Getting Inside the Black Box: Understanding How Jail Diversion Works - Re-released June 2020 | Originally published August 2010 - Policy Research ...

Page created by Maurice Cooper
 
CONTINUE READING
Getting Inside the Black Box:
Understanding How Jail Diversion Works
    Re-released June 2020 | Originally published August 2010
This document was originally developed in August 2010 by Policy Research
Associates. It has been rebranded for distribution to the field, but the content has not
been changed. This document should be viewed as a reference resource. For more
information, please contact pra@prainc.com.

Overview                                             expect a reduction in arrests, fewer jail days, and
                                                     lower charge levels for subsequent arrests.
Over the past 20 years, jail diversion for persons
with mental illness and co-occurring substance       Over 18 years, there has been dramatic
use disorders has become a widely accepted           program growth, from 52 programs identified
part of the criminal justice system. The frequent    in the initial 1992 national survey (Steadman,
contact with police by people with unmet mental      Barbera & Dennis, 1994) to now some 560
health needs and the high rates of mental and        programs operating across 47 states, based on
substance use disorders among correctional           current estimates.
populations have created broad support for
diversion across criminal justice, health, and        By connecting justice-involved people
advocacy lines. Jail diversion programs provide           with a serious mental illness to
a way to redirect high-risk individuals from           comprehensive and effective mental
justice settings into community-based services
                                                       health treatment in the community,
and supports, often with judicial supervision.
                                                        individuals would be stabilized and
The Targeted Capacity Expansion                      communities could expect a reduction in
Jail Diversion Program                               arrests, fewer jail days, and lower charge
                                                           levels for subsequent arrests.
The federal government has supported the
development and expansion of jail diversion
programming nationwide since 1992. After             Convening the Experts
the initial 1997 Jail Diversion Knowledge            In January 2010, a small, diverse group of
Development Application (KDA) demonstration          researchers, policymakers, and jail diversion
project, expansion efforts included                  practitioners convened in Bethesda, MD, to
authorizations for the 2001 Targeted Capacity        assess what conclusions could be derived from
Expansion (TCE) initiative and 2002-2007 TCE         the TCE Jail Diversion cross-site evaluation
for jail diversion programs, followed by the 2008    project data. Present were representatives from
13-state Jail Diversion and Trauma-Recovery:         Policy Research Associates, Inc., the Council
Priority to Veterans initiative.                     of State Governments (CSG), and Westat;
The New Freedom Commission on Mental Health          Federal representatives; program evaluators;
(2004) recommended jail diversion as a public        psychiatrists; peer specialists; and criminal
health and public safety strategy. By connecting     justice professionals from the bench, prosecution
justice-involved people with a serious mental        and defence. The group was charged with the
illness to comprehensive and effective mental        task of critiquing findings, using data from 14
health treatment in the community, individuals       post-booking TCE I programs.
would be stabilized and communities could

                                                                              Getting Inside the Black Box | 2
Major Findings                                                           functioning with baseline reductions of -0.7
                                                                         and -0.78 at 6 and 12 months from a mean
The TCE data showed the clearest impact                                  2.02 baseline (scale of 1-4). The Colorado
of jail diversion in the areas of drug and                               Symptom Index (CSI) demonstrated an
alcohol use, functionality in daily living, re-                          average 30 percent improvement in symptom
arrest history and jail days, and timely service                         reduction and well-being ratings.
linkage. Across each of these categories,
data showed improved outcomes for clients                                Public safety improvements were observed
involved in a diversion program.                                         in 12-month data, with a 53-percent
                                                                         decrease in arrests post-enrollment and a
Drug and alcohol use dropped dramatically                                corresponding reduction in jail days from 52
during the first 6 months. Self-report of any                            days pre-enrollment to 35 days at 1 year post-
alcohol use dropped by more than 50 percent,                             enrollment. Across charge history, 46 percent
while use of alcohol to intoxication and illegal                         of clients diverted on misdemeanor charges
drug use both decreased 70 percent from                                  and 49 percent of those diverted on felony
baseline, with the decrease mostly sustained                             charges experienced no further arrests during
at 12 months.                                                            the following year, so that charge severity itself
                                                                         made no difference to the likelihood of future
Assessment of individual improvement and
                                                                         arrest or charge severity. Overall, diverted
capacity for independent living showed equal
                                                                         clients had 44 percent fewer arrests and 33
improvement: the daily living/role functioning
                                                                         percent fewer jail days (Case et al., 2009).
scale demonstrated improvements in

