Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 2 of 164

BALTIMORE CONSENT
DECREE MONITORING TEAM
FIRST COMPREHENSIVE
RE-ASSESSMENT

September 30, 2020

 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 3 of 164
TABLE OF CONTENTS
OVERVIEW3

INTRODUCTION 4

EXECUTIVE SUMMARY 8

FINDINGS23

Training25

Misconduct Investigations And Discipline 30

Technology37

Staffing, Performance Evaluations And Promotions 40

Stops, Searches, Arrests And Voluntary Police-Community Interactions 44

Impartial Policing 50

Use Of Force 53

Transportation Of Persons In Custody 56

Interactions With Individuals With Behavioral Health Disabilities And In Crisis 59

First Amendment-Protected Activities 67

Interactions With Youth And Coordination With Baltimore School Police 74

Community Policing And Engagement 78

Sexual Assault Investigations 82

Recruitment, Hiring And Retention 86

Officer Assistance And Support  90

SUMMARY OF MONITORING TEAM ACTIVITIES 95

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 2
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 4 of 164
OVERVIEW
THE MONITORING TEAM’S KEY FINDINGS:
Overall
„ The Consent Decree is working as designed, with BPD and the City making substantial progress on foun-
dational reforms in policy revisions, training and planning due to the steady, stabilizing work of the Con-
sent Decree Implementation Unit and the committed leadership of Commissioner Harrison, his Deputy
Commissioners, a new Training Academy commander, a new Chief Technology Officer and others
„ BPD’s substantial progress on foundational reforms has given the Monitoring Team greater confidence in
BPD’s capacity to achieve compliance
„ While promising, the threshold reforms have not yet translated into consistent, observable change on the
street, as training on the reforms is not complete and BPD’s outdated record-keeping systems continue to
render inaccessible or unusable the key data required for rigorous performance assessment

Specific Areas
„ BPD has successfully revised most Consent Decree-mandated policies
„ BPD has fortified its Training Academy by adding sworn and civilian staff, moving to modern facilities at
the University of Baltimore, and adopting new training methods that employ adult learning principles
„ BPD has conducted effective, Consent Decree-mandated in-class training and/or e-learning on use of force,
impartial policing, stops, searches and arrests, body-worn camera use, and sexual assault response
„ Upcoming training initiatives are ambitious, particularly during a pandemic, and will present implemen-
tation challenges
„ BPD’s partnership with community stakeholders on the Collaboration Planning and Implementation Com-
mittee has generated an impressive body of work—new crisis intervention policies and training curricula,
new data requirements for crisis events, an analysis of the gaps in the City’s behavioral health support
system (“gap analysis”)—and established a practical guide to systemic reform
„ Other agencies must match BPD’s commitment to implementing the recommendations in the gap analysis
„ BPD’s tempered, lawful response to recent large-scale protests is a preliminary, if limited, indicator of the
Consent Decree’s positive impact
„ The overhaul of BPD’s internal investigations and discipline functions is slow but progressing
„ Staffing needs, especially in Patrol and Public Integrity, are acute and continue to require urgent action
„ Necessary improvements in BPD’s antiquated IT systems and historically deficient IT governance struc-
ture are gaining steam after some delay, but remain a year or more from completion and thus hamstring
the ability of the Monitoring Team and BPD itself to conduct Consent Decree-required analysis of police
encounters, particularly stops, searches and arrests
„ Implementation of a robust community policing model has been halting and, even with a sound new com-
munity policing plan, will be extremely challenging given community mistrust, the persistence of violent
crime and Patrol officer shortages

The Next Six Months

„ BPD will provide an array of Consent Decree-mandated training on various subjects
„ BPD will finalize nearly all remaining policies requiring revision
„ BPD will begin developing new records management and learning management systems, enhance its
workforce management system, and identify requirements for an Early Intervention System
„ The Monitoring Team will conduct qualitative reviews of use of force incidents and quantitative assess-
ments in areas where reliable BPD data is available

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 3
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 5 of 164

INTRODUCTION

THE CONSENT DECREE

In May 2015, the Civil Rights Division of the United States Department of Justice (“DOJ”)
initiated an investigation of the Baltimore Police Department (“BPD”). The investigation, completed
in 2016, found that BPD was engaged in a pattern-or-practice of constitutional violations, including
using excessive force, infringing on the First Amendment freedoms of speech and assembly, and stopping,
searching, and arresting people in violation of the Fourth Amendment and based on their race. After
making these findings, DOJ entered into negotiations with BPD and the City of Baltimore in an effort
to settle the parties’ differences. BPD and the City did not admit DOJ’s allegations, but they recognized
that the allegations raised long-standing issues of considerable importance to City residents. As a result,
BPD and the City agreed to resolve DOJ’s allegations through a Consent Decree. The Consent Decree is
a court-approved settlement agreement between DOJ, the City and BPD. United States District Court
Judge James K. Bredar is the judge who approved the Consent Decree. Judge Bredar now oversees the
Consent Decree’s implementation. Because the Consent Decree is a court order, Judge Bredar has the
power to enforce its provisions and ensure that BPD and the City do what it requires.

The Consent Decree obligates BPD and the City to adopt a comprehensive set of reforms designed to
promote fair and constitutional policing, rebuild BPD’s relationships with Baltimore’s communities, and
ensure public safety. The Consent Decree prescribes corrective action in a number of areas, including:
community engagement; community policing; stops, searches, arrests, and voluntary police-community
interactions; impartial policing; interacting with people with behavioral health disabilities and in crisis;
use of force; interactions with youth; transportation of persons in custody; First Amendment protected
activities; handling of reports of sexual assault; technology; supervision; misconduct investigations and
discipline; coordination with Baltimore City School Police; recruitment, hiring, and retention; staffing,
performance evaluations, and promotions; and officer assistance and support.

The Consent Decree, in short, requires transformational institutional change. BPD will achieve
compliance with the Consent Decree and free itself from Court oversight when it demonstrates not only
that it has successfully implemented all of the required foundational improvements in policies, training,
technology and operations, but that those improvements have translated, measurably and sustainably,
into constitutional, community-oriented policing.

Achieving transformational change in a large police department does not happen overnight. As the
Consent Decree envisions, it takes time, and it requires adherence to a rigorous, methodical reform
process. In each area of the Consent Decree that addresses how officers discharge their duties (e.g.,

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST RE-ASSESSMENT | SEPT. 30, 2020 4
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 6 of 164
stops/searches/arrests, use of force, and transportation of persons
in custody, to name a few), BPD first must draft and adopt revised
policies. Then BPD must develop and conduct training on those
THE CONSENT
revised policies. At the same time, to ensure that the new policies DECREE, IN
and the new training take root, BPD must revamp vital components SHORT, REQUIRES
of its infrastructure. For instance, BPD must overhaul its technology
to become a modern, data-driven, efficient police force, must fortify TRANSFORMATIONAL
its system of internal investigations and discipline to enhance officer INSTITUTIONAL
accountability, must improve the training and supervision of rank-
and-file officers to ensure lawful, effective job performance, and
CHANGE.
must deploy its officers and improve recruiting and retention so as
to simultaneously enhance public safety and promote community-
oriented policing. It is only after officers have been trained on the
new policies, and after infrastructure upgrades are well underway, that community members can expect
to see sustained, tangible changes in the conduct of BPD officers. The Consent Decree contemplates that
this process will take several years or more.

THE MONITORING TEAM

On October 3, 2017, Judge Bredar appointed a Monitoring Team to assist him in overseeing implementation
of the Consent Decree. The Monitoring Team consists of a lead monitor, Kenneth Thompson, and a team
of experts in policing and police reform, civil rights enforcement, psychology, social science, organizational
change, data and technology, and community engagement. Serving as an agent of the Court, the Monitoring
Team plays three principal roles: arbiter, technical advisor, and facilitator. As arbiter, the Monitoring
Team oversees the day-to-day efforts of BPD and the City to comply with the reforms the Consent
Decree requires. The Monitoring Team reviews, provides feedback on, and ultimately recommends Court
approval or disapproval of the changes BPD makes in its policies, its training and, ultimately, its policing
practices. As technical advisor, the Monitoring Team draws upon decades of collective experience to
provide BPD with technical assistance, including advice about national best practices, to help guide BPD
toward satisfying the requirements of the Consent Decree. As facilitator, the Monitoring Team seeks
to ensure that all stakeholders from within BPD and across Baltimore’s diverse communities have the
opportunity to participate in the reform process (CD 442).1

The Court and the Monitoring Team are not alone in overseeing BPD’s implementation of the requirements
of the Consent Decree. DOJ continues to play an active role. As the plaintiff in the lawsuit that produced
the Consent Decree, DOJ retains the right to enforce the Consent Decree when BPD fails to comply with
its terms. Accordingly, like the Monitoring Team, DOJ is assessing BPD’s progress toward compliance
and lets BPD, the Monitoring Team and the Court know when it believes BPD is making progress and
when it believes BPD is not. In addition, like the Monitoring Team, DOJ provides technical assistance to
BPD as BPD works toward compliance. The reform process under the Consent Decree thus involves four
fully-engaged entities: BPD, the City, the Monitoring Team/the Court, and DOJ.

1 All citations to a specific paragraph of the Consent Decree follow the text that relies on that paragraph and appear in
parentheses containing “CD” and the number of the cited paragraph. Thus, the citation above, which is to Paragraph 442 of the
Consent Decree, follows the relied-on provision of Paragraph 442 and appears as “(CD 442).”

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 5
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 7 of 164
THIS REPORT
Every two years, the Monitoring Team is required to prepare a “comprehensive re-assessment to determine
whether the material requirements of [the Consent Decree] have been achieved,” to recommend any
modifications to the Consent Decree that are necessary to achieve or sustain compliance when “changed
circumstances or the unanticipated impact (or lack of impact)” of any requirement are interfering with
compliance, and to “address areas of greatest achievement and the requirements that appear to have
contributed to this success, as well as areas of greatest concern, including strategies and Technical
Assistance, for accelerating Full and Effective Compliance.” (CD 469-70).

This report is the Monitoring Team’s first comprehensive re-assessment. It incorporates the findings from
our first four semi-annual reports (issued in July 2018, January 2019, July 2019 and January 2020), as
well as our findings regarding the performance of BPD and the City since January 2020. Consistent with
the requirements of Paragraph 469 of the Consent Decree, this report explains (1) the Monitoring Team’s
evaluation of the progress BPD and the City are making toward compliance with the requirements of the
Consent Decree in each area of reform; (2) the challenges BPD and the City will continue to face as they
strive to achieve compliance with Consent Decree requirements in each area; (3) the areas of greatest
achievement and the Consent Decree requirements that we believe have contributed to BPD’s and the
City’s success, as well as the areas posing the greatest challenges and the steps we believe BPD and
the City must take to meet them; and (4) the Consent Decree requirements that BPD and the City are
expected to address in the immediate future.

Beginning with this report, the Monitoring Team will not only provide a narrative assessment of the
progress BPD and the City are making in each area of reform, but a compliance score, too. The compliance
scores in each area are reflected in the spreadsheet in Exhibit 1 and included in the body of this report.
The compliance scoring framework is as follows:

0 - Not Assessed: The Monitoring Team has yet to assess if the City/Department has made progress or
complied with the requirement.
 
1 - Not Started: The City/Department has not yet demonstrated progress toward implementing the
requirement, possibly in order to work on other, necessary projects. 
 
2 – Planning/Policy Phase: The City/Department is addressing the planning and/or policy provisions for the
requirement.
 
3 – Training Phase: The City/Department is addressing the training provisions for the requirement, based on
approved policy.
 
4 – Implementation Phase: The City/Department is in the implementation phase for the requirement, having
developed any required plan or policy and conducted any required training, but has not yet demonstrated
compliance with the requirement.
 
4a – Implementation - Not Assessed: The City/Department has initiated the implementation phase
for the requirement, but the Monitoring Team has not yet assessed the City/Department’s progress in
implementation.

4b – Implementation - Off Track: The City/Department is not making satisfactory progress toward compliance
with the requirement.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 6
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 8 of 164
 
4c – Implementation - On Track: The City/Department is making satisfactory progress toward compliance with
the requirement.
 
4d – Implementation - Initial Compliance: The City/Department has demonstrated compliance with the
requirement but has not sustained compliance for the required time period specified in paragraph 504 of the
Consent Decree.
 
5 – Sustained Compliance: The City/Department has complied fully with the requirement and has
demonstrated sustained compliance by consistently adhering to the requirement over time, as defined in
paragraph 504 of the Consent Decree.

When reading this report and examining the compliance scores, again bear in mind that achieving
transformational change in a large police department takes years and requires adherence to a rigorous
process for reform. Thus, two and a half years into the reform process, which began with the approval
of the First-Year Monitoring Plan in February 2018, BPD and the City have put in place many of the
essential building blocks for sustainable change; they have, in fact, come a long way. But they still have
a long way to go. That fact is reflected in their compliance scores. In most areas of reform, BPD and the
City do not score above a “3”—they have not yet completed the “training phase” of reform—though in a
couple of areas, including use of force, they score a “4a” and have moved to the “implementation phase.”
See Exhibit 1.

This does not mean BPD and the City are off track. To the contrary, they are mostly on track. Prior to
March 2020, when the coronavirus pandemic required cessation of in-service training for four months,
as well as a reallocation of certain BPD resources, BPD and the City met the vast majority of the
deadlines and satisfied most the requirements in the Monitoring Plans. Even amid the pandemic and the
constraints it has imposed, BPD and the City have largely kept pace. Nevertheless, as the Monitoring
Plans implicitly acknowledge, most Consent Decree requirements simply cannot be satisfied in three
years. BPD, its officers, the City, community members, the Monitoring Team and DOJ need sufficient
time and opportunity to focus on each area of reform, and on each requirement within each area, to ensure
that change is real and enduring. Change that is rushed, haphazard and superficial is not sustainable
and does not qualify as true reform.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 7
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 9 of 164

EXECUTIVE SUMMARY

With the approval of the First-Year Monitoring Plan in February 2018, BPD and the City started down
the long road to Consent Decree compliance. Over the past two and a half years, they have covered a lot
of ground and made good progress. They have finalized most key policies, are devising and implementing
core training lessons, and have completed studies and implementation plans on technology, staffing,
officer recruitment/hiring/retention, youth diversion, community policing, the City’s behavioral health
system, and BPD’s relationship with Baltimore School Police. These foundational reforms are not fully
completed, particularly in the area of training, which has entered a critical phase. That is significant
because revised policies cannot take effect until officers have been trained on them. Nevertheless, the
foundational reforms are advancing toward completion at an appropriately aggressive pace, even in the
face of a global pandemic that has taxed BPD’s resources and altered normal operations.

The Monitoring Team has previously observed that the Consent Decree’s foundational reforms, while
time-consuming and labor-intensive, are the easy part of the reform process. The hard part—sustainable,
tangible improvement in organizational and officer performance—lies ahead. But the fact that policy
revisions, training and planning are comparatively easy does not make them any less vital. Without
clear policy to guide officers and hold them accountable, without effective training to ensure adherence to
policy on the street, and without detailed plans for methodically addressing organizational deficiencies
in components ranging from technology to internal affairs, the Consent Decree’s long-term goal of
constitutional, community-oriented policing cannot be achieved.

From the beginning, the Monitoring Team has had little doubt about BPD’s commitment to pursuing
this goal. It showed its commitment by diving headlong into drafting policies, conducting studies and
preparing plans. Yet because of the overwhelming systemic challenges BPD faces, the Monitoring Team
questioned BPD’s capability to reform itself. Over the past year, our concerns have begun to recede.
We have grown more confident that BPD’s capability to reform matches its commitment to reform. Our
growing faith stems from two interrelated organizational developments: the fast maturation of a well-
structured, fully accountable Consent Decree Implementation Unit (“CDIU”) and the installation of a
police commissioner and command staff that is devoted to action, not mere words.

CDIU, whose creation is mandated by the Consent Decree (CD 481), deserves considerable credit for
BPD’s progress to date. BPD went through three different police commissioners and revolving command
staff between October 2017, when the Monitoring Team was appointed, and April 2019, when current
Commissioner Michael Harrison took over. Amid these head-spinning changes, CDIU was a stabilizing
force. Engaging other BPD components as needed, it pressed forward with key policy revisions, important
planning documents, and a new approach to training. There was no slow-rolling or delay. Despite the

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 8
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 10 of 164
instability in leadership, BPD met nearly every single Monitoring Plan deadline for dozens of deliverables
over the first 18 months (all except the deadlines for the two deliverables—the now-completed Staffing
and Community Policing Plans—that relied directly on the vision of a permanent commissioner). This
early progress surpassed what most other agencies under consent decrees have been able to accomplish
in a similar amount of time.

In the Monitoring Team’s view, CDIU has been able to push

BPD forward in the face of significant challenges because it has NEVERTHELESS, THE
had a sufficient number of qualified personnel to do the work, FOUNDATIONAL
has assigned each staff member to a specific area of the Consent
Decree, and has held that member accountable for the deliverables
REFORMS ARE
in that area. In other words, from the beginning, CDIU has been ADVANCING TOWARD
well-resourced and well-managed. In drafting the Consent Decree, COMPLETION AT
the City, BPD and DOJ were astute to mandate the creation of
a permanent, properly supported BPD unit to manage Consent AN APPROPRIATELY
Decree compliance. AGGRESSIVE PACE,
EVEN IN THE FACE OF
CDIU’s work has proceeded apace under Commissioner Harrison.
In fact, while BPD was making progress before Commission A GLOBAL PANDEMIC
Harrison took over, the breadth and depth of the reform enterprise THAT HAS TAXED
has appreciably accelerated under his command. The evidence is
readily apparent to the Monitoring Team. Having successfully BPD’S RESOURCES AND
steered the New Orleans Police Department through the early ALTERED NORMAL
and middle stages of its own consent decree before he came to OPERATIONS.
BPD, Commissioner Harrison has surrounded himself with highly
competent, committed deputies and other top-level command who
have begun revamping key departmental functions. For example:

„ The Training Academy has been transformed, as described in more detail below.
„ Previously stalled improvements in technology and IT governance, essential to modernizing BPD
and achieving Consent Decree compliance, are gaining steam. Commissioner Harrison hired an
experienced Chief Technology Officer, who in turn has hired new technical personnel and reorganized
the Technology Section based on well-defined assignments of responsibility. Among other things,
the Section has adopted a more vigilant approach to IT governance and project management; begun
implementing a workforce management system to facilitate scheduling and resource deployment and
a learning management system to track and deliver officer training; and, crucially, has hired a vendor
to design and install a new records management system (“RMS”) scheduled to be completed next
year. The new RMS will move BPD from cumbersome, siloed, paper-reporting systems to integrated,
electronic, field-based reporting, which will increase officer efficiency, facilitate more effective
supervision, and enable necessary but previously impracticable analysis of Department-wide, district,
unit and individual officer trends in performance and constitutional policing metrics.
„ There are early—if still inconclusive—signs that the Performance Standards Section, which
conducts internal inspections and audits, has started to function the way it should to monitor
compliance with law and policy. The Section routinely performs transportation vehicle equipment
inspections and random transport event audits. This year, it published its first annual analysis of
responses to First Amendment-protected activity and completed its second annual analysis of sexual
assault investigations. It is also currently finishing up its first quarterly analysis of arrests that result
in individuals being released without charges.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 9
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 11 of 164
„ Commissioner Harrison is acting aggressively to address a culture that has been overly tolerant of
misconduct and poor performance. Sending a strong message about the importance of reform, he
elevated to the position of Deputy Commissioner both the head of the Public Integrity Bureau
(“PIB”), which investigates misconduct, and the civilian head of the Compliance Bureau, which is
responsible for Consent Decree implementation. To bolster the coordination and efficiency of the reform
effort, Commissioner Harrison placed CDIU, the Training Academy, the Technology Section, and the
Performance Standards Section within the same Division—the Compliance Division. Additionally,
one of the first policies Commissioner Harrison adopted and has piloted (Draft Policy 321, also not
required by any Monitoring Plan) authorizes minor allegations of misconduct to be promptly resolved
through acceptance of responsibility and negotiated resolution. The policy promises to increase PIB
efficiency, ensure greater accountability and improve officer morale. The serious challenges BPD
faces in the area of accountability cannot be overstated. They are explained in the next section. But
Commissioner Harrison and his commanders have recognized the imperative of culture change and
have begun attempting to achieve it.
„ To underscore the relationship between ethical policing and officer wellness and support, BPD will
soon be among the first major city police agencies to conduct training and adopt a program on peer
intervention/“active bystandership” called Ethical Policing Is Courageous. Though not required
by the current Monitoring Plan, BPD is also proactively implementing policies mandating a duty to
intervene and prohibiting retaliation for reporting misconduct.
A Consent Decree is only as good as the people implementing it. In BPD’s case, because of CDIU,
Commissioner Harrison and his commanders, the reforms the Consent Decree envisions are advancing.
Thus, at this point, the Consent Decree is working. Its provisions do not need to be amended to accomplish
its objectives. The implementation process should be permitted to unfold as designed.

To be clear, this does not mean—at least not yet—that the Consent Decree ultimately will succeed.
Rather, it means that, thirty months in, early-stage threshold reforms have taken shape and are showing
that BPD has the capability to reform. These reforms have not yet translated into widespread changes in
officer conduct. Indeed, because BPD’s antiquated reporting and record-keeping systems effectively render
key data inaccessible or unusable, the Monitoring Team has not yet been able to meaningfully evaluate
whether such changes are occurring. Nevertheless, the work BPD has done on the Consent Decree’s
foundational reforms is beginning to show that BPD can fix its structural and operational shortcomings
and can provide constitutional, accountable, community-oriented policing to the residents of Baltimore.

Certain areas of the Consent Decree have stood out thus far. Some stand out because they vividly
demonstrate BPD’s capacity for sustainable reform. These include training, responding to individuals
in crisis, and First Amendment-protected activity. Other areas stand out because they continue to
pose serious challenges to the prospects for sustainable reform. These include accountability (which
encompasses misconduct investigations, discipline and supervision) and community policing. We address
these areas below. We explain which features of the Consent Decree are working as designed, as well as
those BPD should focus on if it is to meaningfully progress toward compliance.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 10
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 12 of 164
AREAS OF GREATEST ACHIEVEMENT
|TRAINING
The Monitoring Team reported in July 2019 that BPD’s “rapid progress toward establishing a training
program consistent with national best practices is perhaps the most promising development to date.”
ECF No. 220-1. That was 14 months ago. The performance of BPD’s Education and Training Section—the
Training Academy—has only continued to improve since then, even with a four-month interruption in
in-service training because of the pandemic.

The amount of Consent Decree work the Academy has done over the past two years is considerable. It
is detailed elsewhere in this report. It includes developing and delivering department-wide e-learning
and in-class instruction on use of force, stops/searches/arrests, interviews/interrogations, transportation
of persons in custody, impartial policing, body-worn camera use, and responding to reports of sexual
assault, as well as the ongoing development of e-learning and in-class curriculum on numerous other
Consent Decree topics, to be delivered continuously through the end of 2021.

What distinguishes BPD’s work on training, however, is not simply its quantity. Rather, it is the
qualitative improvement BPD has made in a short time in both the management of its training function
and the quality of instruction. The rapid improvements BPD has made compare favorably to the slower
pace of improvement in other departments under consent decrees.

In the Monitoring Team’s view, the Training Academy’s accomplishments are principally attributable to
two developments: the appointment of additional dedicated personnel, including new leadership, and a
commitment to a modern, more dynamic approach to officer education. The Consent Decree requires both
(CD 292-93, 295-97).

From the beginning, the Monitoring Team made clear to BPD that it would need to increase the number
of instructors assigned to the Academy. There was simply no way BPD could satisfy the intensive training
requirements of the Consent Decree without dramatically increasing the number of full-time instructors.
BPD has delivered. In spring 2019, prior to the first tranche of Consent Decree classroom training, which
covered use of force and impartial policing, BPD assigned ten new police instructors to the Academy.
In early 2020, prior to the beginning of the second major tranche of Consent Decree classroom training,
which covers stops, searches and arrests and additional aspects of impartial policing, BPD hired several
additional law instructors. Recently, it brought on two new curriculum writers, civilian employees who
will be essential to the development of the raft of new training programs scheduled to be delivered over
the next 12 - 15 months.

Importantly, BPD has coupled the augmentation of its Academy staff with the installation of highly
qualified leadership. Commissioner Harrison promoted Martin Bartness to Major and tapped him to
head the Academy. A 23-year BPD veteran, Major Bartness is an experienced, respected leader with
a proven devotion to the mission of modernizing and professionalizing police education. In addition to
promoting Major Bartness, BPD hired a first-ever civilian academic director, Gary Cordner. Director
Cordner is an accomplished academic and police professional with years of experience developing high-
quality, evidence-based education programs for law enforcement agencies. Director Cordner now oversees
the development of new curriculum, including curriculum on Consent Decree topics.

Under the leadership of Major Bartness and Director Cordner, BPD has adopted a new approach to
training. BPD has switched from providing officers all annual in-service training in a single, lengthy block

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 11
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 13 of 164
to offering shorter blocks on different topics throughout the year. This staggered approach, concentrating
on one or two topics at a time, slows things down and facilitates increased retention of training material.
It also allows for greater staffing/scheduling flexibility, reducing the operational impact of taking officers
away from their assigned duties for training for longer periods.

Even more important, the Training Academy is embracing adult learning principles. In so doing, it has
taken advantage of programs and resources from other agencies. For instance, on several occasions,
different Academy personnel have traveled to Los Angeles for programming on adult learning at the
LAPD Academy, a leader in the field. BPD’s new model utilizes e-learning to introduce officers to
departmental policies, followed by scenario-based classroom instruction designed to teach officers how
to apply the policies in practice. Officers may not commence the classroom component until they pass
an online test that accompanies the e-learning. Classroom instruction, previously dominated by static
lectures and dense PowerPoint presentations, now features
dynamic, interactive problem-solving exercises that incorporate
WHAT DISTINGUISHES videos from actual police encounters, written scenarios, in-
BPD’S WORK ON person role playing and digital simulations. Before going live, the
classroom component undergoes a piloting process. The piloting
TRAINING, HOWEVER, process includes sessions with officers, as well as sessions with
IS NOT SIMPLY ITS community members. Notably, BPD formed a Community
Training Review Committee to vet and provide feedback on
QUANTITY. RATHER, draft training curricula.
IT IS THE QUALITATIVE
IMPROVEMENT BPD BPD has used the new format to deliver critical Consent Decree
training on use of force, stops/searches/arrests, transport,
HAS MADE IN A SHORT interviews/interrogations, and impartial policing. It will continue
TIME IN BOTH THE to use the format to deliver training in other Consent Decree
MANAGEMENT OF ITS areas. Based on the Monitoring Team’s first-hand observations,
the results have been encouraging. Academy instructors have
TRAINING FUNCTION gained increasing confidence and ability in delivering the
AND THE QUALITY material through facilitated discussion, as opposed to lectures.
They have also become increasingly adept at addressing difficult,
OF INSTRUCTION. pointed questions from their peers and conveying the rationale
for policies, including Consent Decree-mandated policies, that
have met with skepticism among some participants. Moreover,
officers participating in classroom learning appear engaged.
They are generally not sitting in their chairs with their arms folded or looking at their cell phones
(and instructors are getting better at re-engaging them when they do). Officers are taking part in the
discussions and sharing their views. That is reflected in the results from the surveys officers have taken
at the end of each course. The vast majority have indicated that they value the change in approach and
find it effective. For instance, for the training on stops/searches/arrests, which has now been delivered
to roughly three-quarters of the Department, roughly 85% of officers agree or strongly agree that the
training is engaging, while only 5% disagree or strongly disagree.

In short, the significant strides the Academy has made mean that officers are receiving better, more
practical training on core policing activity. The new training now must translate into practice on the
street.

The Monitoring Team expects the quality of training to continue to improve as instruction using adult
learning concepts becomes further ingrained in Academy operations. The Monitoring Team also expects
the quality of learning to improve as a result of the Academy’s recent move to the University of Baltimore.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 12
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 14 of 164
Early on, the Monitoring Team emphasized the need for BPD to upgrade its training facility, which was
housed in a decrepit old school in Northwest Baltimore. The use of that facility conveyed to officers
and recruits that BPD did not prioritize training, did not value their professional development, and did
not recognize policing as a vocation that requires specialized knowledge and skill. The new facilities
at the University of Baltimore—which are expansive, modern, clean, and well-equipped for the use of
technology–appropriately convey the opposite message.

Although the Academy has made great progress over a short

period, there remain areas for improvement. Classroom instruction AT A GLANCE
sometimes can be uneven, particularly at the start of a course. This
potentially indicates the need to fortify the piloting process. More
fundamentally, it suggests that BPD should consider establishing AREAS OF GREATEST
an in-house facilitator development program, similar to the
program certain Academy instructors have attended at LAPD. The ACHIEVEMENT
quality of curriculum writing, though improving, also must get
better. Curriculum development continues to require substantial
technical assistance from the Monitoring Team and DOJ. The „ TRAINING
recent addition of new curriculum writers should help. Relatedly,
early drafts of training curriculum for specialized units (e.g., sex „ RESPONDING TO
offense unit investigators) have not employed adult learning INDIVIDUALS IN
techniques as readily as Department-wide curriculum drafted
within the Academy. Going forward, BPD must ensure that the
CRISIS
development of specialized curriculum is closely coordinated with „ FIRST AMENDMENT-
Academy personnel and that such curriculum is delivered with or
under the supervision of Academy personnel. PROTECTED
ACTIVITY
Finally, although the Academy has had sufficient staff to deliver
the first two tranches of classroom Consent Decree training and
to prepare curriculum for others, the Monitoring Team is concerned that the Academy’s future staffing
needs will not be met. The Staffing Plan issued earlier this year recommends increasing professional
civilian staff at the Academy by 38 and decreasing sworn staff by 24. Reductions in BPD’s FY21 budget
have slowed efforts to create new civilian positions in line with the Plan. At least in the short run, BPD
will not be able to implement the Plan’s recommendations for the Academy until next fiscal year at
the earliest. As a result, BPD will not be able to civilianize certain Academy functions (and will not be
able redeploy sworn officers to Patrol or PIB, where they are needed) and will continue to operate at a
personnel deficit, even if not one as large as before.

|RESPONDING TO INDIVIDUALS IN CRISIS

In addition to BPD’s progress on training, among the most reassuring developments has been the
groundwork done to address officer interactions with individuals with behavioral health disabilities and
in crisis. In a relatively short period, BPD and the City have satisfied most of the threshold requirements
in this area: appointing a Crisis Intervention Team coordinator; developing a crisis intervention plan;
identifying BPD’s data collection needs on encounters with individuals in crisis and creating a new crisis
intervention form to record such data; revising three different crisis intervention policies; creating four
different sets of training curricula on behavioral health awareness (for recruits, officers, dispatchers/911
call-takers, and specialized CIT officers (the last of which is still in development)); and producing a
painstakingly detailed “Gap Analysis” that identifies the shortcomings in Baltimore’s behavioral health
system and recommends solutions for addressing them.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 13
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 15 of 164
The catalyst for all these achievements has been the coordination of the City, its nonprofit partner
Behavioral Health System Baltimore (“BHSB”) and BPD with community stakeholders on the
Collaborative Planning and Implementation Committee (“CPIC”). The expansion of CPIC, and
the coordination of CPIC with BPD and the City, are required by the Consent Decree (CD 104). CPIC
includes relevant City and State officials, advocacy groups such as Disability Rights Maryland and
the National Alliance on Mental Illness, individuals with lived experience, community mental health
providers, substance use services providers, local hospitals, advocates, and committed philanthropists.
CPIC’s leadership represents key components of the crisis intervention system and includes the CEO of
BHSB and representatives of the Mayor’s Office. CPIC has different committees, each responsible for a
category of Consent Decree requirements (e.g., policy, training, Gap Analysis, data).

The policies, training curricula, studies and data needs analysis that BPD and the City have generated
under the Consent Decree were all prepared by CPIC. In other words, all of them are the result of close
collaboration among BPD, City officials and community members. At the beginning of each monitoring
year, CPIC has devised a workplan to ensure that all Monitoring Plan deadlines for all deliverables in
the area of behavioral health are satisfied and that there is adequate time within those deadlines for
the community stakeholders on CPIC’s committees to contribute. Using this organizational framework,
CPIC has produced an impressive body of work in just over two years. That work is a testament to what
BPD and City officials can achieve by working together with devoted, informed community members to
improve policing and behavioral health support in Baltimore.

CPIC has fostered meaningful discussion about other matters as well. Even before the Gap Analysis was
completed, the City, BHSB and BPD were exploring ways to integrate different programs that provide
support for individuals in need, drawing on the collective experience of CPIC’s members to do so. These
disparate programs focus respectively on homelessness, substance abuse, and mental illness, but all
share the common goal of diverting individuals in crisis from the criminal justice system to the health
care system. The City, BHSB and BPD are aware that individuals can struggle with multiple issues. With
CPIC’s input, they have been looking into how to develop a comprehensive, integrated support system.

The discussions around integrating behavioral health services in Baltimore have included a closer
review of BPD’s dedicated crisis intervention programs. These programs include the Crisis Response
Team (CRT), the Law Enforcement Assisted Diversion (LEAD) Team, and what was formerly known
as the Homeless Outreach Team (HOT). CPIC members have expressed concerns that these programs
present the risk that program officers, despite good intentions, might violate the rights of the programs’
intended beneficiaries during street encounters, intimidate and thus unintentionally discourage them
from seeking help, or paternalistically override their decisions to forgo treatment. BPD has shown a
willingness to engage in constructive discussions around these concerns. In fact, in early 2020, partly in
response to these concerns, BPD eliminated HOT and assigned its officers to CRT. CPIC members have
discussed moving to a system that ensures CRT responds to calls involving individuals in crisis and seeks
to connect them with support services when appropriate, but does not affirmatively initiate contact like
HOT did, leaving such contacts to non-police outreach workers.

The Monitoring Team has participated in CPIC discussions regarding BPD’s crisis intervention programs
and the City’s support systems. We have been impressed with the conviction of community members,
the City and BPD to resolve their differences in pursuit of common goals. While the conversations have
not always been easy, they have been vital, as they have fostered working relationships among key
stakeholders in a single forum and have laid the foundation for what is now CPIC’s central mission:
implementing the recommendations in the Gap Analysis.

It is important to understand that implementing these recommendations is not the work of BPD alone.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 14
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 16 of 164
BPD is, of course, responsible for strengthening its crisis response function. But the point of the Gap
Analysis is to identify and shore up deficiencies in the City’s behavioral health system so that BPD is
relied upon less often to resolve matters involving individuals in crisis and so that, when it does respond,
it has bona fide options for diversion to support services. Thus,
the responsibility for implementing the recommendations in
CPIC HAS PRODUCED the Gap Analysis falls most heavily on City agencies, which
must begin to match BPD’s commitment to CPIC’s work. Those
AN IMPRESSIVE BODY recommendations include providing 24/7 around-the-clock
OF WORK IN JUST OVER crisis centers and expanding the presently limited availability
TWO YEARS. THAT of mobile crisis teams, furnishing behavioral health outpatient
services in primary care settings, establishing an integrated
WORK IS A TESTAMENT “no wrong door” approach to behavioral health and substance
TO WHAT BPD AND abuse services, recruiting and retaining competent behavioral
CITY OFFICIALS CAN health workers, particularly in community-based settings,
and developing permanently supported housing programs.
ACHIEVE BY WORKING
TOGETHER WITH By paying heed to the City’s broader mental health care needs,
the Consent Decree was prescient. It recognizes that police
DEVOTED, INFORMED should not be asked or expected to fill the gaps in behavioral
COMMUNITY MEMBERS health support systems, and it embraces the reforms that
TO IMPROVE POLICING many have been advocating during the protest movement this
summer. Because of the Consent Decree, Baltimore already
AND BEHAVIORAL HEALTH has the blueprint for “re-imagining” policing.
SUPPORT IN BALTIMORE
Both the City and BPD have a lot of hard work ahead. The
City and its partner, BHSB, must set about implementing the
demanding recommendations in the Gap Analysis, which will
require a sustained, long-range commitment. BPD, for its part, still must train its officers and dispatchers
on behavioral health awareness, assemble a Crisis Intervention Team to respond routinely to calls
involving individuals in crisis and, above all, handle encounters with individuals in crisis in accordance
with policy to reduce the number of outcomes involving the criminal justice system. Nevertheless, the
Consent Decree is working as planned so far. The early-stage accomplishments and continuing maturation
of CPIC, with buy-in from the City, BHSB and BPD, have set the stage for addressing the formidable
challenges ahead.

|FIRST AMENDMENT-PROTECTED ACTIVITY

Because BPD has been focused on foundational reforms and is still developing the technological capacity
to collect comprehensive data needed for assessment, there is not yet substantial evidence that the
Consent Decree is having an observable or measurable impact on the street. This underscores the reality
that transformational change in a major city police department does not happen overnight. Recently,
however, the first signs of observable progress began to emerge. Although the evidence is anecdotal and
limited, it is noteworthy.

Based on dozens of hours of Monitoring Team observation over a two-week period, BPD responded
admirably to this summer’s protests calling for police reform in the wake of the killing of George Floyd
by Minneapolis police officers. Community organizers deserve substantial credit for keeping the protests
peaceful, but BPD also deserves credit for diligently complying with law and policy. BPD performed more
capably than most of its peer agencies around the country and, in fact, was hailed as a national model for
honoring First Amendment rights while maintaining public safety.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 15
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 17 of 164
BPD officers allowed community members to assemble in public spaces and march on public streets and
sidewalks without intervention; deliver speeches denouncing law enforcement and criticize officers face-
to-face without restriction or retaliation; and make video and audio recordings of police activity without
impediment. As Judge Bredar observed at the latest public hearing in July, “I take all of it as a measure
of progress. The evidence is there. Baltimore didn’t burn. A lot of cities did.”

As explained in more detail in the body of this report, the Monitoring Team attributes BPD’s performance
to three developments. First, protest leaders were well-organized, committed to non-violence, and willing
to share information needed for BPD to maintain public safety while respecting the First Amendment.
Second, BPD commanders lowered the temperature and avoided
provocation by permitting the protests to proceed unimpeded;
refraining from unnecessary arrests for minor infractions;
allowing temporary traffic disruptions; deploying officers largely
COMMUNITY
in their ordinary uniforms rather than in riot gear; keeping BPD’s ORGANIZERS DESERVE
civil disturbance unit away from the protest activity but ready SUBSTANTIAL CREDIT
to deploy in case the protests turned violent; and encouraging
officers to calmly engage in discussion with protestors and FOR KEEPING THE
march alongside them. Third, BPD rigorously adhered to an PROTESTS PEACEFUL,
effective Incident Command System, which provided officers
clear direction both in advance and as events unfolded. The
BUT BPD ALSO DESERVES
ICS prevented confusion on the street and led to coordinated, CREDIT FOR DILIGENTLY
measured police action. COMPLYING WITH
The Monitoring Team believes these developments are traceable LAW AND POLICY.
to the Consent Decree. Under the Consent Decree, BPD officers
have undergone training on use of force, and BPD has adopted
revised, comprehensive policies on First Amendment-protected
activity, which include new, detailed protocols for responding to mass demonstrations. Though not yet
formally active because Department-wide training has not yet been conducted, command staff used those
policies as a guide. The policies emphasize not only the imperative of permitting unimpeded assembly
and speech except in limited circumstances where they imminently threaten public safety, but the
importance of working with community members to ensure peaceful protest activity. Further, BPD has
placed renewed emphasis on the necessity of utilizing an Incident Command System to respond to both
ordinary crime scenes and major events. In the Monitoring Team’s review of BPD’s response to the death
of Detective Sean Suiter in late 2017, one of the missteps we observed was the breakdown of the ICS.
The training delivered to all commanders in response to the Monitoring Team’s review included refresher
instruction on ICS.

BPD’s response to the recent protests is yet additional evidence that, so far, the Consent Decree is working.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 16
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 18 of 164
AREAS POSING THE GREATEST CHALLENGES
|ACCOUNTABILITY (MISCONDUCT INVESTIGATIONS, DISCIPLINE AND SUPERVISION)
The Consent Decree will not be a success unless and until BPD establishes a robust culture of
accountability. A culture of accountability means the Public Integrity Bureau openly accepts, properly
classifies, and timely and thoroughly investigates complaints of misconduct; the Department metes out
discipline fairly and impartially; officers intervene to stop their peers from engaging in misconduct; and
sergeants and lieutenants rigorously monitor officer performance, appropriately rewarding exceptional
work and swiftly correcting missteps. For years, there has been no culture of accountability at BPD. The
City spends millions annually in settlement payouts for lawsuits alleging misconduct. The Department
is still reeling from the Gun Trace Task Force scandal. Maryland courts recently affirmed the dismissal
of a dozen administrative cases of major misconduct because PIB’s investigations were not completed on
time. With numbing regularity, the press reports serious allegations of officers abusing their power—by
planting evidence, using excessive force, making unlawful arrests in retaliation for questioning police
action, falsifying official reports, and even enlisting colleagues to extort a contractor for a reimbursement
for shoddy work on a personal home improvement. There will inevitably be violations of policy in police
agencies; hence the need for internal affairs units. But when officers feel as emboldened to cross the line
as some BPD officers appear to have been, there is an institutional problem.

And this is just what the community reads about in the paper. Over the past 30 months, the Monitoring
Team has observed first-hand the flaws in BPD’s accountability structure. PIB is and remains severely
understaffed—short more than three dozen investigators, according to BPD’s recent Staffing Plan.
To compound the problem, PIB investigators are inadequately trained to conduct internal affairs
investigations, which are different than ordinary criminal investigations. The natural consequence of
understaffing and poor training is there are too many cases for too few investigators, serious allegations
are not investigated promptly, and investigative findings are often inadequately supported. Further,
PIB supervisors do not timely review investigations upon completion, and data on outcomes is not yet
adequately collected or analyzed to gauge unit efficacy and departmental trends.

The Monitoring Team recently completed its first comprehensive compliance review of misconduct
investigations. The results were sobering, if predictable. The review focused on investigations from 2018,
which was prior to full implementation of early-stage reforms. Among other things, the Monitoring Team
found:

„ An unacceptable percentage of investigations were of poor quality.

„ Over 40 percent of the reviewed investigations were “administratively closed,” an outcome the Consent
Decree forbids (CD 344.k.iv.).
„ Case files were often in disarray, and investigations were not documented in a consistent, uniform
manner, leading to a lack of clarity in some cases about the factual issues presented and the
investigative steps taken.
„ Communications with involved parties were poor, with complainants receiving notice of receipt of
their complaints in less than one-quarter of all cases; officers received proper notice that they were
subjects of complaints at a similarly low rate.
„ Well over half of investigations were not completed within the Consent Decree’s 90-day deadline, with
extensions of time never requested or approved.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 17
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 19 of 164
„ Complainants were interviewed in less than half of cases, as were subjects, and fewer than a quarter
of cases had complete video or audio records of all witness interviews.
„ Investigations failed to flag related operational and performance issues that were exposed, such as
training deficiencies, repetitive problematic behavior, and potential policy shortcomings.
The public perception of BPD’s ability to police itself matches what the Monitoring Team has found. The
Monitoring Team’s community and arrestee surveys, completed within the past 15 months, reveal that
Baltimoreans generally do not believe that BPD holds its officers accountable for misconduct. Officers,
too, are frustrated. In focus groups last summer, one of the main reasons officers cited for low morale was
that, because PIB takes too long to investigate cases, they are often left in limbo for extended periods,
even for unfounded complaints or minor administrative infractions that should be resolved quickly.

Beyond PIB’s shortcomings, BPD’s supervisory systems are in need of repair. As the Staffing Study
completed in 2018 found, unity of command has too often been absent. That was particularly true within
the Patrol Division where, due to staffing schedules, officers could go days, even weeks, without working
the same shift as their assigned sergeant. Although the new patrol schedule seeks to address this problem,
it was a major problem for years, adversely affecting not only accountability but officer wellness, as officers
cannot be sure about the support they have. Further, performance evaluations have been historically
uneven. And because BPD has an outdated, ineffectual record-
keeping system, supervisors have never conducted systematic
AT A GLANCE
assessments of the performance of officers under their command. On
paper, BPD has an early intervention system, which is supposed to
AREAS POSING identify officers with emerging performance issues so that prompt
corrective action can be taken. But without adequate, accessible
THE GREATEST data, the system has never functioned well.
CHALLENGES
The Consent Decree prescribes solutions for BPD’s myriad
deficiencies, and BPD has started slowly down the path of reform. It
„ ACCOUNTABILITY has designed a new, more refined system for classifying misconduct
complaints, and recently completed a model intake/classification/
„ COMMUNITY investigations manual for PIB investigators and supervisors,
which is coupled with protocols for sharing information with the
POLICING Civilian Review Board. Relatedly, PIB has streamlined its case
assignment system. BPD authorized a long-needed independent
investigation of the causes of the Gun Trace Task Force scandal,
which is underway, and drafted a best-practices policy ensuring that all BPD members comply with
their constitutional obligation to disclose potential exculpatory evidence in criminal cases, which will be
implemented once officers are trained on it. Additionally, as noted above, Commissioner Harrison has
made efforts, even outside the Consent Decree, to transform PIB.

