Printer Friendly

Predicting desistance or persistence in the substance-abusing, mentally ill, supervised offender population using Laub and Sampson's (2001) social controls and individual factors theory.

This study investigated the relationship that social controls and individual factors have on the persistence rate in the substance-abusing, mentally ill, supervised offender population (N = 886). The data predicted 83% of persistence.

Keywords: desistance, substance-abusing mentally ill, offenders


Following the initial movement to deinstitutionalize the mentally ill during the 1960s and 1970s (Farabee & Shen, 2004; Klerman, 1977), the number of incarcerated inmates serving time with a classification of mentally ill or chronically mentally ill with a history of substance abuse or dependence doubled in the United States (James & Glaze, 2006; Maloney, Ward, & Jackson, 2003; Steven, 2000). Clients diagnosed with both a mental health and substance abuse condition (co-occurring disorders, n.d.) are commonly referred to as substance-abusing mentally ill, or SAMI (Telias, 2001). This population has a number of concerns, including a high prevalence of relapse, medication nonadherence, low treatment engagement rate, and numerous emergency room visits in comparison with other clients diagnosed with only a mental illness (Flo et al., 1999; Telias, 2001). SAMI offenders are often neglected, receive little or no treatment, may be accused of malingering, and are treated as disciplinary problems by the court system and correctional staff (Abramsky & Fellner, 2003).

Trans-institutionalization is the process of shifting SAMI offenders from community-based mental health care to correctional institution care (Steven, 2000). Trans-institutionalization developed as a result of decreased funding to community mental health providers and a change in the attitudes of criminal justice policy decision makers, the corrections industry, and the mental health field regarding effective therapeutic interventions and services for the SAMI population (Lamb, Weinberger, & Gross, 2004; Steven, 2000). Following the trans-institutionalization of the mentally ill into prisons throughout the United States, jail and prison administrators were obligated to accommodate the influx of mentally ill offenders by converting inmate dorms and, in a few cases, entire prisons from a general inmate holding facility to a therapeutic treatment facility (Abramsky & Fellner, 2003; Groom, 1999; Maloney et al., 2003). Prisons and municipal jails in the United States progressively hired licensed mental health staff to develop therapeutic programs that specifically targeted the special needs of mentally ill inmates (Adams & Ferrandino, 2008; Lurigio, 2001). Such programs not only involved medication management and therapeutic interventions, but also involved therapeutic centers/therapeutic communities that specifically focused on mental health management and substance abuse prevention (Groom, 1999; Wang, Owens, Long, Diamond, & Smith, 2000).

Despite these intervention efforts to accommodate the special needs of the SAMI inmate population, their rate of retention and reincarceration within the correctional system continues to be, on average, higher than that of offenders with no history of mental illness (Maloney et al., 2003; U.S. Department of Health & Human Services, 2012). Furthermore, SAMI offenders are more likely to receive longer prison sentences; are less likely to be paroled back into the community; and, when paroled, are more likely to have their parole status revoked and be returned to prison on either a parole technical violation or new criminal conviction (Draine & Solomon, 2001; Groom, 1999; National Research Council of the National Academies, 2007).

New taxonomies, such as desistance and persistence, emerged in the criminal justice and mental health fields that characterize the SAMI population's nonoffending/offending behaviors. First coined by Laub and Sampson (2001), desistance is viewed as a process in which an individual's offending behaviors subside or abruptly stop. Persistence is an individual's continuation or repetition of offending behaviors, especially after experiencing the negative consequences associated with the undesirable behaviors. According to Laub and Sampson, changes in either desistance or persistence are associated with social controls, such as community supervision, community linkages, and individual factors.

