+ The Development & Science of the APRAIS Neil Websdale + - - PDF document

The Science and Practice of Community Informed Risk Assessment

January 25, 2018 2‐3:30pm CST

Neil S. Websdale, Ph.D., Director, Family Violence Institute, NAU and Greg Giangobbe, MA, Law Enforcement Coordinator, Family Violence Institute, NAU. Hosted by Kathleen J Ferraro, Ph.D., Director of Training & Curriculum, Family Violence Institute, Prof. of Sociology, Northern Arizona University.

Funding for this project was made available through the US Department of Health and Human Services, Grant #90EV0440‐01‐00. The viewpoints contained in this document are solely the responsibility of the author(s) and do not represent the official views or policies of the department and do not in any way constitute an endorsement by the Department of Health and Human Services.

Thank you for joining us today!

+

The Development & Science of the APRAIS Neil Websdale

+

APRAIS: Arizona intimate Partner Risk Assessment Instrument System

 Piloted with STOP TA funds in Yavapai County, Arizona  The outcomes used as a springboard to create a statewide, uniform

risk assessment tool and protocols

 Also presented the APRAIS model in two other states (Hawaii,

Montana)

+

What happens before an IPH? Some social patterns

 IPH profoundly gendered but race, class, ethnicity, geo-social

location also mediate

 About 50% of IPHs in US have prior systems contact but low

collaboration, communication, coordination

 About 50% female victims appear to die in relative isolation with no

• r few “system contacts”

 IPH stylized with telltale histories but much knowledge remains

hidden from the community and not shared among systems, agencies, & stakeholders

+ Telltale case histories and the research literature

 40 years of descriptive statistics on IPH (mostly female victims)  Used extensive research to inform the development of the questions

and protocols (Campbell et al., 2003 [cross sectional; one point in time]; Snider et al., 2009 [tracking over time]; Messing et al., 2015 [tracking over time])

 Focus of Yavapai and APRAIS: risk of severe re-assault or near lethal

violence

+ One recent important study we used as a touchstone

 Messing et al., 2015- Non-equivalent groups quasi-experimental field

trial using three groups

 LAP associated with an increase in protective actions & a decrease in

the frequency & severity of violence among this sample of IPV survivors

 Jill Messing ran our seven questions through her sample of 619 cases

• f female victims who responded to questions about risk and repeat

violence at approximately 7 months follow up

+ Telltale signs, antecedents, risk markers

 Prior history of IPV (weapons use; strangulation especially serial;

escalating violence, attempts to control, & emotional harms; beating during pregnancy: previous attempts to kill; forced sex; entrapment; capable of killing)

 Separation/emotional estrangement  Extreme jealousy linked to violence  Depression/suicidal potential

+ Telltale signs, antecedents, risk markers

 Alcohol & drug abuse  Stepchildren in the home  Compromised masculinity/humiliated fury

+

Key developments in AZ regarding risk assessment in IPV cases

 DVFRT development, 2008 - present  Risk conferences, trainings, 2010 - present  Bridging gap between DVFRT work, national risk assessment

developments, & research into IPH – clusters of risk markers seem to matter, Ontario DVFRT – 75% IPHs with 7+ markers

 Nevertheless, there remains the intractable & enduring presence of

IPHs

+ Key developments in AZ regarding risk assessment in IPV cases

 Growing recognition of the need to share IPV risk information  How do we put risk information before the court at IA?  Survivor data: need to notify of discoverability of information shared

in risk assessment

 The development of the APRAIS was more haphazard than may

appear

+ Statutory Considerations

 ARS 13-3967: Release on bailable offenses before trial:  B. Judicial officer shall take into account all of the following:  5. The results of a RA or lethality assessment in a domestic violence

charge presented to the court

+

Other factors relevant to release decisions under ARS13-3967

 1. Views of the victim – many questions  2. Nature and circumstances of the offense charged – many  4. Evidence that the accused poses a danger to others in the

community - Q12 – Has he/she threatened to harm people you care about?

 9. Whether the accused is using any substance if its possession or

use is illegal – Q11. Does he/she use illegal drugs?