                                               Jail Diversion Logic Model1
               Stage 1                                         Stage 2                                         Stage 3
                                                                                                                Improved Mental
Diversion             Identify and                  Linkage              Comprehensive/                         Health/Individual
                      Enroll People in                                   Appropriate                            Outcomes
                      Target Group                                       Community-
                                                                         Based Services

                                                                                                                  Improved Public
                                                                                                                  Safety Outcomes

                  Input                                       Black Box                                          Output
1. From Case, B., Steadman, H.J., Dupuis, S., & Morris, L. (2009). Who succeeds in jail diversion programs for persons with mental illness?
   A multi-site study. Behavioral Sciences and the Law, 27(5).

                                                                                                         Getting Inside the Black Box | 3
Data analysis identified 3 outcome predictors            evaluate diversion programs and those diverted
for future criminal activity: lengthier prior arrest     to them as homogenous groups, there remains
history; gender (with women less likely to               an extraordinary level of heterogeneity among
reoffend); and more illegal drug use.                    programs, dispositional practices, treatment
                                                         services provided, and individual performance.
Overall, the data demonstrated improvement in
mental health outcomes, with reduced symptoms
and improved well-being, and improvement in               Ultimately, in analyzing the findings,
public safety outcomes, with reduced rearrest             the 2010 expert group concluded the
rates, lower charges, and fewer jail days. These         data provide enough evidence to define
data also suggest the predominant factor related         the key ingredients within Stage 1 and
to public safety outcomes is past criminal
                                                           2 of the Jail Diversion Logic Model
behavior. However, prior arrest history is, by itself,
an insufficient determinant of future risk. Other        necessary to create a competent system
compounding risk factors must be considered               capable of meeting the Stage 3 public
and the treatment and supports occurring within              health and public safety goals.
the black box of the jail diversion process must
be examined.                                             In concept, the Jail Diversion Logic Model
                                                         hypothesizes a causal relationship that symptom
Beyond Data: The Black                                   control would result in reduced recidivism
Box of Jail Diversion                                    for justice-involved individuals with a mental
                                                         illness. However, research has shown mental
In science, the “black box” is an entity or system       illness is not the dominant cause of arrest. In
that can be viewed solely in terms of its input,         one metaanalysis, only 4 percent of arrests in a
output, and transfer characteristics, without any        sample of jail divertees with mental illness were
knowledge of its internal workings.                      the direct effect of mental illness although 14.3
In the context of jail diversion, much is known          percent were indirectly related. Substance abuse
about the demographics, charge level, and                was the direct cause of arrest in 22.5 percent
treatment needs of people going in, and study            of cases, with only 8.6 percent indirectly related
data reveal a fair amount about service retention        (Bonta, Law, & Hanson, 1998). Additional research
and re-arrest history among people coming out.           suggests mental illness may be only a modest
The black box represents variables or “change            factor for recidivism and reveals justice-involved
components”—from evidence-based services                 people with mental illnesses meet many of the
to the perceived role of coercion in criminal            “central eight” leading risk factors for future
justice supervision—that may provide clues               criminal behavior (Andrews et al., 2006).
as to what works and why. Despite efforts to

                                                                                 Getting Inside the Black Box | 4
…engage in offending and other forms of
                                                                         deviant behavior not because they have
“Central Eight” Risk Factors
                                                                         a mental disorder, but because they are
It is important to note high “central eight” risk
                                                                        poor. Their poverty situates them socially
scores are shared by both individuals with a
mental illness and those without, suggesting an                          and geographically, and places them at
alternative view of the root of the problem for                           risk of engaging in many of the same
frequent criminal justice contact. Some people                           behaviors displayed by persons without
with serious mental illnesses may:                                      mental illness who are similarly situated
                                                                              (Fisher et al., 2006, pg. 553).