Change, however, is not yet observable. There is some anecdotal evidence that BPD’s oversight functions—
PIB, District commanders, and the Performance Standards Section (body-worn camera review unit)—are
gradually beginning to identify misconduct when it occurs and are responding as warranted by taking
non-disciplinary corrective action, opening an internal investigation, or initiating a criminal referral.
But this is far from the norm. That is not because BPD is not trying hard to achieve change. Rather, it
is because, in this area, change is especially difficult and slow to take root. BPD has multiple layers of
accumulated dysfunction to clear away.

At this point, the formula for change is to stay the course and give BPD an opportunity to do what the
Consent Decree prescribes. It begins with providing training to PIB investigators, which is set to take place

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 18
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 20 of 164
in 2021. PIB investigators still have not received specialized instruction in internal investigations. The
new PIB investigations manual, as good as it is, will only collect dust until PIB investigators are trained
on its provisions. There has been some frustration getting this training off the ground. The pandemic
has slowed development, and BPD had to switch to a new consultant after spending months attempting
to reach an arrangement and develop a curriculum with another. The completion of training should
accelerate improvement in the quality of investigations and the integrity of disciplinary recommendations.
Something as straightforward as ensuring that all investigators work off of a set task list and maintain
case files in a standard, uniform fashion—the Monitoring Team has never seen uniformly organized files
in its PIB case reviews—would go a long way toward consistent completion of timely, complete, well-
supported investigations, findings and recommendations.

Peer intervention (EPIC) training, set to be delivered this fall and winter, is also vital, coupled with
the finalization of new policies mandating a duty to intervene and prohibiting retaliation for reporting
misconduct. A program requiring officers to stop their peers from engaging in misconduct and to intervene
when they do, EPIC is inherently about changing culture and improving community trust. Indications
are that it is proving successful in the New Orleans Police Department, also under a consent decree. It
holds the same promise for BPD.

A number of other mandated accountability reforms are on deck:

„ A new policy on performance evaluations

„ Revisions to the full suite of disciplinary policies
„ Integration of civilians into disciplinary trial boards
„ Implementation of a revamped Field Training Officer program
„ Expansion of the oversight work of the Performance Standards Section
„ Establishment and implementation of a plan for collecting the data needed to complete the statistical
reports and internal PIB audits required by Paragraphs 402 and 405 of the Consent Decree
„ Development of an e-based Early Intervention System that is fully integrated with IAPro (BPD’s
misconduct database) and the nascent Record Management System
This is a lot. BPD needs to have the chance to conduct training and implement all these other measures
before the Monitoring Team is able to determine whether the Consent Decree’s accountability provisions
are accomplishing what they are designed to achieve. Put differently, it is still too soon to tell whether
BPD is failing to implement the Consent Decree’s accountability reforms as urgently and effectively as it
must or whether the Consent Decree fails to address accountability reform in all the right ways.

Establishing a stout system of accountability very likely poses the greatest challenge to the prospects of
the reform enterprise. Unless first-line supervisors and field training officers embrace the change that
is afoot, patrol officers will not internalize it and constitutional policing will not become a consistent
reality on the street. Unless PIB dramatically improves its operations, certain officers will continue to
feel emboldened to break the rules while the rest will continue to suffer from low morale. And unless
BPD starts to prove that it can police itself, it will not win the community’s trust. BPD needs to get
accountability reform right. It has taken baby steps so far. In the coming year, it must start taking full
strides.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 19
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 21 of 164
|COMMUNITY POLICING
Together with constitutional policing, developing an effective community policing model is the Consent
Decree’s principal objective. It depends on building and maintaining the trust of community members so
that, through collaborative problem-solving, crime can be prevented, reduced and solved more readily.
Establishing this model will transform how BPD provides police services. Right now, it is far from clear
that BPD will be able achieve such transformation.

Currently, BPD patrol officers spend most of their time responding to calls for service. Officers have little
time for proactive relationship-building. Under the Community Policing Plan, published in April 2020,
that is supposed to change. The Plan envisions that officers will spend up to 40% of their time on proactive,
community-oriented policing and 60% on calls for service. The 40% figure, which applies to all officers,
includes engagement with community members through informal interactions (pursuing unplanned
interpersonal interactions between calls), formal programming (participating in neighborhood events
and community meetings and establishing partnerships), and problem-solving (proactively working
with community members to address recurrent localized issues and
prevent future neighborhood disorder).
OVER THE PAST
The challenges to implementing this appropriately ambitious plan
are profound. First, as the Monitoring Team’s community and
30 MONTHS, THE
arrestee surveys show, satisfaction with and trust in BPD are MONITORING TEAM
low. A majority of respondents do not believe that BPD effectively HAS OBSERVED
fights crime or respects or tries to cultivate good relationships with
community members. A majority also believe that BPD officers do FIRST-HAND THE
not care about their communities, expressed reluctance to contact FLAWS IN BPD’S
BPD about most crimes, and reported observing BPD officers engage
in misconduct, such as racial profiling, excessive force, and using
ACCOUNTABILITY
verbally abusive language. Much lower percentages of participants STRUCTURE.
have more positive views on these subjects. Because Baltimoreans
have diminished faith in BPD, engagement and problem-solving
are going to be hard. That said, many survey participants said they
would like to see improved relations with BPD, expressed a willingness to engage, and feel comfortable
communicating with BPD when necessary.

Second, violent crime persists in Baltimore. The persistence of violent crime makes community members
more skeptical of the police’s ability to prevent and solve crime and, in the City’s most challenged
neighborhoods, more fearful of engagement because of a fear of retaliation.

Third, as the Staffing Plan has established, BPD needs to significantly enhance Patrol Division staffing
to realize the goal of having officers spend 40% of their time on proactive, community-oriented activity.
As of this summer, according to the Staffing Plan, BPD needs over 175 additional patrol officers and over
30 additional sergeants. Simply put, and as the Community Policing Plan acknowledges, BPD will not
be able to realize its community policing objectives absent a meaningful increase in the number of patrol
officers.

In the face of these challenges, BPD’s progress toward the Consent Decree’s community policing
requirements has been halting. The development of the Community Policing Plan—the threshold reform
measure in this area—took over a year longer than initially scheduled. Prior to the pandemic, it stood
out as among the very few Monitoring Plan deliverables that BPD did not complete on time. As noted,
the delay was due in large part to the serial changes in commissioners prior to the appointment of

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 20
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 22 of 164
Commissioner Harrison. A community policing plan naturally should reflect a commissioner’s vision.
Nevertheless, completion of the Plan still took longer than anticipated because the initial draft was
inadequate—little more than a repackaging of BPD’s 2019 crime reduction plan. BPD essentially had to
go back to the drawing board.

The Plan now in place is sound. It provides a good foundation for

IN THE FACE OF THESE achieving Consent Decree compliance. Indeed, the International
CHALLENGES, BPD’S Association of Chiefs of Police recently featured the Plan as a
model in one of its publications. However, the Plan will have
PROGRESS TOWARD to be continually reassessed and updated. That is particularly
THE CONSENT DECREE’S true given the events of the past several months in the wake of
COMMUNITY POLICING the killing of George Floyd. Those events have called attention
to inadequacies in policing and underscore the need to re-think
REQUIREMENTS HAS community-based strategies for maintaining public safety. In the
BEEN HALTING. “Responding to Individuals in Crisis” section above, we stress
the necessity of improving the City’s behavioral health system
in order to improve BPD’s responses to individuals in crisis. An
effective community policing strategy might similarly require a
more holistic approach, incorporating City resources—not just
BPD resources—to address localized issues through collaborative problem-solving.

Because BPD only recently completed the Community Policing Plan, it has not yet been required to complete
the other Consent Decree deliverables for community policing. Thus, as in the area of accountability, it
is too soon to say whether BPD is falling short or whether the Consent Decree’s community policing
provisions need to be modified. However, there are clearly reforms BPD must implement to advance more
quickly toward the Consent Decree’s community policing goals. Because effective community policing
genuinely depends on complying with the requirements in every other area of the Consent Decree, BPD’s
“to do list” could encompass the entire agreement. To take a prominent example, BPD will not earn
the community’s trust, essential for effective community policing, until it shows progress toward the
Consent Decree’s requirements for accountability and constitutional policing. Nevertheless, there are
more immediate concrete measures BPD must take. They include the following:

Training | BPD has yet to deliver training on community policing. No community policing model can be
implemented without providing officers specific guidance on the type of engagement and problem-solving
that is expected of them. In recognition of the centrality of adopting a community policing model to
Consent Decree compliance, community policing training is now being prioritized and has been moved up
on the training schedule. Curriculum development is underway, e-learning should begin in early 2021,
and in-class instruction should begin in the second quarter of 2021.

Implementation of Community Policing Plan | The Community Policing Plan calls for incorporating
community policing assessment criteria into field training evaluations and performance evaluations after
training is delivered. It similarly calls for establishing data collection requirements and data-driven
measurements for community policing and, correspondingly, including reports on community policing
data in weekly Comstat meetings. It also envisions incorporating community policing performance into
promotion criteria. BPD should move forward with these performance measurements as quickly as
possible once training is completed.

The Community Policing Plan requires the development of localized Neighborhood Policing Plans to
address the issues unique to each City neighborhood afflicted by high crime. A piloting process is supposed
to begin in two districts in the coming months, all commanders and Neighborhood Coordination Officers

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 21
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 23 of 164
are supposed to undergo training on the plans in mid-2021, and plans are supposed to be deployed in
every district by the end of 2021. Because implementing these micro-plans is an important feature of the
Community Policing Plan, BPD must proceed with alacrity. The calls for reform both nationwide and in
Baltimore have only amplified the urgency of prompt implementation.

Consistent with BPD policy, the Community Policing Plan requires fair and impartial policing and
adherence to the tenets of procedural justice. Supervisors must ensure that the officers under their
command treat the people they encounter with simple respect, including when taking law enforcement
action.

During a community-wide call in August 2020 to explain what happened when several blocks in the
Eastern District had to be barricaded for over 24 hours because an armed man had fired his gun in
the presence of family members, was experiencing a crisis, and refused to come out of his home, one
resident complained that two officers stationed on a nearby corner brushed her off when she asked when
unimpeded access to her home might be restored. The Chief of Patrol immediately jumped in to assure
the resident that the officers’ response was unacceptable and asked for the officers’ names and the time
and location of the encounter so that he could address them directly. That is precisely the type of decisive
command- and supervisor-level conduct that conveys to officers the importance of procedural justice,
assures community members that BPD is listening to them, and lays the groundwork for an effective
community policing strategy.

Equipment/Patrol Cars | BPD has developed recruitment, hiring and retention plans under the Consent
Decree. To have enough patrol officers to foster community policing, it must continue to implement them.
One of the concrete measures BPD can take (and has begun to take) to attract and retain quality officers
is to provide them the equipment they need to do their jobs effectively. The completion of a new records
management system is vital. By facilitating electronic field-based reporting, the new RMS will reduce
the amount of time officers must spend on writing reports and increase the amount of time available for
community policing activity. Equally if not more important, BPD must continue replacing its aging fleet
of patrol cars with new ones. In the focus groups the Monitoring Team conducted last summer, one of
officers’ central complaints was the condition of their patrol cars. They indicated that a simple way to
improve job quality and officer morale was to ensure their cars—which are effectively their office space—
are clean, modern and fully functional.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 22
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 24 of 164

FINDINGS

Since February 2018, when their work under the First-Year Monitoring Plan began, BPD and City
leadership have made evident progress toward satisfying the Consent Decree’s foundational requirements.
Notably, BPD and the City achieved this progress despite instability in BPD and City leadership during
the first 12 to 18 months of reform and notwithstanding certain unavoidable delays in implementation
caused by the coronavirus pandemic. The accomplishments of BPD and the City in the face of these
challenges is attributable to the consistent commitment of BPD’s various commanders to the reform
process and the hard work of BPD’s Consent Decree Implementation Unit and the functions that support
it, especially the Training Academy.

BPD has almost completed the initial round of policy revisions required by the Consent Decree. Under
the Consent Decree, BPD has issued upwards of 50 new policies. These policies provide clearer, more
comprehensive guidance to BPD officers.

BPD has prepared and begun conducting training on many of these new policies. It drafted and delivered
e-learning and in-class instruction on use of force and related aspects of fair and impartial policing. It
prepared and delivered e-learning, and is in the midst of classroom training, on stops, searches and
arrests and related aspects of fair and impartial policing. It drafted and is now providing instruction
on behavioral health awareness and crisis intervention to recruits. It furnished e-learning to all officers
on body-worn camera use and responding to reports of sexual assault. It has finalized and has begun
classroom instruction for all officers on behavioral health awareness/crisis intervention and sexual
assault investigations, and will soon begin specialized instruction on behavioral health awareness/crisis
intervention for dispatchers and 911 call-takers. The classroom components of all of these new training
programs incorporate adult learning principles, replacing static, lecture-based instruction with dynamic,
scenario-based, problem-solving exercises.

In addition to drafting policies and conducting training, BPD and the City have made at least preliminary
progress toward implementing the organizational and structural improvements required by the Consent
Decree. They have completed a Technology Resource Study and a Technology Resource Plan and are now
carrying through with the Plan. They have conducted studies and/or finalized plans on staffing, hiring,
recruitment, community policing and engagement, diversion of youth from the criminal justice system,
ameliorating the City’s behavioral health system, and refining BPD’s memorandum of understanding
with Baltimore School Police. And critically, BPD has begun the long, difficult process of transforming its
historically ineffectual Public Integrity Bureau.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 23
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 25 of 164
Finally, BPD has made initial strides toward developing the capacity for self-assessment and self-
correction. It is now regularly conducting monthly equipment audits of its transport vehicles. It has
begun conducting quarterly audits of a random sample of transport events. And it has issued its first
two annual reports regarding the efficacy and integrity of its sexual assault investigations; published an
inaugural report its responses to First Amendment-protected activities, including large-scale protests,
video recording of police conduct, and simple criticism of police action; and is finalizing its initial report
on performance issues revealed by arrests that result in individuals being released without charge.

Despite the achievements of BPD and the City over the past two and a half years, the challenges ahead
remain voluminous and acute. Beyond the need to successfully complete training on numerous revised
policies, BPD must stand up a well-functioning system of internal investigations and discipline so that
officers understand there will be real consequences for misconduct—and for passively standing silent
in the face of it. BPD must carry through with modernizing its IT systems and, once those systems are
operational, ensure that its officers appropriately use them, including their time-saving, data-liberating
electronic field-based reporting capability. BPD must provide essential reinforcements for its Patrol
Division and Public Integrity Bureau through reassignment of officers from other units, civilianization
of certain administrative functions, and improved recruitment and retention. The City must fill the
yawning gaps in its behavioral health system so that BPD officers are not always expected to be the first
responders to individuals in crisis. Ultimately, BPD must change its culture. It must get its officers, and
especially their supervisors, to own the reform project for themselves so that constitutional, community-
oriented policing is the Departmental norm. There is no other way to rebuild community trust and, in the
end, no other way to fight crime effectively.

As required by Paragraph 469, this section of the report sets forth the Monitoring Team’s findings
regarding “whether the material requirements of [each area of the Consent Decree] have been achieved.”
For each area, the Monitoring Team explains (1) the Consent Decree’s requirements over the long term,
(2) the progress BPD has made along the arc of compliance and the numeric compliance score BPD is
receiving (3) the challenges facing BPD, and (4) immediate next steps. The Monitoring Team is not
evaluating BPD’s progress toward satisfying each and every paragraph and each and every requirement
within each area of the Consent Decree. That kind of report card would not reveal much, if anything,
about BPD’s performance at this stage in the reform process. The Monitoring Team instead assesses
BPD’s progress on the specific Consent Decree requirements that are included in the First-, Second- and
Third-Year Monitoring Plans in each area, and then describes the road ahead.

This section begins in the area that has been at the heart of the reform effort over the past year: training.
It then addresses the areas of the Consent Decree that present among the most pressing threshold
challenges facing BPD: misconduct investigations and discipline, technology, and staffing, followed by
areas where DOJ found or expressed concerns about a pattern or practice of constitutional violations,
including stops, searches and arrests, impartial policing, use of force, and transportation of persons in
custody. This section concludes by assessing BPD’s progress in other areas of the Consent Decree.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 24
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 26 of 164
TRAINING
No changes to BPD policies, practices, or procedures can be implemented without making them clear to
officers. As the Monitoring Team has observed in prior reports, even the best policies amount to nothing
if officers do not receive appropriate training on how to comply with them.

The Consent Decree therefore provides that “proper, effective, and comprehensive training is a necessary
prerequisite to constitutional policing” (CD 291). It outlines specific training requirements for stops,
searches, and arrests (CD 67–68); crisis intervention (CD 106–08, 112–113); use of force (CD 166–68);
transportation of persons in custody (CD 238); First Amendment-protected activities (CD 251); sexual
assault investigations (CD 259); supervision and management (CD 303, 308–10); and misconduct
investigations (CD 409–15). BPD must also overhaul its Field Training Officer Program for new Academy
graduates. (CD 301–02). Even in areas where the Consent Decree does not provide detailed requirements
for training, the full implementation of new policies and procedures requires similarly rigorous, high-
quality training to translate paper into practice.

This report elsewhere details the important progress that BPD has made in training on specific subjects,
including training on use of force; stops, searches and arrests; impartial policing; behavioral health
awareness and crisis intervention; investigation of reports of sexual assault; and body-worn camera use.
This section addresses BPD’s progress toward improving and enhancing its overall training function.

BPD has begun developing a first-rate training program for its officers. It is successfully transforming
from a static, lecture-based approach to an approach that incorporates interactive e-learning coupled with
facilitated, scenario-based classroom discussion—a far more dynamic method of instruction that provides
officers the practical skills they need to solve the problems they confront on the street. Although BPD still
has a distance to travel toward full and successful implementation of this forward-thinking approach to
officer training, the Monitoring Team remains encouraged that BPD leadership and Academy staff are
eagerly embracing it.

BPD has made substantial progress toward satisfying the Consent Decree’s general training requirements
regarding Academy facilities, staffing, operations, and approach to instruction. BPD also has developed
and delivered training in certain Consent Decree areas, including use of force and body-worn camera use.
However, BPD must successfully complete the transformation of its approach to training, and it must
continue to collaborate with the Monitoring Team, DOJ and the community to develop and deliver new
training curricula in all other areas of the Consent Decree.

Accordingly, even though BPD is on its way to compliance with the Consent Decree’s general
training requirements, significant work remains before it reaches sustained compliance. Its
compliance score in the “Training Generally” category is “4a” (implementation not yet assessed).

COMPLIANCE SCORE:

4A
TRAINING GENERALLY Implementation-
Not Yet Accessed

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 25
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 27 of 164
|AREAS OF PROGRESS
Implementation of Training Requirements

As noted elsewhere in this report, BPD has provided officers important training on core policies revised
under the Consent Decree, particularly in areas in which DOJ found a pattern or practice of constitutional
violations. In November 2019, BPD certified that nearly all eligible officers successfully completed
e-learning and classroom training on its new use of force policies, as well as new policies addressing
fair and impartial policing (UOF/FIP I). Further, in late 2019, BPD provided e-learning, and beginning
in February 2020 began classroom training, on new policies on stops, searches, and arrests and related
aspects of fair and impartial policing (SSA/FIP II). After the COVID-19 pandemic interrupted this in-
person training in March 2020, BPD re-started the training in July 2020 and plans to complete it by the
end of this year. For the e-learning that preceded in-class instruction in both of UOF/FIP I and SSA/FIP
II training, officers were required to score 100% on tests administered at the end of the e-learning in
order to become eligible for the classroom component.

In addition to this vital UOF, SSA and FIP training, BPD has provided and confirmed completion of
classroom training for field training officers (note that the FTO program, including training, is now
undergoing revision) and Department-wide e-learning on both body-worn camera use and responding
to reports of sexual assault. BPD is increasingly using its e-learning platform as a means of not only
introducing new policies in advance of in-person training and thereby making the in-person training
more impactful and effective, but to provide ongoing training to ensure continuing skill development and
adherence to performance expectations.

Even as the COVID-19 pandemic prevented in-person officer training for several months in 2020, BPD
has continued to develop e-learning and/or classroom training programs in the following Consent Decree
areas:

„ Peer invention; (Ethical Policing is Courageous – EPIC)

„ Sexual assault investigations
„ Supervisory review of stops, searches and arrests
„ Vehicle pursuits
„ First Amendment-protected activities
„ Behavioral health awareness and crisis intervention
„ Misconduct investigations
„ Community Policing
„ Stops, searches and arrests for lesser offenses
BPD’s continued, intensive work over the last six months will help to ensure that officers receive training
on core Consent Decree policies with pandemic-related delays that are as short as possible.

Staffing, Resources and Facilities

At the start of the Consent Decree process, BPD’s Training Academy was “severely understaffed” and
significantly under-resourced. ECF No. 178-1 at 17. As recently as early 2019, there were only three full-

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 26
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 28 of 164
time instructors available to develop and conduct in-service training required by the Consent Decree.
ECF No. 279-1 at 31. BPD’s rapid leadership turnover throughout 2018 directly contributed to “BPD’s
delay in providing adequate resources and personnel to the Training Academy.” Id. at 15.

In the past year and a half, BPD has begun investing appropriately in officer training:

„ In early 2019, before UOF/FIP I classroom training began, BPD provided the Academy with ten
additional full-time instructors. As the Monitoring Team has audited the performance of these
instructors in UOF/FIP I and SSA/FIP II, we have been encouraged by the development and progression
of their teaching capabilities.
„ In the second half of 2019, BPD hired a civilian Academic Director who is an accomplished academic
and police professional with years of experience in developing high-quality, evidence-based education
programs for law enforcement agencies.
„ In the first part of 2020, BPD added two curriculum writers.
„ In the first part of 2020, BPD added several new civilian law instructors to assist with SSA/FIP II
training.
The fortification of the Academy’s staff is encouraging. As explained in the Challenges Ahead section
below, BPD expects to fulfill numerous training objectives simultaneously in the near future. To state
the obvious, BPD would not be able to fulfill those objectives without the new personnel it has brought
in to the Academy.

The Consent Decree requires BPD to update its training facilities (CD 292). In compliance with that
requirement, BPD has moved its Training Academy from “its long-time, deficient facilities to more modern
facilities at the University of Baltimore.” ECF No. 279-1 at 31. As of April 2020, all training functions had
transitioned to the new facilities. It is hard to overstate the significance of this development. As both the
Court and the Monitoring Team have emphasized, modern training facilities are vital to establishing a
first-rate police force. With attractive, state-of-the-art facilities, BPD conveys to both recruits and officers
that policing is a profession and that the Department takes seriously their development as professionals.

Adoption of New Training Paradigm

Prior to the Consent Decree, BPD officers received training in the form of unengaging, lecture-based
instruction that covered all topics within a single in-class block of time each year. ECF No. 178-1 at 17-
18. Further, BPD developed its training curricula with minimal or no input from other components of the
Department or the community.

Under the Consent Decree, BPD has invested substantial time and effort in implementing a fundamentally
different approach to officer training. BPD has hired civilian personnel to coordinate curriculum
development and provide legal instruction. Academy staff have learned from other departments that
have incorporated contemporary adult learning principles into their training programs. Most important,
the Academy has embraced these principles for its own training programs. It has adopted a model that
employs integrated, scenario-based learning focused on officer decision-making, problem-solving and
skill development rather than the static, passive receipt of information.

To ensure that Consent Decree-mandated policies can be implemented as quickly as possible, and to
keep officers more actively engaged in their own professional development, BPD also has moved from
offering one long block of in-service training each year to holding in-service training in smaller blocks

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 27
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 29 of 164
throughout the year. Further, by using a new, comprehensive, multi-stage pilot process, BPD is providing
Academy instructors the opportunity to revise and fine-tune training curricula before they are rolled out
Department-wide.

Significantly, BPD is, for the first time, developing training on core subjects in collaboration with
community members. As explained above, it is providing for public and officer feedback on training
curricula while it is still in draft form. Additionally, BPD has created a Community Training Review
Committee, “a panel of community representatives to provide ongoing feedback and guidance on training
initiatives.” ECF No. 279-1 at 30. CTRC attended and provided feedback at pilot training sessions on
both UOF/FIP I and SSA/FIP II. As BPD continues to transform its training approach, it will “utilize
community members more directly, both to help devise the curriculum and to participate in the training.”
Id.

|CHALLENGES AHEAD
Balancing Training Imperatives

As previously reported, the training demands of the Consent Decree continue to increase. ECF No. 279-
1 at 31. As the slate of training curricula identified above is finalized, the Training Academy’s staff will
need to balance a greater number of concurrent training initiatives than ever before. BPD will need to
ensure the Academy has the resources and support it needs to provide the high-quality training that the
Consent Decree requires. This will be particularly challenging in the face of the civilian hiring freeze for
Fiscal Year 2021, which delays BPD’s plans to civilianize various Academy functions and re-assign sworn
officers to the Patrol Division.

Long-Term Resources and Capacity

The Monitoring Team also has reported that BPD will need adapt to a more intensive model of in-
service training as “the new normal.” Id. at 32. Providing such training—which is based on facilitated
discussion, problem-solving and decision-making that incorporates scenarios, simulations, role-playing,
and performance assessments—requires substantial time and effort and, therefore, substantial staff. For
that reason, the Consent Decree requires the Academy to maintain “an adequate number of qualified
instructors.” (CD 293, 296.)

BPD is presently operating under budgetary constraints, especially because of the pandemic’s effect on
the City’s financial resources. In circumstances like these, it may be tempting to reduce investment in
officer training and professional development. But if the Consent Decree’s reforms are to be implemented,
that temptation must be resisted. The work of the Academy is indispensable to achieving sustainable
compliance with the Consent Decree. Even as the City and BPD must weigh competing priorities, they
must recognize that BPD’s ability to provide community-oriented, constitutional policing will depend on
its ability to provide officers with the requisite training, based on the model that the Academy has been
working diligently to adopt.

|THE NEXT SIX MONTHS

The COVID-19 pandemic continues to make it difficult to forecast with certainty the training requirements
BPD will be able to satisfy in the near future. Nevertheless, BPD has resumed SSA/FIP II and, using
a modified schedule with reduced class sizes, plans to complete it by the end of the year. At the same
time, BPD will begin Consent Decree-required training in other critical areas. The Ethical Policing is
Courageous (“EPIC”) training, which addresses the need for peer intervention and “active bystandership,”

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 28
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 30 of 164
is slated to begin in November 2020, with all officers trained by March 2021. Virtual training on patrol
response to both behavioral health crises and reports of sexual assault began earlier this month, with
all officers trained by March 2021. Classroom training on defensive tactics should begin in fall 2020.
Specialized in-class training on sexual assault investigations for members of the Sex Offense Unit should
be held next month (October 2020).

In the coming months, BPD will deliver e-learning on other Consent Decree subjects, including First
Amendment-protected activity, deadly force, and proactive policing.

Finally, BPD will develop classroom and e-learning curricula for community policing and stops, searches
and arrests for low-level “quality of life” offenses and specialized training on misconduct investigations
for Public Integrity Bureau and Civilian Review Board investigators.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 29
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 31 of 164
MISCONDUCT INVESTIGATIONS AND DISCIPLINE
The new policies BPD has adopted in other areas of the Consent Decree will mean little if BPD does not
hold all officers accountable for following them. As Paragraph 329 of the Consent Decree explains, “[a]
robust and well-functioning accountability system in which officers are held to the highest standards of
integrity is critical to BPD’s legitimacy and a priority of the Department.” The need for BPD to repair
its internal affairs function, now housed in the Public Integrity Bureau (“PIB”), is thus at the heart of
the Consent Decree.

As the Monitoring Team observed in previous reports, the Misconduct Investigations and Discipline
section of the Consent Decree is the longest and most comprehensive, spanning 87 paragraphs and 38
pages. It covers the location, independence, resources and authority of PIB (CD 330-34); the process for
receiving complaints, classifying them, and communicating with complainants about them (CD 335-42);
requirements for conducting fair, thorough, reliable misconduct investigations and making misconduct
determinations (CD 343-58); requirements for handling and referring allegations of criminal misconduct
(CD 359-71); the lodging of disciplinary charges, the administration of disciplinary hearings, and the
imposition of discipline (CD 372-88); the process for community-centered mediation as an alternative to
investigation for certain minor allegations of officer misconduct affecting civilians (CD 389-91); record-
keeping for misconduct investigations (CD 392-95); measures for ensuring transparency, including
issuance of quarterly public reports of aggregate data (CD 396-405); a testing program designed to evaluate
the efficacy of the civilian complaint intake process (CD 406-08); and training of OPR investigators and
supervisors (CD 409-15).

The ultimate goals of the Consent Decree’s provisions implicating PIB are the full, fair, objective, and
timely investigation of all potential officer misconduct; the rigorous review of all misconduct investigations;
and an impartial, transparent, uniform process for the imposition of discipline and corrective action
where appropriate.

In each of the compliance categories addressing misconduct investigations and discipline, BPD has made
reasonable progress toward implementing preliminary Consent Decree requirements. It bears noting,
however, that as compared to most other areas of the Consent Decree, BPD’s progress toward substantial
and effective compliance has been modest. That is not because BPD has not done the work it has been
required to do under the Monitoring Plans, nor is it because BPD has failed to show the commitment
required to achieve compliance. Rather, it is because, in this area, BPD has an extremely tough row to
hoe: BPD must completely overhaul an internal affairs system that has been broken for years and that,
because of its dysfunction, bears considerable responsibility for the community’s lack of faith in the
Department.

In the area of “Misconduct – Intake,” BPD’s compliance score is “2” (policies/planning), as it has
nearly completed policy revisions, but not quite. In the area of “Misconduct – Investigations,”
BPD’s compliance score is “3” (training), as it has completed policy revisions and has begun
drafting training curriculum for PIB investigators. In the areas of “Misconduct – Discipline”
and “Misconduct – Transparency,” BPD has only recently begun revising policies, so its
compliance score is “1” (not yet assessed).

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 30
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 32 of 164

COMPLIANCE SCORE:

2
MISCONDUCT – INTAKE Policies/
Planning

COMPLIANCE SCORE:
MISCONDUCT –
INVESTIGATIONS
3 Training

COMPLIANCE SCORE:
MISCONDUCT –
DISCIPLINE
1 Not Yet
Assessed

COMPLIANCE SCORE:
MISCONDUCT –
TRANSPARENCY
1 Not Yet
Assessed

|AREAS OF PROGRESS
Policies and Procedures

Intake and Classification | Baltimore residents and DOJ identified deficiencies in BPD’s misconduct
complaint intake and classification process as areas of serious concern during DOJ’s investigation. Over
the past two years, BPD has taken a number of important steps toward adopting a uniform, reliable,
user-friendly complaint intake and classification system.

At the outset of the reform process, BPD made filing complaints easier. Previously, individuals who
wanted to lodge misconduct complaints against police officers had to make their complaints in person, sign
them, and have them notarized. BPD’s revised policies have removed these impediments. Additionally,
BPD changed its website to facilitate the acceptance of complaints online and via email and established
a dedicated telephone number for individuals to call to make complaints.

In 2018, BPD collaborated with the Monitoring Team and DOJ to finalize revisions to Policy 306, which
addresses complaint intake, complaint classification and communication with complainants, and PIB’s
Classification Protocols. The policy and protocols were finalized and approved in December 2018. See
ECF No. 175.

In late August 2020, after over a year of collaboration, BPD completed the intake and classifications
section of the PIB manual for investigators. See ECF No. 336. This section of the manual covers the front
end of the internal affairs investigation process, vital to ensuring that complaints are promptly acted
upon, uniformly classified, and handled from the outset with the seriousness they deserve. Neither this
section of the manual nor Policy 306 nor the investigations section of the manual, discussed below, will
be fully active until PIB investigators are trained on it.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 31
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 33 of 164
Finally, BPD and the Civilian Review Board (CRB) have created a Unified Complaint Form, to be used by
individuals filing misconduct complaints with BPD or CRB. BPD is developing an e-learning curriculum
to deliver training on the use of the form in preparation for Department-wide rollout late this year or
early next year. BPD will pilot test the form in the near future. The form is already in use by the CRB.

Investigations | In addition to improving policies and procedures for misconduct complaint intake
and classification, BPD has worked diligently to improve policies and procedures governing misconduct
investigations—investigations that, historically, have moved too slow, lacked thoroughness, and reached
unjustified or inadequately justified conclusions.

After months of intensive collaboration with the Monitoring Team and DOJ, and two rounds of
public comment from community members (including the Civilian Review Board), BPD finalized its
investigations manual for PIB, which the Monitoring Team approved in November 2019. See ECF No.
263. The completion of the investigations manual was a milestone achievement. The manual provides
a complete guide to internal affairs investigations for PIB investigators, supervisors and commanders.
It establishes a comprehensive investigations procedure that, if followed, will yield thorough, fair and
timely investigations that produce appropriate disciplinary recommendations. The Monitoring Team’s
subject matter experts believe the manual is among the most comprehensive and effective internal affairs
investigations manuals in the country. However, it will not be fully implemented until PIB investigators
are trained on it.

Expedited Resolution of Minor Misconduct Allegations | In mid-2019, shortly after arriving at

BPD, Commissioner Harrison developed a policy for addressing complaints of minor misconduct on an
expedited basis. The new policy, Policy 321, permits District commanders, rather than PIB investigators,
to handle such complaints and allows officers to obtain prompt case closure at the District level by
admitting responsibility in negotiated settlements. The Monitoring Team approved the new policy in
September 2019. See ECF No. 216.

Policy violations eligible for prompt, negotiated resolution include reporting late to roll call, failure to
appear in court or for a medical appointment, and loss of BPD property other than a firearm; they do
not include alleged misconduct in officer interactions with members of the public (e.g., excessive force,
false arrests, harassment). Policy 321 does not remove PIB from the disciplinary process for minor policy
violations. PIB will continue to receive and classify complaints at the front end of the process and obtain
reports and retain records of negotiated resolutions at the back end in order to ensure consistency in the
application of the policy.

By allowing District commanders, rather than PIB, to address minor policy violations that call for
supervisory correction, Policy 321 promises to advance several Consent Decree objectives. First, it should
reduce the number of minor misconduct cases assigned to PIB investigators, which in turn should free
PIB investigators to focus on more serious complaints and enable all misconduct complaints, minor and
serious alike, to be resolved more quickly. Officers who engage in misconduct should more promptly be
held accountable, and community members with bona fide complaints should receive swifter justice.
Second, by speeding up complaint resolution, officers who are the subject of complaints that can be
promptly resolved as unfounded, unsustained or sustained through negotiated resolution should not
be left in limbo and kept ineligible for transfer or promotion for unreasonable periods of time, as has
been the case in recent years due to PIB backlogs. Given that the duration of misconduct investigations
has been a primary cause of officer dissatisfaction, Policy 321 could help boost officer morale. Third, by
enabling District commanders to resolve policy violations appropriate for supervisory correction, Policy
321 should fortify BPD’s command structure and supervisory performance in each district and unit.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 32
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 34 of 164
Neither BPD nor the Monitoring Team have begun to evaluate whether, in fact, Policy 321 is accomplishing
these objectives. BPD will begin to do so shortly and will include the results in its forthcoming quarterly
PIB reports.

PIB-CRB Information Sharing Protocols | In March 2019, BPD successfully finalized complaint
intake and classification protocols for PIB and the Civilian Review Board (“CRB”). These protocols identify
the responsibilities of both PIB and CRB for intake and classification of misconduct complaints and for
exchanging information at the outset of complaint investigations. The issuance of the protocols was an
important step toward improving civilian oversight of BPD. PIB and CRB must interact and coordinate
with each other to fulfill their respective statutory roles, but until the adoption of these protocols, PIB
and CRB had never developed a formal process for information sharing.

Under the new protocols, CRB must be furnished prompt, consistent, electronic access to PIB data
through IAPro, BPD’s records system for misconduct investigations and discipline. To facilitate the new
relationship between PIB and CRB, BPD also has assigned a CRB liaison. The Monitoring Team has not
yet assessed the quality and effectiveness of the new relationship between PIB and CRB. It is too soon to
tell whether the protocols can fully repair the historically troubled relationship between PIB and CRB,
but the Monitoring has preliminarily observed that they appear to be fostering improved cooperation.

Having completed the intake/classification information sharing protocols last year, BPD and CRB
proceeded to draft and complete information sharing protocols for misconduct investigations in 2020. See
ECF No. 336. As with the information sharing protocols for intake/classification, the information sharing
protocols for investigations promise to ensure that PIB and CRB are working with the same information
when each conducts its own independent investigations of misconduct allegations.

Disclosure of Exculpatory Evidence Policy | In September 2019, the Monitoring Team approved
a new BPD policy, Policy 1809, on exculpatory evidence disclosure requirements. See ECF No. 246.
The policy is intended to ensure that all BPD members comply with their obligation under federal and
Maryland law to disclose potential exculpatory and impeachment evidence in criminal cases. The policy
requires BPD members to identify and disclose to the prosecution, as soon as possible following the
initiation of criminal charges in state or federal court, evidence that is favorable to a criminal defendant
because it may disprove the guilt of the defendant, may cast doubt on the credibility of a witness for the
government, or may show the defendant should receive less severe punishment.

Once Policy 1809 becomes active, which will occur after officers are trained on it, BPD will join a small
but growing number of law enforcement agencies that have formal procedures concerning the disclosure
of exculpatory evidence. These are requirements that not only respect the constitutional and statutory
rights of the accused, but also promote public safety by seeking to ensure the integrity of criminal
prosecutions, which can be compromised if tainted by the nondisclosure of exculpatory evidence.

Independent Investigation of the Root Causes of GTTF Scandal

The Monitoring Team and the Court itself made clear that BPD needed to retain an outside, independent
team of investigators to thoroughly examine the Gun Trace Task Force scandal and attempt to discern
how far it reached, how it evolved, and how BPD could have allowed it to occur without any internal
repercussions for the officers involved. In response, BPD hired Michael Bromwich and his firm in 2019
to conduct to conduct such a review. Mr. Bromwich has a credentialed history of doing such work. The
former Inspector General of the Department of Justice, he served as monitor of the Metropolitan Police
Department in Washington, D.C. when MPD underwent DOJ-monitored reforms similar to the reforms
BPD is now undertaking, and he has conducted several high-profile investigations involving misconduct
by law enforcement officers. Mr. Bromwich’s independent investigation is underway.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 33
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 35 of 164
|CHALLENGES AHEAD
Quality of Investigations

Improving the quality of PIB investigations is a central goal of the Consent Decree. In 2019-20, the
Monitoring Team conducted its first comprehensive compliance review of PIB investigations. The review
covered a significant sample of PIB investigations conducted in 2018. The report detailing the Monitoring
Team’s compliance review is attached as Exhibit 2.

The review serves as a preliminary “baseline” assessment of the quality of PIB investigations. It is
intended to show BPD’s performance as of 2018, prior to implementation of the major reforms required
by the Consent Decree. As of 2018, PIB had sought to implement certain improvements but was still
contemplating core improvements. BPD has conceded that improving the quality of PIB misconduct
investigations was a work in progress in 2018.

As a pre-reform “baseline” assessment, the Monitoring Team’s review did not consider the current quality
of BPD’s internal investigations. Thus, it does not reflect the impact of the systemic changes, summarized
above, that have been underway since 2018.

Because the Monitoring Team reviewed PIB investigations from a pre-reform period, our overall
conclusion is unsurprising: PIB will need to dramatically improve on its performance from 2018 before its
investigations are of sufficiently high quality to comply with the Consent Decree. The Monitoring Team’s
specific findings are as follows:

„ The overall quality of BPD’s misconduct investigations in 2018 was low. Some 40 percent
of misconduct investigations in 2018 were of “fair” or “poor” overall quality. Many cases (42 percent)
were “administratively closed” without a disposition on the merits—a disposition prohibited by the
Consent Decree—and cases that resulted in a specific disposition often failed to adequately outline the
factual basis for that disposition.
„ The condition of misconduct case files was poor. Documentation of investigations was chaotic
and inconsistent, leading to great uncertainty in a number of instances about both the factual issues
presented and the investigative steps taken.
„ Communication with involved civilians and officers was poor. Individuals making complaints
to BPD were sent written notice that their complaint was received in only one-quarter of cases.
Implicated BPD personnel received notification in less than one-quarter of cases.
„ Inefficiencies and poor management resulted in missed timelines and wasted efforts in
minor cases. It appeared that some 60 percent of investigations were not completed within the
Consent Decree’s 90-day deadline, with extensions of time never requested. Several cases remained
pending for extended periods of time without explanation, justification or approval.
„ Investigators too often failed to interview complainants and other necessary witnesses.
Complainants were interviewed only 41 percent of the time; subject employees were interviewed in 44
percent of cases. Where interviews occurred, the quality of those interviews appeared encouraging,
but only 22 percent of cases had complete audio or video records of all witness interviews.
„ Investigations failed to flag related operational and performance issues exposed by the
investigation, such as training deficiencies, repetitive problematic behavior, and potential
policy shortcomings.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 34
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 36 of 164
Although the Monitoring Team identified some encouraging data points—e.g., complaints of excessive
force, unlawful search and seizure, and First Amendment violations were both relatively rare and
well-investigated—our findings confirm the need for PIB to transform the way it investigates officer
misconduct.

Staffing

The challenges facing PIB remain severe. PIB needs additional investigators—the Staffing Plan estimates
a need for three dozen more—to eliminate the case backlog and ensure timely, thorough, fair resolution of
misconduct allegations. The new policy on negotiated resolution of minor misconduct allegations should
help. But the problems caused by investigative delays, including the absence of swift accountability and
the depletion of officer morale, will not be cured until BPD assigns and trains additional, qualified PIB
investigators.

Training

The completion of the PIB investigations manual in November 2019 was to be closely followed by the
development of a training curriculum designed for PIB investigators, supervisors and commanders. The
initial draft of the curriculum was due to Monitoring Team and DOJ in early February 2020, to be followed
by several months of collaboration and public comment, with a completion date of July 1, 2020, and a
start date of July 15, 2020. The deadlines for this crucial training had to be extended for two reasons.
First, the work restrictions necessitated by the COVID-19 delayed preparation of the curriculum. Second,
for financial and other logistical reasons, BPD elected to not move forward with the program developed
by Roger Williams University, which it had originally selected.

BPD has now identified a skilled subject matter expert to assist in the development of the PIB investigator
training curriculum in the coming months. The Monitoring Team and DOJ will collaborate with BPD
throughout the curriculum development process. Under a newly revised schedule, the curriculum is to be
finalized and the training set to take place in the first quarter of 2021.

In addition to training for PIB investigators, BPD is presently developing a curriculum for Department-
wide training to implement a program on officer intervention called Ethical Policing is Courageous
(“EPIC”). EPIC replicates a well-received initiative implemented in the New Orleans Police Department.
It provides officers with practical skills and techniques to employ when facing situations in which they must
intervene to prevent or stop misconduct and problematic performance by a fellow officer. Commissioner
Harrison views EPIC training as an important part of the culture change BPD is undertaking. The
Monitoring Team and Judge Bredar agree. Currently, BPD projects beginning in-person EPIC training
in October 2020. As with many other training deadlines, public health realities may compel changes in
the projected timeline.

Technology and Reporting

BPD must begin devising an Early Intervention System (“EIS”), which will permit supervisors to identify
and intervene to correct the behavior of officers who begin to display a tendency for violating policy. As
discussed in greater detail in the Technology section of this report, BPD is still more than a year away
from implementing a fully functional, modern EIS. But because the EIS will depend upon source data from
other systems that are being upgraded or overhauled (e.g., IAPro and the Records Management System),
BPD must begin to establish the data requirements for the EIS now, as it embarks on implementation of
the Technology Resource Plan issued last year.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 35
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 37 of 164
BPD also must develop a plan for collecting the data needed to complete the statistical reports and
internal PIB audits required by paragraphs 402 and 405 of the Consent Decree.

Deputy Commissioner Brian Nadeau has solidified his leadership of PIB. Despite the serious challenges
PIB continues to face, the Monitoring Team has growing confidence that, with D.C. Nadeau in charge,
BPD is not only committed to PIB reform, but capable of achieving it.

