Tool to predict recidivism in federal inmates could make more prisoners eligible for early release

Passed in 2018, the First Step Act sought to address re-entry challenges for inmates in the federal prison system. The legislation called for developing an assessment tool to identify inmates for release who had the lowest likelihood of recidivism. A new study assessed how the tool was developed and is used, finding that a greater proportion of inmates could reduce their risk and become eligible for early release over time if they participated in a re-entry program and did not incur infractions. This finding has implications for efforts to reduce prison populations during the COVID-19 pandemic.

The study, by researchers at the University of Nebraska at Omaha, Baylor University, the Federal Bureau of Prisons, and Washington State University, appears in

Justice Quarterly

, a publication of the Academy of Criminal Justice Sciences.

“This federal initiative represents a substantial opportunity to reverse the tide of a decades-long trend of growing rates of incarceration,” according to Zachary Hamilton, associate professor of criminology and criminal justice at the University of Nebraska at Omaha, who led the study.

The First Step Act sought to reduce recidivism in federal prisons by giving lower-risk individuals the opportunity to earn time credits and participate in programs to improve success of re-entry into their communities. In this study, researchers determined that a tool being used by the Bureau of Prisons to identify inmates with the lowest risk of violent and nonviolent recidivism included items that did not predict recidivism.

They expanded the existing tool to create a new one, called PATTERN (the Prisoner Assessment Tool Targeting Estimated Risk and Needs). PATTERN, launched in January 2020, and was designed to be gender-responsive and customized to the Bureau of Prisons population. Among the factors measured by PATTERN are age, prior convictions, work programming, drug treatment while incarcerated, criminal history, history of escapes, and education.

Using PATTERN, roughly half the population sampled was identified as being immediately eligible for early-release time credits. A substantial portion of the other half could become eligible if they participated in re-entry programs, did not incur infractions, and exhibited positive behavior change when reassessed, the study found. In all, almost 72 percent of men and about 96 percent of women could be eligible for early-release credits during the course of their incarceration, according to the study’s findings.

“If more programming to boost re-entry success were available in federal prisons and more prisoners participated in this programming, especially those in high-risk categories, the proportion of offenders eligible for early-release credits would rise,” suggests Grant Duwe, non-resident scholar at Baylor University and director of research at the Minnesota Department of Corrections.

The study concluded that PATTERN demonstrates one of the highest levels of predictive performance, outpacing that of all contemporary assessments researchers reviewed. In addition, PATTERN was validated on the same population on which it was developed and tested on a large development sample, yielding more stable and reliable estimates. Finally, the tool was tailored to specific outcomes (general and violent recidivism) and groups (males and females), which improves its predictive performance. Tests for bias revealed that PATTERN further reduced race/ethnicity disproportionality.

The researchers identified limitations of the study, including constraints regarding the pool of items used to predict recidivism and tight timing due to a requirement to meet legislative schedules. They recommend that the Bureau of Prisons continue to expand the standardized set of needs-based items collected for all offenders, adding items to improve PATTERN’s accuracy and potential for case management. The authors also point to the importance of creating optimized tools for correctional systems, making use of agency-specific data to create strategies that will reduce the prison population while prioritizing public safety.

###

The research was supported by the National Institute of Justice.