“Central Eight” Risk Factors2
                          Risk Factor                                                              Need
 History of criminal behavior                                        Build alternative behavior
 Antisocial personality disorder                                     Problem-solving skills, anger management
 Antisocial cognition                                                Develop less risky thinking
 Antisocial peers                                                    Reduce association with criminal others
 Family and/or marital discord                                       Reduce conflict, build positive relationships
 Poor school and/or work performance                                 Enhance performance, rewards
 Few leisure or recreation activities                                Enhance outside involvement
 Substance abuse                                                     Reduce use
2. From Andrews, D. A., Bonta, J., & Wormith, J. (2006). The recent past and near future of risk and/or need assessment. Crime and
   Delinquency, 52(1).

                                                                                                        Getting Inside the Black Box | 5
Understanding the Black Box and                       and (2) outreach that could significantly reduce
                                                      noncompliance and technical revocations.
Fine-Tuning a Model
                                                      TCE data tell us little about effective service
Viewing the logic model stages as opportunity for     ingredients: range of services, evidence-based
cause and effect, jail diversion data have a fairly   practices, and the level of intensity at which they
robust effect in meeting public health and public     need to be provided. Researchers believe service
safety goals via engagement, treatment, and           answers would help define the process and
supervisory strategies in Stage 1 and 2, including:   changes that occur in the black box and provide
assessment, admission decisions, individualized       a standardized model for replication. However,
planning, intervention choices, community-            it may be a moot question since the evidence
based supervision, and peer support. Despite          suggests individualized plans and the dynamics
this, people still cycle back into contact with the   of the supervisor-client relationship are, in fact,
justice system, which suggests the issue is not       the keys to success (Skeem et al., 2007; Skeem
only a lack of access to services but a need for      et al., 2011).
(1) access to evidence-based “competent care”

What Is in the Black Box of Jail Diversion?

               Input                       What may occur in                        Output
                                             the black box
Inclusionary criteria               Judicial supervision and          Mental health symptom control
                                    mandated treatment                or reduction
Voluntary admission                 Comprehensive needs               Lower costs
                                    assessment, including “central
                                    eight” risk factors
Men/Women                           Service linkage and               Fewer arrests
                                    ancillary supports
Misdemeanor/felony charges          Person-centered, individualized   Reduction in charge severity
                                    planning, choice
Prior arrest histories              Tailored treatment and            Fewer jail days
                                    service intensity
Mental health diagnosis             Trauma-informed care              Improved quality of life
PTSD/trauma history                 “Change agent” providing “firm    N/A
                                    but fair” community supervision

                                                                               Getting Inside the Black Box | 6
Next Steps and Opportunities                                              Drawing on this and what we know from risk
                                                                          assessment and violence studies, the TCE
Collectively, the data provide a framework for                            data strongly suggest we are moving toward
future directions in policy and practice. The                             a more valid model of “what works” to reduce
data support the establishment of jail diversion                          criminal behavior. By introducing criminogenic
programs on the grounds of public health, public                          factors to the discussion, we see the similarities
safety, and individual success. It shows we can                           in risk factors for recidivism across offenders
be reasonably accurate in distinguishing people                           with and without mental illnesses. This finding
more or less likely to reoffend and the range and                         indicates a successful jail diversion model,
intensity of their service needs. However, prior                          in symmetry with reentry planning, hinges on
arrest history alone should not be interpreted as                         integrated, clientoriented community services and
a preclusion to diversion but instead as a helpful                        supports the argument for the use of evidence-
indicator of greater risk factors.                                        based practices throughout mental health and
                                                                          correctional settings.

     Toward a More Valid Model of “What Works” to Reduce Criminal Behavior3
                                                                                                  RNR: Risk-Need-Responsivity Model
   Diversion/Sentence to
                                                                                                  ACT: Assertive Community Treatment
   Mandated Treatment
                                                       Mental Health Treatment                    IDDT: Integrated Dual Disorder Treatment

                                                       e.g., ACT, IDDT, others
                                                       shown to affect                              Small subgroup
   EBP in Corrections
                                                       psychiatric outcomes.
   e.g., RNR, firm but
   fair relationships,
   others shown to affect                                                                                   Fewer new crimes
                                                              Everyone else
   recidivism.                                                                                              and new victims

3. Skeem, J., Peterson, J., & Silver, E. (2011). Toward research-informed policy for high risk offenders with serious mental illness. In B.
   McSherry & P. Keiser (Eds.), Managing High Risk Offenders: Policy & Practice. New York: Routledge.