In the short run, the Monitoring Team remains concerned that the benefits derived from our previous
in-person discussions with PIB staff will begin to erode if pandemic-induced restrictions on in-person
meetings remain in place for a prolonged period. Even if these measures must remain in place, however,
the Monitoring Team is committed to finding ways to innovate, to maintaining our close working
relationships with PIB staff, and to continuing our rigorous oversight of PIB.

|THE NEXT SIX MONTHS

In the next reporting period, in collaboration with the Monitoring Team and DOJ, BPD will complete and
disseminate its first Quarterly PIB Report (CD 402) compiling comprehensive aggregate data regarding
misconduct complaints, both internal and external. The Monitoring Team has reviewed an initial draft,
which is of high quality. (The Monitoring Team has not yet conducted its own assessment of the data
under Paragraph 469.n. of the Consent Decree).

BPD will continue to work closely with the Monitoring Team and DOJ on BPD’s comprehensive review
of its policies pertaining to discipline. The scope of this review is significant, as there are numerous
policies implicating discipline that need to be reviewed and updated or drafted and approved. Included
among these policies is BPD’s core policy on discipline, which includes BPD’s disciplinary matrix. The
matrix prescribes disciplinary outcomes based on the severity of a violation and the officer’s history of
discipline.

As noted, BPD will develop a training curriculum for PIB investigators, supervisors and commanders.
BPD, the Monitoring Team and DOJ have established a bi-weekly call intended to prioritize the
development of the curriculum. In addition, BPD will develop the curriculum for and deliver EPIC
training.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 36
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 38 of 164
TECHNOLOGY

Paragraph 267 of the Consent Decree requires BPD to “provide its officers with the Technology necessary
to implement the Material Requirements of this Agreement . . ..” Paragraphs 268-278 then set forth the
Consent Decree’s technology requirements.

BPD and the City previously completed and submitted a Resource Study (CD 268, 270), which identified
current BPD systems, described the current state of those systems, and made preliminary recommendations
for improvements. Following completion of the Resource Study, BPD and the City prepared and published
a Resource Plan (CD 269-70, 272), which the Monitoring Team approved on December 1, 2018. See ECF
No. 164. The Resource Plan, which is being updated annually (CD 275), addresses how BPD will provide
the necessary computer equipment and access required for personnel to discharge their duties, acquire a
centralized records management system, and ultimately develop an Early Intervention System (“EIS”).
BPD is required to use its best efforts to implement the Resource Plan (CD 274).

The Technology provisions of the Consent Decree also require BPD to disclose to the public the acquisition
of certain new equipment or activity to be used in enforcement activities (CD 278). Further, data collection
and data analysis are required in nearly every area of the Consent Decree, not only to enable the Monitoring
Team to assess compliance, but to enable BPD leadership to better manage the Department. BPD must
review and analyze data in a number of subject areas, including investigatory stops and detentions (CD
41), vehicle stops (CD 46), stops, searches and arrests (CD 82-86), use of force (CD 211-217), transport of
persons in custody (CD 232), and misconduct investigations (CD 392), among others.

Without technology improvements, BPD will not be able to achieve compliance with the Consent Decree.
As the Consent Decree states, compliance is truly “dependent upon BPD acquiring or developing the
appropriate technology.” (CD 267).

After some delay, BPD is now diligently implementing the Technology Resource Plan. With the hiring
of a new Chief Technology Officer and additional professional staff, BPD has fortified its IT governance
structure. Further, BPD has now selected a vendor, secured funding, and begun working in earnest to
design and develop a new Records Management System (“RMS”), which is indispensable to achieving
compliance in nearly every area of the Consent Decree.

In short, BPD has completed the planning process and moved into the implementation phase of
compliance. However, because it has only recently begun implementation, its compliance score
in the Technology Modernization category is “4a” (implementation – not yet assessed).

COMPLIANCE SCORE:
TECHNOLOGY
MODERNIZATION
4A Implementation –
Not Yet Assessed

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 37
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 39 of 164
|AREAS OF PROGRESS
IT Staffing and Governance

BPD hired a Chief Technology Officer and additional IT staff, including two Project Managers, a Business
Analyst, two Developers, an Enterprise Architect, a Fiscal Analyst and a Systems Integration Specialist.
Further, the CTO has begun to establish proper IT governance. He promptly assessed his unit’s
operations, identified the unit’s needs, and implemented foundational workflow management structures
and processes that have long been absent.

Records Management System

On September 30, 2019, BPD released a Request for Information (RFI) for a new Record Management
System (“RMS”). After an extensive evaluation of responses to the RFI, BPD identified Axon Enterprises
to provide the new RMS. A statement of work was developed, and implementation is beginning.

IAPro

IAPro is the system BPD uses to track both use of force incidents/investigations and misconduct
investigations/discipline. Though it took much longer than expected, BPD was able to complete its
upgrade of IA Pro, and is now on the most current, standardized version of the platform. This will ensure
that BPD is able to support and maintain the system going forward, and leverage all of the features and
functionality available to IAPro customers.

Workforce Management System

BPD has identified and deployed and is currently configuring a department-wide work schedule
management application (e-Resource), which is nearly complete.

Learning Management System

BPD has identified a vendor to provide Learning Management System (“LMS”) services.  The acquisition
package has been completed and was submitted for City approval. The LMS is on track for full
implementation by October 2020.

Body-Worn Cameras

BPD acquired new body-worn camera technology and the related policy was approved. While initial
deployment was delayed as a result of COVID-19, it is now in progress.  

Transport Vehicle Cameras

BPD has begun to install new cameras in its transport vehicles. The new cameras will make it less
cumbersome to upload and review recordings of transport events.

Policy regarding Public Disclosure of Technology Acquisitions

BPD is required to publicly disclose the acquisition of certain new equipment or technology to be used
in enforcement activities (CD 278). In furtherance of this requirement, BPD finalized Policy 606, which
establishes the procedure for such disclosure in September 2019. See ECF No. 239. The policy is among the
many measures BPD is adopting under the Consent Decree to enhance the transparency of its operations.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 38
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 40 of 164
 |CHALLENGES AHEAD
While BPD has taken initial steps to address its technology deficits, it still has a long way to go to fully
realize a modern and sustainable IT environment. BPD must continue to plan for and sustain its IT
infrastructure needs to integrate data silos, eliminate duplicative data entry, improve data quality, and
collect and maintain data thoroughly. Otherwise, BPD will fail to achieve compliance with the numerous
Consent Decree requirements that depend upon BPD’s ability to capture, maintain, synthesize and
analyze comprehensive electronic data regarding its law enforcement activities. As explained elsewhere
in this report, BPD’s long-standing technological deficiencies:

„ Make it more difficult for supervisors to assess, commend and correct deficiencies in the performance
of their officers
„ Increase the amount of time officers must spend writing reports
„ Lead to under-reporting of non-arrest police encounters, especially investigative stops and weapons
pat-downs
„ Diminish the amount of time officers have for community policing and additional enforcement activity
„ Make Consent Decree-required analyses of police encounters (particularly stops, searches and arrests)
impracticable, if not impossible
„ Constrain BPD’s ability to assess performance by officer, unit, District and Department-wide
„ Limit public transparency of BPD enforcement activity
If used properly, a new RMS featuring electronic field-based reporting should address each of these
issues.

BPD must be vigilant about ensuring three things for each and every IT initiative. First, each initiative
must be accompanied by clear, consistent policy and training. Second, to ensure proper accountability
and maximum effectiveness of each IT initiative, BPD must consider how that initiative, as well as the
data it generates and retains, will be audited, reviewed or evaluated for a solution. Third, BPD must
keep its eye on the long-range development of an Early Intervention System. EIS implementation will
come toward the back end of BPD’s modernization timeline because BPD must first capture quality data
in each source system. Accordingly, the data requirements for BPD’s EIS must be integrated into each
IT solution implemented under the Resource Plan. What that means is that BPD must determine which
data it needs for its EIS as it configures the data requirements for the RMS, and any other system that
will contain data upon which the EIS will rely.

|THE NEXT SIX MONTHS

Over the next reporting period, BPD will focus on the implementation of the RMS and the deployment of
the new body-worn cameras. RMS implementation, in particular, is extremely complex and will impact
nearly every area of the police department. BPD must commit sufficient time and resources to ensure
the proper coordination of the project.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 39
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 41 of 164
STAFFING, PERFORMANCE EVALUATIONS AND
PROMOTIONS
The Consent Decree requires BPD to complete a comprehensive Staffing Study to determine the
appropriate number of sworn and civilian personnel needed to effectively provide police services, enable
supervision, and satisfy the requirements of the Consent Decree (CD 428). The Consent Decree further
requires that, based on the Staffing Study, BPD must develop a Staffing Plan that will ensure a sufficient
number of deployed personnel to, among other things: implement and sustain effective community and
problem-oriented policing; conduct timely misconduct investigations; supply sufficient patrol officers to
each District without resorting to drafting (i.e., forced overtime), except in unforeseeable circumstances;
promote unity of command when feasible; provide a sufficient number of supervisors; and account for
BPD’s and the City’s existing and projected resources (CD 429). BPD must implement the Staffing Plan,
but may do so in a phased manner that reflects the City’s and BPD’s fiscal resources (CD 430).

As for performance evaluations and promotions, the Consent Decree obligates BPD to have supervisors
meet with officers to discuss their annual performance reviews, which must include written discussions
of the officers’ performance during the rating period, any areas for growth and achievement, and any
areas requiring further training and supervision (CD 431). Direct supervisors must use a formalized
system to document annual performance evaluations for each officer and quarterly evaluations of
probationary employees (CD 432). In addition to these formal evaluations, supervisors must meet with
their subordinates on an ongoing basis to discuss performance and must document their communications
regarding performance challenges and areas for growth (CD 433). The Consent Decree further requires
BPD to conduct performance evaluations of each supervisor (from first line supervisor through commander),
which will include assessments of ability and effectiveness in conducting performance reviews, including
monitoring, deterring and addressing misconduct by officers they supervise (CD 434). Finally, BPD must
ensure its promotional system has clear criteria prioritizing effective, constitutional, and community-
oriented policing as factors for promotion (CD 435).

In December 2018, BPD produced a Staffing Study compliant with Paragraph 428. BPD completed the
Staffing Plan required under Paragraph 429 in February 2020.

In a significant, positive break from past practice, BPD has conceptualized a new promotions regime
under Paragraph 435, including a formal application and committee review process. Policy 1738, focused
on command promotions for the ranks of Captain and Major, was recently approved. See ECF No. 340.
Because Paragraph 435 also references “promotional systems,” meaning promotions to all ranks, the
development of a promotions policy for sergeants and lieutenants is currently underway.

BPD, DOJ and the Monitoring Team have also started work on BPD’s performance evaluation system.

BPD is making reasonable progress on the foundational requirements in these areas. On Staffing
(which is coupled with Recruitment and Retention in the scoring framework, see Exhibit 1),
BPD’s compliance score is “4a” (implementation – not yet assessed), as BPD recently completed
its comprehensive Staffing Plan but has not yet progressed far enough into the implementation
phase to warrant an “on track” or “off track” score. On both Employee Performance Evaluations
and Promotions, BPD’s compliance score is “2” (policy phase). BPD has responsibly complied
with Monitoring Plan deadlines but has not yet finalized policies in either compliance category.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 40
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 42 of 164

COMPLIANCE SCORE:

4A
STAFFING Implementation –
Not Yet Assessed

EMPLOYEE COMPLIANCE SCORE:

2
PERFORMANCE Policies/
EVALUATIONS Planning

COMPLIANCE SCORE:

2
PROMOTIONS Policies/
Planning

|AREAS OF PROGRESS
Staffing Plan

After several delays discussed in prior reports, BPD completed its comprehensive Staffing Plan in February
2020. See ECF No. 292. The Plan details the staffing needed to meet BPD’s anticipated workload and to
comply with the Consent Decree. Key elements from the Plan, available here, include the following;

„ BPD needs 956 police officers, 133 sergeants and 27 lieutenants assigned to the Patrol Division in
order to satisfy BPD’s call-for-service demands and still reserve 40% of officers’ time on activities
other than responding to calls-for-service, proactive community policing in particular.
„ The Public Integrity Bureau needs an additional 39 investigators to address the current backlog in
complaint investigations and to meet the requirements of the Consent Decree. 
„ Sergeants should be assigned the same day-off group as their subordinates and an additional 34
sergeants should be assigned to Patrol. Presently, sergeants have limited contact and time with the
officers they supervise because of the existing patrol work schedule. This prevents unity of command
and makes effective supervision exceedingly difficult.
„ Certain specialized units should be consolidated, while others should have fewer sworn personnel or
should be eliminated, and certain administrative functions should be performed by civilians, freeing
up sworn personnel for other assignments.
„ Civilianization of certain functions is required to increase the number of sworn officers on the street
and to bring needed specialized skills into the Department. 
„ BPD continues to lose officers through retirement and other forms of separation faster than it is
adding new officers. Increasing the size of recruit classes, increasing the number of recruit classes,
and reducing the Academy attrition rate would increase the number of new sworn officers. So, too,
would reducing the officer attrition rate, which BPD must figure out how to address.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 41
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 43 of 164
The City and BPD must now work toward meeting the requirements in the Plan. BPD’s budget, which
was recently approved for Fiscal Year 2021, reflected some of the recommendations in the Plan around
consolidating and eliminating certain specialized units. In particular, as of July 1, officers from  the
Aviation, Mounted, Marine and traffic-related units were transferred to Patrol, where the need for sworn
personnel is most acute. The transferred officers will continue to perform their specialized functions, but
on a limited basis, with the exception of the officers from the Mounted and Marine Units, which were
eliminated in the FY 2021 Budget.

Further, by moving the Training Academy to the University of Baltimore, BPD is able to accommodate
more recruits at once. If hiring improves—something BPD has not yet sustainably achieved—BPD will
be able to bring on a greater number of new officers more quickly.

Performance Evaluations and Command Promotions

The Monitoring Team and DOJ are collaborating with BPD to revise its performance evaluation system
for lieutenants, sergeants and officers. This includes assessing and upgrading the performance evaluation
policy, manual and form. Due to delays caused by the coronavirus pandemic, the work recently began and
is expected to continue until the end of the year. Once the work is completed, BPD will turn to revising
its performance evaluation system for command ranks, Captain and above.

BPD has finalized the development of a command promotions policy, Policy 1738, which covers the ranks
of captain and major. The Monitoring Team approved this policy in September 2020 See ECF No. 340.
The adoption of this policy is an important development. It establishes that promotion into a command
rank will be based exclusively on merit—not on who an officer knows, as has often been the case in
the past—and thus conveys that transparency, fairness, impartiality, and procedural justice are just
as important to BPD’s internal operations as they are to BPD’s interactions with community members.
Policy 1738 formalizes the practice that Commissioner Harrison implemented on his arrival. Since then,
no officer has been promoted to captain or major outside the competitive, merit-based process prescribed
by the new policy.

Equally important, BPD must develop a promotions policy for ranks below captain. The work on that
policy began in 2020, with a scheduled date of completion, following public feedback, in late October 2020.

|CHALLENGES AHEAD
The fiscal impact of COVID-19 on Baltimore’s revenues and the resulting impact on BPD’s budget present
a major obstacle to full implementation of the Staffing Plan, at least in the short run. The Staffing Plan calls
for a total of 675 civilian personnel. This requires an increase of 144 civilian employees. Civilianization of
certain functions is essential in two ways. First, civilians free up sworn members for the Patrol Division,
where BPD needs more officers. Second, professional civilian members will bring needed expertise to the
Department. This is especially true in the critical function of training, where the Staffing Plan calls for
additional instructors. Despite BPD’s need for additional civilian employees, the City recently imposed
a civilian hiring freeze and cut 37 vacant civilian positions from BPD’s budget. These reductions will
naturally affect BPD’s ability to achieve the civilian hiring goals established in the Staffing Plan. The
Training Academy is likely to be one of the areas hardest hit. The Staffing Plan recommends increasing
professional staff by 38 and decreasing sworn staff by 24. This effort is currently on hold.

Without being able to hire civilians to move certain officers to the street, and with only a limited number
of officers who can be moved to the street through the elimination or consolidation of specialized units,
retaining existing sworn personnel and hiring new sworn personnel will be the primary ways for BPD to

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 42
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 44 of 164
meet the sworn staffing targets identified in the Staffing Plan. The Plan establishes that BPD will need
2,785 sworn members. To handle 911 calls for service and engage in the community policing activities
called for in the Community Policing Plan, the Patrol Division, as noted, will need a total of 956 officers,
133 sergeants, and 27 lieutenants across BPD’s nine districts, as noted above. As of July 8, 2020, there
were 777 officers, 99 sergeants, and 27 lieutenants assigned to Patrol. Thus, as of July 8, 2020, to
achieve the Staffing Plan’s goals, BPD had to place 179 additional officers and 34 additional sergeants
in Patrol. Additional officers are also critically needed in the Public Integrity Bureau, where misconduct
investigations remain inadequate and take far too long.

Hiring the additional officers recommended in the Staffing Plan will be a tall order. Recruitment and
hiring have been improving but hiring has barely been able to keep pace with attrition thus far in
2020. Further, although the economic dislocation caused by COVID-19 could increase the number of
applications and thus the number of officers hired, that remains uncertain, and recruiting and retention
efforts could be adversely affected by recent anti-police protests. Achieving the staffing levels established
in the Staffing Plan is among BPD’s most severe challenges.

|THE NEXT SIX MONTHS

BPD will continue work on a promotions policy for ranks below Captain and will seek finalize its
performance evaluation systems.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 43
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 45 of 164
STOPS, SEARCHES, ARRESTS AND VOLUNTARY POLICE-
COMMUNITY INTERACTIONS
In recognition of the importance of the Consent Decree’s requirements on stops, searches, arrests and
voluntary police-community interactions (“SSA”), the Consent Decree’s provisions addressing those
interactions are extensive. They compel BPD to revise its policies and training curricula; provide thorough
prescriptions for communicating with individuals, performing field interviews, and conducting stops,
weapons pat downs, searches and arrests; and establish detailed training, documentation, supervisory,
and data collection and review obligations (CD 29-86).

Over the past 30 months, BPD has made reasonable progress toward satisfying the Consent Decree’s
preliminary SSA requirements:

„ BPD has revised two different sets of SSA policies, is working toward completion of a third and final
set covering lesser misdemeanor offenses, and has made adjustments to the already-revised sets to
reflect both minor changes in the law and the philosophical orientation of BPD’s new leadership.
„ BPD also has developed and is in the midst of delivering comprehensive e-learning and in-class SSA/
FIP II training; recently developed e-learning for supervisors on supervisory review requirements for
stops, searches and arrest; and prepared lesson plans and provided training to both command staff
and officers in response to the Monitoring Team’s findings regarding BPD’s stops, searches and pat-
downs of civilians in Harlem Park following the shooting death of Detective Sean Suiter.
„ On data collection and analysis, BPD’s antiquated IT systems have severely hindered progress, but
BPD recently has developed and begun implementing a methodology for auditing and reporting on
arrests that result in “release without charge” dispositions.
The deadlines for certain SSA deliverables under the Third-Year Monitoring Plan have been delayed
because of resource redeployments necessitated by the COVID-19 pandemic, as well as public safety and
security demands arising from sustained, large-scale protest activity in May and June 2020. Nevertheless,
BPD’s work on SSA requirements—including policy revisions, resumed classroom training, and the
inaugural arrest data audit—has continued.

Because BPD has largely completed policy revisions and continues to provide core SSA training,
its score in each of the SSA compliance categories is “3” (training phase). These categories
include (i) Stops, Field Interviews and Voluntary Contacts, (ii) Searches, (iii) Arrests, and (iv)
Review and Supervision.

STOPS, FIELD COMPLIANCE SCORE:

3
INTERVIEWS AND
Training
VOLUNTARY CONTACTS

COMPLIANCE SCORE:

3
SEARCHES
Training

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 44
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 46 of 164

COMPLIANCE SCORE:

3
ARRESTS
Training

COMPLIANCE SCORE:
REVIEW AND
SUPERVISION
3 Training

|AREAS OF PROGRESS
Policies

In October 2018, after extensive collaboration with the Monitoring Team and DOJ and receipt of public
feedback, BPD finalized revisions to its core SSA policy, Policy 1112, titled “Field Interviews, Investigative
Stops/Detentions, Weapons Pat-Downs and Searches,” as well as five additional policies: 1007 (Search and
Seizure Warrants), 1013 (Strip Searches and Body Cavity Searches), 1106 (Warrantless Arrest Procedures
and Probable Cause Standard), 1109 (Warrantless Searches) and 1505 (Foot Pursuits). Two months
later, with additional feedback from the Monitoring Team, DOJ and community members, BPD finalized
six additional SSA policies, including Policies 906 (Traffic Citations), 1002 (Securing & Interviewing
Witnesses), 1104 (Arrest Warrants), 1105 (Custodial Interrogations), 1108 (DUI/DWI Arrest Procedure),
and 1114 (Persons in Police Custody).

BPD spent the early part of 2020 focused on revisions to SSA policies that address low-level offenses,
including “quality of life” offenses. These policies include: Criminal Citation Procedures (803), Civil
Citation Procedures (808), Marijuana-- Uniform Civil Citation (809), Misdemeanor Theft Procedures
(812), Video Surveillance Procedures (1014) and Lesser Offenses and Alternatives (1018). These policies
are intended to properly prioritize enforcement actions that preserve public safety while simultaneously
promoting fair, non-arbitrary practices that foster the community’s trust. The principles reflected in these
policies are currently being reinforced through SSA/FIP II training, particularly in the lesson plan on
“most effective, least intrusive response.” That lesson plan teaches officers to exercise their discretion to
take the least intrusive law enforcement action consistent with preserving public safety when confronted
with an individual who may be committing a lesser offense, such as loitering, open container, trespassing,
failure to obey, disorderly conduct, and marijuana possession.

Because this final set of SSA policies is integral to advancing the Consent Decree’s objective of community-
oriented policing, now reflected in BPD’s Community Policing Plan, the deadline for issuing drafts of these
policies for public comment was extended from the second monitoring year until this year to accommodate
the extension of the deadline for preparation of the Community Policing Plan. The initial public comment
period on these policies ran from February – March 2020. BPD received both internal and community
feedback. The COVID-19 pandemic then interrupted the process, requiring postponement of additional
work. The Monitoring Team, BPD and DOJ have been discussing resumption of the process in the near
future and expect that BPD will solicit additional public feedback and finalize the revisions by the end
of 2020.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 45
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 47 of 164
Training

The central component of BPD’s SSA work over the past nine months has been the development and
delivery of both e-learning and classroom training on the first two sets of BPD’s revised SSA policies. The
training covers policies on voluntary contacts, field interviews, stops, arrests, interviews/interrogations,
weapons pat-downs, and searches, as well as policies on foot pursuits, most effective/least intrusive
response, custody, transport, and booking. The training also incorporates instruction on previously
approved policies on fair and impartial policing (“FIP”), including Policy 317 (Fair and Impartial Policing)
and Policy 720 (Interactions with LGBTQ Individuals). With input from the Monitoring Team and DOJ,
the approach BPD has taken to integrating instruction on FIP polices with instruction on SSA policies
is similar to the approach BPD took to integrate FIP policies into the successfully concluded training on
use of force.

BPD drafted and finalized an effective, six module e-learning curriculum relatively quickly. Officers must
complete the e-learning and pass tests for each module before becoming eligible for in-class instruction.

BPD experienced some difficulty with the development of the in-class curriculum component of SSA/
FIP II training. That was not unexpected, as BPD continues to adapt to adult-oriented, scenario-based
instructional methods. It is far more challenging to develop training curricula employing these methods
than it is to develop lecture-based curricula. BPD went through several iterations of the SSA/FIP II
classroom curriculum before getting it right. Following two public comment periods, input from the new
Community Training Review Committee following a pilot session, several additional pilot sessions with
officers, and additional input from the Monitoring Team and DOJ, BPD finalized the curriculum in early
February 2020. The curriculum successfully conveys the new SSA and FIP policies through interactive
activities, scenarios and role plays that, as in the real world, layer in the requirements of multiple policies
at once.

Classroom instruction began on February 10, 2020, shortly after the curriculum was approved.
Unfortunately, after five weeks, the training had to be halted because of the pandemic. It restarted
in July 2020. Originally scheduled for 36 officers in each of two classrooms for the two-day program,
physical distancing restrictions have required BPD to reduce class sizes to 24 officers. As a result, BPD
will not be able to complete the training for all officers until the end of 2020.

The hiatus between mid-March and mid-July did not go to waste. The first several weeks of instruction
revealed that certain SSA policies were not as clear as they needed to be or had to be modestly adjusted
for practical reasons. These included policies addressing securing/interviewing witnesses, strip searches,
custodial interrogations, and warrantless searches. (The pilot sessions in January 2020 similarly exposed
a handful of policy provisions in need of clarification.) After lengthy and intensive consultation with the
Monitoring Team and DOJ, BPD is making the necessary changes in policy and has adjusted the training
curriculum accordingly.

Based on the Monitoring Team’s observations, the quality of instruction rapidly improved over the first
two weeks of SSA/FIP II training and, even after the four-month break, has remained at a high level,
even with new instructors brought on board after the break. The course has benefited considerably
from the participation of law instructors (including some recently hired), who have ably complemented
the police instructors. Generally, course content has been delivered effectively, including in the face of
classroom pushback, and most officers appear engaged in the discussions of the course material. The
results of the surveys that the Training Academy furnishes to officers at the end of the course corroborate
this observation, as the vast majority—nearly 85%—have come away from the course either agreeing
or strongly agreeing that they know how to apply BPD’s SSA policies and believe the course material

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 46
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 48 of 164
is engaging, while only a small number, roughly 5%, either disagree or strongly disagree with those
positions. Many officers also have provided positive narrative comments about the quality of the training
(even if some officers have expressed skepticism about certain aspects of the revised policies).

As much as any of the reform measures implemented thus far, BPD’s revised SSA policies reflect the
culture change that the Consent Decree is intended to achieve—that is, a change toward policing that
is simultaneously pro-active, community-oriented and constitutional. Training on BPD’s revised SSA
policies is thus a crucial feature of the reform process. The fact that the Training Academy appears to
be hitting its stride with the classroom component of SSA/FIP II training bodes well not only for future
Consent Decree-mandated training courses, but for the reform effort overall.

In addition to Department-wide SSA training, BPD supervisors will soon receive e-learning instruction
on supervisory review of stops, searches and arrests. The e-learning will reinforce the fact that, because
constitutional policing and restoring community trust are dependent on lawful SSAs and proper SSA
reporting, constitutional policing and restoring community trust are dependent on sergeants and
lieutenants embracing and enforcing BPD’s SSA policies.

|CHALLENGES AHEAD
Training

Although the Academy is successfully providing core SSA/FIP II training, despite the COVID-induced
stoppages, there is a substantial amount of SSA-related training left to develop and deliver over the
next six to nine months. As noted in the Training section above, the Academy must draft and provide
e-learning and curricula for two days of classroom instruction for the third tranche of core Consent
Decree training, which will be delivered later in 2021 and focus on the biases that can affect officers’
interactions with civilians, including their decisions regarding stops, searches and arrests. In addition,
the Academy must develop and deliver training curricula on lesser offenses to reinforce BPD’s policy on
“most effective, least intrusive response.” That training, whose development is underway, is slated for
spring-summer 2021 and will be part of Department-wide instruction on community policing. Even in
the absence of a pandemic, preparing this curricula and providing this classroom instruction would tax
BPD resources. In the midst of a pandemic, it will be even more difficult, especially given all of the other
training courses the Academy is expected to develop.

Reporting on Stops, Searches and Arrests

In each semiannual report, the Monitoring Team has identified the serious deficiencies in BPD’s reporting,
collection and maintenance of data on stops, searches and arrests. Although a new record management
system featuring electronic field-based reporting is scheduled to come online within the next year,
the absence of complete, reliable SSA data has made it impossible for BPD to satisfy Consent Decree
requirements regarding the collection and analysis of such data. Moreover, it has rendered the Monitoring
Team unable to begin thoroughly examining BPD’s compliance with Consent Decree requirements for
stops, searches and arrests. The likely result of the enduring deficiencies in BPD’s IT systems will be the
extension of the life of the Consent Decree, as it will likely take longer for BPD to demonstrate sustained
compliance with the Consent Decree’s SSA requirements than if BPD had the capacity to adequately
collect and analyze SSA data from the beginning.

Paragraph 86 requires BPD to provide quarterly reports on the percentage of investigatory stops that
produce evidence of a crime, the percentage of weapons pat-downs the result in the seizure of unlawful
weapons, and the percentage of searches that result in the seizure of contraband. Data on these “hit

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 47
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 49 of 164
rates” can often indicate whether a police agency, a particular district, a particular unit or a particular
officer is conducting stops, pat-downs and searches with proper legal justification. BPD produced a draft
of its first Paragraph 86 report in 2019, consistent with the Second-Year Monitoring Plan. Because of
the dearth of reliable, meaningful SSA data, BPD had to scrap that draft, which did not contain any of
the required information. What BPD ultimately published was a report that did not include the required
information, but instead identified all of the gaps in SSA data that BPD needs to address in order to be
able to generate that information. At the suggestion of the Monitoring Team and DOJ, BPD requested that
the Court approve amendments to the Second-Year Monitoring Plan that relieved BPD of its obligation
to produce quarterly SSA reports until such time as BPD can properly report the information required
by Paragraph 86. The Court granted the request. Until BPD implements its new records management
system, and until officers are trained on and reliably utilize the system’s electronic field-based reporting
capacity, BPD will not be able to produce Paragraph 86 reports.

As the Monitoring Team has previously reported, the deficiencies in the reporting, collection and
maintenance of SSA data are most evident in BPD’s data on investigative stops. In short, stops are
dramatically underreported and, even when reported, they are recorded in BPD’s antiquated record
management system either extremely late or not at all. The Consent Decree obligates BPD to accurately
record and maintain data on stops, and requires not only BPD, but the Monitoring Team, to evaluate that
data on a regular basis. Accurate data on stops is arguably among the most important data for BPD to
maintain under the Consent Decree. DOJ’s findings emphasized that BPD was engaged in a pattern or
practice of unconstitutional stops, and that alleged pattern or practice is a primary reason for community
mistrust.

In 2019, BPD explored the feasibility of revising and possibly consolidating its paper report forms
to facilitate accurate reporting of stops in the short-term, prior to implementation of the new record
management system. The Monitoring Team and DOJ collaborated with BPD in this effort, providing
BPD with sample paper forms from other departments and offering edits to a revised form that BPD
prepared. Ultimately, however, BPD determined that developing a new form that would be used for only
a short period was impracticable.

Although the dearth of reliable, available SSA data has hamstrung the development of BPD’s capability
for self-evaluation and self-correction, as well as the Monitoring Team’s ability to evaluate BPD’s SSA
performance, there is one potential bright spot. In recent months, consistent with Paragraphs 75-79
of the Consent Decree, BPD’s Performance Standards Section has reviewed available arrest data from
the State’s Attorney’s Office (SAO) and Central Booking, and with input from the Monitoring Tam and
DOJ, has developed a methodology for identifying and analyzing arrests that, according to the SAO,
result in “release without charge” (RWOC) dispositions. BPD is currently preparing its first quarterly
report on RWOC arrests and will publish it before the end of this year. Reports on RWOC arrests will
follow every three months. The purpose of the RWOC analysis is to determine which RWOC arrests
were made without probable cause or inadequate documentation of probable cause, and then to use that
information to determine the need for additional training, non-disciplinary corrective action, or referral
for disciplinary or criminal investigation.

|THE NEXT SIX MONTHS

In the next reporting period, as noted, BPD will complete SSA/FIP II training for all officers, draft training
on community policing and lesser misdemeanor offenses, and complete SSA Policy Set III.

In addition, BPD will continue to work with the Monitoring Team and DOJ to determine whether, in
advance of the development of the new records management system, it can utilize any of the soon-to-

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 48
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 50 of 164
be-activated capabilities of the E-tix system to begin to collect and aggregate at least some of the SSA
data that the Consent Decree requires BPD and the Monitoring Team to evaluate—e.g., data on the
percentage of vehicle searches that result in the seizure of contraband. E-tix is a State-run system used to
record information from traffic stops. With the approval of the Monitoring Team and DOJ, BPD also will
begin to use it to record information from field interviews, as well as information on searches conducted
during traffic stops.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 49
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 51 of 164
IMPARTIAL POLICING
Paragraph 87 of the Consent Decree asserts that “policing fairly and without bias is central to promoting
broad community engagement and building partnerships between law enforcement and community
members that are an important part of effective policing.” To that end, the Consent Decree requires BPD
to: document the demographic category of all individuals who are stopped, frisked, searched, arrested or
make a complaint (CD 88); adopt policies that require fair, impartial, nondiscriminatory policing (CD 89);
establish an impartial policing training curriculum and properly train officers, with community input,
to perform their duties in a nondiscriminatory manner (CD 90-94); and consider whether officers engage
in nondiscriminatory policing in evaluating performance and making hiring and promotion decisions
(CD 95). Paragraphs 90-94 specify requirements for training members to carry out impartial policing
properly.

In the first two and a half years of the reform process, consistent with what the first three
Monitoring Plans have required, BPD has satisfied the policy revision provisions of the Consent
Decree for impartial policing and is nearing fulfillment of the training provisions. Because
BPD remains in the training phase of reform, its compliance score is “3” (training phase).

COMPLIANCE SCORE:

3
IMPARTIAL POLICING
Training

|AREAS OF PROGRESS
Policies

BPD worked with the Monitoring Team, DOJ and Baltimore residents to overhaul two policies that
require fair and impartial policing. The first is the core policy, Policy 317 (Fair and Impartial Policing).
The second is the policy addressed to interactions with LGBTQ individuals, Policy 720 (Interactions with
Lesbian, Gay, Bisexual, Transgender and Queer/Questioning Individuals). Revisions to both policies
were completed in August 2018.

Revised Policy 317 provides clear directives to officers to ensure fair, respectful, nondiscriminatory
treatment of community members, and expressly notes its close connection with policies addressing specific
law enforcement actions, such as stops, weapons pat downs, searches and arrests. Similarly, revised
Policy 720 provides key guidance on adherence to the requirements of both the Fourth and Fourteenth
Amendments. In so doing, it seeks to ensure that officers stop, pat down, search and arrest individuals,
including LGBTQ individuals, only with proper justification and that officers refrain from taking law
enforcement action against individuals based on perceived sexual orientation or gender identity. Policies
317 and 720 are referenced and incorporated in other, more specific policies governing law enforcement
actions, including BPD’s revised policies on stops, searches and arrests.

Training

In 2018, after attempting to craft a one-day standalone training on fair and impartial policing, BPD, with
prompting from the Monitoring Team and DOJ, made a key decision: it would weave training on fair and

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 50
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 52 of 164
impartial policing (“FIP”) into training on both use of force and stops, searches and arrests. The idea
was that encounters involving stops, searches, arrests and uses of force are when otherwise abstract FIP
concepts need to be applied on the street. Based on that decision, BPD is now conducting FIP training in
three different tranches—one integrated with use of force training (UOF/FIP I), another integrated with
training on stops, searches and arrests (UOF/FIP II), and a third that focuses squarely on FIP concepts
but will also include refresher training on use of force and SSA (FIP III). It is also worth noting that
training concerning active bystandership (called “Ethical Policing Is Courageous,” or “EPIC,” training),
which is currently in development, will similarly incorporate FIP principles.

As explained in the Training section above, BPD has now successfully integrated FIP concepts into its
Consent Decree-mandated training on use of force and stops, searches and arrests. BPD completed training
all officers on use of force/FIP in October 2019. BPD’s Department-wide training on stops, searches and
arrests/FIP (SSA/FIP II) is ongoing, having been interrupted by the coronavirus pandemic. BPD has now
resumed in-person training, but with smaller class sizes to ensure proper physical distancing, it will
not finish training all officers until late this year or early next year. Notably, based on officer feedback,
the pilot sessions of SSA/FIP training exposed ambiguities in several policies, and the first two weeks of
live SSA/FIP training exposed several additional policy shortcomings. As a result, with the approval of
the Monitoring Team and DOJ, BPD revised the pertinent policy provisions. The revisions have made
the associated training clearer. That will benefit BPD officers, who must comply with the policies on
the street, and will in turn benefit community members who encounter BPD officers in stop, search and
arrest situations.

In the next several months, BPD will begin the process of designing the curriculum for the third tranche
of Department-wide FIP training, FIP III, which will be delivered later in 2021, following training on
community policing and lesser misdemeanor offenses.

All FIP in-service training has incorporated the new, adult-oriented modes of learning that the Academy
is utilizing—i.e., facilitated problem-solving exercises based on videos, written scenarios, and live
simulation. The effective delivery of FIP training using a new method of instruction, while at the same
time fully integrating it with training on the primary types of encounters requiring application of FIP
principles, has been one of BPD’s most noteworthy accomplishments so far.

|CHALLENGES AHEAD
BPD’s long-term challenge is the same as it was when the Monitoring Team published its first semiannual
report in July 2018. As we said then:

One of the central findings of DOJ’s investigation was that BPD was engaged in a pattern-
or-practice of stopping, patting down, searching and arresting African Americans based
on their race. Revamping policies and implementing revised training curricula on both
impartial policing and stops, searches and arrests are important first steps in ensuring
that BPD officers treat people fairly and equitably, without regard to personal traits like
race, national origin, gender expression, and disability. But these are only first steps. As
with so many aspects of policing covered by the Consent Decree, the question is whether,
in practice, BPD officers interact with community members respectfully and refrain from
discrimination in the performance of their duties. The answer to that question will not come
quickly. It will depend on the adoption of revamped data collection systems that permit
easier, more comprehensive analysis of officer conduct; routine quantitative review and
analysis, over time, of the data gathered through those systems, once they are established;
routine qualitative review of field reports and body worn camera footage; and a proven

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 51
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 53 of 164
change in culture that values and rewards the quality of police work, rather than the number
of arrests. This will be the central task of the Monitoring Team—and more importantly, of
BPD itself, especially supervisory officers—in subsequent years.

ECF No. 126-1 at 70-71.

|THE NEXT SIX MONTHS

In the next six months, BPD will complete SSA/FIP II training for all members and begin to develop a
curriculum for FIP III training.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 52
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 54 of 164
USE OF FORCE
The Consent Decree obligates BPD to ensure that its officers resolve incidents without using force when
possible, employ de-escalation techniques to minimize the need to use force, avoid unnecessary injury or
risk of injury to officers and civilians when force is necessary, stop other officers from using excessive force,
report all uses of force, and be held accountable for using unreasonable force (CD 124). To accomplish
these objectives, the Consent Decree’s section on Use of Force contains requirements regarding policies
on use of force (including weapons-specific policies) (CD 125-65), training on use of force (CD 166-68),
reporting, reviewing and investigating use of force incidents (CD 169-210), and collecting, analyzing and
reporting data on use of force incidents (CD 211-17).

In the first two years of the monitoring plan, BPD successfully completed revisions to its use of force policies
and completed both e-learning and classroom training on the revised policies. BPD filed a certification of
completion with the Court in November 2019. See ECF No. 260.

Because of these developments, BPD has made marked progress toward satisfying the Consent
Decree’s requirements on use of force. It has fulfilled the threshold requirements—policy
revisions and training—and has moved into the implementation phase of compliance. Inasmuch
as assessments have not yet begun, its compliance score is “4a” (implementation – not assessed).

COMPLIANCE SCORE:

4A
USE OF FORCE Implementation –
Not Yet Assessed

However, the Monitoring Team should obtain at least a preliminary understanding of BPD’s performance
in the near future. As explained below, our initial comprehensive use of force compliance review will
begin in the next few weeks.

|AREAS OF PROGRESS
Policies

In 2018, BPD successfully completed revisions to its use of force policies. The policies include the core
policy on use of force, Policy 1115, as well as 13 other policies, which cover, among other things: de-
escalation; reporting, reviewing and investigating uses of force; and proper use of different weapons. The
policies BPD revised are as follows:

„ Policy 409 (Firearms Regulations)

„ Policy 412 (Patrol Rifle Program)
„ Policy 414 (Less Lethal Munitions and Chemical Agents)
„ Policy 710 710 (Level 3 Use of Force Investigations – SIRT)
„ Policy 719 (Conducted Electrical Weapons—CEW)
„ 724 (Performance Review Board)

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 53
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 55 of 164
„ 725 (Use of Force Reporting, Review and Assessment)
„ 1005 (Non-Uniform Policing Standards)
„ Policy 1111 (Batons/Impact Weapons)
„ Policy 1107 (De-escalation)
„ Policy 1115 (Use of Force)
„ Policy 1118 (Oleoresin Capsicum Spray).
„ 1503 (Emergency Vehicle Operation and Pursuit Policy)
„ 1602 (Canine Procedure)
BPD continues to examine and update the policies as needed to provide clarity to officers and ensure best
practices. For instance, in late 2019, as explained in the Youth Interactions section of this report, BPD
adjusted certain policies to account for circumstances involving youth.

Training

As previously reported, BPD successfully completed Department-wide training on its revised use of force
policies (UOF/FIP I) in October 2019. This was the first tranche of Consent Decree training. It included
several e-learning modules requiring test scores of 100% for completion, followed by two full days of
in-service learning that incorporated facilitated discussion and problem-solving with scenarios (video
and written), role plays, and animated simulation. All of the in-service lessons—whether in class, on
the simulator, or in the gym—demonstrated the tactical benefits of de-escalation and provided hands-
on guidance on using only force that is necessary, reasonable and proportional to the threat presented.
According to surveys completed at the end of the course, officers rated the training favorably, with most
indicating that it was a vast improvement over prior use of force training.

The training has proven extremely valuable to the reform process—not only for providing vital instruction
on use of force, which is obviously a key area of the Consent Decree, but also for enhancing the capacity
of the Training Academy. The use of force training program required BPD to hire more instructors;
introduced BPD to the process of creating a training program that incorporates both e-learning and an
intensive, scenario-based in-class curriculum; and allowed Academy instructors to hone their teaching
skills and gain more confidence in the classroom, particularly in addressing difficult, pointed questions
from their peers. The experience the Training Academy gained from developing and delivering high
quality training on use of force is informing training on other Consent Decree subjects.

Performance Review Board

The PRB reviews serious use of force incidents with the objective of improving policies, tactics, training,
and supervision where performance deficiencies are observed. In recent years, the PRB has not served
its intended purpose. It often has not reviewed events until many months (sometimes a year) after they
take place, and PRB members have been inappropriately reluctant to conduct meticulous, meaningful
reviews out of concern that their findings could result in disciplinary action against the involved officers.

BPD has initiated a pilot program with revisions, which require a more rapid review of serious events,
usually within 30 days of occurrence. The model being piloted includes a panel consisting exclusively of
Deputy Commissioners, reviews of events that are much more recent, ideally no more than 30 days old,
and an emphasis on the ameliorative purposes of the PRB—i.e., identifying needed improvements in
policies, tactics, training and supervision.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 54
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 56 of 164
The first pilot took place in early December 2019. In the following months, the PRB reviewed additional
events, including officer involved shootings. Unfortunately, due to the COVID-19 pandemic, PRB meetings
had to be canceled for several months. They are now resuming. Based on what it observed at PRB reviews
prior to the pandemic, the Monitoring Team has been encouraged by the pilot program and expects to
see it continues to progress.

BPD is also in the process of revising its PRB policy, Policy 724, to incorporate the reforms being tested
in the pilot.

Monitoring Team Use of Force Assessment

The Monitoring Team will soon begin its first comprehensive review of use of force. We are examining
a randomly drawn sample of use of force incidents from 2018-2019. In the lead-up to our review, we
collaborated with BPD and DOJ to develop an assessment tool designed to capture the requirements of
the use of force provisions of the Consent Decree, including the provisions addressing de-escalation and
appropriate force, reporting on uses of force, and supervisory review and analysis of uses of force. The
Monitoring Team expects to conclude its review by the end of the year or early 2021.

|CHALLENGES AHEAD
BPD must now hold officers and supervisors accountable to the revised use of force policies on which they
have been trained. Department-wide, that will require diligent collection and analysis of information
on both use of force incidents and the adequacy of use of force reporting. BPD does not have a track
record, much less an impressive track record, of self-evaluation on these subjects, and its data collection
capabilities remain deficient, as explained elsewhere. BPD thus has a steep hill to climb to satisfy the
Consent Decree’s requirements on use of force data collection and analysis.

