New system assesses parolee risks

Current Oregon tool has been used since the '70s and has many critics



PORTLAND — Oregon probation and parole officials will begin using a new tool next month to help them better predict the chances that convicted felons will commit new crimes after they’re released from prison.

The system is called the Oregon Public Safety Checklist and is similar to the actuarial tools that insurance companies use to set premiums. It will make an estimate based on the person’s age, gender and criminal history, replacing an outdated and less-reliable assessment that the state has used for more than 30 years.

A pilot study is under way to determine how the new risk model can be used by judges at sentencing, or by prosecutors and defense lawyers negotiating a plea deal. The Marion County jail is using the new tool to help decide which inmates to release when there’s overcrowding. Multnomah County is considering using it to decide whom to recommend for release from custody pending trial.

Yet prosecutors are concerned that the tool is faulty and misidentifies dangerous offenders as low risk. They point to the low-risk scores the tool gives convicted Woodburn bank bomber Bruce Turnidge or predatory sex offender Jeffrey Cutlip, who is now accused of two homicides in Portland in the 1970s.

“Everyone in the business of supervising offenders has said an actuarial tool is really the cornerstone in the modern world,” Craig Prins, executive director of the Oregon Criminal Justice Commission, told The Oregonian.

Since 1979, the state has relied on something called the Oregon Case Management System to measure the risk that offenders may commit new felonies within three years of their release from prison or their completion of probation. But its reliability has been low compared with screening tools used in other states. It also has been time-consuming, requiring county parole or probation officers to spend up to an hour researching an offender’s prior criminal history.

Under the new method, a parole or probation officer or even a judge types information identifying the offender into a website, and the computer generates a quick analysis of the offender’s demographic information and criminal history, culled from the state’s electronic databases. It produces three scores: the risk the offender will be convicted of a new felony within three years of release; the person’s risk of being arrested for a property crime within five to 10 years of release; and his or her risk of arrest for a crime against a person within five to 10 years of release.

State officials caution that the new tool has its limitations.

It does not include out-of-state or federal arrests or convictions. And it can’t be used to measure risk levels for first-time offenders, who have no conviction history. It’s only an initial screen and not the best tool for evaluating specialized caseloads such as sex offenders.

Statistically, the state found the new checklist is more reliable. The old case management system was accurate about 63 percent of the time; the new checklist has a 78 percent accuracy rate, state officials said.