Community supervision is a branch of the Department of Corrections that asserts legal control of the SAMI-supervised offender (SAMI-SO). Community supervision usually lasts 1 to 5 years, or until the SAMI-SO individual is formally dismissed and is no longer obligated to serve the remainder of his or her sentence (National Research Council of the National Academies, 2007). As a form of social control, the purpose of community supervision is to monitor and control or reduce persistent criminal behaviors (Skeem, Emke-Francis, & Louden, 2006). SAMI-SO individual needs include substance abuse testing and monitoring, employment, housing procurement for homeless supervisees, and multiple linkages for hard-to-find services (National Research Council of the National Academies, 2007; Slate, Roskes, Feldman, & Baerga, 2003).

Community linkages involve using a holistic approach to treat offenders and match their needs to specific community services. Community services can include psychiatric care, substance abuse counseling, housing procurement, rehabilitation, vocational and mental health counseling, and essential services from nurses and peer counselors who also assist in monitoring the behaviors of SAMI-SO individuals, if necessary (Morrissey, Meyer, & Cuddeback, 2007; Slate et al., 2003). What differentiates community linkage services as a form of social control from other community treatment provider programs is the relationship that community linkage services have with community supervision staff. According to Draine and Solomon (2001), a community linkage services provider becomes an extension of community supervision by way of supervision agreements and stipulations between the community supervision staff and the SAMI-SO individual to attend mandated mental health and/or substance abuse linkage services. Because of this relationship, a community linkage provider, such as an addictions counselor, can monitor and report any changes in the mental health status or program participation of the SAMI-SO individual to their community supervisor, which may prevent or stop continued persistent behaviors before they result in the SAMI-SO individual being returned to prison.

Laub and Sampson (2001) suggested that desistance stems from a variety of developmental, psychological, and sociological processes, such as medication adherence, years of education, previous and current mental health treatment, current diagnosis, mental health classification, and number of months in a correctional facility. Additional individual factors can include the quality of housing subsequent to release from a correctional institution to resumption of using mood-altering illegal substances (Farabee & Shen, 2004; Gagliardi, Lovell, Peterson, & Jemelka, 2004).

Despite this collection of hypothesized predictor variables, only one study to date (Delaney, 2011) has investigated the manner in which community supervision, community linkages, and SAMI-SO individual factors were associated with desistance and persistence. Specifically, Delaney (2011) demonstrated that persistence was related to the intensity of supervision; supervisors' training in mental health matters; community linkage; active involvement in substance abuse and/or mental health treatment; and the supervisee's age, education level, and housing status. Yet to be investigated is the amount of variance in desistance and persistence these predictor variables account for when combined. Therefore, the purpose of this study was to test Laub and Sampson's (2001) theory of persistence/desistance and extend Delaney's findings using simultaneous multiple regression analysis to determine the best combination of predictors of persistent incidents. Specifically, the predictor variables used in this study were supervisory mental health training status, linkage to community-based service providers, active involvement with mental health treatment, employment status, ethnicity, mental health classification, housing classification, supervision classification, time spent on active community supervision, age, years of formal educational experience, time spent incarcerated, and positive drug screens/intoximeter results.



The SAMI-SO population was operationally defined, identified, and selected using the mental health classification system developed by the Ohio Department of Rehabilitation and Corrections (ODRC; Urban Institute, 2008). Two groups of offenders are referred to as C1. The first group consists of any offender that meets the criteria for severely mentally ill (SMI)--has significant impairment in thought, mood, judgment, behavior, and no capacity to recognize reality or cope with ordinary demands of life within the prison environment, as manifested by substantial pain or disability. The second group of C1 offenders consists of any offender that may have any psychiatric diagnosis that entails some impairment in functioning or acuity, as demonstrated in a pattern of high-risk behaviors. The third group of offenders, classified as C2, consists of any offender that does not meet the criterion for SMI, but has a psychiatric diagnosis; receives mental health services, including psychotropic medication; and has no impairment in functioning or acuity, as demonstrated in a pattern of high-risk behaviors.

The population characteristics used in selecting case samples for this study were adult male offenders 18 years of age or older that have previously been incarcerated in a correctional institution; have a previous or current history of mental illness; demonstrate substance use or dependence; have a previous history of treatment, and are no longer on active community supervision.


