Content uploaded by Michael Claude Smith
Author content
All content in this area was uploaded by Michael Claude Smith on Dec 17, 2015
Content may be subject to copyright.
Abstract—This paper presents an analysis of and recom-
mendations for improving the relationship between the mental
health and criminal justice systems in Charlottesville and Al-
bemarle County, Virginia.
The project team u sed data analysis, detailed process model-
ing, and stakeholder discussions to identify three major prob-
lems in the current system for managing the needs of people
with mental illness, also referred to as consumers. First, en-
counters between consumers and law enforcement or mental
health personnel resulted in unnecessary safety risks. Second,
the limited resources of both the criminal justice and mental
health systems were consumed from ineffective responses to
consumers’ needs. Third, wh en an individual with mental ill-
ness moved from one agency to another, they often experienced
gaps in treatment which caused a crisis situation to develop,
requiring police involvement.
To address these problems, the project team recommends
that the city and county re-align their existing resources into a
collaborative Crisis Intervention System (CIS). The CIS would
seek to improve the coordination between agencies in the
criminal justice and mental health systems; it would be able to
handle both the acute needs of consumers in crisis and mini-
mize the potential for crisis situations to develop by providing
long-term stability. In addition to using these systems’ limited
resources effectively and efficiently, the development of a CIS
would enhance the quality of life for the region’s consumers
and benefit the community as a whole.
This project was conducted by Univ ersity of Virginia stu-
dents as an evaluation team for the Crisis Intervention Team
(CIT) Taskforce, a d iverse group of local representatives from
both the mental health and criminal justice systems who are
advocates for improving the relationship between the systems.
I. INTRODUCTION
N 2005, over half of all inmates in United States correc-
tional facilities have had or currently have a mental health
Manuscript received April 9, 2007.
M. M. Ordonez is with the Systems and Information Engineering De-
partment, University of Virginia, Charlottesville, VA 22904 USA (e-mail:
mikeordonez@virginia.edu).
K. M. Worrest is with the Systems and Information Engineering De-
partment, University of Virginia, Charlottesville, VA 22904 USA (e-mail:
kworrest@virginia.edu).
M. S. Krauss is with the Systems and Information Engineering Depart-
ment, University of Virginia, Charlottesville, VA 22904 USA (e-mail:
msk9m@virginia.edu).
L. N. Tietje is with the Systems and Information Engineering Depart-
ment, University of Virginia, Charlottesville, VA 22904 USA (e-mail:
Lauren.tietje@gmail.com).
R. Bailey is with the Systems and Information Engineering Department,
University of Virginia, Charlottesville, VA 22904 USA (phone: 434-924-
5393; fax: 4343-982-3972; e-mail: rrbailey@virginia.edu).
M. C. Smith is with the Systems and Information Engineering Depart-
ment, University of Virginia, Charlottesville, VA 22904 USA (e-mail:
mcs5f@virginia.edu).
problem, which is defined as a recent history of hospitaliza-
tion or symptoms of a mental illness; less than 40% of these
inmates received psychiatric treatment prior to their incar-
ceration [1], [2]. This high-maintenance population con-
sumes the limited resources of the criminal justice system
when their treatment needs could often be met more appro-
priately in the community. After an inmate’s release, inade-
quate transitions back to the community can disrupt treat-
ment and increase the likelihood of a crisis situation to de-
velop and possible arrest. For this reason, there is a high re-
arrest rate among individuals with mental illness. Trying to
reduce the number of mentally ill in prisons and jails has
become a major policy issue across the United States.