More than anything, however, holding officers accountable to the new policies will require acceptance,
supervision and enforcement by sergeants and lieutenants. The new policies will not take hold until
these supervisors diligently review force incidents and force reporting, praise officers for upholding
policy, counsel officers whose performance is inadequate, and discipline officers for misconduct. Upper
management will need to do their part by making sure that first-line supervisors properly review use of
force incidents and reporting and by adjusting policy and training when needed.

|THE NEXT SIX MONTHS

In the next reporting period, the PRB pilot program will continue. In addition, BPD will be developing
the third tranche of its core Consent Decree training on use of force, stops, searches, arrests, and fair
and impartial policing (FIP III). As explained elsewhere, this tranche will focus on fair and impartial
policing concepts but will incorporate refresher training on use of force. Finally, the Monitoring Team
will continue with its first compliance review on use of force.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 55
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 57 of 164
TRANSPORTATION OF PERSONS IN CUSTODY
Ensuring the safety of individuals in custody is among the most important obligations of any law
enforcement agency. The Consent Decree requires BPD to: (1) equip all transport vans with seatbelts,
holding straps located along the rear area of each seat that individuals being transported may grip for
security during transport, and transport vehicle cameras (TVCs), and also equip all transport cruisers
with seatbelts (CD 223-24); (2) inspect transport vehicles monthly and create logs to memorialize the
inspections (CD 225); (3) establish and adhere to appropriate procedures for transporting prisoners
(including using seatbelts, straps, and TVCs) (CD 226-33), (4) establish and adhere to protocols for
documenting and comprehensively auditing prisoner transport events (CD 234-37), and (5) revise policies
and training curricula to ensure safe, effective prisoner transport (CD 238).

BPD has made reasonable progress on each of these requirements. It has revised policies, is presently
conducting training, has properly equipped its transport vehicles, is conducting required equipment
audits, recently began auditing the required number of transport events, is generally determining during
its audits whether vehicles are properly equipped and transport officer are complying with policy, and is
taking remedial action when there is non-compliance.

Because the coronavirus pandemic has prevented BPD from completing Department-wide
training on transport procedures on time, BPD remains in the training phase of compliance
in the Transportation of Persons in Custody category and thus has a compliance score of “3”
(training phase). That score may understate BPD’s progress because, with the equipment and
transport audits it is regularly conducting, BPD has begun implementation of operational
requirements.

COMPLIANCE SCORE:
TRANSPORTATION OF
PERSONS IN CUSTODY
3 Training

|AREAS OF PROGRESS
Transportation Equipment

During the first year of monitoring, BPD installed in its transport vehicles all equipment required by
the Consent Decree. All transport vans have been equipped with seatbelts for each seat, holding straps
located along the rear area of each seat that individuals may grip for security, and TVCs to allow live
monitoring of every transported occupant. In addition, all transport cruisers have been equipped with
seatbelts. Accordingly, BPD is satisfying the requirements of Paragraphs 223 and 224.

Transportation Equipment Audits

As required by Paragraph 225, BPD continues to perform monthly inspections of all transport vehicles and
creates logs to memorialize those inspections. The logs, which are used to verify the continued presence
and functionality of all required equipment, are available for inspection by the Monitoring Team and
DOJ. The Monitoring Team has routinely reviewed the logs. They appear complete and show that BPD
is working hard to ensure that its transport vehicles remain properly equipped. However, BPD needs
to take more care to ensure the accuracy of all documentation—specifically, the correlation between the

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 56
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 58 of 164
cover memo for the weekly inspection sheet and the inspection sheet itself. At times, while the cover
memo has identified no vehicles requiring preventative maintenance, or one or two, the accompanying
inspection the check sheet has identified more. To avoid confusion, the cover memo should accurately
reflect the inspection sheets.

Transport Event Audits

Beginning in 2019, to move toward satisfying the transport procedures provisions of the Consent Decree
(CD 226-37), BPD’s Inspection Unit began a pilot program for auditing transport events. The pilot
consisted of a smaller number of audits than the Consent Decree requires. The pilot helped BPD identify
effective workarounds for technological deficiencies that initially hampered BPD’s ability to gather all
of the information required for complete audits (described in prior reports) and otherwise helped BPD
determine its capacity to perform the total number of audits required.

Over the past year, the Inspection Unit has adjusted the scoresheet used to conduct the audits. Adjustments
have continued with the adoption of a weighted scoring system designed to give accurate evaluations of
compliance with BPD policy.

Beginning in January 2020, the Inspection Unit has been able to perform 20 event audits per month—
two from each of the nine patrol districts and two from the Warrant Apprehension Task Force—for a
total of 60 event audits per quarter. This is 15 more than the Consent Decree requires. In addition, the
Inspection Unit has been performing the required 27 random field audits, or spot checks, per quarter. All
of the audits are documented, and the reports are available for review by the Monitoring Team and DOJ.

For the most part, according to the audit reports, transport officers are doing what they are supposed to do
under Paragraphs 226-233 of the Consent Decree and BPD policy, though auditors have observed certain
deficiencies. The most common deficiency identified has been the failure to properly search prisoners
before and after they are placed into transport vehicles and before and after custody is exchanged. This
topic is covered in the curriculum for the Department-wide training on stops, searches and arrests that is
underway. The Monitoring Team will be watching closely to see if, as a result of the audits and training,
BPD’s audit scores in this area improve. The most recent audit showed some improvement.

Although the Monitoring Team has not yet begun its periodic random reviews of audited transport events
to determine the accuracy and comprehensiveness of the audits, the audits themselves, on their face,
generally seem thorough. Auditors appear to be properly reviewing transport events for compliance with
the requirements of Paragraphs 226-233 of the Consent Decree and BPD policy. However, the Monitoring
Team has observed some inconsistency in scoring methodology among audit reports. In at least some
of the reports, it seems different auditors have given different ratings for the same thing. For example,
cruisers (unlike vans) do not have transport vehicle cameras (TVCs), so for cruiser transport audit form
questions asking whether the transporting officer marked on the TVC when the prisoner exited/entered
the transport area and whether the prisoner made any evidentiary statements captured on the TVC, the
answer should be “N/A” for not applicable. While most auditors are recording the proper answer, at least
a few are instead scoring those questions as “1” for “compliant.” To maintain uniformity in audit scoring
methodology, BPD must ensure that all auditors fully understand and apply the audit rating system.

Policies

Consistent with Paragraph 238, BPD completed revisions to its transport policies in August 2018. These
include the core transport policy, Policy 1114 (Persons in Police Custody), as well as Policy 825 (Transport
Vehicle Camera) and Policy 1511 (Vehicle Inspection and Maintenance).

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 57
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 59 of 164
Training

Also consistent with Paragraph 238, BPD developed and delivered Department-wide e-learning on
transport procedures and additionally developed a module on transportation of persons in custody as
part of SSA/FIP II training. As explained elsewhere, this in-service training was interrupted by the
COVID-19 pandemic but has resumed.

|CHALLENGES AHEAD
As with many areas of compliance, COVID-19 has modestly affected progress in the area of transport. In
addition to interrupting Department-wide classroom training, which would have been completed by now,
the pandemic has made transport more cumbersome. For the protection of prisoners and officers alike,
masking precautions have been implemented, and for transport vans, social distancing measures are
being taken, meaning that vans cannot hold as many prisoners and must make more runs.

BPD must finalize the weighted audit scorecard it created and now uses to gauge Consent Decree and
policy compliance for every audited transport event. The weighting system is designed to accurately
capture transport performance.

Otherwise, BPD must continue to address any compliance and compliance assessment issues that the
audits expose. For instance, based on the audits, BPD has found that it must develop a method for
auditors to verify that transport officers frequently check on prisoner welfare (e.g., by having officers
“mark” the TVC when they visually check subjects or by having them audibly record their inquiries into
subjects’ welfare on their body worn cameras).

|THE NEXT SIX MONTHS

In the next reporting period, BPD will complete the interrupted transport training for all officers. In
addition, BPD will continue to work with the Monitoring Team and DOJ to finalize the transport event
audit scorecard. Finally, as explained in the Technology section above, BPD will upgrade its TVC system.
The current system requires footage to be downloaded directly into a database from a hard drive in the
transport vehicle. This process is time-consuming and cumbersome, and it has delayed auditors’ review
of the footage. In the coming months, BPD will begin to upload footage remotely in Evidence.com, the
same system that stores BWC footage.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 58
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 60 of 164
INTERACTIONS WITH INDIVIDUALS WITH
BEHAVIORAL HEALTH DISABILITIES AND IN CRISIS
The Consent Decree reinforces BPD’s “commit[ment] to responding to individuals with behavioral
health disabilities or in crisis in a manner that respects individuals’ civil rights and contributes to
their overall health and welfare.” Paragraph 96 envisions that BPD will accomplish this goal by using
appropriate crisis response techniques. Such techniques will help prevent situations that could lead
to the unreasonable use of force, promote the utilization of the health system for individuals with
behavioral health disabilities and in crisis, and diminish the inappropriate utilization of the criminal
justice system for such individuals. Paragraph 96 thus challenges BPD not only to provide effective law
enforcement responses to events involving individuals with behavioral health disabilities and in crisis,
but to participate in the development of an effective community strategy for improving the City’s support
system for such individuals.

The Consent Decree identifies a series of requirements to accomplish these objectives. They include the
expansion of the Collaborative Planning and Implementation Committee (“CPIC”), which works with
BPD to improve crisis response (CD 104-05); an annual work plan to accomplish the requirements of the
Consent Decree (CD 96, 105); an assessment of the gaps in the City’s behavioral health system coupled
with recommendations for solutions (“Gap Analysis”) (CD 97); maintenance of a Crisis Intervention Team
(“CIT”) whose officers have primary responsibility for responding to incidents involving individuals in
crisis (CD 101-03, 110, 119); development of a Crisis Intervention Plan and CIT Officer Selection Process
to ensure the efficacy of the CIT (CD 120); appointment and training of a Crisis Intervention Team leader
(CD 115-18); training for all officers on responding to individuals with behavioral health disabilities
and in crisis, and specialized training for CIT officers and dispatch personnel (CD 106-13); revision of
policies, including a police emergency dispatch policy, for responding to incidents involving individuals
in crisis (CD 98, 114); and identification of quantitative and qualitative performance measures for the
CIT program and collection of data needed to make those assessments (CD 121-22). Over the long-term,
BPD will analyze this data and issue quarterly public reports gauging its performance in responding to
individuals in crisis.

As previously reported, BPD and the City have satisfied a number of the Consent Decree’s preliminary
requirements, including expanding CPIC membership, appointing a CIT Coordinator, completing (with
CPIC) a crisis intervention plan and a plan for selecting CIT officers, creating a form to track data on
responses to individuals in crisis, revising policies addressing crisis intervention for both officers and
dispatchers, completing and publishing the Gap Analysis, and finalizing training curricula on behavioral
health awareness and crisis intervention for both recruits and officers that exceeds the Consent Decree’s
hours requirements. Within the past eight months, BPD also has completed specialized training curricula
on behavioral health awareness and crisis intervention for dispatchers and is nearing completion of
specialized in-class training curricula for CIT officers.

Because BPD has completed core policy revisions but has yet to deliver training on behavioral
health and crisis intervention to officers and dispatchers or specialized training to CIT
officers, its compliance score in both the Behavioral Health-General and Behavioral Health-
CIT Officers categories is “3” (training phase). For the Behavioral Health-System Coordination
& Improvement category, its compliance score is “4a” (implementation phase – not assessed),
because it has completed the Gap Analysis but has not yet begun meaningful implementation
of the recommendations in the report.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 59
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 61 of 164

COMPLIANCE SCORE:
BEHAVIORAL HEALTH -
GENERAL
3 Training

COMPLIANCE SCORE:
BEHAVIORAL HEALTH -
CIT OFFICERS
3 Training

COMPLIANCE SCORE:
BEHAVIORAL HEALTH -

4A
SYSTEM COORDINATION & Implementation –
IMPROVEMENT Not Yet Assessed

|AREAS OF PROGRESS
BPD and the City have shown a serious commitment to working with community members to determine
how to strengthen behavioral health support systems and improve officers’ capacity to respond to
individuals experiencing behavioral health crises. Over the past two and a half years, work in this area
has required a significant amount of time not only from City and BPD personnel, but from community
stakeholders, advocates, individuals with lived experience, and behavioral health professionals, all of
whom are volunteering with CPIC. Because of the diligence and commitment of CPIC members, the City
and BPD have taken crucial first steps (though only first steps) toward improving the City’s behavioral
health system and providing BPD with the tools to respond more appropriately to individuals in crisis.
Progress has been slowed to a degree because of the COVID-19 pandemic. In particular, as in other areas,
certain deadlines under the Third-Year Monitoring Plan have had to be extended. But productive work
has continued nonetheless.

CPIC

The City and BPD satisfied their initial obligation under Paragraph 104 by expanding the composition of
CPIC to include relevant City and State officials, individuals with lived experience, community mental
health providers, substance use services providers, local hospitals, advocates representing a wide range
of individuals with disabilities, and committed philanthropists. A review of membership and community
representation has continued each year.

CPIC has become a vital advisory body to the City and BPD. The progress made to date—on policies,
training, data collection, identifying systemic flaws, and recommending systemic improvements—is the
result of CPIC’s diligent, collaborative efforts. The efforts of CPIC are all the more impressive because
the participation of community members is voluntary.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 60
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 62 of 164
Work Plans

Under each Monitoring Plan to date, BPD and CPIC have completed a Work Plan. The Work Plan
establishes a detailed schedule for satisfying specific Consent Decree requirements in a particular
monitoring year. Each year, the process of developing the Work Plan has become more sophisticated.
In the first two years of monitoring, the City, BPD and CPIC consistently met established deadlines.

For the current monitoring year, the Work Plan was developed in face-to-face meetings throughout
during the first part of the year. As the challenges posed by the pandemic grew, CPIC transitioned to
meeting by phone and video, ultimately completing the Work Plan that way. Through their contributions
to CPIC’s four subcommittees, Behavioral Health Services of Baltimore (“BHSB”), BPD’s Training and
Compliance Divisions, local advocacy groups such as the Baltimore National Alliance on Mental Illness
(“NAMI”) and Disability Rights of Maryland (“DRM”), and representatives of non-profit organizations,
behavioral health providers, hospital systems, the public school system, and City and State government
helped complete the Work Plan.

Consistent with the Third -Year Monitoring Plan, the Work Plan prescribes:

„ A series of measures intended to prioritize and eventually implement the recommendations set forth
in the previously approved Gap Analysis report, which identifies and proposes solutions to remedy
deficiencies in the City’s behavioral health systems and BPD’s practice concerning individuals with
behavioral health disabilities;
„ An on-going review of behavioral health-related policies previously developed by BPD with CPIC
assistance;
„ An analysis of a sample of police interactions with people with behavioral health disabilities, designed
to identify systemic barriers and solutions to the delivery of care, including, for each interaction, what
precipitated the crisis, what services could have prevented the crisis, how police became involved, how
the response to the crisis could be improved, and what can be done to prevent the crisis in the future;
„ A review of the sources and types of data needed to provide a quarterly report on behavioral health
crisis events, intended to achieve the goals of paragraph 96 of the Consent Decree Paragraph 96,
which requires BPD to use appropriate crisis response techniques and reduce inappropriate use of the
criminal justice system for individuals with behavioral health disabilities;
„ Commencing annual in-service behavioral health training for all officers for the current year and
developing an initial plan for continuing education for next year;
„ Completion of training curriculum for police dispatchers and delivery of the training; and
„ Development of training curriculum for Crisis Intervention Team officers.

CIT Program

BPD has established a solid foundation for a Crisis Intervention Team. A CIT Coordinator was appointed
to head the team in 2018. The first Coordinator accepted another assignment, and BPD was quick to seek
community input and appoint a replacement. Under the Coordinator’s leadership, BPD prepared a Crisis
Intervention Plan to address staffing needs, as well as a CIT Officer Selection Plan. Both were approved
by the Monitoring Team. The Crisis Intervention Plan used police dispatch data to estimate the overall
number of crisis calls in each BPD district and then projected the number of CIT required for the Patrol
Division. The CIT Officer Selection Plan used the CIT Officer requirements in the Consent Decree to
create a set of qualifications and outlined a systematic selection process for CIT officers.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 61
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 63 of 164
Crisis Intervention Policies

In July 2019, with active input from CPIC, BPD finalized revisions to three policies addressing crisis
intervention for both officers and dispatchers: Policy 712 (Crisis Intervention Program), Policy 713
(Petitions for Emergency Evaluation and Voluntary Admission), and Policy 715 (Behavioral Health Crisis
Dispatch). The revisions were drafted by CPIC’s Policy Subcommittee and finalized following coordination
with City agencies (e.g., the Fire Department, which has responsibility for 911 dispatchers), collaboration
with the Monitoring Team and DOJ, and input from community members.

Behavioral Health Data

Consistent with Paragraph 122, CPIC’s Data Subcommittee revised and expanded the form BPD uses to
record interactions with individuals in crisis. The revised form is comprehensive. CPIC received input on
it from experts at local universities, local advocacy organizations, and individuals with lived experience.
The Monitoring Team approved the new form in December 2018. Since then, BPD has been working with
CPIC’s Data Subcommittee to pilot the form in the field and shepherd the form’s data requirements into
new IT systems BPD is presently developing.

CPIC’s Data Subcommittee has been evaluating available data on BPD officer encounters with individuals
in crisis in order determine which performance measures it can reliably analyze and include in the
reports required by Paragraph 122. In particular, the subcommittee has been systematically reviewing
available reports and outcome studies on BPD’s behavioral health intervention programs, including
BPD’s Crisis Response Team (CRT), Law Enforcement Assisted Diversion (LEAD) program, and now-
defunct Homeless Outreach Team (HOT).

In addition to performing this review, CPIC’s Data subcommittee is examining data on behavioral health
calls in each police district and attempting to extrapolate the call volume to the district’s population. The
results are intended to determine the number of CIT officers needed in each district.

The subcommittee is also attempting to understand BPD’s current hospital referral practices involving
individuals in crisis by examining hospital emergency room data. The goal is to determine whether there
are opportunities for diversion to community-based alternatives. Initial examination of the data suggests
a great deal of variability in emergency room referrals. The variability, however, might not be related to
the number of behavioral crisis events in surrounding neighborhoods. Speculatively, the utilization data
could suggest the importance of clarifying admissions criteria for each hospital so that patrol officers
have adequate guidance on making emergency room referrals.

When BPD conducted new behavioral health training in FY2018 and FY2019, it added a module on
behavioral health resources that highlighted the Baltimore Crisis Response Inc.’s crisis information and
referral (CIR) line and also directed officers to a resource app on their phones. The new training was
delivered not only to recruit classes, but to officers during in-service classes. CPIC’s Data subcommittee
reported that, as compared to the four years prior to the training, officer calls to the CIR hotline in the
period since the training began have increased significantly. This strongly suggests that BPD officers are
open to and interested in the availability of community resources.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 62
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 64 of 164
Behavioral Health Awareness (“BHA”) Training Curricula

Recruit Training | The BHA recruit training curriculum, approved in September 2019, is currently
being delivered to recruit classes. It consists of thirteen modules taught over 24 hours. Topics include: an
introduction to BPD’s crisis strategies, an exercise in understanding auditory hallucinations, an overview
of mental illness, an introduction to BPD new behavioral health policies, an introduction to working
with individuals in suicidal crisis, understanding working with youth, understanding neurocognitive
impairments (dementia), a personal story of mental illness from the National Alliance on Mental Illness,
an introduction to neurodevelopmental disorders (intellectual and developmental disabilities), substance
use disorders, de-escalation strategies, emergency evaluations/ voluntary admissions, and incident-based
scenarios. See ECF No. 247 for a complete description of the curriculum.

By requiring 24 hours of instruction for recruits, CPIC and BPD chose to include instruction that exceeds
the 16 hours required in the Consent Decree. See ECF No. 2-2, ¶ 112.b. Increasing the hours of BHA
instruction required of new recruits reflects BPD’s commitment to ensuring that its officers have a solid
understanding of behavioral health and are more adequately equipped to interact with individuals with
behavioral health disabilities.

In-service BHA Training for Officers | The Monitoring Team approved in-service BHA training
curriculum for all officers in November 2019. See ECF No. 262. To accommodate the virtual training
necessitated by the pandemic, the curriculum had to be modified. The revised curriculum is presently
being delivered remotely Department-wide, with scheduled completion in March 2021. The curriculum
consists of six modules. These modules cover a range of topics, including understanding mental health,
an overview of substance abuse, working with individuals in suicidal crisis, voluntary admissions and
emergency petitions, an in-depth examination and discussion of BPD behavioral crisis events based on
body-worn camera footage, and a comprehensive overview of BPD’s revised policies relating to interactions
with individuals with behavioral health disabilities and in crisis. In modifying and adapting the approved
curriculum to a remote learning environment, BPD worked closely with BHSB, community experts, the
Monitoring Team and DOJ to create an experience that mirrored the success of earlier training. 

The training began several weeks ago. Initial participant feedback has been very encouraging.

911 and Dispatch Training | The training curriculum for 911 call-takers and dispatchers was approved
in May 2020. See ECF No. 314. Training will take place in the fourth quarter of 2020. Developed in
coordination with the Fire Department, it will require approximately six weeks to provide classroom
instruction to all 911 call-takers and dispatchers. The training consists of nine sections taught over eight
hours. The training includes a history of mental health treatment; a review of BPD’s Crisis Intervention
Team Program; an exercise in understanding stigma in behavioral health, a review of trauma-informed
practice; an overview of behavioral health based on realistic scenarios; a panel discussion of the peer and
family perspective on behavioral health; engaging and de-escalation over the phone; self-care for 911
specialists and dispatchers; and a comprehensive discussion of BPD’s behavioral health crisis dispatch
policy.

As with other new training programs, the 911 call-taker and dispatcher training employs adult learning
techniques, such as group discussion, realistic videos, applied learning exercises, and role-playing
scenarios. The instruction is being delivered by teams of behavioral health professionals, experienced
police instructors, advocates, and individuals with lived experience.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 63
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 65 of 164
CIT Officer Training | CPIC and BPD are presently working to develop specialized training for CIT
officers, with input from the Monitoring Team and DOJ. Barring delays caused by the pandemic, the
Training Academy plans to conduct this training in the first part of 2021.

Gap Analysis and Recommendations

The Gap Analysis, approved in December 2019, provides an essential guide to achieving the Consent
Decree’s complementary goals of bolstering Baltimore’s behavioral health system, diminishing reliance
on BPD officers to handle calls for service involving individuals in crisis, and ameliorating the interactions
that BPD officers continue to have with individuals in crisis. The Gap Analysis is available here.

The City, BPD, BHSB and CPIC worked with a nationally respected research firm, Health Services
Research Institute (“HSRI”) to conduct the Gap Analysis. The report identifies unmet needs in the City’s
behavioral health system and makes over 40 specific recommendations to address those needs. Those
recommendations focus on the establishment of a fully integrated, systemic approach to behavioral health
care. Among the features of the system are:

„ 24/7, around-the-clock crisis centers

„ a vastly increased number of mobile crisis teams
„ a centralized mechanism for managing crisis response
„ an oversight committee (including CPIC) to promote a “No Wrong Door” approach to behavioral health
support that integrates mental health and substance abuse services (including outpatient services in
primary health care settings) and workforce programs
„ fortified workforce recruitment, retention and competency for behavioral health specialists, especially
in community-based settings
„ financial support for peer-run organizations
„ a wide-range of improved housing programs, including subsidized housing and permanent supportive
housing programs
„ collection and analysis of BPD data to determine why trained crisis intervention officers are not
responding to a greater number of behavioral health calls and
„ improved training for BPD officers, including training on the availability of community-based
alternatives to law enforcement action (which is currently underway).

The Gap Analysis observes that among the many benefits of a more complete, more integrated behavioral
health system will be a reduction in the need for BPD to serve as the primary—often the only—first
responder available for individuals in crisis.

|CHALLENGES AHEAD
System Coordination and Improvement

The Consent Decree’s requirement for a Gap Analysis (CD 97) is not yet fully satisfied. To meet the
requirement, BPD must evaluate a meaningful sample of officer encounters with individuals in crisis.
With an eye toward identifying solutions to the deficiencies observed in the Gap Analysis, the evaluation

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 64
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 66 of 164
of each encounter should include what precipitated the crisis, what services could have prevented the
crisis, how police became involved, and what could be done to prevent a similar crisis in the future.
Because of BPD’s data collection deficiencies, BPD was not able to include such an evaluation in the Gap
Analysis report. In the coming year, BPD and CPIC will develop a process for gathering the required data
and performing the evaluation.

The Gap Analysis calls for far-reaching change. Implementing its recommendations will be a long-term,
intensive enterprise. Establishing an integrated system of community-based care will require cooperative
planning across multiple City and state agencies and non-profit institutions. The funding required to
develop this system will be significant and will need to draw on multiple sources. Additionally, because
of Baltimore’s history of “siloing” programs within the behavioral health system, there remains the
risk that new programs will continue to be “siloed” and will not be incorporated into an integrated,
cohesive system. As Judge Bredar explained at a public hearing in July 2020, and as the Monitoring
Team explained earlier on Facebook (see Summary of Monitoring Team Activities section below), the
early July 2020 police-involved shooting of an armed man experiencing a behavioral health crisis in
Northeast Baltimore concretely exposed the flaws the Gap Analysis identified, as well as the urgent need
to implement its recommendations.

In the short run, to capitalize on the momentum the Gap Analysis has created, the City and BPD must
move expeditiously to shore up the resources they devote to behavioral health issues:

„ The City’s primary behavioral health care partner, BHSB, has limited funding, cannot always capably
implement new programs or strong mechanisms for quality control, and although possessing highly
qualified personnel in some areas, suffers from a lack of capacity and technical skill in other areas.
The pandemic has strained its resources further. BHSB needs greater support from the City to carry
out its mission and, at the same time, the City must demand greater accountability from BHSB.
„ Like BPD, which only recently has established consistent leadership on CPIC, the City must provide
leadership on CPIC. The City’s Law Department has provided strong support for the behavioral health
requirements of the Consent Decree, especially as to BPD, but leadership from other City agencies
is also required to address the comprehensive needs identified in the Gap Analysis. These include
City agencies responsible for health, housing, homeless outreach, youth support, criminal justice, and
innovation.
„ BPD needs to get its revised CIT program off the ground. Newly appointed CIT leadership must
stabilize CIT, continue to actively support CPIC and other community stakeholders, and select enough
qualified CIT officers to meet Departmental needs.
Despite these major challenges, Baltimore has a track record of innovative programming and a wealth of
professional talent. The City also enjoys a reputation for conducting incisive studies involving behavioral
health and developing service models that appear to hold promise, even if they have not endured. The fact
that HSRI reviewed 38 different analytical reports regarding behavioral health care delivery in the City
is a testament to the work that already has been done here. The challenge will be to make the response
to the Gap Analysis more effective and more sustainable than the responses to the other studies. The lack
of success of previous efforts to reform the City’s behavioral health system underscores the critical need
for leadership and accountability in implementing the recommendations in the Gap Analysis.

Training

BPD and the City continue to face challenges under the Consent Decree that are more immediate than
implementing the long-term recommendations in the Gap Analysis. Most prominently, BPD is scheduled

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 65
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 67 of 164
to deliver four different behavioral health training programs in the coming year—for recruits, officers,
911 call-takers/dispatchers, and CIT officer candidates. This would have taxed the Training Academy’s
resources even if there were no pandemic. With the pandemic, BPD’s capacity for conducting the required
training is stretched even further.

Much of the behavioral health curriculum relies on interpersonal skill development and connecting the
officers to treatment providers and those with lived experiences. It is difficult to establish meaningful
relationships remotely. Nevertheless, BPD’s Training Academy leadership, CPIC’s Training Subcommittee
and BPD’s Compliance Division have worked hard to create a productive remote learning environment.
They are experimenting with remote learning for recruits and are revising other BHA curriculum to adapt
to the realities of the pandemic. This has not been easy. BPD was informed that it would not be permitted
to adapt the previously approved “Mental Health First Aid for Public Safety” (MHFA) curriculum to
remote learning and also may not use certified community instructors remotely. As a result, BPD has
drawn from other approved curricula to develop the BHA training necessary for all officers to meet
the State of Maryland’s annual education requirements. If necessary, BPD will also consider adapting
approved curricula for other trainings being held in abeyance.

Data collection

As previously reported, BPD has not historically collected and maintained all the data that the Consent
Decree requires to analyze officer interactions with individuals in crisis. While BPD has developed a new
crisis intervention data form that should capture the required information, the Monitoring Team and
BPD itself will not be able to comprehensively assess compliance with Consent Decree requirements until
BPD officers routinely use the forms and BPD is technologically capable of maintaining and analyzing
data from the forms in a modernized record management system.

|THE NEXT SIX MONTHS

In the next reporting period, as noted, BPD will continue conducting BHA training for recruits and
initiate training for veteran officers, 911 call specialists, and dispatchers. CPIC and BPD will finalize the
training curriculum for CIT officers. BPD and CPIC will also conduct their on-going review of policies
affecting individuals with behavioral health disabilities; an analysis of a sample of police interactions
with individuals with behavioral health disabilities, required to complete the Gap Analysis; and an
assessment of the sources and types of data needed for quarterly reports on crisis events, which are
required by the Consent Decree.

The City, BPD, BHSB and CPIC members also will continue their work to implement the recommendations
in the Gap Analysis. Last fall, Disability Rights Maryland (a CPIC member) worked with BHSB to conduct
a two-day workshop to begin the work. Presenters from other cities spoke about their own successful
efforts to establish integrated behavioral health systems, while local advocates and individuals with
lived experienced shared their ideas and voiced their concerns about the City’s ability to carry through
on the promises of the Gap Analysis. Other advocacy groups such as NAMI, the Bazelon Center, the
NAACP Legal Defense Fund and over 20 additional agencies and organizations also participated in the
workshop and have joined with DRM to support the recommendations in the Gap Analysis. Adding to the
momentum for reform created by the workshop and the Gap Analysis, the State is working on providing
additional reimbursements to the hospital system to strengthen crisis intervention supports state-wide,
including in Baltimore. The time is thus ripe for decisive action to fortify Baltimore’s behavioral health
system, lest the Gap Analysis become just another report on the shelf.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 66
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 68 of 164
FIRST AMENDMENT-PROTECTED ACTIVITIES
As the Consent Decree and BPD’s revised policy on First Amendment Protected Activity explain, the
exercise of First Amendment rights is fundamental to democratic governance because it promotes the free
exchange of ideas. The preservation and protection of First Amendment rights is also vital to maintaining
public trust in the rule of law because it fosters transparency and accountability in government functions,
including policing (CD 239).

For these reasons, the Consent Decree requires BPD to protect several different First Amendment rights:
the right to free speech and expression, which includes the right to criticize law enforcement and engage
in speech in the presence of law enforcement without being subject to retaliation (CD 240-44); the right
to freely organize and participate in lawful public assemblies (CD 245); and the right to observe and
record the actions of BPD officers in the public discharge of their duties (CD 247). The Consent Decree
also protects First Amendment rights by prohibiting the warrantless search and seizure of recorded video
and images, except in limited circumstances (CD 249-50). The Consent Decree prescribes protection for
all of these constitutional rights by obligating BPD to revise its policies and training programs (CD 239,
244, 246, 251); require supervisory approval for dispersing assemblies, seizing recording devices and
recordings, and arresting individuals engaged in expressive activity (CD 252-54); and conduct annual
assessments of its practices relating to First Amendment-protected activity (CD 255).

BPD fully and effectively revised its core First Amendment policies in 2018. However, its
compliance score remains at “2” (policy/planning phase) because, due in part to coronavirus-
induced delays, it has not completed revisions to standard operating procedures for its Mobile
Field Force, which responds to public assemblies when required. BPD is also still in the process
of finalizing e-learning curriculum for training on its core First Amendment policies, so it has
only begun the training phase of compliance. Nonetheless, BPD has initiated self-assessments
of its responses to First Amendment-protected activities, having published its inaugural
assessment this year.

COMPLIANCE SCORE:
FIRST AMENDMENT-
PROTECTED ACTIVITIES
2 Policies/
Planning

|AREAS OF PROGRESS
Policy Revisions

In 2018, BPD completed comprehensive revisions to its two policies addressing First Amendment-
protected activities—the core policy, Policy 804 (First Amendment Protected Activity) and the policy
regarding recording of police activity, Policy 1016 (Public Observation and Recording of Officers). See
ECF No. 140.

Revised Policy 804 is now much more tightly organized, comprehensible, and directive than earlier
versions. It expressly delineates what officers are required to do to protect First Amendment rights
and what they are prohibited from doing. It also clearly sets forth Departmental obligations for
training and data collection and analysis on First Amendment events. Moreover, rather than focusing

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 67
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 69 of 164
exclusively on policing public assemblies, as the prior version did, Policy 804 now provides directives on
respecting individuals’ right to criticize police officers and police activity outside the context of a protest
or assembly. This is vital, as DOJ’s pattern-or-practice findings principally criticized how BPD officers
detained, arrested and used unreasonable force against individuals who criticized or insulted officers—
not how BPD officers handled public demonstrations. Finally, revised Policy 804 provides clearer and
more comprehensive direction on preserving the right to public assembly, including express protocols for
preparing for planned assemblies.

Policy 1016, which addresses the right to observe and record officers in the line of duty, is similarly
improved. Although Policy 1016 did not require as many substantive changes as Policy 804, the revised
policy provides more comprehensive guidance to officers and is better organized. The revised policy includes
detailed procedures for obtaining recorded material thought to be relevant to criminal investigations,
clearly delineating what officers and supervisors may and may not do to secure such evidence.

In the past year, BPD has begun revising standard operating procedures (“SOP”) for its Mobile Field
Force. The Mobile Field Force consists of over 200 specially trained officers from across the Department.
The Mobile Field Force is called upon when needed to monitor, maintain order, and protect the right to
assemble and speak at both planned and spontaneous public demonstrations. If demonstrations turn
violent, the Mobile Field Force intervenes to preserve public safety while respecting First Amendment
rights.

BPD currently has a policy (Policy 413) that generally addresses its Mobile Field Force. However,
during the first monitoring year, when BPD was revising its First Amendment-related policies, BPD,
DOJ and the Monitoring Team agreed that it would be more fitting for Policy 804 to incorporate the
general provisions of Policy 413, and for a comprehensive set of SOPs to replace Policy 413 and provide
detailed tactical guidance to Mobile Field Force officers about how to carry out their responsibilities while
preserving First Amendment rights. BPD, the Monitoring Team, and DOJ determined that rescission of
Policy 413 and replacement with SOPs made sense because the policy is not so much a policy as it is an
outline for a tactical procedural manual. Policy 413 will not be rescinded until the SOPs are finalized and
implemented.

In 2019, BPD began work on the SOPs. BPD produced an initial draft modeled after SOPs for well-
functioning civil disturbance units in other jurisdictions, including the District of Columbia. After
receiving feedback from the Monitoring Team and DOJ, BPD agreed to revise the draft so that it more
closely tracks the Mobile Field Force SOPs for the Maryland State Police, which has adopted a more
progressive, European model of mobile field force deployment. Neighboring jurisdictions with which
BPD often partners to monitor large crowds are undertaking a similar review of their mobile field force
operations.

Under the original Third-Year Monitoring Plan, BPD was scheduled to have nearly completed the SOP
revisions by now. The coronavirus pandemic interrupted those plans. As a critical part of the revision
process, BPD officials were supposed to observe and take part in State Police Mobile Field Force training
this spring. The training had to be cancelled and rescheduled because of the pandemic. Additionally,
because of the pandemic, officers tasked with revising the SOPs have had to spend increased time deployed
to operations. BPD officials will attend the next scheduled State Police training. Once that occurs, BPD
will be fully prepared to revise the SOPs and, accordingly, satisfy the remaining policy requirement for
First Amendment protected activities.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 68
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 70 of 164
Self-Evaluation

In February 2020, BPD submitted its first annual audit of the Department’s responses to First
Amendment-protected activities (CD 255). The purpose of these audits is to implement “corrective action
or improvement measures” where deficiencies are identified (CD 256). BPD’s report is available here.

The challenge in devising a means to audit BPD responses to First Amendment conduct is to ensure that
the instrument evaluates not only BPD’s responses to larger, planned assemblies and internal affairs
complaints based on alleged First Amendment violations, but also the responses of officers to individuals
who exercise their First Amendment rights in the course of routine police encounters by criticizing
police conduct, including the officer’s actions. Evaluating BPD’s responses to such routine exercises of
First Amendment rights is critical because those encounters are far more frequent than either larger
assemblies or encounters that spawn internal affairs complaints. Further, as mentioned above, DOJ’s
investigation found a pattern or practice of First Amendment violations based in large part on improper
retaliatory action, including bringing unwarranted criminal charges or using unreasonable force.

To capture BPD’s responses to these more routine exercises of First Amendment rights, BPD, the
Monitoring Team and DOJ agreed that BPD would use its Performance Standards Section to review
body-worn camera footage of events that involve allegations of disorderly conduct. While disorderly
conduct cases obviously do not encompass all situations in which officers might be confronted with
First Amendment conduct, and not all disorderly conduct cases implicate First Amendment rights, the
parties concluded that, at least initially, such cases might prove useful for exploring whether officers are
retaliating against individuals for exercising their rights.

Based on this agreement, BPD’s first annual audit report (see ECF No. 287-1) evaluates (1) cases involving
internal affairs complaints based on alleged First Amendment violations for 2019, (2) BPD’s response to
two larger First Amendment events in 2019—one involving a protest of the establishment of a university
police force at Johns Hopkins, and another involving a speech given in Baltimore by Donald Trump,
and (3) 2018 cases involving disorderly conduct arrests that were determined to have implicated First
Amendment conduct (a number of disorderly conduct cases, of course, did not). The audit determined
that:

„ Under the Public Integrity Bureau’s revised complaint classification system, adopted in December
2018, there was one First Amendment-related complaint lodged in 2018 and five in 2019. Because
the Public Integrity Bureau had not fully resolved any of the complaints by the time the audit was
completed, and because prior complaints alleging First Amendment violations were not reliably
classified as such under the old system, the audit did not include any evaluation of recent BPD actions
that prompted complaints of First Amendment violations.
„ Based on review of a statistically representative sample of body-worn camera footage, BPD officers
respected protestors’ First Amendment rights to peaceably assemble, speak, and record law
enforcement activity and fully complied with Departmental policy in responding to the large-scale
protests at Johns Hopkins and President Trump’s speech.
„ BPD arrest rates for disorderly conduct complaints in 2018 were extremely low (.08%), which
potentially suggests general compliance with the First Amendment (even though it is hard to know
without reviewing a sample of those non-arrest cases, which BPD will do in future audits, and even
though not all disorderly conduct events implicate First Amendment rights, as noted above). But of
the 30 disorderly conduct arrests reviewed, three (10%) involved First Amendment violations and
were thus referred for disciplinary investigation. Two of those cases appeared to involve arrests in
response to criticism of police actions, and one appeared to involve an arrest in response to video
recording of police actions.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 69
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 71 of 164
Additionally, although not reflective of First Amendment compliance, a substantial number of the other
27 reviewed cases appeared to involve other policy violations, including arrests likely made without
probable cause, statements of probable cause missing critical factual information, including information
supporting probable cause, and arrests that unnecessarily drained Departmental resources (e.g., an
arrest for spitting on a sidewalk) because the incidents should have been resolved short of arrest with a
citation or a warning, consistent with BPD’s policy on the “most effective, least intrusive response.” One
key observation was that, in a number of probable cause statements, the arresting officer supported the
disorderly conduct charge by describing the gathering of people, but failed to explain how the defendant’s
behavior (as opposed to, e.g., the presence of police or pre-incident socializing in the location) caused the
gathering, which is a key element of disturbing the peace.

Of the 28 reviewed cases resulting in adult charges for disorderly conduct, only two resulted in convictions.
One resulted in release without charges for failure to satisfy the probable cause standard, 21 were
dismissed for unexplained reasons, and four were placed on the stet docket.

The Monitoring Team is encouraged by the work BPD did in its fledgling audit. The audit presents a
fair, if necessarily incomplete, picture of BPD’s response to First Amendment conduct. It identifies the
additional information required to present a more complete picture. It isolates certain cases that warrant
disciplinary or supervisory review. It reinforces the need to address previously-known shortcomings in
First Amendment compliance—e.g., incomplete understanding of the law of disorderly conduct (at least
among some officers) and inadequate documentation of facts supporting probable cause that an individual
has committed a crime and has not simply engaged in constitutionally protected activity. It highlights
previously-known shortcomings in other areas—e.g., unacceptable delays in resolving internal affairs
complaints, low-level offenses that should be addressed with warnings or citations rather than arrests,
and either the absence of probable cause or shoddy documentation of probable cause for arrests for low-
level offenses.

Based on these observed shortcomings, the audit specifies areas of emphasis for training. Further, the
review of large-scale protest activity reveals how a solid Incident Command Structure and detailed,
properly executed operational plans can simultaneously safeguard First Amendment rights, promote
constitutional policing, and ensure public safety.

In short, the audit is a promising start. It demonstrates that BPD is capable of engaging in internal analysis
and self-correcting behavior, both of which are essential to becoming a high-functioning, Constitution-
abiding police agency. That said, with assistance from the Monitoring Team and DOJ, BPD must continue
to try to find a more accurate way to identify police encounters implicating First Amendment protections,
if possible. While disorderly conduct cases provide a useful set of such encounters, they are an incomplete
set.

Response to Recent Widespread Protests against Police Brutality

In Baltimore, as in many other cities, there have been widespread protests against police misconduct in
the wake of the brutal killing of George Floyd by Minneapolis police officers. The Monitoring Team has
observed first-hand BPD’s response to the protests. On eight different days and nights between May 29
and June 13—all the days and nights that featured significant protest activity in that period—Monitoring
Team members stationed themselves in both BPD’s command center and on the ground. Those in the
command center were able to review BPD’s incident actions plans and observe how BPD commanders,
using an established Incident Command System, directed officers on the street to respond to the protests.
From live video feeds broadcast throughout the command center, Monitoring Team members also were
able to observe how BPD officers responded to the directives they received and otherwise carried out their

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 70
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 72 of 164
duties. At the same time, members of the Monitoring Team’s community engagement arm were on the
street, interacting with protestors and obtaining a ground-level view of BPD’s response.

Thus far, based on its dozens of hours of observation, the Monitoring Team believes BPD’s response to the
protests has complied with both the Constitution and the Consent Decree. BPD has allowed community
members to assemble in public spaces and march on public streets and sidewalks without intervention;
deliver speeches denouncing law enforcement and criticize officers face-to-face without restriction or
retaliation; and make video and audio recordings of police activity without impediment. BPD’s actions
demonstrate that it has learned important lessons from certain failures in performance in 2015, when it
reacted to the unrest that gripped the City in the wake of Freddie Gray’s death. The Monitoring Team
attributes BPD’s accomplishments to date to several developments.

First, the community members who have led the protests have been well-organized and committed to non-
violence, and they have shared information needed for BPD to simultaneously protect First Amendment
rights and preserve public safety. Violence interrupters have appeared at demonstrations to prevent
violence before it occurs and to intervene when it does. For example, during early demonstrations, on
two occasions when protest participants hurled items (including small exploding fireworks) at police
officers, Monitoring Team members observed peaceful protestors identify the participants, detain them,
and take them to the police. Protest organizers’ devotion to non-violence undoubtedly has made BPD’s
management of the protests less fraught.

Second, through the visible accommodations it has made on the street, BPD has clearly conveyed to
community members that it intends to honor protestors’ First Amendment rights. BPD generally has
avoided engaging in provocative conduct. It has not made unnecessary (even if technically lawful)
arrests. Further, with limited exceptions for certain officers ringing City Hall and a small line of officers
ordered to block off a street on which County Police were effectuating a murder arrest the night of
May 31, BPD officers have policed the demonstrations in their regular uniforms. They have generally
eschewed riot gear, which is often unnecessarily provocative, as shown in other cities with mass protests.
Outfitted Mobile Field Force units have remained ready to respond if violence erupts, but they have kept
away from the demonstrations, attracting no attention, in order to minimize the potential for conflict.
Similarly, some officers deployed around City buildings where demonstrators have gathered, including
BPD headquarters and City Hall, have kept helmets and shields nearby if the need to use them arises,
but have assiduously sought to avoid using them.

To further defuse tensions, BPD officers have often calmly engaged in discussion with protestors,
occasionally marched with them, and sometimes taken a knee alongside them. Such disarming conduct
has contributed to the peacefulness of the demonstrations. Ideally, it will bolster BPD’s ongoing efforts
to rebuild community trust.