A survey of the six state parole regions covering a 5-year period from January 1, 2005, to June 30, 2009, disclosed a total accessible case sample of 7,110 community mental health linkage referral cases. The proportional sampling method (Fraenkel & Wallen, 2003) used resulted in an oversampling of C1 cases (n = 505, 57%) and C2 cases (n = 381, 43%), which is more than enough to establish a confidence level of 95% and a minimum margin of error or confidence interval of 5%. Using this method of oversampling more than compensates for any missing data or incomplete files and eliminates the problems associated with both sampling selection and geographic bias.

The participants' average age was 40 years (SD = 10.30, range = 22-75). The sample consisted of 344 (38.9%) African Americans, 533 (60.2%) European Americans, and 9 (1%) Hispanics. The number of years of educational experience ranged from 5 to 16 years, with a mean of 11 years (SD = 1.406). Subsequent to being released to parole supervision, 376 (42.4%) offenders were supervised at the intensive supervision classification (four supervision contacts per month). Four hundred and thirty five (55.9%) participants were supervised at the basic classification (two supervision contacts per month). Additionally, 13 (1.5%) participants were supervised at the basic low classification (one supervision contact every 3 months), and only two (0.2%) participants were supervised at the monitored time classification (one initial supervision contact only). Additionally, 151 (17%) participants received special supervision by a mental health caseload specialist, whose primary function is to supervise mentally ill offenders. The remaining 735 (83%) participants were supervised by a general caseload supervision parole officer. Eight hundred and fourteen (91.9%) SAMI-SO participants were linked to a community treatment provider via the correctional institution, the adult parole authority's offender services network coordinator, or the supervising parole officer to a community treatment provider in the community for the purpose of mental health, substance abuse, or psychiatric care. Only 72 (8.1%) participants were not linked to a community treatment provider. While on active community supervision, 551 participants (57%) received treatment (mental health, psychiatric, or substance abuse) from a community treatment provider. Three hundred and seventy-five (42.3%) participants were not actively involved in treatment. Two hundred and twenty-five (25.9%) participants were employed, and 671 (74.6%) participants were unemployed while on active community supervision.

Upon release to active community supervision, 409 (46.2%) SAMI-SO individuals were placed in a permanent residence with a family member, in a preincarceration residence, or with a close friend of the family. Another 356 (40.2%) offenders were released to a temporary residence, such as a halfway house, three-quarter-way house, transitional housing, or a treatment facility. The remaining SAMI-SO individuals, (n = 121, 13.7%) were placed in a homeless shelter or were homeless without a permanent residence. The number of months served on active community supervision ranged from 2 to 62, with a mean of 24 (SD = 16.408). The total number of months of incarceration ranged from 2 to 444, with a mean of 56 months (SD = 61.267). The mean number of positive drug alcohol screens was 1.42 (SD = 1.992, range = 0-18). The number of persistent incidents for all 886 cases reviewed ranged from 0 to 6, with a mean of 1.39 (SD = 1.74).

The predictor variables were coded in the following manner: active treatment (1 = yes, 2 = no); age (continuous); ethnicity (1 = African American, 2 = European American, 3 = Latin American); employed (1 = yes, 2 = no); housing (1 = permanent, 2 = temporary, 3 = homeless); linked (1 = yes, 2 = no); mental health case specialist (1 = yes, 2 = no); mental health classification (1 = C1, 2 = C2); months on supervision (continuous); number of months in prison (continuous); positive drug or alcohol screens (continuous); previous treatment (1 = yes, 2 = no); supervision classification (1 = intense, 2 = basic, 3 = basic low, 4 = monitored time); and years of education (continuous).