                                                                                                            Getting Inside the Black Box | 7
Policy Research Associates, Inc.                       Case, B., Steadman, H. J., Dupuis, S., & Morris,
                                                         L. (2009). Who succeeds in jail diversion
345 Delaware Avenue, Delmar, New York 12054              programs for persons with mental illness? A
www.prainc.com | pra@prainc.com                          multi-site study. Behavioral Sciences and the
                                                         Law, 27: 661-674.
About Policy Research                                  Fisher, W. H., Roy-Bujnowski, K. M., Grudzinskas,
Associates, Inc.                                          A. J. Jr., Clayfield, J. C., Banks, S. M., & Wolff.
Policy Research Associates, Inc. (PRA) is a               N. (2006). Patterns and prevalence of arrest
Women-Owned Small Business (WOSB) that                    in a statewide cohort of mental health care
is a national leader in behavioral health and             consumers. Psychiatric Services, 57(11):1623-
research. PRA offers four core services that help         1628.
individuals with behavioral health needs achieve       New Freedom Commission on Mental Health.
recovery. In partnership with its sister non-profit,     2004. Subcommittee on criminal justice:
Policy Research, Inc., PRA offers policy, research,      Background paper. Rockville, MD: Department
technical assistance, and training services,             of Heatlh and Human Services. Pub. No. SMA-
primarily in the content areas of behavioral health;     04-3880.
homelessness; income supports; criminal justice;
juvenile justice; education; service members,          Skeem, J., Eno Louden, J., Polasheck, & Cap,
Veterans, and their families; recovery support;          J. (2007). Relationship quality in mandated
consumers and family members; trauma;                    treatment: Blending care with control.
wellness; and systems mapping.                           Psychological Assessment, 19, 397-410.
                                                       Skeem, J., Manchak, S., & Peterson, J. (2011).
Suggested Citation                                       Correctional policy for offenders with mental
Policy Research Associates, Inc. (2020). Getting         disorder: Creating a new paradigm for
Inside the Black Box: Understanding How Jail             recidivism reduction. Law and Human Behavior.
Diversion Works (2nd ed). Delmar, NY: Author.            35:110-126.

References                                             Skeem, J., Peterson, J., & Silver, E. (2011).
                                                         Toward research-informed policy for high risk
Andrews, D. A., Bonta, J., & Wormith, J. (2006).         offenders with serious mental illness. In B.
  The recent past and near future of risk and/           McSherry & P. Keiser (Eds.), Managing High
  or need assessment. Crime and Delinquency,             Risk Offenders: Policy & Practice (111-121).
  52:7–27.                                               New York: Routledge.
Bonta, J., Law, M., & Hanson, K. (1998). The           Steadman, H. J., Barbera, S., & Dennis, D. (1994).
  prediction of criminal and violent recidivism          A national survey of jail diversion programs for
  among mentally disordered offenders: A                 mentally ill detainees. Hospital and Community
  metaanalysis. Psychological Bulletin, 123,             Psychiatry, 45(11):1109–1113.
  123-142.

                                                                                  Getting Inside the Black Box | 8
Acknowledgments
                           Forum Participants
                     January 14, 2010 | Bethesda, MD

   Stephen Bush                                        Stephanie LeMelle

   Kevin Casey                                         Ann-Marie Louison

 Victoria Cochran                                      Joseph Morrissey

  Debra Ferguson                                       Mary Anne Myers

  Doyle Forrestal                                         Fred Osher

    David Fuller                                         Travis Parker

Micah Haskell-Hoehl                               Susie Davidson Powell

   Kirk Heilbrun                                         Merrill Rotter

 H. Richard Hopper                                      Jennifer Skeem

  Chandria Jones                                         Bonita Veysey

  Kathleen Lacey                                        Douglas Wilson

J. Steven Lamberti                                       David Wright

  Steven Leifman

                                                               Getting Inside the Black Box | 9
You can also read