Third, on each day of protest, BPD has utilized a sound Incident Command System, typically headed by
the Chief of Patrol, and has meticulously prepared incident action plans to address different contingencies
and assign dozens of officers to specific duties. Especially because BPD has coordinated with and led other
agencies in managing the protests, a strictly observed and executed ICS has eliminated confusion on the
street and produced measured, proportionate and effective police action. Commanders have emphasized
the prudence of allowing protestors to peacefully occupy public spaces and march through the streets
without interference, even when assembling and marching has caused temporary traffic disruptions.
Commanders have similarly emphasized the need for restraint in the face of direct verbal provocation
and have deftly handled potentially volatile events. For instance, prior to the main protest activities on
May 31, officers checked locations along anticipated march routes and identified and removed palates of
bricks and plastic bottles containing accelerant, which individuals with plans for property destruction

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 71
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 73 of 164
appeared to have placed strategically near storefronts. To take another example, with the help of violence
interrupters and under the ICS, officers located an individual standing in the middle of the street with a
long gun slung over his back and capably ushered him away from the crowd to an isolated location where
his detention, which was lawful, would not risk a disturbance.

It bears noting that BPD has proven capable of monitoring the protests without compromising police
functions in other parts of the City. Each week, through their participation in BPD Comstat meetings,
Monitoring Team members have observed how BPD has simultaneously managed the protests while
continue to deploy necessary personnel and resources to each police district.

The overall propriety and effectiveness of BPD’s response to recent, large-scale protest activities does
not mean that BPD has reached compliance with the Consent Decree’s provisions for responding to mass
demonstrations. The conclusion here is “so far, so good.” There is still a way to go—sustained appropriate
responses to demonstrations, Department-wide training on First Amendment conduct, completion of
the Mobile Field Force SOPs, new Mobile Field Force training, and more. Nevertheless, the Monitoring
Team is cautiously optimistic that BPD’s recent performance does mean that the Department is turning
a corner. Many of BPD’s big-city peers have not managed demonstrations and contained violence and
property destruction as effectively, and some of them have resorted to disproportionate, sometimes injury-
inflicting force and mass arrests. BPD has not only performed well when measured against these other
agencies; it has performed well when measured against the pertinent standards for measurement—the
Constitution and the Consent Decree.

|CHALLENGES AHEAD
As previously reported, the biggest challenge facing BPD remains ensuring that its officers respect First
Amendment rights not during mass demonstrations, but in their daily interactions with City residents. In
its investigation, DOJ found that BPD officers were engaged in a pattern or practice of violating the First
Amendment by responding to oral criticisms, insults and non-violent provocations with unreasonable
force. Similar conduct, at least in isolated pockets, has continued into the Consent Decree period.

Within the past two years, two different officers, Officer Arthur Williams and Sergeant Ethan Newberg,
have been charged with crimes for using unreasonable force and bringing criminal charges in retaliation
for the exercise of basic free speech rights. The Monitoring Team has reviewed BWC footage and incident
reports from these cases, and it appears that these officers in fact responded to non-violent, constitutionally
protected provocations with force, criminal charges, or both. Officer Williams was convicted of the charges
against him in 2019. Sergeant Newberg has not yet been tried. But the case against him, which alleges
nine different instances of unlawful retaliatory conduct over the span of just ten months (July 2018 –
May 2019), is particularly troubling, because as an experienced first-line supervisor with 24 years on the
job, he has been setting an example for potentially dozens of patrol officers.

BPD’s commendable response to the recent protests suggests that systemic improvement may be afoot.
The protests include speakers who condemn the police and demonstrators who directly, and sometimes
aggressively, criticize officers face-to-face. To the Monitoring Team’s knowledge, none of these scores of
encounters have provoked unlawful retaliation by way of either force or arrest. By repeatedly exercising
restraint in the face of oral provocations, and by receiving consistent reinforcement from commanders about
the need for such restraint, hundreds of officers should be developing a reflexive habit of accommodating
free speech rights. If that is in fact what is happening, officers’ experiences with the recent demonstrations,
together with lessons learned from the Williams and Newberg prosecutions, could set BPD on a path to
sustained compliance with the Consent Decree’s core First Amendment requirements.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 72
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 74 of 164
|THE NEXT SIX MONTHS
BPD must develop and deliver a training curriculum to ensure that officers respect First Amendment
protections. Policies 804 and 1016 were revised nearly two years ago, but because of the immense burdens
placed on BPD’s training academy, recently exacerbated by the Covid-19 pandemic, officers have not
yet received training on them. As a result, they are not yet formally active. To enable them to become
effective as soon as possible, BPD is finalizing an e-learning curriculum that will cover all essential
policy provisions. All officers will be required to complete the e-learning within the next two months. The
e-learning will be supplemented with in-class instruction next year.

In addition to delivering e-learning, BPD will finalize revisions to its Standard Operating Procedures for
the Mobile Field Force.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 73
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 75 of 164
INTERACTIONS WITH YOUTH AND COORDINATION
WITH BALTIMORE SCHOOL POLICE
The Consent Decree requires BPD to alter its approach to how it interacts with youth. The Consent
Decree obligates BPD officers to account for the personal characteristics (age, size, developmental/mental
status, disability status and maturity) of youth they encounter and, where practical, use alternatives to
arrest (e.g. warn and release, counseling, referral to community services and resources, warnings, civil
citations) in order to divert youth from criminal justice system (CD218). To accomplish this goal, the
Consent Decree requires the City to conduct a comprehensive assessment of its effort to reduce youth
involvement in the juvenile and criminal justice systems (“Youth Assessment”) (CD219). It requires BPD
to revise its policies and training as needed, and conduct training in order to properly guide officers in
their interactions with youth (CD220-21). The Consent Decree envisions that, in preparing the Youth
Assessment, the City will obtain input from a collaborative consisting of City Officials, BPD representatives
and community stakeholders, including community organizations with experts in the field, academic and
youth advocates (CD 219).

The Consent Decree also contains several provisions addressing BPD’s relationship with the Baltimore
School Police (“BSP”). In particular, Paragraph 417 of the Consent Decree requires BPD to conduct an
initial assessment of its memorandum of understanding (“MOU”) with BSP and evaluate how BSP has
used BPD’s authorization to exercise law enforcement powers throughout the City. The assessment should
include an analysis of data reflecting the frequency with which BSP officers respond to calls, make stops,
searches, and arrests, and use force under the MOU. BPD will use the assessment to identify deficiencies
and opportunities for improvement, amend the MOU as needed, implement other appropriate corrective
action, and document the changes it makes. Following the initial assessment and amendment of the
MOU, BPD will conduct a biennial evaluation of its coordination with BSP, and make any modification
needed to ensure effective coordination with BSP.

Thus far, consistent with Monitoring Plan requirements, BPD has completed the Youth
Assessment and finished revising some youth-related policies while continuing to work on
others. Because BPD remains in the policy revision phase of reform, its compliance score in the
Youth Interactions compliance category is “2” (policy/planning phase).

COMPLIANCE SCORE:

2
YOUTH INTERACTIONS Policies/
Planning

In the School Police compliance category, BPD has completed the required assessment of its
MOU with BSP and executed a new MOU that attempts to fix the shortcomings in the prior
MOU. What remains in this category are biennial assessments of BPD’s coordination with
BSP, the first of which is approximately 18 months from now. BPD’s compliance score is “4a”
(implementation – not yet assessed).

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 74
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 76 of 164

COMPLIANCE SCORE:

4A
SCHOOL POLICE Implementation –
Not Yet Assessed

|AREAS OF PROGRESS
Youth Diversion Assessment

In April 2019, with input from the City’s Youth Assessment Advisory Board, other community
stakeholders, the Monitoring Team and DOJ, BPD published the Youth Diversion Assessment. The
assessment, prepared by the Center for Children’s Law and Policy, presents a comprehensive evaluation
of diversion opportunities for Baltimore youth. Relying on meticulous data analysis and interviews with
law enforcement personnel, community advocates, and a number of youth, the Assessment included data
on arrests and juvenile court processing of youth, barriers to diversion at different stages in the juvenile
justice process, and thoughtful, achievable recommendations for reducing the number of youth entering
each stage of the juvenile justice system while at the same time preserving public safety. The assessment
is available here.

Policies Affecting Youth

Under the Third-Year Monitoring Plan, BPD began implementing a three-phase process to revise
policies affecting interactions with youth. The first phase, now completed, entailed developing Policy
1207 on youth interrogations and revising previously approved use of force policies to account for youth
interactions. The second phase, currently near completion, involves revising Policy 1202, the core policy
on youth interactions. The revisions include BPD’s process for pre-arrest diversion. The third phase will
require BPD to work with the Mayor’s Office and City agencies to develop a broader diversion program
that will provide officers with meaningful options for diversion to community-based providers.

As previously reported, BPD finalized Policy 1207, the new youth interrogations policy. This was a
significant accomplishment, as the policy, for the first time, places formal limitations on interrogating
individuals under age 18. Under Policy 1207: all youth must have a parent/legal guardian and/or an
attorney present during an interrogation, with a preference for having both a parent/legal guardian and
an attorney present; all youth 15 and under must consult with an attorney prior being advised of their
Miranda rights; officers may not use deception when questioning youth; and BPD’s advice and waiver
of rights form has been modified for youth suspects so that it informs youth of their rights in more
comprehensible language.

In addition to issuing Policy 1207, BPD completed revisions to Policy 1115, its previously approved core
policy on use of force, to account for youth interactions. The revisions prohibit using pain compliance and
control techniques on youth unless deadly force is threatened. They further prohibit using force against
youth who are restrained. BPD’s review also resulted in changes to Policy 719 (Conducted Electrical
Weapon), Policy 1111, (Batons), and Policy 1118 (Oleoresin Capsicum (OC) Spray). BPD officers are now
restricted from using these instruments on youth who are “preteen or younger.”

This year, BPD has been working in close collaboration with youth advocacy organizations, the Monitoring
Team and DOJ on phase two of its youth policy reforms, Policy 1202, titled “Youth Interactions.” This
policy will provide BPD officers general guidance on interactions with youth, including a reshaped process

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 75
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 77 of 164
for pre-arrest diversion, and will establish the framework for completing the third phase of youth policy
reforms, which will establish a broader diversion program to prevent youth from unnecessarily entering
the criminal or juvenile justice systems. A draft of Policy 1202 will be issued for public comment in the
fourth quarter of 2020. The policy should be finalized in the first quarter of 2021.

Evaluating and Revising BPD’s MOU with Baltimore School Police

In April 2019, BPD published its assessment of its MOU with BSP. The assessment examine how BSP
was exercising law enforcement powers off of school property under the MOU (CD 417). The report on
the assessment can be found here.

The assessment reaffirmed that BSP was not adequately recording and maintaining the data required for
evaluation purposes, including data on the frequency with which BSP officers exercised law enforcement
powers under the MOU, as well as data on calls for service, incidents, stops, arrests and uses of force
involving BSP officers acting under the MOU. The assessment recommended addressing these gaps in
data collection in a new MOU. The assessment also recommended that the new MOU contain clear
guidance to BSP officers on reporting uses of force off school property, as well as clear guidance to BPD
officers on taking and investigating complaints involving actions of BSP officers off of school property.

BPD and BSP finalized a new MOU on March 31, 2020. The new MOU incorporates the recommendations
from the Youth Assessment for improved data collection and use of force reporting and satisfies the
Consent Decree requirement to incorporate “policies and protocols governing the proper investigation
of civilian complaints involving BSP officers exercising law enforcement powers pursuant to the MOU.”
(CD 418).

|CHALLENGES AHEAD
BPD must complete phases two and three of the youth policy reform process. Because of the coronavirus,
work on Policy 1202 has been delayed and the deadlines in the Third-Year Monitoring Plan have been
extended. Nonetheless, BPD, the Monitoring Team, DOJ and community stakeholders continue to refine
the draft policy.

Once Policy 1202 is completed, BPD and the City will need to work diligently to establish a comprehensive
youth diversion program, which should incorporate the recommendations from the Youth Assessment.
Coupled with Policy 1202, which will establish criteria for mandatory or presumptive diversion for
certain offenses, the youth diversion program will advance the Consent Decree’s principal goal of
providing youth with community-based alternatives to arrest when appropriate.

BPD also must train officers on its new youth policies when they are completed. Certain aspects of these
policies are already being trained. For instance, BPD has incorporated key provisions of Policy 1207
(Youth Interrogations) into the “Interrogations and Interviews” module of its ongoing Department-wide
training on stops, searches and arrests, and also has included limitations on on-scene interviews of youth
in the “Crime Scenes and Witnesses” module. Nevertheless, comprehensive training on youth policies
will be required. The Third-Year Monitoring Plan calls for BPD to begin to develop such training in the
coming months, with delivery in 2021. Given the recent, extended closure of the Training Academy, the
temporary redeployment of certain Academy personnel to address staffing shortages, and the temporary
delay in finalizing Policy 1202, it is unlikely that BPD will be able to comply with this timeline. It is
nonetheless anticipated that BPD will be able to deliver the training in 2021.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 76
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 78 of 164
|THE NEXT SIX MONTHS
In the next reporting period, BPD will finalize Policy 1202 on youth interactions, begin drafting training
curriculum on youth interactions and, together with the City, start in on the development the youth
diversion program.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 77
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 79 of 164
COMMUNITY POLICING AND ENGAGEMENT
COMMUNITY POLICING
Community policing is a philosophy that promotes organizational strategies that support the systematic
use of partnerships and problem-solving techniques to proactively address the immediate conditions that
give rise to public safety issues such as crime, social disorder, and fear of crime. (Community Policing
Defined, (Washington, DC Office of Community Oriented Policing Services, 2014.)

Effective community policing depends on the trust and cooperation of community members. In many of
Baltimore’s most challenged neighborhoods, mistrust in the police runs high and cooperation with the
police is infrequent. BPD thus faces considerable challenges in engaging in effective community policing
in some of the City’s most underserved communities.

One of the overarching goals of the Consent Decree is to meet those challenges. The Consent Decree
contains specific requirements intended that will affect the way BPD officers interact with community
members when taking law enforcement action. These requirements—regarding, e.g., use of force; stops,
searches and arrests; fair and impartial policing; First Amendment protected activities; and interactions
with youth and individuals with behavioral health disabilities or in crisis—are addressed in the preceding
sections of this report. The Consent Decree begins, however, with certain broad requirements intended to
promote both community policing and community engagement.

Preliminarily, the Consent Decree requires issuance of a new mission statement that integrates
community-oriented principles into BPD “management, policies and procedures, recruitment, training,
personnel evaluations resource deployment, tactics and accountability systems” (CD15). BPD finalized a
new mission statement in July 2018. See ECF No. 119, and then revised that statement after Commissioner
Harrison was appointed. See ECF No. 239.

In addition to a new mission statement, the Consent Decree outlines the kind of community policing
training BPD officers must receive (CD16-17), as well as the data BPD should collect (CD18). The Consent
Decree also requires the City and BPD to develop community engagement plans (CD19), to obtain input
from community groups on policies, practices, training, engagement programs and enforcement strategies
(CD20), to develop a community outreach program to educate and communicate with City residents about
the Consent Decree (CD21), to publish annual reports on BPD’s community policing efforts (CD22), and
to use the results of community surveys to inform policies, training and practices (CD25).

To date, beyond revising its mission statement, BPD has developed a Community Policing Plan.
It has begun drafting a community policing training curriculum but has not yet conducted
training. Accordingly, its compliance score in the Community Policing category is “3” (training
phase).

COMPLIANCE SCORE:

3
COMMUNITY POLICING
Training

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 78
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 80 of 164
|AREAS OF PROGRESS
Community Policing Plan

A concrete, comprehensive Community Policing Plan is critical to BPD’s reform efforts, as defines how
BPD provides police services to Baltimore’s diverse communities. Frequent changes in leadership during
the first year of monitoring delayed development of the Community Policing Plan. There were further
delays after the appointment of Commissioner Harrison, who needed time to ensure that the Plan reflected
his vision for satisfying the community policing requirements of the Consent Decree. After receiving
Monitoring Team, DOJ and community feedback on several drafts, BPD finalized and published the
Community Policing Plan in April 2020. See ECF No. 304. The Plan, available here, describes BPD’s
strategy for addressing crime while proactively engaging the community in a manner that builds trust
and legitimacy. It prescribes organizational redesign, building community partnerships, problem-solving,
building analytical capacity, and district crime plans. Among other things, the Plan establishes a goal for
all patrol officers to devote 40% of their time on activity other than calls for service, including pro-active
community policing. The International Association of the Chiefs of Police asked BPD to write an article
on the Plan as a model for community policing. See the article here.

|CHALLENGES AHEAD
Training

The COVID-19 pandemic has delayed development and delivery of training on community policing. Under
the revised training schedule, the curriculum will be developed in the coming months with community
input, and finalized by the end of the year, with classroom instruction set to begin in spring 2021. As
explained in other sections of this report, this training, on top of the other courses being developed, is
going to test the Training Academy’s capacity.

Staffing

The Staffing section above describes the difficulty of adopting the community policing strategies prescribed
in the Community Policing Plan. To reach the goal of having patrol officers spend 40% of their time
on proactive problem-solving and community policing activity, BPD will need to add over 170 patrol
officers and over 30 sergeants. Given these estimates, staffing shortages will hamper and delay the full
implementation of the community policing plan. It will also affect the ability of District Commanders to
establish continuity of assignment by having the same officers work in the same area every day.

Earning the Community’s Trust

One of the keys to establishing an effective community policing model is building community trust.
The Consent Decree explicitly recognizes this imperative, as earning the community’s trust is one of its
central, overarching goals. The results of the initial round of surveys the Monitoring Team conducted in
2019—a community survey and a custodial arrestee survey—demonstrate just how much BPD must do
to meet that goal. The surveys were designed to measure community attitudes toward BPD. The surveys
will be replicated in future years to determine whether those attitudes are changing.

Community Survey | The Monitoring Team partnered with the Institute for Urban Research at
Morgan State University (“IUR”) to devise, conduct, analyze the results of, and prepare a report on the
community survey. IUR completed data collection for the community survey at the end of June 2019.
IUR interviewed 645 individuals from across Baltimore. It used several different methods for recruiting

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 79
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 81 of 164
participants. The results reflect, through narrative reporting, what a large and diverse sample of people
in Baltimore think about BPD and policing generally. Among IUR’s findings:

„ A majority of participants based their opinions on perceptions, not personal experience, as a majority
rarely or never encounter BPD officers.
„ Satisfaction with and trust in BPD are low, although participants want to improve relationships with
BPD, expressed a willingness to engage with BPD, and feel comfortable communicating with BPD
when necessary.
„ A majority of participants do not believe that BPD effectively reduces crime and keeps people safe.
„ A majority of participants do not believe BPD has a good working relationship with the community, do
not observe BPD taking the time to meet members of the community/neighborhood, do not personally
know the names or faces of BPD officers, and get nervous when they see a BPD officer.
„ A majority of participants do not believe the respect between BPD officers and community members
is reciprocal, maintaining that BPD officers do not show respect toward community members, while
community members treat BPD officers respectfully.
„ A majority of participants have observed BPD officers engage in racial profiling, use excessive force,
or use verbally abusive language.
„ A majority of participants do not believe BPD officers are held accountable for misconduct.
The full report, titled “The Community’s Experiences and Perceptions of the Baltimore City Police
Department,” can be found here. It was published earlier this year.

Custodial Arrestee Survey | Researchers from the University of Toronto’s Munk School of Global
Affairs and Public Policy and Rose Street Community Center, a local organization that assists returning
citizens, conducted the custodial arrestee survey and published a report of its findings at the beginning
of September 2019. Based on the researchers’ interviews of 70 individuals in Baltimore’s central booking
facility shortly after their arrests, the report found that “most detainees we interviewed judge policing on
the basis of their experiences before arrest [rather] than during it.” Consistent with the findings of the
community survey, the researchers determined that over half of the interviewees thought BPD was doing
either a “bad” or “terrible” job, while fewer than 20% thought BPD was doing a “good” or “excellent” job,
and roughly 25% said BPD’s work was neither good nor bad. Importantly, according to the researchers,
this largely negative appraisal was not the result of prejudice against police officers or bias against
policing in general. Rather, “much of [the interviewees’] dismay about policing in Baltimore today stems
from a sense that the police ‘don’t care’ about their community, despite the persistence of social problems
they believe the police can help fix.” The full report, titled “Experiences and Perceptions of the Police in
Baltimore,” can be found here.

COMMUNITY ENGAGEMENT
Consistent with Paragraphs 19-21, BPD has sought to engage community members in the reform process.
BPD’s Consent Decree Implementation Unit has organized or attended numerous community events,
often with sworn BPD personnel in attendance; has disseminated draft policies, training curricula and
reform plans for community input, which BPD has incorporated; and has published annual reports
describing BPD’s community engagement efforts. But re-establishing community trust in BPD will be a
years-long process, one that is currently being hindered by a global pandemic. Whether BPD’s community
engagement work helps to build bridges to City residents, particularly those most distrustful of BPD,
remains to be seen.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 80
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 82 of 164
Over the past 30 months, among other things, BPD’s Consent Decree Implementation Unit and other
BPD personnel have:

„ Complied with the requirement to conduct two community briefings on the Consent Decree in each
police district each year, and held City-wide workshops about the Consent Decree
„ Established a Community Training Review Committee to participate in training curriculum piloting
and provide feedback
„ Visited dozens of community fairs, community association meetings, and community partner
collaborations
„ Conducted officer focus groups, one at MICA and one at Morgan State, and 12 officer/civilian focus
groups with Police Foundation assistance
„ Attended and made presentations at roll calls in each District
„ Completed a community engagement inventory with Police Foundation assistance through interviews
of District command staff and other personnel who engage in community outreach
„ Participated in meetings with Civilian Oversight Task Force membership
„ Canvassed neighborhoods to distribute information on reforms and upcoming Consent Decree
meetings,
„ Utilized social media, email, online surveys, and media outlets to share and collect information about
reform efforts.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 81
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 83 of 164
SEXUAL ASSAULT INVESTIGATIONS
The Consent Decree requires BPD to enhance the trust of victims of sexual assault, to strengthen its
response to and investigations of reports of sexual assault, and to combat gender bias (CD 257). To
achieve these goals, the Consent Decree requires BPD to revise the policies and procedures for responding
to and investigating reports of sexual assault (CD 258); provide initial and on-going annual training to
support the revised policies and procedures (CD 259); ensure through proper supervision and internal
oversight that reports of sexual assaults are thoroughly investigated (CD 260, 262, 263); ensure that
officers transport victims to a medical facility for a forensic exam in all instances in which a forensic
exam is warranted and the victim consents (CD 261); enhance its collection, analysis and reporting of
data regarding the nature and extent of sexual assault crimes (CD 264); and share information about
its sexual assault investigations with other law enforcement agencies, the public, and the Sex Offense
Unit (CD 265). The City and BPD will ensure that their policies and protocols with the Sexual Assault
Response Team (SART) enable them to engage in periodic reviews of services provided by BPD and to
review samples of open cases and those classified as unfounded (CD 266).

As previously reported, BPD satisfied the threshold requirement to revise both its policy on sexual assault
investigations (Policy 708) and its standard operating procedure on such investigations. See ECF No. 152.
Since then, BPD made minor, technical revisions to the policy to ensure consistency with other policies.
See ECF No. 266. BPD also successfully created and finalized a new policy on member-involved sexual
misconduct. See ECF No. 269. Additionally, BPD developed and completed e-learning training for all
officers on responding to reports of sexual assault, see ECF Nos. 242 & 267, and has devoted considerable
time and effort to preparing a draft training curriculum for Sex Offense Unit (“SOU”) investigators, as
well as a draft virtual classroom curriculum for all officers to complement the completed e-learning.
Finally, BPD has produced two annual self-reports on sexual assault investigations, which summarize
investigations data from 2018 and 2019, respectively. See ECF No. 235 & 317. As explained below, the
2019 report shows some progress.

However, much work is left to be done. Once training for SOU investigators and virtual classroom training
for all other officers is completed, which should occur by early 2021, BPD will need to demonstrate
adherence to Policy 708 and its requirements for victim-centered, trauma-informed responses to reports
of sexual assault. Tracking compliance with Policy 708 will require reviews and assessments of officer
responses based on reports, body worn camera footage, and investigator case files. Tracking compliance
with Policy 708’s provisions on data collection and analysis will be dependent on the effective utilization
of a new record management system, which is still in development.

In view of the work done to date and the work that remains, BPD has made reasonable progress
toward satisfying the Consent Decree’s requirements. At this point, BPD’s compliance score
in the sexual assault investigations category is “3,” as it has completed policy revisions and
remains in the training phase of reform.

COMPLIANCE SCORE:
SEXUAL ASSAULT
INVESTIGATIONS
3 Training

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 82
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 84 of 164
|AREAS OF PROGRESS
Officer Training on Policy 708 and Related SOPs

As of November 25, 2019, 2,085 sworn personnel had completed e-learning training on responding to
reports of sexual assault. The training presented officers not only with changes in procedure for responding
to such reports, but also with changes in BPD’s philosophy regarding how police officers should view and
address alleged sexual offenses, offenders and victims.

Follow-up classroom training on victim-centered, trauma-informed approaches to reports of sexual assault

is planned to begin later this calendar year and conclude in March 2021. Originally, the training was
scheduled to in-person. However, due to the limitations on in-class instruction caused by the coronavirus,
BPD, the Monitoring Team and DOJ agree that BPD may conduct the classroom training virtually.
It will be paired with virtual classroom training on responding to individuals experiencing behavioral
health crises. These subjects are sufficiently important that, given the completion of policy revisions in
both areas, BPD, the Monitoring Team and DOJ believe classroom training, even if virtual, should be
delivered as soon as practicable.

Policy on Member-Involved Sexual Misconduct

BPD’s Policy 322 (Member-Involved Sexual Misconduct) was approved and filed with the Court in
December 2019. See ECF No. 269. This policy memorializes BPD’s zero tolerance policy regarding
member-involved sexual misconduct and establishes responsibilities and guidelines for conducting both
criminal investigations of allegations of such misconduct under Policy 708 and internal administrative
investigations of such misconduct by BPD’s Public Integrity Bureau. The policy will not become effective
until personnel in the Sexual Offense Unit and Public Integrity Unit are trained on it. BPD’s intent is to
produce and deliver e-training this calendar year.

Training Curriculum on Investigations for Sexual Assault Investigators

In recent months, BPD and the Baltimore City Sexual Assault Response Team (which includes both public
officials and community advocates) have worked closely with the Monitoring Team, DOJ and experts in
sexual assault investigations to develop training curriculum for SOU Unit investigators. That curriculum
is now finalized. Consistent with Consent Decree requirements, it stresses a victim-centered, trauma-
informed approach to sexual assault investigations, with a focus on the alleged offender’s behavior. The
draft curriculum is based on the new, previously approved standard operating procedure for BPD’s Sex
Offense Unit. Training is scheduled for the week of October 12, 2020.

Reports on Sexual Assault Investigations

Paragraph 264 of the Consent Decree requires BPD to collect and analyze data regarding its sexual assault
investigations every year. The data should include: the number and nature of sex offenses reported; the
number and demographics of the alleged offenders; the number and demographics of the alleged victims;
the resolution of sex offense cases; and the processing of forensic medical exams.

In its second annual report, completed in June 2020, BPD demonstrated improvements in the amount and
type of data reported, though BPD still did not satisfy all of Paragraph 264’s requirements and the data
reported has not been independently verified. See ECF No. 317. The report, available here, contains 2019
data on Child Abuse Unit’s (CAU) sexual assault investigations, 2019 data regarding multiple victims
and multiple alleged offenders, and 2019 data on the reporting of Sexual Assault Forensic Exam (SAFE)

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 83
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 85 of 164
kit requests and processing times. BPD promises to continue to improve the accuracy, comprehensiveness
and clarity of this data. In the report, BPD states that it will begin tracking co-occurring crimes, work
with the State’s Attorney’s Office to obtain data on case outcomes, and explore developing technology to
track data on LGBTQ victims when available based on self- reporting.

Among the highlights from the 2019 report are:

„ The median processing time of SAFE kits puts BPD in-line with proposed State legislation to process
SAFE kits in under 150 days from the date of request.
„ In 2019, BPD sent 54% of the cases to the State’s Attorney’s Office for review as compared to 44% in
2018.
„ In 2019, BPD had two unfounded Uniform Crime Reporting cases as compared to 15 in 2018.

|CHALLENGES AHEAD
COVID-19 restrictions pose the greatest challenge to ensuring that BPD officers, including SOU
investigators, are fully trained on new policies and that their actions comply with Consent Decree
requirements. While patrol officers will undergo virtual, web-based training on victim-centered, trauma-
informed approaches to reports of sexual assault, these concepts can be difficult to grasp and even more
difficult to put into action. The Monitoring Team fully understands the need to complete training as soon
as practicable in the face of the pandemic but has concerns that e-learning and web-based training will
not be sufficient to accomplish the behavioral changes in responses to reports of sexual assault that the
Consent Decree envisions.

As noted, SOU investigators will receive their training in October. At that point, the Monitoring Team
must begin to assess the adequacy of BPD’s sexual assault investigations. Together with BPD and DOJ,
the Monitoring Team must develop and agree to an assessment instrument and methodology that not
only captures what an SOU investigator does, but how—and how well—they do it. There are unique
challenges to assessing whether an investigator used a victim-centered, trauma-informed, offender-
focused approach by reviewing a case file.

BPD also must confront staffing shortages in the SOU. The Staffing Plan calls for 24 SOU detectives.
Currently, there are only ten. Detectives’ caseloads will partly determine how effective they are in
investigating sexual assaults and meeting the requirements of the Consent Decree.

Data collection and analysis challenges also persist. Until now, BPD has never sought to collect the data
necessary for the analyses required by the Consent Decree. Although BPD showed some improvement
in its June 2020 report, its current technology limitations, including its inadequate case management
system, continue to prevent it from gathering and analyzing a significant amount of the data required by
the Consent Decree. The Monitoring Team and DOJ are conferring with BPD about how its reports can
more fully address Consent Decree requirements until the deficiencies in its case management system
are remedied.

Finally, the Monitoring Team, BPD and DOJ agree that a victim survey measuring satisfaction with
BPD’s responses to and investigations of sexual assaults would be helpful. Developing a valid and reliable
survey instrument and conducting a survey without adding to victims’ trauma could prove difficult.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 84
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 86 of 164
|THE NEXT SIX MONTHS
Training for SOU investigators is tentatively scheduled for the week of October 12, 2020. Virtual web-
based training for other officers should be completed by March 2021.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 85
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 87 of 164
RECRUITMENT, HIRING AND RETENTION
The Consent Decree recognizes that BPD’s recruitment and hiring program, as well as its efforts to retain
officers, are in need of significant improvement. BPD’s Staffing Plan, discussed above, reinforces the
importance of these improvements—i.e., the importance of increasing the number of officers, particularly
in Patrol and PIB—to the success of the reform effort. The Consent Decree obligates BPD to: (1) develop
and implement a Recruitment Plan with “clear goals, objectives and action steps for attracting and
retaining a quality work force that reflects the diversity of the Baltimore Community” (CD 420-22);
(2) review and reform its hiring processes (CD 423-25); (3) develop and implement a Retention Plan to
“identify challenges and recommend solutions to improve BPD’s retention of employees” (CD 426); and
(4) routinely assess its recruitment, hiring, and retention practices (CD 427).

BPD has satisfied the threshold requirements of the Consent Decree—a hiring report, a
Recruitment Plan, and a Retention Plan. It is now in the implementation phase, e.g., refining
standard operating procedures for its Recruitment Section, adopting a new interview process
that includes community members on interview panels, and adopting the National Testing
Network’s “FrontLine National” exam as the first stage in BPD’s officer selection process. BPD’s
compliance score in the Staffing, Recruitment & Retention category is “4a” (implementation –
not yet assessed).

COMPLIANCE SCORE:
STAFFING, RECRUITMENT
& RETENTION
4A Implementation –
Not Yet Assessed

|AREAS OF PROGRESS
On-Going Evaluation of Hiring SOP

In November 2018, after an in-depth review of its hiring processes, BPD finalized an initial hiring
report, as required by Paragraph 423. See ECF No. 169. At the same time, BPD adopted new standard
operating procedures for its Recruitment Section. Since then, BPD, the Monitoring Team and DOJ have
continued to engage in regularly scheduled meetings to review, assess, and discuss modification to the
SOP. Discussions and resulting modifications have addressed, but have not been limited to, background
investigation procedures. For instance, in 2019, BPD discovered that a number of potential hires who
were otherwise well qualified for employment were being prevented from advancing in the hiring process
because of their failure to register for the Selective Service by age 26. Although failure to register is a
violation of federal law, the Department of Justice and Selective Service agreed to suspend prosecutions
of non-registrants in 1988, more than 30 years ago. BPD determined that barring candidates from
employment based on a failure to register would both deprive the Department of otherwise qualified
candidates and retain a hiring requirement with potentially disparate effect. BPD has now modified the
SOP to allow candidates that have not registered previously to obtain a waiver through the selective
service “information letter” exception process, rather than automatically disqualifying them.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 86
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 88 of 164
Background Checks

Historically, delays in background checks have been a primary reason for delays in hiring. To prevent
these delays and expedite the hiring process, BPD retained an outside vendor, Kentech Consulting, to
assist with the administration of background checks. Thus far, BPD’s partnership with Kentech has
proven successful in reducing delays in the hiring process.

RecruitStat

In 2019, BPD fully implemented RecruitStat, the procedure for reviewing the Department’s progress
on recruiting, hiring and retention on a weekly basis. In the first quarter of 2020, due to changes in
Recruitment Section leadership and the redeployment of resources required by the COVID-19 pandemic,
RecruitStat was put on hold. Because RecruitStat is an effective accountability mechanism for BPD’s
staffing needs, the Monitoring Team expects the Recruitment Section to restart RecruitStat meetings
shortly.

Recruitment and Retention Plans

In 2018, indicative of its priority of improving recruitment, BPD exceeded the requirements of the First-
Year Monitoring Plan by issuing a Recruitment Plan, which was not due in Year One. Over the past
two years, as discussed above and detailed in our prior reports, BPD has been implementing the Plan
by adopting certain measures to bolster recruitment. BPD’s progress on recruitment has been aided by
its partnership with the Mayor’s Office on Innovation, which has helped to identify the root causes of
inadequate staffing and to establish strategies for addressing the shortage.

In the last part of 2019, BPD completed the Retention Plan required by Paragraph 426. See ECF No.
274. The process of developing the Plan included opportunities for input from both rank and file officers
and community members. The plan is available here. Among other things, it calls for the adoption of
a process that identifies reasons for officer resignations prior to pension eligibility. It also advocates a
number of incentives for experienced officers.

|CHALLENGES AHEAD
Achieving a Positive Hiring-to-Attrition Ratio

In 2019, despite its efforts to recruit, hire and retain more officers, BPD lost more officers to attrition
than it hired: 187 vs. 156. Thus far in 2020, due primarily to a large number of retirements in January,
the statistics have improved, but not as much as hoped. Attrition still outpaces hiring, though the gap
has narrowed. As of the end of August, BPD had hired 134 officers and lost 147 to attrition, for a net loss
of 13 officers. See ECF No. 338-2.

The persistence of a negative hiring-to-attrition ratio this year may be attributable to interruptions
in operations caused by a combination of two events: the pandemic, which has affected operations
Department-wide, and the protests following the killing of George Floyd, which required the weeks-
long redeployment of Recruitment Section officers to Patrol. Because of both the temporary reduction in
Recruitment Section personnel and pandemic-induced safety precautions, interviewing and processing
of applicants (who often must travel to complete parts of the background investigation process) had
to be postponed.  This created a ripple effect, delaying processing of additional applicants. Further,
during the delays, BPD may have lost candidates, who may have selected another police agency, grown
disinterested, or found non-police employment.   

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 87
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 89 of 164
The Recruitment Section is again fully staffed, but officers will periodically work Patrol to make up for
budget cuts in overtime. There is also great concern that another surge in coronavirus infections within
BPD will severely hamper recruitment due not only to another possible reduction in Recruitment Section
personnel, but also to the reduced capacity of external partners, such as the National Testing Network,
which administers the hiring exam.

The Recruitment Section also has identified a structural obstacle to recruitment: a funding shortfall for
recruiting out-of-state candidates. While BPD is properly focused on recruiting home-grown candidates,
it must also recruit from other regions in order to meet hiring goals. The funding shortfall puts BPD at
a competitive disadvantage. For instance, the Washington, DC Metropolitan Police Department offers to
relocate new hires to temporary dormitory-style housing and provides a housing stipend afterward. The
Recruitment Section often hears from applicants—particularly younger, college-educated applicants with
student loans—that they require such assistance.

Despite these challenges, the Recruitment Section continues to work diligently to recruit and hire a
diverse cross-section of candidates, representative of the community BPD serves. For instance:

„ New Section leadership has focused on assembling the right team and better management practices.
„ The Section has adopted a triage methodology, which prioritizes likely successful candidates:
applications from candidates without criminal records, compromised work histories, or other likely
disqualifiers are segregated and closely monitored, resulting in expedited hiring times, while
candidates with more challenging backgrounds are advised of ways to improve their chances so that,
if qualified, they are provided equal opportunity.
„ The Section has identified a need to evaluate the efficacy of its recruitment strategies—to determine
what exactly is working best so that it can allocate its resources appropriately. Such self-evaluation
is crucial right now, given the imperative to hire new officers and BPD’s resource and budgetary
constraints. BPD’s ability to perform such self-evaluation will be contingent on the hiring of a
recruitment analyst, which is currently on hold due the civilian hiring freeze.
„ A new Section strategy prioritizes outreach to other public safety partners, like fire departments. This
strategy has resulted in a recent hiring bump. Section leadership believes there is a significant benefit
to focusing on this pool of potential applicants, as they tend to be civic-minded, have already been
subjected to government background checks, and understand what it means to work in an organization
similar to a police agency.
„ The Section has adapted to the times by updating recruitment methods. Virtual job fairs, college
messaging boards, candidate “mentoring” through e-mail dialog, and simple improvements in customer
service have become the order of the day.
These measures likely explain why the hiring-to-attrition ratio is improving this year, even in the face
of the daunting challenges of the past six months. Obviously, the numbers must improve far more
dramatically if BPD is to meet its staffing goals. Yet there is at least some evidence that a turnaround is
underway. Before drawing even tentative conclusions, the Monitoring Team awaits the year-end totals.

Retention

The overarching goal of BPD’s Retention Plan is simple and direct: to retain an engaged, productive, and
healthy workforce. To accomplish this goal, BPD must affirmatively support the desire of its members
to complete their careers, or at least of majority of their careers, with BPD. Consistent with the main
takeaways from recent officer focus groups, such affirmative support broadly requires the following:

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 88
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 90 of 164
„ Addressing, through planning and policy, the staffing and deployment issues that affect officer morale.
„ Adhering to a fair, transparent promotional process and committing to the development of officer
leadership skills.
„ Implementing burden-reducing practices, updating technology and equipment (especially patrol cars),
and improving the physical work environment.
„ As described elsewhere in this report, BPD has begun implementing measures to achieve each of
these objectives, as required by the Consent Decree.

|THE NEXT SIX MONTHS

BPD will continue to seek to make the hiring process more efficient, implement the Retention Plan, and
use RecruitStat to track progress and identify ways to improve its recruitment, hiring and retention
practices.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 89
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 91 of 164
OFFICER ASSISTANCE AND SUPPORT
Under the Consent Decree, BPD must adopt several important measures to support the health and
wellness of its officers. The Consent Decree requires BPD to: provide, review and revise, as needed, an
Employee Assistance Plan (“EAP”) that furnishes no- or low-cost counseling and mental health wellness
services (CD 436-437); develop peer support services (CD 438); offer all officers a voluntary mental health
evaluation before returning to duty after a traumatic incident (CD 439); develop well-being protocols
to be utilized during officer deployments in periods of civil unrest (CD 440); and develop protocols for
annually assessing the efficacy of all of BPD’s officer assistance programs (CD 441).

In the two and a half years of monitoring, BPD refined its EAP (CD 436-437), its traumatic and high-
stress incident protocols (CD 439-440)., and its peer support program policy (CD 438). Further, BPD’s
Officer Safety and Wellness Section (“OSW”) has actively offered support services to BPD officers. BPD is
in the process of developing a methodology for measuring the efficacy of its support programs as part of
the Third-Year Monitoring Plan (CD 441).

Having completed the policy revisions required by the Consent Decree BPD’s compliance score
in the Officer Assistance and Support category is “3” (training), because BPD is in the process
of developing Ethical Policing Is Courageous Training, which is an officer wellness program.
Apart from the development of EPIC, however, BPD has progressed into the implementation
phase in this category, as its reinvigorated officer wellness services and programs are
operational.

COMPLIANCE SCORE:
OFFICER ASSISTANCE
AND SUPPORT
3 Training

|AREAS OF PROGRESS
Policies

In 2018, BPD completed required revisions to its EAP policy, Policy 1703. See ECF No. 151. Since then,
OSW has conducted outreach to educate officers about the EAP, including by posting information in each
district and on social media, sending materials to supervisors, and speaking at roll-call trainings and
“pop up” health fairs.

In 2018, BPD also finalized revisions to its traumatic and high-stress incident protocols, Policy 1731. See
ECF No. 144. These are wellness protocols for (1) “all officers returning to duty following a traumatic
incident (e.g. serious injury, officer-involved shooting, officer-involved accident involving fatality, or all
other uses of force resulting in death or serious injury)” (CD 439), and (2) “officer deployments during
public demonstrations or civil unrest” (CD 440). BPD has trained responsible personnel on the protocols
through e-learning.

In November 2019, BPD completed revisions to its peer support program. See ECF No. 258. BPD now
has a number of officers trained by the International Critical Incident Stress Foundation to respond
to critical and traumatic incidents as members of BPD’s Peer Support Team. The Peer Support Team
provides emotional support to officers and their families and are on-call to respond to incidents as needed.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 90
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 92 of 164
Early Intervention, Support and Guidance

Over the course of 2019, OSW completed 56 early intervention sessions and 217 support and guidance
sessions. For the first half of 2020, OSW completed 18 early intervention sessions and 507 support and
guidance sessions.

BPD utilizes Behavioral Health Services Baltimore (BHSB) to assist in furnishing mental health
services to BPD officers and their family members. BHSB is a central participant in the Collaborative
Planning Implementation Committee, discussed in the Interactions with Individuals with Behavioral
Health Disabilities section above. BHSB assists OSW in furnishing critical incident stress debriefings
to officers involved in shootings and other critical incidents, consistent with Policy 1731. OSW, BHSB,
or peer support program personnel provide follow up support. BHSB also assists OSW in providing field
counseling to officers and in hosting health fair “pop-ups.” According to BHSB, 6.5% of BPD employees
now utilize EAP services, which is 20% higher than the national average.

To reduce its reliance on BHSB in responding to critical incidents, OSW is now developing in-house
capacity. OSW will certify several of its staff in Learning Exercise Design for critical incident response
through the Federal Emergency Management Institute.

As the Monitoring team and DOJ collaborate with BPD to develop an assessment methodology for officer
wellness initiatives, OSW has undertaken an online survey of BPD officers. The survey helps to identify
the strengths of OSW programs, as well as any weaknesses. As of July 30, 2020, 1136 officers had
responded. The results are encouraging. Over 95% of respondents indicate that they know BPD has an
officer wellness program. Additionally, 18% of respondents report having utilized BPD’s officer wellness
resources—a higher percentage than reported earlier by BHSB. A full discussion of these survey results
will be available when the final assessment report is published in the fourth quarter of 2020.

Officer Wellness Programming

In 2019, OSW conducted weekly “Managing Stress in the Workplace” presentations reaching over 2,100
sworn members. Additionally, approximately 100 civilian employees were provided stress reduction
training during several “lunch and learn” sessions. OSW has continued to make these presentations in
2020.

The OSW survey indicates that these presentations are needed, as the results indicate that officers
experience stress from several sources, including supervision, and witnessing or experiencing physical
altercations or injury in the line of duty.

OSW conducted 10 pop-up health fairs in 2019. It has conducted additional fairs in 2020. These workplace
events present BPD members with resources related to physical fitness, mental health, and financial well-
being. The resources made available during health fairs include nutritional services, meal preparation
services, physical therapy, fitness training, acupuncture, meditation, yoga, sound therapy, and financial
planning.

OSW is now adjusting certain wellness programming to accommodate the restrictions on in-person
interaction necessitated by the pandemic. For starters, OSW plans to hold online video conferences (via
Zoom) to answer officers’ questions about available resources. It also plans to develop videos that contain
interviews with members who have had positive experiences with the health and wellness team. These
videos will be distributed through BPD communications channels, including PowerDMS and district
television.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 91
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 93 of 164
Continued in-person programming will include presentations at new employee orientations, which are
intended to reinforce the idea that participation in officer wellness and safety initiatives should be a
routine part of every officer’s career. OSW also intends to extend its services to BPD family members,
recognizing that family members are the first line of support.