+ Other key elements of the APRAIS

 Focus on the “How” & acknowledging we cannot predict outcomes  APRAIS addresses complexities of IPV and the need to better

triage/identify dangerous cases

 Basic logic – past behavior the most promising albeit imperfect

predictor of future behavior

 Balancing victim safety v. rights of the accused

+ Key elements of the APRAIS

 CIRA: Tiers 1 & 2 (mandatory/optional). Embedded in

communities/multiple stakeholders

 Respect for victim autonomy, dignity, liberty rights  Statutorily supported + administrative mechanisms + law

enforcement/advocacy readiness

 December 13, 2017 Chief Justice Scott Bales of the AZ Supreme

signed off on the changing of the rules of criminal procedure to add the Form 4(C) (Effective April 2, 2018)

+ Key elements of the APRAIS

 The APRAIS provides risk information gathered by police from

victims/alleged victims to the court at IA

 It does not contribute information to the trial or sentencing phase  It provides police, advocates, and victims/alleged victims with more

information

 Potentially provides victims/alleged victims with

time/space/resources to maneuver

+ Other key elements

 AZ GOYFF funded another 4-year APRAIS project commencing

January 2018

 FVI at NAU to work with the Arizona Peace Officers Standards and

Training Board (AZPost) to train officers across the state to implement the uniform tool should their agencies elect to deploy it

+

Risk Assessment Tool and Protocols Risk Assessment Tool and Protocols

Criminal Justice Criminal Justice DV Advocates and Survivors DV Advocates and Survivors CCRT and DVFRT CCRT and DVFRT Public Health Public Health

Yavapai Risk Assessment Project Tier 1 APRAIS Questions

Question Yes No Decline

• 1. Has the physical violence increased in frequency or

severity over the past six months?

• a. Alternate wording: Is the pushing, grabbing, hitting, or
• ther violence happening more often?
• 2. Is he/she violently and constantly jealous of you?
• 3. Do you believe he/she is capable of killing you?
• 4. Have you ever been beaten by him while you were

pregnant? (e.g. hit, kicked, shoved, pushed, thrown, or physically hurt with a weapon or object)

• 5. Has he/she ever used a weapon or object to hurt or threaten

you?

• 6. Has he/she ever tried to kill you?
• 7. Has he/she ever choked/strangled/suffocated you?

If this has happened more than once, check here Totals

+ Risk classification and scoring

 Three categories: risk (0 or 1 yes/7); elevated risk (2 or 3 yes/7); high

risk (4+/7)

 Elevated risk and high risk cases trigger optional links with

advocacy and a law enforcement follow up

 The behaviors you answered “yes” to have been present in very

dangerous situations

 “Elevated” - 6 times more elevated risk of severe re-assault or near

lethal violence when compared to those with fewer than 2 risk factors present

 “High-risk” – 10.5x

+ Relative and Absolute Risk

 Relative risk: A victim answers 4+/7 APRAIS “yes” and has a 10.5X

greater chance than someone answering less than two “yes” of experiencing severe re-assault or near lethal violence within the next 7 months (see parallel statement below)

 Absolute risk: Among victims who answer 4+/7 “yes” on the APRAIS

tool roughly 15% will experience severe re-assault or near lethal violence within 7 months (Messing, OK data, comparing 4 v. 0 or 1 “yes” responses)

+ Relative and Absolute Risk

 Both statements are accurate but if you hear the relative risk without

hearing the absolute risk we may develop an exaggerated sense of longer term risk of severe re-assault/near lethal violence

 Fair justice implications?  Defendant freedom v. victim safety  Victim/alleged victim choice/agency  Tempers the tendency to alarm/paralyze victims?

+ Case Law: U.S. Supreme Court on RA

 Barefoot v. Estelle 463 U.S. 880 (1983)

 Expert testimony on dangerousness may not always be correct  Indeed, defense team argued psychiatric predictions of future

dangerousness were wrong “most of the time”

 Nevertheless, such testimony is admissible and ought be subject to

the adversarial process

+ Amicus brief in Estelle from APA

 Noted the “unreliability of psychiatric predictions of long-term future

dangerousness is by now an established fact within the profession”

 “The APA’s best estimate is that two out of three predictions of long-

term future violence by psychiatrists are wrong”

 The Supreme Court “The court does not dispute this proposition”