The criterion variables were persistence (continuous) and desistance (0). Because of the unique nature of this study, the criterion (persistence) was coded as a zero (0). Any continuous numbers above zero to the highest number found in the data were considered as either a persistent incident or incidents. Given that a persistent incident or incidents were accumulative, ranging from one to perpetuity, a persistence scale appropriate for each predictor variable relative to persistent incidents was constructed, with a means plot, after each variable was analyzed. The following markers were analyzed, measured, and coded as either a single persistent incident or incidents in meeting the criterion persistence, which is a continuous variable: new offenses or arrest; reincarceration (number of persistent incidents associated with prison sanction time for post-release control offenders [1-270 days], or revocation for parole offenders); number of persistent incidents associated with being sentenced to county or municipal jail time (12 months or less); resentencing on a new number resulting from a new conviction; number of persistent incidents related to noncompliance or violation of supervision conditions; number of unsuccessful treatment failures or noncompliance with treatment; number of incidents associated with drug use or abuse; number of written/verbal sanctions or violation sanction process hearings resulting from a supervision violation; number of persistent incidents associated with noncompliance with a treatment provider, such as failure to attend scheduled meetings, attendance, treatment disruption, and medication nonadherence; number of linkage failures; and any identified persistent behaviors that interfere with normal functioning or active community supervision.


A multiple regression analysis was conducted to examine the relationship between social controls and individual factors on persistent behavioral patterns in the SAMI-SO population during a period of active community supervision. The three categories of predictor variables were supervision, linkages, and individual factors. The supervision variables were type of community supervision classification, supervisor's mental health training status, and length of time on active community supervision. The linkage variables included whether or not a SAMI-SO was linked to a community-based service and whether or not he was receiving mental health treatment while on community supervision. Individual variables included age, ethnicity, education level, employment status, mental health classification, housing classification, length of time in prison, and alcohol and other drug screens.

The simultaneous multiple regression means, standard deviations, and within-class intercorrelations are listed in Table 1. The combination of variables to predict persistent incidents (supervision level, mental health case specialty, months on supervision, community linkages, active in treatment, age, ethnicity, years of education, employed, previous mental health treatment, mental health classification, housing status, months incarcerated, and positive drug or alcohol) status was statistically significant, F(14,871) = 170.68, p [less than or equal to] .001, and indicated that a combination of the predictor variables was significantly related to the criterion variable persistent incidents. The beta coefficients are presented in Table 2. High supervision level, mental health case specialty, months on supervision, community linkages, active in treatment, ethnicity, years of education, employed, previous mental health treatment, housing status, months incarcerated, and positive drug or alcohol screens significantly predicted persistent incidents in this model when all 14 predictor variables were included. The adjusted [R.sup.2] value was .729. This indicates that 73% of the variance in the criterion variable persistent incidents could be predicted from the predictor variables and was explained by this model.


This research study focused on 14 predictor variables that may be good indicators for determining the relationship between desistance and persistent behaviors in the SAMI-SO population during a period of active community supervision. This study extends Delaney's (2011) research and represents the first empirical study of its kind that focuses exclusively on the combined relationship that social controls and individual factors have as potential catalysts in determining desistance or persistent behaviors specific to the SAMI-SO population.

The results of the multiple regression analysis indicated that combinations between the above variables and persistent incidents are good predictors in evaluating the SAMI-SO population during an active period of supervision. Additionally, the results of a simultaneous multiple regression indicate that seven of the 14 predictor variables (supervision classification, months on supervision, active treatment, age, years of education, mental health class, and positive drug or alcohol screens) were very strong indicators in measuring desistance or persistent behaviors. Specifically, the data indicate that supervised SAMI-SO individuals are more likely to recidivate if they have fewer required monthly contacts with their supervising officer, are not actively engaged in treatment, and are significantly impaired in functioning because of a pattern of high-risk behaviors associated with their mental illness. Furthermore, the younger and less educated a SAMI-SO individual is, and the more positive substance abuse screens a SAMI-SO individual has, the more likely the individual is to be persistent. These statistical results suggest that a combination of all 14 predictor variables is a good model that can be duplicated for the purposes of evaluating or researching the SAMI-SO population during an active period of community supervision.


