Minnesota Judicial Branch Logo

Pretrial Release Initiative

The Minnesota Judicial Council's Pretrial Release Initiative, one of the Strategic Initiatives of the Branch, is aimed at studying evidence-based tools for use by judges making pretrial release decisions. 

Under Minn. Stat. § 629.74 Minnesota Judicial Coucil approves pretrial evaluation forms to be used in each county.  Minnesota Judicial Council policy 524: Pretrial Release Evaluation, adopts the use of a statewide pretrial evaluation form and Minnesota Pretrial Assessment Tool (MNPAT) and directs the use of pretrial risk assessment tools in Minnesota District Courts. 

The assessment tool is meant to ensure that judges have the most predictive and least biased information, providing accurate, objective, and useful information for pretrial release decisions.

The assessment tools will be validated to ensure effectiveness and fairness in release decisions.

The Minnesota Judicial Council established the Pretrial Release Initiative Implementation Steering Committee to implement the statewide pretrial risk assessment tool and form.  Implementation was completed in late 2018. The Steering Committee included public and private attorneys, probation representatives and law enforcement, court administration, and judges.

 

GovDelivery SubscribeSign up to receive updates


This page provides updates on the rollout and maintenance of the Minnesota Pretrial Evaluation Form and Assessment Tool (MNPAT). 

Updates were made to the Pretrial Release Initiative website on August 10, 2021:
  1. Fast Facts - MNPAT Validation Study was added to the Resources tab.
  2. Updates were made to the Judicial Bench Card on the Resources tab. 
On October 31, 2019, the Main Charge Scoring Guide and MNPAT Pretrial User Guide were updated. The Main Charge Scoring Guide was revised to clarify the wording on the 6 point 609.2231 offense.

The following were updates to the MNPAT Pretrial User Guide in the 10/31/2019 update:
 
  1. Language was added to the “Refused/Unknown Fields” section instructing agents to indicate in the collateral comments text when a scored MNPAT variable could not be scored. The name of the variable followed by “could not be scored” should be the only description provided. Reasons why a variable is not scored should not be included. Including this language on the form will encourage judges to ask questions from the bench about the particular areas in question. This change creates alignment and consistent data with anticipated future changes to the CSTS pretrial module.
  2. The age at first delinquency adjudication section was further revised to clarify that offenses used to score this section can also be used to score criminal conviction history, if they meet the parameters outlined in the criminal conviction history definitions.
The following were updates to the MNPAT Pretrial User Guide in the June 13, 2019 update:
 
  1. The bench warrants table was updated to clarify that juvenile failure to appear bench warrants should not be counted.
  2. Language was added to the county of residence field to refine how to enter the location of defendants who do not reside in Minnesota (enter the county followed by “Non-Minnesota County”). If the client’s county of residence is not known, enter “Unknown.”

The following were updates to the MNPAT Pretrial User Guide in the April 4, 2019 update:

  1. Language was added to the processing and overview section clarifying the purpose of completing the MNPAT and outlining scenarios outside of a typical arrest in which the MNPAT should be completed.
  2. Language was added to the collateral contacts and victim sources section clarifying that information from the collateral or the victim can change the score for the current problematic chemical use section or the additional questions for validation. All other discrepancies should be noted in the comments.
  3. Tribal stipends and unemployment benefits were added to the list of public income assistance examples which should be scored as 0 points.
  4. The age at first delinquency adjudication section was updated to state that offenses used to score this section "may or may not" be used to score the criminal conviction history section.
  5. The criminal conviction history section was updated to indicate that cases pending sentencing (e.g. pleaded guilty, sentencing hearing upcoming) should not be counted in the score.
  6. The failure to appear bench warrant status of "Recalled Inactive" was changed to "Count," as this is the new court standard to use when inactivating warrants.
  7. The bench warrants section was updated to clarify that generic bench warrants should not be counted.
  8. The failure to appear bench warrants notes were updated to indicate that failure to appear bench warrants that come from family court should be counted.
  9. The pending case field description was updated with a definition of targeted misdemeanors.
In December 2018, modifications were made to MNPAT resources found on the Resources tab.  Specifically, the MNPAT User Guide, Main Charge Scoring Guide, Offenses Requiring an Pretrial Evaluation, and Person-related Convictions List have been updated. 

The following were updates to the Pretrial User Guide in the 12/11/18 update:
 
  1. A new section titled “Updating the CSTS Pretrial Module” was added to clarify what information needs to be added to the module. This includes information that may be added to the form during or before the hearing (e.g. main charged offense, case number, etc.).
  2. The criminal history section added a bullet indicating that all juvenile adjudications (except EJJ) should not be scored or listed. This was moved out of the table to be clear this direction applies to all criminal conviction history.
  3. A sentence was added to the No Modification to the MNPAT section clarifying that narrative information, including description why a field is scored, should not be included on the MNPAT.
  4. The “Bench Warrants – Fail to Appear at a Hearing” section was updated to clarify what FTA bench warrants should be counted:
    1. Issued Active/Served FTA bench warrants should be counted
    2. Do not count FTA warrants that are issued as a result of a missed initial hearing from a summons.
    3. Do not count FTA warrants if the defendant failed to appear as a result of being held in custody in anther jurisdiction, including ICE custody.
  5. The “Refused/Unknown Fields” section was updated with direction to not include information about a client refusing to answer questions on the MNPAT.
  6. The collateral comments section was updated to provide clarity about what information can be used to verify the client’s responses.
  7. The “Homeless or Three or More Address Changes in Past Year” section was edited, changing “has been” to “is” for describing if a client is in custody in a correctional facility or living in a temporary situation.