+ State courts on RA

Pettingill v. Pettingill, 480 S.W

.3d 920 (KY 2015) – Kentucky Supreme Court upheld (6-0) a lower court decision to grant OP (order of protection) based partially on the appropriate employment of the judge’s knowledge of risk markers

+ State courts on RA

 State v. Loomis, 881 N.W

.2d 749 (WI 2016) –RA cannot be used as the determinative factor in deciding whether an offender can be supervised safely and effectively in the community

 RA may not be used to determine whether to incarcerate an offender

• r the severity of the sentence

+ Wisconsin Supreme Court in Loomis

 Five written warnings for judges in situations where PSIs incorporate

COMPAS algorithmic assessment tool

 1. Proprietary nature of COMPAS prevents the disclosure of how risk

scores are calculated

 2. Unable to identify high risk individuals – rather population groups  3. No cross validation for the Wisconsin population (national data

• nly)

+ Loomis

 4. Studies raised questions about over-identification of minorities as

“high-risk”

 5. COMPAS was developed specifically to assist the Department of

Corrections in making post-sentencing determinations

 The court expressed a desire to instill general skepticism about the

tool’s accuracy and

 Targeted skepticism with regard to the tool’s assessment of risks

posed by minority offenders

+ APRAIS GOALS

 Produce standardized & evidence-based RA tool and protocols for

law enforcement & advocacy

 Create a shared language of risk to inform CJS decisions regarding

bail, conditions of release, supervision, sanctions, & treatment

 Provide education to inform case handling and public awareness on

a broader scale, e.g. public health screening

+ APRAIS GOALS

 Create an addendum to the Form 4 (law enforcement release

questionnaire) through which law enforcement can communicate IPV risk to the court

 Work with law enforcement regarding existing reporting

mechanisms and possible long term realignment

+ APRAIS Protocols

 Conduct the risk assessment after the on-scene investigation is

completed

 Officer override option for professional judgment  Intended for IPV only  Questions are optional and asked of male and female victims

+ APRAIS Protocols

 To the alleged victim: we are assessing “potential danger”  You have to make your own decisions  To the alleged victim: RA discoverable  Body cameras and notification of discoverability?

+ Early Outcomes

 Victim declination rates low – 10-30%  Glendale PD LAP declination rates around 50%  More detailed police reports  Insufficient data about possible influence in court, e.g. bail setting  Connects victims/alleged victims with services  Victim, officer, & public safety

+ Early Outcomes

 At PPD and PVPD, a significant number of calls resulted in a decision

not to administer the tool. Possible reasons:

 Distraught victims that officers perceive having an inability or

unwillingness to complete the tool

 Victim intoxication  The inability of officers to contact the victim at the time of the report

(i.e. victim GOA)

+ Early Outcomes

Third party calls for service where the alleged conflicting

parties’ deny any IPV

Officer determination that the call comprised a “verbal”

dispute and that no crime had occurred

Officer unwillingness to administer the tool regardless of the

circumstances

Survey of victims in focus groups suggest the vast majority

(86%, N=42) want to be notified of the discoverability of RA information

+ Early Outcomes

 When the tool is administered it seems to:  Increase officer understanding of the case or situation  Contribute to more detailed police reports  Increase the appearance of officer compassion toward victims  Officers frustrated by not knowing case outcomes (e.g. prosecution –

fruit of their labor)

+ Questions

 Contact information:  Neil Websdale  Family Violence Institute  Northern Arizona University  Neil.Websdale@nau.edu  928-637-4510

APRAIS Training

Implementation process / Victim safety and autonomy / Stake‐ holder buy in

Training Process And Issues

• Relationship building with specific organizations
• Time to train Officers
• Acknowledge concerns of officers– There is always pushback
• Not over‐selling APRAIS – it assists – it is not a cure all for IPV

Training Process And Issues

• Why – Why do risk assessment at all, why APRAIS and not something else?
• Benefits – What benefits are there to the Patrol Officer and the Victims
• Feedback loops – How do we get information back to the Patrol Officer, Victims and

Survivors

• Target influential personnel in the organization – Those that affect the culture of the
• rganization

Training Process And Issues

• We stress and re‐emphasize officer safety
• We stress the victim has autonomy in the process – Their safety is paramount
• We stress due process for the accused
• Multi‐Prong / Holistic Training approach

