, with results reported in figures 3 through 5.
Figure 3 presents logistic regression results predicting the proportion of subjects who were drug-free during the 30-day period immediately prior to the 18-month follow-up interview. Drug-free status is defined as no report of any drug use in the previous 30-day period, and no positive urinalysis at the 18-month interview. Because this analysis includes only those demonstrably on the street at the time of the 18-month interview, there are fewer cases than the analyses in figures 1 and 2 (which included even those in prison at 18 months, because they had an opportunity for rearrest or relapse earlier). The analysis controlled for the same baseline variables as in figures 1 and 2. As indicated in figure 3, 51% of the CREST clients and 68% of the KEY-CREST clients were drug-free during the month prior to the 18-month follow-up contact. Both of these percentages are significantly greater than the COMPARISON group (p < .01). And, although the difference is not significant due to the small sample size, the KEY group appears to do better on the past 30-day criterion than the COMPARISON group. This result, combined with the earlier observation of the bi-polar distribution, suggests that if KEY-only clients managed to stay out of trouble early after release, they may do better than the COMPARISON group. An alternative mechanism for measuring degree of relapse that occurs at postrelease and post-treatment is reported in figure 4. The data represent analyses of quantitative reports of frequency of drug use, rather than the dichotomy of use or no use. As noted earlier, respondents reported their use of drugs on a six-point scale ranging from 0 (no use) to 6 (use several times daily). The maximum value of these frequency reports of illegal drug use at any time during follow-up is reported on the vertical axis in figure 4. The mean values for each group are controlled for the same background variables as in the previous figures, but linear regression techniques were used to generate these estimates.
As illustrated in figure 4, the mean value for the COMPARISON group was 3.29 (3 = drug use about once a week) as compared with 2.16 ( 2 = 1-3 times a month) for the CREST group and 1.12 (1 = less than once a month) for the KEY-CREST group. Again, significant differences (p < .01) in drug use are evident between the COMPARISON group and the CREST and KEY-CREST groups.
Similarly, the impact of treatment group on the use of a gateway drug to relapse-alcohol to point of intoxication-is also evident in this type of analysis, as seen in figure 5. The estimates vary from an average of "about monthly intoxication" for the COMPARISON group to only a fraction above "no intoxication" for the KEY-CREST group.
It is, of course, not strictly appropriate to apply standard tests of significance to both random and purposively selected comparison groups in the same analyses. However, the analyses have attempted to compensate by controlling for some of the known group differences. In separate analyses not reported here, the potential effects of previous involvement in treatment and length of time in treatment were tested by including these variables in the multivariate model. These variables proved non significant in the context of treatment group assignment and did not change the pattern or magnitude of group differences.
Overall, across the five outcome measures, the most striking effect is the consistent benefits of the transitional TC treatment in a program like CREST and the even greater benefits for those who have both primary and secondary treatment (KEY-CREST). Both the CREST and KEY-CREST groups do significantly better on each of five outcome measures examined here, relative to the COMPARISON group. However, for none of the five outcome measures is the KEY group statistically distinguishable from the COMPARISON group.
By improving the program outcome assessment with appropriate multivariate controls and incorporating a sufficient length of follow up time, a consistent and persuasive pattern of results emerges supporting the continuum of TC treatment for the reduction of drug use and criminal recidivism. These data provide evidence for the effectiveness of the TC continuum extending beyond just an in-prison program. The outcome data especially support the value of a reentry work release TC like CREST. Both groups assigned to treatment in CREST had significantly improved drug and arrest outcomes, even when other plausible predictors were statistically controlled, including actual time in drug treatment. The group exposed to the prison TC only, the KEY group, was not significantly different from the COMPARISON group, suggesting that in-prison TC experience only may not produce sufficient positive program effects with respect to drug and arrest outcomes, although some in this group may do quite well and be more likely to seek treatment after release.
The effectiveness of the TC continuum model has not gone unnoticed. On September 10, 1996, in a press release on drug-abuse treatment for state prisoners from the Department of Justice, the effectiveness of the KEY-CREST model was referenced as a model program for use by other states (Department of Justice 1996). The following day, in his address announcing an appropriation of $27 million for residential drug treatment for prisoners, President Bill Clinton also referenced the effectiveness of the Delaware model (ABC 1996).
This research was supported by Grants DA06124 and DA06948 from the National Institute on Drug Abuse.
ABC World News Tonight 1996 News story, September 11. Anglin, M.D,, and Y. Hser
1987 Addicted women and crime. Criminology 25 (May), 359-397. Anglin, M.D., and Y. Hser
1990 Treatment of drug abuse. In Drugs and Crime, ed. M. Tonry and J.Q. Wilson, 393460. Chicago, IL:University of Chicago Press.