The Implementation Steering Committee began meeting in March 2018 and met for the last time in November 2018. The committee made the following decisions and recommendations: 
 
  • The Court Services Tracking System (CSTS) was selected as the case management system where data from the Minnesota Pretrial Release Evaluation Form will be entered and stored for use in the validation study.
  • The modifications needed to the CSTS system to ensure data is entered and stored properly will be available in February 2019, but a mail-merge form can be created in CSTS until that time.  The form can be found on the Resources tab.
  • The Minnesota Pretrial Evaluation Form (the Minnesota Pretrial Assessment Tool (MNPAT) is part of the form) will be implemented in each county by December 1, 2018. Counties can implement the MNPAT any time after October 5, 2018 and before December 1, 2018.  Before using the MNPAT, all stakeholders must receive training.  Training materials, including an implementation training guide, can be found on the Resources tab.
  • The Committee acknowledged that each county may have unique implementation issues.  To ensure a smooth rollout of the MNPAT, each county should have a "County Leader" who leads the rollout.  That County Leader can use training materials to assist in the rollout, including an implementation training guide, found on the Resources tab.
  • Judge and Probation district champions have also been identified to help the County Leaders.
In January 2018, the Minnesota Judicial Council approved the Minnesota Pretrial Assessment Tool (MNPAT).

Anoka, Cass, Hennepin, Sherburne, and Wright counties adopt alternative tools that meet the statewide standards. Validation is required for these tools. All other counties will use the statewide tool, MNPAT. 

The MNPAT will be validated through the State Court Administrator’s Office to promote consistent risk analysis. Alternative tools require approval from the Minnesota Judicial Council. Districts or counties using an alternative tool are required to meet the statewide standards for validation, according to Minnesota Judicial Council Policy 524: Pretrial Release Evaluation.

Minnesota Pretrial Release Evaluation Form and Assessment Tool (MNPAT)
Minnesota Pretrial Questionnaire
 
Pretrial Evaluation: The collection and analysis of a defendant’s information on factors impacting his/her likelihood to appear and risk to public safety.  May be conducted by local corrections or their designee (pretrial services, jail administration or law enforcement). 
Also known as:
  • bail evaluation
  • bail study
  • pretrial bail evaluation
Pretrial Evaluation Form: Template for recording factors of a particular defendant and the current charge used to inform the pretrial release decision.  A form does not assign a score or recommendation for pretrial release decisions and does not assign a weight for different factors.
Also known as:
  • pretrial evaluation form
  • bail form
  • form
Pretrial Release Decision: Release decision by court, or designated authority, that a defendant will be released on his/her own recognizance (ROR), released on conditions (CR), or released on bail/bond (bail).

Pretrial Release Process: The full process used in a court location (involving pretrial services and the court) from arrest/summons, including the pretrial evaluation, the pretrial release decision, and the actual release or detention of a defendant.

Pretrial Failure: An outcome that a pretrial risk assessment seeks to predict and minimize.  The most common outcomes are 1) likelihood of failure to appear (FTA) for court appearances in the present case and 2) likelihood of new crimes (e.g. arrest, charge) committed during the pretrial window (release to disposition).

Pretrial Risk Assessment Tool: A research-based (actuarial) instrument comprised of predictive factors that are weighted and scored to provide a recommended pretrial release decision (based on categorized risk of pretrial failure).  The tool is used to assist the court, or its designated authority, in making pretrial release decisions. 
Also known as:
  • tool
  • scale
Validation: A study of the effectiveness of a tool at predicting the outcome it seeks to predict on a specific population.

Racial Bias (in the Pretrial Risk Assessment Tool context): the tendency to treat those in a racial group (or groups) differently or unfairly. Racial discrimination is a result of racial biases, which can be explicit (conscious preference for a race), or implicit (preference for a race that operates outside of our awareness). 

Racial Disparity (in the Pretrial Risk Assessment Tool context): Racial disparity occurs when a racial group’s proportion of a particular statistic is over-represented as compared to its percentage of the population
For example: the percentage of those arrested as a proportion of the population.

Race Neutral (in the Pretrial Risk Assessment Tool context): a policy or score, for example, classifies individuals without regard for their race.


 
The Minnesota Judicial Branch has begun its first post-implementation validation of the Minnesota Pretrial Release Evaluation Form and Assessment Tool (MNPAT).

The MNPAT is a tool provided to judges to ensure they have accurate, objective, and useful information available to them for pretrial release decisions. The MNPAT was implemented in December 2018 as a result of the Judicial Branch Pretrial Release Initiative.

Judicial Council Policy 524 requires that the MNPAT be validated on a regular basis to ensure accuracy and fairness in predicting pretrial release failure. The MNPAT and its validation study support the vision of the court system being fair, consistent, and free of discrimination.

The validation is being conducted using data from cases that were heard between April 1, 2019, and March 31, 2020. Most of the cases heard during that time frame have now reached disposition.

The validation study is using the collected data to evaluate whether the MNPAT is accurately predicting a defendant's pretrial risk without bias (i.e. failure to appear and pretrial crime). In addition to quantitative analyses, the validation study is coordinated with a Validation Committee comprised of judges and partner agency stakeholders from across the state to gain their input on the utilization of the MNPAT and potential changes. The partner agency stakeholders include representatives from County Probation, Community Corrections, Department of Corrections, Minnesota County Attorneys Association, the state Public Defender’s Office, Tribal Courts, and research scientists.

At the completion of the validation study, a final report with recommended changes will be presented to the Judicial Council.