Training Process And Issues

• We re‐emphasize the interdependence that is essential for effective partnerships

with the community and the justice system

• Front end = Patrol Officers
• End users = Detectives, Command staff, Community stakeholders, Prosecutors,

Defense Attorneys, Court Personnel & Judges

• Delivery methods – Getting information out to 174 agencies in Arizona

Training Process And Issues

• Front end user vs end user ‐ Design training to patrol
• We define an objective for the officers
• Training must be realistic

Training Process And Issues

• We emphasize that the APRAIS tool is part of the evolution of professional policing

and the guardianship of the communities officers serve

• APRAIS allows us to connect to the community in ways we haven't before

Why Do Risk Assessment At All ?

• Historically domestic violence investigations may have only scratched the surface of

identifying the danger that looms for many victims

• IPV risk assessment allows LE to look deeper into the dynamics of an intimate

partner relationship and possibly uncover potential danger signs to raise the awareness in victims, triage resources for them, and help hold offenders accountable

What Is APRAIS ?

• Arizona intimate Partner Risk Assessment Instrument System (APRAIS)
• It provides risk information gathered by police from victims/alleged victims of

intimate partner violence to the court at the initial appearance of the accused

• It does not contribute information to the trial or sentencing phase – only at the

initial appearance

Purpose Of The APRAIS Instrument

APRAIS provides criminal justice personnel and advocates with an evidence‐based tool and protocol to gather information which helps identify potential risk of severe re‐ assault, as well as lethal or near lethal, intimate partner violence (looking at the most severe / violent incidents) & possible future threats to officers

Why APRAIS And Not Something Else?

• Used extensive research to inform the development of the questions and protocols
• ARS 13‐3967 B5 ‐ judges shall consider RA scores when setting bail, etc
• On December 13, 2017 the Arizona Supreme Court amended Rule 41 of the Arizona

Rules of Criminal Procedure

• The court effectively added a uniform or standardized statewide risk assessment

tool to the law enforcement release questionnaire in cases of intimate partner violence = APRAIS

Will This Change How Law Enforcement Investigates Domestic Violence ?

• Possibly…however this Form IV rule change will not:

– Dictate policy or procedures for law enforcement investigations – Agencies will still be free to ask any questions they deem important to their investigations & communities (through tier 2)

Objective Of APRAIS

• We hope that using the APRAIS tool along with policies and operational protocols

that emphasize the importance of good, solid police work in Intimate Partner Violence cases, can in many cases, result in a successful intervention for the victims

• ver time
• The researchers will track outcomes to learn more

When Can AZ Law Enforcement Begin To Use The APRAIS Tool?

• (Voluntary) Implementation begins April 2nd, 2018 – Information will go out to

agencies and AZPOST will add APRAIS information to the basic training curriculum at the Academies

• Agencies may choose to implement APRAIS on IPV cases including misdemeanors
• Agencies have the flexibility to implement as they see best for their communities

with community safety as the principal focus

• Guiding principles of APRAIS are to be maintained

Unique Relationship With AZPOST

• Coordinating together as a team for information, training and support for agencies
• Have the authority of AZPOST supporting the tool
• AZPOST providing logistical and technical support

‐Training space ‐On‐line learning modules ‐Training videos ‐Continuing credit hours for officers

Stakeholder Buy‐In

• Meeting formally and in‐formally with various groups to explain and gain buy‐in for

the tool ‐ Police Chiefs ‐ Sheriffs ‐ Police trainers ‐ Other CJS players and community partners –This includes victims

Influential Personnel In LE Organizations

• We want to target those personnel who influence an organization’s culture both

formally and informally

• Formal – Chiefs and their command staff
• Informal leaders of the organizations

‐Recruit Training Officers at the Academies ‐DV instructors at the Academies ‐Field Training Officers ‐Mentoring Officers

Training Methods

• We utilize a variety of training methods to connect the officers to the information

they need to be successful in the field

• In‐Person training
• Train the Trainers
• On‐line training modules
• On‐line videos

We recognize the importance of reaching smaller agencies and the unique challenges they face

