Practice Profile: Pretrial Interventions for Ensuring Appearance in Court

Evidence Rating for Outcomes

Crime & Delinquency | Multiple crime/offense types

Justice Systems or Processes | Failure-to-Appear

Date: This profile was posted on January 10, 2018

Practice Goals

Pretrial release refers to a process in which adult defendants are released to the community, rather than detained in jail, prior to the disposition of their cases. To ensure that defendants who are released before disposition show up to scheduled court dates, jurisdictions have used the following three interventions: 1) court-date reminder notifications; 2) bonds as a condition of release; and 3) pretrial supervision, during which defendants are released and supervised in a non-institutional setting. The goals are to ensure that individuals attend their scheduled court appearances and to maintain public safety.

Practice Components

Judges determine whether an individual will be released during pretrial, usually based on the assessment of an individual’s likelihood of appearing in court and other considerations of public safety. Each of the three interventions has unique components.

Court-Date Reminder Notifications. Court-date reminder notifications may come in the form of postcards or letters, telephone calls (made by officers or through an automated system), and text messages. In addition to variations in the form of the notification, the content of the notification can also vary. For example, some notifications may only provide the court date, time, and location. Other notifications may explain the consequences of failing to appear in court such as a bench warrant being issued (Bechtel et al. 2017).

Bonds. Different types of release by bonds are available to individuals. One type is surety bond, or commercial bond, where money is lent to individuals on release by a bond dealer, usually for a fee. If the individual does not appear in court, the bonding company is liable for the full amount of bail. Another type is called deposit bond, where individuals post a small fraction of the bail amount (such as 10 percent) to the court.

If the individuals fail to appear in court, they lose the deposit and are held liable for the full value of the bond. Individuals may also be released on unsecured bonds, which does not require a deposit to be released; however, they will be liable for the full amount if they do not appear in court (Helland and Tabarrok 2004).

If individuals do appear in court, they typically receive back any bond or deposit already paid to the court. Using bonds as a condition of release is based on the idea that bonds serve as a deterrent for failing to appear at court because individuals will strive to avoid bond forfeiture or the threat of pursuit from a bonding company (Betchel et al. 2017).

Pretrial Supervision

The decision to release defendants to pretrial supervision is often based on the determination of a defendant’s level of risk (of not appearing in court and/or reoffending before the court date). Some jurisdictions will use certain factors to determine a defendant’s risk classification, such as the seriousness of the current offense or the likelihood of flight or rearrest (Goldkamp and White 2006). Other jurisdictions may use validated risk assessments, such as the Risk Prediction Index, to classify a defendant’s risk level (Robinson et al. 2012).

Defendants may be released to the custody of a pretrial supervision or services department (if one exists in the jurisdiction), or to other agencies, such as probation or parole offices, that are responsible for post-conviction supervision.

Pretrial supervision practices vary widely by jurisdictions. Pretrial supervision could include location monitoring; electronic monitoring; halfway house placement; substance abuse testing; or participation in treatment programs for alcohol/drug or mental health issues, or for sex offenses (VanNostrand and Keebler 2009). Pretrial supervision programs also vary in their requirements and protocols, such as standards for graduation, noncompliance, and termination (Hahn et al. 2016).

Meta-Analysis Snapshot
	Literature Coverage Dates	Number of Studies	Number of Study Participants
Meta Analysis 1	1976-2014	14	0

Meta Analysis 1

Bechtel and colleagues (2017) conducted a meta-analysis to examine what is known about the effectiveness of pretrial interventions, including court notifications, bonds, and pretrial supervision. A search was conducted on the Pretrial Justice Institutes website, along with searches of literature, using terms such as “pretrial”, “bond” and “defendant”, “bail”, or “pretrial risk assessment”. The authors identified over 800 manuscripts and included studies that met the following criteria: 1) the manuscript had to use data from a jurisdiction in the United States; 2) the publication year had to be 1970 or later; 3) the manuscript had to report on at least one of the following outcomes: failure to appear during pretrial release, arrest for new criminal behavior during pretrial release, bond revocation during pretrial release, or a combination of these outcomes; and 4) the manuscript had to compare at least two groups.

A total of 16 studies were included in the review of pretrial interventions, including two studies that specifically looked at court notifications, nine studies that specifically looked at bond types, and three studies that specifically looked at pretrial supervision (Note: Although one study looked at drug/alcohol testing, and one study looked at multiple types of pretrial interventions, these program types were not included in the CrimeSolutions review). The total number of defendants included in the 16 studies was 385,097 (however, the total number of defendants from the studies only on court notifications, bond types, and pretrial supervision was not provided). In all the studies, those defendants with enhanced conditions were considered the treatment group and those defendants experiencing typical conditions or conditions that involved lesser constraints were considered as the comparison group (for example, defendants who were required to post secured bond were considered the treatment group, while those defendants who were not required to post a second bond were considered the comparison group). Most of the studies did not report characteristics of the samples, including age, gender, and race. Four studies reported age of the study participants; the average age was 32.6 years. Four studies reported gender of the participants; 81.0 percent were male. Five studies reported information on race; the samples were mixed in terms of race of the defendants. Eleven studies reported charge severity of the study sample: one study included only misdemeanor charges, three studies included only felony charges, and seven studies included misdemeanor and felony charges. The two studies on court notifications were conducted in Multnomah County, Oregon, and in Nebraska.

For each study, one effect size for each potential outcome was calculated, using the largest sample size that was associated with the most rigorous test (i.e., matched sample versus unmatched sample) and the longest follow-up period. For all studies, weights and effect sizes (which were odds ratios and r values) were calculated using standard procedures. A methodological scale was used to control for study quality in multivariate analyses predicting effect sizes.