Roads Diverge: Long-Term Patterns of Relapse, Recidivism and Desistance for a Cohort of Drug-Involved Offenders (original) (raw)

Research indicates that a large percentage of inmates released from prison back into their communities will be rearrested. Particularly vulnerable are those who have past histories of drug addiction. Except for intensive experiences with long term aftercare programming, there appears to be very few programs that significantly increase the probability of (re)integrative success for ex-offenders attempting to become conforming members of society. Other evaluation studies examining the efficacy of particular programs in reducing recidivism and drug use (NIDA, 2012, Pendergast, 2009) have been promising, however, we still know very little about the underlying mechanisms that produce change in offending patterns. Unlike evaluation studies examining the efficacy of particular programs, the primary goal of this project was to increase our understanding about the mechanisms and processes of desistance from crime and drug use among current urban, largely minority and increasingly female criminal offenders. Using a mixed methods research design, this research follows former drug-involved offenders for over 20 years post release from prison. The project was guided by Paternoster and Bushway’s identity theory of desistance (2009), which relies on the concept of identity that is theorized to provide direction for an individual’s behavior. The identity theory of desistance emphasizes the individual identity as reflexive, interpretive, and as such, premised on human agency. METHODS The project features a multi-method design and unfolded in two phases. The sample for this study originated from a previous sample used to evaluate the efficacy of Therapeutic Communities (TC) in reducing recidivism and relapse for drug involved offenders in 1989. In Phase I of the present study, official arrest records were obtained for the original 1,250 offenders through 2008 from both official Delaware and NCIC data sources. From these data, race and gender specific offending trajectory models were estimated. In Phase II used the trajectories as a sampling frame to select 304 respondents for in-depth interviews. The goal of the interviews was to examine the processes and mechanisms that led to persistence or desistance from crime and drugs. MAJOR FINDINGS The NCIC data uncovered an average of 2 more arrests for cohort members compared to the Delaware data. However, the trajectory models that resulted from both sources of arrest data were essentially the same, with trajectory group membership across both data sources identical in over 95% of the cases. For the total sample, a five trajectory group model fit the data best, with three groups differentially increasing their rate of offending and then leveling off to near zero, and an additional two groups increasing at different rates but remaining more criminally active throughout the time period compared to the desisting groups. Supporting the identity theory of desistance, interviews revealed that the vast majority of offenders who had successfully desisted from both crime and drug use first transformed their “offender identity” into a “non-offender working identity.” This was true for both race and gender groups interviewed. This cognitive process was typically motivated by respondents realizing that if change did not occur, they would likely become what they feared, either dying an addict, dying in prison, dying alone, or some other horrible fate. To behaviorally conform to their new “non-offender” identity, respondents used various tools including changing their “people and places” by seeking out noncriminal ties and staying away from locations that triggered their drug use or criminal behavior. Although treatment usually did not result in immediate desistance for most, the vast majority who eventually “got clean” relied on the tools learned in treatment programs when they were ready to use them. Religion also was cited as a tool by many that was used to establish pro-social support networks that reinforced their new identities. Contrary to life-course theory, partnership and parenthood did not appear to be “turning points” for the majority of our respondents. However, when they were ready to get clean, rekindling relationships with adult children or grandchildren was an extremely important factor for many of our respondents. Similarly, while getting a good job upon release did not deter the majority of addicts from relapse upon release from prison, it is clear that finding stable employment that paid a living wage is extremely difficult for this sample of drug involved offenders and that deciding to get clean and/or maintaining a “non-offender” identity is more difficult when access to a meaningful quality of employment is beyond reach or nonexistent.