Multi‐ Prong / Holistic Training

• NAU Family Violence Institute (FVI) developing specific training programs and
• utreach for the community of players involved in IPV
• Law Enforcement ‐ Patrol, Detective, Command staff, Trainers, Recruits
• Prosecutors
• Defense attorneys
• Judges
• Advocates and community partners

Importance of Patrol

• We emphasize that Patrol plays a critical part in the success of the tool
• We need their ongoing feedback on how the tool is working and any issues that

arise in the field

Importance of Patrol

• The Patrol Officer is one of the few professionals in the Criminal Justice system who

might see and hear what goes on in the confines of violent/abusive homes

• As with most cases Patrol plays a critically important role in the criminal justice
• system. Patrol officer behavior and the application of the APRAIS tool might improve

victim safety and offender accountability

Officer Safety

• Officer safety and scene security is emphasized
• Officers determine if and when to proceed with the tool

Victim Safety And Disclosure

• We emphasize victim autonomy

‐ We do not order / tell victims what they must do ‐ We provide information and assistance to help them make decisions

• We disclose discoverability of information to the victims

‐ We are open and upfront that information may be seen by defense council and defendants

• Victims don’t always know that what they tell the officers is discoverable –so it’s

important that we let them know so they are fully informed

Victim Safety And Disclosure

(Full Disclosure to Victim) (Allow Victim to determine course of action) Victim participation voluntary Responses discoverable Your partner might find out what you tell me It also means a judge has to consider risk assessments if setting bail

Victim Safety And Disclosure

• When addressing the need to ask the APRAIS questions of the victim on scene, the
• fficer should be having a dialogue expressing not only the tool’s importance, but

also explaining the options available to the victim. For example: “I am going to be asking questions about what’s happened to you today as well as in the past with your partner in order to gather more information about the potential level of danger in your relationship. We are doing this to hopefully improve your family safety and that of our community.”

Victim Safety And Disclosure

• We emphasize to the officers that they should be fully engaged when interacting

with the victim

• We want the officers to be good listeners, make eye contact, have a conversation

with the victim and to demonstrate and behave in a caring manner ‐ We want them to be true guardians

• We know it’s important for victims to feel that officers care and it’s important for
• fficers to understand that

What Benefit To Me (Patrol)

• Reduce some repeat calls to the same location – over time
• Gain additional information on violence potential – Officers are often injured or

killed on these type of calls

• Impact the Aggravated Assault / Homicide rate in relation to IPV

‐ looking to impact the most serious / violent incidents

• What would even a 5% reduction in aggravated assault look like? Patrol time /

Detective time and resources used to investigate

What Benefit to Me (Patrol)

• Provide victims with potentially life saving information and build rapport while on

scene

• We hope to gain additional information over time to assist in keeping victims safe
• Another step in community policing by being a part of a community informed risk

assessment – multiple stakeholders –who's mission is to impact IPV and provide victims with information and options to keep them safe

Learning More Information About Case Outcomes And The Fruits Of Patrol Officer Labor

• We want to ensure police and community stakeholders are in the loop
• Multiple community partners, both law enforcement and non‐law enforcement are

involved and avenues are being identified to ensure there is good information flow

• n the impact of employing the tool
• Note: APRAIS involves community feedback loops from survivors/focus group

feedback

Evolution of Community Policing

• We emphasize that this is a natural evolution of community policing
• New community partnerships are created
• Community informed risk assessment

Evolution of Community Policing

• Better rapport with IPV victims
• We want to highlight and continue to push the community policing aspect of the

APRAIS tool and the guardianship officers should have with the communities they serve

This Is The Natural Evolution Of Policing

• LE constantly changes and we adapt to better serve and protect the community:

Knowledge, tactics, equipment, training & community liaison

• What was active shooter / mass killing event protocol 15 years ago?
• What has been the evolution of DV response: Pictures of injury / recordings /

strangulation‐choking

What Is Your Purpose ?

• Is the way we respond and the actions we take making victims safer?
• Does the way we respond identify those who are most dangerous and cause the

most harm?

• What messages are we sending and reinforcing with our actions or lack of action?

Questions

• Contact information:

– Greg Giangobbe – Family Violence Institute – Northern Arizona University – greg.giangobbe@nau.edu – Phone: (928) 523‐1276