Skip to content
A Cleveland Municipal Court employee sorts through defendant files scored with risk-assessment software. (Dake Kang/​The Associated Press)

Toledo, Ohio — Officials in Lucas County released new data showing that more individuals are being released on their own recognizance, pretrial crime is down, and significantly more people are reporting for court hearings since the county began using a data-driven pretrial risk assessment tool known as the Public Safety Assessment (PSA). The PSA is being used, or is in the process of being implemented, in 30 cities and states across the country to help judges make consistent and risk-based decisions about whether to release or detain defendants prior to trial. Specifically, the Lucas County data show that: 

  • The number of releases without the need for bail nearly doubled. The percentage of pretrial defendants released by the court on their own recognizance, meaning they did not have to post bail, jumped from 14 percent before the county began using PSA to almost 28 percent today.
  • Pretrial crime is down. The percentage of pretrial defendants arrested for other crimes while out on release has been cut in half — from 20 percent before the county began using the PSA to 10 percent today. In addition, the percentage of pretrial defendants arrested for violent crimes while out on release has decreased — from 5 percent before the county began using the PSA to 3 percent today.
  • More defendants are returning to court. The percentage of pretrial defendants who skipped their court date has been dramatically reduced — from 41 percent before the county began using the PSA to 29 percent today.
  • The PSA is race and gender neutral. In Lucas County, black and white defendants are being released at equal rates. Unlike some risk assessments, the PSA does not take into account factors that some have argued could be discriminatory such as a person’s ethnic background, income, level of education, employment status, or neighborhood.

The Public Safety Assessment is an important part of our effort to ensure that we are using our jail to house the right people — those who pose a risk to public safety or of not returning to court,” Judge Gene Zmuda of the Lucas County Court of Common Pleas explained. It is helping our judges make more effective decisions about who should be in jail and who can be safely released.”

Lucas County first began using the PSA in January 2015 as part of an integrated response to a federal court order capping the jail population at 403 inmates. The Lucas County Court of Common Pleas sought to ensure that the jail’s limited space was being occupied by the defendants who were the most likely to commit new crimes or skip court if released — and that lower-risk defendants were being supervised in the community in ways that are consistent with the risk posed.

In addition to assisting judges in determining which defendants to release and which to detain before trial, the PSA is helping Lucas County Sheriff John Tharp and the courts in Lucas County manage and reduce the jail population at a level that is at or below the federal cap. Prior to implementation of the PSA, many defendants charged with lower-level crimes were released immediately after booking because the jail did not have space for them. Others were held in jail for some time but were then let out under an emergency release” protocol in order to keep the jail population below the cap. Because these releases were based primarily on what a defendant was charged with, rather than the level of risk posed, many defendants with a long record of reoffending or skipping court were quickly let out, while lower-risk defendants remained in jail. As a result, more than four of every 10 defendants didn’t return for their court dates, and one in five was rearrested while their original case was still pending. With the PSA in place, Lucas County is now able to make these decisions based on risk and has seen the dramatic reductions in pretrial crime, violence, and missed court dates cited above.

Working with Dr. Marie VanNostrand, one of the nation’s leading researchers and practitioners in pretrial criminal justice, Lucas County has overhauled its pretrial system. Cases are now being processed more efficiently, with the number of cases resolved at a defendant’s first appearance more than doubling from 8 percent to 17 percent. Because of the system improvements, judges are now able to designate low-risk defendants for release and high-risk defendants for detention (with an attendant increase in pretrial detention from 17 percent of defendants to 23 percent), rather than having the emergency-release protocol drive these outcomes. As a result, the number of emergency releases has plummeted from 39 percent of pretrial defendants to 5 percent.

The PSA was developed by the Laura and John Arnold Foundation (LJAF), a philanthropic organization that made the risk assessment available free of charge. It provides information that judges can consider when deciding whether to release or detain a defendant prior to trial. The PSA uses neutral, reliable data to produce two risk scores: one predicting the likelihood that an individual will commit a new crime if released pending trial, and another predicting the likelihood that he will fail to return for a future court hearing. Scores fall on a scale of one to six, with higher scores indicating a greater level of risk. The PSA also flags defendants that it calculates present an elevated risk of committing a violent crime.

There are early indications from other communities that the tool is helping to reduce crime and jail populations in those locations, but the results in Lucas County are especially striking, according to LJAF Vice President of Criminal Justice Matt Alsdorf.

We applaud the criminal justice team in Lucas County for their leadership and foresight in identifying ways that the Public Safety Assessment could help manage the Lucas County Jail population,” Mr. Alsdorf explained. The results are very positive and are a further indication that the tool is making it possible for communities to address their specific needs in ways that make the system more effective, efficient, and fair.”

The PSA was created using the largest, most diverse set of pretrial records ever assembled — 1.5 million cases from approximately 300 jurisdictions across the United States. Researchers analyzed the data and isolated factors that most often exist for defendants who commit a new crime, commit a violent crime, or fail to return to court if released before trial.

The factors are: 

  • Whether the current offense is violent
  • Whether the person had a pending charge at the time of the current offense
  • Whether the person has a prior misdemeanor conviction
  • Whether the person has a prior felony conviction
  • Whether the person has prior convictions for violent crimes
  • The person’s age at the time of arrest
  • How many times the person failed to appear at a pretrial hearing in the last two years
  • Whether the person failed to appear at a pretrial hearing more than two years ago
  • Whether the person has previously been sentenced to incarceration

Though these neutral factors can help judges gauge the risk that a defendant poses, they do not impede a judge’s discretion or authority in any way. The decision about whether to release or detain a defendant always rests with the judge regardless of the scores produced by the risk assessment.