Studies show that mentally ill inmates have higher rates of
substance abuse or dependence, more prior convictions, and
more complications after arrest than other inmates [1]. Such
complications include more violations of facility rules, vic-
timizations, and injuries than those without mental health
problems. In the Charlottesville area, people with mental
illness, also referred to as consumers, have a high rate of
involvement with the criminal justice system. This is due to
the lack of continuous mental healthcare available from the
agencies comprising the criminal justice and mental health
systems. The Albemarle-Charlottesville Regional Jail and
the Charlottesville, Albemarle County, and University of
Virginia Police Departments comprise the criminal justice
system resources studied by the project team. The key men-
tal health resources analyzed included the University of Vir-
ginia Emergency Department, Region Ten Community Serv-
ices Board, and On Our Own. The coordination of all seven
entities is required to provide a continuum of care to con-
sumers and handle crisis situations safely.
Police officers, usually the first responders to a mental
health crisis, often lack the training needed to understand
mental illness or de-escalate the situation. During encounters
between officers and distraught consumers, misunderstand-
ings can result in feelings of disrespect, physical injuries, or
death. Since officers are often unable to recognize signs of
mental illness or are unaware of the resources available to
them, consumers are often arrested on minor charges, result-
ing in incarceration instead of treatment.
If the officer decides not to arrest because he recognizes
behavior indicative of a mental illness, he or she must hold
the individual in custody for a psychiatric evaluation, which
can take up to eight hours. Long wait times frustrate the of-
ficers since they are taken away from what many believe to
be their primary responsibility—patrolling the streets. Also,
the condition of the person in mental crisis can be exacer-
Michael M. Ordonez, Kara M. Worrest, Mimi S. Krauss, Lauren N. Tietje, Reid Bailey, Michael C. Smith
Mental Health Resources and the Criminal Justice System:
Assessment and Plan for Integration in Charlottesville, Virginia
I
bated by the chaotic nature of the emergency department
waiting room. This situation endangers police officers, hos-
pital staff, the consumer, and other patients.
If a person with mental illness is arrested, they have the
option of receiving mental health medications while in jail.
When an individual is being released from jail, only infor-
mal coordination currently exists between the jail and com-
munity mental health resources. Upon release, the consumer
is concerned with finding housing, food, and employment,
meaning that filling expensive prescriptions is a low priority.
Without assistance, most released inmates will stop taking
mental health medications and relapse.
While each organization provides great service in its area
of expertise, Charlottesville’s current system for handling
mental health crises is not as effective as it should be; sev-
eral agencies which care for people with mental illness lack
the necessary coordination between one another for success.
This project assesses the current use of mental health and
criminal justice systems resources in handling situations
involving mental illness. To address the problems in the cur-
rent system, the team suggests the realignment of existing
resources into a Crisis Intervention System (CIS) capable of
managing the care of acute crisis situations while also pro-
viding a continuum of treatment to consumers. The CIS
would also use resources effectively and efficiently through
the integration of key agencies. To achieve this goal, the
team evaluated the current system to identify problems and
then developed alternatives to improve the current system.
The final product of this work is a redesigned Crisis Inter-
vention System and a plan for implementation in Charlottes-
ville and Albemarle County.
II. GOALS AND OBJECTIVES
A. Goals
The initial goal of the project was to assess the need for a
crisis intervention site in Charlottesville and Albemarle
County. Such a site would function as a psychiatric emer-
gency room and be designed to facilitate the completion of
mental health evaluations. However, the goal of the team
quickly evolved beyond this simple needs assessment. Based
on preliminary analysis, the team recognized the importance
of addressing a much larger issue: the interactions between
the criminal justice and mental health systems. The team’s
goal then became the focused on the development of a Crisis
Intervention System (CIS); a CIS seeks to improve the de-
livery of mental health services to people in crisis and to
coordinate mental health resources with those of the criminal
justice system. Through this alignment of resources, the CIS
would address the acute symptoms of individuals in psychi-
atric crisis while also minimizing the chance for crises to
develop by providing long-term stability through a contin-
uum of mental health care. Furthermore, the CIS would re-
align existing services in the most effective and efficient
way possible, to maximize the return from limited resources.
B. Objectives
Five objectives were developed to measure the city and
county’s achievement of a CIS, shown in Figure 1. Each of
these main objectives was then broken down into several
measurable sub-objectives.