The results of this study provide both practical and theoretical implications. Because these data indicate an inverse relationship between recidivism and contacts with their supervising officer, counselors who work with SAMI-SO individuals on supervision are encouraged to advocate on behalf of their clients to increase the frequency of contacts with their supervising officer. We also suggest that counselors and counseling agencies place increased emphasis on retaining SAMI-SO clients in the counseling relationship. Policies regarding discharging clients after a certain number of no-shows may need to be reexamined in light of the finding of the inverse relationship between active engagement in treatment and persistence. In addition to policy adaptation that extends the time to engage or recapture SAMI-SO clients, counseling agencies with a significant SAMI-SO population may wish to create special treatment engagement teams whose chief responsibilities are to engage or reengage SAMI-SO clients in the counseling process. These teams could use motivational enhancement strategies and community linkage staff to reduce or eliminate unique barriers that might exist among this population of clients. For the SAMI-SO population in this study, there was a link between their high-risk behaviors and positive substance abuse screens and their rates of persistence. These findings suggest that counselors should conduct thorough risk assessment and substance use assessments at the onset of and throughout the counseling process. Likewise, for those SAMI-SO clients who exhibit high-risk behaviors and/or substance use issues, relapse prevention planning should begin early on in the counseling relationship. Finally, because less educated SAMI-SO clients were more likely to persist than their more educated peers, counselors may wish to engage their SAMI-SO clients in discussions about how improving their educational level may fit into their long-term career goals.

Furthermore, these results suggest that counselors should shift their attention away from looking at the recidivism rate as the single marker for determining success or failure to focusing on individual characteristics of the SAMI-SO population relative to social controls and individual factors in determining individual needs that contribute to success. Such redirection may encourage the development of new treatment models that encompass the whole human by providing services that bridge the gaps between mental health and substance abuse services, psychiatric care, case management services, employment, housing, education, and most important, successful linkages. By bridging these gaps, mental health specialists, criminologists, and social scientists might be able to develop an effective therapeutic model that can meet the specific needs of the SAMI-SO population. Such a model could potentially improve this population's overall functioning in the community and interrupt their propensity to repeatedly cycle through the criminal justice system.

We further recommend a change in the manner in which courts and corrections systems measure the success or failure of community-based treatment services. One of the main benefits of measuring the viability of a community treatment program on desistance or persistence measures instead of recidivism could be to alter the funding streams from less viable treatment programs to treatment programs that have better outcomes in providing services to the SAMI-SO population.

Finally, a number of Laub and Sampson's (2001) theoretical constructs did not contribute to the prediction of recidivism. Flowever, their identification of specific markers that contributed to desistence behavior helped in the development of this model. We recommend that future researchers replicate these methods to determine if our findings were unique to this sample or to identify new markers that could enhance this model. The model created from this research study could have broad implications when used as a formula to make clinical predictions about who will and will not be persistent in the future. This model could help counselors determine which of their clients would benefit from closer or more frequent supervision and how to strategically allocate services or resources to reduce the risk of persistent behavior. Counselor supervisors can use this model in determining which offenders should be assigned to which counselors. For example, counselor supervisors are encouraged to use the model to predict clients that are at greater risk to persist and then assign these higher risk clients to more experienced counselors who have the training and familiarity with this special population.


This study is limited in scope to an analysis of secondary data from the ODRC database of SAMI-SO individuals released from community supervision. The homogeneity of the case samples' ethnicity (mostly African Americans and European Americans), although reflective of the majority of inmates in Ohio's correctional system and community supervision, constrains the generalizability of these results to dissimilar ethnic groups. Furthermore, the population of interest is limited to offenders classified in the database as having a history of mental health issues and substance-abusing behavior and excludes other offenders with dissimilar characteristics who are on community supervision. A retrospective correlational research approach restricts one from making causal statements about the nature of the relationships between the predictors and criterion variables. Finally, because no two community supervisors manage their caseload the same way, successful or unsuccessful outcomes relative to supervision compliance, community linkages, or treatment engagement will vary among SAMI-SO individuals.

