Evaluation of Prison-Based Drug Treatment in Pennsylvania, 2000-2001 (ICPSR 3540)

Data Collection Notes View help for Data Collection Notes

Hide

Study Purpose View help for Study Purpose

The purpose of this study was to examine multiple treatment process measures and post-release outcomes for inmates who participated in Therapeutic Community (TC) drug treatment programs or comparison groups provided by the Pennsylvania Department of Corrections at five state prisons. The project attempted to examine more closely the relationships among inmate characteristics, treatment process, and treatment outcomes than previous studies had done in order to explore critical issues in prison-based drug treatment programming and policies. The study aimed to address gaps in the current knowledge about prison treatment programs. There were several questions that the researchers focused on: (1) How was quality of program implementation related to treatment outcomes? (2) How were drug and alcohol needs assessments conducted, and how did needs assessments influence treatment process and outcomes? (3) Were inmates with different levels and types of need matched with appropriate treatment? (4) To what degree did appropriate assessment and program placement moderate treatment effects? (5) What kinds of inmates were best suited for Therapeutic Communities? (6) What magnitude of treatment effect was to be expected from TC if appropriate extraneous influences were controlled for in analyses?

Study Design View help for Study Design

Researchers examined in-treatment measures and multiple post-release outcomes for inmates who participated in TC drug treatment programs or comparison groups at five state prisons, Graterford, Houtzdale, Cresson, Waymart, and Huntingdon. Matched comparison groups were made up of TC-eligible inmates who participated in less intensive forms of treatment (e.g., short-term drug education and outpatient treatment groups) due to a shortage of intensive treatment slots at the five institutions. The comparison groups were constructed based upon known predictors such as drug dependency, need for treatment, and criminal history. Process and outcome measures incorporated a range of institutional measures (e.g., misconducts), intermediate measures (e.g., attitudinal and behavioral change, participation in treatment), and post-release measures (e.g., drug relapse, rearrest and reincarceration, employment, levels of parole supervision). In order to create adequate comparison groups, formal classification assessments and drug and alcohol assessments were conducted on all inmates. The average scores on the Pennsylvania Department of Corrections Screening Instrument (PACSI) and the Texas Christian University (TCU) Drug Screen were examined for inmates in each program to determine how many inmates fell into low, medium, and high need categories. Because a shortage of space precluded intensively treating all those who were assessed with a high need for alcohol or other drug (AOD) treatment, a large pool of eligible offenders assigned to less intensive forms of treatment (e.g., outpatient treatment, drug education, or ancillary groups) was accessible for inclusion in the matched comparison sample. Included in the treatment sample were all current TC residents as of January 1, 2000. New subjects were added to the study as they were admitted to treatment programs. For each inmate admitted to an AOD treatment program during the study period, researchers collected the treatment start date, the discharge date, and the treatment outcome. Between January 1 and November 30, 2000, data on all inmates admitted to or discharged from alcohol or drug treatment programs were collected on a monthly basis. Monthly tracking was continued throughout the study to determine treatment outcomes (e.g., successful vs. unsuccessful). TC clients were asked to complete additional self-report measures that tapped psychological constructs and inmate perceptions of the treatment experience, and TC counselors were asked to complete periodic reassessments of each inmate's participation in treatment. TC clients were asked to complete the Resident Evaluation of Self and Treatment (REST) form and TC counselors were asked to complete the Counselor Rating of Client (CRC) form. Self-reports of treatment process and psychological functioning were gathered within 30 days after admission, again after six months, again at the end of 12 months, and again at discharge if the inmate remained in TC longer than 12 months. Counselor ratings of inmate participation in treatment were similarly gathered one month, six months, and 12 months following admission to treatment. Standardized drug scores were calculated based on the inmate's TCU Drug Screen score, if available, or the inmate's PACSI score, if the TCU was not available. Because the TCU is a 0-9 scale, and the PACSI is a 0-10 scale, scores were transformed into standardized z-scores. After release, both treatment and comparison groups were tracked over time to monitor rearrest, reincarceration, drug use, and employment. Inmates were informed that their participation in the research was voluntary, and they were asked to sign a Subject Consent Form. The researchers retained the responsibility to maintain the security and confidentiality of all information obtained from the inmates during all data collection and analyses. At the time of the creation of this dataset, 247 TC inmates and 809 Comparison inmates had been released from prison, with follow-up periods extending up to 18 months.

Sample View help for Sample

Inmates in five Pennsylvania prisons that were placed in Therapeutic Community (TC) drug treatment programs made up the TC treatment groups. Comparison treatment groups were made up of TC-eligible inmates who participated in less intensive forms of treatment (e.g., short-term drug education and outpatient treatment groups) due to a shortage of intensive treatment slots at the five institutions. Included in the sample were all current TC residents as of January 1, 2000. Between January 1 and November 30, 2000, new subjects were added to the study as they were admitted to treatment programs.

Universe View help for Universe

All inmates in five Pennsylvania state prisons who had a high need for alcohol and drug treatment in 2000.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Description of Variables View help for Description of Variables

Measures can be broken down into the following four categories and their sources: (1) Inmates Background Factors were collected from the Pennsylvania Additive Classification System (PACT), the Pennsylvania Department of Corrections Screening Instrument (PACSI), and the TCU (Texas Christian University) Drug Screen. Variables for include offenders' age, psychological status such as depression and anxiety, and assessed need for alcohol and other drug treatment. (2) Institutional Indicators: Impacts Internal to the Prison Environment were collected from the Department of Corrections Misconduct Database, research and program records, and TCU Resident Evaluation of Self and Treatment (REST) forms. Variables include misconducts and levels of program participation. (3) Intermediate or "Proximal" Outcomes: Reductions in Risk for Drug Use and Criminal Behavior were collected from research and program records, TCU Counselor Rating of Client (CRC) forms, and TCU Resident Evaluation of Self and Treatment (REST) forms. Variables include treatment process (e.g., length of time in treatment, degree of involvement in treatment), inmate ratings of treatment program features, participation in therapeutic groups, counselor attitudes and behavior, resident attitudes and behavior and counseling sessions, psychosocial functioning (self-esteem, depression, anxiety, decision-making), and social functioning (e.g., hostility, risk-taking, and social conformity). (4) Post-Release Indicators: Inmate Behavior Upon Release from Prison were collected from the Pennsylvania Board of Probation and Parole, Pennsylvania state police records provided by the Pennsylvania Commission on Crime and Delinquency (PCCD), and the Department of Corrections inmate records system. Variables include relapse to drug use (type of drug, frequency of use), rearrest or warrant issued (number and type of offenses, survival rate), reincarceration (number and type of offenses, survival rate), parole violation or revocation, and post-release employment status (e.g., full-time, part-time, unemployed and able to work, unemployed and unable to work).

Response Rates View help for Response Rates

Response rates for TC samples (i.e., completion of Consent, Drug Screen and REST forms) were 86 percent. The response rates for inmates in the Comparison samples (Subject Consent forms and TCU Drug Screen forms) were 51 percent.

Presence of Common Scales View help for Presence of Common Scales

Likert scales were used in the TCU Resident Evaluation of Self and Treatment (REST) form and the TCU Counselor Rating of Client (CRC) form. Also used in the study are the Pennsylvania Additive Classification System (PACT), the Pennsylvania Department of Corrections Screening Instrument (PACSI), and the Texas Christian University (TCU) Drug Screen.

Hide

Original Release Date View help for Original Release Date

2003-06-19

Version History View help for Version History

Hide