Pretrial Services

The use of smart and effective pretrial services can help reverse jail overcrowding by allowing more people who have been charged but not yet convicted of crimes to await trial or disposition in the community. For some, pretrial services provide extra support and monitoring to ensure a defendant’s successful release. The decision of pretrial release is a difficult one and there are no guarantees; however, evidence-based tools and tested practices are facilitating the successful release of people who have been charged with lower level offenses.

Risk assessment tools

When an individual has been arrested and charged with an offense, courts have to decide whether to release them. They must, in other words, assess risk. In the pretrial context, risk refers to the likelihood that, if released, an individual will either fail to appear in court or commit a crime while awaiting disposition.

Until recently, jurisdictions largely used cash bail or a monetary bond schedule or “consensus based” decision-making to determine the conditions of a defendant’s release before trial. Years of studies show that these strategies do not actually mitigate risk and disproportionately punish those who cannot afford bail. In recent years, some states and localities have turned to a more objective decision-making process by using risk assessment tools to guide pretrial release decisions. 

Pretrial risk assessment tools are similar in some ways to the predictive models used by insurance companies. Using data from thousands of prior cases, developers identify which factors have the strongest correlation with risk. Criminal justice risk assessments often include factors like the current charge, any prior offenses and whether someone failed to appear (FTA) in court in the past. Based on robust statistical correlations, developers then assign weights to each factor. That enables the calculation of a risk score, classifying a defendant as low, medium, or high risk if released. 

Various risk assessment models are in use, employing different sets of variables. Some of these are shown in the table below: 

Table 1: Factors in Pretrial Risk Assessment Tools

Virginia Colorado Kentucky Federal Florida Ohio
Current change
X
X
X
Pending change
X
X
X
X
Previous convictions (misdemeanor and/or felony)
X
X
X
X
Previous FTA
X
X
X
X
X
Violet conviction
X
X
Residency (length, ownership, contribute, etc.)
X
X
X
X
X
Employment/student status
X
X
X
X
Current/history of drug or alcohol abuse
X
X
X
X
X
Working phone
X
X
Age (current or at first arrest)
X
X
X
X
Active warrant
X
X
Mental health
X
X
On probation/parole
X
Education
X
Citizenship/foreign relations
X
Marital status
X
Previously incarcerated in jail or prison
X
X

Risk scores are meant not to replace judiciary discretion, but to act as an objective, guiding tool for pretrial decision makers. With these tools in use, more people have been let out of jail before trial while maintaining public safety.

 
"It's definitely changed the way I decide. When I was a young judge, I relied on intuition and community expectations. Now I know with verified information what is statistically likely."
—Kentucky Circuit Court Judge David Tapp on the Public Safety Assessment pretrial tool
 

Risk scores and racial bias

Risk assessment tools are intended to reduce bias in court decisions about pretrial detention. A three-year pilot study in Yakima, Washington using risk assessment to guide pretrial release decisions greatly reduced disparities for Latinos, as outlined in the Bail Reform section

While there are some examples of risk assessment tools reducing disparities, some fear that these algorithms perpetuate racial biases in the criminal justice system. A privately developed risk assessment tool in Florida was shown to perpetuate racial disparities among black and white defendants. A major concern is that some risk assessment tools rely on socioeconomic factors that correlate with race, such as education level, neighborhood of residence, and employment—a practice that is not recommended by experts. Additionally, because they are more heavily policed, black and Latino people are more likely to have prior arrests or convictions that will increase their risk scores. 

Taking these challenges into account, a recent publication by Harvard Law School recommends that policy-makers prioritize transparency, accountability and oversight, and a holistic approach to fairness in considering risk assessment tools. If transparent and well monitored, risk assessment tools have the potential to reduce racial disparities in the criminal justice system by using data rather than subjective decision-making which is much more susceptible to implicit bias. 

Harvard’s Primer on
Bail Reform
Benefits, challenges and recommendations for
risk assessment.
See Page 20 of this primer.