Directions for Future Research

Several hypotheses flow from the observations made here and would benefit from further inquiry. First, with respect to the age findings, we hypothesize that the younger a SAMI-SO individual is, the more community-based linkage services he needs to stabilize while on active community supervision. Second, in measuring mental health classifications on persistence, we hypothesize that the difference in persistent incidents between C1s and C2s is the result of C1s being identified early in prison or after being released to community supervision, which results in a referral and active treatment involvement in the community. Early identification, referral, and active treatment involvement appears to positively influence the number of persistent incidents involving the SAMI-SO population during an active period of community supervision. Third, in measuring housing classification on persistent incidents, we hypothesize that the difference in persistent incidents relative to the housing classifications are a result of the stability versus instability associated with the three housing classifications. That is, SAMI-SO individuals who are residing in a permanent residence are more likely to be mentally, emotionally, and socially stable than SAMI-SO individuals who are residing in either temporary housing or a homeless shelter, which is less stable and consistent. Fourth, in measuring positive drug screens or intoximeter tests produced by SAMI-SO individuals on persistence, we hypothesize that the difference in persistent incidents may be closely associated with active treatment involvement, supervision status, and other predictor variables that may have a positive impact on maintaining desistance.

Finally, future studies could broaden desistance research by incorporating other statistical methods, such as surveys involving both SAMI-SO individuals and community supervision staff. Surveying SAMI-SO individuals while they are on active community supervision could help to identify specific needs that, once addressed, could improve their ability to function normally in society. In addition, surveying community supervision staff could assist in the development of policies or procedures that help them to become more effective community supervisors with hard-to-supervise SAMI-SO individuals. Future research does not necessarily have to focus on the 14 variables identified in this study, because there are other areas or markers in desistance research that could be examined. These areas of interest could be supervision relationships, cultural transference, transference associated with SAMI-SO individuals that have an alternate lifestyle, treatment provider and patient relationships, and medication and treatment compliance concerns.

DOI: 10.1002/j.2161-1874.2014.00020.x

Received 09/24/12

Revised 12/27/12

Accepted 01/23/13


Abramsky, S., & Fellner, J. (2003). Ill-equipped: U.S. prisons and offenders with mental illness. In J. Saunders (Ed.), Human rights watch (pp. 1-201). New York, NY: Human Rights Watch.

Adams, K., & Ferrandino, J. (2008). Managing mentally ill inmates in prisons. Criminal justice and Behavior, 38, 913-917.

Co-occurring disorders, (n.d.). In Psychology Today diagnosis dictionary. Retrieved from

Delaney, R. B. (2011). A retrospective study on the relationship among social controls and individual factors as indicators in predicting desistance or persistence in the substance abusing mentally ill supervised offender population. Dissertation Abstracts International: Section B. Sciences and Engineering, 72(10), 6464.

Draine, J., & Solomon, P. (2001). Threats of incarceration in a psychiatric and parole service. American Journal of Orthopsychiatry, 71, 262-267.

Farabee, D., & Shen, H. (2004). Antipsychotic medication adherence, cocaine use, and recidivism among a parolee sample. Behavioral Science and the Law, 22, 467-476.

Fraenkel, J. R., & Wallen, N. E. (2003). How to design and evaluate research in education (5th ed.). New York, NY: McGraw-Hill.

Gagliardi, G. J., Lovell, D., Peterson, P. D., & Jemelka, R. (2004). Forecasting recidivism in mentally ill offenders released from prison. Law and Human Behavior, 28, 133-155.

Groom, B. (1999). Handling the triple whammy: Serious mental illness, substance abuse and criminal behavior. Corrections Today, 61, 114-119.