Improving agency coordination consists of improving the
transition between agency programs and increasing the
awareness of treatment options within other agencies.
Improve relationship between mental health community
and the community at large consists of increasing the
knowledge of available resources, reducing community
stigmas through education, and continuously informing the
community of CIS developments.
Improve quality of life for people with mental illness in-
volves increasing mental health service utilization and ac-
cess to medications, decreasing substance abuse, improving
Fig. 1. Crisis Intervention System Objectives. This objective tree shows the 5 key objectives of a CIS, which would measure the success of Charlottes-
ville and Albemarle County’s crisis intervention.
overall mental condition, decreasing homelessness, and de-
creasing violence and victimization.
Increase overall safety during crisis situations involves
reducing the use of unnecessary force by police officers,
improving de-escalation techniques, and reducing the num-
ber of injuries incurred by all parties present (police offi-
cer(s), consumer, family members, bystanders, etc.).
Improve data collection and usage entails the creation of a
single database to capture data needed for CIS evaluation,
converting all agencies to electronic data collection, use of
consistent terms on forms across all agencies, and collection
of data necessary for system evaluation.
III. TECHNICAL APPROACH
A. Process Flow Models
The team developed an understanding of how the current
system functions by creating process flow models, and then
using them as the basis for stakeholder discussions to iden-
tify problems in the system. These flow models use decision
and action nodes to show how a consumer moves through a
single agency or from one agency to another. Models were
created to outline current procedures for each of the system
agencies, starting from when a police officer arrives on
scene to handle a mental health crisis. These models were
based on protocols and the experiences of individuals work-
ing within each system, including police officers, emergency
department doctors, and staff at the regional jail. The process
flow models create a visual understanding of the decisions
an officer makes during a mental health crisis, and how that
effects where the consumer goes. For example, Figure 2
outlines the steps and decisions that occur after a police offi-
cer issues an Emergency Custody Order (ECO) and takes the
person to the University of Virginia Hospital.
B. Interaction Models
While the process flow models provided insight into how a
consumer interacts with different organizations, they did not
capture how different agencies interacted with one another.
Interaction models were created to highlight the relation-
ships between organizations. Each model focused on a single
agency and created a visual flow of how this agency inter-
acted with each of the other agencies to serve a consumer in
crisis.
The interaction model in Figure 3 diagrams the interac-
tions between the jail and each of the other major organiza-
tions in the current system. The solid lines represent rela-
tionships between two agencies, such as the flow of services
or movement of consumers from one agency to another. The
dotted lines represent new relationships formed as a result of
the team’s work. Two such changes shown in Figure 3 in-
clude On Our Own’s peer-support and recovery education
programs at the jail and the community resource card used
by police to refer consumers to the area’s mental health re-
sources.
C. Data Analysis
In addition to the process flow and interaction models, the
project team conducted a thorough analysis of available data.
Emergency Custody Orders (ECOs) and Temporary De-
tention Orders (TDOs) are orders issued that give law en-
forcement the authority to take a person with mental illness
into custody for a mental health evaluation. Involuntary hos-
pitalization results if the evaluator at Region Ten finds that
the individual meets certain criteria. Analysis performed on
these data sets examined the relationships between the dif-
ferent police departments in the area and consumer facilities,
such as the UVA Hospital, Region Ten, and the Martha Jef-
ferson Hospital.
Fig. 2. Police Officer—University of Virginia Emergency Department. This process model outlines the processes that occur once an individual is taken to the
University of Virginia Emergency Department to receive a psychiatric evaluation.
The Regional Emergency Communications Center pro-
vided data on police officers responses to calls involving
mental illness over the past three years. The team looked at
officer response times, the number of officers on a case, and
the costs associated with the man-hours spent on each case.