 
The Laura and John Arnold Foundation’s Public Safety Assessment (PSA) tool was developed using data from 1.5 million cases from about 300 jurisdictions. The tool uses nine factors to separately evaluate an individual’s likelihood of failing to appear, committing new criminal activity, and committing new violent criminal activity.

The tool's weighting factor is transparent, and the foundation strives to continually evaluate and improve PSA while building a community of learning.
 

Charlotte-Mecklenburg, North Carolina 

Pretrial Services in Charlotte-Mecklenburg, North Carolina began using the Arnold Foundation’s Public Safety Assessment (PSA) tool in 2014. In two years, the county jail population decreased by 40 percent with no increase in reported crimes.

Kentucky

Since 2013, Kentucky has used the Arnold Foundation’s PSA. In the first six months, the foundation reports re-arrests of released individuals decreased to 8.5 percent from 10 percent, with 70 percent of inmates released, up from 68 percent. 

Pretrial supervision

If an individual is considered to pose some risk if released before trial, pretrial service divisions can use supervision strategies to mitigate risk while allowing defendants to get back to their daily lives. It is important to note that individuals who have been deemed low risk do not need to additional supervision and these practices should be reserved higher risk individuals. The Pretrial Justice Institute (PJI) defines pretrial supervision as a law enforcement program designed to ensure that an arrested person released before trial will both appear in court and pose no threat to public safety

If monitoring is deemed necessary, pretrial supervision allows people to get out of jail with accountability and support to improve their outcomes while in the community. Pretrial supervision can include check-ins (in person or over the phone), drug testing and, at times, electronic monitoring. Pretrial services also use tools such as court date reminders, which are helpful for even lower risk individuals. Pretrial supervision allows jurisdictions to release and monitor people who pose no threat to public safety and who would benefit from the support. Many local jurisdictions partner with probation agencies that already perform similar monitoring duties to deliver their pretrial services while other jurisdictions have established their own pretrial services agencies. 

Many jurisdictions have had great success in reducing jail populations through pretrial services. 

Smart Pretrial

PJI and the Bureau of Justice Assistance (BJA) launched the Smart Pretrial initiative to create more fair and effective pretrial practices. The Smart Pretrial tests innovations in six counties: Alachua County, Florida; Lucas County, Ohio; Marion County, Indiana; Mecklenburg County, North Carolina; Santa Clara County, California; and San Francisco, California.

 
PJI’s Smart Pretrial initiative prioritizes three goals: public safety, court appearance, and the appropriate use of release, release conditions, detention, and public resources.
 

New Jersey

In 2017, New Jersey overhauled its pretrial justice system to largely eliminate the practice of cash bail. Learn about the state’s successes and challenges in the Bail Reform section

Santa Clara County, California

The county’s Office of Pretrial Services oversees the release of defendants awaiting trial and has roots dating to 1969

The program is considered a great success. A Sacramento Bee op-ed by Erwin Chemerinsky, dean of the University of California Berkeley Law School, reports that 95 percent of those released appear in court and 99 percent remain arrest-free. 

Criminal Justice Reform Information Center
Information about New Jersey’s Pretrial Services

Santa Clara’s Pretrial Services facilitates two types of release programs: Release on Own Recognizance (ROR) and Supervised Own Recognizance (SOR). 

ROR permits defendants to be released from jail without paying bail, contingent on their promise to appear in court. As people accused of a crime are presumed innocent until proven guilty, ROR was a long-standing norm before the rapid growth of pretrial detention. In Santa Clara, ROR release is authorized for people with strong ties to the community who pose little flight risk and no danger to public safety. 

SOR release is a program unique to Santa Clara County that permits defendants who are not granted ROR to be released from jail with supervision by the Office of Pretrial Services. 

In December 2017, Santa Clara County’s offices of Pretrial Services, the public defender, the sheriff and Re-entry Services collaborated to launch the “No Cost Release” multilingual campaign to inform detainees and their families about pretrial release programs, criminal defense attorneys and re-entry services, all of which are available free of charge. The campaign emphasizes the options offered by the Office of Pretrial Services, which allows defendants to be released from jail pending trial if they participate in an ROR or SOR program.