The purpose of the study was to answer the following three questions regarding the three state sponsored alcohol treatment programs:
Do the alcohol treatment programs adhere to evidence based practices that have documented success in addressing substance abuse?
Do the alcohol treatment programs deliver treatment in a manner consistent with program protocols thereby demonstrating program fidelity?
Are alcohol treatment program graduates more or less likely to re-offend compared to a comparison group of non-program participants?
Montana, Ohio, and Texas were selected for the following reasons:
Each of these states operates a DUI/DWI program targeted directly at drinking and driving violations within at least one of their correctional institutions.
Each of these states delivers alcohol recovery services vested in the efficacy of cognitive-behavioral change, and normative and representative programming which facilitates generalization of the finds.
These states agreed to participate in this study ensuring inmates, staff, and administrators availability for recorded interviews, admittance to the treatment environment for observation of the program, and full and complete access to electronic data records allowing outcome analysis.
The qualitative component of this study consisted of a review of the treatment modality materials describing program content, the services delivery plan, and on-site interviews to evaluate the degree of evidence based practice represented in the program designs and level of program fidelity. On-site in-depth interviews were carried out with facility administration and program staff, focus group interviews with a sample of treatment group participants, and observation of treatment services at the sites. Montana treatment program was visited twice during the first year of the study and a follow-up visit in the second year of the study. Ohio treatment program was visited only in the first year of the study, two visits each to the treatment facility and post-treatment at a halfway house adding a longitudinal qualitative element. For Texas access to the treatment program was denied by the Texas Department of Criminal Justice after a lengthy review.
The quantitative data collection component addressed outcome performance based on a quasi-experimental design comparing the recidivism rates of treated offenders to those of a non-treated comparison group. Data on recidivism and individual level characteristics were provided by each state's correctional agency.
Montana qualitative data
Participants were selected randomly, five participants per session totaling 30 interviewees per visit for a total of 60 interviewees in year one and 30 in year two of the study.
Montana Department of Corrections (MDC) data
From July 2006 to May 2012 1,275 individuals entered the program, of which 908 were assigned to the West facility. All East facility participants were removed from further consideration to match the qualitative element of this study, which was conducted exclusively at the West facility. The 908 West participants were categorized into two separate groups for analysis purposes. Group 1 includes individuals who received some treatment, but did not complete the treatment program (n = 114). Group 2 represents all individuals who graduated from the program (n = 794). After removing cases with missing information, group 1 contained 106 cases and group 2 760 cases.
Ohio qualitative data
Two visits to the correction center were made and the entire population of the program participants, seven and five, were interviewed. Another 10 participants were interviewed following post-treatment and release to a halfway house in two separate visits.
Ohio Department of Rehabilitation and Corrections (ODRC) data
The Ohio Department of Rehabilitation and Corrections (ODRC) provided data on program participants and other ODRC members under supervision between the Fall of 2009 and September 2012. Group 1 were DUI offenders under the supervision of the ODRC, but based on program eligibility criteria were deemed ineligible for program participation. Group 1 consisted of 1,192 individuals for which information on relevant variables was available in 1,137 cases (4.6% missing data). Group 2 members met the admission criteria, entered the program, and successfully completed the 90-day treatment plan. Group 2 consisted of 46 individuals for which information on relevant variables was available in 45 cases (2.2% missing data). Group 3 includes ODRC inmates that met the criteria for treatment, but were not enrolled in the program and ultimately did not receive any treatment. Group 3 consisted of 112 individuals for which information on relevant variables was available in 106 cases (5.4% missing data).
Texas Department of Criminal Justice (TDCJ) data
Data was provided by the Texas Department of Criminal Justice (TDCJ) on three different groups of individuals from March 2008 to December 2011. Group 1 were inmates meeting criteria for inclusion, but not selected for treatment based on TDCJ assessment. Group 2 were participants that successfully completed the program. Group 3 were individuals who began, but did not complete treatment programming. The TDCJ originally provided information on 4,080 offenders across all three of these groups during the study period, of which 2,840 graduated from the treatment program.
Qualitative data: Male offenders, staff, and administrator in state sponsored alcohol specific treatment programs in prisons in Montana, Ohio, and Texas.
Quantitative data: Offenders considered or accepted for state sponsored alcohol specific treatment programs in prisons located in Montana, Ohio, and Texas.
Ohio Department of Rehabilitation and Corrections
Texas Department of Criminal Justice
Montana Department of Corrections
administrative records data
Montana Department of Corrections (MDC) data (n = 1,275) contains 61 variables on: offender unique id; treatment facility; whether case is valid; treatment location; indicator of program completion; indicator of program failure; offender age at start of program, offender gender, race, and marriage status; offender criminal history counts, incident counts, and whether they had a violent criminal history; number of DUI counts pre-treatment and DUI offenses post treatment; number of in prison infractions; time to program; time in program; time from program; time in MDC; time at risk; end of study period; time to failure; reason for failure; placement after failure; MDC and program admit, exit, and return dates; and total count of criminal history, criminal history pre-treatment, criminal history post-treatment, incidents, violent offenses, violent offenses pre-treatment, violent offenses post-treatment, DUI, and infractions.
Ohio Department of Rehabilitation and Corrections (ODRC) data (n = 1,350) contains 53 variables on offender unique id, group, program participant, more than one year on release, more than two years on release, program start date, successful discharge date, exit date from ODRC, end date of data, days from release to end date of data, first return date, number of days out from ODRC until recidivism, return type, type of violation, days from ODRC entrance to program entrance, days in program, time in prison, days in post treatment to release from ODRC, any violation, violation within one year, violation within two years, offender race, offender sex, offender education, offender marriage status, offender criminal history counts, offender infraction counts, and offender grouping.
Texas Department of Criminal Justice (TDCJ) data (n = 4,080) contains 122 variables on: offender unique id; group; offender's program status at end of study; TDCJ entrance date; program entrance date; days from TDCJ entrance to program entrance; program completion date; days in program; departure date; time in prison with no treatment; days from post treatment to release from TDCJ; end date of data; if more than one year on release; if more than two years on release; any violations; release codes such as probation, parole, mandatory supervision release, hold release, or died; days from release to first, second, third, fourth, fifth, and sixth violation; allegations for administrative decision, disposition date, disposition code, and codes for administrative decision for first, second, third, fourth, fifth, and sixth violation; transfer date and return to TDCJ for first, second, third, fourth, fifth, and sixth violation; reason for entry to program; new receive on this charge; new longer sentence on previous charge; whether probation, parole, or mandatory supervision revocation; offender age and age at release; offender race, marital status, number of children, and education level; number of previous TDCJ visits; whether first, second, or third or more visits to TDCJ; number of major and minor infractions; custody status; program change date; other type of release; and whether violation within one or two years