The Emergency Department from the UVA Hospital pro-
vided data on all psychiatric cases it served in the first two
quarters of 2006. This data allowed the team to identify the
arrival times of cases to the department, the cost of treat-
ment, and the amount paid to the hospital. This data is valu-
able in determining the cost of mental health treatment for
Charlottesville, as the University Emergency Department is
a public institution which must serve patients regardless of
their ability to pay for treatment.
The Albemarle-Charlottesville Regional Jail provided two
documents for analysis: statistics from the pharmacy and
statistics from the jail’s mental health department. The
pharmacy documents included the number of prescriptions
filled, the number of inmates receiving prescriptions, and the
costs of medication. Data from the mental health department
supplied the number of evaluations completed, the number
of inmates taking psychotropic medication, and the percent-
age these inmates represented of the total jail population.
The analysis of this data determined the amount of money
and resources invested by the jail into treating inmates with
mental illness.
D. Development of Observations through Discussion
In addition to conducting data analysis and modeling the
relationships between agencies, the project team met with
groups representing each of the system’s diverse
stakeholders. Through discussions with these stakeholders,
the team came to understand the full complexity of the sys-
tem and the importance of improving interactions between
the criminal justice and mental health systems. Each
stakeholder represented an entirely different view of the cur-
rent system and expressed a distinct vision of how the ideal
system would work. The challenge for the project team was
to combine all of these perspectives and needs into a single
system that would address all of the major problems identi-
fied in the existing system. Hence, the stakeholder discus-
sions played a significant role in developing the team’s ob-
servations and improvement recommendations.
Additional observations resulted when the team discov-
ered that data on the current system was lacking or of poor
quality. This posed a problem for assessing the current use
of resources in Charlottesville and Albemarle County. More
importantly, without an accurate analysis of the region’s
current resource usage, the success of changes implemented
by the city and county would have no starting point from
which to gage their progress. While the team analyzed the
data that was available, they also recommended what im-
provements in data collection and usage were needed in or-
der for the system’s progress to be evaluated in the future.
IV. FINDINGS AND OBSERVATIONS
While each of the agencies provided effective mental
health services on their own, the project team discovered
that the transition of a consumer from one agency to another
left too much room for treatment gaps to occur. Although
this was difficult to capture with available data, in discus-
sions with stakeholders based on the process flow and inter-
action models, the importance of improving these inter-
agency transitions was revealed. When gaps in treatment
occur, the individual with mental illness is likely to relapse
and result in a crisis situation. This represents a major set-
back in mental health treatment for the consumer and a large
expense for the criminal justice and mental health systems.
A. Interaction between the Jail and Community Mental
Health Resources.
On average, 14% of the inmate population of the Albe-
marle-Charlottesville Regional Jail received psychotropic
(mental health) medication. The cost per inmate for these
medications was about $171, which amounted to nearly
$13,000 spent by the jail each month to treat mental illness.
As a form of mental health treatment, psychotropic medica-
tions are only effective if used consistently. When individu-
als are released from jail, they receive three days of medica-
tions and a one month prescription. However, from
stakeholder discussions, the team found that upon their re-
lease from jail, finding food, housing, and employment takes
precedence over acquiring expensive medications. If their
prescription is not filled within the first three days of their
release, the individual’s mental illness may relapse. The gap
in treatment caused by these situations squanders all of the
treatment and money invested by the jail prior to the indi-
vidual’s release.
B. Interaction between the Police Departments and Re-
gion Ten
Data analysis on the police officer disposition locations in-
dicated that police officers do not frequently take individuals
in crisis to Region Ten for a psychiatric evaluation. Only 3%
Fig. 3. Interaction Model for the Jail. Interactions between key agen-
cies of the criminal justice and mental health systems were captured
by using lines to represent the flow of individuals or services from
one agency to another.
of cases with reported locations were taken to Region Ten,
whereas the UVA Hospital accounts for 30% of all cases.
