Pretrial Risk Assessment

Risk Assessmet Header

What Is Pretrial Risk Assessment?

A pretrial risk assessment instrument or tool provides an objective analysis of whether an arrested person is likely to appear in court and not get rearrested if released before trial. Using a pretrial risk assessment tool reduces bias and subjectivity in court decisions about who should be detained before trial and which conditions, if any, should be required of those who are released.

Most states authorize courts to consider only two factors when deciding whether an arrested person should be released or detained: court appearance and public safety.1

A typical risk assessment instrument is a questionnaire or form that identifies information that categorizes defendants as having lower, medium, or higher risk for pretrial failure. Most arrested people fall into lower and medium risk categories, as the chart below illustrates.

Risk Category Public Safety Rate Court Appearance Rate Percent of Defendants
 1 (lower) 95% 95% 20%
 2 85% 85% 49%
 3 77% 77% 23%
 4 (higher) 51% 51% 8%
The Colorado Pretrial Assessment Tool (CPAT)—Executive Summary


How Does Pretrial Risk Assessment Work?

Developers of evidence-based pretrial risk assessment tools analyze data from recently closed cases to determine factors that correlate with pretrial failure. Many tools are developed using a specific county’s or state’s data; one tool—the Public Safety Assessment—was developed using data collected from a national sample.

Since some factors are more predictive than others, tools “weight” relevant factors when tallying the final score. Pretrial risk assessment tools should be re-tested on a regular basis to ensure they remain valid for the population with which they are used.

While no pretrial assessment tool can erase racial and ethnic disparities elsewhere in the criminal justice system, evidence-based assessments are designed to be race- and gender-neutral and regular testing can ensure they continue to have no disparate impact.

PJI recommends that jurisdictions select tool that are open to review, transparent about their factors and weighting, and race- and gender- neutral. They should also publish aggregate data on how their tool performs—including indicators of racially disparate impact.


How Should the Tools Be Used?

Results from pretrial risk assessments are meant to inform—not replace—the court’s discretionary decision making. Court officers—prosecutors, defenders, and judges—consider pretrial risk assessment results alongside other factors, such as the current charge and the arrested person’s individual circumstances.

The results of a pretrial risk assessment can also help courts determine any conditions that might be needed to assure that a released individual will appear in court and stay out of trouble. Best practices call for combining pretrial risk assessment tools with a release matrix that uses a person’s score and charges to suggest—not mandate—the least onerous conditions that will assure pretrial success.

Using the least onerous conditions is not only required by law—people are presumed innocent until proven otherwise—but it is also supported by research. Over-supervising lower- and medium-risk people during the pretrial period has been shown to make them more likely to be arrested and less likely to return to court.

Conditions of release may include

  • Providing current phone numbers for court date reminder texts
  • Phone or in-person check-ins
  • Curfews
  • Stay-away orders
  • Referrals to health assessments for voluntary services
  • Electronic monitoring (in rare cases)


What are the Limits of the Tools?

Pretrial risk assessment instruments should NOT be used to automatically detain or to predict likelihood of guilt. Nor should they be used in any trial-based or post-trial decisions. No tool predicts how an individual person will behave; they show only how that person compares to others who are similarly situated and provide evidence of how those others behaved in the past.

Although it may be tempting to combine pretrial risk assessment with a needs assessment, most experts agree this is inappropriate. Courts have limited authority to compel legally innocent people to participate mandated services, and many of these—substance use counseling, for example—require an admission of guilt that defies the fundamental presumption of innocence.


More Resources

For more information on the science of pretrial risk assessment, please see the following resources.

Issue Brief: Pretrial Risk Assessment - PJI 2015

Race and Risk Pretrial Assessment Overview - PJI 2016

Race and Gender Neutral Pretrial Risk Assessment, Release Recommendations, and Supervision: VPRAI and Praxis Revised - Luminosity 2016

A Discussion of Pretrial Risk Assessment: Perpetuating or Disrupting Racial Bias?

A Framework for Pretrial Justice: Essential Elements of an Effective Pretrial System and Agency - NIC 2017

Demystifying Risk Assessment, Key Principles and Controversies - CCI 2017

1. New York State courts are not authorized to consider danger when making pretrial detention decisions.