Ball, J.C., and A. Ross
1991 The Effectiveness of Methadone Maintenance Treatment. New
Ball, J.C., J.W. Shaffer, and D.N. Nurco
1983 Day-to-day criminality of heroin addicts in Baltimore: A study in the continuity of offense rates. Drug and Alcohol Dependence 12:119-142. Brown, B.S., ed.
1979 Addicts and Aftercare: Community Reintegration of the Former Drug User. Beverly Hills:Sage Publications. Chaiken, M.R.
1989 In-prison Programs for Drug-involved Offenders. Washington, DC: National Institute of Justice. Chavaria, F.R.
1992 Successful drug treatment in a criminal justice setting: A case study. Federal Probation 56(March):48-52. Condelli, W.S., and R.L. Hubbard
1994 Client outcomes from therapeutic communities. In Therapeutic Community: Advances in Research and Application, eds. F.M. Tims, G. De Leon, and N. Jainchill, 80-98. Rockville, MD:National Institute on Drug Abuse Research Monograph No. 144,
De Leon, G.
1984 Program based evaluation research in therapeutic communities. In Drug Abuse Treatment Evaluation: Strategies, Progress, and Prospect" eds. F.M. Tims and J.P Ludford. 69-87. Rockville, MD:National Institute on Drug Abuse Research Monograph No. 51. De Leon, G.
1994 Therapeutic communities. In The American Psychiatric Press Textbook of Substance Abuse Treatment, eds. M. Galanter and H.D. Kleber. Washington, DC:American Psychiatric Press. Department of Justice
1996 Residential substance abuse treatment for state prisoners fact sheet,
September 11. Washington, DC:Office of Justice Programs. Field. G.
1992 Oregon prison drug treatment programs. In Drug Abuse Treatment in Prisons and Jails, eds. C.G. Leukefeld and F.M. Tims, 142-155. Rockville, MD:National Institute on Drug Abuse Research Monograph No. 118. Forcier, M.W.
1991 Substance abuse, crime and prison-based treatment: Problems and prospects. Sociological Practice Review 2(2):123-131. Gossop, M., L. Green, G. Phillips and B. Bradley 1990 Factors predicting outcome among opiate addicts after treatment. British Journal of Clinical Psychology 29:209-216. Hall, S.M., B.E. Havassy, and D.A. Wasserman
1990 Commitment to abstinence and acute stress in relapse to alcohol, opiates, and nicotine. Journal of Consulting and Clinical Psychology 58(2): 175-181. Hawk, K.
1993 Drug treatment in prisons. 1993 Congressional Summit on U.S. Drug Policy, May 7 Washington, DC. Hooper, R.M., D. Lockwood, and J.A. Inciardi
1993 Treatment techniques in corrections-based therapeutic communities. The Prison Journal 73(3/4):290-306. Hosmer, D.W., and S. Lemeshow
1989 Applied Logistic Regression. New York:Wiley and Sons. Hubbard, R.L., M.E. Marsden, J.V. Rachal, H.J. Harwood, E.R. Cavanaugh, and H.M. Ginzburg
1989 Drug Abuse Treatment: a National Study of Effectiveness. Chapel Hill:University of North Carolina Press.
1979 Heroin use and street crime. Crime and Delinquency 25:335-46. Inciardi, J.A.
1992 The War on Drugs II: The Continuing Epic of Heroin, Cocaine, Crack, Crime, AIDS, and Public Policy. Mountain View, CA:Mayfield Publishing Co.
Inciardi, J.A., ed.
1993 Drug Treatment and Criminal Justice. Newbury Park, CA:Harcourt Brace.
Inciardi, J.A., D. Lockwood, and S.S. Martin 1994 Therapeutic communities in prison and work release: Some clinical and policy implications. In Therapeutic Community: Advances in Research and Application, eds. F.M. Tims, G. De Leon, and N. Jainchill, 259-267. Rockville, MD:National Institute on Drug Abuse Research Monograph No. 144.
Inciardi, J.A., D.C. McBride, and J.E. Rivers 1996 Drug Control and the Courts. Newbury Park, CA:Sage. Inciardi, J.A., and A.E. Pottieger
1986 Drug use and crime among two cohorts of women narcotics users: An empirical assessment. Journal of Drug Issues 16:91-106. Inciardi, J.A., and A.E. Pottieger
1991 Kids, crack, and crime. Journal of Drug Issues 21:257-270. Inciardi, J.A., and A.E. Pottieger
1994 Crack cocaine use and street crime. Journal of Drug Issues 24:273-292. Johnson, B.D., P. J. Goldstein, E. Preble, J. Schmeidler, D.S. Lipton, B. Spunt, and T. Miller
1985 Taking Care of Business: The Economics of Crime by Heroin Users. Lexington, MA:Lexington Books. Leukefeld, C.G., and F.M. Tims (eds.)