Using the t-test for proportions at a significance level of
0.01, there is a significant difference (t-value = 23.073, p-
value < 0.001) between the number of cases taken to Region
Ten and the UVA Hospital. The UVA Emergency Depart-
ment being overcrowded could be a result of the lack of co-
ordination between the police departments and Region Ten.
The number of beds available for mental health treatment
decreased by 86% from 1955 to 1999, meaning that over-
crowded hospitals cannot provide an adequate number of
beds to individuals in need [3]. Region Ten facilitates an
average of 32 involuntary hospitalizations each month, and
an additional 15 voluntary hospitalizations. In some cases,
an individual presents a serious risk but cannot be hospital-
ized because no beds are available. The criteria for an invol-
untary hospitalization are that the individual is an imminent
danger to himself or others or the individual is so seriously
mentally ill as to be unable to care for himself. During the 21
month period beginning in January of 2005, in four separate
months Region Ten could not find beds for as many as four
people in serious need of hospitalization. This endangers
both the public and the person with mental illness, and rep-
resents a critical problem in the current system.
C. Police Department Costs
Based on the number of officers per case and the time
spent once an officer is dispatched, 9,660 labor hours were
spent on cases involving mental illness the last three years.
Using the average hourly salary of $20 for a police officer,
this accounts for $193,000. In 25% of the cases police offi-
cers spend more than two and half hours on a case, which
amounts to $128,000 in labor costs. This means that 66% of
the mental health salary costs are being spent on only 25%
of the cases. It must be noted that the provided data only
included cases marked with the mental illness code and is
expected to be an under representation of the actual number
of cases involving mental illness. The amount of money
spent on workman’s compensation is also not included in
this analysis because the police departments do not record
the number of workman’s compensation cases that are a
result of interactions with people with mental illness.
D. University of Virginia Emergency Department
From discussions with stakeholders, the project team
found that the Emergency Department (ED) is hectic, loud,
and crowded, which can trigger an individual with mental
illness to become agitated and cause a disturbance. Because
the facility was not built with these specific episodes in
mind, hospital equipment is accessible and can be thrown or
used in a potentially violent way, which puts the consumer,
police officers, hospital personnel, and other patients at risk.
Not only does the ED present serious safety risks, but the
long wait times frustrate police officers because they are
kept off their patrol until the consumer’s mental health
evaluation is complete. In addition to safety concerns and
officer wait times, the ED must also handle the problem of
losing money on most psychiatric cases. Since the Univer-
sity’s hospital is public, the ED must provide treatment to
patients regardless of their ability to pay. Data analysis
shows that the hospital loses an average of $83 per psychiat-
ric case since 34% of psychiatric patients have no insurance.
V. RECOMMENDATIONS
The project team’s strongest recommendation to the city
and county is the re-alignment of existing mental health
agencies and the criminal justice system to form a Crisis
Intervention System. The objectives of the CIS provide clear
guidance for the direction in which the region must move to
improve safety, use resources effectively, and enhance the
delivery of treatment to consumers.
The team also identified specific areas upon which the CIT
Taskforce should focus their efforts in the immediate future.
The first area was in improving the transition back to the
community of inmates with mental illness upon their release
from jail. The remainder of the team’s recommendations
focuses on ways to improve data collection in the region.
Not only is such data needed to complete the evaluation of
the current system, but also so that the impact of the CIT
Taskforce’s changes can be evaluated in the future.
A. Jail and Community Mental Health Resources
To prevent wasting the costly mental health services pro-
vided to inmates by the jail, the transition after the inmates’
release back to the community has to be improved. The team
recommends the creation of two liaison partnerships to aid
the consumer in this transition. The first liaison would be
created at On Our Own and the second at Region Ten, both
community mental health resources. As an individual’s re-
lease date approaches, the jail’s mental health staff will meet
with the inmate to discuss ways to continue treatment once
released and explain how the liaisons can help. If the indi-
vidual accepts the help of the liaisons, then the jail’s mental
health staff will notify each liaison of the inmate’s upcoming
release. Upon release, the On Our Own liaison will focus on
helping the individual to obtain medication and navigate the
community resources available to find housing, food, and
employment. Meanwhile, the Region Ten liaison will be
responsible for helping the consumer to access Region Ten’s
programs and get new prescriptions as quickly as possible.