The Monitoring Team has been impressed with OSW’s accomplishments over the past two years. We
are equally impressed that OSW continues to strive for improvement. It is likely that OSW’s efforts to
provide wellness services to officers and to educate officers about the availability of those services explain
why BPD members focus on self-care at higher rates than the national average for law enforcement
personnel.2

|CHALLENGES AHEAD
Making Operational Changes to Improve Morale

On behalf of the Monitoring Team, the Crime and Justice Institute conducted a series of eight focus groups
of BPD personnel from May 21 – 23, 2019. The focus groups were intended to gauge officer opinions on
a range of topics, including how BPD is falling short in providing them the direction and support they
need to do their jobs effectively and what BPD can do to improve. The eight groups consisted of (1)
two patrol officer groups of mixed race and ethnicity, but gender specific—one male, one female—to
allow for comparison by gender, (2) three patrol officer groups, separated by race/ethnicity—one African
American, one White, one Latinx—in order to make comparisons across race/ethnicity, and (3) detective,
sergeant, and lieutenant groups of mixed race/ethnicity and gender. Officers from every shift and every
district were represented. In total, a total of 68 sworn personnel—40 patrol officers, nine detectives, ten
sergeants, and nine lieutenants—participated.

The Monitoring Team published CJI’s final report on the officer focus groups at the end of July 2019.
Among the report’s key findings:

„ Officers believe a majority of residents supports and trusts them, though support varies significantly
by district, and individuals most likely to interact with the police—both offenders and victims—are
less trustful and supportive.
„ Officers are unclear about BPD’s expectations around community policing and, because they spend
so much time responding to calls for service, they have little or no time for pro-active policing. Officers
favor a return to post assignments, which give them a chance to engage with and get to know the
residents of one neighborhood.
„ Command staff and supervisors have not adequately communicated Consent Decree objectives and
requirements to rank-and-file officers. More clarity and better communication are needed. A frequently
cited example is the lack of clarity around the use of force reporting policy—specifically, which actions
constitute a reportable “Level 1” (minor) use of force.

2 Kuhns, Joseph B., Edward R. Maguire, and Nancy R. Leach. 2015. Health, Safety, and Wellness Program Case
Studies in Law Enforcement. Washington, DC: Office of Community Oriented Policing Services. See https://cops.usdoj.gov/
RIC/Publications/cops-p332-pub.pdf

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 92
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 94 of 164
„ Morale is low due to deficiencies in leadership, supervision and staffing. Rank-and-file officers feel
disconnected from command staff, and the lack of a permanent Commissioner for a lengthy period
resulted in the absence of clear and consistent direction. Officers believe that commanders are
more concerned about BPD’s image than their welfare. Officers also maintain the Department is
understaffed: they are often forced to work overtime (“drafted”), are generally overextended, and
have no time for pro-active or community policing. According to focus group participants, BPD could
improve morale by more consistently recognizing superior performance; effectively communicating
changes and expectations down the chain of command; improving equipment (especially patrol cars),
technology and facilities; improving the efficiency of the internal affairs process, which takes far too
long; and addressing staffing issues by, e.g., eliminating certain specialized units and redeploying
the assigned officers to Patrol, hiring civilians for administrative positions, and terminating certain
deployment strategies that take officers off the street.
„ Participants believe that, due to staffing problems, officers are often promoted too quickly. They also
believe that, historically, promotions and transfers have been based on “who you know,” rather than
on “how well you do.”
The full report, titled “Feedback from the Field: A Summary of Focus Groups with Baltimore Police
Officers,” can be found here.

Recently, BPD has taken measures to address some of the concerns expressed in the focus groups. It has
developed an objective, uniform, transparent system for promotions for the ranks of Captain and Major;
created and distributed a “pocket guide” and e-learning material with key Consent Decree requirements,
including the requirements for reportable Level 1 uses of force; opened a new, modern training facility,
upgraded facilities in certain districts, and developed a plan for upgrading its aging vehicle fleet; produced
a staffing plan that envisions increasing the number of officers in the patrol division not only by bolstering
recruitment, but by civilianizing certain functions currently performed by sworn members and reducing
the number of specialized units; adopted a community policing plan that will call for officers to spend
40% of their time on pro-active, community-oriented policing and 60% responding to calls for service; and
implemented a new policy for prompt negotiated resolution of minor rule violations, which promises to
free up investigators to concentrate or more serious violations and thus expedite the disciplinary process
in all cases.

EPIC

As explained elsewhere, BPD has initiated EPIC, a program to train officers to be “active bystanders,”
to intervene with fellow officers to prevent and address misconduct. EPIC is, at root, an officer wellness
program, designed to keep officers from getting into situations that could adversely affect their physical
and mental health. The successful implementation of EPIC could reap substantial benefits for officer
wellness.

Self-Assessments

As required by Paragraph 441, BPD must conduct regular assessments of the effectiveness of its officer
wellness services. It is presently collaborating with the Monitoring Team and DOJ to develop an
appropriate methodology. Meaningful self-assessments are needed to identify and correct deficiencies in
OSW services and improve the quality of OSW performance.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 93
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 95 of 164
Expanding the Use of Wellness Services and Practices

To maximize its effectiveness, OSW would like to proactively identify officers who are in need of wellness
services and either do not recognize it or are reluctant to acknowledge it. Historically, a high number of
officers receive services only after their health problems become critical.

When BPD’s new record management system becomes operational, OSW hopes to utilize it to examine
various data potentially indicative of health problems and conduct affirmative outreach to officers
identified in the data.

Additionally, OSW is seeking to develop ways to identify officers through peer referral, which would
enable colleagues to alert OSW confidentially about potential issues. Referrals could be initiated as a
result of anecdotal comments in the workplace or observations of an officer’s conduct. OSW hopes to
identify officers via confidential referral through EPIC, which, by educating officers about the value of
accountability, trains them to detect potential physical or behavioral health issues in their peers. OSW
also hopes BPD will retain in-house mental health professionals to facilitate early detection. Offering
confidentiality, in-house providers could attend roll calls to raise awareness, deploy in BPD’s health and
wellness vehicle to make direct contacts, and field requests from a confidential referral app to provide
treatment.

In addition to proactively identifying officers who would benefit from wellness services, OSW seeks to
incorporate wellness practices into BPD’s standard operating procedures. Many of the OSW’s challenges
persist not because it lacks resources, but because commanders and first-line supervisors do not perceive
the services it provides as essential to fighting crime. As a result, OSW is often called in late, after a
member is already experiencing the after-effects of a traumatic incident. BPD should integrate wellness
practices, such as work breaks and nutrition, into the deployment procedures of various units, especially
high stress units (e.g., shooting investigations).

|THE NEXT SIX MONTHS

BPD will continue to prepare a methodology for assessing the effectiveness of its officer support programs.
The methodology should be finalized in the next six months.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 94
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 96 of 164

SUMMARY OF
MONITORING TEAM
ACTIVITIES
Since its appointment in October 2017, the Monitoring Team has sought to fulfill each of its prescribed
roles under the Consent Decree—technical advisor, arbiter, and facilitator. The Monitoring Team’s work
in each role is summarized below. The details of the Monitoring Team’s work, recorded on time sheets for
each Monitoring Team member in 1/10 hour increments, are reflected in the Monitoring Team’s approved
invoices, which are available on the Monitoring Team’s website at https://www.bpdmonitor.com/
monthly-statements.

The Consent Decree provides that the Monitoring Team will be paid $1,475,000 per year in fees and
expenses. For the first 33 months of its work (October 2017 through June 2020), the City paid the
Monitoring Team $4,189,363.00 in fees and $175,572.68 in expenses. In addition, from October 2017
through June 2020, the Monitoring Team contributed pro bono services for its work on the Consent
Decree in an amount equal to $1,843,247.80, meaning that 29.7% of the Monitoring Team’s work during
the 33 months was at no cost to the City.

ENGAGEMENT WITH STAKEHOLDERS

|COMMUNITY ENGAGEMENT
From the beginning, the Monitoring Team has engaged in active, affirmative community outreach. A core
group of Team members are devoted to community engagement: lead monitor, Ken Thompson; deputy
monitors Seth Rosenthal and Chuck Ramsey; lead community liaison Ray Kelly; community engagement
coordinator Darnyle Wharton; and community engagement specialist Jessica Drake. This group meets
once every week to debrief and plan community engagement activity.

As the Consent Decree requires, the Monitoring Team has held community forums once every three
months, rotating from one part of the City to another. During the COVID-19 pandemic, the Monitoring
Team held its community forum virtually on Facebook Live.

In addition to holding the required community forums, the Monitoring Team and its community engagement
team have been meeting with community members where they live. Monitoring Team members have
attended or convened dozens of community meetings in different parts of the City, including meetings of
neighborhood associations, faith-based organizations, civic leaders, advocacy organizations, and affinity
groups. The meetings are intended to inform community members about the Consent Decree process, to
obtain input on improving the process, and to listen to their views about BPD.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 95
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 97 of 164
The Monitoring Team hosts bimonthly Facebook Live sessions. During these sessions, community members
are given the opportunity to post questions online and obtain real-time answers from Monitoring Team
leadership. One of the last Facebook Live session, held in June 2020 during the COVID-19 pandemic
and in the wake of the killing of George Floyd in Minneapolis, received over 1,700 views. Additionally,
in the first year of monitoring, community engagement team members participated in several “corner
crawls,” canvassing neighborhoods and engaging in discussions with residents about the Consent Decree.
Although temporarily halted during the pandemic, but restarted this past summer, lead community
liaison Ray Kelly also issues a monthly newsletter called “The Monthly Monitor.” The newsletter is
emailed to the Monitoring Team’s distribution list and linked from the Monitoring Team’s Facebook
and Twitter accounts. The newsletter provides information about recent and upcoming developments
under the Consent Decree, with a focus on opportunities for community members to engage in the reform
process.

Perhaps the most essential aspect of the Monitoring Team’s community engagement effort is the
utilization of a team of neighborhood liaisons. There is one liaison in each of the City’s nine police districts.
Overseen by the team’s lead community liaison, Ray Kelly, and community engagement coordinator,
Darnyle Wharton, the neighborhood liaisons educate their neighbors about the Consent Decree and
the work of the Monitoring Team and serve as the Team’s initial points of contact for information and
opinions about the performance and conduct of BPD officers, which the Team will need to fully assess
BPD’s compliance with the Consent Decree. The neighborhood liaisons hold regular “office hours” at local
libraries and community centers, attend community meetings and events in their districts, and canvass
their neighborhoods to educate community members about the Consent Decree. The neighborhood liaison
team is one of a kind: no other consent decree monitoring team has one.

In addition to conducting affirmative, localized outreach to inform and hear from community members
about BPD and the reform process, the Monitoring Team has pursued targeted engagement with
community members around specific Consent Decree requirements, eliciting written community
input on proposed BPD policies and training programs. Under each of the annual Monitoring Plans,
the Monitoring Team built in a community feedback component. (BPD and DOJ also have their own
feedback mechanisms). As the Monitoring Team’s Third-Year Monitoring Plan submission explains: “for
each policy undergoing revision, and for each training curriculum being developed, the Third-Year Plan
gives community members two separate opportunities for feedback—the first after BPD collaborates
with the Monitoring Team and DOJ and produces an initial draft, the second after BPD addresses and
incorporates feedback from that initial Comment Period and produces a final draft.” ECF No. 290 at 15.

Over the past thirty months, the Monitoring Team, BPD and DOJ sought and received meaningful
public comment on dozens of policies covering every area of the Consent Decree, including use of force;
stops, searches and arrests; First Amendment-protected activities; misconduct investigations; youth
interactions; responding to reports of sexual assault; and interactions with individuals with behavioral
health disabilities. Community members similarly provided input on draft curricula for training on use
of force; stops, searches and arrests; fair and impartial policing; behavioral health awareness and crisis
intervention; responses to reports of sexual assault; and body-worn camera use. Community members
also have provided feedback on draft reports and plans for improving City services and BPD operations,
including a Community Policing Plan; a “Gap Analysis” identifying the deficiencies in the City’s behavioral
health systems; an assessment of BPD’s memorandum of understanding with Baltimore School Police; a
Technology Resource Plan; and Staffing, Hiring and Retention Plans.

To elicit community feedback on these deliverables, the Monitoring Team posted and received comments
in response to surveys on its website, received detailed letters and e-mails from community members and
organizations, and welcomed more informal oral feedback from community members. The Monitoring

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 96
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 98 of 164
Team shared whatever feedback it received with BPD. In turn, BPD revised each deliverable in response
to all feedback provided (that is, feedback provided to BPD, the Monitoring Team and DOJ), collaborated
with the Monitoring Team and DOJ to ensure that the revised drafts properly reflected that feedback, and
then published a final curriculum, report, plan or policy following approval by DOJ and the Monitoring
Team.

In 2019, BPD formed and began utilizing a Community Training Review Committee consisting of
community members who attend and provide feedback on in-service training programs while they are
in development. Thus far, the CTRC has participated in and provided feedback on in-service training
programs on use of force, stops, searches and arrests, and fair and impartial policing. Monitoring Team
members and DOJ representatives attended and provided feedback at subsequent pilot sessions involving
both Training Academy instructors and other officers.

|COMMUNICATION WITH THE PARTIES

Since its appointment, the Monitoring Team has communicated with BPD, the City and DOJ multiple
times on a daily basis—in in-person meetings, in conference calls, and by email. Monitoring Team members
have worked exhaustively with the parties to make sure BPD and the City produce all the deliverables the
Monitoring Plans have required. By way of example, since January 2020, even with the limitations posed
by the pandemic, the Monitoring Team and DOJ have collaborated with BPD and provided extensive oral
and written comments and written line edits on the following deliverables, among others:

„ Drafts of in-class curriculum for training on stops, searches and arrests, as well as pilot tests of that
training
„ Drafts of e-learning and in-class curriculum for training on behavioral health awareness and crisis
intervention for recruits, officers, specialized CIT officers, and 911 call-takers and dispatchers
„ Drafts of in-class curriculum for both Sex Offense Unit investigators and patrol officers on investigations
of reports of sexual assault
„ Drafts of e-learning on First Amendment-protected activities
„ Drafts of revised policies, including policies on misconduct investigations and discipline, youth
interactions, and stops, searches and arrests
„ Draft BPD reports on responses to First Amendment-protected activities, sex crimes investigations,
and arrests resulting in releases without charge; and
„ Draft community policing and staffing plans

|POLICE ENGAGEMENT
In addition to conferring daily with members of BPD’s Consent Decree Implementation Unit, City Law
Department attorneys representing BPD, and BPD command staff, the Monitoring Team engages directly
with BPD members. Monitoring Team members have established relationships with union leaders and
spent substantial time at BPD’s Training Academy (where recruit, in-service, and field training officer
training are conducted) and Public Integrity Bureau (which investigates allegations of officer misconduct).

The Monitoring Team also has established and meets periodically with an informal group of rank-and-
file officers to obtain their candid feedback on the Consent Decree, the positive attributes of BPD, and the

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 97
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 99 of 164
challenges facing BPD. Although temporarily halted because of the coronavirus pandemic, Monitoring
Team members—and Judge Bredar himself—also periodically go on ride-alongs with BPD officers who
regularly furnish their views on the Consent Decree and policing in Baltimore. Moreover, in May 2019,
as discussed below, the Monitoring Team conducted formal focus groups of BPD officers, detectives, and
supervisors to listen to their thoughts and gather their ideas about effective reform.

Soon after our appointment, the Monitoring Team established a protocol for notification and potential
response to critical incidents involving BPD officers, such as officer-involved shootings. The notification
is immediate and allows for local Monitoring Team members or out-of-town members in Baltimore to go
to the scene to observe BPD’s response. The notification protocol has been utilized every time there has
been an officer-involved shooting over the past 30 months, and a Monitoring Team member, usually lead
monitor Ken Thompson, always has responded to the scene.

|MEETINGS WITH THE COURT

The Monitoring Team’s leadership, including Ken Thompson, Seth Rosenthal, Chuck Ramsey, Hassan
Aden and Theron Bowman, communicate regularly with Judge Bredar—in person, by telephone, and by
email—to update him on developments, to advise him, and to take direction.

Early on in the reform process, Judge Bredar determined that each month he would hold a three-hour
working session with the Monitoring Team and the parties to discuss developments and challenges in a
specific area of the Consent Decree. In recent months, the meetings have covered not only BPD’s progress
on specific topics (including, e.g., sexual assault investigations, crisis intervention, and stops, searches
and arrests), but developments in what the Court and the Monitoring Team view as the indispensable
structural elements of the reform process: misconduct investigations and discipline (i.e., integrity and
accountability), training, technology, and staffing.

ASSESSMENTS AND TECHNICAL ASSISTANCE

Since the inception of the First-Year Monitoring Plan in February 2018, the Monitoring Team has
assessed BPD’s performance and assisted BPD in revising its policies in every area of the Consent
Decree; developing and providing training on the revised policies; completing studies and plans involving
technology, community policing, staffing, recruitment, retention, the City’s behavioral health system,
and BPD’s memorandum of understanding with the Baltimore School Police; and conducting reviews of
BPD’s performance in certain areas.

Because the early years of the Consent Decree have necessarily focused on implementing foundational
reforms, the Monitoring Team has spent much of its time assisting with and evaluating policy revisions,
training curriculum, and training delivery. As BPD moves toward completing training on its revised
policies, and especially as BPD completes implementation of new technology enabling more comprehensive,
more reliable data collection and analysis, the Monitoring Team will shift its focus to compliance reviews
and outcome assessments.

|POLICY REVISIONS
Drawing on their expertise and knowledge of national best practices, the Monitoring Team has assessed
and advised BPD on revisions to approximately fifty policies covering every area of the Consent Decree.
As explained in more detail in the Findings section below, those policies address:

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 98
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 100 of 164
„ Use of force
„ Stops, searches and arrests
„ Fair and impartial policing
„ Misconduct investigations and discipline
„ First Amendment-protected activities
„ Transportation of arrestees
„ Youth interactions
„ Interactions with individuals with behavioral health disabilities and in crisis
„ Responses to reports of sexual assault
„ Disclosure of exculpatory evidence in criminal cases
„ Officer wellness
„ Officer peer support

|TRAINING
As BPD has finalized revisions to its policies, the Monitoring Team has spent an increasing amount of
time observing and assessing BPD’s development and delivery of training curricula. The first tranche of
Consent Decree-mandated classroom training, which covered use of force and related aspects of fair and
impartial policing, was delivered to all sworn BPD personnel (with minor exceptions that have now been
addressed) between June and October 2019. The second tranche, which covers stops, searches and arrests
and related aspects of fair and impartial policing, began in February 2020, was interrupted for several
months because of the coronavirus, and re-started (with smaller class sizes due to physical distancing
restrictions) in mid-July 2020, with an expected completion date of December 2020. Throughout these
in-service training courses, Monitoring Team members and DOJ attorneys have provided extensive
feedback to Academy supervisors and instructors. Each observed training day is typically followed with
a debrief among Academy supervisors and instructors, Monitoring Team members, and DOJ members.

The Monitoring Team worked closely with BPD and DOJ to develop both e-learning and in-class training
curricula on use of force, stops, searches, arrests and related aspects of fair and impartial policing. Officers
had to complete e-learning and pass on-line tests on these topics before they could participate in the in-
class components. The Monitoring Team pored over drafts of the curricula, participated in extensive
working sessions to revise them, and observed and critiqued pilot testing of the in-class curricula.

The Monitoring Team also has collaborated with BPD and DOJ to develop e-learning and classroom
curricula for training on behavioral health awareness and crisis intervention and responses to sexual
assault, as well as e-learning curricula on First Amendment-protected activities and body worn camera
use.

|FOUNDATIONAL ASSESSMENTS AND REFORM PLANS

Some of the foundational work required by the Consent Decree entails assessing BPD’s present capacity
to implement reforms and, where BPD falls short, developing a plan for ensuring that those reforms are
achievable. Over the past 30 months, the Monitoring Team has reviewed, worked with BPD, the City and
DOJ to develop, and evaluated the following:

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 99
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 101 of 164
„ A technology plan, which serves as the blueprint for modernizing BPD’s IT systems so that report-
writing is less time-consuming and cumbersome for officers, supervision is more efficient and data-
driven, and rigorous analysis of Departmental trends is finally feasible.
„ A comprehensive staffing plan, which seeks to address BPD’s personnel shortcomings and needs.
„ Plans for improving BPD’s efforts to hire and retain officers
„ A community policing plan designed to ensure that BPD officers do not spend all of their time
responding to calls for service and instead spend adequate time on pro-active policing, engaging with
community members and earning their trust.
„ A detailed critical assessment, or “Gap Analysis,” of City programs devoted to assisting individuals
with behavioral health disabilities and histories of substance abuse.
„ An assessment of BPD’s memorandum of understanding with Baltimore School Police. The objective
of the assessment is to improve BPD’s coordination with BSP and to modify the MOU accordingly.

|SURVEYS
As described in detail in the Findings section, the Monitoring Team thus far has conducted an initial
community survey, an initial custodial arrestee survey, and an initial set of officer focus groups.

The Monitoring Team and the Institute for Urban Research at Morgan State University are currently
planning a second community survey. The methodology is being refined to include a true random sample
intended to replicate the demographics of Baltimore. It will continue to include a purposive component
designed to reach marginalized populations. The second survey has had to be delayed because of the
pandemic, which potentially limits in-person interviewing. The Monitoring Team, the City, BPD and
DOJ are in the process of making adjustments to the methodology to account for these limitations.

|COMPLIANCE REVIEWS AND OUTCOME ASSESSMENTS

The Consent Decree requires the Monitoring Team to conduct both compliance reviews and outcome
assessments. Compliance reviews are qualitative evaluations of BPD performance in different areas of
the Consent Decree. They are conducted with an eye toward determining how far BPD has come, and how
far it still needs to go, to achieve compliance with Consent Decree requirements (CD 454). For instance,
over time, are the quality of internal investigations improving, are uses of force increasingly well-justified
and well-reported, and are investigative stops more routinely supported by well-articulated reasonable
suspicion and arrests more routinely supported by well-articulated probable cause?

Outcome assessments, by contrast, are quantitative assessments designed to determine whether the
reforms required by the Consent Decree in each area are having a tangible, measurable impact—whether,
independent and apart from BPD’s progress toward compliance with Consent Decree requirements,
policing is changing in the real world (CD 456). For instance, are the policy revisions and training in the
area of stops, searches and arrests producing a greater percentage of investigative stops that turn up
evidence of prosecutable criminal activity and a lesser percentage of weapons pat downs and searches
that turn up no guns or contraband? Or are the policy revisions and training on use of force, which
emphasize de-escalation, leading to fewer encounters requiring more serious Level 2 and Level 3 uses of
force?

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 100
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 102 of 164
While providing technical assistance and evaluating policy revisions, training and plans for organizational
change (e.g., in staffing, supervision, and technology) make up much of the Monitoring Team’s work in
the early years of the Consent Decree, compliance reviews and outcome assessments will make up most
of the work in later years, once the foundational reforms are in place.

The Consent Decree process is now at the point when the Monitoring Team will begin focusing
increasingly on measuring progress through compliance reviews and outcome assessments. Presently,
however, measuring progress is effectively limited to qualitative evaluations of BPD performance. That is
because the manifest deficiencies in BPD’s IT systems and data collection practices substantially restrict
evaluations that rely on quantitative data. Thus, while the Monitoring Team has begun to conduct
qualitative compliance reviews in certain areas, it will be hampered in its ability to conduct many of the
outcome assessments required by paragraph 459 of the Consent Decree, as well as compliance reviews
that entail reliance on quantitative data. At this stage, due to the unreliability, incompleteness or total
absence of relevant data, most such evaluations would be of limited value and extremely difficult to
perform. The unfortunate result is that, in many areas, the Monitoring Team will not be able to establish
meaningful baselines against which to measure BPD’s progress for approximately another 12 to 18
months, after BPD finishes upgrading its Record Management System and officers begin reliably using
electronic field-based reporting. That delay will inevitably prolong the life of the Consent Decree, as BPD
must be able to show sustained improvement in each area in order to achieve substantial and effective
compliance.

Over the past year, the Monitoring Team conducted its first comprehensive compliance review of internal
investigations by the Public Integrity Bureau. The review covered a randomly drawn sample of the
investigations PIB conducted in 2018. A summary of the Monitoring Team’s conclusions from this
compliance review are included in the Misconduct Investigations and Discipline section below, and the
full report is included in Exhibit 2.

The Monitoring Team will soon begin comprehensive compliance reviews of use of force incidents. The
use of force review will include a randomly drawn sample of force cases from 2018-19. For each case, the
Monitoring Team will evaluate the appropriateness of the force used, the accuracy and completeness of
the reporting, and the quality of supervisory review.

Due to the above-mentioned deficiencies in BPD’s IT systems and data collection practices, the Monitoring
Team’s work on outcome assessments has been preparatory thus far. In the first year of monitoring, the
Monitoring Team and statistical expert Dr. Ralph Taylor invested significant time in understanding
which relevant data BPD has available and which, if any, of the required outcome assessments would be
worthwhile or even possible. In December 2018, due to competing professional commitments, Dr. Taylor
had to resign from the Monitoring Team. After an extensive search, the Monitoring Team contracted with
the Crime and Justice Institute at Community Resources for Justice (“CJI”) to help design and perform
its outcome assessments. The Monitoring Team and CJI are partnering with Dr. Gabriela Wasileski of
the University of Baltimore, a statistical and social science expert. The Monitoring Team has worked
extensively with its new experts to chart a course for completing the outcome assessments required by
the Consent Decree.

One area of note is stops, searches and arrests. DOJ’s investigation found a pattern or practice of race-based
and otherwise unlawful stops, searches and arrests. It is not an overstatement to say that reforming BPD’s
stop, search and arrest practices is at the heart of the Consent Decree. Indeed, the fractured relationship
between BPD and many communities is thought to be largely attributable to BPD’s past practice of
making stops, searches and arrests without adequate legal justification. Accordingly, the availability and
integrity of BPD’s data on stops, searches and arrests has been, and continues to be, of special interest to

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 101
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 103 of 164
the Monitoring Team. As the Monitoring Team has previously reported, this data, particularly on stops
and searches, is both notoriously hard to access (requiring manual review of scanned paper reports) and
incomplete (due to underreporting by BPD officers). Given the data’s inadequacy, the Monitoring Team
and the parties have frequently discussed, and will continue to discuss, the feasibility and the utility of
evaluating stop, searches and arrests prior to the introduction of a new records management system.

Unfortunately, however, the quantitative review of the data that the Consent Decree requires—to
determine aggregate performance and investigate disparate impact—is probably impracticable at
this point. A qualitative compliance review of some subset of reports on stops, searches or arrests—to
determine officer compliance with policy on conduct and reporting, plus adherence to Consent Decree
requirements on supervision—may be feasible, but to make it so, the Monitoring Team will first have to
find a way to eliminate or reduce the cumbersomeness of the exercise. To take one example, to determine
the frequency with which investigative stops turn up evidence of a crime and the adequacy of the
documented justifications for such stops, the Monitoring Team would have to read through thousands of
incident reports in hard copy simply to determine which ones feature investigative stops, because BPD’s
current record management system does not currently have a way of segregating encounters initiated by
investigative stops.

|DIAGNOSTIC PERFORMANCE REVIEWS

As part of the work it must do to assess BPD’s compliance, the Monitoring Team also has been conducting
diagnostic analyses of BPD’s performance in discrete matters. The purpose of these analyses is not to
formally gauge BPD’s compliance with the Consent Decree, but rather to get a sense of how certain core
functions are currently being performed and, if they are not being performed effectively or in compliance
with the Consent Decree, to provide BPD guidance on how to improve performance. The ultimate objective
is for BPD to meaningfully engage in its own after-action assessments so that when it finds problems, it
will self-correct and take remedial action on its own, without prompting from the Monitoring Team or
DOJ.

The Monitoring Team conducted one such diagnostic analysis by evaluating the propriety of BPD’s
interactions with civilians in the Harlem Park neighborhood following the shooting death of BPD Detective
Sean Suiter. See ECF No. 126-1 at 56-67 (explaining concerns about how BPD officers performed and
documented stops, searches and arrests). In December 2018, as a result of the Monitoring Team’s findings,
as well as the findings of an Internal Review Board that conducted a separate assessment of BPD’s
investigation of Detective Suiter’s death, BPD prepared and delivered a full-day training for all command
staff on both constitutional requirements for conducting stops, searches and arrests and proper use of an
Incident Command System for responding to significant events. In addition, BPD prepared and delivered
to all officers a mandatory e-learning training program on BPD policies and constitutional requirements
on stops, searches and arrests through its PowerDMS system. The Monitoring Team and DOJ provided
input on the lesson plan for the command staff training, attended that training (as did Judge Bredar),
and gave BPD feedback following the training. The Monitoring Team and DOJ also provided input on the
e-learning lesson plan for the officer training on stops, searches and arrests.

Since its assessment of the Harlem Park lockdown, the Monitoring Team has routinely performed similar
assessments. For instance, we show up at the scene, participate in briefings, and scrupulously review
reports and body-worn camera footage of police-involved shootings. A recent example is the July 2020
police-involved shooting of a man experiencing a behavioral health crisis in the basement of a home
in Northeast Baltimore, which occurred in the context of both the recent, Consent Decree-mandated
“Gap Analysis” identifying shortcomings in the City’s behavioral health system and nationwide protests
calling for allocating adequate resources to behavioral health care. As the Monitoring Team reported on
its Facebook page:

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 102
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 104 of 164
From the beginning, the Monitoring Team has been closely following last week’s shooting of
a Baltimore man experiencing a behavioral health crisis. Lead monitor, Ken Thompson, was
on the scene immediately following the shooting. Several team members, including the team’s
behavioral health expert Dr. Randy Dupont, watched the body worn camera footage, listened
to the 911 call and dispatch recordings, and reviewed the police reports in the immediate
aftermath of the incident. We also have been conferring regularly with Judge Bredar. And we
paid close attention to today’s public release of body-worn camera footage.

While Dr. Dupont and Monitoring Team leadership continue our 360 degree assessment of the
incident [which the Monitoring Team plans to include a future report], one major question it
raises is whether it resulted from gaps or deficiencies in the City’s behavioral health system
that were identified in a comprehensive, Consent Decree-mandated analysis published late last
year. According to The Sun, ten days before the incident, the man who was shot was hospitalized
during another behavioral health crisis. The Monitoring Team has yet to see evidence that
there was adequate behavioral health follow-up after the man’s discharge or adequate follow-
up investigation to determine whether the man had additional weapons. That kind of follow up
is critical, as it can reduce the risk that incidents like this will occur.

One of the goals of the Consent Decree is for the City to do the hard work of plugging the gaps
in its behavioral health system so that these events became less frequent. As the Monitoring
Team has reported, the City recently has started down this road in earnest, but it still has a
long distance to travel.

In addition to this and other police-involved shootings, the Monitoring Team actively reviews non-lethal
incidents involving BPD officers. In the past year, such incidents have included an arrest on Pennsylvania
Avenue that was originally believed to involve an assault by bystanders on the BPD sergeant who made
the arrest but recently resulted in criminal charges against the sergeant for making false statements
regarding the incident; an officer’s alleged use of a chokehold during a drug arrest; and several incidents
that led to the indictment of a BPD sergeant for allegedly abusing his authority by using excessing force
and making false arrests. With these incidents, the Monitoring Team evaluates how BPD handles them
just as closely as whether the involved officers engaged in misconduct. In even the most well-run law
enforcement agencies, individual officers will periodically engage in misconduct. The health of an agency
is determined by how effectively the agency responds.

In May and June 2020, as reported in the Executive Summary and First Amendment-Protected Activity
Section above, the Monitoring Team conducted an intensive diagnostic evaluation of BPD’s response to
sustained demonstrations against police brutality in the wake of the killing of George Floyd in Minneapolis.
We spent dozens of hours live-monitoring BPD’s performance from BPD’s command center and in the
street. Our conclusions are included in the First Amendment-Protected Activities section below.

Finally, the Monitoring Team actively monitors approximately twenty specific PIB investigations at a
time, receiving regular monthly updates from PIB supervisors. Once the investigations are concluded,
the Monitoring Team and DOJ assess whether they were conducted and resolved properly, with an eye
toward providing remedial guidance if they were not. In one case from 2018, which involved allegations
that an officer exaggerated certain facts in his trial testimony in a drug case, the Monitoring Team
and DOJ found shortcomings in PIB’s investigation and conclusions and recommended certain remedial
training for PIB investigators as a result. If the Monitoring Team observes similar shortcomings in other
investigations, it will include its observations and recommended actions in future reports.

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 103
Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 105 of 164
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 106 of 164

BALTIMORE CONSENT DECREE MONITORING TEAM | FIRST COMPREHENSIVE RE-ASSESSMENT | SEPT. 30, 2020 105
Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 107 of 164

EXHIBIT 1
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 108 of 164

Implementation
Sustained
Policy/Planning Training Initial
Not Yet Assessed Off Track On Track Compliance
Section Compliance
Community Oversight Task Force
Community Policing
Stops, Field Interviews & Voluntary Contacts
Searches
Arrests
Stops, Searches & Arrests: Review & Supervision
Impartial Policing
Behavioral Health: General
Behavioral Health: CIT Officers
Behavioral Health: System Coordination & Improvement
Use of Force: General
Use of Force: Reporting & Supervision
Interactions with Youth
Transportation
First Amendment
Sexual Assault Investigations
Technology Modernization
Policies Generally
Training Generally
Supervision: Field Training
Supervision: Duties & Training
Supervision: Early Intervention System
Misconduct: Intake
Misconduct: Investigations
Misconduct: Discipline
Misconduct: Transparency
School Police
Staffing, Recruitment & Retention
Employee Performance Evaluations
Promotions
Officer Assistance

Legend Not Applicable Not Started In Progress Complete

Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 109 of 164

EXHIBIT 2
Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 110 of 164

Summary of Aggregate Results:

Monitoring Team Preliminary Baseline Assessment
of 2018 OPR Investigations

September 30, 2020

1
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 111 of 164

EXECUTIVE SUMMARY

This report summarizes the results of the Monitoring Team’s review of the Baltimore Police
Department’s (“BPD” or “the Department”) 2018 misconduct investigations, conducted at the time
by the Department’s Office of Professional Responsibility (“OPR”), which is now called the Public
Integrity Bureau (“PIB”). The report is intended to establish a preliminary baseline for assessing
the quality of misconduct investigations over the life of the Consent Decree.

This report is technical and details various statistical analyses throughout. As we are writing for
several different audiences – the Court, the public, the City of Baltimore, BPD, the Department of
Justice, and even academics – the report attempts to present the required information as simply,
but as thoroughly, as possible.

This assessment is referred to as a “preliminary baseline assessment” for a few reasons. First, the
methodology that the Monitoring Team deployed, with the agreement of the Parties, was designed
to get a working overview of where BPD stood on misconduct investigations as of 2018 – a point
at which BPD had tried to make some important changes but was still contemplating implementing
many other, core reforms. Consequently, many case files were reviewed by just one Monitoring
Team reviewer. In the future, when there is reason to believe that BPD’s performance may be
closer to the level of compliance, a more rigorous methodology is likely to be appropriate.

Second, this assessment is not an assessment of the current quality of the Department’s internal
affairs investigations. It does not capture many systemic changes that have already been made.
Instead, this report captures where the Department was in 2018 with regard to the requirements of
the Consent Decree for misconduct investigations. Considered in this light, the findings of this
assessment are wholly unsurprising – and, perhaps to some extent, underwhelming.

The main findings of this report include the following:

 The overall quality of BPD’s process for investigating and making findings in
misconduct investigations remains in need of significant improvement. Some 40
percent of OPR investigations in 2018 were determined by Monitoring Team experts to be
of poor or fair quality (on a five-point scale from poor to excellent, defined in detail below).
Although a majority (61 percent) of cases were rated as “good” or above, more cases were
rated as “poor” or “fair” in quality (40 percent) than “very good” or “excellent” (23
percent), indicating a weighting towards the lower half.

Even among the cases that were “good,” “very good,” or “excellent,” systemic deficiencies
indicate that BPD will need to travel a significant distance to reach compliance with the
Consent Decree’s many provisions relating to misconduct investigations. For instance,

2
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 112 of 164

almost 42 percent of the cases we reviewed were “administratively closed” without a

disposition (a determination on the merits). Of the cases that made it to disposition, a full
81 percent did not have narratives that “include a precise description of evidence that either
justifies or fails to justify officer’s conduct based on investigator’s independent review of
facts/circumstances of the incident.” Material inconsistencies in the evidence were
resolved adequately in only 34 percent of cases, and determinations about the credibility
of witness statements were properly made in only 35 percent of cases.

 The condition of misconduct case files was poor. The documentation of investigative
materials was chaotic and inconsistent. As a result, reviewers had great difficulty in making
their assessments and, unfortunately, frequently had to use a score of “unable to determine”
when auditing the many specific requirements of the Decree.

 The quality of classification and intake of complaints was poor. The Classification and
Intake Manual developed through the Consent Decree process has not yet been
implemented. However, using that as a measure for this report, the classification and intake
deadlines, although often difficult to discern from the files, were not met in half of cases
we reviewed. In addition, the most serious alleged rule violation was used to determine
classification and OPR jurisdiction in only two-thirds of cases.

 Allegations of excessive force, unlawful search and seizure, First Amendment

violations, and criminal acts were rare and typically well-investigated.

 Communication with complainants was poor overall. Reviewers could certify that
complainants were sent written notice of OPR’s receipt of their complaints in only a quarter
of cases and were sent status updates only a third of the time. However, in contrast to the
DOJ Findings, which emphasized that OPR investigators discouraged complainant
participation in the investigation, reviewers did not identify any such cases. This is
encouraging.

 Communication with employees was likewise poor. BPD provided timely notification
to personnel who were the subjects of complaints in less than a quarter of cases. Reviewers
did not identify any cases where the notification jeopardized the investigation.

 Cases were inefficiently managed, which resulted in consistently missed timelines and
wasted efforts on minor cases. Despite the poor condition of case files, reviewers were
able to determinate that 60 percent of investigations were not completed within the 90 day
deadline. Extensions of time were never requested.

3
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 113 of 164

 Other performance issues surfacing during the investigation, such as repetitive

problematic behavior, training issues, or the need to explore policy revisions, were
rarely considered.

 Investigators frequently failed to interview complainants and other necessary

witnesses. Complainants were interviewed only 41 percent of the time; subject employees
were interviewed in 44 percent of cases. On the positive side, reviewers found bias, leading
questions, and concerns about investigator behavior very rarely in both complainant and
subject employee interviews. BPD employee witnesses were identified in about a third of
cases, but of those, only 61 percent were actually interviewed. Civilian witness interviews
were very rare, as were interviews of non-BPD law enforcement witnesses. Only 22 percent
of cases had complete audio or video records of all witness interviews.

 Across misconduct investigations, complainant injuries were uncommon. Injuries

occurred in only 6.4 percent of cases; officer injuries occurred less than one percent of the
time and never due to subject actions. Complainant injuries were documented, and
photographs taken, in a majority of cases.

Overall, there is much that needs to improve before misconduct investigations at BPD will
reach compliance with the terms of the Consent Decree. As noted repeatedly in this report, the
quality of BPD’s misconduct investigations in 2018 is not unexpected. Although there may be
some improvement when comparing investigations in 2018 and investigations from 2010 through
2015, the period of DOJ’s investigation of BPD, BPD is still in the process of making major
changes in its practices for addressing and investigating allegations of misconduct. Since 2018,
the Department has been working hard on implementing these changes, including producing new,
high quality Intake, Classification and Investigations Manuals for the Public Integrity Bureau
(“PIB,” which is what OPR is now called).

The findings in this report confirm the need for BPD to overhaul the way that it addresses officer
misconduct and sets the level for gauging future improvements.

I. INTRODUCTION

This report summarizes the results of the Monitoring Team’s assessment of the Baltimore Police
Department’s misconduct investigations in 2018, conducted by the Department’s Office of
Professional Responsibility(“OPR”), which at the time was the primary internal affairs unit for the
Department. The purpose of the evaluation was to assess the Department’s baseline level of
performance, during an era prior to implementation of the significant reforms required by the
Consent Decree between the City of Baltimore, BPD and the United States (the “Consent Decree”).
From these baseline results, progress over time can be measured.

4
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 114 of 164

The Consent Decree expressly recognizes that “[a] robust and well-functioning accountability
system in which officers are held to the highest standards of integrity is critical to BPD’s
legitimacy, and a priority of the Department.” Dkt. No. 2 ¶ 329. Accordingly, it sets forth
procedural and substantive requirements for BPD’s internal investigations function, with specific
emphasis on complaints implicating constitutional issues, such as False Arrest, Excessive Force,
and infringement on First Amendment-protected activities. See Dkt. No. 2 at 112 et seq. The
Decree clearly defines what a well-functioning accountability system looks like:

[O]ne in which BPD: openly and readily receives complaints reported by civilians
and officers and fully, fairly, and efficiently investigates them; supports all
investigative findings by the appropriate standard of proof and documents them in
writing; holds accountable all officers who commit misconduct pursuant to a
disciplinary system that is fair, consistent, and provides due process; and treats all
individuals who participate in BPD’s internal disciplinary process—including
complainants, officers, and witnesses—with respect and dignity. Id.

DOJ identified systemic failures in BPD’s investigation and adjudication of officer misconduct,
including: “discouraging individuals from filing complaints; poor investigative techniques;
unnecessary delays; minimal review and supervision; and a persistent failure to discipline officers
for misconduct, even in cases of repeated or egregious violations.” Findings Letter at 139. Specific
issues included discouraging members of the public from filing complaints, misclassifying
complaints to minimize them, administratively closing complaints prematurely, and failure to
investigate complaints in a timely manner. Id., 140 et seq. These deficiencies, combined with the
failure to properly supervise investigations and inadequate civilian oversight, resulted in systemic
failures in accountability at the BPD.

At the time of the Consent Decree, the Department’s Office of Professional Responsibility
conducted investigations of officer misconduct. Consequently, the Decree’s specific requirements
attach to OPR. BPD Police Commissioner Michael Harrison has since restructured the department
to create the Public Integrity Bureau (“PIB”), which investigates allegations of misconduct for
BPD. At the time of this review, OPR1 had primary responsibility for investigating allegations of
misconduct.

1
Currently, PIB investigates allegations of misconduct and its efficacy will be assessed in
subsequent reports. For purposes of this report, the Monitoring Team uses “OPR” to refer to the
investigative arm of BPD, as that was the structure at the time the reviewed cases were completed
and to make clear that this is not a review of the current structure. Again, this whole report is a
baseline.
5
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 115 of 164

Separately, Baltimore’s Civilian Review Board (“CRB”) reviews allegations of misconduct that
fall within its statutory purview: complaints lodged by members of the public regarding police
officers’ (1) abusive language, (2) false arrest, (3) false imprisonment, (4) harassment, and (5)
excessive force. Those allegations are defined in the Code of Public Local Laws of Baltimore City
(P.L.L.), §§ 16-41(b)–(f).

Deputy Commissioner Brian Nadeau assumed leadership of PIB and the OPR function in
September 2019 and has been working toward overhauling PIB.

On November 18, 2019, the Monitoring Team approved a new Investigations Manual for PIB and
filed it with the Court. Dkt No. 263. That manual addresses the processes, standards, and
benchmarks to which PIB investigators must adhere as they conduct an investigation. A separate
manual on the intake and classification of civilian complaints, the PIB Intake and Classification
Manual, has also been filed with the Court.

While a part of the Third-Year Monitoring Plan, this assessment comes at a very early point in the
process of overhauling the accountability systems at BPD, especially as the bulk of cases that the
Monitoring Team reviewed were completed in 2018 – before BPD began making any meaningful
changes to its accountability systems. Therefore, it is important to recognize that this assessment
is not designed to test whether BPD has successfully implemented reforms to their accountability
systems. Instead, this assessment establishes a general baseline to measure whether the changes
that are currently underway actually result in improvements.

BPD will have successfully adopted of a robust accountability system not simply when it achieves
compliance with the many specific, technical requirements of the Decree, but when officers and
community members alike develop confidence in its ability to fairly identify, investigate, and
address conduct that violates Departmental rules.

II. METHODOLOGY

The Monitoring Team has done its best to set clear guidelines for reviewers, to train reviewers,
and to hold itself accountable for its results. Accordingly, this report, including its qualitative
findings, is heavily data-driven and technical. In this section, we explain the methodological
decisions we made that generated the data on which our conclusions rely.

The methodology outlined here was developed for the purposes of conducting a preliminary
baseline assessment. The Parties and Monitoring Team agreed to the following methodology in
order to identify a clearer understanding of where BPD stood with respect to the quality of
misconduct investigations in 2018 – before many Consent Decree reforms were implemented – in
the most effective and efficient manner. An expanded methodology is anticipated as the

6
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 116 of 164

Monitoring Team charts progress over time and begins to assess whether BPD’s various reforms
have been implemented sufficiently to constitute eventual compliance with the provisions of the
Decree.

A. Development of the Methodology

The methodology for this assessment was developed in collaboration with the City of Baltimore,
BPD, the Department of Justice and the Monitoring Team. The sample size and the survey
instrument were subject to discussion and agreed on by the parties, with full opportunity to consult
with external statistical experts and provide feedback during the development process. The
reviewers were provided with the data collection instrument and the coding instructions. As
discussed in more detail below, because the case files were chaotic and inconsistent, there was no
uniform way for reviewers to obtain the required information from the case files. Instead, reviewers
had to carefully examine each case file to obtain whatever information they could in order to
complete the survey instrument.

B. Sample Size and Composition of Cases

As of June 2019, the Department reported a total of 1,372 completed OPR investigations initiated
in 2018, in addition to 112 cases with a separate status of “suspended” and 5 cases with a status of
“initial.” Since these 117 cases were not really “completed,” they were eliminated from the initial
sample.

BPD sorted the completed cases into several “incident types,” purportedly based on their origin
(internal or external), the nature of the alleged violation (failure to appear in court, ethics, firearm
discharges, serious incidents), and the particular unit responsible for the investigation (Command,
Ethics, SIRT (Special Investigation Response Team Investigations (“SIRT”), or Internal Affairs).

Specifically, the incident types BPD used2 were:

 Command Investigations (“CIU”) – minor investigations that were handled by command

staff (not OPR) because BPD determined they did not warrant a full internal affairs
investigation
 Court Failure to Appear (“FTA”) – officer failed to appear in court as scheduled
 Ethics – complaint alleges misconduct by OPR personnel, corruption, or on-going
misconduct, or requires proactive investigation

2
BPD has modified the classification process in its recently completed PIB Intake &
Classifications manual. PIB now includes three investigative entities – Internal Affairs, Ethics, and
SIRT – and uses three complaint designations – Serious Misconduct, Other Misconduct, and
Service Complaint.
7
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 117 of 164

 External Complaints – a complaint of misconduct made by a member of the public

 Firearm Discharges – an investigation into the discharge of a firearm
 Internal Complaints – a complaint made by a BPD employee
 SIRT – investigations of serious use of force incidents

To the extent that the explanations or definitions of these incident types are unclear, it should be
noted that these were previously developed by the Department – and are distinct from categories
that BPD will use under the Consent Decree in the future.

In their evaluations, Monitoring Team reviewers found that these classifications were
inconsistently applied and subject to overlap, and the distinctions were often unhelpful. For
example, “External Complaints” could include ethics or firearms complaints – and could
potentially be handled as a Command Investigation. CIU complaints entailed the least discernibly
consistent investigation process, while FTA cases were the most consistent and thorough, as they
only addressed a simple, discrete issue – whether the officer appeared in court. Given these
conceptual shortcomings in incident type classifications, the value of sampling according to
incident type was simply to ensure that a comprehensive cross-section of cases was evaluated. The
table below provides the total breakdown of 1,372 “completed” cases based on BPD-designated
incident type:

1,122 Command Investigations

64 Court Failure to Appear
18 Ethics
105 External Complaints
4 Firearms Discharge
55 Internal Complaints
4 SIRT Investigations

To secure the appropriate representation for each case category, the final sample was randomly
selected as follows:

N Margin of Category
Error
89 ±10% CIU
39 ±10% FTA
18 ±0% Ethics
51 ±10% External Complaints
4 ±0% Firearms Discharge
36 ±10% Internal Complaints
4 ±0% SIRT Investigations
241 ±8% TOTAL

C. Cases Reviewed by Multiple Reviewers to Examine Inter-Rater Reliability

8
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 118 of 164

A basic question for any assessment with multiple reviewers is whether the reviewers are
interpreting the questions consistently and providing consistent responses. To test this, a sub-
sample of the 241 total cases (N=40) was randomly selected for review by two independent
reviewers. For these 40 cases, the two reviewers separately reviewed and gathered data from this
sub-sample of logged and completed misconduct investigation files. Note that the 18 “Internal
Investigations” cases contained a cross-section of Ethics, External Complaints, Firearm
Discharges, Internal Complaints, and SIRT investigations.

18 Internal Investigations
9 Court FTA
13 CIU

These 40 cases were integrated into the overall sample size, but as they were double-reviewed, the
value of each case was weighted at .5, which gave proportionate effect to each individual case and
avoided double counting these cases.

The purpose of establishing a sub-set of cases to be reviewed by two, rather than just one, reviewer
was to be able to evaluate whether anything about the nature of the assessment instrument or about
specific reviewer biases may have affected the results. In more formal terms, the double-review
of some cases allow for the computation of inter-rater reliability.

D. Measuring Inter-Rater Reliability

The inter-rater reliability of the double-reviewed cases was measured as a percent agreement.
Proportion or percentage of agreement is the most straight-forward and widely used way to assess
consistency/reliability of scores on categorical variables even though levels of agreement can arise
purely by chance. In other words, some agreement is expected because two raters could come to
the same conclusion accidentally rather than because they actually agreed on the established code.
Thus, throughout this report, where possible, an “agree” column in tables shows the percent of
cases in which reviewers came to the same conclusion on any given question. The proportion of
agreement – the “agree” column in the various tables presented in this report – demonstrates the
consistency of responses provided by two independent reviewers. It should be noted the “agree”
column is not reported for questions with a small sample size – less than eight – or questions
related to demographic information because the small size makes the statistical test less valid.

The Consent Decree places the burden on the City and BPD to demonstrate that it is affirmatively
complying with the various requirements of the Consent Decree. With respect to misconduct
investigations, the Decree requires BPD to “support all investigative findings by the appropriate
standard of proof and document them in writing.” Dkt. No. 2 at ¶¶ 329, 343. Therefore, where
Monitoring Team reviewers said they were “Unable to Determine” whether an assessment

9
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 119 of 164

criterion had been met, those responses are combined with “No” responses to show BPD’s overall
noncompliance with a criterion. For purposes of assessment, then, the question is whether the City
can demonstrate a sufficient number of “Yes” responses, making the relative percentage of “Yes”
and “Not Yes” (i.e. “No” or “Unable to Determine”) the central inquiry.3

The Monitoring Team is reporting the percentage agreement as a way of understanding the results,
and the relative weight to assign to any one finding. Low agreement on questions for cases that
were reviewed by multiple reviewers can indicate a problem with the question being asked, a
problem with reviewer understanding of the question, a problem in readily determining an
appropriate answer based on the thing being reviewed, or that the question simply cannot be readily
answered.

For the purposes of gauging progress and certify compliance going forward, any disagreement
between raters will be addressed and reconciled, including with the assistance of a third reviewer.
For purposes of this preliminary baseline assessment, the parties and the Monitoring Team agreed
that devoting the resources to resolving disagreement was unnecessary for the purposes of this
report. Again, this assessment was geared toward establish a baseline in an area where the Parties
expected, for a variety of reasons, that BPD would will need to make much additional progress
going forward. Where low rates of agreement were found on a given sub-question, the Monitoring
Team, in collaboration with the parties, will reassess the survey instrument and the instructions
provided to reviewers to ensure that future reviews yield more robust results.

As discussed throughout this report, the state of the case files varied greatly – some were complete
and easy to navigate; some were chaotic and even internally inconsistent. Different case types,
e.g., CIU, FTA, or internal complaint cases, all had different formats and were not always alike
within each category. As one reviewer noted:

 Again, I note that the case summaries can be confusing. For example, dates don't
always align with the cover sheets and reports. This seems to be strictly an
administrative issue. It would be more efficient to eliminate the written checklist
and sign-off sheet and go with a clearer integrated, automated solution.

Another reported:

 It continues to be difficult to make accurate assessments on timelines as reports are

often not dated and summaries are inconsistent and confusing.

As OPR investigators did not investigate cases or document their investigations in a uniform
manner, it was not possible to train Monitoring Team reviewers “how it is done.” Reviewers were

3
On cases where “no” was the desired answer, this was reverse-coded as “no” and “not no.”
10
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 120 of 164

required to approach each case, figure out how it was structured, and investigate the materials to
score the required evaluation criteria. Once PIB establishes a consistent intake, classification and
investigation process, and especially once all PIB actions are electronically stored, future
reviewers will be able to be trained specifically where in an investigation file to locate the
information needed to score each evaluation requirement in the assessment instrument. This will
lead to faster reviews and a higher percentage of reviewer agreement across the board.

Much of the disagreement between reviewers was found in sub-questions. The workflow of the
survey instrument offered different options based on responses to questions. For example, if a
reviewer indicated there were problems with an interview, the instrument provided options for
exactly what the concern was. While there was good agreement between reviewers about whether
an interview was problematic, the reasons for the concern varied between reviewers.

Sub-questions also exposed a complication in the methodology for determining percentage of

reviewer agreement. Because the double-reviewed cases are compared against each other, if
reviewers disagreed on a core question, that disagreement was necessarily carried over into the
sub-questions. For example, in Q40, reviewers were asked whether the involved employee was
interviewed. There was 67.5% percent agreement between reviewers on that question. Q42 (“Did
the employee decline to be interviewed”) only appeared as an option for reviewers who answered
“no” to Q40; otherwise the instrument skipped the question. This appropriately controlled relevant
follow-up questions for the single-reviewed cases. However, for double-reviewed cases, this
resulted in missing values for reviewers that answered “yes” to the threshold question because that
reviewer was not presented with the sub-question. The software read the comparison of a missing
value with any other answer as disagreement. To fix this, we removed cases in sub-questions with
missing values. Although this methodology lowered the sample size for the double-reviewed cases,
it allows us to more accurately present the level of agreement and to overcome the limits of the
logic in the software.

All of this leads to the conclusion that although the Monitoring Team includes percent agreement
for transparency, readers should not over-focus on this metric and lose the forest for the trees.
Throughout the report, the percent agreement was typically above 60 percent, with some outliers.

E. Testing the Survey Instrument

In addition to the level of agreement between reviewers, the Monitoring Team explored the quality
of the data collection instrument, seeking to determine whether the survey instrument was effective
at answering the questions it was designed to answer. Thus, we computed a measure of internal
consistency reliability, called the Kuder-Richardson (“KR-20”). The KR-20 measures the
consistency of the score(s) of individual items on the data collection instrument. It can be
interpreted similarly to Cronbach’s alpha. For each section computed, the “alpha” is presented and

11
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 121 of 164

interpreted as a proportion of variance. The value of alpha tells us how closely related a set of
items are as a group (do they measure the same phenomenon). Ideally, the alpha values should
range between .70 and .90. An alpha of 0.6 and below would be considered poor reliability in most
research situations; above .90 suggests redundancy in the questions. Generally, the alpha was good
for the Monitoring Team’s survey instrument. Areas where a low alpha is found will be
reconsidered for the next iteration of the misconduct assessment.

Examining the results for this assessment, the Monitoring Team identified some questions that
were either duplicative or in conflict with other questions. For example, Q33 was linked to Q31,
but should have been linked to Q32 and limited to cases where there was an actual refusal to be
interviewed. Similarly, Q50, Q57, and Q63, which also asked about documenting refusals to be
interviewed, lacked the necessary precursor question of “Did the Complainant Grant Permission
to Audio Record?” As such, the results have no value and are not reported, as they will only add
confusion. Questions 33, 50, 57, and 63 are therefore excluded. The Monitoring Team will correct
this survey error in the next review.

III. RESULTS

This report is logically divided into sections based on issue area. Some questions have been
rearranged from their original order in the data collection instrument for reader clarity (although
the charts include the question numbers for reference). The answers to the questions, both
numerical and percentage of the sample size (N), are presented, and where appropriate, followed
by the percent agreement (Agree) between reviewers for the double-reviewed cases and by the
alpha value.

Sub-questions are indented in the tables to show their dependency on prior answers. The number
of cases (N) within the sub-questions are also presented.

Again, the purpose of this report is to test the waters, not draw significant conclusions about the
quality of OPR (now PIB) investigations. The results are not unexpected, as new polices and
training have not yet been implemented.

The Monitoring Team notes generally here that, across some areas of evaluation and auditing, the
reported “inter-rater agreement” rates for those cases that were evaluated by two reviewers are
lower than the Team would typically want to see. Overall, we believe that this is related, first, to
the often chaotic and incomplete nature of the files – with documents seemingly randomly included
or omitted – and similar pieces of information being reported differently on different forms or
different reports. Second, and especially within this context of unstructured and incomplete case
files, reviewers used different, good-faith approaches to gauge the misconduct investigation in
light of various requirements that BPD was, in many instances, not yet trying to satisfy in its

12
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 122 of 164

process – which led to diverging conclusions on some particular elements of the evaluation. The
Team is optimistic that better, more complete case files structured with eye at compliance with
Consent Decree requirements will allow reviewers, using an expanded methodology, to better
operationalize areas of inquiry – yielding higher levels of agreement on some of the dimensions
where there is less agreement than desired in this initial evaluation.

OVERALL CASE EVALUATION

We begin with a discussion of the Monitoring Team’s overall case evaluations. Although these
evaluations are, to some extent, the least precise and the most subjective, an understanding of the
overall quality of the investigations helps to guide an understanding of the more specific
evaluations presented in the remainder of the report.

The Consent Decree requires that misconduct investigations be objective, comprehensive, and
timely. Dkt. No. 2 ¶ 329. Accordingly, reviewers were asked to rate the overall quality of the
investigations on a five-point scale, using the following definitions:

5 – Excellent – The investigation complied with all Consent Decree requirements and BPD
protocols, and investigators made reasonable attempts to follow all leads and answer all
material questions. The investigation was fair, thorough, objective, and timely.
4 – Very Good – The investigation complied with most Consent Decree requirements and
BPD protocols and investigators made reasonable attempts to follow all leads and answer all
material questions.
3 – Good – Although some aspects of the investigation could be improved, the identified flaws
did not appear to materially or unduly impact the quality of the overall investigation. The
resulting investigation provided sufficient information to evaluate the incident but could be
improved.
2 – Fair – Several aspects of the investigation could be improved. Identified flaws materially
impacted the quality of the overall investigation, and the resulting file provided insufficient
information to evaluate the incident.
1 – Poor – All or nearly all aspects of the investigation could be improved. The investigation
failed to establish sufficient information to support an evidence-based evaluation of the
incident due to investigative deficiencies, material omissions, or other issues.

Reviewers rated 40 percent of investigations as Poor or Fair. Slightly more than one-third (37
percent) of cases were rated Good, indicating that although the identified flaws tended not to
materially or unduly affect the quality of the overall investigation, the investigative quality could
still be improved. Less than one-quarter of investigations (23 percent) were rated either Very Good
or Excellent.

13
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 123 of 164

(Q92) Based on your overall review of the investigation, what was the overall quality of
the investigation?
5 – Excellent 6.5 (2.70%)
4 – Very Good 49 (20.33%)
3 – Good 88.5 (36.72%)
2 – Fair 51 (21.16%)
1 – Poor 46 (19.09%)

CIU cases, which were essentially ad hoc investigations conducted by BPD’s chain of command,
were rated “poor” and “fair” more often than other case types. This is likely because these chain
of command investigations were not being conducted in a manner that even attempted to comply
with Consent Decree requirements or the requirements of a full-blown investigation.

A plurality, some 41 percent, of investigations that were not FTA or CIU cases were “good,” and
another nearly 35 percent were “excellent” or “very good.” It must be noted at the outset, however,
that even among cases that were “excellent” or “very good” per the standards and definitions
articulated above, many investigations did not satisfy all of the specific requirements of the
Consent Decree that have been memorialized in policies and manuals for investigators since the
time of the investigation.

At the same time about one-quarter (25 percent) of these cases were of “poor or “fair” quality.”
The fact that approximately one out of every four non-FTA, non-CIU investigations was deficient
in fundamental ways indicates to the Monitoring Team – in conjunction with the many specific
findings presented throughout this report – that BPD has a distance to travel toward compliance in
the area of misconduct investigations.

14
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 124 of 164

Scaled Review by Case Type (% of cases)

60

50
50

40.54
40 35.96

29.78 31.08
30 25.84

19.5
20 17.07
15.77

9.769.01
10 7.3
3.663.6
1.12
0
Poor Fair Good Very good Excellent

FTA Internal Investigations CIU All cases

Across case types, investigations that were rated very good or excellent stood out, and reviewers
had little difficulty making these determinations. Reviewer summaries of these cases were
generally clear and unambiguous. Some examples include:

 The investigation was thorough and complete. Investigator effectively reviewed the
actions of the officers starting at his initial contact with the suspect, subsequent foot
pursuit and use of force to take the suspect into custody. A copy of the Conducted
Electrical Device policy was provided in the investigative packet. The evidence
provided within the report demonstrated the officer was in compliance of the policy
and supported the investigative findings of unfounded.

 The Investigators conducted a thorough and very complete investigation. The

Investigators did not limit the scope of the investigation to solely the use of force but
investigated all the actions (police contact/pursuit) leading up to the suspect’s arrest.
The file is comprehensive and well done.

The inadequacies in cases rated fair or poor were similarly clear. Reviewers found deficiencies
both in the quality of the case files and the overall diligence of the investigators:

 The case file includes a number of related documents, including an incident report,
juvenile custody report, and an Axon report that suggests that there may be BWC video
of the interactions in question. However, there is no real investigative report here—
simply an IAPro summary that logs some very cursory investigative efforts. In short,

15
Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 125 of 164

there is no evidence that an investigator meaningfully considered the complainant's

claims, or the basis for "closing" the case without taking action. Although the nature
of the complaint may well be outside the purview of OPR (essentially a claim that
charges should not have been filed/pressed in a given case), there is insufficient
justification provided in any materials for closing the case with minimal investigative
action/effort.

 I would not categorize this as an investigation. It is a confusing data entry. It is

impossible to ascertain the nature of the complaint or whether the complaint is directed
at a BPD employee. I'd like to discuss the BPD data entry systems. This was an
automated entry, but a useless one. If they were required to enter something, it should
have included more information; for example, the nature of the complaint and whether
or not caller identified the subject as a BPD employee.

 The summary and packet were included in the file, but the investigator's report was not
present. However, there was a BWC violation and the conclusion was Unfounded. I
suspect this was because the officer had requested re-training on BWC prior to the audit
that revealed the violation. As there was no investigation in the file, I am unable to
conclude one way or another whether the investigation was adequate. Based on the
packet, there was enough evidence to conclude that a violation occurred, however, so
I am unclear how they got to unfounded.

 The case was closed with little to no investigation. There is no evidence they tried to
locate the complainant to call them on the phone. The summary shows a notice was
mailed to the complainant, but the case shows closed two days after it was opened. Also
the investigator identified on the BWC footage that the officer refused to provide his
partner's name so the complainant could lodge a complaint with all the officer's name.
Nothing was done related to this issue and the complaint was closed.

 The Consent Decree requires that investigators will conduct objective, comprehensive,
and timely administrative investigations. This investigation failed to meet these
thresholds. Specific allegations clearly outlined in the complainant’s letter are ignored
and not addressed. The investigators reliance on the Body Camera as singular
determining factor when the complainant outlines several allegations is not an effective
investigative process and can limit the objectivity. There is no evidence provided that
demonstrated the 911 recorded call was reviewed. The Officer called the Complainant
via his phone, no evidence was gathered to address his conduct during the call. The
complainant was not formally interviewed. The Complainant called a supervisor who
did not respond but called the complainant after she left the location, this was never

16
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 126 of 164

addressed. The investigation failed to identify the policy cited and recognize that the
confusion by the Complainant started with inconsistent information by BPD personnel.

Reviewers found that investigators had a strong tendency to rely on BWC footage or documentary
evidence and forgo witness interviews. Often investigators would simply gather information until
they believed they had enough information to conclude the investigation, rather than fully
following the investigation to its conclusion. Complainant interviews were frequently neglected.
Even when witness interviews occurred, frequently they were not included in the case file, either
as recordings or transcripts, so it was difficult to assess quality. For example, in certain cases where
the failure to conduct witness interviews compromised the investigation, the reviewers reported:

 This was a generally solid investigation of a PO trainee caught cheating by several of

his peers. However, this was only a "good" investigation because officers who saw the
cheating were not interviewed; instead, written statements were used.

 As this case involved the conflicting testimony of two parties, additional interviews
should have been conducted to resolve if possible. No attempts were made to interview
the officers, the responding supervisor, the involved parties, except for the father of the
involved female, who was not present during the dispute. While the video confirmed
there was likely no false arrest or false statement, this investigation could have been
much stronger.

 The investigation was missing interviews of the subject officer, the CO that assaulted
the suspect, and the responding Sgt. Instead it relied on BWC evidence and written
statements. The officer on scene did exactly what he should have - called for a
supervisor and professionally interacted with the suspect who had been assaulted by
the CO, but failed to raise the issue of criminal assault. The responding Sgt did not do
a good job of investigating the CO, for which there was already PC for assault; the
officer on scene did not discuss this with the Sgt either. The OPR investigator attributed
statements to the CO that I cannot find in the file.

BWC evidence was especially persuasive to investigators and frequently supported the employees’
version of events. Reviewers frequently noted that evidence lost when an officer did not turn on
the BWC would have been helpful to the resolution of the case. Some examples include:

 The core of this investigation was sound. The complainant's claims are definitively
refuted by the BWC evidence. The primary deficiency was the failure to identify, and
interview, the many other responding officers/law enforcement who are seen on BWC.
It appears that only the primary officer was interviewed. Although interviews with
hospital staff were audio-recorded, no recording of the primary involved officer

17
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 127 of 164

interview appears in the reviewed case file. The investigator's questioning was not bad,
though some reliance on leading questions in a few instances was unnecessary.

 Investigation was through and complete. Investigator effectively used BWC footage
combined with several witness statements and medical records to support findings.

In the few cases that alleged rudeness and unprofessional behavior, there was a tendency for the
investigators to minimize the officer’s behavior. As such, opportunities to address procedural
justice issues in call handling were missed. For example:

 The investigator and the Field Supervisor improperly characterized the officer's
conduct as professional when in the reviewer’s opinion it was not professional. He
insulted the complainant, commented on her parenting skills and lack of discipline
within the home. He also became angry after a minor use of force and was yelling at
the mother when he left the apartment. None of this was noted in the report. Moreover,
the report fails to discuss the allegation of conduct unbecoming and how the officer's
conduct did not violate this rule.

 The IA investigation was fine and the investigators very professional. It's just
unfortunate that a mother who was warming her car to bring her children to school was
ultimately arrested for refusing to sign a citation. I acknowledge that she was agitated
and behaved poorly, but I'm not certain an arrest would have been necessary if the
officers handled the situation a bit differently.

Reviewers noticed significant deviations between case cover sheets and the rest of the case file
concerning dates of investigative activities, including the start date.

Reviewers often noted the seeming lack of interest in exploring issues the cases presented beyond
the particular rule violations alleged, including tactical issues. In one case, which involved an
officer chasing a suspected armed suspect into a building alone, the investigation did not explore
the pursuit or the tactical issue of following a suspect into a closed space without backup. In
another case, where a corrections officer punched a BPD prisoner, there was no discussion by the
officer on scene, the responding sergeant, or the investigators about the possibility of criminal
charges, even though clear probable cause existed for assault. In contrast, there were cases where
reviewers were impressed that investigators took the time to explore issues not directly implicated
by the complaint.

Reviewers found several cases where minor infractions escalated into serious uses of force (one
was an officer-involved shooting), where although the actual use of force was reasonable,
necessary, and proportional, investigators should have given some consideration to whether

18
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 128 of 164

tactics, training, or policy should be adjusted. In one case, a mother waiting for her child was
arrested for failing to sign an infraction – all technically legal – but the incident could undoubtedly
have been handled in a way to reduce conflict and avoid that unfortunate result. In fairness, the
complainant was abusive and argumentative, but that is exactly the type of encounter that police
officers must handle daily. Another example:

 [I]t was clear to me that this was a thorough investigation and the officer was justified
in using deadly force when confronted with deadly force. Some may question,
however, whether the appropriate laws and/or policies are in place in terms of pursuit
(in this case air and foot) of suspects eluding police following minor infractions.4

There were a few cases where the inconsistencies in statements likely crossed the line to
dishonesty, which was not investigated further. For example:

 Based on the investigative report and the evidence provided it seems to be evident that
the involved officer lied to the investigator and on her interrogatory. She stated she was
on vacation but the records show her working and there was no slip for vacation time
on the said date.

Finally, turning to inter-rater reliability for this section, as there was a range of five possible ranked
responses, and as discussed above, there is a useful demarcation line to divide the responses into
poor/fair and good/very good/excellent, it is more helpful to look at the distribution of responses
rather than just the percent agreement.

Overall, there was complete agreement between reviewers in a third of cases (32.5%). More than
a third of cases were within one rank of another (35.5%). As such, there was close or perfect
reviewer agreement in 68 percent of all cases.

Similarly, dividing the cases into poor/fair and good/very good/excellent, there was reviewer
agreement in 70 percent of cases. An additional 7.5 percent of cases were split between fair/good,
which shows reviewer consistency even if the results were on opposite sides of the split. Generally,
because this means that, 32 percent of the time, reviewers disagreed by two ranks or more. the

4
BPD Policy 1505, Foot Pursuits, was drafted and approved by the Court in late 2018, and provides
guidance on balancing the interest of public safety with the risks of foot pursuits. Specifically, the
policy directs members: “Members may engage in Foot Pursuits with suspects only when there is
[reasonable articulable suspicion] to believe that the suspect has committed, is committing, or is
about to commit a crime (with the exception of those instances identified below under prohibited
actions) and when members reasonably believe that there is a valid law enforcement need to detain
the suspect that outweighs the threat to safety posed by pursuit. The decision to initiate or continue
such a Foot Pursuit, however, must be continuously re-evaluated in light of the circumstances
presented at the time.”
19
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 129 of 164

Monitoring Team would like to see higher agreement in this dimension in the future. If this were
a compliance assessment, the Monitoring Team would devote the resources to examining what
caused the disagreement between reviewers and would resolve that disagreement by using a third
reviewer.

However, at least for purposes of this preliminary baseline assessment, the Monitoring Team is
satisfied that the level of agreement sufficiently captures the general, past state of the OPR
investigations and provides guidance on where BPD should focus its efforts going forward.

COMPLAINT SOURCE/TYPE INFORMATION

We first consider where, how, and by whom complaints were received. Anonymous complaints
were rare – less than two percent of the sampled cases. While many police departments have some
concerns around accepting anonymous complaints due to worries about the possibility of false
allegations from community members or even other officers, the rarity of such complaints in
Baltimore should be reassuring. The rarity of anonymous complaints may suggest that community
members are not uniformly reluctant to identify themselves to the accountability system because
of fears of retaliation. It could also mean that individuals who would otherwise be afraid of the
consequences of making a complaint are staying away from BPD entirely, not even giving an
anonymous complaint. Other data points and further study will be necessary to understand what
the low number of anonymous complaints might mean.

Over 55 percent of the cases reviewed originated with complaints coming from outside the
Department. Almost 43 percent were internally generated. Many of the internally-generated cases
came from semi-automatic reporting systems, such as failure to appear in court (FTA) cases, which
accounted for slightly less than half of the internal referrals.

Most complaints (71 percent) were received by mail, phone, or email. Only 13 percent of
complaints were made in-person. This finding supports the on-going need to develop PIB’s
technological systems for complaint intake and tracking, as most cases are not walk-ins.

Complainant type (Q6) Internal External Anonymous

103.55 (42.95%) 133.5 (55.39%) 4 (1.66%)

Who received the complaint (Q7) OPR Employee Non-OPR Employee

146 (60.58%) 95 (39.42%)

5
We assigned a weighted value of .5 to double-reviewed cases in order to include them in the
overall sample, which explains the 103.5 cases meeting this category.
20
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 130 of 164

Who was the Non-OPR Other BPD Other BPD CRB Other
Employee who received Employee Supervisor
the complaint? (Q8) 32.5 (35.33%) 51(55.43%) 1.5 (1.60%) 8.5 (9.24%)

In person Not in-person (post Other

How was the complaint mail, phone, web etc.
received? (Q9) 18.5 (13.45%) 97.5 (70.91%) 21.5 (15.64%)

Monitoring Team reviewers were asked to generally characterize the facts of the complaint,
independent of how the case was categorized by BPD. Their free-text summaries of complaints
revealed a wide variety of allegations. Some of the most frequently identified complaints were:

 Dereliction of duty (not taking necessary action) was the most common complaint,
occurring in approximately 1/5 of all cases. Complainants said that officers were lazy, came
too quickly to conclusions, disregarded evidence, and were not just. This category captured
general dissatisfaction with service.

 Failure to appear in court was also very common, especially as the Monitoring Team
sampled “FTA” cases. These cases were typically automatically generated by the court
liaison and quickly adjudicated, with officers frequently admitting to the violation.

 Harassment complaints were relatively common. Complainants felt targeted by police,

either because they had multiple contacts, because they believed the interaction was unjust,
or because they believed they were profiled. Two cases alleged racial discrimination.

 Internal administrative complaints were common. This category included issues such as
leaving a post, losing identification or other equipment, exceeding permitted overtime,
reporting late, or questionable use of leave. It did not include integrity issues or direct
insubordination, which were alleged rarely.

 Use of force was the next most frequent, which is discussed in more detail below.

 Abuse of authority included claims such as running criminal or juvenile records for
personal reasons, possession of case materials for personal reasons, attempting to change
departmental systems to show an accident was non-preventable, or using status as an
officer for other benefits.

 Allegations of rudeness were common, ranging from a bad attitude to overt hostility.

21
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 131 of 164

 Body worn camera (BWC) violations were also common and typically surfaced as a
secondary complaint revealed during an investigation, although there were several stand-
alone cases where officers failed to record.

 Other, less frequent allegations included loss of property and retaliation.

COMPLAINANT DEMOGRAPHICS

Demographic information about individuals making misconduct complaints was not reliably
captured by OPR. Given that many of the Consent Decree requirements for the collection of such
information were not yet incorporated in policy or training at the time the complaints were made
and investigations were conducted, the lack of consistency and completeness is unsurprising.

Nevertheless, in some instances where information about a complainant’s gender, race, or ethnicity
were not captured, Monitoring Team reviewers could discern this information from available
evidence (e.g., by reviewing body-worn camera footage showing the complainant).

Most cases (91 percent) involved a single complainant. Distribution of complainant sex was almost
equal (48 percent female) and (41 percent male), with 11 percent unknown. The frequent inability
of reviewers to determine race (74 percent) or ethnicity (92 percent) from the files renders any
conclusions about demographics impossible. Moving forward, under new policies, new manuals,
and new training, PIB will need to consistently and clearly record demographic information to
allow for analysis.

(Q21) Whether at the outset of the complaint or Yes No

during the course of the investigation, was more than 10 (8.5%) 106 (91.38%)
one complainant identified?

(Q22) Complainant’s Sex Female Male Unknown

60 (47.81%) 51.5 (41.04%) 14 (11.16%)

(Q23) White/Caucasian Black/African- Hispanic/Latino Unknown

Complainant’s American
Race 8 (6.51%) 23 (18.69%) 1 (0.82%) 91 (73.98%)

(Q24) Hispanic/Latino Not Hispanic or Other Unknown

Complainant’s Latino
Ethnicity 0 8 (6.53%) 0 97 (92.38%)

CRIMINAL CASES AND CONSTITUTIONAL ISSUES

22
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 132 of 164

Few internal investigations involved associated criminal investigations: only three cases involved
any type of criminal investigation. Of those, one was an officer-involved shooting, for which the
State’s Attorney declined criminal charges; one involved officers indicted on federal charges
unrelated to the misconduct under administrative investigation; and one was an un-sustained
domestic violence referral from an outside jurisdiction. In all of these cases, the Monitoring Team
reviews gave the investigations an overall investigative quality rating of Very Good6, which may
imply that proper internal resources were directed at these potentially more serious cases.

(Q10) Was a criminal investigation of one of Yes No

more BPD members conducted that addressed 3 (1.24%) 238 (98.76%)
the complaint?

The Consent Decree requires the Monitoring Team to focus not just on misconduct investigations
generally but on some specific types of complaints – including “the frequency of civilian
complaints to OPR and CRB alleging unlawful arrests,” (Paragraph 459(c)(ii)); use of force
complaints, (Paragraph 459(d)(ii)); and complaints relating to First Amendment protected
activities, (Paragraph 459(j)(ii)). Overall, cases involving such complaints were relatively
infrequent among the cases that were reviewed: excessive force (7.47 percent, or 18 cases), false
arrest (2.28 percent or 5.5 cases), and First Amendment activities (1.24 percent or 3 cases).

(Q19) Did the complaint Yes No UTD Agree

involve allegations, whether
made originally by the
complainant or uncovered
during the investigation,
relating to:
Use of Force (e.g., excessive 18 (7.47%) 219 (90.87%) 4 (1.66%) 38/40
force or failure to report (95.00%)
force)?
First-Amendment-protected 3 (1.24%) 235 (97.51%) 3 (1.24%) N/A7
activity?
Unlawful/False Arrest? 5.5 (2.28%) 231.5 (96.06%) 4 (1.66%) 39/40
(97.5%)

Of the use of force cases, only one had overlap with false arrest; none overlapped with First
Amendment allegations. None of the use of force allegations were sustained by the Department.

6
Question 92 asked reviewers to rate the overall investigation on a scale of Poor, Fair, Good, Very
Good, and Excellent.
7
There were no double-reviewed cases with First Amendment allegations.
23
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 133 of 164

The Monitoring Team assigned overall investigative quality ratings of Good, Very Good, or
Excellent to the investigations involving use of force.

The Consent Decree requires that every case result in a particular disposition that is accompanied
by a standard of proof. Dkt. No. 2 at ¶344(k):

i. “Unfounded,” where the investigation determines, by clear and

convincing evidence, that the alleged misconduct did not occur or did
not involve the accused officer;
ii. “Sustained,” where the investigation determines, by a preponderance of
the evidence, that the alleged misconduct did occur;
iii. “Not Sustained,” where the investigation is unable to determine, by a
preponderance of the evidence, whether the alleged misconduct
occurred;
iv. “Exonerated,” where the investigation determines, by a preponderance
of the evidence, that the alleged conduct did occur but did not violate
BPD policies, procedures, or training;
v. BPD shall discontinue use of the disposition “administratively closed.”

The Monitoring Team found that the evidence in all but one of the use of force cases met the
required standard of proof to support the conclusion – and the evidence in the one case that failed
to meet that standard was solely due to the investigator not addressing officer rudeness; in other
words, it was not related directly to the force allegation. Given that the Monitoring Team found
that the conclusions in only 61 percent of all cases were supported by evidence that met the
pertinent standard of proof, the use of force cases ranked comparatively well.

The demographic breakdown of complainants in use of force cases does not allow for the ready
understanding of either who brings force complaints or whether there are sharp discrepancies in
how force complaints are addressed based on race. The race of only two complainants (both
African-American) could be identified; the race of complainants in the remaining cases was
unknown.

Of the three First Amendment cases, none of the cases involved direct action against any person
participating in First Amendment activities. For example, one was an internal complaint that
officers had wrongfully accused a Major of not lowering the American flag in memoriam of
President George Bush. Allegations were not sustained in one case and were determined to be
unfounded in two other cases. The Monitoring Team rated two investigations as Very Good
overall. One was rated Fair, primarily due to a poor intake focused on technical issues rather than
the spirit of the complaint (complainant wasn’t the owner of the car that was towed).

24
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 134 of 164

Of the five false arrest cases, one was judged exonerated, three were unfounded, and one was not
sustained by the Department. The Monitoring Team found that the evidence in all five cases met
the requisite standard of proof. In terms of overall investigative quality, two were ranked Very
Good, two were Good, and one was Fair. With respect to the case ranked Fair, the investigative
issues giving rise to the ranking were unrelated to the determination of whether a false arrest had
occurred. In that instance, a complainant’s daughter had admitted to kicking out the windows of a
car, which was sufficient to establish probable cause for the involved arrest.

REVIEW/CLASSIFICATION/TIMELINES

Intake and classification of complaints is critically important to ensure proper, consistent

Departmental responses to complaints and proper recognition of the severity/complexity of
complaints. Appropriate intake and classification help guarantee that similar complaints will be
handled in a similar manner, which supports the internal and external legitimacy of the process –
again, the goal is that the accountability systems are perceived by all as fair.

The Department of Justice identified systemic problems with intake and classification in their
findings letter:

BPD lacks meaningful accountability systems to deter misconduct. BPD does not
consistently classify, investigate, adjudicate, and document complaints of
misconduct according to its own policies and accepted law enforcement standards.

Findings Letter at 128.

The Consent Decree has significant requirements governing Intake, Classification, and
Communication with Complainants, including revising the relevant policy and manual and
ensuring that the complaint process is open and accessible, that complaints may be made
anonymously and through any common means of communication, that all complaints are accepted
and properly classified according to the allegations, and that all classifications and changes to
classification are tracked in a database. See Dkt. No. 2 at ¶¶ 335–42.

The PIB Intake and Classification Manual has been updated to ensure that PIB will accept
anonymous complaints, that complaints will be accepted by email, phone, mail, or walk-in, and
that allegations will be timely classified in IA Pro (the designated computer software program used
for internal affairs investigations) and based on the options in that system. From the moment a
complaint is received, there is now clarity about how it will be categorized and managed. Strict
adherence to these new policies will help with consistency and accuracy of classification.

25
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 135 of 164

However, the investigations reviewed in this assessment occurred prior to the finalization of the
new manual. Consequently, this baseline review essentially grades the Department against
standards, expectations, and requirements that they were not yet trying to meet in 2018. For that
reason, the Monitoring Team identified a number of core problems with the classification of new
complaints and adherence to Consent Decree-required investigative timelines.

In some instances, then, the Monitoring Team was able to definitively conclude that BPD had not,
in fact, adhered to the Consent Decree’s requirements for intake and classification. In an even
greater number of instances, however, because the investigative files were often chaotic – lacking
uniform format or structure or missing important information – Monitoring Team reviewers could
not determine, one way or another, whether various requirements had been met.

(Q25) Review & Yes No UTD Agree

Classification
Initial classification
determination made within 119 (49.38%) 65.5 (27.18%) 56.5 (23.44%) 20/40
72 hours of OPR being (50.0%)
notified of allegation of
misconduct?
Complaint assigned to OPR
investigator within 72 hours 101.5 (42.12%) 98.5(40.87%) 41 (17.01%) 21/40
of OPR being notified of (52.5%)
allegation of misconduct?
All appropriate allegations
identified and listed, 166.5(69.09%) 46.5 (19.29%) 28 (11.62%) 21/40
including those that, if true, (52.5%)
would violate BPD policy
but are not affirmatively
identified by the complaint
or complainant?
All relevant policy/manual
sections accurately 144 (59.75%) 66.5 (27.59%) 30.5 (12.66%) 32/40
identified and listed? (80.0%)
Most serious policy
violation appropriately used 162 (67.22%) 39 (16.18%) 40 (16.60%) 34/40
for purposes of (85.0%)
classification?
Most serious policy
violation appropriately used 160 (66.39%) 36.5 (15.15%) 44.5 (18.46) 35/40
(87.5%)

26
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 136 of 164

to determine whether OPR

will investigate?
Alpha (α ) 0.855

Only about half (49 percent) of cases were timely classified within 72 hours, as required by the
Consent Decree. Somewhat less than half (42 percent) were assigned to an investigator within 72
hours of an investigator being notified of misconduct. At the intake stage, more than two-thirds
(69 percent) of cases had all allegations properly identified. Sixty percent of cases properly
identified the specific BPD policies that were implicated by the complaint. In more than two-thirds
(67 percent) of cases, the most significant violation was appropriately used to classify the
investigation.

The relatively low agreement between reviewers for the questions about timeliness suggests that
the number of cases that complied with Decree requirements may actually have been lower.
Because the case files were in such disarray, it appears that reviewers used different dates, found
in different parts of the file, as plausible “start dates” from which to chart timelines. As such, the
Monitoring Team does not have strong confidence in the specific results presented in the first three
rows above, but the overall concept – that BPD was not clearly recording timelines, investigators,
or capturing all appropriate allegations – is wholly consistent with the reviewers’ experiences
grappling with the case files. Going forward, PIB will need to ensure that the dates of core
administrative actions are reliably captured and uniformly documented in a standard format.

In contrast, there was solid agreement between reviewers about whether all relevant policy
violations were identified and whether the most serious allegation was used for classification
purposes.

The items that measured intake and classification have a reasonably good alpha reliability
coefficient (0.855). Thus, we can comfortably conclude that, for intake and classification, the
evaluation instrument is measuring what we are trying to measure.

Communication with Complainants

To ensure transparency, the Consent Decree contains requirements that complainants receive
periodic updates from the Department. These requirements are not directed at internal complaints,
but at external complaints from members of the public in order to provide good customer service
and improve legitimacy. As such, the sample of cases examining communication with
complainants is limited to those cases identified by the Monitoring Team as “external.”8 Within

8
This should not be confused with BPD’s use of “external” as an incident type. As previously
discussed, other incident types, such as CIU, could start with a complaint by a member of the
public. External cases here were identified by the Monitoring Team.
27
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 137 of 164

these cases, communication was poor overall. OPR provided complainants with written notice that
it had received their complaints in only a quarter (27 percent) of cases. Likewise, reviewers were
able to find evidence that OPR provided complainants with periodic updates about the status of
their cases in only 36 percent of cases. Because of the low number of overall cases in which
complainants received communications from BPD, the Monitoring Team found that
communications were professional and respectful in just 38 percent of cases.

Inter-rater reliability was approximately 60% agreement across this question set. Although the
alpha was lower (0.64) than other sections, this is likely because the sequence of questions about
complainant communication combined pure auditing questions (“was there notice within 7 days?”)
with more qualitative ones (“were communications with complainant professional and
respectful?”) The Monitoring Team will consider the workflow of the data gathering instrument
in future compliance reviews.

(Q26) Yes No UTD Agree

Notification/Communication
with Complainant(s)
Written notice of receipt 36 (26.97%) 76.5 (57.30%) 21 (15.73%) 12/18
within 7 days of OPR’s receipt (66.67%)
of a complaint?
OPR mail or email periodic 48 (35.96%) 68 (50.94%) 17.5 12/18
updates to complainant on (13.11%) (66.67%)
status of
complaint/investigation?
Communications with 50.5 (37.97%) 17.5 (13.16%) 65 (48.87%) 11/18
complainant professional and (61.11%)
respectful?
α = 0.641

Of the relatively few cases that included communications with the complainant, only slightly more
than half (53 percent) of the communications included basic information like the tracking number,
the case number, and the allegations being investigated. Fifty-four percent of communications
informed complainants how they could follow up to find out case status. As with most of the
review results around technical or timing questions, the state of the case files led to some
disagreement among reviewers, albeit with a sample size below the reporting threshold. As such,
the below results are better interpreted as thematic than as precise quantifications of the files. The
Monitoring Team will expect to see improvement in this area as BPD adopts new policies and
procedures.

28
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 138 of 164

For this set of questions, there was a moderate level of disagreement between reviewers in the
double-reviewed cases, which again is likely due to confusion over where in the files this
information could be found and/or internal inconsistency within the files.

The questions that measured communication with the complainant have an acceptable alpha value
of 0.782. Thus, we can conclude that these questions adequately measure the communication with
the complainant.

The overall conclusion here is that communications with complainants need to be greatly improved
for consistency and content, but investigators are not discouraging complainants from making or
pursuing complaints.

(Q27) Details re: Written Notice to Yes No UTD

Complainant
Did receipt include tracking number, 18.5 (52.86%) 3.5 (10.00%) 13 (37.14%)
other relevant case numbers, and
allegations being investigated?
Did notice inform complainant how 19 (54.29%) 2.5 (7.14%) 13.5 (38.57%)
he/she may inquire about complaint
status?
Did notice contain any language that 0 22 (62.86%) 13 (37.14%)
could reasonably be construed as
discouraging participation in the
investigation (such as warning
against providing false statements or
a deadline by which complainant
must contact the investigator)?
α = 0.782

Communication with Involved Officers & Employees

Timely notification to the subject officer only occurred in a quarter (24 percent) of cases, though
interrater reliability scores were low for this question. Notice to supervisors of involved officers
or employees occurred only one out of ten times (12 percent of cases). The Decree’s requirement
that an officer’s supervisor facilitate an officer’s appearance before investigators and document
such facilitation appeared to be followed in only 12 cases – or about 5 percent of the time overall.