1988 Compulsory Treatment of Drug Abuse: Research and Clinical Practice. Rockville, MD:National Institute on Drug Abuse Research Monograph No. 86.
Leukefeld, C.G., and F.M. Tims, eds.
1992 Drug Abuse Treatment in Prison and Jails. Rockville, MD:National Institute
on Drug Abuse Research Monograph No. 118. Lockwood, D., and J.A. Inciardi
1993 CREST Outreach Center: A work release iteration of the TC model. In Innovative Strategies in the Treatment of Drug abuse, eds. J.A. Inciardi, F.M. Tims, and B.W. Fletcher, 61-69. Westport, CT:Greenwood Press. Martin, S.S., C.A.Butzin, and J.A. Inciardi 1995 Assessment of a multistage therapeutic community for drug involved offenders. Journal of Psychoactive Drugs 27:109-116.
McLellan, A., and A.I. Alterman
1991 Patient treatment matching: A conceptual and methodological review with suggestions for future research. In Improving Drug Abuse Treatment" eds. R.G. Pickens, C.G. Leukefeld, and C.R. Schuster, 114-135. Rockville, MD:National Institute on Drug Abuse.Research Monograph 106. Reno, J.
1993 Testimony. 1993 Congressional Summit on U.S. Drug Policy. May 7, Washington, DC. Rouse, J.J.
1991 Evaluation research on prison based drug treatment programs and some policy implications. The International Journal of the Addictions 26(1): 29-44.
Simpson, D.D., J. Savage, and M.R. Lloyd
1979 Follow-up evaluation of treatment of drug abuse during 1969 to 1972.
Archives of General Psychiatry 36:772-780. Simpson, D.D., and S.B. Sells
1982 Effectiveness of treatment for drug abuse: An overview of the DARP
research program. Advances in Alcohol and Substance Abuse 2:7-29. Speckart, G. and M.D. Anglin
1986 Narcotics use and crime: An overview of recent research advances.
Contemporary Drug Problems, Winter:741-769. Tims, F.M., G. De Leon, and N. Jainchill, eds. 1994 Therapeutic Community: Advances in Research and Application. Rockville,
MD:National Institute on Drug Abuse.Research Monograph No.144. Wexler, H.K., G.P. Falkin, D.S. Lipton, A.B. Rosenblum, and L.P. Goodloe
1988 A Model Prison Rehabilitation Program: an Evaluation of the "Stay'n Out" Therapeutic Community. New York: Narcotic and Drug Research, Inc. Wexler, H.K., G.P. Falkin, and D.S. Lipton
1990 Outcome evaluation of a prison therapeutic community for substance abuse treatment. Criminal Justice Behavior 17(1):71-93. Wexler, H.K. and R. Williams
1986 The "Stay'n Out" therapeutic community: Prison treatment for substance abusers. Journal of Psychoactive Drugs 18(July-September):221-230. Yablonsky, L.
1989 The Therapeutic Community: A Successful Approach for Treating Substance Abusers. New York:Gardner Press.
James A. Inciardi, Ph.D., is the Director of the Center for Drug and Alcohol Studies at the University of Delaware, Professor in the Department of Sociology and Criminal Justice at Delaware, Adjunct Professor in the Department of Epidemiology and Public Health at the University of Miami School of Medicine, a Distinguished Professor at the State University of Rio de Janeiro, and a Guest Professor in the mt of Ph at the Federal University of Rio Grande do Sul in Porto Alegre, Brazil. Steven Ss Martin, MA, is an Associate Scientist at the Center for Drug and Alcohol Studies and coprincipal investigator on the "Ongoing Studies" project. He has authored or co-authored more than 40 articles on the epidemiology, etiology, prevention, and treatment of substance abuse. Clifford A Butzin, Ph.D., is also an Associate Scientist at the Center for Drug and Alcohol Studies He was previously Chief of the Research Consulting Division at Wilford Hall Medical Center in San Antonio, and before that he was on the faculty at Duke University. Robert ML Hooper, Ph.D., Executive Director of Substance Abuse Treatment, Correctional Medical Systems, has been involved in justice prog nationally for the past 25 years. Over the last 8 years he has developed intensive abuse treatment programs with the Delaware model receiving national attention. Lana D. Hanison, Ph.D., is the Associate Director and a Senior Scientist in the Center for Drug and Alcohol Studies at the University of Delaware. She has worked on the three largest epidemiological drug surveys in the United States, the National Household Survey on Drug Abuse, the monitoring the Future survey, and the Drug Use Fog program Adddress correspondence and reprint requests to James A Inciardi, Comprehensive Drug Research Center, 1400 N.W. lOth Ave. (D-93), Miami, FL 33136.…