Ho, A. P., Tsuang, J. W., Liberman, R. P., Wang, R., Wilkins, J. N., Eckman, T. A., ... Shaner, A. L. (1999). Achieving effective treatments of patients with chronic psychotic illness and comorbid substance dependence. American Journal of Psychiatry, 156, 1765-1770.

James, D. J., & Glaze, L. E. (2006). Mental health problems of prison and jail inmates. Washington, DC: U.S. Department of Justice, Office of Justice Programs. Retrieved from

Klerman, G. L. (1977). Better but not well: Social and ethical issues in the deinstitutionalization of the mentally ill. Schizophrenia Bulletin, 3, 617-631.

Lamb, R. H., Weinberger, L. E., & Gross, B. H. (2004). Mentally ill persons in the criminal justice system: Some perspectives. Psychiatric Quarterly, 75, 107-126.

Laub, J. H., & Sampson, R. J. (2001). Understanding desistance from crime. Crime and Justice, 28, 1-69.

Lurigio, A. J. (2001). Effective services for parolees with mental illness. Crime & Delinquency, 47, 446-161.

Maloney, M. P., Ward, M. P., & Jackson, C. M. (2003). Study reveals that more mentally ill offenders are entering jail. Corrections Today, 65, 100-103.

Morrissey, J., Meyer, P., & Cuddeback, G. (2007). Extending assertive community treatment to criminal justice settings: Origins, current evidence, and future directions. Community Mental Health Journal, 43, 527-544.

National Research Council of the National Academies. (2007). Parole, desistance from crime, and community integration. Washington, DC: The National Academies Press.

Skeem, J. L., Emke-Francis, P., & Louden, J. E. (2006). Probation, mental health and mandated treatment. Criminal Justice and Behavior, 33, 158-184.

Slate, R. N., Roskes, E., Feldman, R., & Baerga, M. (2003). Doing justice for mental health and society: Federal probation and pretrial services officers as mental health specialists. Federal Probation, 67, 1-10.

Steven, R. (2000). The deinstitutionalization of the mentally ill and growth in the U.S. prison population: 1971 to 1996. University of California, Berkley. Retrieved from http://socrates.berkeley. edu/~raphael/raphael2000.pdf

Telias, D. (2001). Dual diagnosis. International Journal of Psychosocial Rehabilitation, 5, 101-110.

Urban Institute. (2008). Evaluation of the Ohio Department of Rehabilitation and Correction and Corporation for Supportive Housing's Pilot Program. Washington, DC: Urban Institute. Retrieved from

U.S. Department of Health and Human Services. (2012). USDHHS Report: Interventions for adults with serious mental illness who are involved with the criminal justice system. Rockville, MD: Agency for Healthcare Research and Quality. Retrieved from reports/?pageaction=displayproduct&productid=1259#6499

Wang, E. W., Owens, R. M., Long, S. A., Diamond, P. M., & Smith, J. L. (2000). The effectiveness of rehabilitation with persistently violent male prisoners. International Journal of Offender Therapy and Comparative Criminology, 44, 505-514.

Rodney B. Delaney, John M. Laux, Nick J. Piazza, Martin H. Ritchie, Department of Counselor Education, and Morris Jenkins, Department of Criminal Justice, all at The University of Toledo. This article is based on the dissertation research conducted by Rodney B. Delaney, now at the Department of Adult Psychiatry, MetroHealth Hospital, Cleveland, Ohio. Correspondence concerning this article should be addressed to John M. Laux, Department of Counselor Education, The University of Toledo, MS 119, 2801 West Bancroft Street, Toledo, OH 43606 (e-mail:
Means, Standard Deviations, and Intercorrelations
for Persistent Incidents and Predictor Variables