(Q28) Communication Yes No UTD Agree

with Involved
Officer(s)/Employee(s)

29
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 139 of 164

Timely notification to all 57.5 (23.86%) 114.5 (47.51%) 69 (28.63%) 23/40

involved employee(s)? (57.5%)
Timely notification to
supervisor of officer(s) 29.5 (12.24%) 92 (38.17%) 119.5 (49.59%) 32/40
under investigation? (80.00%)
Officer’s supervisor(s)
facilitating officer’s 12 (4.98%) 101.5 (42.12%) 127.5 (52.90%) 36/40
appearance documented in (90.0%)
writing?
α = 0.762

Where there was a notification to an employee, regardless of whether it was provided on the
Decree-required timeline, it was not found to jeopardize the investigation (such as by including
too many facts prior to an interview). Notices comported with due process and the law. Only one
case involved a notice containing information that might have unnecessarily jeopardized the
investigation.

Reviewers determined that fewer than 8 percent of notices prohibited officers from speaking to
witnesses, reviewing reports, or taking other actions that could jeopardize the investigation.

(Q29) Details re: Initial Yes No UTD Agree

Notification to
Officer/Employee(s)
(N=57.5 from Q28)
Did notice jeopardize the 0 51.5 (89.57%) 6 (10.43%) 3/3
investigation? (100.00%)
Did officer only receive notice
prior to being formally 17.5 (30.70%) 26 (45.61%) 13.5 (23.68%) 1/3
interviewed by OPR, and not (33.33%)
earlier?
Did notice comport with due 44 (76.52%) 2 (3.48%) 11.5 (20.00%) 3/3
process and the law? (100.00%)
Did the notice contain the 43.5 (75.65%) 4.5 (7.83%) 9 (15.65%) 2/3
nature of the investigation? (66.66%)
Did the notice contain any
information that might have 1 (1.77%) 47.5 (84.07%) 8 (14.16%) 2/3
unnecessarily jeopardized the (66.66%)
investigation?
Did notice prohibit officers
from speaking to 4.5 (7.83%) 38 (66.09%) 15 (26.09%) 3/3

30
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 140 of 164

witnesses/complainants, (100.00%)
reviewing police reports,
reviewing body camera footage,
or taking other actions that
could jeopardize the
investigation.
α = 0.762

Investigative Timeliness

The Consent Decree requires that internal investigations be completed within 90 days of the
initiation of the investigation unless an extension is specifically requested and approved, in
writing, by the OPR Director. Dkt. No. 2-2 at 122.

The state of BPD investigative files, which often featured various and sometimes conflicting dates,
made determining the date on which investigations started difficult. Fewer than two out of five (39
percent) investigations were completed within the Decree-required 90-day timeframe, with
extensions virtually never sought.

(Q70) Regarding Yes No UTD Agree

investigative timeliness
Investigation completed 95 (39.42%) 142.5 (59.13%) 3.5 (1.45%) 38/40
within 90 days of initiation of (95.00%)
the investigation?
Extension requested at any 0.5 (0.21%) 181.5 (75.31%) 59 (24.48%) 25/40
point during the (62.50%)
investigation?

INVESTIGATION: COMPLAINANT INTERVIEW & INJURIES

The Consent Decree requires that reasonable efforts be made to identify and interview any “civilian
complainant or witnesses in person at a time and place that is convenient and accessible for the
complainant or witness.” Dkt. No. 2-2 at ¶ 343. Additionally, audio or video recording shall occur.
Id. at ¶344(d).

Here, unlike the previous section, the Monitoring Team did not restrict the analysis of complainant
interviews to external cases because the complainant in internal cases generally should be
interviewed as part of a quality investigation unless there are reasons not to do so. FTA cases could
be considered as an exception. Such cases are essentially auto-generated when an officer fails to
appear in court, and the record is completed with computer tracking – such that there is not

31
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 141 of 164

necessarily a designated individual who is the source of the complaint. Consequently, there is no
reason to interview the supervisor who administratively referred the complaint. In contrast, if an
employee makes a misconduct referral on another employee, the complainant employee will need
to be interviewed even though the referral is internal. Nevertheless, because the terms of the
Consent Decree generally address only instances where there is an individual – either a member
of the public or member of the Department – identified as the “complainant,” this report includes
FTA cases within the results regarding complainant interviews. The Monitoring Team expects that
the specific protocols that the Department is adopting via misconduct investigation policies and
manuals will help to clarify the appropriate mechanisms for addressing the issue surrounding the
nature of complainants and the utility of interviewing them in FTA cases going forward.

Complainant interviews occurred in only 41 percent of cases. Of cases where no interviews were
conducted, some involved the failure of a complainant – sometimes repeatedly – to appear for
scheduled interviews or to respond to investigator inquiries. For many others, however, there
simply was no evidence that investigators attempted to conduct an interview. At first glance, it is
counterintuitive that reviewers only agreed about whether there was a complainant interview in 26
of 40 cases. On a straightforward and critical question like “was the complainant interviewed,”
generally the Monitoring Team would want to see higher agreement between reviewers. However,
again, the state of the files was wholly inconsistent. There were many cases in which the
investigator referred to a complainant interview, but there was no evidence (transcript, recording,
or even analysis) that the interview actually occurred or was incorporated into the findings. This
sort of inconsistency in the case files led to somewhat inconsistent results by reviewers. The overall
takeaway, that BPD needs to interview all complainants when possible and clearly document the
information gathered from the interview, remains clear.

Where interviews were conducted, only about half of them were confirmed to be audio recorded.
The consistent failure to put either a transcript or audio recording into the case record, even when
an interview was said to have been conducted, compromised the ability of reviewers to make
thorough assessments of the quality of the interviews. Video recording was only used in three
cases. No case suggested a need for translation or accommodations.

(Q31) Complainant Yes No UTD Agree

Interview9

9
Question 31, sub-question 7 asked whether there was a reported injury in the complainant’s
interview and Q35 asked whether photographs of the injury were taken, or a medical release
requested. As these questions were duplicative of Q66 and Q67, which more appropriately
explored injuries in the context of the entire case, rather than in the specific context of the
interview, the Monitoring Team removed these questions. Sub-question 7 was also misaligned
32
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 142 of 164

Complainant 99 (41.08%) 136 (56.43%) 6 (2.49%) 26/40

interviewed? (65.00%)
Interviews occurred at 6/8
time/place convenient and 88.5 (36.72%) 99 (41.08%) 53.5 (22.20%) (75.00%)
accessible for witness?
Permission requested to 36 (14.98%) 140.5 (58.30%) 64.5 (26.76%) 6/8
record interview? (75.00%)
Interview video- 3 (1.24%) 201 (83.40%) 37 (15.35%) 8/8
recorded? (100.00%)
Interview audio- 43.5 (18.05%) 154.5 (64.11%) 43 (17.84%) 6/8
recorded? (75.00%)
Was there a request or 5/8
apparent need for 0 223.5 (92.74%) 17.5 (7.26%) (62.50%)
translation or
accommodation?
Were reasonable steps 5/8
taken to gather evidence, 136.5 (56.64%) 75.5 (31.33%) 29 (12.03%) (62.50%)
documents in a timely
manner?
Was there appropriate 26/40
focus on retrieving 63.5 (26.35%) 69 (28.63%) 108.5 (45.02%) (65.00%)
perishable items?
Were all contacts logged? 148 (61.41%) 45 (18.67%) 48 (19.92%) 6/8
(75.00%)
Was complainant 6/8
interview thorough and 51 (21.16%) 77.5 (32.16%) 112.5 (46.68%) (75.00%)
unbiased?
α = 0.780

Although reviewers were unable to determine whether permission to record an interview was
requested in 27 percent of cases, and confirmed that no permission was requested in 58 percent of
cases, when permission was in fact requested, complainants consented in 97 percent of cases—all
but one. It is not clear from the record why the one complainant refused to grant permission, as the

with the other sub-questions in Question 31, as it asked about content, not process. The alpha was
significantly improved by removing the sub-question.
33
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 143 of 164

refusal was not documented or a summary generated, but it does not generally appear that
complainants object to having their interviews recorded.

Yes No
(Q32) Did complainant grant permission to record the
interview? (N=36 from Q31) 35 (97.14%) 1 (2.86%)
(Q34) Re: the request or apparent need for translation
or accommodation, was such 0 0
translation/accommodation provided?

When complainant interviews were conducted and appropriate documentation included in the case
file, Monitoring Team reviewers found those interviews generally adequate. When identifying
specific deficiencies in investigations, reviewers could select one or more possible choices.
Overall, in 90 percent of cases where an interview was conducted and there was sufficient
documentation to reach conclusions, reviewers identified no significant deficiencies. Nine cases
(9 percent of cases with complainant interviews) had one deficiency; and two cases (2 percent of
cases with complainant interviews) had three deficiencies listed.

Reviewers consistently flagged the cases that were deficient, even if they gave slightly different
reasons for the inadequacies. For example, the only case that identified “Leading Questions” also
included “Appearance of Bias” and “Relevant questions left unanswered.” That case was double-
reviewed and the second reviewer chose “Leading Questions” only. Nevertheless, both reviewers
flagged deficiencies. In that case, the second reviewer was left with the impression that the
investigator was attempting to discourage the complainant from continuing with her complaint
(she was undeterred). Discouraging a complainant – which was identified as a serious issue in the
Department of Justice Findings Letter – could be described in many different ways. However, the
Monitoring Team is confident that deficiencies were flagged, even if the basis differed.

Monitoring Team reviewers found that questioning reflected bias in three cases. In one case, the
complainant raised a concern about the assigned investigator having a conflict of interest with the
complainant. The complainant and the investigator were interviewed about that allegation and then
the case was administratively closed the same day with little explanation.

(Q37) Indicate any deficiencies or issues identified during each complainant interview
Leading questions noted 1
Inadequate questioning 5
Appearance of bias 3
Inconsistencies not addressed 0
Relevant questions left unanswered 5
Concerns about investigator demeanor 1

34
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 144 of 164

INVESTIGATION: WITNESSES: SUBJECT/EMPLOYEE INTERVIEW

The subject employee was identified by the complainant in 66 percent of cases. Where the
employee was not specifically identified in the complaint, investigators made all reasonable efforts
to identify the subject employee about 62 percent of the time.

When subject employees were identified by the complainant, they were interviewed in only 44
percent of investigations. As officers rarely declined to be interviewed – that occurred only once
across reviewed cases – the Monitoring Team’s hypothesis is that the failure to interview was the
investigator’s decision.

Yes No Agree
(Q38) Did the complainant 158 (65.56%) 83 (34.44%) 33/40
specifically identify the subject (82.50%)
officer’s name(s)?
(Q39) Did investigator(s) make all 51.5 (62.05%) 31.5 (37.95%)
reasonable efforts to identify the 5/7
officer(s)? (N=83 from Q38) (71.43%)
(Q40) Involved employee(s) 106.5 134.5 27/40
interviewed? (44.19%) (55.81%) (67.50%)

(Q42) Regarding the employee not Yes No UTD Agree

being interviewed (N=134.5 from
Q40)
Did employee(s) decline to be 1 (0.75%) 104 (78.49%) 27.5 14/14
interviewed? (20.75%) (100.00%
)
Could employee(s) be 78.5 (59.25%) 33.5 (25.28%) 20.5 9/14
identified/located? (15.47%) (64.29%)

Where interviews of officers occurred, reviewers found that more than half (59 percent) were
adequate in that investigators reasonably pursued all relevant lines of investigative inquiry.

Bias (less than two percent of interviews), leading questions (four percent), actual contamination
(less than one percent), and evidence of pre-interview questioning (less than two percent) were not
common. OPR interviewers infrequently inquired into the possibility of witness contamination
(only 10 percent of interviews). In nine of the ten interviews where investigators inquired about
contamination, they also were found to have properly controlled for attorney interference –

35
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 145 of 164

suggesting that some interviewers, in some cases, were able to take the type of appropriate
investigative precautions necessary under the Consent Decree.

(Q41) Details of Yes No UTD Agree

Subject/Employee Interviews
(N=106.5)
All relevant lines of
investigative inquiry 63 (59.43%) 15 (14.15%) 28 (26.42%) 8/13
reasonably and adequately (61.54%)
pursued?
Possible bias noted (e.g.,
automatic preference for 2 (1.90%) 70.5 (67.14%) 32.5 (30.95%) 12/13
officer, disregard of (92.31%)
complainant based
on criminal history or guilty
plea)?
Leading questions used in the 4.5 (4.27%) 63 (59.72%) 38 (36.02%) 11/13
interview(s)? (84.62%)
Evidence of potential
contamination of employee 1 (0.94%) 73 (68.87%) 32 (30.19%) 12/13
accounts? (92.31%)
Interviewer inquiry as to
potential contamination of 10.5 (10.00%) 55 (52.38%) 39.5 (37.62%) 10/13
employee accounts? (76.92%)
Evidence suggesting that pre-
interview 2 (1.90%) 75.5 (71.90%) 27.5 (26.19%)
questioning/discussion 12/13
occurred prior to the (92.31%)
interview?
Interviewer prevent union
representation from disrupting 17.5 (16.51%) 51 (48.11%) 37.5 (35.38%) 9/26
interview(s)? (69.23%)
α = 0.951

(Q43) Indicate any deficiencies or issues identified during each subject/employee

interview
Leading questions noted 2
Inadequate questioning 7
Appearance of bias 1
Inconsistencies not addressed 1

36
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 146 of 164

Relevant questions left unanswered 9

Concerns about investigator demeanor 0

INVESTIGATION: CIVILIAN WITNESSES

The Consent Decree requires that reasonable efforts be made to identify and interview all witnesses
and interview any “civilian complainant or witnesses in person at a time and place that is
convenient and accessible for the complainant or witness.” Dkt. No. 2-2 at ¶343. Additionally,
audio or video recording shall occur. Id. at ¶344(d).

Efforts to identify, locate, and interview civilian witnesses were not uniform in those cases where
civilian witnesses were implicated, occurring in less than half (48 percent) of cases where experts
determined that the efforts would have been applicable. Of those cases where civilian witnesses
were identified, the civilians were interviewed only 38 percent of the time.

When civilians were interviewed, interviews were generally conducted at a time and place
convenient for the civilian witness (78 percent), typically audio-recorded (65 percent), and
generally adequate (78 percent). Remarkably, reviewers found no instances of bias, potential
contamination, or evidence of any pre-interview questioning. Leading questions only surfaced in
two cases. The main takeaway is that civilian interviews need to be pursued in all cases where
such witnesses are identified, and that the record keeping of those interviews needs to be improved
significantly.

(Q45) Effort to Contact Yes No UTD N/A

Civilian Witnesses
Sustained and reasonable 62 (25.73%) 58 (24.07%) 10 (4.15%) 111 (46.06%)
efforts made to contact and
interview witnesses
All witness contact efforts 70 (19.05%) 41.5 (17.22%) 21.5 (8.92%) 108 (44.81%)
logged?

Yes No Agree
(Q47) Were any civilian witnesses
identified or reasonably implicated by 43.5 (18.05%) 197.5 27/40
the complaint, canvass, or (81.95%) (67.50%)
investigation?
(Q48) All identified civilian witnesses 16 (37.65%) 26.5 (62.35%) 1/ 2
interviewed? (50.00%)

37
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 147 of 164

(Q49) Details re: Civilian Yes No UTD Agree

Witness Interviews
(N=16 from Q48)
Interviews occurred at 12.5 (78.13%) 0 3.5 (21.88%) 1/2
time/place convenience and (50.00%)
accessible for witness?
Interviews video-recorded? 0.5 (3.13%) 11.5 (71.88%) 4 (25.00%) 1/2
(50.00%)
Interviews audio-recorded? 10.5 (67.74%) 2 (12.90%) 3 (19.35%) 0/2
All relevant lines of
investigative inquiry 12.50 (78.13%) 0 3.5 (21.88%) 0/2 (0%)
reasonably and adequately
pursued?
Possible bias noted? 0 11.5 (74.19%) 4.5 (25.81%) 1/2
(50.00%)
Leading questions used in 1.5 (9.38%) 10.5 (65.63%) 4 (25.00%) 1/2
the interview? (50.00%)
Evidence of potential
contamination of civilian 0 14 (87.50%) 2 (12.50%) 1/2
witness accounts? (50.00%)
Interviewer inquiry as to
potential contamination of 0 10.5 (65.63%) 5.5 (34.38%) 1/2
employee accounts? (50.00%)
Evidence suggesting that
pre-interview 0 13 (81.25%) 3 (18.75%) 1/2
questioning/discussion took (50.00%)
place prior to the interview?
All witness interviews
sufficiently memorialized for 12.5 (78.13%) 1.5 (9.38%) 2 (12.50%) 0/2
purposes of a full and (0%)
complete investigative file?
α = 0.851

(Q51) Indicate any deficiencies or issues identified during each civilian witness interview
(N=16 from Q48)
Leading questions noted 2
Inadequate questioning 0
Appearance of bias 0
Inconsistencies not addressed 0

38
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 148 of 164

Relevant questions left unanswered 0

Concerns about investigator demeanor 0

INVESTIGATION:
BPD OFFICER/EMPLOYEE WITNESS INTERVIEWS

The Consent Decree requires that:

BPD shall require its employees to cooperate with administrative investigations,

including appearing for an administrative interview when requested by a BPD
investigator, and providing all relevant documents and evidence under the person’s
custody and control.

Id. at ¶ 347(d). Additionally, the Decree prohibits interviews of BPD employees from using leading
questions and from discouraging the provision of a full account, and requires that BPD employee
interviews be thorough and un-biased. Id. at ¶ 349.

BPD employees were implicated as possible witnesses in about one-third (35 percent) of cases. Of
those, the employees were interviewed 61 percent of the time. As most of the employees identified
could be located (62 percent) and none refused to testify, it again is concerning that investigators
decided not to regularly interview such witnesses.

Audio recording was the most common means of capture at 74 percent of cases; about 4 percent
were video recorded. However, as many of the audio recordings or transcripts were not included
in the case files, reviewers frequently were unable to make determinations about the quality of
interviews.

Most interviews of BPD employee witnesses were generally sound. No significant deficiencies
were noted in 94 percent of cases where civilians were interviewed.

Yes No UTD Agree

(Q53) Were any BPD
officer/employee witnesses identified 84 (34.85%) 157 (65.15%) 0 32/40
or reasonably implicated by the (80.00)
complaint or investigation?
(Q54) Were the employee(s) 51.5 (61.31%) 32.5 (38.69%) 0 8/9
interviewed? (88.89%)

(Q55) Regarding the employee(s) not Yes No UTD Agree

being interviewed (N=32.5 from Q54)

39
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 149 of 164

Employee(s) decline to be interviewed? 0 21.5 (66.15%) 11(33.85%) 3/3

(100.00%)
Employee(s) able to be 20 (61.54%) 6.5 (20.00%) 5.5 3/3
identified/located? (16.92%) (100.00%)

(Q56) Regarding Interview(s) of Yes No UTD Agree

Involved Officer(s)/Employee(s)
(N=51.5 from Q54)

Interviews video-recorded? 2 (3.96%) 39.5 (78.22%) 9 (17.82%) 5/5

(100.00%)
Interviews audio-recorded? 37 74.00%) 5 (10.00%) 8 (16.00%) 4/5
(80.00%)
All relevant lines of investigative
inquiry reasonably and adequately 32.5 (63.11%) 4.5 (8.74%) 14.5 3/5
pursued? (28.16%) (60.00%)
Possible bias noted (e.g., automatic
preference for officer, disregard of 1 (1.94%) 36.5 (70.87%) 14 (27.18%) 5/5
complainant based on criminal history (100.00%)
or guilty plea)
Leading questions used in the 1 (1.94%) 34.5 (66.99%) 16 (31.07%) 5/5
interview? (100.00%)
Evidence of potential contamination of 1 (1.94%) 36.5 (70.87%) 14 (27.18%) 5/5
employee accounts? (100.00%)
Interviewer inquiry as to potential
contamination of 4.5 (8.82%) 30 (58.82%) 16.5 3/5
employee accounts? (32.35%) (60.00%)
Evidence suggesting that pre-interview
questioning/discussion took place prior 0 36 (70.59%) 15 (29.41%) 4/5
to the interview? (80.00%)
Interviewer prevent union
representation from disrupting 8 (16.16%) 24 (48.49%) 17.5 3/5
interview (35.35%) (60.00%)
α = 0.933

(Q58) Indicate any deficiencies or issues identified during the interview of

Officer(s)/Employee(s) (N=51.5 from Q54)
Leading questions noted 1
Inadequate questioning 2

40
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 150 of 164

Appearance of bias 0
Inconsistencies not addressed 0
Relevant questions left unanswered 1
Concerns about investigator demeanor 0

INVESTIGATION: WITNESSES: LAW ENFORCEMENT (NON-BPD) WITNESS

INTERVIEWS

Non-BPD law enforcement witnesses were identified as potential witnesses in fewer than 10
percent of cases. Those witnesses were interviewed about one-third of the time. This meant that
only eight cases had non-BPD law enforcement witness interviews.

Most (81 percent) of non-BPD law enforcement witness interviews reasonably pursued all relevant
lines of investigative inquiry, and there was no evidence of bias, leading questions, or
contamination. One case showed evidence that there was pre-interview questioning. No non-BPD
law enforcement witness refused to be interviewed.

Yes No Agree
(Q60) Were any non-BPD law enforcement
witnesses identified or reasonably 23.5 (9.75%) 217.5 (90.25%) 35/40
implicated by the complaint or (87.50%)
investigation?
(Q61) All identified non-BPD law
enforcement witnesses interviewed? (N=23.5 8 (34.04%) 15.5 (65.96%)
from Q60)

(Q62) Non-BPD Law Enforcement Yes No UTD

Interview Details (N=8 from Q61)

Interviews video-recorded? 0 7 (87.50%) 1 (12.50%)

Interviews audio-recorded? 7 (87.50%) 0.5 (6.25%) 0.5 (6.25%)
All relevant lines of investigative inquiry
reasonably and adequately pursued? 6.5 (81.25%) 0 1.5 (18.75%)
Possible bias noted 0 6.5 (81.25%) 1.5 (18.75%)
Leading questions used in the interview? 0 6.5 (81.25%) 1.5 (18.75%)
Evidence of potential contamination of
witness accounts? 0 6.5 (81.25%) 1.5 (18.75%)
Interviewer inquiry as to potential
contamination of employee accounts? 0 5.5 (68.75%) 2.5 (31.25%)

41
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 151 of 164

Evidence suggesting that pre-interview

questioning/discussion took place prior to 1(13.33%) 5 (66.67%) 1.5 (20.00%)
the interview?

(Q64) Indicate any deficiencies or issues identified during the interview of law enforcement
(non-BPD) Witness (N=8 from Q61)
Leading questions noted 0
Inadequate questioning 0
Appearance of bias 0
Inconsistencies not addressed 0
Relevant questions left unanswered 0
Concerns about investigator demeanor 0
Other (write in) 1

INVESTIGATION: PHYSICAL EVIDENCE, DOCUMENTATION OF INJURIES

Complaints of injury or apparent injuries occurred in 6.4 percent of cases. The Monitoring Team
confirmed that there were no injuries reported in 92 percent of cases—the vast majority. This
includes cases where use of force by a BPD officer was at the core of the complaint, as well as
cases where the complaint was about something else but an injury nonetheless occurred during the
interaction at issue. Of the cases involving use of force-related allegations, more than half (56
percent) involved some complaint of injury by the complainant.

In the few cases that involved a complainant injury, the injuries were documented in about three-
quarters of cases, photographs were taken about two-thirds of the time, and medical records
requested about half of the time. It is unclear why this information was not included, but moving
forward, PIB will need to document subject injuries more consistently and/or clearly indicate why
investigators were unable to do so.

Officer injuries were involved in only two cases, both of which were documented. Importantly,
neither case involved injury inflicted by a suspect.

 A supervisor was attempting to cut a suspect out of a locked patrol car – the officer
locked the keys inside – and was injured in the process.

 An officer was attacked by a dog responding to a call. The dog was shot.

Yes No UTD
15.5 (6.43%) 222.5 (92.32%) 3 (1.24%)

42
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 152 of 164

(Q66) Was there a complaint of injury,

or a readily apparent injury?
(Q68) Were officers injured? 2 (0.83%) 236 (97.93%) 3 (1.24%)
(Q69) Were officer’s injuries
documented? 2 (100%) 0 0

(Q67) Regarding the complainant of Yes No UTD

injury
Complainant injuries documented 12 (77.42%) 2.5 (16.13%) 1 (6.45%)
Photographs taken? 9.5 (61.29%) 5 (32.26%) 1 (6.45%)
Medical records release requested? 8 (51.61%) 4.5 (26.03) 3 (19.35%)

INVESTIGATIVE REPORT

The Consent Decree, paragraph 351, clearly lays out the requirements for investigative reports for
misconduct investigations:

At the conclusion of each investigation, misconduct investigators will prepare an

investigation report. The report will include:

a. A narrative description of the incident, including a precise description

of the evidence that either justifies or fails to justify the officer’s conduct
based on the misconduct investigator’s independent review of the facts
and circumstances of the incident;
b. Documentation of all evidence that was gathered, including names,
phone numbers, and addresses of witnesses to the incident. In situations
in which there are no known witnesses, the report will specifically state
this fact. In situations in which witnesses were present but
circumstances prevented the misconduct investigator from determining
the identification, phone number, or address of those witnesses, the
report will state the reasons why. The report also will include all
available identifying information for anyone who refuses to provide a
statement;
c. Documentation of whether officers or other BPD employees were
interviewed, including audio and video and a transcript of those
interviews, if available;
d. The names of all other BPD employees who witnessed the incident;
e. The misconduct investigator’s evaluation of the incident, based on his
or her review of the evidence gathered, including a determination of
whether the officer’s actions appear to be within BPD policy, procedure,

43
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 153 of 164

regulations, orders, or other standards of conduct required of BPD

officers;
f. In cases where credibility determinations must be made, explicit
credibility findings, including a precise description of the evidence that
supports or detracts from the person’s credibility;
g. In cases where material inconsistencies must be resolved between
complainant, officer, and witness statements, explicit resolution of the
inconsistencies, including a precise description of the evidence relied
upon to resolve the inconsistencies;
h. If a weapon was used, documentation that the officer’s certification and
training for the weapon were current; and
i. Documentation of recommendations for non-punitive corrective action
or misconduct charges.

Reviewers found that, in 72 percent of cases, the investigative report narrative included a
“description of incident that includes precise description of evidence that either justifies or fails to
justify the officer’s conduct based on investigator’s independent review of facts/circumstances of
the incident.” This requirement requires a comprehensive assessment of a report’s analysis of the
evidence bearing on an investigator’s findings, and the Monitoring Team does not have confidence
in these results, which seem very high given the generally low quality of the investigations
reviewed.

Yes No Agree
(Q72) Narrative with description of incident
that includes precise description of evidence 173.5 (71.99%) 67.5 (28.01%) 31/40
that either justifies or fails to justify (77.5%)
officer’s conduct based on investigator’s
independent review of facts/circumstances
of the incident?

To probe these topline findings further, the Monitoring Team excluded cases which were
administratively closed (which, by definition should not have included such a specific narrative).
Of the cases that were definitively found not administratively closed, a full 81% did not have
narratives meeting the standard. The Monitoring Team also found that FTA cases (which reviewers
anecdotally indicated were more often complete) had the highest percentage of narratives; the CIU
cases (which often had the most chaotic files) had the lowest.

Narrative with description (Q72)

Q82e Allegation closed or Yes No Total (N= (241)
administrative closed
Yes 40 (39.6%) 61 (60.4%) N=101

44
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 154 of 164

No 25.5 (19.0%) 108.5 (81.0%) N=134

UTD 1.5 (25%) 4.5 (75%) N=6

(Q75) Did the investigative report include

documentation that officer’s certification and 3.5 (43.75%) 4.5 (56.25%)
training for the weapon were current?

70 percent of cases included sufficient documentation of evidence gathered, which is remarkably

high given how incomplete many of the case files were. In 15% of cases where there were no
known witnesses, there was not sufficient documentation to support the case resolution.

Only 22% of cases contained complete records of audio, video, or transcript evidence for all
witnesses. The paucity of complete interview records greatly limited the reviewers’ ability to
review cases and make determinations of adequacy based on a comprehensive, complete
investigative record. Additionally, while reviewers were able to agree on whether there was
sufficient documentation to support the case overall, agreement was poor on the more nuanced
questions about sufficiency of evidence. Reviewers agreed on whether there was sufficient
documentation of the evidence gathered 78% of the time, but agreement dropped into the 50%
range for the more detailed questions. The alpha of 0.964 for this section suggests redundancy in
the questions as well, so the Monitoring Team attributes these inconsistencies to repetitive
questions and the confused state of the files reviewed. The Monitoring Team will consider revising
this work flow in the next compliance review.

(Q73) Did the Yes No UTD N/A Agree

investigative
report provide:
Sufficient
documentation of 169 (70.12%) 54 (22.41%) 2.5 (1.04%) 15.5 (6.43%) 31/40
all evidence (77.50%)
gathered (including
names, phone
numbers,
addresses of
witnesses, etc.)?
Sufficient
documentation if no 43.5 (18.05%) 36 (14.94%) 3.5 (1.45%) 158 (65.56%) 23/40
known witnesses? (57.50%)
Sufficient
documentation of 33.5 (13.90%) 18 (7.47%) 5.5 (2.28%) 144 (59.75%) 23/40
reason(s) why (57.50%)
witnesses were
unable to be
identified or have

45
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 155 of 164

contact information
collected?
Sufficient
documentation of 32.5 (13.49%) 27 (11.20%) 12.5 (5.19%) 169 (70.12%) 28/40
identification of (70.00%)
anyone who refused
to provide a
statement?
Sufficient
documentation of 123 (51.04%) 49 (20.33%) 3.5 (1.45%) 65.5 (27.18%) 23/40
whether (57.50%)
officers/other BPD
employees were
interviewed?
Sufficient 69.5 (28.84%) 35.5 (14.73%) 7.5 (3.11%) 128.5 (53.32%) 20/40
documentation of (50.00%)
names of all other
BPD employees
who witnessed
interviews?
Audio, video,
and/or transcripts of 53.5 (22.20%) 109 (45.23%) 7.5 (3.11%) 71 (29.46%) 22/40
all witness (55.00%)
interviews
included?
α = 0.964

61 percent of cases provided sufficient evaluation of the incident based on the evidence; similarly,
evidence was adequately examined in 62 percent of cases. Perhaps more significant, reviewers
concluded that a full quarter of cases did not have a sufficient evaluation of the incident.

Overall, material inconsistencies were resolved adequately in only 34 percent of cases and
credibility determinations about witness statements were properly made in only 35 percent of
cases. Similarly, officer statements were critically evaluated in slightly more than 37 percent of
cases and were not well evaluated in almost a quarter of cases. There was prior deception or
untruthful statements by witnesses in 134 cases, but investigators in only six of those cases (4%)
actually took this information into account. Overall, this shows a lack of rigorous analysis of
evidence, even though a majority of evaluations (61%) were passable.

Additionally, throughout this section, very low agree rates appear that cast doubt on the reliability
of the results. Overall, it appears that reviewers were simply throwing up their hands in frustration
with their ability to tease information out of the files. Any section below with an interrater
reliability below 60 percent should be taken as a general comment, not any kind of specific
quantitative finding. The Monitoring Team believes that questions with lower agreement rates
46
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 156 of 164

suffered from a problematic interplay of unorganized, highly chaotic case files and uncertainty
among reviewers about how various elements of the presented file may, or may not, have satisfied
Consent Decree requirements.

(Q76) Evaluation of Yes No UTD N/A Agree

the Incident in the
Investigative report
Sufficient evaluation
of incident, based on 147 (61.00%) 60 (24.90%) 17 (7.05%) 17 (7.05%) 26/40
review of gathered (65.00%)
evidence, as to
whether within
policy, procedures,
regulations, orders,
or other standards of
required conduct?
Explicit credibility
findings, including 76.5 (31.74%) 76 (31.54%) 10 (4.15%) 78.5 24/40
precise description (32.57%) (60.00%)
of evidence
supporting/detractin
g from a person’s
credibility?
Explicit resolution
or discussion of 82 (34.02%) 55 (22.82%) 7 (2.90%) 97 10/26
material (40.25%) (38.46%)
inconsistencies,
including precise
description of
evidence relied upon
to resolve the
inconsistencies?
Evidence adequately
examined and 149.5 (62.03%) 51 (21.16%) 17.5 (7.26%) 23 (9.54%) 29/36
described? (80.55%)

47
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 157 of 164

Credibility
determinations about 85.5 (35.48%) 47 (19.50%) 27.5 (11.41%) 81 9/31
statements based on (33.61%) (29.03%)
independent,
unbiased, and
credible evidence?

Officer statements
critically evaluated? 89.5 (37.14%) 54 (22.41%) 23.5 (9.75%) 74 14/34
(30.71%) (41.18%)
Past
deception/untruthful 6 (2.49%) 89 (36.93%) 39 (16.18%) 107 7/26
statements by (44.40%) (26.92%)
witnesses,
complainants,
officers taken into
account?
Importance of 141 (58.51%) 47.5 (19.71%) 17 (7.05%) 35.5 22/37
evidence reasonably (14.73%) (59.46%)
weighed?
Testimonial 74 (30.71%) 41.5 (17.22%) 11 (4.56%) 114.5 6/26
evidence reasonably (47.51%) (23.08%)
weighed?
All relevant BPD
officer activity in the 120 (49.79%) 46.5 (19.29%) 31 (12.86%) 43.5 20/36
incident and any (18.05%) (55.56%)
evidence of potential
misconduct

48
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 158 of 164

uncovered, whether
or not part of the
original allegation,
fully investigated
and evaluated?
Sufficient 64.5 (26.76%) 51 (21.16%) 9 (3.73%) 116.5 9/29
documentation of (48.34%) (31.03%)
recommendations
for non-punitive
action or misconduct
charges?
Sufficient
documentation of 30 (12.45%) 53 (21.99%) 11 (4.56%) 147 1/21
whether additional (61.00%) (4.76%)
training, counseling,
or intervention was
recommended?
α = 0.963

Inconsistencies in the evidence were encountered in almost a quarter (23 percent) of cases. Where
there were such inconsistencies, they were not resolved some 59 percent of the time. Moving
forward, BPD will need to ensure that investigative reports expressly address and resolve material
inconsistencies.

Yes No Agree
(Q77) Were there material inconsistencies 55 (22.82%) 186 (77.18%) 32/40
between complainant, officers, witnesses, (80.00%)
statements and/or amongst the evidence?
(Q80) Were the inconsistencies 22.5 (40.91%) 32.5 (59.09%) 4/6
appropriately resolved? (66.67%)

(Q78) What was the nature of the inconsistencies

Inconsistencies among statements 29.5
Video inconsistent with statement(s) 11
Inconsistent evidence 6.5
Other 8

(Q79) Did the investigator make and the investigative report state all reasonable efforts
to resolve material inconsistencies?
All inconsistencies reasonably addressed 17

49
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 159 of 164

Inconsistencies not reasonably addressed 38

Finally, weapons were seldom used during incidents (3.32%, or eight cases), but in that small
sample, the current state of the officer’s certification for use of that weapon was included in only
three cases. The rarity of use of weapons makes it difficult to draw conclusions for this
assessment measure, but clearly certification should be included in all cases moving forward.

(Q74) Was a weapon used or involved during the

incident? 8 (3.32%) 233 (96.68%)

INVESTIGATIVE FINDINGS

Of the cases that resulted in disposition, the evidence met the standard of proof required to justify
the disposition10 in 61 percent of cases; the evidence in more than one-quarter (27 percent) did not
meet the standard. Also, as expected because these requirements were not implemented at the time
of review, very few investigations considered patterns in officer behavior (16.4%), prior
unsustained complaints (6%), or officer training records (less than 5%). Agreement and alpha for
these questions were very good. Moving forward, PIB will need to significantly improve the
quality of both investigations and investigator analysis to ensure that, in every investigation, the
evidence meets the standard of proof required to justify the disposition. Further, PIB must adhere
to Consent Decree requirements by examining patterns and past behaviors.

In approximately seven out of ten cases (72 percent), investigators expressly identified and
recommended a specific disposition. In 33 percent of cases, however, the finding that investigators
was “closed” or administratively closed.” This is remarkably consistent with the findings in the
original investigation by the Department of Justice, which found that “BPD supervisors
administratively closed 33 percent of all allegations received from 2010 through 2015—ensuring
that the allegations would result in no further investigation or officer discipline.” Findings Report

10
As previously discussed, the relevant findings and standards of proof are:
“Unfounded,” where the investigation determines, by clear and convincing evidence, that the
alleged misconduct did not occur or did not involve the accused officer;
“Sustained,” where the investigation determines, by a preponderance of the evidence, that the
alleged misconduct did occur;
“Not Sustained,” where the investigation is unable to determine, by a preponderance of the
evidence, whether the alleged misconduct occurred;
“Exonerated,” where the investigation determines, by a preponderance of the evidence, that the
alleged conduct did occur but did not violate BPD policies, procedures, or training.

50
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 160 of 164

at 142. Administratively closing cases without adequate investigation or without reaching

disposition undermines the legitimacy the BPD’s system of accountability.

Reviewers said that 68 percent of allegations closed without disposition had documented
supervisory approval for closure. The Monitoring Team does not have confidence in this figure.
We believe that it is the result of (1) the chaotic nature of the reviewed files and (2) a lack of
uniform operationalization among reviewers with respect to how the Consent Decree’s
requirements for documentation will look like in practice. The Team suspects that some reviewers
inferred approval from the fact that a case was documented as closed within the IAPro database.
This is not necessarily the case.

(Q82) Investigative Yes No UTD Agree

Findings
For each misconduct
allegation, did the 172.5 (71.58 %) 60.5 (25.10%) 8 (3.32%) 35/40
investigator expressly (87.50%)
identify and recommend
a disposition of
“unfounded,”
“sustained,” “not
sustained,” or
“exonerated” explicitly
identified and
recommended by the
investigator?
Did findings consider 39.5 (16.39%) 151.5 (62.86) 50 (20.75%) 36/40
patterns in officer (90.00%)
behavior based on
disciplinary history?
Did findings consider 14.5 (6.02%) 168.5 (69.92%) 58 (24.07%) 33/40
prior complaints in (82.50%)
which allegations were
not sustained?
Did findings consider 11.5 (4.77%) 173 (71.78%) 56.5 (23.44%) 39/40
officer training records? (97.50%)
For any allegation, was
the disposition "closed" 78.5 (32.57%) 156.5 (64.94%) 6 (2.49%) 29/40
or “administratively (72.50%)
closed” used?

51
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 161 of 164

Based on your review of

the investigation, did the 147 (61.00%) 64 (26.56%) 30 (12.45%) 30/40
investigator’s (75.00%)
recommended
disposition of each
allegation and/or finding
meet the required level
of proof?
("Preponderance of the
evidence" for all
dispositions other than
"unfounded," for which
required level of proof is
"clear and convincing
evidence").
α = 0.752

Yes No Agree
(Q83) Was the approval for the use of
the "closed" or "administratively 54 (68.79%) 24.5 (31.21%) 6/7
closed" disposition documented in (85.71%)
writing? (N=78.5 from Q82e)

ADMINISTRATIVE REVIEW OF COMPLETED INVESTIGATION

The Consent Decree requires administrative review of completed misconduct investigations.

Cases that are managed by supervisors at Districts must be forwarded through the chain of
command to the District Commander to determine whether the evidence satisfies the standard of
proof that justifies the recommended disposition, whether there are investigative deficiencies,
and whether the investigation is complete and accurate. Dkt. 2-2 at ¶354(a).

All investigations are ultimately referred to the Director of OPR, to:

[R]eview the report to ensure that the report is complete, that it meets the
requirements of BPD policy and this Agreement, and that the findings are supported
by the appropriate standard of proof;

52
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 162 of 164

[O]rder additional investigation when it appears that there is additional relevant

evidence that may assist in resolving inconsistencies or improving the reliability or
credibility of the findings . . . .

Dkt. 2-2 at ¶354(b).

Only about 44 percent of cases were timely forwarded through the chain of command for review.
At the chain of command review, patterns in officer behavior based on disciplinary history were
considered in only 14 percent of cases. Prior complaints that did not result in discipline and
training records were almost never considered (5.6%).

(Q85) Yes No UTD Agree

Was completed investigation
timely forwarded through 105 (43.57%) 98 (40.66%) 38 (15.77%) 30/40
investigator’s chain of (75.00%)
command to Director of
OPR?
Did review/findings consider 33 (13.69%) 148.5 61.41%) 59.5 (24.69%) 38/40
patterns in officer behavior (95.00%)
based on disciplinary
history?
Did review/findings consider 13.5 (5.60%) 162 (67.22%) 65.5 (27.18%) 35/40
prior complaints in which (87.50%)
allegations were not
sustained?
Did review/findings consider 13.5 (5.60%) 160 (66.39%) 67.5 (28.01%) 40/40
officer training records? (100.00%)
Was the disposition "closed" 80 (33.20%) 150.5 (62.45%) 10.5 (4.36%) 27/40
or "administratively closed" (67.50%)
used?
α = 0.801

Yes No Agree
(Q86) Was approval for "closed" or 43.5 (55.06%) 35.5 (44.94%) 3/6
"administratively closed" disposition (50.00%)
documented in writing? (N=80 from
Q85e)

In addition to all the specific investigation requirements, the Consent Decree requires BPD to
“develop a process to ensure that completed misconduct investigations are evaluated for policy,

53
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 163 of 164

training, tactical or equipment concerns, including any recommendations for how those concerns
will be addressed.” CD at ¶352. Specifically, BPD should evaluate whether:

a. An assessment of whether the law enforcement action was in compliance

with training and legal standards;
b. Other tactics were more appropriate under the circumstances;
c. The incident indicates a need for additional training, counseling, or other
non- disciplinary corrective actions; and
d. The incident suggests that BPD should revise its policies, strategies, tactics
or training.

Reviewers found that these assessments were rarely evident in the investigation files, with only 35
percent of cases clearly indicating that the conduct under investigation was in compliance with
training and legal standards. Consideration of tactics (5.8%), training or counseling (11.4%), and
changes to policy (7.8%) was infrequent.

POLICY, TRAINING, & TACTICAL CONCERNS

(Q88) Yes No UTD Agree
Law enforcement action 84.5 (35.06%) 76 (31.54%) 80.5 (30.40%) 31/40
was in compliance with (77.50%)
training and legal
standards?
Other tactics were more 14 (5.81%) 159.5 (66.18%) 67.5 (28.01%) 38/40
appropriate under the (95.00%)
circumstances?
Incident indicates need for 27.5 (11.41%) 154.5 (64.11%) 59 (24.48%) 35/40
additional training, (87.50%)
counseling, or other non-
disciplinary corrective
actions?
Incident suggests that BPD 19 (7.88%) 169.5 (70.33%) 52.5 (21.78%) 36/40
should revise its policies, (90.00%)
strategies, tactics, or
training?
Α = 0.912

The Consent Decree also imposes supervisory requirements for misconduct investigations. In
addition, it imposes restrictions intended to protect against conflicts of interests for investigators,
disqualifying as investigators and supervisors PIB staff who either witnessed the incident that is

54
 Case 1:17-cv-00099-JKB Document 342-1 Filed 09/30/20 Page 164 of 164

the subject of the investigation or had a business or personal relationship with the subject, the
complainant. or a witness.

Meetings between OPR supervisors and investigators on investigative progress were only
documented in about 4 percent of cases. Cases were only reassigned in 5.6 percent of cases, though
the vast majority of those reassignments were documented.

Conflicts either with involved personnel or personnel with a personal or financial conflict of
interest were identified in less than 1 percent of cases.

CONFLICTS OF INTEREST
(Q89) Yes No UTD
Meetings documented between OPR 10.5 (4.36%) 174 (72.20%) 56.5 (23.44%)
supervisors and investigators to
evaluate progress of an investigation
documented?
Was an employee who was involved in 2 (0.83%) 192 (79.67%) 47 (19.50%)
or a witness to the incident conduct or
review the
investigation arising from the incident?
Did an employee with (1) an external 1 (0.41%) 185 (76.76%) 55 (22.82%)
business relationship, or (2) a personal
relationship with a principal or witness
in the investigation conduct or review
the misconduct investigation?
Was the investigation reassigned to 13.5 (5.60%) 223.5 (92.74%) 4 (1.66%)
another investigator at any point?

Yes No
(90) Was the re-assignment documented? 11 (81.48%) 2.5 (18.52%)

55