Although specific names cannot be tracked between the
jail system and the community mental health resources, data
can be collected to determine the utilization of these pro-
grams upon release. If the jail collects data on the number of
individuals that say they will meet with Region Ten and Re-
gion Ten compiles data on the number of individuals that are
seen upon release from jail, then the two agencies can com-
pare these numbers to indicate the number of individuals not
continuing their mental health treatment. As the liaison pro-
gram improves, the team expects consumers to spend more
time in the community with fewer crises. A decrease in the
number of re-arrests of consumers who used the program or
an increase in the amount of time between release and re-
arrest would both indicate the success of the improved tran-
sition.
B. Data Collection Needs for Police Departments
In 24% of cases involving mental illness, the available data
had no recorded end location, limiting the benefits of the
data analysis. The team recommends that all three police
departments establish protocols outlining the exact way for
disposition locations to be entered.
First, it should be required for all cases that are closed to
include the location, which would eliminate the problem of
having 24% of the location data missing. Second, a uniform
location coding system should be imposed that will limit the
disposition categories to six locations—Arrest, UVA Hospi-
tal ED, Region Ten, Martha Jefferson Hospital, Settled on
Street, and Other. A uniform coding system would allow for
better, more robust analysis since the number of assumptions
would be reduced. Lastly, after an ECO is issued, the out-
come of the psychiatric evaluation should be recorded,
which will indicate if the individual was released or if the
individual was voluntarily or involuntarily hospitalized.
Once data is collected using these protocols, the project
team recommends that additional analysis be completed in
order to gain a better, more accurate insight of the disposi-
tion locations.
C. University of Virginia Emergency Department
To ease the impact on the ED of serving psychiatric clients
while operating well above capacity, the police departments
must improve their use of Region Ten’s services. The lack of
police officer awareness of community resources was re-
vealed during discussions with stakeholders and represents a
simple starting point for improvements. The ED can contrib-
ute to the evaluation of the system as a whole by tracking the
wait times for its psychiatric cases. This will provide data
from which to gauge the impact of changes implemented to
decrease officer and consumer wait times.
VI. CONCLUSION
The project team assessed the current use of resources in
response to mental illness by the criminal justice and mental
health systems in the Charlottesville area. Three problems
identified in the current system were unnecessary safety
risks, the over-consumption of limited resources by mental
health situations, and gaps in treatment which led consumers
to relapse. To address these problems, the team recommends
the city and county realign existing resources into a Crisis
Intervention System (CIS) capable of managing the care of
acute crisis situations while also providing a continuum of
treatment to consumers. The CIS would also use resources
effectively and efficiently through the integration of key
agencies. Uniform and electronic data collection is a key
component of an effective CIS. Without proper data collec-
tion, evaluation of the system cannot be completed.
ACKNOWLEDGMENT
The project team would like to thanks Tom von Hemert
and the members of the CIT Taskforce for their time and
dedication to improving the emergency mental health system
in Charlottesville and Albemarle County, Virginia. The team
would like to thank Nathan Veldhuis and the members of the
Mental Health Law Clinic at the University of Virginia Law
School for their cooperation and coordination with the
evaluation team.
REFERENCES
[1] James, D. J., & Glaze, L. E. (2006). Mental Health Problems of
Prison and jail inmates. Bureau of Justice Statistics: Special Report.
[2] Hails, J., & Borum, R. (2003). Police Training and Specialized Ap-
proaches to Respond to People with Mental Illnesses. Crime & Delin-
quency, 49(1), 52-61.
[3] Kupers, T. A. (1999). Prison madness: The mental health crisis be-
hind bars and what we must do about it. San Francisco: Jossey-Bass
Publishers.