Variable                   M        SD         1         2

Persistent incidents      1.39      1.74    .58 **    .27 **

Predictor variable
 1.   SL                  1.59      0.53     --       .37 **
 2.   MH case             1.83      0.38               --
 3.   Months             24.36     16.41
 4.   Community           1.08      0.27
 5.   Active TX           1.42      0.50
 6.   Age                40.24     10.30
 7.   Ethnicity           1.62      0.51
 8.   Years of ed        11.02      1.41
 9.   Employed            1.75      0.44
10.   Prev. MH TX         1.00      0.05
11.   MH class            1.43      0.50
12.   Housing             1.67      0.70
13.   Months incar.      56.03     61.27
14.   Positive drug       1.42      1.99

Variable                     3         4         5         6

Persistent incidents      .12 **    .26 **    .80 **

Predictor variable
 1.   SL                  .00 *     .21 **    .63 **
 2.   MH case             .05       .11 **    .26 **
 3.   Months               --       .06 *     .06 *
 4.   Community                      --       .33 **
 5.   Active TX                                --
 6.   Age                                                  --
 7.   Ethnicity
 8.   Years of ed
 9.   Employed
10.   Prev. MH TX
11.   MH class
12.   Housing
13.   Months incar.
14.   Positive drug

Variable                     7         8         9        10

Persistent incidents

Predictor variable
 1.   SL
 2.   MH case
 3.   Months
 4.   Community
 5.   Active TX
 6.   Age               -.15 **    .11 **    .16 **   -.02 *
 7.   Ethnicity           --      -.03      -.13 **   -.01
 8.   Years of ed                   --      -.15 **   -.07 *
 9.   Employed                                --       .03
10.   Prev. MH TX                                       --
11.   MH class
12.   Housing
13.   Months incar.
14.   Positive drug

Variable                    11        12        13        14

Persistent incidents

Predictor variable
 1.   SL
 2.   MH case
 3.   Months
 4.   Community
 5.   Active TX
 6.   Age
 7.   Ethnicity
 8.   Years of ed
 9.   Employed
10.   Prev. MH TX
11.   MH class             --      .09 *     -.05 *     .40 **
12.   Housing                       --        .02       .19 **
13.   Months incar.                            --      -.06
14.   Positive drug                                      --

Note. N = 886. SL = supervision level; MH case = mental
health case specialty; Months = months on supervision;
Community = community linkages; Active TX = active in
treatment; Years of ed = years of education; Prev. MH TX =
previous mental health treatment; MH class = mental health
classification; Housing = housing status; Months incar. =
months incarcerated; Positive drug = positive drug or

* p <  .05. ** p < .001.

Simultaneous Multiple Regression Analysis Summary
for Study Variables Predicting Persistence

Variable                                 B   SE B      P

Supervision level                     0.25    .08     .08 **
Mental health case specialty          0.04    .09     .08
Months on supervision                 0.07    .02     .07 **
Community linkages                   -0.04    .12    -.06
Active treatment                      1.78    .10     .51 **
Age                                  -0.08    .03    -.46 *
Ethnicity                            -0.05    .63    -.01
Years of education                   -0.05    .23     .03 *
Employed                              0.64    .77     .16
Previous mental health treatment      0.55    .65     .15
Mental health classification          0.97    .79     .28 **
Housing status                        0.25    .48     .10
Months incarcerated                   0.00    .01     .12
Positive drug or alcohol              0.11    .18     .13 **
Constant                             -3.13    .77

Note. N = 886. [R.sup.2] = .73; F(14, 871) = 170.68.
* p < .05. ** p < .001.
COPYRIGHT 2014 American Counseling Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Delaney, Rodney B.; Laux, John M.; Piazza, Nick J.; Ritchie, Martin H.; Jenkins, Morris
Publication:Journal of Addictions & Offender Counseling
Article Type:Report
Date:Apr 1, 2014
Previous Article:Evaluating addictions counseling programs: promoting best practices, accountability, and advocacy.
Next Article:Predicting criminal recidivism following drug court: implications for drug court practice and policy advocacy.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters