BOP Special Housing Unit Independant Assessment, CNA Analysis, 2014

• Feb. 8, 2016 • Locations: United States of America • Topics: Control Units/SHU/Solitary Confinement

Download original document:
Brief thumbnail

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

Federal Bureau of Prisons: Special
Housing Unit Review and
Assessment
Key Contributors:
Kenneth McGinnis, Dr. James Austin, Karl Becker, Larry Fields, Michael
Lane, Mike Maloney, Mary Marcial, Robert May, Jon Ozmint, Tom Roth,
Emmitt Sparkman, Dr. Roberta Stellman, Dr. Pablo Stewart, George Vose,
and Tammy Felix
December 2014

Distribution limited to Federal Bureau of Prisons

Acknowledgements
The project team acknowledges and thanks Director Charles E. Samuels, Jr. and the
executive staff of the Bureau of Prisons for their assistance and support in conducting an
independent assessment of the restrictive housing programs within the agency. This
review could not have been successfully completed without Director Samuels’ strong
personal support of the scope of the review and the goals and objectives of this project.
His personal involvement and interest in the independent review of the operations and
programs of this aspect of the Bureau was critical to completing the assessment. In
addition, the project team thanks the wardens and staff of the institutions visited by the
team for their outstanding support and willingness to fully open their operation to outside
scrutiny.
Finally, this project could not have been possible without the support and deep
involvement of the National Institute of Corrections and Acting Director Robert M.
Brown, Jr. and his staff. In particular, the CNA project team is grateful to Shaina Vanek
for her commitment to and coordination of the project and for facilitating
communication and the exchange of information between the project team and the
Bureau staff. Without her personal involvement and the initiative she took to find
resolution to literally every challenge that arose during the project, this report could not
have been completed.
Distribution
Distribution limited to Federal Bureau of Prisons. Specific authority: Order Number
DJBP0700NAS079
Distribution authority is controlled by Federal Bureau of Prisons.
This document represents the best opinion of CNA at the time of issue.

Approved by:

December 2014

Monica Giovachino
Managing Director, Safety and Security
CNA, Institute for Public Research

Abstract
This report provides an independent, comprehensive review of the Federal
Bureau of Prisons’ operation of restrictive housing and identifies potential
operational and policy improvements. Specifically, it provides a comprehensive,
detailed evaluation of the Bureau’s use of restrictive housing, including the
following key areas: national trends and best practices in the management of
restrictive housing units; profile of the Bureau’s segregation population; Bureau
policies and procedures governing the management of restrictive housing; unit
operations and conditions of confinement; mental health assessment and
treatment within restrictive housing units; application of inmate due-process
rights; reentry programming; and the impact of the use of restrictive housing on
system safety and security. The report also evaluates the impact of the
restrictive housing program on the federal prison system and places the
Bureau’s use of segregation in context with professional standards and best
practices found in other correctional systems.
The findings and recommendations contained in this report are based on the
information and data collected while conducting site visits to the Bureau’s
restrictive housing units and facilities from November 2013 through May 2014.
Any operational changes or new written policies implemented by the Bureau
after completion of the site visits regarding their use of restrictive housing are
not reflected in this report. Some such changes were in process or were
scheduled for implementation after the completion of the site visits.

This page intentionally left blank.

Executive Summary
The Federal Bureau of Prisons (Bureau) uses restrictive housing for serious
infractions of institutional and system-wide rules governing inmate conduct, such as
engaging in violent, aggressive behavior against other inmates and staff. Restrictive
housing is also used for inmates who cannot be safely managed in a general
population setting, or who have been otherwise determined to be a security threat.
There are three categories of restrictive housing used by the Bureau: Special Housing
Units (SHU), Special Management Units (SMU), and the United States Penitentiary,
Administrative Maximum (ADX) in Florence, Colorado.
The Bureau’s Program Statements governing the three types of segregated housing
units indicate that all three types of housing have a similar function and purpose
which is to “separate inmates from the general inmate population to protect the
safety, security, and orderly operation of Bureau facilities, and to protect the public.”
The specific placement criteria and conditions of confinement vary for each type of
segregated housing unit as does the type of inmate housed in each of the respective
units.
As of November 2013, approximately 5 percent of the entire Bureau’s prisoner
population was being housed in one of these restrictive housing populations with the
vast majority in the SHU status (see Figure 1). Shortly thereafter the number of
inmates held in SHU housing began to decline. Similarly the SMU population began
to decline in the summer of 2013. The Bureau population as a whole has also been
slightly reduced from a peak of 217,815 to its current population of 212,283 as of
December 12, 2014. This level of use of restrictive housing is consistent with that
experienced by most state correctional systems.
Much of the decline is attributable to a reduction in the SHU population and, in
particular, inmates who have been assigned to protective custody or are serving
disciplinary segregation sanctions. The Bureau was able to provide detailed SHU
population statistics beginning in February 2013. At that time, the count was 10,262
in over 100 facilities and has steadily declined since then reaching 8,939 by June
2014. This is a reduction of 31 percent from the 13,000 reported count of the SHU
population in 2011. There have been no reductions in the ADX populations.

iii

Figure 1.

Number of ADX, SMU and SHU Inmates February 2013 – June 2014

Below we summarize key findings from our review:
The general conditions of confinement in restricted housing units are consistent
with national regulations and standards.
Management of the SHU’s is complicated by the high percentage of inmates that
have requested protection from other inmates, often due to gang related issues.
The Bureau does not have adequate non-punitive protective custody housing
units that have equivalent levels of programs and privileges as general
population inmates.
Backlogs in inmates awaiting transfer to the next program level negate the
intent of the program design and decrease the motivation to change behavior.
Mental health services in restrictive housing require improvement in three
specific areas: 1) proper mental health diagnoses; 2) more effective treatment;
and 3) providing sufficient psychiatric staffing.
The lack of time parameters for completion of disciplinary hearings results in
substantial variation among facilities in the amount of time served in
segregation for similar offenses, and can result in disproportionately long
sanctions.
There is no formal Bureau-wide reentry preparedness program specific to
restrictive housing and inmates in these settings have very limited access to
reentry programming.

Bureau information systems do not effectively track the number and movement
of inmates within the restrictive housing units.
There are additional opportunities available to the Bureau to further lower the SHU
and SMU populations by adopting the recommendations outlined in this report.
Primary approaches to further reduce the restrictive housing population include:
Establish a time limit on the amount of time that an inmate can be held in
investigative status;
Allow credit for time served in SHU upon determination of disciplinary
sanction;
Establish a housing option separate from SHU for inmates in protection status
(protective custody);
Continue rigorous review of referrals to restrictive housing;
Reduce the time period for completion of the SMU program from the present
18-24 months to 12 months and compress the four levels to three levels by
combining Level 3 and Level 4 and allowing more differentiation between the
conditions of confinement between the levels; and
To ensure appropriate treatment for seriously mentally ill inmates, a complete
review of all inmates assigned to ADX, SMU and SHU should be completed by
the Bureau to identify all inmates who should be transferred to a secure mental
health program similar to the ones being developed at USP Atlanta and USP
Allenwood.

This page intentionally left blank.

Contents
Chapter 1:

Introduction ........................................................................................................ 1

Project background ............................................................................................................... 1
Methodology ........................................................................................................................... 2
Facility selection ............................................................................................................. 3
Operational assessment—facility site visit protocol .............................................. 4
Inmate samples and interviews................................................................................... 6
Mental health interviews .............................................................................................. 8
Quantitative data analysis—individual level ............................................................ 9
Quantitative data analysis—aggregate level ........................................................... 10
Composition of the Bureau of Prisons special housing .............................................. 11
Special housing units .................................................................................................. 11
Special management units ......................................................................................... 14
Administrative maximum facility program ............................................................ 18
ADX and SMU population trends .............................................................................. 20
Summary of the Bureau’s recent initiatives and segregation capacity .................... 22
Development of the reintegration housing unit .................................................... 23
USP Atlanta mental health unit ................................................................................. 24
Chapter 2:

National trends and use of segregated housing .....................................25

Definitions of segregation and types of inmates in segregation .............................. 28
Key litigation issues ............................................................................................................ 31
Standards regarding segregation ..................................................................................... 33
Segregation sentencing structures .................................................................................. 35
Impact of segregation systems on institutional safety ............................................... 36
Mental health issues in segregated housing .................................................................. 38
Step-down programs .......................................................................................................... 43
Virginia ........................................................................................................................... 44
Mississippi ..................................................................................................................... 44
Washington .................................................................................................................... 46
Maine............................................................................................................................... 46
Reductions in segregation populations .......................................................................... 47
Mississippi ..................................................................................................................... 47
Colorado ......................................................................................................................... 48

vii

Reentry and prerelease programming ............................................................................ 48
Summary ............................................................................................................................... 50
Chapter 3:

Analysis of the restrictive housing populations ....................................51

Population characteristics ................................................................................................. 51
Primary offenses and sentences ............................................................................... 52
Medical and mental health care levels ..................................................................... 54
Security and special management issues ................................................................ 56
Time in restrictive housing ........................................................................................ 58
Disciplinary conduct of the SMU and ADX ............................................................. 58
Recidivism rates of ADX and SMU inmates ............................................................ 63
Trends in assault rates and lockdowns ................................................................... 64
Summary of major findings ....................................................................................... 66
Chapter 4:

Due process ...................................................................................................... 67

Federal due process standards......................................................................................... 68
Inmate discipline ................................................................................................................. 70
Disciplinary process observations ............................................................................ 75
Special housing units.......................................................................................................... 79
Placement ....................................................................................................................... 80
Placement of protection cases in administrative detention................................ 81
Protective custody........................................................................................................ 82
Weekly review meetings.............................................................................................. 85
Segregation reviews ..................................................................................................... 86
Special management units................................................................................................. 88
Referral and assignment ............................................................................................. 88
Progression through program phases ..................................................................... 94
Observations on progression..................................................................................... 97
Due process and ADX ...................................................................................................... 102
Administrative remedy program .................................................................................. 105
Chapter 5:

Mental health assessment and treatment .............................................. 111

Assessment process ........................................................................................................ 113
Inmate interviews and case review trends .................................................................. 115
Clinical analysis and observations ............................................................................... 118
Mental health diagnostic process .......................................................................... 118
Mental health treatment .......................................................................................... 126
Other site-specific observations ............................................................................ 129
Chapter 6:

Reentry ........................................................................................................... 133

viii

Reentry programs for inmates in the general population ....................................... 134
Reentry program evaluation .......................................................................................... 138
Program services in restrictive housing ............................................................... 139
Reentry program services in restrictive housing................................................ 141
Reentry program observations ............................................................................... 146
Chapter 7:

Restrictive housing operations and conditions of confinement ..... 151

Organizational structure ................................................................................................ 151
Special housing unit management......................................................................... 151
ADX Florence unit management ............................................................................ 153
Restrictive housing management structure observations ................................ 154
Correctional operations staffing levels and approach ............................................. 155
Special housing unit staffing .................................................................................. 155
Special management unit staffing ......................................................................... 158
Control unit staffing at Florence ADX .................................................................. 160
Staffing observations ............................................................................................... 162
Staff training, curriculum, and approach .................................................................... 163
Bureau of Prisons employee development manual ............................................ 165
Annual refresher training ........................................................................................ 165
Specialized training .................................................................................................. 166
Special housing unit specific training................................................................... 167
Special housing unit training curriculum ............................................................ 168
Staff training observations...................................................................................... 169
Security systems and practices ..................................................................................... 170
Physical plant characteristics ................................................................................. 171
Patrol requirements .................................................................................................. 173
Inmate movement practices.................................................................................... 174
Impact of gang issues and separation practices........................................................ 178
Post orders ......................................................................................................................... 180
Use of force and critical incidents ................................................................................ 180
Types of force ............................................................................................................ 182
Use of force observations ........................................................................................ 187
Use of force at USP Lewisburg ................................................................................ 187
Conditions of confinement ............................................................................................ 188
Adequacy and allocation of program space ............................................................... 204
Statutory and Bureau requirements ...................................................................... 205
Existing potential program space .......................................................................... 207
Observations on conditions of confinement ....................................................... 209
Program reviews ............................................................................................................... 210
Program review observations ................................................................................. 211
Reporting .................................................................................................................... 212

Operational review observations ........................................................................... 213
Chapter 8: Summary of conclusions ................................................................................. 215
Extent of segregation ....................................................................................................... 215
Conditions of confinement ............................................................................................ 217
Impact of the “separatee” status .................................................................................. 218
Investigation, protective custody, and pending transfer inmates in SHU ............ 219
Inmates awaiting administrative actions .................................................................... 219
Protective custody inmates in special housing units ............................................... 220
The special management unit program ....................................................................... 221
Mental health services ..................................................................................................... 222
Crowding and segregation.............................................................................................. 224
Due process ....................................................................................................................... 225
Reentry ............................................................................................................................... 227
Information system needs .............................................................................................. 228
Summary of findings and recommendations ............................................................. 228
Statistical analysis of restrictive housing ............................................................ 228
Due process ................................................................................................................ 229
Mental health ............................................................................................................. 231
Reentry ........................................................................................................................ 233
Conditions of confinement ..................................................................................... 234
Appendix: Research team bios ........................................................................................... 237

List of figures
Figure 1.
Figure 2.
Figure 3.
Figure 4.

Number of ADX, SMU and SHU Inmates February 2013 – June
2014 ................................................................................................................... iv
SMU and ADX populations, January 2004 – July 2014 .......................... 21
ADX and SMU populations and inmate assault rate, 2004-2013 ......... 65
Inmate disciplinary process......................................................................... 71

This page intentionally left blank.

xii

List of tables
Table 1.
Table 2.
Table 3.
Table 4.
Table
Table
Table
Table

5.
6.
7.
8.

Table 9.
Table 10.
Table 11.
Table 12.
Table 13.
Table 14.
Table 15.

Table 16.
Table 17.
Table 18.
Table 19.
Table 20.

SMU and ADX facilities and populations selected for site visits,
November 2013 ................................................................................................ 3
SHU facilities and populations selected for site visits, December
2013 .................................................................................................................... 4
SHU and SMU prisoner interviews attempted and completed by
facility................................................................................................................. 8
Inmates interviewed and evaluated regarding mental health status
by facility ........................................................................................................... 9
SHU populations by status, February 2013 – June 2014 ....................... 14
SMU populations by facility, December 2013 .......................................... 18
Bureau segregation populations by segregation type, 2004–2014...... 22
Demographic attributes of ADX, SMU, SHU, and other Bureau
populations, November 23, 2013 ............................................................... 52
Primary offenses of ADX, SMU, SHU, and other Bureau
populations, November 23, 2013 ............................................................... 53
Sentence and time served attributes of ADX, SMU, SHU, and other
Bureau populations, November 23, 2013 ................................................. 54
Medical and mental health care levels of ADX, SMU, SHU, and other
Bureau populations, November 23, 2013 ................................................. 56
Security and other management issues of ADX, SMU, SHU, and
other Bureau populations, November 23, 2013 ...................................... 57
Length of time in ADX, SMU, and SHU status, November 23, 2013 .... 58
Disciplinary reports for current ADX and SMU populations, as of
November 23, 2013—total number of reports ........................................ 59
Disciplinary reports for ADX and SMU populations, as of
November 23, 2013—reports in the 12 months before and after
restrictive housing placement ..................................................................... 59
Types of disciplinary infractions by ADX inmates, as of November
23, 2013 ........................................................................................................... 60
Types of disciplinary infractions by current SMU inmates, as of
November 23, 2013 ....................................................................................... 61
In-prison recidivism rates for inmates released from ADX and SMU . 64
Prison recidivism rates for inmates released from SMU by time in
SMU ................................................................................................................... 64
Changes in ADX and SMU populations, assaults, and lockdowns,
2008–2013 ....................................................................................................... 65

xiii

Table 21.
Table 22.
Table 23.
Table 24.
Table 25.
Table
Table
Table
Table

26.
27.
28.
29.

Hearing dismissal rates ................................................................................ 76
SMU referrals and outcomes, January 2013 – February 2014 .............. 92
Remedy/SMU filings by reason, December 19, 2012 – December
19, 2013 ......................................................................................................... 108
Remedy SHU filings by disposition type, December 19, 2012 –
December 19, 2013 ...................................................................................... 108
Symptoms of a mental health disorder among prison and jail
inmates........................................................................................................... 112
Cases interviewed and assessed ............................................................... 115
Key health care attributes of assessed inmates .................................... 117
Key conclusions of independent psychiatric review ............................ 118
Mental health treatment staff at reviewed facilities............................. 124

xiv

Glossary
ACA
ADX
ART
ASCA
BOP
CHS
CFR
DHO
DR
DSCC
ESL
FCI
GAO
GED
LCP
MDOC
NIC
OC
ORE
PRD
PREA
RAC
RHU
RPP
RRC
SAM
SHU
SIS
SMI
SMU
SRO
SSU
UDC
USP
VADOC

American Correctional Association
Administrative Maximum Security Institution
Annual Refresher Training
Association of State Correctional Administrators
Federal Bureau of Prisons
Criminal History Score
Code of Federal Regulations
Disciplinary Hearing Officer
Disciplinary Report
Designation and Sentence Computation Center
English as a Second Language
Federal Correctional Institution
Government Accountability Office
General Educational Development
Life Connections Program
Mississippi Department of Corrections
National Institute of Corrections
oleoresin capsicum
Office of Research and Evaluation
Program Review Division, Federal Bureau of Prisons
Prison Rape Elimination Act
reentry affairs coordinator
reintegration housing unit
Release Preparedness Program
residential reentry center
special administrative measure
special housing unit
Special Investigative Service
Serious Mental Illness
special management unit
segregation review official
special security unit
unit disciplinary committee
United States penitentiary
Virginia Department of Corrections

This page intentionally left blank.

xvi

Chapter 1: Introduction
Project background
The use of high-security, restrictive housing units, also known as segregation units,
by prison systems to manage dangerous or problematic offenders has received
increased scrutiny in recent years. Virtually all state correctional systems, as well as
most large local jail systems, use these units as a disciplinary tool and as a means to
manage offenders who may need to be kept separate from general institutional
populations. These units are typically characterized by very limited out-of-cell time
and reduced access to privileges such as phone calls, visits, and personal property.
The Federal Bureau of Prisons (Bureau) uses restrictive housing for serious
infractions of institutional and system-wide rules governing inmate conduct, such as
engaging in violent, aggressive behavior against other inmates and staff. Restrictive
housing is also utilized for inmates who cannot be safely managed in a general
population setting, or who have been otherwise determined to be a security threat.
There are three categories of restrictive housing used by the Bureau:
Special housing units (SHU)
Special management units (SMU)
The administrative maximum (ADX) facility in Florence, Colorado
The Bureau has developed a comprehensive set of policies and procedures that
govern the operation of these restrictive housing units. In order to assess the
effectiveness of these policies and the consistency of restrictive housing operations
with accepted national standards and best practices, the Bureau sought an
independent, outside review of the restrictive housing program. Accordingly, on
January 29, 2013, the Bureau issued RFQ AS0139-2013, Special Housing Unit Review
and Assessment. The stated objective of the RFQ was to select a contractor to
conduct an independent, comprehensive review of the Bureau’s operation of
restrictive housing and identify potential operational and policy improvements. After
an extensive evaluation process including a technical review of qualifications by the
National Institute of Corrections (NIC), CNA was selected to conduct this assessment.
Upon completion of background checks of project team members, CNA received

notice to commence work on the project on September 19, 2013. Interviews and
fieldwork at Bureau facilities began in November 2013 and continued through May
2014. Data analysis and follow-up reviews were completed by November 2014.
The CNA project team was composed of eight former state correctional system
directors, four former deputy correctional system directors, two psychiatrists, and
two PhD-level criminal justice system researchers. All team members had substantial
experience in the management and evaluation of restrictive housing units.
The following report provides a comprehensive, detailed evaluation of the Bureau’s
use of restrictive housing, including the following key areas:
National trends and best practices in the management of restrictive housing
units
Profile of the Bureau’s segregation population
Bureau policies and procedures governing the management of restrictive
housing
Unit operations and conditions of confinement
Mental health assessment and treatment within restrictive housing units
Application of inmate due process rights
Reentry programming
Impact of the use of restrictive housing on system safety and security
The report evaluates the impact of the restrictive housing program on the federal
prison system and places the Bureau’s use of segregation in context with
professional standards and best practices found in other correctional systems.

Methodology
The overall research approach consisted of a wide variety of qualitative operational
assessments as well as quantitative methods that provided a comprehensive review
of the Bureau’s current restrictive housing practices. In this section of the report,
these methods are described.

Facility selection
The first step in structuring the analysis was selection of the facilities to be included
in the review. The three Bureau facilities that house inmates in SMU and ADX
status—Florence, Lewisburg, and Allenwood—were designated for comprehensive
site visits. At the time the study commenced, there were approximately 2,100
inmates in the SMU and ADX units at these facilities. Table 1 shows the population
breakdowns at these facilities at the beginning of the project. Inmates under special
administrative measures (SAMs) located at the ADX were excluded from this study
under the terms of the contract.
Table 1.

SMU and ADX facilities and populations selected for site visits,
November 2013

Facility

SMU

ADX

Control

USP Florence
ADX Florence

Totals

645

742

322

USP Lewisburg

786

USP Allenwood

249

1,680

322

2,099

Totals
Source: Bureau/NIC.

The next step was the selection of facilities that house the much larger SHU
populations. A list of these facilities was provided by the Bureau/NIC, outlining each
facility’s SHU population and geographic location. The 14 private facilities that hold
nearly 1,700 SHU inmates on any given day and the various metropolitan correctional
centers were excluded from the study.1 Table 2 lists the facilities selected for
comprehensive, on-site SHU reviews. The USP Hazelton facility, which houses female
SHU inmates, was included in the study in order to assess conditions in female
restrictive housing. Currently, there are no female inmates in SMU or ADX status,
largely because there is not a sufficient number to create specialized SMU or ADX
female units.
The sites selected were the more secure USP facilities with the exception of the
federal correctional institution (FCI) Butner Medium II. Concentrating on higher
security facilities that contained significant SHU populations offered the most value
to the project, given the project budget and the time constraints. However, the
geographical mix of the facilities combined with the size of the sample they provide

Bureau Secure Housing and Inmate Discipline Quarterly Report, June 30, 2013, p. 13.

ensures sufficient numbers to make this review valid and representative of the
Bureau as a whole.
Table 2.

SHU facilities and populations selected for site visits, December 2013

Facility
FCI Butner Medium II

Region
Mid-Atlantic

USP Coleman I

Southeast

USP Hazelton (females)
USP Terre Haute

Total SHU
population

Disciplinary

Administrative

184

148

Mid-Atlantic

North Central

206

187

USP Tucson

West

143

104

USP Victorville

West

256

210

USP Florence

North Central

898

211

663

Total
Source: Bureau/NIC.

Operational assessment—facility site visit protocol
The project team conducted an assessment of segregated housing unit operations at
each facility. The teams assessed operational performance and compared policy
compliance with applicable statutes; the Code of Federal Regulations (CFR); Bureau of
Prisons program statements, policies, and operational memoranda; and American
Correctional Association (ACA) standards. Each project team member has experience
and expertise in the areas they were assigned to evaluate. The team included former
directors of corrections, former prison wardens, program supervisors, psychiatrists
with experience in mental health treatment and correctional medicine, an attorney,
and project researchers.
Initially, the reviews addressed the facility mission, goals, and objectives of the
restrictive housing unit. In advance of the site visits, a document review was
conducted to determine the availability of data required for the assessment process.
Each facility that was assessed is accredited by the Commission on Accreditation of
the ACA, and the latest Visiting Committee report that led to the accreditation was
assessed as part of preliminary data gathering. The team also reviewed any pending
litigation and court orders that affected operational performance, policies, and/or
operating procedures.
Each site visit included two to three days on site with observations of facility
operations on all three shifts. CNA team members were able to access all areas
housing segregated inmates and interview any staff member at the facility to gather
information for the facility assessment. At the end of each site visit, the warden and

associated executive staff received a preliminary briefing on major findings and
possible recommendations.
During the site visits, inmates from pre-selected representative samples were
interviewed in conjunction with a review of their case files. A detailed description of
the sampling process follows in the next section of this report. File reviews were
conducted to determine if Bureau policies regarding due process issues were
followed. The inmates’ disciplinary records were also examined to better understand
the basis for placement in restrictive housing.
Specific areas of review by the consultants included, but were not limited to, the
following:
Physical plant—assessment of unit design and the availability of services within
the context of the design. Cell size and adequacy of cell furnishings were
examined as well as the present cell occupancy versus the intended cell design.
Conditions of confinement—adequacy of recreation space, amount of out-of-cell
time, quality and quantity of meals, clothing, access to hygiene products,
correspondence privileges, access to legal services, visitation, access to
commissary, and access to showers and sanitary facilities. As part of this
process, facility records on out-of-cell time, recreation, meals, showers, haircuts,
telephone use, and visitation were reviewed.
Staffing levels—unit staffing plans were assessed to determine the adequacy of
staffing to meet unit demands and workload. As part of the analysis, records of
actual deployment on posts in the staffing plan were reviewed to ensure that
staffing levels were consistent with the post plan. The consultants reviewed up
to three months of records, known as staffing rosters, to make their judgments.
The staffing assessment included a review of the manner in which the Bureau
supervises staff and assigns staff to work in the units and the staff rotation
schedule.
Staff training—evaluations of Bureau training programs that prepare staff to
work in the correctional environment, as well as training specific to the
management of restrictive housing units. Special attention was paid to training
in the use of force, use of chemical agents, self-defense, and unit operations, as
related to restrictive housing operations. Staff training attendance records were
reviewed to determine if the staff in need of such training was receiving it.
Use of force—analysis of federal regulations and policies relating to the
authorized use of force, including a review of six months of data on use-of-force
incidents and a close examination of a random selection of use-of-force
incidents, including review of video footage of the incidents.

Disciplinary procedures and due process—compliance with federal regulations
and the program statements concerning inmate discipline, including an
examination of disciplinary records documenting compliance with due process
requirements.
Operational requirements—review included job descriptions, post orders,
policies and procedures, supervision and patrol requirements, the application of
restraints, documentation of unit activity such as logs, inmate movement
procedures, and visits by institutional staff including managers, supervisors,
healthcare professionals, and program staff. The frequency of critical incidents
and use-of-force incidents was also a part of this assessment.
Classification procedures and compliance—review included periodic progress
reviews, interviews with unit team members, and a review of inmate case files.
Compliance with segregation reviews was also reviewed.
Programming/reentry—institutional practice relating to the availability of and
inmate participation in programming. A review of practices providing inmates
with a program release preparation prior to their release was also conducted.
Access to medical services—examination of records documenting medical staff
presence in the restrictive housing units and access to care inmates receive. The
operational assessment did not include a review of the quality of medical care.
Mental health care—review of the management of inmates with mental health
issues, particularly those classified as seriously mentally ill. The project
reviewed medical and mental health records and conducted interviews with
inmates and clinical staff.

Inmate samples and interviews
As noted above, representative samples of inmates at selected facilities served as the
focus of the data collection and analysis. By closely examining these inmates through
case file review, observations, and interviews, the project team obtained a better
understanding of the nature and effects of restrictive housing within the Bureau. A
considerable amount of time was also spent interviewing staff at each facility and at
the Bureau executive level to gain their perspectives. Restrictive housing operations
were also assessed through a review of a wide range of Bureau documents and
reports.
The design of the study called for sampling of inmates currently housed at the
selected facilities. Once selected, all data on the inmate was collected and evaluated
on each of the factors listed above. The Bureau created an electronic spreadsheet for

each facility approximately one week prior to the site visit. Based on that
spreadsheet, a random number was assigned to each listed inmate and samples of
approximately 25 or more were selected. The goal was to ensure that at least 20
inmates would receive an in-depth assessment, including interviews and case file
reviews.
A central part of the assessment was a private interview with each inmate outside of
his/her cell. Security procedures typically required that all inmates were escorted by
two officers to the interview room in restraints. With the exception of USP Lewisburg,
the interviews were conducted with just the inmate and interviewer.2 All inmates
were required to sign a standardized informed consent form that was developed by
the NIC, the Bureau, and CNA.
The interviews were based on a structured format that sought responses from the
inmates on the following topics:
Extent of Bureau incarceration
Extent of placement in SHU/SMU/ADX confinement
Basis for placement and due process issues
Conditions of confinement in restrictive housing, including time out of cell,
access to programs, and privileges
Medical and mental health status and care
Staff and/or inmate abuse
Recommendations
Table 3 summarizes the number of inmates who were selected and those who were
actually interviewed. In order to attempt to meet the minimum number of 20 inmates
per site, it was necessary to supplement the random selected sample with inmates
who were not on the list but were willing to be interviewed. These “supplemental”
sampled inmates may have biased the effort to generate representative samples in
unknown ways. Limited analysis is provided in the report to determine how the
inmates chosen for the random sample and inmates actually interviewed differed
from the total SHU/SMU/ADX populations. Due to significant problems with the data
requests, which are detailed later in this report, it was not possible to directly
2
For some interviews, facility leadership required that a facility staff person sit in on the
interview for security reasons.

identify the interviewed inmates and compare them with the total population in
restrictive housing at each facility and across the Bureau.
Overall, there was a 70 percent interview completion rate, which, given the security
requirements associated with the interviews, was better than expected. The lowest
rates were at the Lewisburg and Tucson facilities, where less than 50 percent of the
sampled inmates expressed a willingness to be interviewed. The average number of
interviews across the 12 facilities or units was 22, which met the overall goal of
having at least 20 interviews and cases analyzed.
Table 3.

SHU and SMU prisoner interviews attempted and completed by facility
Completed
No

Yes

Total

Completion rate

Allenwood

74%

Florence

88%

Florence ADX

68%

Lewisburg

49%

SMU subtotal

136

67%

ADX

81%

Butner

62%

Coleman

78%

Florence

55%

Hazelton

97%

Terre Haute

80%

Tucson

47%

Victorville

84%

SHU subtotal

178

250

71%

Grand total

115

269

386

70%

Facility
SMU

SHU

Source: CNA/JFA Institute.

Mental health interviews
In addition to the inmate interviews listed above, separate lists were generated for
inmates to be privately interviewed by the project team psychiatrists. Prior to their
site visit, the roster generated by the Bureau was provided to the research team. This
allowed the researchers to identify inmates by level of mental health condition. (Four
levels exist, with Level 1 reflecting no significant mental health illness.) A sample was

then produced a few days in advance of the site visit. There was some very limited
duplication of the inmate and mental health evaluations that were completed. The
overall response rate for the mental health reviews was slightly higher than for the
inmate reviews (81 percent). The interviews were designed to allow CNA consulting
psychiatrists to offer their professional assessment.
Table 4 summarizes the completion rate for the mental health interviews by facility.
Table 4.

Inmates interviewed and evaluated regarding mental health status
by facility
Facility

Not completed

Completed

Completion rate

Allenwood

100%

Atlanta

100%

Butner

89%

Coleman

56%

Florence

89%

Hazelton

73%

Lewisburg

93%

Terre Haute

64%

Tucson

62%

Victorville

100%

Total

180

81%

Source: CNA/JFA Institute.

Quantitative data analysis—individual level
The study also sought to conduct a more comprehensive assessment of the SMU and
SHU population using the Bureau data systems. Like most if not all state correctional
data systems, the Bureau’s data systems are not designed to directly measure the
status of restrictive housing in terms of admissions to SHU and SMU, release from
these same statuses, and the current restrictive housing populations. For this and
other reasons, securing the necessary data and readying them for statistical analysis
took much longer than originally projected.3

3
Another reason for this delay was the decision by the Bureau to not allow the CNA
researchers to have regular and informal contacts with the Bureau’s Office of Research and
Evaluation (ORE) staff. Instead formal meetings had to be convened and emails transmitted
through the Contracting Officer Representative (COR) at NIC. This process resulted in lengthy

Beginning in November 2013, the Bureau first attempted to provide such information
for the SMU and ADX populations. This was done by creating multiple large data files
that captured all movements in and out of SMU and ADX status since 2009. These
files were then merged and analyzed to create a snapshot of the SMU and ADX
populations based on the absence of a release movement from that status. This was
achieved in May 2014.
A similar effort was made to create cohorts of SMU and ADX releases. These cohorts
were needed to calculate the length of stay in SMU and ADX status. Our goal was also
to conduct a recidivism study to determine what percentage of SMU and ADX
releases were returning to restrictive housing and for what reasons. After many
efforts to create such release cohorts, Bureau researchers opted to create the cohorts
for CNA using their considerable and intimate knowledge of the Bureau SENTRY data
system. These cohorts were finally established in July 2014, leaving little time to
conduct the analysis within the project schedule.
Until very recently, there have been no data system capabilities within the Bureau to
evaluate the much larger SHU population. The new special housing unit application
system is a stand-alone data system that has not been used by the Office of Research
and Evaluation (ORE) staff for evaluation purposes. The ORE team made several
attempts, without success, to create files from this system that could then be used to
create SHU release and snapshot data files. It was decided in June 2014 that the
Bureau would only be able to create a current SHU population listing with which
more complete data could be merged. It was also mutually agreed that a SHU release
cohort could not be produced for this study. The SHU snapshot file was produced in
July 2014, which did not permit sufficient time to conduct a comprehensive analysis
of that population as originally intended.

Quantitative data analysis—aggregate level
The study design also proposed to review trends in the number of Bureau inmates in
restrictive housing over time. The shift in Bureau policies that has led to a significant
increase in the SMU population, large transfers of the SMU population from Florence
to Lewisburg, and a significant decline in the SHU population have all significantly
affected the Bureau’s restrictive housing population over time. We also requested
that the Bureau provide the numbers of assaults on staff and inmates over time as
well as the number of lockdowns occurring each month.
delays in receiving answers to data questions and data files. The typical time frame for
receiving such data files for similar studies in Ohio, Mississippi, Oklahoma, Kentucky, Colorado
and Georgia has been 30-60 days.

Data on the SMU and ADX populations going back to 2004 were provided in a timely
manner. Data on staff and inmate assaults were provided in May 2014. The number
of lockdowns per month was not being collected on a systemic basis until 2008, and
those data were provided in June 2014.
Similar data on the number of inmates in SHU status per month were available in a
“dashboard” report that was available only in a portable document format (PDF).
These data were not retrieved in a spreadsheet format until July 2014, when ORE was
able to manually transfer the information for the CNA team.

Composition of the Bureau of Prisons special
housing
As described earlier, the Bureau operates three types of segregated housing units:
special housing units (SHUs), special management units (SMUs), and the
administrative maximum security (ADX) institution in Florence, Colorado. The Bureau
also operates communications management units. The conditions of confinement in
these units are similar to general population in that inmates are allowed to
participate in out-of-cell activities for up to 16 hours per day. The communications
management units were excluded from the scope of work of this study.
The Bureau’s program statements governing the three types of segregated housing
units indicate that all three have similar functions and purpose: to “separate inmates
from the general inmate population to protect the safety, security, and orderly
operation of Bureau facilities, and to protect the public.”4 The specific placement
criteria and conditions of confinement vary for each type of segregated housing unit,
as does the type of inmate housed in each.

Special housing units
As outlined in Program Statement 5270.10, Special Housing Units, the purpose of the
SHU is as follows:
Special Housing Units (SHUs) are housing units in Bureau institutions
where inmates are securely separated from the general inmate
population, and may be housed either alone or with other inmates.

Program Statement 5270.10, Special Housing Units, 28 C.F.R. 541.21.

Special housing units help ensure the safety, security, and orderly
operation of correctional facilities, and protect the public, by
providing alternative housing assignments for inmates removed from
the general population.
The policy further indicates that inmates placed in SHU are in either administrative
detention status or disciplinary segregation status.
Section 541.22 of the program statement describes administrative detention as
follows:
Administrative detention status is an administrative status, which
removes you from the general population when necessary to ensure
the safety, security, and orderly operation of correctional facilities, or
protect the public. Administrative detention status is nonpunitive,
and can occur for a variety of reasons.
The program statement permits placement in administrative detention status for the
following reasons:5
The inmate is awaiting classification or reclassification.
The inmate has been placed in holdover status and is awaiting or in transit to a
designated institution or other destination.
It is determined that the inmate’s removal from general population is necessary
because continued placement in general population poses a threat to life,
property, self, staff, other inmates, the public, or to the security or orderly
running of the institution and one or more of the following—

The inmate is under investigation or awaiting a hearing for possibly
violating a Bureau regulation or criminal law.

The inmate is awaiting transfer to another institution or location.

It has been determined that the inmate requires protective custody based
on the inmate’s request or staff determination that administrative
detention status is required for the inmate’s own protection.

The inmate has been determined to require postdisciplinary detention and
is ending confinement in disciplinary segregation status, and a return to

Program Statement 5270.10, Special Housing Units, 28 C.F.R. 541.23.

the general population would threaten the safety, security, and orderly
operation of a correctional facility, or public safety.
As noted in the above eligibility standards for placement in the SHU, inmates may be
placed in administrative detention status for protection. The program statement
outlines the following circumstances in which this can occur:6
Inmate has been determined to be a victim of inmate assault or threats.
Inmate has been confirmed to be an informant and his/her safety is at risk
because of providing, or being perceived as having provided, information to
staff or law enforcement authorities regarding other inmates or people in the
community.
Inmate has refused to enter general population because of alleged pressures or
threats from unidentified inmates, or for no specific expressed reason.
Based on evidence, staff believes that the inmate’s safety may be seriously
jeopardized by placement in the general population.
The statistical analysis of the SHU population was severely hampered by the inability
of the Bureau to provide an accurate data file of the inmates assigned to and released
from SHU status as well as an accurate snapshot of the current SHU population. At
that time, the count was 10,262 in over 100 Bureau facilities, with the population
steadily declining since then, reaching 8,939 by June 2014. This is a significant
reduction from the self-reported count of the SHU population of over 13,000 in 2011.
As with the SMU and ADX population, a very large percentage (66 percent) of the SHU
population has “separatee orders,” which means they have enemies and/or there are
safety and security concerns that prohibit specific inmates from being housed with
one another. This issue greatly restricts the Bureau’s ability to return SHU (as well as
SMU and ADX) inmates to the general population—even when considering transfers
to other Bureau facilities.
A further complication is the level of crowding that exists within the Bureau. As of
2013, the Bureau stated it was at 137 percent of its rated capacity. The rates of
crowding are even higher at their high- and medium-security facilities (154 percent
and 144 percent, respectively). Since many of the SMU inmates upon their release will
require placement in the high- and medium-security facilities, this level of crowding

Program Statement 5270.10, Special Housing Units, 28 C.F.R. 541.27.

further exacerbates the difficulty of transferring these inmates out of SHU in a timely
manner.
Table 5.

SHU populations by status, February 2013 – June 2014

Date

Total

Disciplinary
segregation

Investigation

Protective
custody

Pending
actions

2/2013

10,262

1,722

4,581

1,882

2,002

3/2013

10,070

1,700

4,516

1,868

118

4/2013

10,235

1,802

4,634

1,906

1,769

124

5/2013

10,086

2,041

4,355

1,854

1,714

122

6/2013

9,915

1,860

4,369

1,825

1,719

142

7/2013

9,821

1,542

4,707

1,787

1,675

110

8/2013

9,808

1,716

4,635

1,687

1,598

172

Other

9/2013

9,696

1,803

4,458

1,709

1,515

211

10/2013

9,530

1,771

4,483

1,613

1,516

147

11/2013

9,483

1,768

4,451

1,718

1,405

141

12/2013

9,434

1,506

4,567

1,562

1,726

1/2014

9,357

1,570

4,283

1,593

1,825

2/2014

9,484

1,424

4,750

1,532

1,718

3/2014

9,177

1,585

4,388

1,432

1,704

4/2014

9,096

1,533

4,336

1,380

1,766

5/2014

8,926

1,508

4,260

1,340

1,716

102

6/2014

8,939

1,376

4,252

1,361

1,802

148

Source: Bureau/ORE.

Special management units
As specified in Program Statement 5217.01, Special Management Units, inmates who
have participated in or had a leadership role in geographical group/gang-related
activity, and/or present unique security and management concerns may be
designated to an SMU, where enhanced and more restrictive management approaches
have been determined to be necessary to ensure the safety, security, or orderly
operation of Bureau facilities, or protection of the public.
The program statement defines the SMU as a nonpunitive program status that may
be appropriate for any inmate meeting the referral criteria as outlined in Section 2 of
P5217.01. The objective of the SMU status as outlined in that document is to enhance
a safe and orderly environment at all Bureau institutions.

Bureau policy permits placement in a SMU for any sentenced inmate whose history,
behavior, or situation requires enhanced management approaches that would ensure
the safety, security, or orderly operation of Bureau facilities, or for protection of the
public. The policy states that one or more of the following criteria must exist to
support consideration for placement in an SMU:
Participated in disruptive geographical group/gang-related activity.
Had a leadership role in disruptive geographical group/gang-related activity.
Has a history of serious and/or disruptive disciplinary infractions.
Committed any 100-level prohibited act.7
Participated in, organized, or facilitated any group misconduct that adversely
affected the orderly operation of a correctional facility.
Participated in or was associated with activity such that greater management of
the inmate’s interaction with other people is necessary to ensure the safety,
security, or orderly operation of the Bureau facilities, or protection of the
public.8
The SMU program has four levels or phases, differentiated by the conditions of
confinement and expected time frames for completion. Completion of all levels is
expected to occur within 18–24 months in the absence of any further behavioral
issues. Level 1 completion is expected within four months. Levels 2 and 3 are
expected to take from six to eight months each, and Level 4 is expected to take two
to four months.
Level 1—minimum stay four months. At this level, interaction between inmates is
minimal (for example, showers and recreation). All inmates are double bunked. The
associate warden is responsible for determining which inmates may be housed or can
participate in activities together, as necessary, to protect the safety, security, and
good order of the institution. Inmates are ordinarily restricted to their assigned cells.
Inmates participate in an institution and unit admission and orientation program as
outlined in the policy on admission and orientation. The goal of the SMU admission
and orientation program is to provide inmates with information regarding the
institution’s operations, program availability, and the requirements for successful

•P.S. 5270.09, Inmate Discipline Program.

Program Statement 5217.01, Special Management Unit, Section 2, Referral Process.

progression through each of the four levels of the program, based upon specific
goals established for each inmate. Institution staff will interact with each inmate on
an individual basis to assess the inmate’s program and counseling needs, discuss the
SMU program objectives and expectations, establish a set of program goals based on
the inmate’s individual needs and the programming available within the unit, and
communicate the requirements of the SMU program. Progression through Level 1 is
based upon the inmate’s compliance with behavioral expectations as established by
institution and SMU staff. A multidisciplinary special management review is
conducted by the unit manager, captain, and associate warden (chairperson) or the
person acting in that capacity. This review includes input from the SMU unit team,
correctional staff, psychology staff, education staff, and other appropriate staff to
determine the inmate’s readiness to progress to the next level. After the initial
programming assessment, Level 1 inmates are reviewed at least every 90 days.
Inmates are expected to progress to Level 2 after four months.
Level 2—minimum stay six to eight months. At this level, interaction between inmates
remains minimal. The associate warden is responsible for determining which inmates
may be housed or participate in activities together, as necessary to protect the safety,
security, and good order of the institution. Inmates are ordinarily restricted to their
assigned cells; however, out-of-cell activities and programming may be increased on
a case-by-case basis depending on behavioral performance. Inmates continue their
involvement in General Educational Development (GED) (or high school equivalency)
or ESL (English as a second language) education, either individually or in a classroom
setting. Initially at this level, inmates may be involved in programs on a self-study
basis; then, individual and small group counseling sessions dealing specifically with
treatment readiness and fundamental communication skills will be required.
The associate warden is responsible for determining which inmates will participate in
group activities. All program activities are intended to reinforce the goal of
coexisting and acting responsibly. Curriculum at this level targets “treatment
readiness skills” (such as basic empathy, attentiveness responding, respect, and
genuineness) to enhance inmate receptivity to the new concepts which they will be
exposed to in Level 3. Small-group counseling sessions, in particular, focus on
treatment readiness and fundamental communication skills. Progression through
this level is based upon the inmate demonstrating the potential for positive
community interaction.
During Level 2, inmates generally program and function separately. Progression to
Level 3, however, requires the inmate to demonstrate the ability to coexist with other
individuals, groups, or gangs. Level 2 inmates are reviewed at least every 90 days and
are expected to progress to Level 3 after six to eight months. Inmates who fail to
make satisfactory progress may be returned to a previous level.

Level 3—minimum stay six to eight months. Inmates at this level begin to interact in
an open but supervised setting with individuals from various groups, including open
movement in the unit with their demonstrated ability to effectively coexist with other
inmates. The associate warden is responsible for determining which inmates may be
housed or participate in activities together, as necessary to protect the safety,
security, and good order of the institution. There are also increased privileges (for
example, commissary and property) at this level for those who accomplish unit goals
and maintain appropriate conduct. Progression through this level is based upon the
inmate’s ability to demonstrate positive community interaction skills. Inmates are
formally reviewed by the unit team every 90 days and are expected to progress to
Level 4 after six to eight months. Inmates who fail to make satisfactory progress may
be returned to a previous level.
Level 4—minimum stay two to four months. At this level, inmates must be able to
demonstrate their sustained ability to coexist and interact appropriately with other
individuals and groups in the unit. The associate warden is responsible for
determining which inmates will participate in group activities. This level
encompasses the inmate’s last two-to-four months in the SMU program. Level 4
inmate reviews are conducted every 30 days and documented in the same manner as
previous reviews.
SMU inmates are reviewed by the unit team in conjunction with regularly scheduled
program reviews as provided in the policy on inmate classification and program
review. The unit team specifically reviews inmates for progression through the levels
of the program. An inmate’s institutional adjustment, program participation,
personal hygiene, and cell sanitation are considered during review for progression to
subsequent levels.
Progression through the program levels is dependent upon time in the specific level,
demonstration of appropriate behavior, and participation in programming goals. A
panel review is conducted at the end of each level to make recommendations
regarding progression. By policy, progression from Level 3 to Level 4 is based on the
“ability of the inmate to demonstrate positive ‘community’ interaction. It must also
be determined the inmate will likely meet the re-designation criteria.” Progression
from Level 4 to a general population facility is “based upon the inmate’s ability to
function in a general population setting with inmates of various group affiliations.”
Staff indicated that most team meetings for SMU inmates are done at the cell door.
There are only slight programming and operational differences between inmates in
Level 3 and Level 4. According to both policy and practice, Level 3 and Level 4
inmates may be housed in the same cell.
At the time that this review was initiated, there were SMU programs operating at USP
Allenwood, USP Lewisburg, and USP Florence. As of November 2013, the SMU census
issued by the Bureau showed a total of 1,680 inmates assigned to SMU status across

these three facilities (table 6). Since then, the Bureau has been transferring SMU
inmates to USP Lewisburg with the intention of eliminating the SMU program at USP
Florence.
Table 6.

SMU populations by facility, December 2013
Facility

SMU population

USP Florence

645

USP Lewisburg

786

USP Allenwood

249

Totals

1,680

Source: Bureau/NIC/ORE.

The Bureau was not able to provide a count of the SMU population broken down by
the four levels described above. USP Allenwood is supposed to contain only Level 3
and Level 4 SMU inmates, while USP Lewisburg and USP Florence house only Level 1
and Level 2 inmates. As discussed later in this report, we encountered inmates at USP
Lewisburg who had graduated from Level 2 but were being retained at that facility,
reportedly due to a lack of capacity at USP Allenwood. Assuming that USP Allenwood
does represent the Level 3 and 4 SMU populations, it is clear from table 6 that
85 percent of inmates in SMU are in Levels 1 and 2. Given the minimum length of
time required to complete Levels 1 and 2 (10–12 months), it may take an extended
period for many of these inmates to be returned to the general population.

Administrative maximum facility program
The ADX facility is located at the Federal Correctional Complex in Florence, Colorado.
The ADX is a high security facility housing maximum-custody-sentenced inmates in
single-occupancy cells. Maximum custody is the highest custody level that can be
assigned to an inmate.
This is the only facility of its type in the Bureau. The stated missions of the ADX are
to (1) assist the Bureau in maintaining the safety of both staff and inmates, while
eliminating the need to increase the security of other open population penitentiaries;
and (2) confine inmates under close controls while providing them opportunities to
demonstrate progressively responsible behavior; participate in programs in a safe,
secure, and humane environment, and establish readiness for transfer to a less
secure institution. The ADX houses inmates who require an uncommon level of
security due to their records of serious institutional misconduct, involvement in
violent or escape-related behavior, and/or who have unusual security needs based on
the nature of their offense. Placement of these inmates at another facility would pose
a risk to the safety and security of the institution, staff, and inmates and the public.

The institution operates three distinct programs throughout its numerous housing
units, which includes the general population and step-down program, control unit
program, and special security unit program. The ADX also has an SHU.
Referrals of inmates for placement at the ADX must be approved by the regional
director, the chief of the Bureau’s Designation and Sentence Computation Center
(DSCC), and the assistant director of the Correctional Programs Division in the
central office. All inmates referred for placement receive a hearing prior to
placement. Inmates may attend the hearing, make a statement, and present evidence
to the hearing administrator conducting the hearing. A mental health evaluation is a
required component of all referrals for placement at the ADX.
The facility census at the time of our review on March 31, 2014, was 410 inmates. Of
those, 75 were assigned to the control unit; 23 to the SHU control overflow; nine to
the SHU segregation unit; 19 to the J unit step-down program; 28 to the H unit
(housing inmates with SAMs); and 256 in general population. The H unit, which
houses the Special Security Unit Program, was excluded from this review under the
terms of the contract.
A snapshot of the census was taken on the first of the month for a 12-month period
from April 1, 2013, to April 1, 2014. During that time, the high census was 447 on
April 1, 2013, and the low census was 411 inmates on March 1, 2014, and on April 1,
2014. For the 12-month period, the average census for the facility was 428.
The three main components of the ADX that were reviewed in this assessment were
the general population step-down units, the special security unit (SSU), and the
control unit. Each of these is summarized below.
General population: Inmates in this portion of the facility meet the basic ADX
placement criteria. The purpose of the program is to monitor the inmate’s
adjustment to general population while providing an increasing level of privileges
and recreation access. The general population and step-down units have a fourphase, 36-month minimum program length. Inmates are gradually placed in less
restrictive housing and program conditions based on their adjustment to their
conditions of confinement.
Special security unit: The SSU houses inmates who have the need for more restricted
conditions or have a SAM authorized by the attorney general. SAMs may be deemed
reasonably necessary to prevent disclosure of classified information that would pose
a threat to national security if disclosed. SAMs include, but are not limited to, placing
an inmate in administrative detention and restricting social visits, mail privileges,
phone calls, and access to other inmates and to the media. While in the unit, each
inmate participates in a three-phase program, with each phase being less restrictive.

Inmates housed in the SSU are reviewed annually to determine if the SAM status
should be renewed or modified.
Control unit: The control unit program, established by Program Statement 5212.07,
provides housing for inmates who are unable to function in a less restrictive
environment without posing a threat to others or the institution. Referral to the unit
is outlined in PS 5212.07 and is reviewed by the regional director in the region which
the inmate is housed. If the regional director concurs with the placement, the referral
is submitted to the regional director of the North Central region, where ADX Florence
is located. The regional director then designates a hearing administrator to conduct a
hearing to review the placement referral. A mental health evaluation is a required
component of the referrals to the control unit, and medical, psychological, and
psychiatric concerns are considered during the review.
The hearing administrator conducts the hearing, which the inmate may attend and
present evidence, call witnesses, and receive the assistance of staff if necessary. The
decision of the hearing administrator is then submitted to the Executive Panel
(warden of ADX Florence, North Central regional director, and assistant director of
the Correctional Programs Division) for final review and placement.
Inmates placed in the control unit are reviewed within four weeks of initial
placement. Subsequent reviews are conducted on a monthly basis by the unit team,
while the Executive Panel is to review each inmate’s status and placement on a
quarterly basis.
The Psychology Services Branch reviews all ADX referrals, a process that has been in
place since 2012. Psychology Services reviews all cases prior to designation to the
ADX. Bureau staff reported that cases have been rejected for ADX placement based
upon this review, but no data were available on the number of cases in this category.

ADX and SMU population trends
The historical trends for both the ADX and SMU populations are shown in figure 2. A
dramatic increase in the SMU population began in 2009 and plateaued at about
2,000; the population recently declined through July 2014. The ADX population has
remained fairly stable at 425 since 2004.

Figure 2.

SMU and ADX populations, January 2004 – July 2014

After a dramatic increase in the SMU program that began in 2009, there have been
recent and significant reductions in the SHU and SMU populations (table 7). Similarly,
the Bureau population has also been slightly reduced, from a peak of 217,815 to
210,961 as of December 25, 2014. However, even with the decline in the Bureau
population, the proportion of inmates in some form of segregation is at 5 percent—
down from an estimated 6.9 percent in 2011. Much of the total decline is attributed
to a reduction in the SHU population and, in particular, inmates who have been
assigned to protective custody or are serving disciplinary segregation sanctions.
There have been no reductions in the number of SHU inmates being investigated or
the ADX populations over this time frame.

Table 7.

Bureau segregation populations by segregation type, 2004–2014
% of total
Bureau
population

Year

Total

SMU

ADX

SHU

Total
segregation

2004

180,328

393

439

2005

187,618

398

469

2006

193,046

472

532

2007

199,618

487

548

2008

201,280

466

564

2009

208,118

894

449

1,343

2010

209,771

1,357

427

1,784

2011

216,632

1,491

451

13,000

14,942

6.9%

2012

217,815

1,942

434

10,262

12,638

5.8%

2013

216,570

1,680

419

9,434

11,533

5.3%

2014

215,324

1,399

409

8,939

10,747

5.0 %

Source: Bureau/ORE.
Note: 2011 SHU population is a self-reported estimate by the BOP. The 2014 populations
are as of June 2014.

Summary of the Bureau’s recent initiatives
and segregation capacity
As the populations have been reduced, there have also been reductions in the
capacity of the program since its formation. Bureau Director Charles E. Samuels, Jr.
reported that an SMU had been closed at FCI Talladega, Alabama, where there was an
80-bed unit for Level 1 and 2 inmates. This SMU was deactivated in February 2013.
The director also reported that there had been plans to open an additional SMU
program at Oakdale, Louisiana, but that plan was withdrawn, again due to the steady
reduction in counts at the existing SMU programs and the absence of growth in
previous years.
The capacity of other units within existing SMU facilities has also been modified
based on the reduced SMU population—for example, the deactivated units at USP
Lewisburg that were observed during the site visit to that facility in January 2014.
G unit, which had a potential capacity of 162 SMU inmates, was closed at the time of
the site visit.
Also during the site visit to Florence, Colorado, the complex warden reported that
the SMU there was being phased out and the SMU inmates transferred to USP
Lewisburg. At the time of our site visit (April 2, 2014), the SMU at USP Florence had a

capacity of 750 and a population of 474. The capacity was split, with 374 beds for
Levels 1 and 2 and 376 beds for Levels 3 and 4. The available capacity was, in
practice, lower than 750, as D unit was being used as a general population unit and
other beds were housing the ADX step-down program. The warden indicated that
those in Levels 3 and 4 would remain at USP Florence until completion of the
program, while those in Levels 1 and 2 would gradually be transferred to USP
Lewisburg.

Development of the reintegration housing unit
The Bureau has also initiated a housing and program option for the protective
custody population, which makes up a significant portion of the overall SHU
population. It was reported that in November 2013 there were 1,437 protection cases
in special housing, with an additional 172 in special housing awaiting placement
back into the general population. In October 2013, the Bureau issued a memorandum
that provided field staff with procedures and criteria for placement of inmates in the
reintegration housing unit (RHU) located at the Federal Correctional Complex at
Oakdale, Louisiana.9
The target population for the RHU consists of male inmates who “consistently refuse
to enter general population at multiple locations” and those who have been
designated through the classification process as protective custody. The stated
purpose of the RHU is as follows:10
Remove inmates from SHUs and provide a less restrictive housing environment.
Address factors that cause inmates to refuse placement in general population.
Develop skills to increase amenability to entering general population.
Reduce continued transfers and SHU placement.
The preferred candidates and outcomes for the RHU include the following:
Target inmates who consistently refuse to enter general population at multiple
institutions and who have a general fear of placement in general population.

Memorandum to All Regional Directors, October 18, 2013, from Acting Assistant Director,
Correctional Programs Division titled FCC Oakdale – RHU Activation Procedures.

Reintegration Housing Unit PowerPoint.

Remove inmates from SHUs and provide a less restrictive housing environment.
Address factors that cause inmates to refuse general population.
Develop skills to increase amenability to entering general population.
Reduce continued transfers and SHU placement.
The initial capacity of the RHU was established as 160 beds with a potential for
future expansion to 320 beds. The RHU provides alternative housing outside the SHU
for protective custody inmates who are not security threat group members and who
could transition to general population housing.

USP Atlanta mental health unit
Finally, the Bureau has established a specialized unit for Mental Health Care Level 3
mentally ill inmates as an alternative to housing in the SMU, SHU, or ADX. This
program, the USP Atlanta Secure Mental Health Step-Down Program, has an initial
capacity of 24–30 and is primarily intended to remove some but not all Level 3
inmates with serious mental illness (SMI) from the ADX and other high-security USPs.
Inmates classified by the Bureau as Mental Health Care Level 4 are housed in the
Bureau’s medical referral centers.

Chapter 2: National trends and
use of segregated housing
Removal of disruptive and violent inmates from the general population and their
placement in separate housing units has been a common practice in prison systems
since their inception.11 In the United States, placement of inmates in solitary
confinement—the most extreme form of segregated housing—has been documented
as early as the 1800s, when administrators believed that silent contemplation led to
reform.12
Although the use and management of segregated housing have changed, the practice
of separating and isolating inmates using special cells or facilities has continued.13
The modern use of segregation and solitary confinement within specialized units and
facilities began to emerge in the 1970s, as prison populations began to rise, spurring
a series of highly publicized riots, prison violence, and increased prison crowding.14 It
was hoped that segregating the most disruptive inmates for extensive periods of
time under extreme forms of security would serve both to deter and to incapacitate
highly disruptive behavior.
By incapacitating disruptive inmates, centralized and specialized segregation units
would allow the vast majority of inmates who were conforming to the prison systems
rules and regulations to carry out their daily routines of work, recreation, and
program participation without the fear of violence or intimidation by more
aggressive inmates. It also allowed the other prisons to avoid lengthy lockdowns and
major disturbances.

C. Riveland. Supermax Prisons: Overview and General Considerations. National Institute of
Corrections Technical Assistance Number 98P4002. Jan. 1999.
12
D.P. Mears and J. Watson. “Towards a Fair and Balanced Assessment of Supermax Prisons.”
Justice Quarterly 23, no. 2, Jun. 2006: 232-270.
13
J.M. Pizarro, V.M. Stenius and T.C. Pratt. “Supermax Prisons Myths, Realities, and the Politics
of Punishment in American Society.” Criminal Justice Policy Review 17, no. 1, Mar. 2006: 6-21.
14
J. Wooldrege. “Research Note: A State Level Analysis of Sentencing Policies and Inmate
Crowding in State Prison.” Journal of Crime and Delinquency 42, no. 3, Jul. 1996: 456-466.
11

Three factors that influenced the rise of segregated housing deserve further
attention: (1) the significant increases in the nation’s state and federal prison
populations, (2) the attending increased crowding, and (3) the increased presence of
organized street and prison gangs.
After many decades of relative stability in the rate of incarceration, state and federal
prison populations began to accelerate in the 1970s, which served to disrupt prison
subcultures. In 1972 alone, 90 prison riots were reported by state and federal
officials.15 As noted by many criminologists, prison administrators depend heavily
upon a cooperative and conforming inmate population.16 Correctional officers are
greatly outnumbered by the inmate population at any given time and are armed with
few, if any, lethal weapons. Control is maintained by their daily interactions with
inmates and by offering differing levels of freedom of movement, privileges, and
activities (work, programs, and recreation) to mitigate the monotony of “doing time.”
However, as the prison populations grew, greater numbers of less experienced people
were needed to work in the expanding field of corrections.
Prison crowding also worsened as policy-makers passed legislation designed to
sentence more people to prison for longer periods of time. It simply was not possible
to build a sufficient number of prisons to accommodate the rising tide of inmates.
Crowding further exacerbated the level of violence in prisons and the need to better
control highly disruptive inmates.
Finally, the presence of modern organized street gangs within the prison population
increased, which served to further disrupt stability.17 Although prison gangs have
long been a prison management issue, the 1970s saw a new development in which
street gangs that were organized outside the prison system began to enter it in much
larger numbers.

Useem, B. & Piehl, A.M. (2008). Prison state: The challenge of mass incarceration. New York: Cambridge
University Press.
Starting with Donald Clemmer’s The Prison Community (New York: Holt, Rinehart, 1958),
social scientists produced an extensive literature on prisoner society as it existed from the
1930s through the 1960s. A few of the studies are Graham Sykes, The Society of Captives
(Princeton, NJ: Princeton University Press, 1958); Rose Giallombardo, The Society of Women
(New York: Wiley, 1966); David Ward and Gene Kassebaum, Women’s Prison (Chicago: Aldine,
1965); John Irwin, The Felon (Upper Saddle River, NJ: Prentice-Hall, 1970); James Jacobs,
Stateville (Chicago: University of Chicago Press, 1977), and; and James Austin and John Irwin,
It’s About Time: America’s Imprisonment Binge, 4th Edition (Palo Alto, CA:

17
Irwin, John. Prisons in Turmoil (Boston: Little, Brown, 1980). Jacobs, James, B. Stateville: The Penitentiary in
Mass Society (Chicago: University of Chicago Press, 1977).

There was also shift from what Ward and Carlson described as the dispersion
approach to the concentration approach to segregation housing.18 The dispersion
approach involved dispersing disruptive inmates and repeatedly transferring them to
different high-security prisons in the hopes of breaking up potential cliques or gangs
and preventing them from recruiting other inmates. It also provided some relief to
staff who had to manage these inmates on a daily basis. However, as prison
populations and the number of disruptive inmates grew, the efficacy of this
approach waned. It has now been replaced by the concentration approach, in which
specially designed facilities are constructed or older facilities renovated to permit the
long-term confinement of disruptive inmates in a highly controlled setting. These
facilities have relatively small housing units with cells that are difficult to damage,
enable staff to turn off water and electricity, and facilitate the use of force when
needed to conduct searches and cell extractions. Specially designed “cages” are
constructed to permit limited access to recreation, case managers, and medical and
mental health staff. The security staff are expected to be specially trained in working
in these prisons or units, which sometimes were known as “supermax” prisons.
The first forms of supermax and high segregation units can be traced to the Bureau’s
opening of Alcatraz in 1934 as a high-security penitentiary for “habitual” and
“intractable” federal inmates. After its closure in 1963, the Bureau experimented with
the dispersion model. With a rising population and increased levels of disruption, the
agency decided to once again concentrate its disruptive inmates at a special highcontrol unit at the Marion Penitentiary, which opened in 1978. In 1983, the deaths of
two officers and an inmate resulted in this prison’s conversion to indefinite
administrative segregation, or lockdown. Marion continued to house this population
until it opened a modern high-security, secure segregation unit called the
administrative maximum penitentiary in Florence, Colorado, in 1994. This unit is
now referred to as the ADX. Following the Bureau’s lead, supermax prisons and
housing units began to spring up in most state prisons systems as well as many of
the largest jail systems.
It must be emphasized that, in most jurisdictions, the proportion of segregated
inmates is relatively small. The last national survey, conducted in 2002, found that,
on average, 5 percent of the state prison population was assigned to some form of

Ward, David A. and N. A. Carlson. “Super-Maximum Custody Prisons in the United States.” Prison Service
Journal, Issue No. 97. April 1994.

administrative or disciplinary segregation.19 That same survey found significant
variation among the states with a range of 1 percent to 16 percent.

Definitions of segregation and types of
inmates in segregation
Segregated housing is known by a variety of names, including (but not limited to)
restrictive housing, segregation, administrative segregation, punitive segregation,
disciplinary segregation, isolation, control unit, special housing unit (SHU), special
management unit (SMU), intensive management unit, security control unit, and
supermax. For purposes of this report and based on several prior studies, the
following definitions of segregation from the prison general population are used:
Protective custody—The purpose is to protect an inmate from threats of violence
and extortion from other inmates. The inmate remains in this status until the
threats have been removed or the inmate is released from prison.
Acute/serious mental health needs—The purpose is to provide intense mental
health treatment to inmates with SMIs. The placement of an inmate and the
treatment plan are determined by the mental health team.
Acute medical needs—The purpose is to provide intense medical care to inmates
with life-threatening medical conditions and/or physical disabilities. The
placement of an inmate and the treatment plan are determined by medical
health professionals, including a psychiatrist or a physician.
Investigation segregation—The purpose is to temporally segregate an inmate
until serious allegations of misconduct or the need for protective custody is
determined. Once the investigative process is completed, the inmate can be
assigned to a segregation status or returned to the general population.
Disciplinary segregation—The purpose is to punish an inmate for a violation of
a major disciplinary rule. The inmate is to be released back to the general
population once the period of disciplinary segregation has been served.

Austin, James and McGinnis, Kenneth. Classification of High-Risk and Special Management :
A National Assessment of Current Practices. National Institute of Corrections. NIC Accession
Number 01946. June 2004.

Administrative segregation—The purpose is to incapacitate an inmate whose
presence in the general population would pose an ongoing threat to inmates
and staff. The placement of an inmate in this status is solely determined by a
limited set of criteria established by correctional administrators.20
In general, an inmate who is suspected of serious misconduct can be assigned to a
segregation unit until an investigation is completed. If convicted of the charges, the
inmate can then be sentenced to a specific period of time in segregation—often
referred to as disciplinary segregation. While most states limit the amount of time
one can serve in segregation, it is possible to accrue more time in disciplinary
segregation based on misconduct committed in the unit. Once the disciplinary
segregation time has been served, the inmate can then be placed in administrative
segregation for an indefinite period of time.
Some inmates do not pose a threat to the security of the prison system, but require
protection from other inmates. They are referred to as protective custody inmates. It
is not uncommon to find these inmates comingled with disciplinary and/or
administrative segregation inmates.21
Placement in these various forms of segregation is wholly determined by correctional
administrators. Segregation is allowed for a variety of reasons; consideration of the
offense committed, the number of infractions, and pending investigations all factor
into placement decisions. Some correctional systems further constrain placement
decisions to include only those instances in which evidence of specific harm or an
escape attempt is present.
Despite their overall low prevalence, highly disruptive offenders have historically
presented significant challenges for prison administrators and staff. Segregation
units by their very nature require higher levels of staffing on all levels (security,
medical, mental health, facility maintenance, programs and recreation, and legal
services). In response to this dynamic, corrections administrators have increasingly
developed and implemented a myriad of population segregation measures designed
to mitigate the impact of highly disruptive offenders within the correctional system.
A key issue is the nature and extent of isolation and solitary confinement. The most
recent and comprehensive survey of current practices by the state and the Bureau
20

H. Metcalf, J. Morgan, S. Oliker-Friedland, J. Resnik, J. Spiegel, H. Tae, A. Work, and B.
Holbrook. Administrative Segregation, Degrees of Isolation, and Incarceration: A National
Overview of State and Federal Correctional Policies. Yale Law School, Public Law Working
Paper No. 301. Jun. 1, 2013. Austin and McGinnis, 2004.

Metcalf, et al., 2013.

was completed by Metcalf et al. in 2013. They found that there was general
agreement that the practice of segregation involves removing inmates from the
general population and restricting their participation in recreation, group meals, and
programmatic offerings.22 The same report found that the degree of isolation varied
across systems, although the national standard is confinement to a cell at least 23
hours per day with very limited access to recreation, visits, and program services.
Recent reviews of segregation in Oklahoma, Kentucky, Illinois, Ohio, New York,
California, and the Bureau of Prisons have found that large numbers of the
segregated population are double celled rather than placed in a single cell, thus
negating the claims of solitary confinement.
In a study examining inmates’ experiences in administrative segregation, O’Keefe
noted a general lack of interpersonal contact, meaningful activity, access to reading
materials, and windows.23 According to Kupers et al., some of the particulars
associated with segregation practices include near 24-hour-per-day cell confinement
alone or with a cellmate, meals eaten in cells, limited trips to a recreation area, and
relatively infrequent noncontact interactions with family and friends.24
A somewhat contrasting view of segregation was presented by Berger et al., who
described cells that generally house two inmates and the ability to communicate with
inmates on either side of single-occupancy cells.25 They also noted that inmates may
talk with one another during recreation and with staff members during rounds.
Furthermore, segregation cells are similar in size to (or larger than) those in general
population, and meals served are similar to those received by general population
inmates. With respect to activities, inmates in segregation can receive mail, make
phone calls, participate in out-of-cell recreation and institutional programing
(including educational and religious services), and possess reading material, but on a
very restricted basis.

Metcalf, et al., 2013.
M.L. O’Keefe. “Administrative Segregation From Within: A Corrections Perspective.” The
Prison Journal 88, no. 1, Mar 2008: 123-143.
24
Kupers, T.A., T. Dronet, M. Winter, J. Austin, L. Kelly, W. Cartier, T. J. Morris, S. F. Hanlon,
E. L. Sparkman, P. Kumar, L. C. Vincent, J. Norris, K. Nagel, and J.McBride. “Beyond Supermax
Administrative Segregation: Mississippi’s Experience Rethinking Prison Classification and
Creating Alternative Mental Health Programs.” Criminal Justice and Behavior 36, no. 10, Oct.
2009: 1037-1050.
25
R.H. Berger, M. P. Chaplin, and R. L. Trestman. “Commentary: Toward an Improved
Understanding of Administrative Segregation.” The Journal of the American Academy of
Psychiatry and the Law 41, no. 1, Mar. 2013: 61-64.
22
23

As will be shown below, inmates in segregation often have more frequent contact
with medical, dental, and mental health service providers, all of whom are often
mandated to make daily rounds of segregated housing units. It can also be argued
that inmates in segregation may also enjoy greater privacy, the ability to eat meals
without interruption, a reduced likelihood of victimization and injury, and possibly a
lower probability of committing infractions that would serve to increase their prison
terms. These so-called benefits of segregation manifest themselves in inmates who
refuse to be released to the general population, instead preferring to complete their
prison sentence in segregation status.

Key litigation issues
There are key legal issues that correctional agencies must address in the operation of
their segregation units. A growing number of cases being filed against correctional
agencies are challenging the constitutionality of segregation. Our intent is not to
provide a comprehensive legal analysis of these issues but to summarize key issues
and court decisions.
Virtually all of the pending and past litigation on the use of segregation focuses on
the placement process, conditions of segregation, and duration of segregation. These
three issues are linked to constitutional requirements of due process and the
imposition of cruel and unusual punishment. The 5th and 14th Amendments to the
US Constitution provide due process protections. The 8th Amendment bans cruel
and unusual punishment.
Fred Cohen’s recent comprehensive review of case law and Supreme Court decisions
showed that the use of solitary confinement and segregation for extended periods of
time is constitutional and does not necessarily violate the 8th Amendment.26 He
noted that there are many cases where the courts have ruled that excessive
deprivation of basic services and conditions of confinement (for example,
objectionable food, insufficient clothing, insufficient heat, lack of lighting, or lack of
mental health services) does constitute cruel and unusual punishment. Further, in
several cases the courts have ruled that correctional officials must provide some
level of due process in determining initial and continued placement in certain
segregation housing conditions.

26
Cohen, Fred. “Penal Isolation: Beyond the Seriously Mentally Ill,” Criminal Justice and Behavior Vol. 35, No. 8,
August 2008, 1017-1047.

Although many cases have direct bearing on these issues, three major cases that are
narrowly related to segregation are frequently referenced by the courts:
Sandin v. Conner, 515 U.S. 472, 484 (1995)
Wilkinson v. Austin, 545 U.S. 209 (2005)
Estate of DiMarco v. Wyoming Department of Corrections., 473 F.3d 1334, 1342
(10th Cir. 2007)
The Sandin case was decided by the U.S. Supreme Court. It involved an inmate in the
Hawaii prison system who received a 30-day sentence to segregation. He was not
allowed to call witnesses on his behalf. The Court ruled that the prisoner was not
able to show that 30 days in segregation within the Hawaii prison system constituted
“atypical and significant hardship on the inmate in relation to the ordinary incidents
of prison life.” Consequently, there was not a protected liberty interest that would
entitle him to due process procedures.
In the Wilkinson case, which was brought against the Ohio Department of
Rehabilitation and Correction and its recently created supermax facility, the U.S.
Supreme Court unanimously held that recently adopted policies were sufficient to
address the due process issues. The issue of cruel and unusual punishment was not
addressed by the Court, as it had been previously settled by the parties.
The DiMarco decision was based on the use of segregation within the Wyoming
Department of Corrections. The case involved an anatomical male living as a female,
who sued the Department of Corrections for being placed in solitary confinement for
her own protection. The Tenth Circuit identified four factors that need to be
assessed to determine if a liberty interest existed relative to the proper use of
segregation:
1. Segregation relates to and furthers a legitimate penological interest, such as
safety or rehabilitation.
2. Segregation conditions of confinement are extreme.
3. Placement in segregation increases the duration of confinement.
4. Placement is indeterminate.
In its ruling, the Tenth Circuit found all four factors justified the placement of Ms.
DiMarco in isolation.
The Austin decision was focused on the Ohio State Penitentiary, which is the state’s
supermax facility. The state had already settled with the plaintiff on conditions-ofconfinement claims, which significantly and positively changed the range of services

and privileges afforded inmates. In particular, a step-down level system was created,
which permitted the vast majority of inmates to work their way out of segregation
and back to the general population.
These reforms left the U.S. Supreme Court to focus on due process issues. When the
Ohio State Penitentiary first opened, the Court observed the following:
There is no official policy governing placement that was in effect, and
the procedures used to assign inmates to the facility were
inconsistent and undefined, resulting in haphazard and erroneous
placements.
In response to the initial litigation, Ohio officials significantly narrowed the criteria
by which a prisoner could be assigned to the Ohio State Penitentiary. Further, the
authority to transfer prisoners to or release them from the Ohio State Penitentiary
could only be approved and authorized by the central classification office. The
Supreme Court held that the new criteria and review procedures did provide
sufficient due process protection. It is worth noting that, with the new policies in
place, the Ohio State Penitentiary segregation population dropped significantly.

Standards regarding segregation
Despite the widespread use of segregation, there is a lack of accepted guidelines or
standards governing its use. In particular, criteria and process for placement in
segregation, conditions of segregation, and criteria and process for release from
segregation vary substantially by state.27 For example, Nebraska reports a total of 19
reasons that would justify segregation placement, including “any other information
regarding the inmate that the classification authority deems appropriate.”28
Conversely, Mississippi allows only five such rationales for segregation placement.
There are differences among the states in due process/notification procedures. Most
states require a formal hearing and some form of written notice to the prisoner prior
to reaching a decision on whether to place him/her in segregation. Such hearings
usually must be held within 14 days of the notice. Very few states allow inmates to
have a legal representative or advocate present at the hearing. Standards for periodic
reviews are even less defined in agency policies. These reviews can occur as often as
every six months, with most states requiring 30- or 60-day reviews. Only a few states
27

Metcalf, et al., 2013.

Metcalf, et al., 2013:6.

report a requirement for a face-to-face interview with the inmate as part of the
review.29 When a face-to-face interview is not mandated, these reviews may simply be
case file paper reviews with a notice sent to the prisoner after the decision has been
reached.
It was not until 2013 that the Association of State Correctional Administrators
(ASCA) Administrative Segregation Sub-Committee examined the issues surrounding
segregation and provided recommendations regarding the use of restrictive
housing.30 The subcommittee’s final recommendations to correctional systems and
administrators on the use of administrative segregation only were published as
follows:

Provide a process, a separate review for decisions to place an offender in
restrictive status housing.

Provide periodic classification reviews of offenders in restrictive status
housing every 180 days or less.

Provide in-person mental health assessments, by trained personnel, within 72
hours of an offender being placed in restrictive status housing, and periodic
mental health assessments thereafter including an appropriate mental health
treatment plan.

Provide structured and progressive levels that include increased privileges as
an incentive for positive behavior and/or program participation.

Determine an offender’s length of stay in restrictive status housing on the
nature and level of threat to the safe and orderly operation of general
population as well as program participation, rule compliance and the
recommendation of the person(s) assigned to conduct the classification review
as opposed to strictly held time periods.

Provide appropriate access to medical and mental health staff and services.

Provide access to visiting opportunities.

Provide appropriate exercise opportunities.

Provide the ability to maintain proper hygiene.

Metcalf et al., 2013: pp 11-13.

Association of State Correctional Administrators, Administrative Segregation Sub-Committee.
Final Restricted Status Housing Policy Guidelines. Aug. 9, 2013.

10. Provide program opportunities appropriate to support transition back to a
general population setting or to the community.
11. Collect sufficient data to assess the effectiveness of implementation of these
guiding principles.
12. Conduct an objective review of all offenders in restrictive status housing by
persons independent of the placement authority to determine the offenders’
need for continued placement in restrictive status housing.
13. Require all staff assigned to work in restrictive status housing units to receive
appropriate training in managing offenders on restrictive status housing
status.
Although the ASCA guidelines provide the first effort to standardize the use and
conditions of segregation, they offer little in terms of specifics. For example,
guidelines 7 and 8 state that visits and recreation are to be provided, but fail to state
the number of visits or recreation periods per month and their minimum duration.
The recommendations are also silent on whether inmates requiring protective
custody should be placed in administrative segregation and under what conditions of
confinement.

Segregation sentencing structures
There are two basic models for committing a person to segregation status that mimic
the indeterminate and determinate criminal sentencing structures. Most states use
relatively short sentences for placing an inmate in disciplinary segregation, which
often can range from five to thirty days. However, if an agency wishes to continue the
segregation status beyond that time frame, they have developed two models.
One model is comparable to an indeterminate sentencing structure, where release
from administrative segregation is not specified and occurs only at the discretion of
the agency. Under this structure, the inmate is committed to segregation with no
specified release date. States including Ohio, Mississippi, and Colorado have used
this model, often in tandem with a step-down, incentive-based program that is
discussed in the next section.
The determinate structure is essentially a segregation sentence that can range from
30 or more days to many years. Under this model, serious disciplinary offenses can
produce long, fixed segregation terms. The only way to mitigate these terms is for
the agency to reduce them at review hearings based on good conduct. States
including Kentucky, New York, and California use this model.

Impact of segregation systems on
institutional safety
The overriding objective of administrative segregation is to protect inmates and staff
by incapacitating high-risk inmates. But it may also serve to deter other inmates from
becoming involved in serious rule infractions or in acts of violence out of fear of
being segregated for long periods of time. With the recent advent of progressive stepdown programs, one can also argue that disruptive inmates are being successfully
treated or rehabilitated while segregated.
It has also been alternately argued that supermax prisons and/or increased use of
administrative segregation units are either excessively expensive or cost-effective.
There is consensus among correctional professionals that segregation units require
higher staff-to-inmate ratios and increased presence of medical and mental health
staff. Some have also argued that segregation increases the likelihood of mental
illness and suicide. However, it may be that segregation units reduce the danger of
violence in the general population, which allows the vast majority of the prison
system to operate with greater efficiency and effectiveness.
Although there have been several descriptive studies of administrative segregation,
Mears31 and Berger et al.32 both suggest that there have been few, if any, credible
studies on its outcomes. The primary methodological issue is establishing a control
or “counterfactual” research design that would answer the basic question of what
would have happened had the use of segregation not increased.
A study of three state prison systems found that the increased use of segregation
had no effect on inmate-on-inmate assaults, but did reduce the incidence of inmateon-staff assaults.33 A subsequent study in Illinois found the same results and offered
evidence that the number of lockdowns in the other facilities was reduced.34

Mears, Daniel P. 2013. “Supermax Prisons: The Policy and the Evidence.” Criminology & Public Policy V. 12,
(4): 681-720.
32

Berger, et al., 2013.

Briggs, Chad S., Jody L. Sundt, and Thomas C. Castellano. 2003. “The effect of supermaximum
security prisons on aggregate levels of institutional violence,” Criminology 41:1341–76.
34
Sundt, Jody L., Thomas C. Castellano, and Chad S. Briggs. 2008. “The sociopolitical context of
prison violence and its control: A case study of supermax and its effect in Illinois.” The Prison
Journal 88:94–122.
33

In Washington state, research found no difference in recidivism rates between
inmates who had experienced its supermax system and a matched group that had
not. However, the same study found that inmates released directly from supermax to
the community had a significantly higher recidivism rate than the matched group (69
percent rearrested for a felony versus 51 percent for the control group).35
Mears and Bales completed what is arguably the most sophisticated quasiexperimental study of supermax confinement in the state of Florida.36 Using complex
matching procedures, they found little if any difference in recidivism rates for
inmates who experienced supermax confinement and those who did not. More
significantly, there were no differences in recidivism rates based on how long a
person was in supermax. In other words, it made little to no difference if an inmate
was confined for 4, 8, 12, or 24 months—the recidivism rates were the same. The
study also tried to replicate the Washington state finding that inmates released
directly from segregation had higher recidivism rates. The authors concluded that it
did not, but they used a measure of “recency” and not direct segregation releases.37
The United States Government Accountability Office (GAO) reported in 2013 that
there are five states (Colorado, Kansas, Maine, Mississippi, and Ohio) where the use
of administrative segregation has been reduced and that there has been no adverse
impact on institutional safety.38 However, there may have been no association
between reducing the size of the segregation population and assault rates,
homicides, or other serious incidents; this does not by itself indicate a causal
relationship.
Conversely, the New York state prison system found a sharp decline in its rates of
assaults as it rapidly expanded the size of its SHU population.39 These data led the
agency to conclude that increasing the size of its SHU population caused the rate of
assaults to decline. The Bureau made a similar claim to the GAO, saying that the rise
in the SMU population was the reason assault rates and the number of lockdowns

Lovell, David, L. Clark Johnson, and Kevin C. Cain. 2007. “Recidivism of supermax inmates in Washington
State.” Crime & Delinquency 53:633–56.
Mears, Daniel and William Bakes. 2009. “Supermax Incarceration and Recidivism.” Criminology 47.4:801-836.

37
Recency referred to inmates who were relatively close (within a few months) to their prison
release dates.
38

U.S. Government Accountability Office. May 2013. Bureau of Prisons: Improvements Needed in Bureau of
Prisons’ Monitoring and Evaluation of the Impact of Segregated Housing. Washington, DC: GAO.

Goord, G.S. 2006. Prison safety in New York. Albany: NYDOCS. Available online at
http://www.docs.state.ny.us/PressRel/06CommissionerRpt/06PrisonSafety Rpt.pdf

had declined.40 Despite these positive associations, it is not clear if there is a causal
relationship between segregation policies and institutional safety.

Mental health issues in segregated housing
Mental illness has become increasingly prevalent in corrections systems. According
to the Bureau of Justice Statistics (BJS), as of mid-2005, 56 percent (705,600) of state
and 45 percent (78,800) of federal inmates had some type of mental health problem
over the past 12 months.41 It should be recognized that the definition of “mental
health problem” was broad and included symptoms such as insomnia, sadness, loss
of appetite, and persistent irritability. The BJS survey also showed proportional
increases in the number of inmates who said they had used prescribed medication
for a mental health problem since admission to prison, used prescribed medication
for a mental or emotional problem, or received mental health treatment.
This is not to say that all or even a majority of these inmates require placement in
specialized treatment programs or housing units. The majority of them suffered
from some form of depression or mania, which is treatable by medication and
counseling. The number of inmates who require special housing due to a severe
mental illness (SMI) is much lower. According to Lovell, Allen, Johnson, and Jemelka,
reviews of clinical studies indicate agreement that 10 to 15 percent of inmates in
state prisons suffer from SMI.42 A 1999 BJS report on the same topic estimated that
16 percent of state and jail inmates and probationers were “mentally ill,” while only 7
percent of Bureau inmates were classified the same way.43
These figures are likely unsurprising to mental health clinicians and correctional
officers who have witnessed the relationship between deinstitutionalization and the
increasing number of mentally ill inmates in the correctional system.44

GAO, 2013, pp. 33-34.

D.J. James and L.E. Glaze, L. E. Mental Health Problems of Prison and Jail Inmates. Bureau of
Justice Statistics Special Report NCJ 213600. Sep. 2006 (revised Dec. 2006).
42
D. Lovell, D. Allen, C. Johnson and R. Jemelka, Ron. “Evaluating the Effectiveness of
Residential Treatment for with Mental Illness.” Criminal Justice and Behavior 28, no. 1, Feb.
2011: 83-104.
41

Ditton, Paula M. Mental Health and Treatment of and Probationers. Washington, D.C.: U.S. Department of
Justice, Bureau of Justice Statistics. July 1999.

A.L.S Brandt. “Treatment of Persons with Mental Illness in the Criminal Justice System: A
Literature Review.” Journal of Offender Rehabilitation 51, no. 8, Nov. 2012:541-558.

The BJS report also found that violations and fight-related injuries are common
among mentally ill inmates. Within the state-level prison population, 58 percent of
those with mental illness, compared with 43 percent of those without, committed
violations or were injured as a result of fighting. Among state and federal prison
populations, mentally ill inmates were also more frequently involved in physical or
verbal assaults on correctional staff or inmates. Mentally ill inmates are often more
disruptive than inmates without mental illness, and disruptive behavior associated
with mental illness often leads to the inappropriate placement of mentally ill inmates
in administrative segregation.
Given the above national statistics, it is not surprising that mentally ill inmates are
more frequently segregated and often spend a longer time in segregation.45 Mentally
ill inmates are also placed in segregation for protective custody reasons and because
of a lack of proper placement options. It has also been claimed that the experience of
segregation for lengthy periods of can result in an inmate decompensating and
developing mental illness(es). Some have suggested that many SMIs worsen due to
the stress of incarceration, presenting risks of self-injury and harm to staff or other
inmates.46
Some courts have held that placement in administrative segregation under certain
conditions is unsuitable for mentally ill, developmentally disabled, and nuisance
inmates.47 For example, in 1995 a federal court in California found that the
placement of certain inmates—those with mental illness such as, borderline
personality disorder, brain damage, mental retardation, chronic depression, and/or
impulse control disorders—in California’s Pelican Bay SHU constituted cruel and
unusual punishment, an 8th Amendment violation.48 In Madrid v. Gomez, the court
examined the placement of mentally ill inmates in the SHU at California’s Pelican Bay
State Prison, a supermax facility, and found that an 8th Amendment violation existed
for
those who the record demonstrates are at a particularly high risk for
suffering very serious or severe injury to their mental health,

J. Skeem, J. Peterson, and E. Silver. “Toward Research-Informed Policy for High Risk
Offenders with Serious Mental Illness.” In B. McSherry and P. Keiser, eds., Managing High Risk
Offenders: Policy and Practice. New York: Routledge. 2011. 111-121.
45

Ford, Julian, Robert L. Trestman, Fred Osher, Jack .E. Scott, Henry J. Steadman and Pamela C. Robbins. Mental
Health Screens in Corrections. National Institute of Justice Research for Practice. May 2007.

47
F.R. Maue. “Management of the Mentally Ill in Administrative Segregation: Legal and
Management Challenges.” Corrections Today, July 2006, pp 46-47.
48
Madrid v. Gomez, 889 F. Supp. 1146 (N.D. Cal. 1995)

including overt paranoia, psychotic breaks with reality, or massive
exacerbations of existing mental illness as a result of conditions in
the SHU. 49
The Madrid court also found that inmates did not undergo psychiatric screening
prior to placement in the Pelican Bay State Prison SHU, and that the provision of
mental health services to SHU inmates with SMI was noticeably lacking.
Although many researchers and experts agree that administrative segregation is
generally not a suitable placement option for inmates with SMI, the evidence on
whether such placement causes deterioration among mentally ill inmates is mixed.50
For example, a 2010 study of administrative segregation in the Colorado corrections
system found that not only did inmates with and without mental illness not
deteriorate while in segregation, but some actually exhibited signs of improvement.51
The authors of the study warned of its limited generalizability and the need for
replication. In supportive reaction to the Colorado study’s findings, Berger et al.
explained that context matters, and that some mentally ill inmates seek placement in
segregation as a way to decrease interpersonal and environmental stimulation.52
These inmates, according to Berger et al., exhibit tendencies (such as self-imposed
isolation) that are similar to individuals with SMI in a community setting. Other
inmates self-select into segregated housing as a means of avoiding foes and reducing
safety risks that they believe exist in general population settings.
Conversely, Haney noted that personal accounts, descriptive studies, and systematic
research published over several decades have substantiated that solitary
confinement is associated with negative psychological effects, and that these effects
are particularly pronounced among mentally ill inmates.53 This evidence, Haney
asserted, is bolstered by findings from other areas of inquiry that demonstrate the

Madrid v. Gomez, 889 F. Supp. 1146 (N.D. Cal. 1995)

B.A. Arrigo and J.L. Bullock. “The Psychological Effects of Solitary Confinement on Inmates in Supermax
Units.” International Journal of Offender Therapy & Comparative Criminology 52, no. 6, Dec. 2008: 622-640.

M.L. O’Keefe, K. J. Klebe, A. Stucker, K. Sturm, and W. Leggett. One Year Longitudinal Study
of the Psychological Effects of Administrative Segregation. National Institute of Justice Document
232973. Oct. 2010.
52
R.H. Berger, M. P. Chaplin, and R. L. Trestman. “Commentary: Toward an Improved
Understanding of Administrative Segregation.” The Journal of the American Academy of
Psychiatry and the Law 41, no. 1, Mar. 2013: 61-64.
51

Haney, Craig. “Prison Effects in the Age of Mass Incarceration.” The Prison Journal 92, no. 4,
2012:1-24.

negative effects of acute sensory deprivation, the effects of elderly loss of social
contact, and the consequences of isolating mentally ill patients.
Researchers have struggled to disentangle the relationship between high rates of
mental illness in the segregation population and the potential role that segregation
may play in the deterioration of mental health status. 54 It may be the case that higher
rates of mental illness in segregation reflect self-selection of mentally ill offenders
into segregation placements. Studies seeking to distinguish and closely examine the
effects of segregation have often been constrained due to small sample sizes, timing
of segregation placement relative to the study period(s), and reliance on quasiexperimental design. (True experimental design in this context is not feasible due to
ethical concerns associated with denial or delay of mental health or medical services
for research purposes.) Despite these constraints, research on mentally ill offenders
and the use of administrative segregation, along with relevant legal guidance, can
help frame discussions regarding appropriateness and application.
In an evaluation of mental health screening tools, Ford et al. explained that mental
health screenings and assessments are necessary to the timely identification and
effective treatment of inmate mental health needs.55 Furthermore, information
provided through screenings and assessments underlies the provision of services to
which inmates are constitutionally entitled and facilitate positive readjustment when
the inmate is released into general population or the community. Peters asserted that
without early identification of mental disorders, inmates are unlikely to seek
treatment, and missed identification of trauma undermines appropriate diagnosis,
which in turn may lead to inadequate participation in treatment, supervision, and
reentry planning.56
The guidance from ASCA on the use of restrictive housing suggests that facilities
should conduct an in-person mental health assessment within 72 hours of an
inmate’s placement in a segregated setting.57 Further, mental health assessments
should be performed by trained personnel and conducted on a periodic basis
following the initial assessment.
Despite the disproportionate share of infractions committed by mentally ill inmates
and the overrepresentation of inmates with mental illness in segregation settings, a
survey of state-level corrections disciplinary processes found that many state

M. L. O’Keefe, et al., 2013.
J. Ford, et al., 2007.
56
R.H. Peters. Assessing Dependence, Comorbidity, and Trauma. Mental Health Law & Policy
Faculty Publications Paper 642. 2013.
57
ASCA, Aug. 9, 2013.
54
55

correctional systems do not have formal policies on the role of mental health in the
disciplinary process.58 In some states, mental health plays no role or merely a minor
role for mentally ill inmates facing disciplinary charges. However, most states’
correctional policies require adequate medical and mental health treatment for
mentally ill inmates placed in segregation. For mentally ill inmates found guilty of
disciplinary infractions, most states have formal or informal policies directing
consultations with mental health professionals regarding inmate disposition.
The ACA National Commission on Correctional Health Care guidelines assert that
health care staff should “immediately review the health care needs of offenders
placed in disciplinary segregation to determine if there is any known health
contraindication to segregation placement.”59 These guidelines also stress that
processes must be implemented to ensure that mentally ill inmates in segregation
undergo continued evaluation that is conducted by a qualified mental health
professional, and significant inmate deterioration—indicating that segregation is no
longer suitable—warrants an alert to correctional administrators.
During the ACA 2013 Winter Conference plenary session entitled “Re-evaluating
Administrative Segregation: The Human, Public Safety and Economic Impact,”
correctional administrators highlighted the need for ACA standards regarding
segregation. Expert panelists identified several approaches that could enhance
compliance with the law,60 including the following:
Defining the types of mental illnesses that are incompatible with segregation,
and having mental health staff conduct screenings.
Providing access to in- and out-of-cell mental health treatment for segregation
inmates.
Creating individualized mental health services plans through multidisciplinary
treatment team collaborations.
Identifying mental illnesses in a timely fashion, because early identification
prevents deterioration.

M.S. Krelstein. “The Role of Mental Health in the Inmate Disciplinary Process: A National
Survey.” The Journal of the American Academy of Psychiatry and the Law 30, no. 4, Dec. 2002:
488–96.
58

59
Association of State Correctional Administrators, Administrative Segregation Sub-Committee. Final Restricted
Status Housing Policy Guidelines. Aug. 9, 2013.

J. Scafuri. “Administrative Segregation: Continuing the Conversation.” On the Line: An Online
Publication of the American Correctional Association, Sep. 2013.

Step-down programs
The ASCA‘s Guiding Principles for Restrictive Housing Status also suggested that
correctional systems provide segregated inmates with the opportunity to progress
through structured levels that link increased privileges to positive behavior and/or
program participation.61 Its recommendation was based on a number of states, as
well as the Bureau, that are actively working to reduce the number of inmates in
segregation through the development and implementation of step-down, intensive
management, and behavioral management programs.62
Metcalf et al. examined some of the step-down programs that attempt to link the
transition out of administrative segregation to achievement of specific goals,
including the completion of behavioral management plans and/or courses. Inmates
in transition programs are selected according to specific, stringent processes. A
minimum stay (generally six months to one year) is a common feature of these
programs, and inmates are made aware that any disciplinary infraction during the
program period will likely lead to additional time in segregation.
Several states, including Colorado, Massachusetts, Mississippi, Virginia, and
Washington, are developing or have developed programs that provide
administratively segregated inmates with increased opportunities for group activities
and therapy while maintaining a safe degree of separation. Connecticut,
Massachusetts, Mississippi, New Jersey, New Mexico, and Virginia have all reported
development of programs that target inmate behavior issues. New Mexico’s program,
for example, is structured around a two-level system that incentivizes positive
inmate behavior through a corresponding measured reduction of restrictions. More
detailed reports follow on the Virginia, Mississippi, Washington, and Maine stepdown programs.

Association of State Correctional Administrators, Administrative Segregation Sub-Committee.
Final Restrictive Status Housing Policy Guidelines. Aug. 9, 2013.

Virginia
The Virginia Department of Corrections (VADOC) is one of several state corrections
departments that have developed an administrative segregation step-down program
using evidence-based practices.63 The program claims to have reduced the
administrative segregation population and lowered prison safety incidents, each by
over 50 percent, and to have decreased inmate grievance filings by 23 percent.64
VADOC also reported a reduction in the use of sick leave and highlighted this finding
as an indication of reduced staff stress and improved morale, although it is not clear
how this assessment was made. The reported size of the Virginia program as of 2013
was 460 inmates and, as of August 2013, approximately two years after inception of
the step-down program, none of the enrolled offenders had returned to
administrative segregation.
The VADOC step-down initiative includes an enhanced classification review prior to
inmate assignment to or placement in segregation.65 Multidisciplinary staff teams and
validated instruments that determine criminal risks, underlying reasons for negative
individual inmate behaviors, and inmate motivators form the foundation of the more
extensive assessment. Throughout this initiative, inmates are provided cognitive
programming opportunities that promote learning and practice of positive behaviors.
The process includes an additional step-down classification security level that
provides a test bed for changed behavior. Segregation inmates who demonstrate that
they can participate appropriately in programs and control behavior can earn
additional responsibility.

Mississippi
In response to litigation regarding the use of segregation, the Mississippi Department
of Corrections (MDOC) changed its classification process and mental health
programming, leading to significantly decreased rates of violence, disciplinary
infractions, and use of force in the segregation unit.66 One of the specific changes
63
“Virginia Recognized for Transforming Highest-Security Prisons.” Virginia Department of
Corrections Press Release, Aug. 7, 2003.
64
“Virginia Step Down Program for Administrative Segregation.” Southern Legislative
Conference Press Release, 2013; These reductions occurred over approximately two years
following implementation of the VADOC Step-Down initiative.
65
Virginia DOC, 2013.
66
T. A. Kupers, T. Dronet, M. Winter, J. Austin, L. Kelly, W. Cartier, T. J. Morris, S. F. Hanlon,
E. L. Sparkman, P. Kumar, L. C. Vincent, J. Norris, K. Nagel, and J.Mcbride. “Beyond Supermax
Administrative Segregation: Mississippi’s Experience Rethinking Prison Classification and

made by MDOC was the development of a step-down unit for inmates with SMI, as
required under the 2007 Presley v. Epps consent decree.
Inmates requiring intermediate mental health treatment, but not inpatient
psychiatric services, are eligible for the unit, which is jointly administered by MDOC
and its medical and mental health contractor. The step-down unit, housed in MDOC’s
segregation facility, was designed to gradually move inmates with SMI from a
segregated setting to a more open one as they demonstrate appropriate behaviors
and the ability to function within an open unit. To be eligible for the step-down unit,
inmates must have a condition that is classified as an SMI. Inmates demonstrating
motivation to succeed in the unit are given priority. Once in the unit, inmates earn
their progress from the segregated tier to the open tier, and finally to general
population upon graduation.
The MDOC step-down unit relies on the “assertive community treatment” approach,
and staff focus on inmates’ “intact faculties, ambitions, positive life experiences, and
strengths of character, and how those buffer against disorder,” rather than on their
mental illness67. Inmates learn about their illnesses and the means for appropriately
addressing anger, impulses, and anxiety, and are rewarded via incentives (such as
additional time alone in activities rooms with media equipment, and use of
additional library materials) for positive behaviors. Although initial group treatment
among inmates who are still segregated includes the use of ankle restraints secured
to the floor, it still provides a necessary opportunity for interpersonal
communication and connectedness. Once inmates have transitioned to the open tier,
they participate in group sessions without restraints.
Staff selection, training, and collaboration are critical to the success of the MDOC
step-down unit. While respecting inmate confidentiality, mental health providers
meet on a weekly basis with correctional staff to ensure the delivery of a high level of
care in a secure environment. Correctional officers assigned to the unit must
complete an intensive mental health training curriculum and upon completion are
given the title of correctional mental health manager. The MDOC step-down unit’s
success is indicated by a reduction in rule violation reports filed against inmate
graduates. An examination of a cohort of 43 graduates revealed an average of 4.7
rule violation reports per inmate in the six months prior to unit admission. During
the cohort’s time in the unit, that figure fell to an average of 1.2 per inmate. In the

Creating Alternative Mental Health Programs.” Criminal Justice and Behavior 36, no. 10, Oct.
2009: 1037-1050.
67

Kupers, et al., p. 6

six months following unit graduation, the cohort exhibited an average of only 0.6 per
inmate.

Washington
Similar to the experiences in Virginia and Mississippi, the limited information
available regarding segregation reduction efforts in Washington indicates decreases
in behavioral incidents. The Washington Department of Corrections, in collaboration
with Disability Watch Washington and the Vera Institute of Justice, developed
“intensive management” or “intensive treatment” programs that allow for structured
group activities as well as a variety of therapy options for inmates housed in
segregation. Inmates must participate to return to general population, and are
assigned to specific programs based on individual mental health and behavioral
assessments. Programming delivery formats include self-directed, cell-front and
classroom, and course offerings include cognitive programming.68 It is now rare for
an inmate in the Washington State corrections system to spend more than 90 days in
segregation, and the proportion of inmates that return to segregation has
decreased.69 In the past, inmates released from segregation came back more than 50
percent of the time. Since the inception of the intensive management program, 131
inmates have graduated and of those, only 24 have returned.

Maine
The Maine Department of Corrections conducts risk assessments for each
administrative segregation inmate, and the information from the assessment is used
to develop individualized behavioral programs. The unit team reviews inmate cases
weekly and determines whether inmates will be provided the opportunity to
participate in group recreation and therapy. Program participation/attendance is
required and can include in-cell as well as individual and group counseling. While in
the program, inmates are under an incentive system that allows them opportunities
for increased amounts of out-of-cell time, more recreation time, fewer restraints, and
access to additional property. Following the policy changes, the Maine Department of
Corrections has not observed increased violence or other problems in the general
population. Additionally, although the policy changes are relatively new, both
behavioral incidents and the amount of time spent in segregation have decreased
since the changes were implemented.

68
69

Association of State Correctional Administrators. Segregation Survey. Jan. 7, 2013.
J. Martin. “State Prisons Rethink Solitary Confinement.” Seattle Times, Jan. 7, 2013.

Reductions in segregation populations
A number of states have reduced their segregation populations. As suggested above,
these declines have occurred by narrowing the criteria for placement in segregation
and/or reducing the period of segregation. The latter is often achieved by
implementing a step-down program or by adjusting the lengths of segregation
sentences. Following are some examples of states that have accomplished such
reductions.

Mississippi
Prior to 2006, Mississippi had approximately 1,000 inmates assigned to its highsecurity unit 32 at the Mississippi State Penitentiary prison complex in Parchman. In
response to ongoing litigation that contested the criteria for placement in
segregation and the conditions of confinement, the Mississippi Department of
Corrections developed a plan to reduce this population over 12 months. The reforms
centered on the following tasks:
1. Develop new criteria to limit the basis for admission to long-term segregation.
2. Using the new criteria, review all inmates in segregation at Parchman to
determine who should be immediately transferred to other facilities.
3. Remove all inmates with a SMI and transfer them to the MDOC’s mental health
facility.
4. For the remaining inmates, create a step-down program that would allow the
inmates to be released to the general population within nine months.
By the close of 2008, the Parchman segregation population was below 100. As of
2014, there are fewer than 230 inmates in long-term segregation out of a total prison
population of 21,148 (or 1 percent of the total prison population).70
The remaining segregation population had been moved by the MDOC to private
prisons located in the state that operate on contract with the MDOC. The most
recent data indicates that a total of 280 long term segregation inmates are
maintained in the system of which 109 are mental health cases. These inmates are

Kupers, T.A., et al., 2009.

housed at the East Mississippi Correctional Center, which is a private prison operated
by the Management and Training Corporation (MTC).71

Colorado
In 2011 there were approximately 1,500 inmates in administrative segregation in the
Colorado Department of Corrections, or 7 percent of the entire prison population.
Since then the number of inmates in restrictive housing has steadily declined.72 This
was accomplished by narrowing the criteria for placement in segregation and
reducing the length of stay. In particular, a more structured step-down protocol was
established that allowed inmates to be released to general population within nine
months if their behavior was compliant.73 This involved a four-level system with
specific rules and privileges associated with each of the four 90-day periods.
After the recommendations were implemented, the segregation population began to
decline, reaching approximately 600 inmates by the end of 2013. Further
programmatic reforms were implemented by the department in 2014, which resulted
in the restrictive population dropping to below 200 (1 percent of the total
population) by December 2014. There have been no associated increases in rates of
institutional violence as the inmate population in restrictive housing has declined.74

Reentry and prerelease programming
The reintegration of inmates who have been in segregation for significant periods
back into the prison general population or the community is an increasingly
important issue. It requires a strategy that balances less restrictive placement of
inmates with institutional or community safety and security. Reintegration requires
adequate assessment and intervention and should afford inmates the time and
opportunity necessary to readjust to more regular human interaction and activity
levels prior to a complete transition.
Inmates transitioning out of a segregation environment are at greater risk of
recidivism than their general population counterparts. For example, within the
71

Mississippi Department of Corrections Fact Sheet, December 1, 2014.

Colorado Department of Corrections, http://www.doc.state.co.us/dashboard-measures.

Based on telephone interview with Colorado Department of Corrections officials.

Colorado Department of Corrections, http://www.doc.state.co.us/dashboard-measures.

Colorado Department of Corrections system, the recidivism rate for administrative
segregation inmates was between 60 and 66 percent, while the rate for general
population was only 50 percent.75 Furthermore, 12 percent of inmates released from
administrative segregation returned to segregation within one year; within two years,
20 percent had returned. One explanation for this trend is the comparative lack of
coping skills observed within the recidivist cohort. In addition, as reported earlier,
40 percent of the Colorado inmates released from segregation were being released
directly to the community with no period of decompression.
Lessons learned from community re-entry programs (such as Reentry Partnership
Initiatives76) provide a foundation for developing programs that transition inmates
from administrative segregation to lower custody levels or the community. Taxman,
Young, and Byrne asserted that a successful reentry program includes an
“institutional phase” characterized by a wide range of programming options
designed to prepare inmates for life within the community.77 Program options
include education, vocational training, life skills training, and individual and/or
group counseling (for example, individual motivational readiness treatment). As
applied to administrative segregation, the institutional phase could be viewed as the
period during which inmates in administrative segregation have been identified for
transition to a lower level of custody and are actively working toward movement out
of segregation.
It should be noted that, despite the widespread validity of the re-entry concept, there
is little evidence to date that such programs have been effective in reducing
recidivism. The national evaluation of 12 adult reentry programs78 found that while
participation in re-entry programs accelerated, there were only modest positive
results in post-release employment, housing, and freedom from substance abuse.
Furthermore, there were no discernable effects on recidivism rates. The problem with
75

A Bill for an Act Concerning Appropriate Use of Restricted Confinement. Sixty-eighth General Assembly of
Colorado. Signed into law Jun. 3, 2011.
Lattimore, Pamela K. and Christy A. Visher. 2013. “Prison Reentry Services on Short-Term
Outcomes: Evidence From a Multisite Evaluation. Evaluation Review, 37(3-4) 274-313.

F. S. Taxman, D. Young and J. Byrne. Targeting for Reentry: Matching 8 Needs and Services to Maximize Public
Safety. National Institute for Justice Document 196491. Aug. 20, 2002.
In 2008, Congress passed the Second Chance Act (P.L. 110-199), an effort to improve
outcomes for people returning to communities after incarceration. This first-of-its-kind
legislation authorizes federal grants to government agencies and nonprofit organizations to
provide support strategies and services designed to reduce recidivism by improving outcomes
for people returning from prisons, jails, and juvenile facilities. The Second Chance Act’s grant
programs are funded and administered by the Office of Justice Programs in the U.S.
Department of Justice. Source: http://csgjusticecenter.org/nrrc/projects/second-chance-act.

these reentry programs was that the “dosage” (length of programming) was
insufficient to produce stronger treatment effects.79
The implications of this study for reentry from segregation to the general population
are not clear. We noted earlier that prior studies suggest that exposure to supermax
conditions does not affect recidivism. There have been no published studies to date
on whether exposure to supermax or other forms of segregation reduce inmate reoffending within the prison system itself. Important unanswered questions are (1)
whether increasing or reducing the duration of segregation is of any value to
institutional safety, and (2) whether an improved reentry regime would help mitigate
that risk.

Summary
The use of various forms of segregation began to increase in the 1970s as prison
population growth began to accelerate. As segregation populations have increased,
so, too, have concerns about their effects on segregated inmates and overall
institutional safety. These concerns have led some states to reconsider their use of
segregation and the conditions of confinement. In particular, states are reviewing
their criteria for placement in segregation, conditions of confinement, and length of
stay. Several states are under consent decrees regarding these issues and other
constitutional matters.
The extent to which mental health issues are associated with these high-security
units has also been questioned. In particular, are inmates placed in such units who
have a diagnosed mental health issue being properly treated and/or should they be
placed in a separate mental health treatment unit?
There is little published research to date on the effects of increased use of
segregation. Published studies suggest that placement in segregation does not have a
positive effect on recidivism rates. States that have reduced segregation populations
have found no adverse impact on institutional safety. Still, many questions persist
among corrections administrators and other stakeholders.

Lattimore, Pamela V and Christy A. Visher. 2013. “Prison Reentry Services on Short-Term Outcomes: Evidence
From a Multisite Evaluation. Evaluation Review, 37(3-4) 274-313

Chapter 3: Analysis of the
restrictive housing populations
In this chapter, we present data on the key attributes of the three major restrictive
housing populations (ADX, SMU, and SHU) in the Bureau. Where available data
permit, we make comparisons between these three populations and the much larger
number of inmates who are not in restrictive housing and who are mostly housed in
the general population.
At the time that this study began, there were approximately 220,000 inmates
incarcerated in the Bureau on any given day. Of that number, only 5 percent were
housed in the ADX, SMU, or SHU. What follows is a review of this small percent of the
Bureau population, their admission and release trends, and their lengths of
confinement.

Population characteristics
Table 8 compares the demographic attributes of the restrictive housing population
and other Bureau inmates as of November 23, 2013. These and other comparisons
are based on snapshot data files produced by the Bureau’s Office of Research and
Evaluation (ORE). There are few major differences between the restrictive housing
populations and the general population, with the following exceptions:
1. The ADX and SMU restrictive housing populations are exclusively male.
2. The SMU population is disproportionately younger.
3. The ADX population is disproportionately older.
4. The SMU population is disproportionately black.
5. All segregated populations are disproportionately U.S. citizens.

Table 8.

Demographic attributes of ADX, SMU, SHU, and other Bureau populations,
November 23, 2013
ADX

SMU

SHU

Other

415

1,675

9,189

207,166

100%

98%

93%

25 or younger

11%

26–35

18%

46%

41%

33%

36–50

46%

42%

39%

41%

51 or older

35%

10%

19%

White

59%

46%

55%

60%

Black

38%

48%

40%

37%

Indian

Asian

16%

28%

30%

35%

Total population
Gender
Males
Females
Current age

Race

Hispanic ethnicity
Citizenship
United States

88%

87%

83%

74%

Mexico

12%

18%

Other

Source: Bureau/ORE.

Primary offenses and sentences
Table 9 summarizes the primary offenses for which inmates have been sentenced.
Members of the ADX population (and to a lesser degree, the SMU and SHU
populations) are far less likely to be sentenced for a drug crime, which is the
dominant offense for the nonsegregated Bureau population. Rather, ADX inmates are
more likely to have been sentenced for the violent crimes of homicide, aggravated
assault, robbery, and possession/use of weapons/explosives.
Regarding the type of sentences Bureau inmates are serving, ADX inmates are far
more likely to have a life sentence (39 percent) than inmates of the SMU (9 percent),
SHU (3 percent), or other Bureau facilities (2 percent). Predictably, the ADX inmates
without a life sentence also have far longer sentences (average of nearly 30 years),

have served longer periods of incarceration (15 years), and have much longer periods
of time left to serve (17 years).
It was also possible to estimate the number of inmates who were scheduled to be
released within 12 months (as of November 23, 2013). For the nonsegregated Bureau
inmates, about 30 percent or 55,430 were to be released in the next 12 months. In
contrast, only 2 percent of the ADX population was to be discharged from prison in
that time frame. The numbers were progressively higher for the SMU (9 percent) and
SHU (21 percent) populations. Overall, over 2,000 inmates in restrictive housing were
scheduled to be released within a year.
Table 9.

Primary offenses of ADX, SMU, SHU, and other Bureau populations,
November 23, 2013

Total population

ADX
415

SMU
1,675

SHU
9,189

Other
207,166

Primary offense

Drugs

12%

31%

35%

47%

Weapons/explosives

12%

28%

21%

14%

Homicide/aggravated Assault

32%

12%

Burglary/larceny

15%

Robbery

16%

13%

Immigration

10%

11%

Fraud/bribery/extortion

Sex offense

Miscellaneous

Missing/unsentenced

Source: Bureau/ORE.

These data have important implications for reentry programs and other related
policies. Clearly, the need for such programs will be significant for the inmates in
SHU and as well as the other nonrestricted inmate population; 57,525 of them were
scheduled for release over the subsequent 12 months, as shown in table 10. This
number should be fairly constant at any time. There is much less of a need for these
programs for the ADX and SMU populations. However, given the long periods of
placement in restrictive housing for these inmates, as described later in this chapter,
the need for reentry programs remains significant for those nearing release in ADX
and SMU housing.

Table 10.

Sentence and time served attributes of ADX, SMU, SHU, and other Bureau
populations, November 23, 2013
ADX

Attribute
Lifers

Inmates
161

SMU

SHU

Other

%
39%

Inmates
154

%
9%

Inmates
286

%
3%

Inmates
4,889

%
2%

Average

Median

Average

Median

Average

Median

Average

Median

Sentence

29 years

25 years

18 years

15 years

11 years

9 years

10 years

8 years

Time served

15 years

14 years

9 years

7 years

5 years

4 years

3 years

Time left to serve

17 years

11 years

9 years

5 years

6 years

3 years

4 years

2 years

Inmates

155

1,931

21%

55,430

29%

Releases within
12 months

Source: Bureau/ORE.

Medical and mental health care levels
The official medical and mental health status of the restrictive housing population is
quite similar to that of the overall Bureau population. As noted earlier, the mental
health care levels established by the Bureau are as follows:80
Level 1—no significant mental health care
Level 2—routine outpatient mental health care or crisis-oriented mental health
care
Level 3—enhanced outpatient mental health care or residential mental health
care
Level 4—inpatient psychiatric care
The medical care levels as defined by the Bureau are as follows:
Level 1—healthy or simple chronic
Level 2—stable or chronic care
Level 3—unstable, complex chronic care
80

PS5310.16, page 8, 5/1/2014

Level 4—medical referral center care required
The Bureau states that the mental health care level system is tied to resource needs,
not diagnoses. Specifically, an inmate with a mental illness may be classified at
mental health care level 1 if his/her treatment needs are minimal, or if he/she does
not need care on an ongoing basis or crisis-oriented care of significant intensity.
Mental health care levels are designed to identify inmates in need of more intensive
mental health resources—for example, individual therapy or residential
programming.
The majority (between 68 and 72 percent) of restrictive housing inmates have been
classified by the Bureau as healthy or only needing simple chronic medical care, and
not requiring any specialized medical treatment (table 11). These proportions are
virtually identical to those of other Bureau inmates (also 72 percent). Of those
requiring medical care, the level of care is mostly at the lowest threshold; between 26
and 30 percent are at “care level 2—stable, chronic care.”
Regarding mental health care level, the proportions of the restrictive population
determined to be in need of care or treatment are even lower, with the vast majority
assigned to care level 1. The ADX has the highest proportion of inmates at mental
health care levels 2 and 3, but they only represent 10 percent of the entire ADX
population. These proportions are comparable to the mental health care levels of the
populations in nonrestrictive housing.
As noted in the literature review, state prison systems have been reporting much
higher proportions of inmates in their segregated population units. The chapter on
mental health care does raise some questions on the accuracy of the mental health
ratings provided by the Bureau’s mental health staff and suggests that a higher
proportion of the segregated population may have a significant mental health
ailment.

Table 11.

Medical and mental health care levels of ADX, SMU, SHU, and other
Bureau populations, November 23, 2013

Medical and mental
health care levels
Total Cases
Medical

ADX

SMU

SHU

Other

415

1,675

9,189

207,166

Level 1

67%

72%

Level 2

30%

28%

26%

24%

Level 3

Level 4

Level 1

90%

94%

87%

93%

Level 2

Level 3

Missing
Mental health

Level 4
Missing

Source: Bureau/ORE.

Security and special management issues
As expected, the Bureau populations in restrictive housing pose special management
issues, which is why they have been placed in ADX, SMU, or SHU. The Bureau’s
classification system is designed to identify those inmates who pose the highest risk
to inmate and staff safety. Table 12 shows that, unlike other Bureau inmates, the
inmates in restrictive housing are classified at the highest security levels. Virtually all
ADX and SMU inmates are assigned to the “high” category. The SHU population is
classified predominantly as high (36 percent) or medium (40 percent) security, while
the rest of the Bureau population is largely classified as medium, low, or minimum
security.

Table 12.

Security and other management issues of ADX, SMU, SHU, and other
Bureau populations, November 23, 2013

Security attribute

ADX

SMU

SHU

Other

Total cases

415

1,675

9,189

207,166

Security level

Unclassified

Minimum

18%

Low

16%

41%

Medium

40%

29%

High

99%

36%

96%

93%

67%

40%

57%
10

52%
11.1

21%
9.3

8%
6.2

Other Issues
Separation required from
specific other inmates
Gang member
Average criminal history score
Source: Bureau /ORE.

The higher number of medium-security custody inmates is largely due to the high
number of protective custody inmates assigned to SHU status and inmates who are
being investigated for possible rules violations.
Virtually all ADX and SMU inmates have “separation” restrictions, which means that
they cannot normally be placed in the same Bureau facility with one or more specific
other inmates. Separation orders are often linked to rival gang affiliations, which are
also a common attribute of the ADX, SMU, and SHU populations. Together, the
separation and gang membership issues complicate efforts to double-cell inmates
within the ADX, SMUs, and SHUs as well as to release them to the general population.
The “Other” population has a lower but still noteworthy proportion of inmates with
separation orders: 40 percent.81
Finally, the restrictive housing populations have significantly higher criminal history
scores (CHSs). The CHS is a measure used by the federal courts to help determine
whether a person should be incarcerated and how long the sentence should be. The
higher the CHS, the more likely the person will be sentenced to prison and/or to a
longer prison term. The higher CHS for the segregated populations means they have
significantly more extensive prior convictions.

“Other” refers to all other Bureau inmates who were not in any form of segregation as of
November 23, 2013.

Time in restrictive housing
The one-day snapshot also makes it possible to measure how long the current
population has been in restrictive housing (table 13). ADX inmates have the longest
periods of continuous assignment, with an average of approximately four years. The
lower median number reflects inmates recently assigned to ADX and the small
number of ADX inmates with extremely long periods of continuous confinement.
The SMU population has an average stay in the program of 277 days. The SHU
population has a relatively short period of confinement of 76 days. The median
numbers are considerably lower for the same reasons as for the ADX population.
The available data did not permit an analysis of length of stay in SHU for the various
subgroups in this status (disciplinary segregation, administrative segregation,
protective custody). SENTRY (the Bureau’s inmate management/database system) and
other Bureau data systems do not track this status in terms of length of stay.
Table 13.

Length of time in ADX, SMU, and SHU status, November 23, 2013
ADX

SMU

SHU

415

1,675

9,189

Average

1,376 days

277 days

76 days

Median

941 days

211 days

40 days

Number of Inmates
Time In restrictive housing

Source: BOP/ORE.

Disciplinary conduct of the SMU and ADX
As suggested by the classification and special management data, inmates in ADX and
SMU have accumulated lengthy histories of misconduct. The average number of
disciplinary reports (DRs) for the ADX population prior to placement in ADX was 15
with a median of 7. For the SMU population, the numbers were even higher, with an
average of 17 and a median of 11 reports prior to placement in SMU.
For both groups, the number of reports since being placed in restrictive housing is
substantially lower. However, these comparisons of disciplinary histories before and
after restrictive housing do not account for the varying amounts of time different
inmates served in the Bureau before and after placement in restrictive housing.
Further, the number of DRs prior to the most recent placement in restrictive housing
includes DRs received during prior commitments to ADX and SMU.

To further isolate the possible effects of restrictive housing on inmate misconduct,
the analysis time frame was limited to the 12 months immediately before and after
placement in restrictive housing (Table 15). Although the 12-month average and
median numbers are lower than those based on total prior history, a suppression
effect does not appear. This means that the conduct of the inmates in restrictive
housing continues at the same rate that was occurring prior to placement in
restrictive housing. While the inmates have been incapacitated, there appears to be
little change on their behavior at least initially during the first 12 months of
restrictive housing.
Table 14.

Disciplinary reports for current ADX and SMU populations, as of November
23, 2013—total number of reports
Number of disciplinary reports

ADX

SMU

Number of inmates

415

1,675

Average

Median

Average

Median

Average

19.5

19.8

Median

DRs prior to placement

DRs after placement

Total

Source: BOP/ORE.

Table 15.

Disciplinary reports for ADX and SMU populations, as of November 23,
2013—reports in the 12 months before and after restrictive housing
placement
Disciplinary reports in past 12
months

ADX

SMU

Number of inmates

415

1,675

Average

5.1

4.3

Median

Average

5.6

4.6

Median

12 months before placement

12 months after placement

Source: BOP/ORE.

The types of infractions in the reports accumulated by ADX and SMU inmates both
prior to and while in restrictive housing status are quite varied but include a

significant number of very serious rule violations. The Bureau has classified all of the
possible violations into four levels of severity: greatest, high, moderate, and low. The
415 ADX inmates had accumulated about 4,500 infractions prior to being placed in
ADX, of which 65 percent were in the “greatest” or “high” levels.
For ADX, of the 50 “killing” incidents that include attempts and aiding and abetting,
23 involved injuries that proved to be fatal. Of the four that occurred after placement
in ADX, only one involved a fatal injury. For SMU, of the 24 “killing” incidents prior
to placement, two involved injuries that proved to be fatal. Neither of the two that
occurred after SMU placement resulted in a fatality. All 26 fatalities were inmates.
The 1,675 SMU inmates had accumulated 25,996 infractions of which 62% were also
in the “greatest” or “high” severity levels. Since being in ADX and SMU, the
accumulated numbers have declined as well as the proportions that are in the
“greatest” and “high” severity levels. This decline is expected given that the time
frame for the “before” period exceeds the time in SHU and ADX. And as noted earlier,
when one controls for time frames (12 months before and after), the number and
rate of DRs have not changed.
It should be emphasized that these results are largely descriptive in nature and not
intended to test the impact of restricted housing on inmate conduct. That type of
analysis was beyond the scope of this project and could not be supported by the data
provided by the Bureau.
Table 16.

Types of disciplinary infractions by ADX inmates, as of November 23, 2013
Before
ADX placement

Total
Greatest severity / 100 level
Killing
Assault with serious injury
Escape
Setting a fire
Possessing dangerous weapon
Rioting
Encouraging riot
Possessing hazardous tool
Refusing drug or alcohol test
Introducing drugs or alcohol
Using drugs or alcohol
Possessing drugs or alcohol

During and after
ADX placement

% of total

4,500
812
50
191

100%
18%
1
4

2,142
350
4
13

100%
16%
<1
<1

10
53
307
4
6
20
95
2
41
17

<1
1
7
<1
<1
<1
2
<1
<1
<1

0
2
33
0
0
7
278
6
3
2

0
<1
2
0
0
<1
13
<1
<1
<1

Before
ADX placement
Other “greatest severity”
High severity / 200 level

During and after
ADX placement

% of total

% of total
<1

2,148

47%

697

33%

Fighting with another person

116

Threatening bodily harm
Engaging in sex acts

503
308

11
7

201
143

9
7

Making sex proposal or threat
Interfering with security devices
Group demonstration

169
28

4
<1

31
2

1
<1

Destroying property over $100
Possessing intoxicants

156
117

4
3

79
24

4
1

31
581
50
55
1,490
118
111
619
165
86
106
66
219
50

<1
13
1.
1
34%
3
3
14
4
2
2
2
5
1%

46
111
13
25
1,083
43
78
506
63
180
50
30
133
12

2
5.
<1
1
51%
2
4
24
3
8
2
1
6
<1%

Refusing alcohol test
Assault without serious injury
Disruptive conduct—high
Other “high severity”
Moderate severity / 300 level
Possessing unauthorized item
Refusing work/program assignment
Refusing to obey order
Being insolent to staff
Failing to stand count
Destroying property $100 or less
Being unsanitary or untidy
Other “moderate severity”
All “low severity” / 400 level

Source: BOP/ORE.
Table 17.
Types of disciplinary infractions by current SMU inmates, as of
November 23, 2013
Before
SMU placement
N
% of total
Total
Greatest severity / 100 level
Killing
Assault with serious injury
Escape
Setting a fire
Possessing dangerous weapon
Rioting/encouraging riot

25,966
4,733
24
547
15
161
2,405
24

100%
18%
<1
<1
<1
<1
9
<1

During and after
SMU placement
N
% of total
5,061
694
2
19
0
23
233
0

100%
14%
<1
<1
0
<1
5
0

Before
SMU placement
N
% of total
Taking hostages

During and after
SMU placement
N
% of total

114
300

<1
1

10
305

<1
6

Introducing drugs or alcohol

Using drugs or alcohol
Possessing drugs or alcohol

541

397

12
47
16
11,402
1,289
1,471
2,254
226
937
100
391
959
159
2,498
100
288
730
9,624
1,015
905
3,144
874
872
422
153
2,239
207

<1
<1
<1
44%
5
6
9
1
4
<1
2
4
<1
10
<1
1
3
37%
4
4
12
3
3
2
<1
9
<1%

1
12
4
2,396
257
467
460
60
217
3
51
7
0
451
35
138
250
1,967
245
200
613
119
239
205
31
560
4

<1
<1
<1
47%
5
9
9
1
4
<1
1
<1
0
9
<1
3
5
39%
5
4
12
2
5
4
<1
11
<1%

Possessing hazardous tool
Refusing drug or alcohol test

Destroying/disposing of item
during search
Interfering with staff—greatest
Disruptive conduct—greatest
Other “greatest severity”
High severity / 200 level
Fighting with another person
Threatening bodily harm
Engaging in sex acts
Making sex proposal/threat
Interfering with security devices
Group demonstration
Destroying property over $100
Possessing intoxicants
Refusing alcohol test
Assault without serious injury
Interfering with staff—high
Disruptive conduct—high
Other “high severity”
Moderate severity / 300 level
Possessing unauthorized item
Refusing work/program assignment
Refusing to obey order
Being insolent to staff
Failing to stand count
Destroying property $100 or less
Being unsanitary/untidy
Other “moderate severity”
All “low severity” / 400 level

Source: BOP/ORE.

Recidivism rates of ADX and SMU inmates
To examine the question of how many inmates released from ADX or SMU
subsequently return to restrictive housing, a cohort of inmates released in 2011 was
created and recorded to determine whether they have returned to restrictive housing.
Although the number of ADX and SMU releases are relatively small, the overall return
to restrictive housing status are also low. Of the 66 ADX inmates released to the
general population, only 9 percent returned to ADX. Of the 585 SMU inmates
released, 19 percent have returned to SMU status (see table 18).
A much higher proportion of the releases do incur at least one additional disciplinary
report during the follow-up period. Of the inmates released from SMU, 84 percent
recorded another disciplinary report, with an average of 3.7 reports.
For the inmates released from ADX, the percentages are similar, 83 percent with a
new disciplinary report, but for a much lower average of 1.6 reports.
It is beyond the scope of this report to explain the higher prison recidivism rates for
the SMU releases. Clearly, there are substantial differences between the attributes of
the two cohorts that are related to institutional misconduct. For, example the
inmates released from SMU are younger than those released from ADX. Older
inmates are less likely to become involved in misconduct, so the maturation effect
may be greater for the ADX inmates. The small number of ADX inmates released
does not permit a multivariate analysis that could adequately control for age and
other factors possibly related to recidivism.
The larger number of inmates released from SMU makes it possible to examine the
bivariate relationship between time in SMU and recidivism (table 19). Most of the
releases are clustered in the 19–24 month range. The prison recidivism rates do not
vary significantly by shorter or longer periods of SMU confinement. Those inmates
who spent more than two years in SMU had a significantly higher rate of return to
SMU but an equivalent rate of new disciplinary reports. Younger SMU inmates also
tended to have shorter periods of SMU placement.
These data suggest that the amount of time in the SMU is not a strong predictor. In
fact, longer periods of time in SMU are associated with higher in-prison recidivism
rates. While the numbers are not large, they do raise the question of whether the
current time frame of 18 months or longer for completing the four-phase SMU
program is excessive. As noted in the literature review, some state systems have
shorter step-down programs, and these data support a move to moderate the current
length of the SMU program.

Table 18.

In-prison recidivism rates for inmates released from ADX and SMU

Attribute

ADX

SMU

Releases

585

Average age at release

44 years

36 years

Average time In segregation

6.8 years

1.7 years

5.4 years

1.6 years

5 (8%)

77 (13%)

Median
Total (%) released directly to streets
% returned to segregation (ADX or SMU)

19%

246 days

469 days

% with DR

83%

84%

Average number DRs after release

1.6

3.7

Average time to segregation return

Table 19.

Prison recidivism rates for inmates released from SMU by time in SMU
Length of Stay

Releases

% return

% new DR

18 months or less

146

16%

69%

19–24 months

346

17%

77%

25 months or more

29%

75%

Source: BOP/ORE.

Trends in assault rates and lockdowns
To what extent have overall rates of inmate assaults and the use of lockdowns
changed as the Bureau’s segregation populations have increased? One of the key
purposes of restrictive housing is to provide greater safety for inmates and staff as
the more disruptive and violent inmates are incapacitated by removing them from
the general population for extended periods of time.
Figure 3 compares the rise in the SMU population with the annual inmate assault
rate. The assault rate, which is low at less than three assaults per 100 inmates, has
remained stable since 2004. There has been a slight decline since the rapid increase
in the SMU population.
However, while the overall assault rate has remained constant, there may have been a
reduction in the high-security populations. Since most of the inmates placed in
restrictive housing were classified at higher security levels and housed at the USPs,
the reduction in assault rates may be limited to those facilities.
Table 20 shows that there has been a decrease in the assault rate at the Bureau’s
high-security prisons, suggesting that the rapid increase of the SMU population did
have an incapacitation effect. Further, there has been a significant reduction in

(1) the number of lockdowns at the various Bureau prisons and (2) the total number
of lockdown days (table 20). Again, no claim of causation can be made, as other
external factors may be related to these trends.

ADX and SMU populations and inmate assault rate, 2004-2013

2,500

10.00

Priosners in ADX or SMU

9.00
2,000

8.00
7.00

1,500

6.00
5.00

1,000

4.00
3.00

500

2.00
1.00

0.00

Assault Rate Per 100 Prisoner Population

Figure 3.

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
SMU
ADX
Assault Rate

Table 20.

Changes in ADX and SMU populations, assaults, and lockdowns, 2008–
2013

Lockdowns

Total
lockdown
days

148

1,210

8.2

111

917

8.3

10.4

129

877

6.8

2.3

9.1

773

8.3

526

6.1

706

9.9

Year

SMU

ADX

Assaults
per 100
inmates

2008

100

471

2.5

Assault rates in
high-security
prisons
10.5

2009

936

447

2.7

10.9

2010

1,398

433

2.6

2011

1,545

450

2012

2,042

434

2.4

8.7

2013

1,675

415

2.2

8.6

Source: BOP/ORE.

Days per
lockdown

Summary of major findings
Inmates placed in ADX, SMU, and SHU status represent a small proportion of the
Bureau’s prison population. They also differ from other BOP inmates on a number of
key attributes. In particular, they are disproportionately male and older, and more of
them are U.S. citizens. They tend to have been sentenced for a more serious/violent
offense and assigned to higher custody levels, to be associated with a gang (either
street- or prison-organized), and to have separation restrictions.
All three populations have substantial time left to serve on their sentences. However,
there are over 2,000 inmates in restrictive housing who will be released within a year,
which suggests the need for reentry services. Differences in sentence lengths, time
served, and time left to serve are especially pronounced among the ADX and SMU
inmates.
The majority of inmates in restrictive housing do not require any specialized medical
treatment and are virtually identical to other Bureau inmates. Somewhat surprisingly,
relative to mental health care level, the proportions of inmates that are reported in
need of care or treatment are even lower and comparable to the mental health care
levels of the nonrestrictive populations.
The SMU and ADX inmates have lengthy disciplinary records, which include repeated
histories of institutional violence and other types of serious misconduct. It does not
appear that initial placement in the SMU and ADX units has any impact on the rate of
misconduct that was occurring prior to placement in restrictive housing.
The vast majority of ADX and SMU inmates released in 2011 were not returned to
restrictive housing status, although most incurred another DR within two years.
There was no strong or consistent relationship between time in SMU and in-prison
recidivism rates, although inmates with the longest placement in SMU had a higher
rate of return to restrictive housing. This may be due to the inmates’ failure to
comply with the SMU program for a substantial period of time and thus their higher
risk of recidivism than that of inmates who complied and completed the program in
a shorter period of time.
The Bureau’s assault rate remained unchanged as the size of the SMU population
dramatically increased. However, there has been an associated decline in the rate of
assaults in the high-security prisons, the number of lockdowns, and the number of
lockdown days.

Chapter 4:

Due process

The project team conducted a comprehensive review of the application of inmate due
process rights in the assignment and management of inmates in special housing at
each site visited, including an evaluation of procedures to protect due process rights
during referral, designation, and throughout the duration of placement within SHU,
SMU, and/or ADX. This assessment reviews the application of the Bureau’s
disciplinary process and assesses its administrative remedy procedures.
The following documents were reviewed as part of the assessment:
Applicable decisions of the United States Supreme Court and circuit and district
courts
P.S. 5212.07: Control Unit Programs
P.S. 5217.01: Special Management Units
P.S. 5270.09: Inmate Discipline Program
P.S. 1315.07: Legal Activities, Inmate
P.S. 1330.17: Administrative Remedy Program
P.S. 5270.10: Special Housing Units
Special Management Unit Handbook
As outlined earlier in this report, data collection and analysis focused on
representative samples of inmates at selected facilities. The Bureau placed no
restrictions on the team’s access to inmate files. By closely examining the selected
inmates through case file review, observations, and interviews, the project team
developed a better understanding of the due process systems associated with
placement into restrictive housing within the Bureau. The file reviews of the sampled
inmates included a review of the basis of placement in restrictive housing, the review
process involved in placement, inmates’ progress through the programs, and the
process used to determine either release or transfer to another facility.

A number of recent disciplinary cases were also reviewed at each facility to more
fully understand the basis of disciplinary sanctions imposed and the compliance
with due process standards in administering discipline. The number of disciplinary
cases varied at each facility, but a review typically covered 30–50 of the most recently
completed disciplinary cases. The project team reviewed each case along with all
supporting documents, including the incident report and associated hearing
documents.
The case file review was supplemented by interviews with inmates, disciplinary
hearing officers (DHOs), and the administrative remedy coordinator at each facility.
Staff at the Central Office who directly managed due process functions were also
interviewed, as were staff at the Designations and Sentence Computation Center
(DSCC).82

Federal due process standards
The United States Supreme Court has held that prison inmates are entitled to certain
procedural protections before they can be deprived of protected liberty interests.
This is known as the right to due process. The Supreme Court holds that, when a
protected liberty interest exists, for example in the case of good-time credits, certain
minimum requirements are due to the inmate before this interest can be infringed
upon.
The inmate’s interest must, however, be considered in view of the fact that he is
incarcerated and the environment is secure in nature. In Superintendent v. Hill, 472
U.S. 445, 454-55 (1985), the Court recognized that the safeguards of due process are
to be considered in light of the “legitimate institutional needs of assuring the safety
of inmates and staff, avoiding burdensome administrative requirements that might
be susceptible to manipulation, and preserving the disciplinary process as a means
of rehabilitation.”
When a prison disciplinary action infringes on an interest protected by the due
process clause (a protected liberty interest), the inmate must receive (1) advance
written notice of the disciplinary charge, (2) an opportunity to call witnesses and
present documentary evidence in his or her defense, and (3) a written statement by
the fact finder of the evidence relied upon and the reasons for the action taken (Wolff
82
The Designations and Sentence Computation Center (DSCC) is located in Grand Prairie, Texas and is responsible
for the review of recommendations for designation to a specific facility, re-designation or transfer to a different
facility, and also review the placement recommendations to SMU and the ADX General Population.

v. McDonnell, 418 U.S. 539 1974). If there is “some evidence” to support the decision,
then due process is satisfied (Superintendent v. Hill). “Some evidence” means “any
evidence on the record that could support the conclusion reached” (Id.)
In regard to timing, there must be at least 24 hours between the time the inmate
receives the written charges and the hearing. This has been determined to be
sufficient time to allow the inmate to prepare a defense.
The inmate has the right to call witnesses and present documents supporting his
defense, as long as doing so is not hazardous to institutional safety. Prison
regulations control when the hearing is held; if the hearing is not held within the
time specified by the regulation, this failure is not in and of itself a violation of due
process. In the court’s view, the relevant inquiry will not concern whether the prison
complied with its own regulations, but rather whether the inmate ultimately received
all he or she was entitled to under the Wolff standard outlined above.
If an inmate needs legal assistance, the prison should provide access to a “reasonably
adequate law library” for conducting legal research. The Supreme Court has declined
to find that inmates have a right to counsel for disciplinary matters (Wolff, 18 U.S. at
570). The Court did indicate, however, that if an inmate is illiterate or suffers from a
mental condition, legal assistance may be appropriate. However, that assistance need
not be provided by an attorney.
In sum, as long as the prison ensures that the requirements of Wolff are met, due
process is viewed as being satisfied.
Finally, even though an inmate may be able to file a grievance regarding a condition
of confinement at any time for any reason, the mere fact that such a grievance
process exists does not create a protected liberty interest. Most such “blanket”
grievance processes are designed to comply with the Wolff standards and can serve
as an additional fail-safe measure to protect against inadvertent violations of due
process rights. However, if an inmate challenged the underlying decision in court, the
court would first determine whether or not the particular issue the inmate raised
implicated a protected liberty interest. If not, then the blanket grievance process
would be unnecessary.
Decisions to place inmates in some type of special housing unit, such as segregation
or restrictive housing, that may result in a deprivation of the inmate’s liberty interest
are protected by the Due Process Clause. Whether such placement amounts to a
violation of a protected liberty interest requires an answer to the following question:

Does the segregation assignment impose an “atypical and significant” hardship on
the inmate “in relation to the ordinary incidents of prison life?”83
It is important to remember that the act of confining an inmate to special housing is
not, in and of itself, a violation of due process and therefore, the Wolff standard does
not apply just because a transfer occurs. The transfer has to amount to an “atypical
and significant hardship” in order to trigger due process protections. While an
inmate may feel burdened by being placed in administrative segregation, if it is for a
short time, this is not generally considered an “atypical and significant hardship.”
On the other hand, courts have also held that even a minor hardship can amount to a
deprivation of liberty if it is imposed for an extended period of time. As federal
courts have applied this rule, it has become apparent that very few instances of
administrative segregation, except those that clearly affect the total duration of the
inmate’s incarceration, will result in infringement of due process. A recent relevant
ruling84 held that if segregation leads ultimately to a longer prison sentence
compared to those served by inmates in the general population, then the segregation
may be “atypical and significant” and due process should be afforded.

Inmate discipline
Inmate discipline within the Bureau is governed by Program Statement 5270.09,
Inmate Discipline Program, as amended July 8, 2011. The provisions of the policy
apply to all people in the custody of the Bureau including in Bureau contract
facilities.
A May 2013, the Government Accountability Office report Improvements Needed in
Bureau of Prisons’ Monitoring and Evaluation of Impact of Segregated Housing
included a simplified summary of the process outlined in Program Statement
5270.09, from staff observation of a potential prohibited behavior through the
referral process for review of the allegation to conclusion of the hearing process.
This summary is included in figure 4.

Sandin v. Conner, 515 U.S. 472, 483 (1995).

Figure 4.

Inmate disciplinary process

The key elements of the inmate discipline program as required by PS 5270.09 are
summarized in the pages that follow.
The disciplinary process is initiated when a staff member believes the inmate has
committed, or observes the inmate committing, a prohibited act. The resulting
incident report must contain all known facts (except when there is information that
must remain confidential). All people present during the incident are to be listed, as
is any physical evidence. The staff member will issue the inmate the incident report
within 24 hours of becoming aware of the incident. The incident report is then
forwarded to the supervising lieutenant.
The 24-hour notice period begins at the point at which staff becomes aware of the
incident or of probable cause to believe that the inmate was involved in an incident.
For example, it would begin on the date that a probable-cause drug test came back
positive, not necessarily on the date that the staff observed inmate behavior that
gave rise to the test.
The lieutenant usually serves as the investigating officer and is responsible for
completing the investigation and providing the inmate with formal notice of the
charges as contained on the second page of the incident report. The investigating
officer may informally resolve the incident report (except for those in the “greatest”
and “high” severity levels). If not resolved informally or referred for external

investigation due to the nature of the incident, the incident report is forwarded to
the unit disciplinary committee (UDC).
The UDC reviews both the incident report and the investigation results, and can
render one of the following decisions:
Determine the inmate committed the prohibited act or identify similar
prohibited acts.
Determine that the inmate did not commit the specified acts.
Refer the incident report to the DHO for further review based on the nature of
the charges (prohibited acts of “greatest” or “high” severity level are
automatically referred to the DHO).
The following are examples of prohibited acts within each severity level:
Greatest—killing, escaping from escort, manufacturing a weapon, rioting,
assaulting any person, including sexual assault, taking a hostage, introducing or
making any narcotics, drugs, alcohol, or related paraphernalia
High—escaping from a work detail or other non-secure confinement, fighting,
threatening with bodily harm, bribery, extortion, making sexual proposals or
threats, engaging in sex acts, stealing, destroying government property
Moderate—indecent exposure, misusing authorized medication, refusing to
work or to accept a program assignment, violating conditions of a community
program, insolence, gambling, possessing money, smoking where prohibited,
communicating a gang affiliation
Low—malingering, using obscene or abusive language, unauthorized physical
contact, interfering with a staff member, conduct which disrupts or interferes
with the security or orderly running of the institution or Bureau
The UDC is ordinarily composed of two or more unit staff members. It is required to
review the incident report and hold a hearing within five working days. It can dispose
of most charges with findings and sanctions, but must refer 100- and 200-level
offenses to the DHO. To render findings and order sanctions for low-level offenses
(300 and 400 levels), two members of the UDC must participate in the hearing. If the
UDC renders findings and sanctions (on low-level offenses), the inmate has no due
process rights, as the low-level sanctions available for such offenses do not implicate
any liberty interest. Even so, inmates may appeal the decision of the UDC and, using
the appropriate section on the incident report, the UDC provides the inmate with
notice of appellate rights.

For minor (300- and 400-level) infractions, the program statement also provides an
informal resolution process. In this process, after the inmate receives a copy of the
incident report, and the unit team and inmate may agree to an informal resolution.
Informal resolution of an incident report requires the consent of both the staff and
the inmate but occurs at the sole discretion of staff. The incident report is placed in
“pending informal resolution” status. If the inmate completes tasks that may be
required of him or her, the status is changed to “informally resolved.” If informal
resolution fails for any reason, the disciplinary process starts again and the incident
report is forwarded to the UDC.
Incidents that involve possible criminal behavior—those involving narcotics,
weapons, assaults, or cell phones—must first be referred to the Federal Bureau of
Investigation or other outside authorities for further investigation and possible
prosecution prior to initiating the disciplinary process. Similarly, cases that require
laboratory testing are deferred until the lab results are obtained. Accordingly, the
incident report may not be fully completed or provided to the inmate until a later
date. Incident reports may also be updated, supplemented, or rewritten and
resubmitted after additional facts become known.
For a referral to the DHO, only one member of the UDC is required. The UDC is
required to provide the inmate with its disposition/notice of charges at least 24
hours prior to the DHO hearing.
The policy lists the prohibited acts that may result in discipline.85 The acts are
divided into four categories based on severity: greatest, high, moderate, and low. The
policy also includes a table that outlines the prohibited acts and indicates the
available sanctions for each act. Sanctions are imposed either by the DHO or the UDC
depending on the severity level.
Inmates have the right to select a staff representative to assist in preparation for the
hearing and to serve as a representative during the hearing. This option was utilized
frequently in the cases we reviewed, and in many instances continuances were
granted to ensure the availability of the requested staff representative.
If evidence indicates that an inmate does not fully understand the nature of the
process or the allegations or cannot assist in his or her own defense, the disciplinary
hearing may be postponed until the inmate is competent to participate. The UDC or

Table 1. Prohibited Acts and Available Sanctions, PS 5270.09

DHO will make this decision based on available evidence, including evidence
presented by the mental health staff. 86
Bureau policy allows for the possibility that an inmate who has been determined to
be mentally ill may not be held responsible for his/her conduct. In that case, the
inmate is cleared but the incident report is retained in the inmate’s file for
informational purposes only.
If the charges in the incident report are sustained, the hearing officer may impose
sanctions that vary depending on the severity level and related circumstances of the
incident as outlined below.
Greatest severity level: forfeit or withholding of earned statutory good time
and nonvested good conduct time up to 100 percent; disciplinary segregation
up to 12 months; monetary restitution or a monetary fine; impoundment of
property; removal from a program, group activity, or job; loss of privileges
including visits, telephone, commissary, movies, and recreation
High severity level: forfeit or withholding of earned statutory good time and
nonvested good conduct time up to 50 percent or up to 60 days, whichever is
less; disciplinary segregation up to six months; monetary restitution or a
monetary fine; impoundment of property; removal from a program, group
activity, or job; extra duty; loss of privileges including visits, telephone,
commissary, movies, and recreation
Moderate severity level: forfeit or withholding of earned statutory good time
and nonvested good conduct time up to 25 percent or up to 30 days,
whichever is less; disciplinary segregation up to three months; monetary
restitution; monetary fine; impoundment of property; removal from program,
group activity, or job; extra duty; loss of privileges including visits,
telephone, commissary, movies, and recreation
Low severity level: forfeit of up to 12.5 percent of good conduct time credit
available for year (on second occurrence of same offense in six months), or
forfeit of up to 25 percent of good conduct time credit for year (on third
occurrence of same offense in six months); monetary restitution or a
monetary fine; impoundment of property; removal from a program, group
activity, or job; extra duty; loss of privileges including visits, telephone,
commissary, movies, and recreation

PS5270.09, Chapter 3, 541.6(b)

The maximum sanctions for disciplinary segregation for these levels are higher than
the authorized levels applied in many state systems. Georgia limits disciplinary
segregation to 30 days. In certain circumstances in Ohio, the 15-day segregation can
be increased to 30 days.87 Other systems limit segregation to 180 days; systems
including those in Illinois, New York, and Kentucky limit segregation for a single
offense to 12 months. However, most states allow consecutive segregation sentences,
which can lead to very long periods of disciplinary confinement.

Disciplinary process observations
In the course of this review, UDC and DHO hearings were reviewed and observed at
each of the locations reviewed. DHO docket summaries were reviewed at each
institution that was reviewed. The docket summaries included information on the
violation code, location, and result of the hearing including the sanction imposed.
The review also included interviewing staff involved in the UDC hearing process and
the DHO assigned to each facility.88 In addition, hearing documents for the most
recent 40–50 DHO cases were reviewed. This included all related materials that were
used in the hearing process, including the original incident report, results of the
investigation of the incident, documentation of the UDC review and referral to the
DHO, all materials associated with the DHO hearing, and the imposed sanctions. In
total over 450 case files were reviewed in the course of the site visits.
The DHO staff members assigned to the facilities do not directly report to the local
institution’s warden or a designated staff. DHO staff members have autonomy to
impose sanctions and do not consult with the operations staff when determining
those sanctions. The DHO staff report up through the organizational structure to the
Bureau’s Correctional Programs Division. Regional DHO staff and central office staff
provide direction, guidance, oversight, and training to the field-based DHO staff, but
there is no direct reporting relationship to the administration of the institution
where they are assigned. This is done in order to ensure the independence of the
hearing process.
At all facilities visited (with the exception of USP Terre Haute, where the assigned
DHO was located in the Kansas City Regional Office and hearings were conducted by

Inmate Disciplinary Process, 56-DSC-01, effective December 10, 2009, Ohio Department of
Rehabilitation and Correction.

All DHO staff were interviewed at the facility assigned with the exception of USP Terre Haute
where the DHO was based in the regional office. In this case the interview was completed
telephonically.

video), the disciplinary hearings were held out of cell in a private location in the
presence of the DHO, consistent with policy and best practices.
FINDING: The DHOs and others involved in the disciplinary process were well
versed in their duties. All appropriate notices and procedures were followed, and
inmates responded respectfully to the process and the decision.
Generally, the DHO at each facility had arranged for a staff representative to attend
or postponed the hearing until the staff representative could be present. In two
cases, although the paperwork clearly indicated that the inmates had been given
appropriate notice and service of charges, when the inmates questioned appropriate
notice and/or service, the DHO re-served the inmates and delayed the hearings.
FINDING: Bureau disciplinary processes and procedures provide substantial and
redundant assurances for due process compliance.
At some facilities, the file review of DHO cases found that a significant percent of
charges filed through incident reports were expunged as a result of the DHO hearing
process (table 21). This is the equivalent of a dismissal or acquittal for the inmate.
Table 21.

Hearing dismissal rates
Facility
USP Lewisburg
USP Allenwood
USP Florence
USP Hazelton
USP Victorville
FCI Butner II

Dismissal rate
12%
12%
22%
6%
7%
15%

Source: BOP.

These dismissal rates are higher than rates observed by project team members at
comparable facilities at the state level, which is an indicator that the process is valid
and ensures due process in the review of charges and allegations. It is also
significant given that the process has multiple review points where the charges can
be dismissed prior to referral to the DHO.
Sanctions issued by the DHO were found to be consistently lower than the allowed
level for each of the severity levels. For example, offenses in the “greatest” severity
level carry a maximum time in disciplinary segregation of 12 months. Of all the case
files reviewed, not one included a sanction of 12 months segregation. For 100-level
offenses such as possession of a weapon the typical sanction included 30–45 days
disciplinary segregation and in many cases was 15–20 days segregation. At USP
Victorville, the typical sanction for an assault or a weapons offense was 30–60 days
disciplinary segregation. At USP Tucson, a typical sanction for possession of a

weapon was 45 days, while an assault resulted in 60–90 days segregation. At USP
Coleman, a weapons offense resulted in 30 days of disciplinary segregation.
Bureau DHOs use the restriction of privileges such as visiting, commissary, and
telephone extensively as a sanction for offenses within all severity levels. Almost
every sanction issued by the DHO at each of the facilities reviewed included a
restriction of one or all of the above privileges. At USP Coleman, weapons offenses
typically resulted in a sanction of 41 days loss of good conduct time; 15 days
placement in disciplinary segregation; or 180 days restriction on visits, commissary,
and telephones. Similar sanctions, with some variance in the amount of segregation
time, were found in virtually all the institutions reviewed. In the cases reviewed, it
was normal for the DHO to impose 180 days loss of commissary, telephone, and
visits. The extensive use of restriction of privileges resulted in the accumulation of
loss of privileges over an extended period. It was not unusual to find inmates who
had lost visit privileges for more than one or two years. During one interview, an
inmate reported that he had lost visit and phone access for seven years. This was
confirmed through review of disciplinary records provided by the DHO. File reviews
confirmed that similarly lengthy restrictions of privileges were common in the
system. The use of these sanctions is an outgrowth of the attempt to find alternative
sanctions to placement in disciplinary segregation.
Sanctions issued by the DHO are effective the date of the hearing and are not made
retroactive to the date of the incident report or the date of referral by the UDC to the
DHO. There is no time limit in policy that governs how quickly the DHO must hold a
hearing after the investigation/UDC process is complete. Scheduling of hearings is at
the discretion of the DHO.
The lack of time limits for completion of disciplinary hearings results in substantial
variation among facilities in the amount of time served in restrictive housing for
similar offenses, and can result in disproportionately long sanctions.
During our review we found institutions that had established informal internal
requirements that the hearing be completed within 14 days of receipt by the DHO.
Another facility had established a guideline of 20 days. In the course of this review,
instances of hearings held more than 30 days after the incident date were not
unusual. Longer delays occurred when cases were continued awaiting the results of
drug tests or other information or referred for further investigation. For example, a
sanction of 30 days in restrictive housing can extend to 90 days or longer since time
served is not credited and the time frame of the sanction is not made retroactive to
the date of original confinement in restrictive housing.
By comparison, most state systems have specific time requirements for conducting a
hearing or issuing a continuance based on need for additional information (such as
investigation results or availability of witnesses). Ohio policy requires that inmates

charged with a rule violation must be scheduled for a hearing as soon as practicable
but no later than seven days, excluding weekends and holidays, after the alleged
violation is reported—unless the hearing is prevented by exceptional circumstances,
unavoidable delays, or reasonable postponements, which must be documented.89
FINDING: Sanctions issued by the DHO are effective the date of the hearing and
are not made retroactive to the date of the incident report or the date of referral
by the UDC to the DHO.
The result was that in many cases a sanction of 30 days segregation in actuality
became a restrictive housing sanction of 90 day or longer since the time served in
segregation was not credited to the sanction or made retroactive to the date of
original confinement in a restrictive housing setting.
RECOMMENDATION 4.1: Establish reasonable time frames in which the hearing
must be scheduled while permitting reasonable continuances while awaiting
investigation reports, drug test results, and other essential information.
RECOMMENDATION 4.2: Establish by policy that a sanction of segregation time
should be issued retroactive to the date of original admission to restrictive
housing, providing credit for time served.
File reviews indicated that sanctions for similar offenses vary from institution to
institution. The policy only specifies the maximum sanction for each severity level,
thus providing the DHO only general guidance for determining appropriate
sanctions. We were also informed that the Bureau has no systemic way to measure
disparities in sanctions across this large system. Over the years, initial DHO training
has reportedly been cut from two weeks to one week to the current three days.
In our visits to the selected facilities, we noted wide disparities in the type and
severity of sanctions imposed for similar offenses. As noted, weapons offenses
sanctions varied from 15 days of restrictive housing plus loss of privileges to 60 or
90 days of restrictive housing plus loss of privileges. The independence of the DHO
staff in determining sanctions also results in some disparity in sanctions. The chief
disciplinary officer of the Bureau90 stated that the central office does not monitor
sanctions for consistency. They do, however, have the ability to identify
inconsistencies from hearing officer to hearing officer and can address those when
necessary and appropriate.
Inmate Disciplinary Process, 56-DSC-01, effective December 10, 2009, Ohio Department of
Rehabilitation and Correction.

Interview conducted on 11/14/2013.

RECOMMENDATION 4.3: Establish a system for monitoring patterns and trends in
the use of disciplinary sanctions among Bureau facilities.
It was also noted through our review of individual cases that different DHOs have
different philosophies about the use of specific types of sanctions. DHOs at some
locations used monetary fines with great success in lieu of restriction of privileges or
extended time in restrictive housing. However, this sanction was not universally
used, and at some facilities appears not to have been used at all, based on the cases
reviewed and the interviews with the DHO assigned to the unit.
PS 5270.09 authorizes the DHO to suspend any sanction for a period not to exceed
six months. In the event a sanction is suspended by the DHO or the UDC, the effect
of the sanction is waived unless the inmate receives a new incident report for a
prohibited act during the period of the suspension. In the event of a new incident the
DHO or UDC will act on the new incident report and retroactively determine and
impose the sanction for the suspended case. Review of case files and interviews with
the DHO indicated that this option is used extensively by some DHOs as a means of
providing motivation to avoid committing future prohibited acts.
Overall, our findings show that DHO staff are adequately trained on the disciplinary
processes and procedures, and that compliance with these requirements is
consistent from facility to facility. The organizational independence of the DHO from
the local administration is rare outside of the Bureau as only a small number of state
systems use this option. This independence was apparent during interviews with
staff both in headquarters and at the local institutional level. This, combined with
the relatively high percent of dismissals and expunged cases as compared to what is
observed in state systems, are indicators of the independent decision making of the
DHO process. However, the review did note issues with consistency in the level of
sanctions for similar offenses between facilities, hearing delays that resulted in
prolonged stays in segregation, and loss of access to basic privileges (visits,
telephone, and commissary) for extended periods of time.

Special housing units
SHUs are housing units in Bureau institutions where inmates are securely separated
from the general inmate population and may be housed either alone or with other
inmates. SHUs help ensure the safety, security, and orderly operation of correctional

facilities, and protect the public, by providing alternative housing assignments for
inmates removed from the general population.91
According to Bureau policy, placement in an SHU is a result of assignment to either
administrative detention or disciplinary segregation status. Administrative detention
is an administrative and nonpunitive status and can occur for a variety of reasons.
Disciplinary segregation is a punitive status that can be imposed only by a DHO.
An inmate may, under Bureau policy, be placed in administrative detention for the
following reasons:
Pending classification or reclassification—for example, in the case of a new
inmate or one whose classification is under review
Holdover status during transfer to a designated institution or other destination
Pending transfer to another institution or location
Removal from general population, where it has been determined that placement
poses a threat to life, property, self, staff, other inmates, the public, or the
security or orderly running of the institution
Being under investigation or awaiting a hearing on charges of violating a Bureau
regulation or criminal law
For protection, whether requested by the inmate or determined by staff to be
necessary
After completion of disciplinary detention when return to the general
population would threaten the safety, security, and orderly operation of the
facility, or public safety

Placement
Each inmate who is placed in the SHU is notified of the basis of the placement when
the lieutenant or other correctional supervisor prepares and issues an administrative
detention order. A separate administrative detention order is required when an
inmate’s status in administrative detention changes.

PS 5270.10, Special Housing Units, Effective date of August 1, 2011.

Administrative detention status: When placed in administrative detention
status, the inmate receives a copy of the administrative detention order within
24 hours detailing the basis for the placement. When an inmate is placed in
administrative detention status pending classification or while in holdover
status, an administrative detention order is not issued.
Disciplinary segregation status: When an inmate is placed in disciplinary
segregation status as a sanction for violating Bureau regulations, notice is
provided to the inmate by the DHO at the end of the discipline hearing.
Placement in the SHU is reviewed by the segregation review official (SRO) based on
the following requirements as outlined in PS 5270.10.
Three-day reviews: Within three work days of placement in administrative
detention status, not counting the day of admission, weekends, and holidays,
the SRO will review the supporting records. Inmates in disciplinary segregation
status do not receive this review.
Seven-day reviews: Within seven continuous calendar days of placement in
either administrative detention or disciplinary segregation, the SRO will
formally review the placement status at a hearing that the inmate may attend.
The inmate must sign a waiver if he or she does not want to participate in this
face-to-face review. Subsequent reviews of the case records will be performed by
the SRO every seven continuous calendar days thereafter.
Thirty-day reviews: After every 30 calendar days
either administrative detention or disciplinary
formally review the placement status at a hearing
Again, inmates must sign a waiver if they do not
face-to-face reviews.

of continuous placement in
segregation, the SRO will
that the inmate may attend.
want to participate in these

Placement of protection cases in administrative
detention
Inmates may be placed in administrative detention as a protection case in the
following circumstances:
The inmate was the victim of an inmate assault, or was being threatened by
other inmates.
The inmate’s safety is threatened because of providing, or being perceived as
having provided, information to staff or law enforcement authorities regarding
other inmates or people in the community.

The inmate refuses to enter the general population because of alleged pressures
or threats from unidentified inmates, or for no expressed reason.
Based on available evidence, staff believe the inmate’s safety may be seriously
jeopardized by placement in the general population.
When an inmate is placed in administrative detention for protection, the warden or
designee (ordinarily the captain) must review the placement within two business days
to determine if continued protective custody is necessary. This review consists of:
Staff investigation: Whenever an inmate is placed in the SHU as a protection
case, whether requested by the inmate or staff, an investigation will occur to
verify the reasons for the placement.
Hearing: The inmate will receive a hearing conducted by the SRO according to
the procedural requirements within seven calendar days of the placement.
Periodic review: If the inmate remains in administrative detention status
following the hearing, he/she will be periodically reviewed as an ordinary
administrative detention case.
This review includes documents that led to the inmate being placed in protective
custody status and any other documents pertinent to the inmate’s protection. In
addition, P5324.08 Suicide Prevention Program mandates that protective custody
inmates be screened for suicidal ideation within 72 hours of being placed in SHU.

Protective custody
Despite the very different purposes of the SHU, all inmates including those in
protection status are exposed to the same security and operational restrictions as
well as the same access to programs and privileges.
As shown in table 5, 15 percent of those housed in SHUs system-wide as of June
2014 were there based on protection needs. The percent varies from institution to
institution, with the percentage in the USP SHUs being significantly higher. An
additional portion of inmates in SHUs are unverified protection cases who refuse
assignment outside of the unit. Some of the protection claims are the result of the
inmate’s own behavior, while others are not validated. However, a significant portion
of these offenders have legitimate protection needs.
FINDING: A disproportionate number of inmates are being housed in the SHUs
based on protection claims.

Often the reason for an inmate’s request for protection is gang related. The number
of inmates that have separatee issues (cannot be housed with specific other inmates)
is significant and affects inmate management. The warden at USP Lewisburg advised
the assessment team that of 748 inmates at that facility, 334 had separatee issues.92
This raises the question of the appropriate way to manage inmates with verified
protection needs. Such inmates are presently housed in administrative detention,
which is identified by Bureau policy as a nonpunitive status. However, they are
assigned to the same housing unit as inmates in punitive segregation. Inmate
movement procedures, including application of restraints, are the same; the
frequency of recreation and telephone access is the same, as was the frequency of
visits at all but two facilities visited. With the exception of minor differences in
personal property allowed and in-cell programming opportunities, the day-to-day
conditions of confinement were not much different. Considering that one status is
nonpunitive and that some individuals are included strictly as a result of being
verified as requiring protection, serious consideration should be given to
reevaluating the day-to-day conditions of confinement for individuals who have been
verified as needing protection.
FINDING: The application of the same security and operational restrictions to the
protective custody population as to others in administrative segregation is
contrary to nationally accepted practices.
This is a complex issue within the Bureau due to the extensive presence of security
threat group members, even in the SHU. Many of those have verified need for
protection as a direct result of their prior involvement with a security threat group,
who are also victims or potential victims that need protection while assigned to the
Bureau. These protection needs should be provided, but in a more normalized setting
than what is presently provided in the SHU.
In numerous states, the conditions of confinement for protection cases have been
altered to parallel that provided to other general population inmates. In these
instances, inmates are housed in SHUs or similar units while the claim for protection
is investigated. Once the need for protection is verified, they are moved to a separate
unit that provides conditions of confinement that are similar if not identical to those
provided to general population inmates. These units operate separately from other
general population units and afford inmates the ability to function in a normalized
prison environment while ensuring they are protected. Kentucky has done this
successfully at the Eddyville facility, and Ohio operates units that replicate general

Interview briefing with Warden, January 21, 2014

population conditions for protective custody inmates. The Bureau is moving in the
same direction with the establishment of the reintegration housing unit (RHU) in
Oakdale, Louisiana.
The RHU was activated in October 2013 with an objective to target male inmates who
consistently refuse to enter general population.93 The facility was opened with an
initial capacity of 160 beds but was increased to 208 in February 2014; it has a
potential future capacity of 320 beds. The actual population of the facility in
September 2014 was 82. The criteria for placement include documentation that the
inmate is classified as a protective custody case, an assessment from psychological
services that the inmate is willing to participate in programming offered at the RHU,
and the following additional criteria:
The inmate has refused to enter general population and this fear is
unsubstantiated (cannot be verified by staff).
The inmate will have been housed in an SHU for more than 12 months.
Medium and high security inmates are considered appropriate.
Inmates whose SHU placement is based on a gang-specific security threat group
assignment are ordinarily excluded.
Those currently in disciplinary segregation are excluded.
Those whose sex offender classification is the basis for placement in SHU will
ordinarily not be assigned to the RHU.
Inmates must be classified in medical and mental health care level 1 or 2.
Inmates should normally have at least 6 months remaining to serve.
Sex offenders who require protection and who meet the established criteria have
been designated for the Sex Offender Management Program at USP Tucson. This is a
normalized general population program and allows this group of protection inmates
to fully participate in programs. The Bureau reported that it operates nine Sex
Offender Management Program facilities, all of which offer an environment which
increases the likelihood a sex offender can remain in general population.

Memorandum of October 18, 2013, FCC Oakdale – RHU Activation Procedures, authored by
the Acting Assistant Director, Correctional Programs Division.

These are the only two general population options for verified protective custody
inmates that the consultant team identified within the Bureau.
RECOMMENDATION 4.4: Expand housing alternatives for inmates in verified
protective custody status that have levels of programs and privileges that are
equivalent to those for general population inmates.

Weekly review meetings
At all the Bureau facilities visited, SHU weekly review meetings were observed. These
meetings are convened and directed by the warden of each facility and are attended
by representatives of the warden’s management team as well as representatives of
each key discipline within the facility. This includes associate wardens, unit
managers, directors of psychological services and medical services, investigative
staff, case management representatives, and security and chaplaincy staff.
The purpose of the meeting is to review the status of each inmate held in the SHU
and determine if continued placement in the SHU is appropriate or whether an
alternative placement can be identified. The structure and approach to these
meetings was similar at each location we observed.
The reviews are guided by a formatted document that contains all the key
information on an inmate including name, reason for placement in the SHU, mental
health status, any medical issues, unit team comments, and status of any program
referrals or redesignations. Also included in the document is a picture of the inmate
so all participants in the meeting can identify the inmate being discussed.
The participants in the review meeting systematically review the status of each
inmate. A designated representative, usually a unit manager, summarizes the
offender’s status, including any changes or pending actions. These can include
pending transfer requests, status of any investigation, mental health or medical
actions, program participation or completion, and general adjustment to the facility
and/or unit. A representative of each discipline presents updated information on the
inmate where appropriate. After discussion, the group with the concurrence of the
warden may take action on the case, such as releasing the inmate from SHU (we
observed that this action occurs frequently), referring the inmate for redesignation,
or accelerating the investigation into the inmate’s case.
The formal discussion is normally followed by the entire group touring the SHU and
addressing any informal or formal requests presented by inmates.

FINDING: The conduct of weekly reviews of SHU placements in a formalized
setting with the facility’s entire management team is an exemplary practice that
ensures ongoing review of the status of inmates and their placement options.
The review team for this project has not observed anything similar to this practice in
terms of frequency of reviews, breadth of participants, and level of discussion on
each and every case as was observed in the facilities reviewed. This process ensures
that inmates housed in the SHU are reviewed at the highest levels on a weekly basis
and that their placement options are evaluated regularly.

Segregation reviews
Policy requires frequent contact and review of those inmates placed in administrative
segregation. Through the SRO reviews and the previously described SHU Weekly
Review process, there is ongoing review of the status of each inmate in relation to
his/her continued placement in the SHU. There is however, inconsistency in how the
SRO reviews are conducted and the scope of these reviews. In some of the facilities
reviewed, the SRO reviews are conducted cell-side and consists simply of an inquiry
as to how things are going and if any problems exist that need to be addressed. This
provides no privacy and little opportunity for the inmate to have a dialogue with
staff on his situation and what may have precipitated the problems that resulted in
his placement in the SHU. This is especially important given the prevalence of
security threat group members housed in the same units as those with protection
claims.
RECOMMENDATION 4.5: Establish a policy standard requiring private, face-to-face
interviews for the segregation review.
The Bureau reports that Program Statement 5270.10 requires face to face meetings at
the 7 and 30 day review. Three day reviews are paper reviews, seven day and 30 day
reviews give the inmate an opportunity to attend the hearing. Those inmates who
refuse to attend the hearing are required to sign a waiver stating they refuse. Some
Bureau facilities provide the opportunity for the inmate to meet privately with the
SRO outside of the cell area and in a private location where there can be an open
discussion of the issues and concerns that the inmate may face. This should become
the standard for the Bureau in conducting the SRO hearings. Other facilities do not
encourage or facilitate the face to face meeting in a private location.
FINDING: The SRO reviews at some of the facilities reviewed appeared
perfunctory and lacked substance in contact and purpose.
In these instances the SRO review becomes a situation of quickly doing “how are you
doing and do you have any problems” rather than a review of the reasons for

placement, program needs, placement issues if any, and verifying that the inmate has
access to medical, mental health issues, etc. It appeared that there was little formal
structure to the purpose and content of the reviews.
The GAO segregation review report issued in May 2013 found that “…the facilities
did not consistently document conditions of confinement and procedural protections
as required under Bureau policy guidelines.” The GAO reported deficiencies such as
missing documentation, monitoring rounds not being consistently conducted, or
inmate review policies not fully implemented.
FINDING: The review of randomly selected inmate records found some omissions
in the maintenance and content of inmate records documenting the placement
rationale in the SHU.94
At one facility reviewed, ten percent (10 percent) of the files reviewed had problems
with filing of these documents in the inmate record file including the absence of
required supporting documents and the existence of a backlog in filing these
documents in the inmate record. This was only observed at one facility during the
course of the review.
RECOMMENDATION 4.6: Develop and deploy an electronic inmate record system
to document SHU placement decisions.
The missing documents were later found as they were available elsewhere, but the
inability to maintain inmate records in a timely and accurate fashion illustrates the
need for the Bureau to substantially improve their recordkeeping, including the use
of electronic means.
FINDING: The requirements that are contained in the policy and procedures that
govern the placement and review of inmates housed in the SHU are consistent
with national standards and afford inmates in these units with due process in
relation to their placement in the units.
The execution of these requirements could be improved through the use of private,
face-to-face, reviews with the inmates of his/her status in the SHU. This is done in
some facilities, but not consistently applied throughout the system.
94
Documents examined and filed include Incident Reports, BP-AO298, Inmate Rights at
Disciplinary Hearing, BP-293, Administrative Detention Orders, BP-308. The Bureau reported
that in August 2014, the SHU application was upgraded to include a warning screen feature
notifying staff when omissions from the records were noted. This upgrade resulted in no
deficiencies for this step in any Correctional Services Program Reviews conducted during the
last quarter

Special management units
According to PS 5217.01, Special Management Units, assignment to a special
management unit (SMU) is considered to be a programmatic assignment. PS 5217.01
states that SMU designation is nonpunitive, and may be appropriate for any inmate
meeting the referral criteria outlined in the program statement.
The conditions of confinement for SMU inmates are more restrictive than for general
population inmates. Inmates are expected to complete the four-level SMU program in
18 to 24 months, at which time they may be re-designated (transferred) to an
appropriate facility. At the time of the initiation of this review SMU programs were
functioning at USP Lewisburg, USP Allenwood, and USP Florence. During the course of
the review, the Bureau began the phasing out of the SMU at USP Florence through the
gradual transfer of inmates in the unit to USP Lewisburg.

Referral and assignment
Per the program statement, designation to a SMU may be considered for any
“sentenced inmate whose interaction requires greater management to ensure the
safety, security, or orderly operation of Bureau facilities, or protection of the public.”
Placement to a SMU requires that the inmate meet any of the following:95
Participated in disruptive geographical group/gang-related activity.
Had a leadership role in disruptive geographical group/gang-related activity.
Has a history of serious and/or disruptive disciplinary infractions.
Committed any 100-level prohibited act, according to 28 CFR part 541, after
being classified as a member of a Disruptive Group pursuant to 28 CFR part
524.
Participated in, organized, or facilitated any group misconduct that adversely
affected the orderly operation of a correctional facility.
Otherwise participated in or was associated with activity such that greater
management of the inmate’s interaction with other persons is necessary to

PS 5217.10, Section 2, effective 11/19/2008.

ensure the safety, security, or orderly operation of Bureau facilities, or
protection of the public.
Referral. If an inmate meets any of the referral criteria the unit team may present a
re-designation referral to the warden. The referral packet consists of a completed
Request for Transfer/Application of Management Variable (EMS-A409), copies of
pertinent Special Investigative Supervisor reports and incident reports, and a cover
memorandum to the warden summarizing the rationale for referral for SMU
designation. If the warden approves the referral, it is submitted to the regional
director.
Hearing on the redesignation request. If the regional director determines that
sufficient evidence exists to convene a hearing, the regional director appoints a
hearing administrator to conduct a hearing into whether the inmate meets the
criteria for SMU designation. The hearing administrator will have been trained and
certified as a discipline hearing officer (DHO), will be an impartial decision-maker,
and will not have been personally involved as a witness or victim in any relevant
disciplinary action involving that inmate.
Interviews with Bureau executive staff indicated that the Bureau decided to use
DHO’s in the SMU review process because they are trained to provide due process
protections in administering the disciplinary systems and they are independent of
the operational chain of command, reporting directly to the regional hearing
administrator and to the Office of General Counsel.
The warden will be notified of the regional director’s decision to conduct a hearing
before the inmate is provided pre-hearing notice. The inmate’s security needs will be
assessed and staff made aware of any additional security precautions.
The appointed hearing administrator completes a notice of the hearing referral and
sends it to the inmate’s current institution. Unit team staff at the institution provides
the inmate with a copy of the notice at least 24 hours before the hearing, and
document delivery to the inmate. If the inmate is illiterate, the delivering staff
member will read the notice verbatim. If the inmate does not speak English, the unit
team staff will make arrangements to provide translation.
The notice advises the inmate of the date and time of the hearing and advises the
inmate of the opportunity to appear at the hearing. The notice also provides a
sufficiently detailed explanation of the reasons for the referral. The notice also
informs the inmate that a nonprobationary staff member will be available to help the
inmate compile documentary evidence and written witness statements to present at
the hearing. The assisting staff member’s responsibility in this role is limited to
assisting the inmate in obtaining copies of documents needed, for example, from his

central file or other reasonably available source(s), or a written statement(s) from
other reasonably available inmates or staff.
The inmate has the opportunity to appear at the hearing, make an oral statement,
and present documentary evidence and written witness statements, except where
contrary to the safety, security, or orderly operation of Bureau facilities, or
protection of the public. The inmate may not call witnesses at the hearing.
Post-hearing findings and decision. The hearing administrator is required to consider
whether the inmate meets the criteria as specified in PS 5217.10 for placement into
the SMU program. Upon completion of the hearing the hearing administrator will
prepare the “Hearing Administrator’s Report on Referral for Designation to a Special
Management Unit” and will submit it to the regional director. The report will provide
a detailed explanation of the reasons for the hearing administrator’s findings.
Upon receipt of the hearing findings the regional director determines whether the
SMU referral is necessary to ensure the safety, security, or orderly operation of the
Bureau or protection of the public. The regional director includes a recommendation
on the placement request and then forwards it to the DSCC in Grand Prairie, Texas,
for final review.
The DSCC reviews all documents related to the case including the hearing
administrator’s report and, after consulting with the assistant director, Correctional
Programs Division, Central Office, will determine whether a SMU referral is approved.
If the SMU referral is approved, the DSCC selects the SMU that best meets the
inmate’s greater management needs. The DSCC will then forward the decision to the
receiving regional director and warden, with copies to the referring regional director
and warden. This review also includes a mandatory review by the Bureau’s
Psychology Services Branch to determine if admission to SMU is appropriate.
Post-decision notice and appeal. The inmate’s copy of the completed report will be
sent to the referring warden, who is to ensure delivery to the inmate. The report
advises the inmate of the opportunity to appeal the decision through the
Administrative Remedy Program, directly to the Office of General Counsel. An
inmate’s appeal of the decision or the hearing administrator’s findings does not
delay designation and transfer to a SMU.
FINDING: The Bureau has established policies and procedures that afford due
process protections to inmates in the referral and assignment to SMU.
Bureau policies and procedures governing referral and assignment to SMU provide
for multiple layers of review and are clearly intended to comply with the principles
of due process. The review of the documents and the process indicate that the

Bureau complies with its own requirements throughout the review and placement
process.
One indicator of the validity of the process of referral and review of requests for
placement in the SMU is the number of individuals who had been referred for SMU
placement since the initiation of the program and of those referred, the number that
have been rejected for placement either by the regional director or by DSCC.
Documentation provided by the DSCC indicated that a total of 5,435 inmates had
been submitted from January 2009 to June 6, 2014. Of the inmates referred, 1,057
(19.5 percent) had been denied placement by the DSCC96. This does not include
denials/rejections that occurred at the regional offices as those records were not
readily available.
A more recent picture of the validity of the referral and review process was obtained
by reviewing the number of referrals by month that have been submitted since
January 2013 through February 2014 and the number that have been rejected by
either the DSCC or the regional director (see table 22).This more recent snapshot of
the review process indicates that 14 percent of the referrals have been denied
placement.

DSCC Initiatives Report of 6-6-14

Table 22.

SMU referrals and outcomes, January 2013 – February 2014

Date
January 2013
February 2013
March 2013
April 2013
May 2013
June 2013
July 2013
August 2013
September 2013
October 2013
November 2013
December 2013
January 2014
February 2014
Total
Rate

Submitted
52
61
53
75
43
68
44
47
12
50
43
59
76
73
756
100%

DSCC
approved
40
52
49
57
37
53
38
42
11
46
40
53
66
63
647
86%

DSCC
denied
11
9
4
18
6
15
6
5
1
3
3
6
10
10
107
14%

Region
denied
6
2
2
3
2
4
3
1
9
5
8
10
8
2
65
9%

DSCC
deferred
1
0
0
0
0
0
0
0
0
1
0
0
0
0
2
<1%

Source: Bureau ORE.

FINDING: The significant level of SMU placement request denials indicates the
review process provides independent assessment beyond the institution level of
the necessity of the SMU placement that reduce the number of admissions by
about 14 percent.
Periodic review. SMU inmates are reviewed by the unit team in conjunction with
regularly scheduled Program Reviews as provided in the policy on Inmate
Classification and Program Review97. The unit team specifically reviews inmates for
progression through the levels of the program. An inmate’s institutional adjustment,
program participation, personal hygiene, and cell sanitation are considered when
reviewing the inmate for progression to further levels.
Redesignation criteria. To be re-designated from SMU status, an inmate must for a
period of 12 to 18 months, abstain from all of the following:
Geographical group/gang-related activity
Serious and/or disruptive disciplinary infractions

P5100.08 Inmate Security Designation and Custody Classification (9112/06)

Group misconduct that adversely affects the orderly operation of a correctional
facility
The inmate must also demonstrate a sustained ability to coexist with other inmates,
staff, and other persons.
Referral procedures. When an inmate has met the re-designation criteria, the unit
team will submit a referral to the warden for designation to the general population. A
review of records indicated that the referral is ordinarily for placement at another
institution.
If an inmate is not recommended by the unit team for re-designation after 24
months, a referral for continued SMU designation must be submitted to the regional
director. If the regional director approves continued SMU designation, the inmate will
receive written notice of the decision and the rationale for it. The inmate may appeal
the decision as provided by the Administrative Remedy Program.
The review of records, interviews with staff and inmates confirmed that the Bureau is
in compliance with SMU referral and assignment policies and procedures both in
form and in substance. However, the following concerns were identified in the review
of this program and its associated processes.
Subjective criteria for admissions. The Bureau placement criteria provides for
discretion and flexibility in assigning inmates to the SMU program. This is similar to
the selection process in similar programs operated in state prison system. Instances
of inconsistent assignments/rejections were found both in the file reviews and in
observing the weekly SHU review meetings. The inconsistencies primarily existed
from region to region. This inconsistency is balanced by the layered review process
and the fact the final placement decision is made centrally.
While case law has indicated that assignment to programs like the SMU is left to the
discretion and judgment of prison officials, inconsistencies in assigning practices
have the potential to risk creating fairness and equity concerns. For example, we
observed SMU inmates being referred for re-designation after a single serious
disciplinary violation, while recommendations for other inmates with repeated
instances of the same violation were rejected – or, in some cases – not initiated by
the local administration. Demonstrable inconsistencies in assignment decisions
create equity issues in the application of the placement of inmates to the SMU
program. The balance between discretion of placement versus creating a consistent
and reliable placement process is a challenge that all systems are presented with in
rendering these types of placement decisions.
Periodic reviews. As noted earlier, SMU inmates are reviewed by the unit team in
conjunction with regularly scheduled Program Reviews as provided in the policy on

Inmate Classification and Program Review. Upon arrival to the SMU, an intake
screening is to be completed within 24 hours of the placement, with follow-up review
to be completed after three days in SMU. Subsequent reviews are then conducted at
seven, fourteen and twenty-eight days. Thereafter, the case manager complete thirty
day reviews using form BP-951, Special Management 30-Day Conditions Review. This
schedule of reviews is completed for the duration of the inmate’s assignment to SMU.
Our observations, file reviews, and interviews with staff and inmates confirmed
compliance with these SMU assignment reviews. However, while the documentation
was available electronically and/or in paper form in the appropriate units, we found
that official inmate records were not updated in a timely or consistent manner.
During the course of observing the reviews we found that in some cases the reviews
were held cell-side. It is most appropriate that all reviews be conducted in a private
setting so that a proper and private review and hearing can be conducted.
FINDING: Scheduled SMU Conditions Reviews were in some cases not conducted
in a private setting, consistent with professional practices, and were not reflected
in the official inmate records in a timely manner.

Progression through program phases
The SMU program consists of four levels, with each level differentiated by conditions
of confinement and anticipated time frames. Completion of all four levels will take at
least 18-24 months with the goal of integrating the inmate successfully into a general
population setting. An assessment of the progression process and the associated
conditions of confinement from one level to the next is addressed in this section of
the report.
To determine the conditions outlined above, we examined several key factors:
Statutory requirements
Bureau program statements
Special Management Unit Inmate Handbook
Bureau performance review reports
Inmate files
Housing unit activity schedules
Manual and
confinement

electronic

systems

designed

document

conditions

On-site observations of operational practices
Staff interviews
Inmate interviews
The project team found that the conditions of confinement for inmates assigned to
the special management unit program were generally more restrictive than the
conditions for inmates assigned to the general population and less restrictive than
those assigned to a special housing unit (SHU). Although the conditions were more
restrictive than in a general population setting, the conditions being provided in SMU
appeared to be consistent with applicable federal regulations. As the inmate
progresses from one level to the next, the program is designed to provide fewer
restrictions on the inmate and more opportunity for programming. The following
general findings for each level within the program are described.
SMU level 1. An inmate in Level 1 may be single or double-celled and is normally
allowed to participate in recreation at the same time as other inmates however the
overall interaction between inmates is minimal. With the exception of recreation
almost all programming is provided while the inmate remains in the cell. The
expected time to complete this level of the program is approximately four months.
An initial programming assessment is completed by staff within the first 28 days of
the inmates’ arrival to the unit and every 90 days thereafter. All of the conditions of
confinement identified in the federal regulations and program statements appeared
to be provided to the inmates.
Progression through Level 1 is based on the inmate’s compliance with behavioral
expectations, completion of treatment assignments and absence of significant
misconduct reports. Inmates meeting the program stipulations are expected to
advance to the next level in approximately four months. At the time of the project
teams site visits, inmates in Level 1 were being housed at USP Florence and USP
Lewisburg. Inmates in Level 1 at USP Florence were housed in separate cells from
inmates in Level 2. However at USP Lewisburg inmates in both levels were housed
together.
The primary differences in conditions of confinement for inmates in this level
compared to Level 2, were inmates in Level 1 are allowed two telephone calls per
month compared to four per month for Level 2 and the topics covered in the in-cell
treatment curriculum for each level were different.
SMU level 2. Interaction between inmates allowed in Level 2 was found to be very
similar to the interaction allowed between inmates in Level 1. With the exception of
recreation, almost all programming is provided while the inmate is in the cell.
Program Statement P5217.01 Special Management Unit states that inmates in this

level have minimal interaction with other inmates and inmates will ordinarily be
restricted to their assigned cells. Level 2 inmates were being housed at USP Florence
and USP Lewisburg. Observations made by team members reflect that operational
practices were generally consistent with the program statement at this level. The
progress of each inmate in Level 2 is formally reviewed by staff at least every 90
days. Progression through Level 2 is based upon the inmate demonstrating the
potential for positive “community” interaction. Inmates meeting the program
expectations of Level 2 are normally double-celled and expected to advance to the
next level in approximately six to eight months.
The primary differences in conditions of confinement between Level 1 and 2 were the
number of telephones calls allowed per month (2 versus 4) and the treatment
curriculum. In addition, at USP Florence, there was an incentive program for inmates
who were on extensive telephone restriction to be allowed a telephone call if their
adjustment and progress in the program was positive. All programming with the
exception of recreation was being provided on an individual basis while the inmate
remained in their cell.
SMU level 3. The Program Statement, P5217.01, Special Management Unit, states in
part the following: “inmates at this level will begin to interact in an open, but
supervised, setting with individuals from various groups, to include open movement
in the unit and frequent group counseling sessions commensurate with the inmate’s
demonstrated ability to effectively coexist with other inmates.” The progress of a
Level 3 inmate is formally reviewed by staff at least every 90 days. Progression
through Level 3 is based upon the inmate’s ability to demonstrate positive
“community” interaction skills. Inmates meeting the program expectations of Level 3
are normally double-celled and expected to complete Level 3 in approximately six to
eight months.
At the time of the project teams site visits, Level 3 inmates were being housed at all
the three SMU program facilities; USP Allenwood, USP Lewisburg and USP Florence.
Inmate interaction and conditions of confinement for inmates in Level 3 at USP
Allenwood and USP Florence were noticeably increased compared to Level 1 and 2.
The increase was primarily the result of the addition of indoor recreation
opportunities being provided and expanded access to telephones, commissary and
visits. Inmates housed at USP Lewisburg in Level 3 appeared to have little increase in
inmate interaction as no indoor recreation was offered. Although inmate interaction
and average out-of-cell time had increased at USP Allenwood and USP Florence when
compared with Level 2, there was no “frequent group counseling” that was being
offered at any of the three facilities.
The lack of frequent group counseling or virtually any group counseling is in
contrast with the language found in the SMU program statement which states “… to
include frequent group counseling sessions.” Group counseling was almost

nonexistent. The only group program being provided at the time of the review was a
“cognitive skills for reentry preparation” group that was being offered to Level 3 and
4 inmates at USP Florence. Facility staff reported that due to construction delays the
group had recently started just prior to the project team’s arrival.
SMU level 4. Program Statement P5217.01 states that “…inmates must be able to
demonstrate their sustained ability to coexist and interact appropriately with other
individuals and groups in the unit.” Inmates in Level 4 are reviewed by staff at least
every 30 days. Progression through Level 4 is based upon the inmate’s ability to
function in a general population setting with inmates of various group affiliations.
Inmates meeting the program expectations of Level 4 are double-celled and expected
to be integrated into the general population after being in the level between two and
four months.
There were very few differences noted in the conditions of confinement between
Level 3 and 4 at USP Florence and USP Lewisburg. The primary difference was the
increased treatment focus on preparation and transitioning out of the SMU program.
At USP Allenwood, there were a few additional differences, including: Level 4 inmates
were allowed to purchase extra clothing (sweat shirt/pants), make up to 300 minutes
of social telephone calls per month (Level 3 150 minutes) and receive additional
visiting hours per month (four). At USP Lewisburg and USP Florence there were no
noted significant differences between the two levels with the exception of the
treatment curriculum.

Observations on progression
At the time of the review, the USP Lewisburg SMU program was considered a Level 1
and 2 facility; however, a number of Level 3 inmates were being housed at USP
Lewisburg in the SMU program. The inmate handbook dated May 2013 states that
“…since bed space is sometimes limited some inmates will complete Level 3 at USP
Lewisburg.” On the second day of the project team’s site visit, there were over one
hundred Level 3 inmates housed at USP Lewisburg. Management personnel reported
that a combination of factors resulted in a higher than normal number of inmates in
Level 3 and 4 being housed in the SMU. The Bureau was unable to identify the
number of Level 3 inmates housed with Level 2 and Level 1 inmates are Lewisburg.
Factors influencing housing at USP Lewisburg included the following:
USP Florence was no longer accepting Level 3 inmates as a result of a revised
agency plan to phase out the SMU program at USP Florence. This decision
temporarily reduced the number of available Level 3 and 4 beds in the SMU
program.

USP Lewisburg had space available in the SMU while USP Allenwood, a Level 3
and 4 facility, had limited beds available to accept additional inmates.
Several inmates in Level 3 and 4 had a scheduled release date into the
community that was within the next few months, and a decision was made to
allow those inmates to remain at USP Lewisburg until their release.
Limited bed space was reported to be available at FTC Oklahoma City which
serves as an administrative security federal transfer center designed to house
holdover inmates in-transit to other facilities.
The transfer review and authorization process including the centralized efforts
by staff at the DSCC, located at the Grand Prairie, Texas, office complex,
reportedly impacts the timeliness of transfers. Staff reported that several
inmates had been approved for Level 3 program placement. However, because
of a lack of alternative space being available and/or inmate separation issues,
numerous inmates were on an approved waiting list pending transfer out of
Lewisburg or were awaiting a decision regarding their transfer. This backlog
appeared to impact the program progression process and the conditions of
confinement received for inmates in Level 3 and 4. The average time on the
waiting list for transfer to USP Allenwood was 60 days as reported by the
associate warden for programs. A delay in transfers to various degrees was
noticed at all three facilities providing a SMU program, however it was most
significant at USP Lewisburg.98
The waiting list for inmates to transfer to USP Allenwood was not based on a
chronological order. Staff reported that command personnel from USP
Allenwood are allowed to select inmates on the approved transfer list who
appear to be the most compatible with the inmate population at USP Allenwood.
As a result, some inmates who had progressed to Level 3 and were approved for
transfer had to wait an extended period of time prior to being transferred due
to the selection process. The selection process appears to be in contrast with
one of the primary goals of Levels 3 and 4, which is to prepare inmates to
coexist in a general population setting with other individuals and groups.
The project team reviewed several inmate files at each of the three SMU programs.
The purpose of the review was to assess the progression process and to determine
whether inmates were advancing in the program as described in the SMU program
statement. As a result of the review, it was determined that in most situations

The Bureau reported that policy requires transfer applications be processed within 60 days.

inmates were progressing through the program in time frames consistent with their
adjustment to the guidelines established for the program. Inmates who were in
compliance with the guidelines were routinely advancing in levels. Inmates who were
not completing assignments and/or were receiving disciplinary infractions were
appropriately not progressing and in some cases regressing (i.e., going from Level 2
to 1 in the program. Overall, staff was consistently monitoring inmate behavior and
the inmates’ progression through the program. Personnel assigned to the unit teams
were found to be very familiar with the criteria for advancement in the program and
they appropriately documented the progress of each inmate on their caseload.
The most significant issue noted in the file review was that when an inmate
completed a particular level, and the completion required a facility transfer, the
timeliness of the transfer was not consistent with maintaining the integrity of the
program. This practice was most prevalent at USP Lewisburg, although the delays
occurred at all three SMU facilities. Although the inmate would advance to Level 3,
they would remain at the facility and not receive similar conditions of confinement
as those inmates in Level 3 housed at USP Florence or USP Allenwood. As a result, the
inmates’ consistent exposure to less restrictive conditions of confinement based on
level completion was often delayed. For example, expanded out-of-cell time, access to
larger recreation areas, increased interaction, access to indoor recreation and access
to additional program activities did not always occur in a timely fashion.
During the course of our review at USP Lewisburg and USP Allenwood, we noted that
delays in movement from Level 2 from Lewisburg to Level 3 at USP Allenwood were
occurring. Inmates advancing to Level 3 and 4 from USP Lewisburg were considered
for transfer to USP Allenwood or USP Florence. At the time of the site visit to
Lewisburg only USP Allenwood was accepting Level 3 inmates from Lewisburg. The
average time on the waiting list for transfer to USP Allenwood was 60 days according
to the associate warden for programs. Similar information was obtained during the
subsequent review of USP Allenwood.
Level 3 inmates awaiting transfer from USP Lewisburg are generally held in housing
units E, F and I. Due to the limited space and separation issues, numerous inmates
who have completed the program and have earned a transfer from the facility are on
a transfer waiting list. Staff reported there were 112 inmates in Level 3 housed in the
SMU. This backlog proves to be a concern and disrupts the smooth flow of inmates
who should be progressing through the system. The waiting list for transfer is not
managed chronologically. USP Allenwood personnel are allowed to select inmates
who are most compatible with the population at that facility, based largely on
separation issues.
Inmates in Level 3 housed at Lewisburg pending transfer or release do not receive the
same programing opportunities as inmates in Level 3 at USP Allenwood. Inmates who
advance to Level 3 at USP Lewisburg earn credit toward completion of Level 3 and are

afforded some of the Level 3 privileges to the extent possible. However, no group
activities are allowed outside of the recreation that is being provided at Lewisburg.
Long delays in advancement to USP Allenwood, are problematic and are somewhat
self-defeating for the SMU program.
Staff at both USP Lewisburg and USP Allenwood reported that the backlog in the
system was in part the result of the limited number of appropriate beds available in
the Bureau. Staff indicated there are few beds available at USP Allenwood, which
provides Level 3 and 4 housing.
FINDING: Current backlogs in inmates awaiting transfer to the next program level
negate the intent of the program design and decrease the motivation to change
behavior. Further, it is inconsistent with the program’s objectives to hold
graduates of Level 2 in a unit that operates with that level’s restrictions rather
than receiving the benefits of advancement to Level 3.
The project team visited each of the SMU programs and observed highly qualified
and trained staff closely monitoring the progress of each inmate assigned to the SMU
program. This monitoring process was reflected by staff’s routine on-site presence in
the units and a review of the inmate files where progression compliance was well
documented. The design of the SMU program held the inmate accountable for his
actions. The inmate was responsible for taking an active role in the program and
through the successful application of self-study and individual participation he was
able to advance through the levels. The program activities that have been established
focus primarily on the development of positive behavior and values that are designed
to assist the inmate in successfully residing in a general population setting.
The project team found that the SMU program was essentially a two phase program
where in each phase there are currently two levels. The first major phase primarily is
what the Bureau refers to as Level 1 and 2 which consists of in-cell programming.
With the exception of recreation and individual access to an electronic law library
kiosk almost all programming is provided to the inmate while he remains in the cell.
The Bureau Program Statement, P5217.01, Special Management Unit, states inmate
interaction with other inmates at Levels 1 and 2 is designed to be “minimal”. The
project team found interaction was indeed minimal and that there was very little
difference in the conditions of confinement between the two levels. The primary
difference was the number of telephone calls allowed per month and the topics
covered in the treatment curriculum.
The second major phase of the program includes Levels 3 and 4. This phase is
designed to focus more on preparing the inmate for general population housing.
Through expanded out-of-cell time, increased interaction with others and exposure
to a revised treatment curriculum the inmate is being presented the opportunity to

100

experience conditions and treatment that will allow him to demonstrate “positive”
community interaction skills. The program statement cites that during these two
levels “frequent group counseling sessions” will be offered. The project team did not
identify any frequent group counseling sessions occurring – either first hand through
observation or through staff/inmate interviews. There was one reentry group that
had recently started at USP Florence however most programming, outside of
recreation, continued to be provided individually while the inmate remained in his
cell. The primary difference in the conditions of confinement between Level 3 and
Four was is in the treatment curriculum provided. The project team did find some
minor differences between the two levels involving frequency of telephones calls or
types of visits provided at an isolated facility, however, this was not the case at all
three facilities.
The expanded focus on developing preparedness to enter the general population was
the key difference observed between Levels 3 and 4. In reality, Level 4 has become an
almost perfunctory step at USP Allenwood. When the inmate graduates from Level 3
and moves to Level 4, the re-designation request is immediately submitted. The
completion of the Level 4 program is almost universally four months, except where
major issues arise with the inmate’s adjustment. In reality, once Level 3 is completed,
the inmate is prepared to return to general population.
Based on other state practices and the observations of the SMU programs, the
program should be consolidated into a three-phase program rather than the current
four-phase program.
In order to achieve this, Level 3 and Level 4 as now constituted in the SMU would be
consolidated into a single program phase. It would operate like the current Level 4.
Level 1 and 2 would continue to operate as presently constituted.
RECOMMENDATION 4.7: Reexamine the SMU levels as they currently operate,
their corresponding conditions of confinement, the length of time in each level,
and their compliance with the SMU program statement. The program should be
consolidated into a three-phase program rather than the current four-phase
program and the minimum length of time to complete the program adjusted
accordingly.
Additional concerns noted by the project team involved the delays in progression
through the levels, specifically when the progression involves a facility transfer.
Inmates requiring a facility transfer in order to continue to advance in the program
were often delayed access to the full scope of conditions that came with the new
level because of the delay. As a result of the delays, extensive backlogs in transfers
develop and the opportunity for access to the full scope of less restricted conditions
of confinement exist. This practice was most evident at USP Lewisburg.

101

As mentioned previously in this section of the report, the lack of frequent group
counseling sessions being offered for Level 3 and 4 inmates was a concern identified
by the review team. The Bureau program statement on SMUs clearly states that in
Level 3 and 4 frequent group counseling sessions are to be included. Group
counseling was limited and the continued practice of providing individual in-cell
programming was primarily being offered to inmates in Level 3 and 4.
Lastly, the progression process identified by the project team raised concerns due to
the inconsistency in conditions of confinement offered to inmates in the same level.
For example, inmates in Level 3 of the SMU program housed at USP Allenwood are
allowed up to 150 minutes in social telephone calls per month, two (2) one-hour
noncontact social visits per month and indoor and outdoor recreation. The same
inmate if housed at USP Lewisburg would be allowed twice the number of social
telephone call minutes per month (up to 300 minutes), more than twice the amount
of social visits per month (five (5) one-hour social visits) and outdoor recreation only.
FINDING: There is a lack of consistency in the conditions of confinement for an
inmate classified at the same level in the same program when housed at a
different facility. This presents concerns regarding the integrity and design of the
level system.

Due process and ADX
In October 2012, a memorandum was issued by the assistant director of the Bureau’s
Correctional Programs Division, outlining a revised referral process for ADX general
population placement.99 The key provisions of the referral process are outlined in
this memorandum and are summarized in the following paragraphs.
Placement of inmates in the ADX general population is at the discretion of the
assistant director of the Correctional Programs Division (CPD). The Executive Panel
(assistant director, North Central regional director, warden of Florence ADX) retains
authority for placement of inmates into the ADX control unit.
Placement criteria: Referrals for placement at ADX–general population must meet
one or both of the following:

99
Memorandum from Assistant Director, Correctional Program Division, Oct 15, 2012,
Administrative Maximum Facility General Population Referral Procedures.

102

Inmate’s placement at other facilities creates a risk to institutional security and
good order or poses a risk to the safety of staff, inmates, others or to the public
safety.
Due to the inmate’s status either before or after incarceration, the inmate may
not be safely housed in the general population of another institution.
The memorandum outlines the factors that are sufficient to warrant consideration
for placement:
The inmate is subject to restrictive conditions of confinement as a result of a
SAM or based on documented information from a government agency that the
inmate was linked to terrorist activities and presents national security
management concerns.
The inmate is subject to restrictive conditions and the sentencing judge
imposes restrictions on contacts by the inmate pursuant to 18 U.S.C. 3582(d) or
a similar statute.
The inmate engaged in any conduct that is prohibited by any federal or state
law in the facility where the inmate was housed.
The inmate has committed two 100 or 200 level prohibited acts within the last
60 months.
After being validated as a member of a Disruptive Group the inmate committed
any 100 level prohibited act.
The inmate has been identified as participating in, organizing, or facilitating any
group misconduct that adversely affected the operation of the facility.
The inmate engaged in behavior that is of such severity that it is determined
that the inmate would be unable to function in a less restrictive environment
without being a threat.
The notoriety of the inmate is such that his well-being would be jeopardized if
placed in a less secure facility.
The inmate has access to resources to the extent that housing in a less secure
facility would pose a higher probability of escape.
There is a detailed outline of the referral and review process for inmates who were
already designated, for inmates referred for initial designation after sentencing, and
for referral of those within the witness security program. The process for review of
these referrals is somewhat similar and requires the appointment of a hearing

103

administrator to conduct a due process hearing. The review process, similar to that
for SMU placement, is a multi-tiered review process that includes reviews by the
warden, regional director, the administrator of the Intelligence and Counter
Terrorism Branch, the chief of DSCC and the assistant director of the Correctional
Programs Division. The Office of Medical Designations and Transportation also
performs a review of the case in relation to mental health and medical needs. The
intent of this review is to exclude inmates from ADX placement if they have serious
mental or medical illnesses. This review is cited in P5310.16. In addition, a
requirement to include psychological testing in the mental health evaluation is
incorporated in this policy along with more detailed guidance regarding the content
of the mental health evaluation.
A hearing administrator will be designated by the national discipline hearing
administrator to conduct a due process hearing for inmates who are referred for
placement at ADX–general population. Notice is provided to the inmate at 24 hours
prior to the hearing. The inmate has the right to be present for the hearing and have
opportunity to make a statement and present evidence to the hearing administrator.
At the conclusion of the hearing, the hearing administrator shall prepare a written
recommendation on whether placement at ADX is warranted.
The report is submitted to the assistant director of the Correctional Programs
Division for review and either acceptance or rejection of the placement. If accepted
the Chief of DSCC is notified of the final decision to initiate the placement.
The process for admission to the ADX control unit is similar. However, the request is
forwarded to the regional director for the North Central region (as the ADX falls
within that region) and the final decision must be jointly approved by the assistant
director for the Correctional Programs Division and the regional director for the
North Central region.
After the adoption of the ADX referral and assignment process as outlined in the
October 13, 2012 memorandum100, all inmates already assigned to ADX were
afforded a hearing, retroactively.101

100
The Bureau reported that they first implemented these procedures on January 1, 2008. They
have since been updated and modified. In November 2009, the agency decided to provide this
due process placement hearing, following the new procedures and criteria outlined in the
guidance memorandums, to all inmates housed at the ADX who had not received such a
hearing.

In our interviews, one ADX inmate claimed that he was never afforded a hearing before or after
his assignment to ADX. Along with ADX Florence staff, we reviewed the inmate’s file and could
find no documentation that the inmate was ever afforded a hearing. ADX Florence staff attempted
101

104

Through the file reviews, interviews with inmates, and review of the operations of
ADX we confirmed that inmates are advancing through the ADX program into the
step down program that can lead to assignment to general population outside of
ADX. At the time of our visit to USP Florence, 18 inmates were in the final stage of
the step down program, with several awaiting assignment to an appropriate unit in
the general population of USPs.
The review of due process for ADX found general compliance with the referral and
admission policies and procedures, including completion of appropriate
documentation.
The review of ADX control unit inmate records revealed missing or incomplete
documentation directly related to the ability to verify due process-type protections in
four of the files reviewed. This is consistent with our review of SMU inmate records
where we noted that due process-type documentation for SMU inmates was missing
from some of those records.
As noted in the assessment of Bureau re-entry programs we found that there is little
education or information sharing for inmates who are completing the lengthy ADX
confinement process and preparing to be either released to the community or
released to general population at a non-ADX facility. This is detailed further in the
reentry section of this report.

Administrative remedy program
The policy and procedures of the Bureau Administrative Remedy program are
outlined in PS 1330.18 dated January 6, 2014.
The purpose of the Administrative Remedy Program as stated in PS 1330.18 is to
allow an inmate to seek formal review of an issue relating to any aspect of his/her
own confinement. An inmate may not submit a Request or Appeal on behalf of
another inmate. As stated in the policy document the objectives of the
Administrative Remedy Program are as follows:

to locate the documentation while we were on-site. Then, they were invited to locate the
appropriate documentation of the hearing and to notify us when it was located. We never received
such notification.

105

Provides a procedure that is available to all inmates by which they will be able
to have any issue related to their incarceration formally reviewed by high-level
Bureau officials.
Provides that each request for review, including appeals, will be responded to
within the time frames specified in the policy.
Provides for a process to ensure that a record of inmate administrative remedy
requests and appeals will be maintained.
Through the administrative remedy process, it is believed that Bureau policies
will be more correctly interpreted and applied by staff.
The Bureau’s Administrative Remedy Program contains no time or subject matter
limits, thus allowing any inmate a separate and unrestricted avenue to raise issues or
seek relief regardless of when the alleged incident occurred.
FINDING: The Administrative Remedy Program provides a redundant level of due
process protection for all Bureau inmates, beyond that provided by many state
departments of corrections.
Using the process, inmates can challenge any aspect of their confinement, including
segregation placement and conditions at any time. For disciplinary segregation
inmates, because their challenge is to the DHO decision, the challenge goes directly
to the regional office via form BP-10. For SMU and ADX inmates the challenge goes
directly to the Office of General Counsel in the Central Office via form BP-11.
The warden is required to appoint one staff member, ordinarily above the
department head level, as the administrative remedy coordinator (coordinator) and
one person to serve as administrative remedy clerk (clerk). The regional director and
the national inmate appeals administrator, Office of General Counsel, is to be advised
of these appointees and any subsequent changes.
To coordinate the regional office program, each regional director is required to
appoint an administrative remedy coordinator of at least the regional administrator
level, ordinarily the regional counsel, and an administrative remedy clerk.
PS 1330.18 establishes timelines for responses from the appropriate Bureau officials.
Administrative Remedy fillings and responses are tracked via the Bureau central
computer system. A listing of all remedy filings for the period from December 19,
2012 to December 19, 2013 for all inmates housed in SMU was provided. A total of
404 filings are listed for this time period for inmates housed in USP Lewisburg, USP
Allenwood, and USP Florence. The report provides a status code for each remedy
request filed. These codes (and their unedited language supplied by the Bureau) are
as follows:

106

ACC (Accepted)—The inmate has properly filed an administrative remedy at the
appropriate level (i.e., the packet that the inmate submitted has all required
documentation for the level that the inmate is filing at) and has properly
exhausted at the lower levels.
REJ (Rejected)—The inmate has not properly filed an administrative remedy at
the level that the remedy was submitted at (i.e., the packet that the inmate
submitted does not have all required documentation for the level that the
inmate is filing at) and/or the inmate has not properly exhausted at the lower
levels prior to submitting the remedy. We do not reject appeals because the
nature of the issue is not valid for a remedy request. We respond to all appeals
even though the inmate may have to use another avenue for their request (i.e.,
Tort Claims), but we do not reject an appeal based on a nonvalid issue.
CLD (Closed Denied)—The inmate will NOT be granted the relief that they are
requesting in their administrative remedy.
CLG (Closed Granted)—The inmate WILL be granted the relief that they are
requesting in their administrative remedy.
CLO (Closed Other)—An appeal can be closed using CLO for various reason
which will be reflected in the status reason code. For example, CLO ISJ improper subject matter; CLO MOT request or appeal is moot; CLO REP request
or appeal is denied as repetitive of previous filing; CLO WDN withdrawn at
inmate’s request; CLO XPL information or explanation for the inmate’s request
is only provided; or CLO OTH, this can be used when an inmate who filed an
administrative remedy has died.
VOID—An appeal can be voided out of the system when the information initially
entered for the inmate is entered incorrectly and needs to be corrected for a
proper record of the appeal submission.
The status of the 404 remedy filings by SMU inmates listed on the documents is
noted in table 23.

107

Table 23.

Remedy/SMU filings by reason, December 19, 2012 – December 19, 2013
Decision

Number

Percent

Accepted

118

29%

Closed—denied

22%

Closed—granted

0.%

Closed—other

18%

Rejected

122

30%

Void

Granted

404

100%

Total
Source: Bureau of Prisons.

The above analysis does not included filings that are informally resolved as provided
by PS 1330.18, paragraph 542.13. The most obvious indicator from this review is the
fact that none of the 404 filings are listed as granted.
Similarly, data on filings for SHU inmates for the same time period were obtained
from the Bureau. For the time period from December 19, 2012 to December 19, 2013
a total of 285 requests were noted on the data sheet. The status of the 285 filings by
SHU inmates are shown in table 24.
Table 24.

Remedy SHU filings by disposition type, December 19, 2012 – December
19, 2013
Decision

Number

Percent

Accepted

101

35%

Closed—denied

Closed—granted

Closed—other

Rejected

147

51%

Void (VOD)

<1%

Granted

285

100.00%

Total
Source: Bureau of Prisons.

As noted with SMU filings, these data do not include filings that have been informally
resolved per the policy. Also similar to the SMU the most obvious indicator is that
none of the requests for relief have been granted.
Each facility AR clerk maintains a separate tracking and accountability system to
track each filing from initiation of the remedy request to investigation and response,

108

through the appeals process. These systems are time sensitive so that reminders and
flags are attached to ensure timely responses.102
The AR clerks also ensure that AR’s are entered into the central Bureau system.
Institutional staff reported that the local tracking systems (spreadsheets) are more
detailed and helpful than reliance on the central tracking system.
Inmates can appeal UDC or DHO decision via the Administrative Remedy Program.
UDC decision is appealed locally (form BP-8) and DHO decision is appealed to the
region (form BP-10).
A 2013 study, “Procedural justice and prison: Examining complaints among federal
inmates (2000-2007),”103 which was conducted by David M. Bierie appears to validate
the Bureau’s grievance system. “Generally speaking, people feel a process is more
‘just’ when their voice is heard before decisions are made, decision makers treat
everyone equally, outcomes are proportionate, and there is a process of appeal or
challenge if they don’t agree with an outcome.” The opposite is also true if the
system is perceived to be unfair; thus, the grievance process plays “a central role in
generating compliance or defiance” by inmates.
The study found that the Bureau’s grievance system is perceived by some inmates as
overly formal and more concerned with procedural practices and deadlines than the
substance of a complaint.
The study also reported that most complaints concerned issues related to discipline,
medical care and staff, with food, housing and use of force at the bottom of the list.
The timelines for responses to PS 1330.18 remedy requests were summarized by the
national inmate appeals administrator during an interview on November 14, 2013.
These timelines are as follows:
Initial filing: 20 calendar days from the date on which the basis of the filing
occurred
Wardens review: Warden’s review is to be completed within 30 calendar days
with a possible 30 day extension with cause

For example, PS 1330.18 requires that grievances to wardens must be answered within 30
days, appeals to the region must be responded to within 30 days and appeals to the central
office must be responded to within 40 days, each with possible 30-day extensions.
102

103
“Procedural justice and prison: Examining complaints among federal inmates (2000-2007),”
by David M. Bierie. Psychology, Public Policy and Law, Vol. 19(1), Feb. 2013.

109

Regional director: Inmate must appeal to the regional director within 20
calendar days of the date Warden signed the response. The regional director
must respond within 30 calendar days with an additional 30 day possible
extension
Central office appeal: Appeals submitted to the central office are to be
responded to within 40 days with a possible 30-day extension with cause
The national inmate appeals administrator reported that in November 2013 central
office was 11 months behind in responding to appeals to central office. It was
reported that this was due to the volume of complaints and the complexity of the
appeals. However, the administrator reported that the office has made a concerted
effort to process restrictive housing appeals in a more expeditious manner, resulting
in less time to address these issues.

110

Chapter 5: Mental health
assessment and treatment
One of key areas of concern in the use of restrictive housing and/or solitary
confinement is the mental health status of the people who are assigned to such
housing units. As suggested in the literature review, it has been found in other
studies that large proportions of the segregated populations suffer from mental
illnesses that either predate admission to restrictive housing and/or develop as a
result of the restrictive housing experience. Regardless of the basis for the mental
health illness, it is essential that inmates with a current mental health illness in
restrictive housing be properly diagnosed and treated.
In this chapter, we review the Bureau’s mental health population that is assigned to
SHU, SMU or ADX facilities and units. As noted in the earlier chapter, only a small
percent of the Bureau’s restrictive inmate population has been identified with a
significant mental illness of some kind. This figure likely represents an under
identification of those inmates who are truly suffering from some form of a mental
health illness.
The most recent mental health status data on state, local jails and the Bureau are
based on a 2005 survey. As shown in table 25, large percentages of all three inmate
populations were found to be diagnosed by the external researchers to have at least
one recent history (past 12 months) of a mental health illness. In terms of current
symptoms, the percentages are somewhat lower but still significant. For the Bureau
the estimate was 31 percent of the inmates having current mental health illness .
symptoms.
Also shown in table 25 are the November 2013 mental health care levels.
Significantly, 93 percent of the population are in Care Level 1 which is the lowest
care level available. This is not to say that some portion of these inmates have no
mental health issues. The Bureau reported that under current policy, inmates
assigned to Care Level 1 can and do have such symptoms but do not raise the level of
elevated treatment beyond medication.

111

Table 25.

Symptoms of a mental health disorder among prison and jail inmates

Symptoms in past 12 months or since admission
Major depressive or mania symptoms

State
prison

Federal
prison

Local
jail

Persistent sad, numb, or empty mood

32.9

23.7

39.6

Loss of interest or pleasure in activities

35.4

30.8

36.4

Increased or decreased appetite

32.4

25.1

42.8

Insomnia or hypersomnia

39.8

32.8

49.2

Psychomotor agitation or retardation

39.6

31.4

46.2

Feelings of worthlessness or excessive guilt

35.0

25.3

43.0

Diminished ability to concentrate or think

28.4

21.3

34.1

Any attempted suicide

13.0

6.0

12.9

Persistent anger or irritability

37.8

30.5

49.4

Increased or decreased interest in sexual activities

34.4

29.0

29.5

Thoughts of revenge

28.4

21.3

34.1

Delusions

11.8

7.8

17.5

Hallucinations

7.9

4.8

13.7

Care level 1

N/A

93%

N/A

Care level 2

N/A

Care level 3

N/A

<1%

N/A

Care level 4

N/A

<1%

N/A

Psychotic disorder symptoms

Bureau mental health care levels

Source: Survey of Inmates in State and Federal Correctional Facilities, 2004, and the
Survey of Inmates in Local Jails, 2002.
Nationally, the Bureau of Justice Statistics estimates that at least 15 percent of the
state inmate population has symptoms of psychotic disorders as compared with the
Bureau’s percentage of 13 percent.104 A very small percentage of the Bureau inmates
with serious mental illnesses are transferred to specialized mental health treatment
facilities located at the Atlanta, Butner, Springfield, Carswell, Devens or Rochester
facilities. In addition, residential mental treatment units addressing a range of
mental health conditions are located at Atlanta, Coleman, Danbury, and Terre Haute.
104
Bureau of Justice Statistics, Mental Health Problems of Prison and Jail Inmates, September
2006, NCJ 213600.

112

The Bureau advised the project team that additional residential mental health
treatment units are planned for activation in 2015 at Allenwood and Florence.
This assessment is intended to evaluate those inmates in the Bureau with a serious
mental health problem to determine if the diagnosis and treatment plans prescribed
by the Bureau are appropriate. Recommendations are made at the end of the chapter
that would serve to improve the current mental health system within the Bureau.

Assessment process
Two board-certified psychiatrists (Dr. Pablo Stewart and Dr. Roberta Stellman) were
retained to conduct independent mental health assessments of inmates assigned to
the various restrictive housing units within the Bureau. Each psychiatrist was given a
structured interview form to be completed on each sampled inmate. The assessment
was based on: (1) a brief review of the existing BOP mental health record retained at
the BOP facility and (2) a face-to-face confidential interview with the inmate.
For each inmate taking part in the review process the CNA retained psychiatrist was
asked to make the following assessments and opinions:
Did the CNA psychiatrist agree with the Bureau mental health diagnosis?
o If no, what was diagnosis is recommended by the CNA psychiatrist?
Did the CNA psychiatrist agree with the Bureau prescribed treatment including
medication and out of cell treatment?
o If no, what is the recommended treatment for the inmate?
Did the CNA psychiatrist believe the inmate was appropriate for placement in
restrictive housing?
o If no, what form of housing is recommended?
Were there any other comments that pertained to this inmate’s mental health
status or treatment appropriate for this person?
For each selected facility, a list of inmates housed in restrictive housing and
scheduled to be assessed was provided by CNA several days in advance of each visit.
Inmates that were sampled were predominantly assigned to Mental Health Care
Levels 2 and higher. As noted in Chapter 3, the specific definitions provided by the
Bureau for the four care levels are as follows:
CARE Level 1-MH—no significant mental health care

113

CARE Level 2-MH—routine outpatient mental health care or crisis oriented
mental health care
CARE Level 3-MH—enhanced outpatient mental health care or residential mental
health care
CARE Level 4-MH—inpatient psychiatric care
In addition to the level of mental health care, the Bureau also provided the date the
inmate was admitted to either SHU, SMU or ADX status in order to calculate the
length of stay in restrictive housing. In sampling the cases, efforts were made to
ensure that people who had experienced lengthy periods of time in restrictive
housing were included in the sample. As such the sample is not a pure random
sample but rather a purposeful sample that was designed to ensure inmates with
various lengths of stay were captured.
CNA also sampled a small number of inmates who were a) assigned to Mental Health
Care Level 1 and b) who had been in restrictive housing for extensive periods of time.
These cases were sampled to determine if some levels of de-compensation in their
mental health status had occurred since being assigned to restrictive housing.
Due to the inmate refusals there were some instances where additional inmates were
evaluated simply based on their willingness to be assessed. These “nonsampled”
cases were included in the overall evaluation to ensure a sufficient number of
inmates were evaluated at each facility.
It is recognized that there has been considerable debate about the reliability of
psychiatric diagnosis between psychiatrists. Previous studies have shown low level of
inter-reliability when two psychiatrists are asked to assess the same mental health
patient.105 However, in this situation the project psychiatrists were not being asked to
develop a full psychiatric assessment to formulate a diagnosis. Rather the task was
to review the current diagnosis and treatment plan to determine whether the two
were consistent with one another. This type of review is commonly done by
supervising psychiatrists who oversee mental treatment units. Both of the project
psychiatrists who conducted these reviews have considerable experience in this area.

Aboraya, Ahmed, Eric Rankin, Cheryl France, Ahmed El-Missiry, and Collin John.
“The Reliability of Psychiatric Diagnosis Revisited: The Clinician's Guide to Improve
the Reliability of Psychiatric Diagnosis.” Psychiatry. January 2006; 3(1): 41–50.

105

114

Inmate interviews and case review trends
Table 26 summarizes the overall sampling results. The 12 facilities/housing units
reviewed had a total of 2,683 inmates in restrictive housing shortly prior to each site
visit. Of that population, only 297 (or 11 percent) were determined by the Bureau to
have some mental health issue. Of these 297 inmates, 180 (or 61 percent) were
interviewed and assessed. For some facilities, the number of cases evaluated
exceeded the number of mental health care inmates at the facility. These higher
numbers reflect inmates who were not designated with a significant mental health
issue but were evaluated to assess the accuracy of the Level 1 mental health status.
Table 26.

Cases interviewed and assessed
Population in
restricted housing at
time of site visit

Mental health care
level 2 or higher

Inmates
interviewed
and assessed

Allenwood SHU

116

Atlanta SMU MH

Facility

Butner SHU

Coleman SHU

171

Florence ADX

368

Florence SMU

493

Florence SHU

Hazelton SHU

Lewisburg SMU

779

Terre Haute SHU

200

Tucson SHU

221

Victorville SHU

221

2,683

297

180

Total
Source: Bureau/JFA

Table 27 shows the medical and mental health care levels for the interviewed
inmates. The most frequent level of mental health care is Level 2 followed by Care
Level 3. The 35 Care Level 1 inmates were those that had been assessed by the
Bureau as not having a substantial mental health problem. However, the CNA
psychiatrists concluded that 7 (or 20 percent) of the 35 cases had a significant
mental health problem.
While the sample size is quite small and may not be completely representative of the
entire Care Level 1 population, it raises the possibility that a proportion (20 percent)
of the entire Care Level 1 population may have a significant mental health issue that

115

has been missed by the Bureau mental health system. This conclusion is supported
by the non–mental health interviews where a significant proportion of the inmates
reported being depressed and/or having medical symptoms related to depression
and anxiety (e.g., loss of weight, inability to sleep).
The Medical Care Levels are also noteworthy with almost half of the sample having
significant medical problems, which placed them in Care Levels 2 and 3. In terms of
their time in restrictive housing at the time of the site reviews, the average overall
time was 242 days. The ADX inmates had the longest average number of days in
restrictive housing at 959 days.

116

Table 27.

Key health care attributes of assessed inmates

Attribute
Total cases

Frequency

% of Total

180

100%

Mental health care status
Mental health care 1

19%

Assessed as care level 1

77%*

Assessed as care level 2

11%*

Assessed as care level 3

9%*

Mental health care 2

114

63%

Mental health care 3

17%

SHU

27%

SMU

108

60%

ADX

13%

Care level 1

47%

Care level 2

38%

Care level 3

Unknown

Restrictive housing status

Medical care status

Average time in restrictive housing

8%
242 days

SHU

107 days

SMU

144 days

ADX

959 days

* Percentages based on the 34 care level 1 inmates. For one case there was no
CNA assessment.

Finally, in general terms, the independent psychiatric review found considerable
disagreement in the core areas of a mental health program. These include: initial
mental health diagnosis, adequate psychiatric staff coverage and coordination with
other mental health staff, provision of adequate out-of-cell treatment services based
on individualized treatment needs, review/modification of prescribed medication,
and the capacity to quickly remove an individual from a segregated environment and
place them in a health services treatment unit for residential treatment.
FINDING: Based on the review of the inmate mental health records and the inmate
interviews, the reviewers disagreed with the BOP diagnosis in nearly two thirds of
the cases reviewed. The review further indicated that the treatment being offered
by the BOP was insufficient or inappropriate in over half of the cases reviewed.

117

CNA believes that approximately one-third of the cases reviewed should not be
assigned to restrictive housing and about 30 percent should be placed in a
specialized mental health program or residential treatment unit similar to the one
implemented at USP Atlanta and found in many state prison systems. These units are
structured clinical environments that provide daily programming and a therapeutic
milieu for individuals who cannot function in general population due to their mental
illness and frequently receive disciplinary reports as a consequence. It is noted that
the Bureau has indicated that, in 2015, the Atlanta program will be expanded with
the establishment of a similar treatment program at USP Allenwood.
Table 28.

Key conclusions of independent psychiatric review
Assessment item

Inmates

Total sample

180

100%

Disagree with bureau diagnosis

114

63%

Inappropriate treatment

53%

Does not require restrictive housing

36%

Needs mental health program

29%

Clinical analysis and observations
The following section provides a more in-depth assessment of the current BOP
mental health system as it is operating in the SHU, SMU and ADX restrictive housing
environments. The intent is to better understand the statistical data presented earlier
in this chapter and to identify problems that are restricting the delivery of effective
mental health services to inmates in the SHU, SMU and ADX housing units.

Mental health diagnostic process
As noted above, there was considerable diagnostic disagreement between the Bureau
mental health staff (which is comprised mostly of psychologists) and the CNA
psychiatrists. Some of the reasons for this level of disagreement are outlined in the
following.
Infrequent updates of initial mental health assessments. The Bureau does complete a
comprehensive psychological assessment at the time of admission to the Bureau.
However, there is not a similarly comprehensive re-evaluation during the course of

118

their incarceration within the restrictive housing units other than monthly progress
notes – and these are completed only for those on the Bureau mental health caseload.
These assessments are usually based on brief, cell-front visits106, which are part of
the mandatory rounds mental health and medical staff must make on at least a
weekly basis. At USP Hazelton when private interviews occurred, they were
performed by psychology interns rotating through the psychology service as part of
their graduate training. Although the use of interns can be beneficial to both the
interns and the agency in terms of service, the use of such interns can lead to a lack
of continuity in programming when the interns rotate off the service, as was reported
by the staff at USP Hazelton.
RECOMMENDATION 5.1: All inmates should be seen in a private setting for a
comprehensive mental health evaluation prior to placement in any restrictive
housing environment.
The initial evaluations should assess the presence of a mental illness and the
determination of whether the inmate can tolerate the conditions of restrictive
confinement. They should also assess the presence of the potential for self-injury
and active signs or vulnerabilities for significant mental health de-compensation.
It is acknowledged that a pre-screening procedure is in place at all Bureau facilities.
Per 5324.08 Suicide Prevention Program, the Suicide Prevention Program Coordinator
is to provide SHU staff with a list of inmates with mental health conditions who may
become dangerous, self-destructive, or suicidal when placed into SHU. The
Correctional Services Supervisor is to immediately notify Psychology Services if one
of these inmates is placed in SHU. In addition P5310.16 mandates a comprehensive
mental health evaluation, to include psychological testing, prior to placement at the
ADX.
Many of the mental health inmates interviewed were in need of and were receiving
psychotropic medications. However, the notes in the case file did not follow a
differential diagnosis pathway. Rather it appears that the inmate receives a
diagnosis, most often by a practitioner without specialty training in psychiatry, and
is treated for that diagnosis alone without consideration that they may actually have
another condition that manifests with similar symptoms but for which the treatment
differs. There is little, if any, consideration of more than one diagnosis or evolving

106
The revised Treatment and Care of Inmates with Mental Illness policy, issued May 1, 2014,
now mandates private meetings with inmates classified as CARE2-MH and above. This policy
change was aimed at addressing an identified concern with cell-front sessions. The policy was
not in effect at the time of the site visits and thus was not assessed by the team.

119

consideration that the initial diagnosis was incorrect and perhaps the person suffers
from another disorder entity requiring alterations in the treatment plan.
The concern is that modification to the initial set of prescribed medications and
treatment plan may be needed due to (1) a comprehensive follow-up diagnoses,
(2) consideration that the initial diagnosis was incorrect, and/or (3) that the person
has developed an illness requiring alterations in the initial treatment plan.
FINDING: The lack of on-going assessments can lead to the absence of a proper
mental health status evaluation.
For example, at USP Tucson there had been a recent suicide in the special housing
unit. The inmate in the adjoining cell (diagnosed with depression and traumatic brain
injury) was interviewed. He expressed significant difficulty in adjusting the death of
his neighbor, which increased his own depression and suicidal ideation. After the
loss, he requested to be put in the restraint cell and have the staff keep his
medications. He was identified as being on the mental health caseload but, at the
time of the site visit, was not being considered for transfer to a treatment unit or
receiving additional mental health services.
RECOMMENDATION 5.2: A complete re-evaluation of the mental health record
should be performed by psychology and psychiatry staff every 30 days. Included
in this review should be a face-to-face interview by a member of the mental health
team in a private setting and the results of this interview included in the reevaluation record.
The re-evaluations should assess the presence of a mental illness and the
determination of whether the inmate can tolerate the conditions for segregated
confinement. They should also assess the presence of the potential for self-injury
and active signs or vulnerabilities for significant mental health de-compensation.
It should be noted that after the completion of the site visits the Bureau issued a
revised Program Statement 5310.16, Treatment and Care of Inmates with Mental
Illness. The provisions of this modified program statement address in policy some of
the issues observed during this review. Specifically, the policy states the following in
Section 8a, Restrictive Housing:
Ordinarily, all critical contacts, regardless of an inmate’s mental health care level,
will, to the extent possible, be conducted in a private area. These include the
following:
Diagnostic assessments
Suicide risk assessments

120

Crisis intervention contacts
Protective custody reviews
Sexual assault prevention intervention
Mental health treatment contacts as indicated by the treatment plan
Any other service that addresses potentially sensitive issues or high-risk
behaviors
Additionally, all inmates with mental illness in restrictive housing units (e.g., SHU,
SMU, ADX) will receive, at a minimum, face-to-face mental health contacts consistent
with the type and frequency indicated by their care level, to the extent feasible. These
contacts take place in a manner that protects an inmate’s privacy to the extent that
safety and security of staff are not compromised. Contacts should be consistent with
the goals of the treatment plan, and are in addition to any critical contacts or
contacts required by policy (e.g., SHU Review).107
Due to the timing of the issuance of this program statement, the CNA team was
unable to assess the impact of the requirements on the actual delivery of services to
those in restrictive housing.
Lack of close coordination between psychology and psychiatric staff. Much of the
treatment being provided to the inmates takes the form of psychiatric medications
that can only be prescribed by a psychiatrist or a psychiatric nurse practitioner who
are very familiar with the patient’s symptoms and prior mental health history.
In the majority of the facilities inspected, the prescribing physician was not a
psychiatrist, which further added to the problems of coordination with the
psychology staff. The electronic medical record system also contributed to the
diagnostic difficulties encountered. Psychology and medical services document in
different electronic records so one must exit one software system and enter another
to try and integrate the treatment for those with mental illness108.
FINDING: A number inmates in restrictive housing demonstrated significant
symptomatology compatible with the presence of a serious mental illness, which
was undetected by the psychology staff. This is based not only on the cases
107

Program Statement 5310.16, Treatment and Care of Inmates with Mental Illness.

The Bureau has reported that as of April 2014, the electronic medical and mental health
records have been integrated and all providers document in BEMR. Due to the date of
implementation this was not verified.
108

121

interviewed by the CNA psychiatrists but also the larger number of Mental Health
Level Care Level 1 inmates who were reporting symptoms of depression, lack of
sleep and loss of weight, and anxiety.
RECOMMENDATION 5.3: A vigorous quality improvement program should be
established.
The Bureau should ensure that the psychologists are adequately identifying all of the
inmates who suffer from mental illness. The existing quality assurance measures in
place, to include remote reviews of the mental health record conducted by the
Psychology Services Branch should be included in an internal evaluation of the
quality assurance programs.
Assessments and contacts are not completed in private confidential settings. A major
system-wide deficiency is the practice of providing the vast majority of clinical
mental health contacts in a nonprivate, cell-side rounds format. According to Bureau
policy, the psychology staff is expected to complete a direct contact with every
inmate in restrictive housing on the mental health caseload at least monthly and
conduct weekly rounds on all inmates in restrictive housing109.
FINDING: Very few of the monthly mental health assessments occur in private
settings on a face-to-face basis.
Instead, these contacts typically occur through the cell door within the presence of a
cellmate and within earshot of the inmates housed in adjacent cells. In this
nonconfidential environment it is unlikely that an inmate will confide vulnerabilities
including suicidality. Therefore, inmates with worsening mental illnesses or the onset
of new symptoms can remain under-identified throughout the course of their
restrictive housing incarceration.
An example of this situation was a young man interviewed at the USP Florence
facility. He was actively delusional, isolative, and demonstrated a full complement of
symptoms compatible with chronic paranoid schizophrenia. Psychology staff
suspected he had a mental illness because of his poor hygiene; yet he had not been
adequately interviewed in a private confidential setting for sufficient time to reveal
his psychopathology.

109
The Bureau noted that P5310.16 Treatment and Care of Inmates with Mental Illness issued
May 1, 2014 mandates private, at least monthly sessions with CARE2-MH inmates and private,
at least weekly sessions with CARE3-MH inmates. In addition, critical contacts, such as suicide
risk assessments, are also to be conducted in a private setting.

122

Another example was encountered at USP Lewisburg, where an overtly psychotic
inmate was assessed by the psychology staff as being antisocial and was not offered
any mental health services.
Lack of psychiatric staff. Inmates who are identified as having a mental health issue
treated by pharmacological methods are most frequently evaluated and followed by a
general health practitioner (sometimes a physician but more often a physician’s
assistant) and not a board certified psychiatrist. There is a clear shortage of
psychiatric physicians throughout the facilities that were visited.
FINDING: The shortage of psychiatric staff in Bureau facilities leads to numerous
problems in both diagnosis and treatment, particularly for the seriously mentally
ill inmates.
Recommendation 5.4: Given the level of disagreement in the assessment and
treatment plan formulation, the Bureau should conduct an inter-reliability test for
its mental health staff to better determine the accuracy of the diagnosis and
treatment plan process.
Most facilities have limited psychiatry hours that are insufficient to assess and treat
those inmates with mental disorders in the system, not just restricted to the SHU or
SMU. Psychologists must prioritize psychiatric review to only a handful of the least
stable mentally ill inmates on their caseloads. Table 29 summarizes current and
vacant mental health treatment positions at reviewed Bureau facilities.

123

Table 29.

Mental health treatment staff at reviewed facilities
Doctoral-level
psychologist

Psychology
interns/
postdoctoral
residents

Treatment
specialists

Social
workers

Psychiatrists

Psychiatric
nurse
practitioners

Contract
psychiatry or
tele-psychiatry

FCC
Allenwood

Tele-psychiatry

FCC Butner

N/A

FCC
Coleman

Tele-psychiatry

FCC Florence

Contract and
tele-psychiatry

USP/SFF
Hazelton

Tele-psychiatry

3 psychologists

Facility

124

USP
Lewisburg
FCC
Terre Haute
FCC Tucson

FCC
Victorville

Vacant positions as
of
October 30, 2014
1 psychiatrist*
2 psychologists
1 social worker*
3 psychiatrists
2 psychologists*
1 treatment specialist
1 psychiatrist
1 psychologist
3 treatment
specialists
1 psychiatric nurse
practitioner
1 psychologist*
1 social worker

Tele-psychiatry

1 postdoctoral
resident
1 psychologist
2 treatment
specialists

Tele-psychiatry

3 psychologists

N/A

Tele-psychiatry

FCC = Federal Correctional Complex.
* Vacant position is newly allocated.

2 psychologists
2 treatment
specialists
2 treatment
specialists

Psychiatric medications were observed in several cases to be prescribed in
subtherapeutic or inadequate doses to treat the identified condition. At USP Terre
Haute, a mid-level provider prescribed a tricyclic antidepressant to an inmate with
serious depression despite this family of antidepressants being potentially lethal in
an overdose situation. When medications were prescribed, inmates were not seen for
review in a timely manner compatible with community and correctional standards of
practice. One inmate at USP Florence had not been seen by any medical provider
regarding his medications since 2012. Some sites, including USP Florence and USP
Atlanta, had no psychiatric presence within the prison whatsoever. All psychiatric
services were provided by tele-psychiatry at those facilities.
Wardens expressed their frustration with their (and the Bureau’s) inability to recruit
and retain psychiatrists. Tele-psychiatry is available at sites without a psychiatrist,
but the hours provided are so limited that the referral system is delayed and
insufficient to meet the health needs of the inmates.
For example, USP Florence has four hours per month of tele-psychiatry time available
for 128 SMU inmates. Time is split between two psychiatrists, one of which has
refused to staff cases with the psychology staff. On average, no more than six to
eight inmates can be seen per tele-psychiatry session. The on-site physician assistant
and nurse practitioner will not follow psychiatric patients. The facility physician sees
inmates on psychiatric medications but frequently will not start medication if the
inmate is not deemed to be “a perfect match for a DSM criteria” or put patients on
drug holidays, a practice that is discouraged because of the likely risk of a relapse.
The assessment team was told that the Bureau brought in their chief psychiatrist
several months ago to try and catch up with inmate care. One inmate summed the
situation up by saying that access to psychiatry at USP Florence was “impossible.”
Similarly, USP Terre Haute has four hours of tele-psychiatry available every 4-6 weeks
and only 23 percent of their SHU mentally ill have been seen by the psychiatrist.
At USP Hazelton, 16 hours of psychiatry services are available per month for 600
female inmates. Given the traditionally high degree of expressed psychopathology in
female incarcerated populations, this amount of dedicated psychiatry time is
insufficient to meet the needs of the population. Understaffing in this area leads to
potential under diagnosis, inadequate treatment, and delayed referral because the
psychiatrist only has time to see the most severely mentally ill inmates.
Tele-psychiatry should be a resource of last resort when there is no psychiatrist
present in a facility. It should not be routinely relied upon for day-to-day psychiatric
care.
For example, in one case at USP Terre Haute, the psychology staff noted there was a
significant delay in the receipt of a progress note by the tele-psychiatrist. In that one
case, no note has yet been received despite the inmate being seen six weeks ago.

125

At USP Florence, the sick call system was reported to be unreliable and not
confidential – both of which are requirements by any national accrediting body and
federal standards that require reliable access to care. This observation was also
reported by inmates at USP Terre Haute.
While on site, the assessment team was notified that USP Tucson had hired a full
time psychiatrist.
RECOMMENDATION 5.5: Psychiatrists need to be more actively involved in the
diagnostic and treatment process.
Currently, the psychiatrists only get involved if a patient is brought to them for
evaluation. The psychiatric staff needs to work more closely with the psychologists
to ensure that all of the mentally ill inmates are properly identified and referred to
treatment. The psychiatrist should also meet with the psychology staff on a monthly
basis to review the medication and case management plan for each Mental Care Level
2 and 3 inmates housed in restrictive housing.

Mental health treatment
Lack of out of cell treatment. Inmates in restrictive confinement rarely receive out of
cell mental health programming within the Bureau. Instead, the primary form of
treatment consists of written materials, such as cognitive behavioral handouts,
delivered to the cell by psychologists.
For example, written homework assignments are passed out to the inmates who are
expected to complete them in their cells and turn them within a few days. These
assignments are then graded by the psychologist and the results are used to
determine whether the inmate can progress to the next level in the SMU or ADX
progressive programs. These workbooks do not constitute mental health treatment.
Efforts to provide out of cell group activities at the expanding USP Florence SMU
were curtailed because of the timing of such activities had to conform to the
Bureau’s approved evidence based plan of approved group therapies. Psychologists
at the site level have not been encouraged to implement innovative group services to
fit the mental health needs of their inmate populations.
There are some notable exceptions that were positive in nature. For example, at USP
Florence, one psychologist was conducting a weekly out of cell group in the SMU.
Psychologists at the ADX Florence have recently begun seeing some of their patients

126

in out-of-cell settings.110 USP Allenwood is another facility where the psychologists
attempt to see their patients in out-of-cell settings. Another exception is the Level-III
treatment program at the USP Atlanta. A very conscientious lead psychologist
directed this program and saw to it that her patients participated in out of cell
treatment activities.
FINDING: Overall most restrictive housing units had no mental health programing
and especially no out of cell programming for any inmates with or without mental
illness.
Some interviewed inmates reported little or no response by psychology to their
requests for individual counseling. Exceptions to this practice were noted at the
United States penitentiaries at Allenwood and Atlanta, as well as at the ADX. As
noted above, facility staff reported they do not have adequate staffing in the facility
to provide more than the minimum required weekly rounds and monthly cell front
checks.
RECOMMENDATION 5.6: A program of regular out-of-cell mental health treatment
should be implemented.
Specifically, it is generally accepted that inmates with serious mental illnesses in
restrictive housing should receive a minimum of 10 hours of unstructured out-of-cell
time and at least 10 hours of structured therapeutic activities. They also should
receive weekly, out-of-cell clinical interviews by their assigned psychologist.
Some interviewed inmates with serious mental illnesses in restrictive housing, with
the noted exception of USP Florence, described their conditions as worsening in
confinement or not improving. These reports by inmates were especially prevalent at
USP Lewisburg which houses the largest number of SMU inmates (almost 800) of
whom approximately 50 are assigned to Mental Health Levels 2 and 3.
Lack of sufficiently trained mental health staff to provide treatment to the segregated
housing units. A consistent finding of the assessment team was that access to mental
health services is directly related to the level of professional expertise by the chief
psychologist as well as the number of mental health staff available at each site. In
several of the facilities reviewed, the project psychiatrist opined that the chief
psychologist did not possess adequate experience or clinical skills to run a
comprehensive mental health program. This is especially problematic in that the
facility psychologists act as gatekeepers to psychiatric and mental health treatment.

CNA was informed by the Bureau that this change has come about due to the ongoing litigation
about mental health system inadequacies.
110

127

FINDING: Almost all facilities reported a lack of mental health staff required to
provide treatment services.
USP Hazelton was down five psychology positions with no psychologist assigned to
the SHU. USP Florence had only three psychologists and one psych tech for a 700inmate population with no dedicated psychologist for the segregation unit. This duty
was split among the psychologists on the staff. In most facilities, providing
treatment services to the restrictive housing units was not the primary focus of the
mental health staff, which has to provide care to the larger general population
inmates with mental health needs.
It is acknowledged that in comparison the Bureau’s mental health staffing levels
exceed that found in many state correctional facilities. As noted previously, the
Bureau employs a large number of doctoral level psychologists – more than 600, a
higher doctoral level psychologist staffing rate than many state systems.
RECOMMENDATION 5.7: The Bureau should complete a clinical staffing needs
analysis.
Based on the results of this clinical staffing needs analysis, the Bureau should then
recruit and retain a sufficient number of psychiatrists to meet agency demands. No
specific number for psychiatrists can be offered until the staffing analysis is
completed. It is acknowledged that there are significant challenges associated with
the recruitment and retention of psychiatrists. It is a fact that there are a decreasing
number of training programs in psychiatry, a decreasing number of applicants for
existing programs, and a decreasing number of graduates from programs – all at the
time of a national shortage of psychiatrists.
Improper assignment to restrictive housing. As noted earlier, there were a number of
interviewed inmates who had demonstrated serious and unstable psychiatric
symptomatology, which should have excluded them from a restrictive housing
setting.
FINDING: This review identified inmates in restrictive housing whose mental
conditions should have precluded them from assignment to these units.
In some cases, the facility’s lack of mental health treatment warranted a transfer to a
specialized treatment program that did not require the level of security that was
operating at the SHU, SMU or ADX units. It is acknowledges that legitimate security
needs can be associated with a small, violent segment of the mental health
population. A heightened level of security may be required for these inmates, even
within a specialized treatment program. This complicates placement options for
these inmates.

128

RECOMMENDATION 5.8: A protocol needs to be established that identifies those
inmates with serious mental illness who should be excluded from SHU, SMU or
ADX housing.
A similar review and re-assessment protocol should be implemented that facilitates
the identification of those inmates who decompensate while in SHU, SMU or ADX
housing.

Other site-specific observations
SMU programming. Psychology contacts with SMU inmates are primarily a selfguided activity based on the delivery of written handouts and homework
assignments. This does not constitute a treatment program driven by an
individualized mental health treatment plan. This behavioral approach for inmates
with only disciplinary problems might be adequate, but is insufficient for individuals
with significant mental health conditions.
The SMU program does not take into account the inmate’s mental illness in
evaluating the inmate’s progress or lack thereof with the four SMU program levels.
For example, one inmate (who would benefit from residential mental health
treatment) had been assigned to the SMU for three years. He has an SMI diagnosis
and a history of repeated self-injury and persistent suicidal ideation. He was almost
to Level 3 in SMU when he overdosed on a potentially fatal antidepressant and was
regressed back to Level 1. This exemplifies a case of someone with a serious mental
illness/personality disorder being punished for his psychopathology, rather than
treated for it. He reported to the CNA psychiatrist that he is only seen during the
weekly rounds despite requesting more intensive counseling for over three years.
Long-term segregation effects at ADX. ADX Florence presented an interesting mix of
mentally ill patients. While there were a significant number of seriously mentally ill
individuals who required care at ADX, there were also a significant number of non–
mentally ill inmates housed at ADX.
A majority of these inmates made it very clear that they wanted to remain in the ADX
Florence and would commit a serious offense to ensure their ongoing housing in the
facility. Several of the inmates interviewed said they would assault someone if they
were told that they were going to be transferred to another Bureau facility. The
reason given was their belief that the yards at the various USP’s were exceedingly
more dangerous and they knew that they would likely have to kill someone on the
yard if transferred out of the ADX.
Among the interviewed inmates, none stated that they wanted to be transferred from
the ADX, which was a tribute to the level of care the inmates are receiving.

129

It should be noted that part of the desire for these inmates to remain at ADX is the
unique and often close relationship these men have with the staff. It was clear from
our observations that ADX staff knew the inmates very well in terms of the basis for
their placement in ADX but also they individual needs and interests.
FINDING: The assessment team encountered no cases where an inmate’s serious
mental illness was due to their prolonged placement in the ADX.
This reluctance to leave ADX Florence may be related to privileges such as reading
materials, television, and recreation activities afforded inmates at ADX and the
professionalism of the security and program staff assigned there. As noted above,
the quality and quantity of the mental health care at ADX has recently improved.
Delays in transferring inmates out of SHUs. Inmates in SHUs can wait for many
months for an opening in a program at another prison. During this time no
additional mental health services are offered which potentially can have significant
adverse effects while the inmate remains under segregation conditions.
RECOMMENDATION 5.9: All inmates who are found to be decompensating from
the effects of restrictive housing should be transferred to the most appropriate
unit for treatment and observation.
RECOMMENDATION 5.10: Inmates with serious mental illness who are not
excluded from restrictive housing should start participating in a treatment
program.
Such programming should consist of a minimum of ten hours of out-of-cell
structured therapeutic activities and an additional ten hours of out-of-cell
unstructured activities should as yard and dayroom time. Within these standards the
treatment programming should be individualized to the inmate’s specific condition
and treatment needs.
Large numbers of protective custody inmates who require mental health treatment but
are not receiving it. As indicated earlier in this report, these inmates are supposed to
be receiving protection from other inmates by the Bureau. However, the restrictive
nature of the SHUs makes it very difficult to afford any form of meaningful mental
health treatment to these inmates. For example, at USP Coleman, there was an
overrepresentation of protective custody inmates being housed in the SHU with only
five hours per week of out-of-cell time.
RECOMMENDATION 5.11: Inmates should not be housed in a SHU for protective
custody but rather, should be in sheltered general population housing.
Innovation at Hazelton to enhance mental health services. At USP Hazelton, the
psychology staff has a “Hot List” which is a binder kept in the officers’ station
updated monthly that lists the inmate’s name, diagnosis and the psychology staff’s

130

concerns regarding the inmate’s risk of behavioral disturbances. The staff there was
also very active in a multidisciplinary meeting with security and classification to aid
in expediting women being progressed out of the SHU. This best practice approach
should be expanded to the other SHUs. It is our understanding that the Bureau has
done so.

131

This page intentionally left blank.

132

Chapter 6:

Reentry

Federal and state corrections facilities held over 1.6 million inmates at the end of
2010 — approximately one of every 201 U.S. residents.111 According to the National
Reentry Resource Center, 708,677 individuals are released to the community
annually from state and Federal prisons and another 9 million are released from
local jails each year.112 Persons involved in the criminal justice system often cycle in
and out of various correctional agencies throughout their lives. The reentry of an
individual into the community without appropriate support and resources is a major
public safety concern. This is a particular concern for those inmates being released
directly from the highly controlled environment of restrictive housing.
As noted in chapter 3, within a year over 2,000 inmates currently in ADX, SMU or
SHU status will be released from the Bureau to the community. A small but visible
number of SMU and ADX inmates (approximately 80) were released directly from
segregation to the community. A higher but unknown number of SHU inmates were
are also directly released to the community from SHU housing units. For these
reasons alone it is important that the Bureau have effective re-entry programs for the
restrictive populations.
Since the passage of the Second Chance Act in 2008, reentry has become a major
policy emphasis for corrections professionals. Hundreds of millions of dollars in
Second Chance and Justice Reinvestment monies have been dedicated to expanding
and improving reentry programs with the goal of reducing recidivism. While
researchers are evaluating the effects of these programs, the Urban Institute has
developed a “What Works in Reentry” clearinghouse to inform the field about
promising and best practices. Four of these guiding principles are as follows:

111 Guerino, P.M., P.M. Harrison, and W. Sabol. Inmates in 2010. NCJ 236096. Washington, D.C.:
U.S.
Department
of
Justice,
Bureau
of
Justice
Statistics,
2011.
http://www.bjs.gov/content/pub/pdf/p10.pdf
112 Beck, A.J. The Importance of Successful Reentry to Jail Population Growth. Presented at the
Urban Institute's Jail Reentry Roundtable, June 27, 2006. www.urban.org/projects/Reentryroundtable/upload/beck.PPT

133

1. Focus on individuals most likely to reoffend. Research clearly indicates that
successful programs begin with carefully sorting offenders according to their
risk --separating those likely to reoffend from those less likely.
2. Base programs on science and ensure quality. Researchers are very clear that
resources must be invested in program models that have the promise of
reducing recidivism.
3. Implement effective community supervision practices. Policies and practices
must provide supervision officers with a broad range of options for swift and
certain sanctions that are proportionate to the violation and appropriate for
the individual under supervision.
4. Apply place-based strategies. Place matters. Ensuring that resources are
available to offenders – where they live – is particularly critical to reducing
recidivism. For individuals to change their behavior, services and supports
crucial to their success must be nearby.
In order to place the Bureau’s approach to reentry planning for inmates in restrictive
housing in context, the project team first conducted a general review of reentry
services for general population inmates as well as the overall delivery of program
services to inmates in restrictive housing. However, the scope of this project limits
this review to the evaluation of the content and delivery of reentry programming to
the restrictive housing population and the findings and recommendations do not
apply to the overall reentry approach for general population inmates.

Reentry programs for inmates in the general
population
The Bureau’s website states that release preparation begins the first day of
incarceration; however, the focus on release preparation intensifies at least 18
months prior to release. The Bureau’s Release Preparation Program includes classes
in résumé writing, job search, and job retention while incorporating presentations by
community-based organizations that help ex-inmates find jobs and training
opportunities after release. In planning for release, the Bureau works with the United
States Probation System providing all pertinent information to the probation officer
that may bear on the safe and effective supervision of the released offender. This
information includes any record of medical, psychiatric, psychological, sex offender,
or substance abuse treatment.
All inmates have the responsibility to develop and submit a suitable release plan for
investigation and verification by the probation office in the district of supervision.
Release plans may include placement in a residential reentry center (also known as

134

community correctional center), normally for a period of up to 180 days, to afford
the inmate a reasonable opportunity to complete development of a suitable release
plan. When no adequate release plan is developed and an inmate will be released to
supervised release directly from an institution, U.S. probation officers may seek
modification of the conditions of release to include a special condition that the
inmate reside at a residential reentry center (RRC) or halfway house, a contracted
facility that provides assistance to inmates nearing release.113
The Bureau typically places appropriate inmates in RRCs prior to release to help
them adjust to life in the community and find employment. RRCs provide inmates
with a structured and supervised environment along with employment counseling,
job placement services, financial management assistance, and other communitybased social services and programs.114 RRCs facilitate inmates’ efforts at
reestablishing ties to the community while allowing staff at the RRC to supervise
inmates’ activities.115
RRCs provide suitable residence, structured programs, job placement, while the
inmates’ activities are closely monitored. The Bureau ensures the provision of mental
health, substance abuse, and sex offender treatment for offenders in RRCs and Home
Confinement through contracts with community-based treatment providers. The
Community Treatment Services section of the Psychology Services Branch is
responsible for establishing and overseeing these contracts, and the associated care
provided.
There are two program components: the Community Corrections Component and the
Prerelease Component.
The Community Corrections Component is designed as the most restrictive option.
Except for employment and other structured program activities, an inmate in this
component is restricted to the RRC. An inmate shall ordinarily be placed in the
Community Corrections Component upon arrival at the RRC. This orientation period
normally lasts for two weeks or until the inmate has demonstrated the responsibility
necessary to function in the community. Based on their professional judgment, the
RRC staff shall determine when an inmate is prepared to advance to the Prerelease
Component.

U.S.
Department
of
Justice,
Bureau
http://www.bop.gov/locations/cc/index.jsp.

113

114

Ibid.

115

Ibid.

135

Prisons,

Community

Corrections,

The Prerelease Component is designed to assist inmates making the transition from
an institution setting to the community. These inmates have more access to the
community and family members through weekend and evening passes. Participating
in community-based transitional services may reduce the likelihood of an inmate
with limited resources recidivating, whereas an inmate who is released directly from
the institution to the community may return to a criminal lifestyle.116
The Bureau provides a comprehensive array of programs that directly or indirectly
support reentry preparation for general population inmates. This programming
includes:
Education. All institutions offer literacy classes, English as a Second Language,
parenting classes, wellness education, adult continuing education, library services,
and instruction in leisure-time activities. Inmates who do not have a high school
diploma or a GED certificate are required to participate in the literacy program for a
minimum of 240 hours or until they obtain the GED. Non-English-speaking inmates
must take English as a Second Language.
Vocational training. Programs are based on the needs of the inmates, general labor
market conditions, and institutional labor force needs. An important component is
on-the-job training, which inmates receive through institution job assignments and
Federal Prison Industries. The Bureau also facilitates post-secondary education in
vocational and occupationally oriented areas.
Behavioral skill building. Parenting classes help inmates develop appropriate skills
during incarceration. Recreation and wellness activities encourage healthy life styles
and habits.
Substance abuse treatment. The Bureau offers four different levels of substance
abuse treatment: (1) education regarding substance abuse and its effects; (2) the
Residential Drug Abuse Program, which is a cognitive-behavioral program delivered
within a modified therapeutic community model where offenders experience living in
a prosocial community; (3) nonresidential drug treatment for offenders who have
short sentences; may not meet the criteria for, or are awaiting an opening in, the
Residential Drug Abuse Program; are transitioning to the community; or have had a
positive urinalysis test; and (4) community treatment services, which as a part of
reentry provides continuity of care for offenders placed in RRCs and on Home
Confinement.
Mental health. The Bureau provides formal counseling and treatment on an
individual or group basis with institutional psychologists, psychiatrists, social
116

Bureau Program Statement Change Notice #7310.04.

136

workers, and treatment specialists. The Bureau operates a series of residential
psychology treatment programs to provide more intensive care for inmates with
serious mental illness. In addition, medical referral centers provide inpatient
psychiatric care for acutely ill inmates.
Sex offender programs. The Bureau offers both residential and nonresidential
programs for inmates with a current or prior conviction for a sex crime (including
sex involving consenting adults such as prostitution or pimping). Inmates may also
be eligible if there was a sexual element in the crime.
Religion. Chaplains facilitate religious worship and sacred scriptural studies across
faith lines in addition to providing pastoral care, spiritual guidance, and counseling.
Religious programming is led by agency chaplains, contracted spiritual leaders, and
trained community volunteers. The Life Connections Program (LCP) and Threshold
Programs offer inmates the opportunity to improve critical life areas within the
context of their personal faith or value system. The LCP utilizes various faith
communities nationwide who serve as support group facilitators or mentors at
program sites and release destinations to enhance community reintegration. Reentry
preparation for inmates not eligible for the residential LCP is also offered through
the Threshold program that also seeks to strengthen inmate community reentry.
Threshold is a nonresidential condensed version of LCP that is active in institutions
throughout the agency.
Programs directly supporting reentry include:
Release planning. Inmates released under federal supervision (i.e., Supervised Release
Parole) must submit release plans for review and approved by the U.S. Probation
Office. Release plans have two primary components: residence and employment.
These plans must be submitted 90-days prior to a release.
Residential reentry centers. RRC placement provides an opportunity to
establish/solidify sound release plans (i.e., residence and employment) prior to
release to the community and to allow a readjustment to community life prior to
release. Consideration and referral for RRC placement should occur well before a
release date.
Release preparation program. The RPP assists inmates in developing plans for their
personal lives and future employment. The program offers six modules concerning
the personal, social, and legal responsibilities of civilian life: (1) Continuity of Care
and Infectious Disease; (2) Resume Writing; Money Management; (3) USPO and
Supervision Requirements; (4) Veterans Outreach (only for veterans); (5) Release
Requirements; and (6) Psychology of Release. Staff indicated that they encourage
inmates who are within 30 months of release to complete the RPP.

137

Reentry affairs coordinators. Reentry affairs coordinators (RACs) are assigned to an
institution or region and are responsible for preparing release readiness materials.
RACs perform orientation for inmates related to the various aspects of reentry,
which includes informing inmates of the requirements and benefits of the Affordable
Care Act, as well as other reentry topics, to include job placement, housing, benefits,
requiring identification, job skills, veteran benefits, Social Security benefits, etc. The
also develop partnerships to foster reentry efforts and continuity of care; serve as
the point of contact for outside agencies - providing training and information; help
identify areas that need to be addressed for the inmate population specific to each
institution and develop resources to address those needs; manage the volunteer
program; and compile data and information to assess reentry efforts.
Regional reentry affairs administrators. The regional reentry affairs administrators
provide direct supervision to the RACs; providing training and oversight, as well as
direct authority, monitoring and tracking for skill development and reentry
initiatives, including volunteer program activities throughout the region. They serve
as liaison with state and regional governmental agencies and organizations to foster
partnerships and develop resources to assist institutions in reentry efforts.
Additionally, they serve as the liaison with the NRB in Central Office.
In addition to these specific programs the Bureau also provides staff training on
reentry services both to new officers and on an in-service basis to current staff.

Reentry program evaluation
There have been very few rigorous studies of the impact of re-entry programs. Those
that have been completed have shown either negative or no effects on recidivism.
Project Greenlight was designed to provide reentry services for New York state
inmates. It was evaluated using a rigorous experimental design with random
assignment. One year follow-up results showed that the experimental group (reentry)
performed worse than inmates who were not exposed to the re-entry program.117
More recently, an initial national evaluation of 12 reentry sites funded by the U.S.
Bureau of Justice Assistance reported that although there was an increase in the
number and type of services provided to soon-to-be-released inmates, the services
did not produce significant differences between the experimental reentry inmates

117

Wilson, J. A., and R. C. Davis. 2006. “Good Intentions Meet Hard Realities: An Evaluation of the Project
Greenlight Reentry Program.” Criminology and Public Policy 5:303–38.

138

and the control group. The researchers believe the lack of an impact can be traced to
insufficient “dosage” of services and the exposure of services to the control group.118
To date, the Bureau has not conducted a formal evaluation of its reentry programs or
a formal recidivism study since 1994.119 This is surprising given the large number of
such studies that were conducted in the 1970s and 1980s which were used to
developed risk instruments and evaluate core treatment programs throughout the
field of corrections. The 1994 study showed that Bureau inmates released in 1987
had a re-arrest rate of 41 percent, which was well below the rate reported in studies
of recidivism among state inmates. This 3-year re-arrest rate was consistent with
previous studies conducted on Bureau inmates.
The lower Bureau re-arrest rate is consistent with the profile of the federal prison
population. In aggregate, Bureau inmates tend to be older, not convicted of violent
crimes and have modest prior criminal histories, and lower rates of mental illness as
compared to state inmates. Collectively these data suggest a large low risk
population that would require minimal reentry services.

Program services in restrictive housing
Program services in restrictive housing units are generally delivered by psychologists
and treatment specialists providing routine mental health services, crisis
intervention, and cognitive-behavioral interventions targeted to inmates’ specific
needs. Cognitive-behavioral interventions offered include anger management, basic
cognitive skills, criminal thinking, values, and the nonresidential drug abuse
program.
The physical design of most restrictive housing units as well as the severe limits on
inmate interaction makes meaningful program delivery difficult. Inmates are
generally not allowed to congregate in a classroom setting and space limitations in
most restrictive housing areas do not provide suitable areas for program delivery. As
a result, program staff services are provided on an individual basis via interactive
journals, books, audio presentations, cell-side visits, and private counseling sessions.
Specific types of restrictive housing programming include:
Bibliotherapy. Psychologists provide inmates with specific self-help books and
articles that target the inmate’s expressed interest/need. The staff defines
118

Lattimore, Pamela V and Christy A. Visher. 2013. “Prison Reentry Services on Short-Term Outcomes: Evidence
From a Multisite Evaluation. Evaluation Review, 37(3-4) 274-313.

119

Harer, Miles D. 1994. “Recidivism Among Federal Inmates Released in 1987.” Washington, DC: Federal
Bureau of Prisons, Office of Research and Evaluation.

139

“bibliotherapy” as “using books to aid people in solving the issues that they may be
facing at a particular time.” Examples include Dialectic Behavior Therapy, Rational
Emotive Therapy, and Chicken Soup for the Soul, etc. Inmates receive a certificate for
participation in the program.
ESL/GED. Small congregate education classes are provided on a very limited basis,
with inmates confined in individual cages or secured to chairs that are bolted into
the floor. For example at USP Allenwood, a new schedule has been introduced which
allows for inmates enrolled in ESL/GED classes to participate in a congregate
educational class for one hour per week. The maximum number of students in a
class at any one time is nine.
Self-study packets. A variety of educational, vocational, mental health, behavioral and
substance abuse treatment programs are provided through self-study packets
(workbooks to complete and be reviewed by staff). These workbooks are distributed
to inmates, collected upon completion, and then evaluated by staff. The SMU
programs at USP Allenwood and USP Lewisburg also use audio programs (referred to
as “radio” programming) with paper tests to verify successful completion of specific
courses.
Life Skills. The psychologists and treatment specialists distribute word games,
puzzles, and workbooks on stress management, communications, anger management
and other related topics. Other self-help programs include Anger Management,
Coping, Drug and Alcohol Abuse. These are booklet-driven programs that require
homework by the inmate. Written feedback is provided cell side. Completion of this
work is documented in the inmate’s electronic medical/mental health record.
The specific programs offered and mode of delivery in restrictive housing varied by
facility. Examples of the different approaches to programming at specific facilities
include:
ADX. The majority of the programming is provided in-cell on an individual selfstudy basis through closed circuit TV’s located within each cell. Within the past
six months, five “therapeutic enclosures” were built in the gymnasium so five
inmates at a time can receive congregate or group programming. Currently, the
psychology department is using the therapeutic enclosures to conduct a reentry
preparation program. One 90-minute group is meeting once a week.
USP Hazelton. Staff reported that their philosophy is that programming is
suspended while inmates are in the SHU because they believe inmates should
lose privileges when they are sent to restrictive housing. Staff also reported that
their practice is to minimize an inmate’s time in the SHU so that they can
resume programming as soon as possible. There are no congregate programs
provided to the inmates housed in SHU.

140

FCI Butner. Education courses are offered in six categories: Math (7 classes);
Science (3); Reading (4); Social Studies (4); Writing (4); and Miscellaneous classes
(4). Inmates can only enroll in one class at a time; coursework is provided on
paper and collected when completed. Inmates are expected to complete each
course in two weeks. If assistance is required, education staff will provide
assistance cell side.
USP Coleman. All programming in the SHU is in-cell and voluntary. There are no
structured programs. Staff will speak with the inmates one-on-one through the
solid steel cell door. Upon request, staff will provide a variety of written
materials. SHU inmates are not allowed to officially enroll in an education
program and are not allowed to take the GED exam.
The methods used to deliver programming in restrictive housing necessarily limit the
type and level of programming offered. For example at USP Allenwood, a Vocational
Trades Instructor offers two Adult Continuing Education courses to the SMU inmates;
a computer course and a heating, ventilating, and air conditioning (HVAC) course,
both of which are in-cell, self-study programs. Each course takes four to six months
to complete. The staff member stated she goes to the SMU for two hours, twice a
week and will go cell to cell or speak to the inmates during their recreation time
while out on the range. The staff member provides a test for each module that must
be completed prior to advancing to the next module. At this point in time, no inmate
has successfully completed either course. While this type of vocational programming
is important for preparing inmates to be productive in the community, this program
may be too difficult to complete by self-study.

Reentry program services in restrictive housing
The need for reentry programming in restrictive housing is predicated on the
likelihood of an inmate’s direct release to the community from restrictive housing, or
imminent release after transfer back to general population. The transition from the
high level of control and restrictions on behavior present in restrictive housing units
to the comparative freedom and lack of structure that inmates face upon release can
be highly disorienting. Appropriate reentry programming can assist inmates in
coping with this huge change in their living circumstances.
FINDING: There was no data available at any of the facilities visited that identified
the number of inmates released directly to the community from restrictive
housing.
However, staff at each facility acknowledged that inmates are being released directly
from restrictive housing. Data on the number of offenders being released would be
valuable to inform facility and Bureau leadership in making policy decisions
regarding the need to provide reentry programming to inmates in restrictive housing.

141

RECOMMENDATION 6.1: The Bureau should routinely track and monitor the
actual numbers of inmates releasing directly from restrictive housing at each
facility monthly.
The issue of inmates releasing from restrictive housing with little or no preparation
is significant. The magnitude of the issue is not fully known since no data was
available on the frequency of this practice. One staff member reported that they do
not need to track that information since their goal is to minimize the time an inmate
spends in restrictive housing. While the goal of shortening the time in restrictive
housing is correct and will help this situation, it ignores the fact that inmates are still
releasing from restrictive housing.
FINDING: Facilities do not provide step-down planning to transition an inmate
from restrictive housing to general population and subsequently to their eventual
release. The prevailing practice is to keep inmates in restrictive housing until
such time as they discharge to an RRC or directly to the community.
With the exception of assigned completion of self-study activities related to reentry,
inmates often abruptly transition from extended stays in restrictive housing to
general population or the community without any meaningful step down
programming. Many of the staff interviewed indicated that this was acceptable and
suggested that it was preferable to release inmates from the SHU, SMU or ADX rather
than first transitioning to general population due to the risk of violence to the
general population. While this may be a sound decision for institutional security, it is
not in the interests of the communities where these inmates are being released.
Inmates spending extended periods of time in confinement with little social
interaction or skill-building programming are seriously unprepared for reentry and
re-socialization.
RECOMMENDATION 6.2: Establish a policy whereby only under extraordinary
circumstances would an inmate discharge directly from a SHU, SMU or the ADX.
To support this policy, the Bureau should require monthly reports from each facility
on all inmates releasing from restrictive housing. This ‘exception’ report should
include the length of time in restrictive housing, specific reentry programming and
preparation provided, and documentation of the reasons why the inmate was
released from restrictive housing as opposed to step down or general population and
other relevant information. Requiring facility staff to provide this information will
help raise awareness of the problem and assist staff in finding ways to make better
decisions about moving inmates to less restrictive settings prior to release. Requiring
this type of report would force facility administration and Bureau leadership to
regularly examine these occurrences and take steps as appropriate to minimize this
occurrence.

142

RECOMMENDATION 6.3: Develop a step-down program with increasing
incentives, more out of cell opportunities and increasing opportunities for
congregate programming.
When inmates in SMU Level 4 go to general population, they are going from
restrictive housing to general population all at once, without any step-down or
transition. Providing a step-down from Level 4 before general population would
provide more meaningful programming and increase inmates’ social skills by
interactions with others, and would also provide incentives for inmates to work
harder on the programs and be less disruptive. The step-down process would also
make it possible for more inmates in Level 4 to try to be tested in more of a
congregate setting but with much less risk or exposure.
The following is a representative summary of reentry programming provided in
Bureau restrictive housing units. The common characteristic across all facilities
visited was the absence of any actual programming of consequence provided to
inmates.
USP Lewisburg. The facility is designed to house only SMU Level 1 and 2 inmates and
offers a Release Preparation Program (RPP) through self-study booklets and audio
programs. Inmates progressing to Levels 3 and 4 are transferred to USP Allenwood.
Staff indicated that inmates are released from USP Lewisburg to the community from
the SMU Level 1 or 2 if they are unable to progress to Levels 3 and 4 due to behavior
issues. Staff reported that some inmates do not want to be moved to Level 3 or Four
and act out just prior to advancing to ensure they will remain in Level 1 or Two until
release. One staff person described it as a way of keeping safe and away from general
population without showing weakness or fear.
For inmates who are to be released from the SMU, case managers begin informally
providing them with reentry services at about 18 months before release. This
includes more phone time to talk with family and their probation officer to prepare
for return to the community. Case managers also begin planning for home placement
and transfer to a RRC or “Public Law” placements in halfway houses. If a Public Law
placement is denied, the probation officer finds a shelter for the inmate to live in
when released. The RAC provides information for inmates related to the
requirements and benefits of the Affordable Care Act, as well as other reentry topics
such as job placement, housing, benefits, requiring identification, job skills, veteran
benefits, Social Security benefits, etc.
USP Allenwood. RPP programming for Level 3 and 4 inmates in the SMU at USP
Allenwood is individualized and consists of six modules: (1) Continuity of Care and
Infectious Disease; (2) Resume Writing; Money Management; (3) USPO and
Supervision Requirements; (4) Veterans Outreach (only for veterans); (5) Release
Requirements; and (6) Psychology of Release. All programs are offered through selfstudy packets or audio program. The RPP is voluntary and completion is not required

143

for Level progression. Staff indicated that they encourage inmates who are within 30
months of release to complete the RPP. In addition to the self-study packets,
psychologists work with inmates that have significant mental health needs to
facilitate linkage to community resources. This facilitation can include telephone
consultations with community resources, providing written materials from the
support services agencies, and release planning with family by telephone. In some
cases, inmates releasing to a metropolitan area such as Washington, DC can be linked
with advocacy agencies that have a specific mission to assist mentally ill inmates
releasing to the community.
USP Florence SHU. There is no coordinated, targeted reentry programming. Case
managers, counselors, psychology staff and education staff provide materials for
self-study upon request.
ADX Florence. Inmates confined to the ADX typically serve extended periods of time
there. All inmates at the ADX are provided release programming and there are no
exclusions due to risk or classification level. The staff reported that they try not to
have inmates release directly from the ADX, however, due to the nature of charges or
institutional adjustment, some inmates do release directly to the street. The case
manager and unit manager complete supervised release plans for inmates. The staff
that works with the ADX inmates noted that because of the nature of the inmates
they are supervising, they would ensure they have contact with the releasing inmate’s
probation officer. For those inmates within 18-24 months of release, halfway house
applications may be submitted to residential reentry managers for review/approval.
Due to the nature of the inmates at the ADX, many are denied placement in
residential release centers. As a result, staff has a lot of communication with the
probation officer in the inmate’s home location. Staff aims to transfer inmates to a
facility closer to their home prior to release however this was characterized by staff
as a significant challenge.
USP Terre Haute. A reentry affairs coordinator prepares release readiness materials,
including Affordable Care Act benefits, job placement, housing, benefits, personal
identification, job skills, veteran’s benefits, Social Security benefits etc. The
coordinator is also responsible for training institutional staff regarding reentry
concepts and procedures. Case managers work with the inmates as they prepare for
release.
USP Hazelton. Although there is no coordinated, comprehensive, targeted reentry
programming available for SHU inmates, case managers work with the SHU inmates
who are within 90 days of release to assist SHU inmates in obtaining a social security
card, birth certificate and driver’s licenses if appropriate and to develop home plans
which may consist of confirming a home address. The case manager sends the notes
on the inmate and the home plan to the probation officer in the home area. The
probation officer will investigate the home location and either approve or disapprove
of the plan. If disapproved, an alternative release and home plan is developed by the

144

case manager working with the inmate. The case managers also provide inmates
releasing to a halfway house or to the street, informational pamphlets including
information on “one stop centers” which are in most large cities that assist
reentering offenders with employment, disability employment, job seeker resources
and veteran resources.
General population inmates at Hazelton are often released to the Court Services and
Offender Supervision Agency and Hope Village in Washington, DC. The Court
Services and Offender Supervision Agency provides video conferences with general
population inmates and their children prior to release as a part of reentry
programming. However this program is not provided to inmates in restrictive
housing.
FCI Butner. Facility case managers develop home plans for each inmate to be released
from restrictive housing. This consists of identifying a home address so the
probation officer can check the residence out for suitability. Halfway house referrals
are completed and sent to residential reentry managers for review/approval although
some inmates releasing from restrictive housing are not eligible for a halfway house
due to the nature of their offenses. Inmates are not enrolled in any community
services until they are in a halfway house or back in the community. The case
manager also assists the inmate in obtaining a social security card, birth certificate
and driver’s license if appropriate.
There is no coordinated, comprehensive, targeted reentry programming otherwise
available. The reentry affairs coordinator assists case managers with reentry
preparation as needed. The unit manager may provide informational pamphlets on
“one stop centers” located in most large cities that assist reentering offenders with
employment, disability employment, job seeker resources and veteran resources. As
in other Bureau facilities, there are no educational classes in restrictive housing with
the exception that if an inmate was previously enrolled in education or requests
programming, ‘push packets,’ self-administered course work, are provided.
USP Coleman. Inmates must request information regarding reentry programs. There
are no structured programs. Once an inmate is within 18 months of discharge, the
case manager will begin release planning, which consists of verifying an address
upon release and ensuring the inmate has appropriate identification to include a
social security card, birth certificate and driver’s license if appropriate. A reentry
affairs coordinator is available to staff and inmates as a resource to help with reentry
services and provides the training to the staff on the topic of reentry during annual
refresher training.
As can be seen from these examples, inmates in Bureau restrictive housing have very
limited access to reentry programming. Services are generally limited to providing
basic information on identification and benefit issues, and referrals to community
programs and services. This stands in stark contrast to range and depth of reentry

145

programming provided to Bureau general population inmates. Ironically, it is the
restrictive housing population that is in the most need of programs and poses the
greatest potential risk in their transition back to the community.

Reentry program observations
Access to program data. Currently, the record of inmates who complete psychology
self-help programs such as Anger Management, Coping, Drug and Alcohol Abuse this
work is documented in the Psychology Data System (PDS) and in the Bureau
Electronic Medical Records System, but neither of these systems are accessible by
case managers and case managers need to be aware of all course work being
completed by inmates120. Inmates are provided with completion certificates and
Psychology Services staff provide feedback to the unit team at the time of the
inmate’s Progress Review.
RECOMMENDATION 6.4: Ensure that when inmates complete psychology self-help
programs such as Anger Management, Coping, Drug and Alcohol Abuse that
completion of these activities are documented in support of reentry planning so
that case managers and counselors are aware of these activities.
Lack of program strategy. Although inmates in restrictive housing receive
information about reentry, there is no formal coordinated, comprehensive, targeted,
specialized reentry program inmates Bureau wide who are unable to participate in
general population reentry programming due to their restrictive housing status.
Some facilities visited had small rooms within the restrictive housing unit that would
enable several inmates to be safely secured at tables to participate in small group
classroom training or even individualized training for the highest risk inmates.
RECOMMENDATION 6.5: Develop and provide a coordinated, comprehensive,
targeted, specialized cognitive reentry programming specifically designed for
inmates in restrictive housing.
While there needs to be a loss of privileges for inmates in disciplinary segregation,
there are many administrative segregation inmates and some disciplinary segregation
inmates who will spend long periods of time in restrictive housing. Currently, those
inmates will not return to general population and therefore not receive any
meaningful reentry preparation programming.

120
Bureau has advised the project team that since April 2014, PDS has been a part of BEMR. The
system is now integrated. Since site visits were complete we were unable to verify this improvement.

146

The Bureau should design instructor led cognitive programming to support reentry
similar to what is being provided to general population inmates. This programming
could be designed as the “Cliff’s Notes” version of reentry services in order to
provide at least the most critical reentry related programming to those inmates at
least 6 – 12 months prior to their release. Such programming does not have to be as
comprehensive as that provided to the general population, however it should be
meaningful.
Such a program could easily be designed by the Reentry Services Division staff to
include both instructor led classes with self-study and homework provided to these
offenders whose risk levels and disciplinary status would enable such controlled
group programming. Direct interaction with the inmate by the instructor would help
prevent cheating where the inmate’s cellmates complete homework assignments.
Considering the types of inmates being released from these units, and often after
extensive periods of time in restrictive housing, this should be a priority.
No substantive programming is provided in restrictive housing units at any level.
Some of the Bureau’s program descriptions refer to self-help reading materials as
‘program’ or ‘therapy’ however they are barely more than reading activities.
Additionally, there is currently no way to ensure an inmate actually did the work if
they are double-celled. Some inmates interviewed reported cheating by having other
inmates do the work so they could get credit for completing the assignments.
Program communication. Most contact with inmates housed in SHUs, SMUs, and the
ADX is cell side. This limits the ability for any meaningful or confidential dialogue.
Inmates are not likely to discuss sensitive or private issues shouting through a cell
door with other inmates and staff present.
RECOMMENDATION 6.6: Provide inmates the opportunity to participate in more
out of cell individual interviews.
Congregate programming. Interaction with a service provider, psychologist etc. is
essential for meaningful learning and aids in preparation for reentry by engaging in
discussion and social interaction. With the exception of the ADX Florence, no other
facilities visited provided any congregate programming. The ADX at USP Florence has
begun using specially constructed “Therapy Enclosures,” on a very limited basis.
While this may not be possible in every facility due to space constraints, it may be
feasible in some facilities.
RECOMMENDATION 6.7: Provide appropriately screened inmates with the
opportunity to participate in small group programs facilitated by a counselor,
case manager, psychologist, education instructor or other treatment provider.
Such programming should only be provided to those inmates the Bureau staff feel
are appropriate based on their disciplinary status, security risks etc. However, there

147

are a significant number of inmates who could be safely programmed in very small
groups secured to tables separated a safe distance from other inmates.
Program design. Research shows that programs for high-risk offenders should focus
on evidence-based practices that target their individual criminogenic needs. Inmates
in restrictive housing are unable to receive the programming available to general
population inmates.
RECOMMENDATION 6.8: Provide programming that identifies and addresses most
significant areas of need for high-risk offenders in order to assist the offender to
successfully reintegration into the community.
If limited programming is made available for selected inmates in restrictive housing
who are not likely to return to general population, the programming provided should
be designed to meet their most significant areas of need based on their risk needs
assessments and other relevant factors.
Training. Staff (including civilian and contractor staff working within the institution)
should receive additional training that reinforces the fact that reentry is not a
program, but rather a philosophy. Staff do currently receive information as part of
annual refresher training (ART) from the Reentry Affairs Coordinator and a brief
module in the in-service training however, in order to ensure a more comprehensive
and successful approach to reentry, particularly in restrictive units, all staff need to
be trained and expected to work collaboratively with each other, understand they
model and reinforce pro-social behavior as a way to encourage inmates to change.
RECOMMENDATION 6.9: Educate staff about the need for inmates in restrictive
housing to receive formal reentry programming if being released from restrictive
housing.
Reentry culture/philosophy. There is no formal Bureau-wide reentry preparedness
program specific to restrictive housing. Each facility visited seemed to have their own
unique programming that they tried to offer within the confines of the restrictive
housing unit. Most staff interviewed did not appear to recognize the need for
programming beyond self-help packet programs and no facility was able to provide
data on the number of inmates actually being released from restrictive housing. The
mindset that it is okay and preferable to discharge from a SHU needs to change.
RECOMMENDATION 6.10: Establish and maintain a culture among all BOP staff,
employees and contractors that recognizes the need for meaningful reentry
programs for “all” inmates in the Bureau of Prisons, including those in restrictive
housing, beginning at new officer/staff training and continuing in every annual
in-service training.
The Bureau of Prisons has made reentry a priority and has taken significant steps to
emphasize the importance and impact of a comprehensive reentry program. The

148

mission statement of the Bureau of Prisons (that is prominently displayed in all
agency communications including the BOP’s internal and external website, or internal
newsletter, and many ad hoc memorandums and publications) is to operate safe,
secure, humane prisons, and to prepare inmates for release. Bureau of Prisons staff
interviewed understand their responsibility to prevent inmates who are released
from returning to criminal activities in the community. Bureau staff (and inmates)
have been told that, “Preparation for release begins on the first day of incarceration”.
In the past few years the focus on reentry has been especially pronounced through
the creation of the Reentry Services Division, an organizational change in the agency
that required support from the attorney general and final approval from Congress.
In addition to establishing this division to coordinate and amplify the agency’s
reentry efforts and message, the director communicates continuously with staff
about this critical aspect of the mission, and he has also communicated directly with
the inmates. Specifically, the director wrote an open letter to inmates on June 19,
2013, expressing hopes that they embrace the opportunities provided to them to
pursue the education, training and treatment needed to succeed when they return to
the community.
While we salute the director and BOP leadership for making reentry a priority, the
reality is that inmates in restrictive housing are not afforded reentry services in any
significant way due to their housing restrictions and because in disciplinary
situations, loss of programs is a consequence of their placement.
Psychology and education staff in the facilities visited were making individual efforts
to provide more program services to this population. It is understood that the
challenges that restrictive housing creates and the associated security concerns are a
priority over programming. However, more effort should be made to provide
meaningful reentry services to inmates in restrictive housing while still maintaining
appropriate security.
Because of the risks that many of these individuals pose, it is essential that the
Bureau do more where possible to improve the reentry into the community in order
to protect the public’s safety.
The project team did observe and interview staff that believed reentry programming
is not possible or appropriate for those inmates in restrictive housing. It is
appropriate to not provide programming to inmates who are in restrictive housing
for short periods who will likely return to the general population where they can
resume programming. This recommendation is aimed at establishing a culture [and
policy] that recognizes the need for inmates in restrictive housing to receive more
meaningful reentry programming if they will likely never return to the general
population and will eventually be released to the community.

149

ADX Releases. Inmates are spending a significant period of time at the ADX and then
placed directly to a halfway house or released directly to the community. Some
inmates interviewed have spent from eight to fifteen years at the ADX and will be
releasing to a halfway house or to the community. The Bureau needs to develop a
step-down or transitional program to prepare inmates for this adjustment.
RECOMMENDATION 6.11: Review of the practice of keeping inmates at the ADX
until halfway house placement or direct release to the community on the inmate’s
release date.

150

Chapter 7: Restrictive housing
operations and conditions of
confinement
The section of the report addresses the Bureau’s approach to management of
restrictive housing units. This analysis includes assessment of organization, staffing,
training, security operations, and conditions of confinement. The Bureau
management approach is relatively uniform from one facility to another; however,
there are some unique characteristics to each facility’s management structure; a
number of which are described in the following chapter.

Organizational structure
Special housing unit management
The organizational structure governing the SHUs is similar across facilities in
structure and management approach. To a degree, the management structure is
bifurcated with security managed by the Bureau’s Correctional Services department
and issues pertaining to classification, programming, minor disciplinary matters and
case management assigned to the unit management teams.
Institutions with SHUs are generally but not always part of a larger complex of
facilities managed by a single complex warden. The complex warden is supported by
additional management staff that may include other wardens of a lower grade
reporting to him or her, as well as associate wardens that are deployed in facilities
throughout the complex. In the facilities that were assessed for this project, the
complex warden was managing a USP.
Within the USP a SHU is managed by the facility’s correctional services division and
overseen by an associate warden. These associate wardens are responsible for
supervising all uniformed staff in the correctional services division.
Direct oversight of the SHUs is the responsibility of the SHU lieutenant, who is the
uniform commander of the restrictive housing unit. These lieutenants supervise day-

151

to-day operations five days per week, and also are designated as the SRO, that
conducts the segregation reviews of the inmates assigned to the unit as required by
Program Statement 5270.10, Special Housing Units.
Only the USP Hazelton Secure Female Facility did not have an SHU lieutenant
assigned to manage unit operations. At Hazelton, the shift or operations lieutenant
supervised the SHU, in addition to his/her other duties, and on occasion an extra
lieutenant was assigned to the unit to supervise operations.
Correctional officers are not selected by management for their posts, but are
awarded their post assignments based on a seniority bidding process that takes place
each quarter of the calendar year. Therefore, SHU staff can turnover once every three
months, depending on bidding for the various posts and the officer’s status in the
seniority system. Lieutenants are assigned by the warden and his/her staff, based on
management prerogative. As outlined above, unit chain of command shifts when the
SHU lieutenant is off duty and the shift commander or operations lieutenant for the
facility takes over supervision of the SHU and the unit officer in charge (SHU#1 post).
The Bureau is invested in a decentralized management program known as Unit
Management. The theory behind unit management is to decentralize operations by
dividing large correctional institutions into more manageable units, where staff work
in close proximity to the inmates they manage. With respect to the SHUs, unit teams
are not specifically assigned to the unit to manage the unit and inmate activity.
Instead the unit team assigned to the inmate prior to placement in SHU will retain
management of the inmate’s case, and will see the inmate in the SHU. Day to day
responsibility rests with the SHU lieutenant or in the cases where there is no SHU
lieutenant on duty, the operations lieutenant.
Other departments of the institution provide services to inmates and regularly visit
the unit. Medical staff enter the unit a minimum of two times per day to administer
medications and conduct triage on inmate medical concerns. Facility chaplains,
educational staff, caseworkers, and counselors provide services at least on a weekly
basis. The shift supervisor or operations lieutenant visits the unit each shift that a
SHU lieutenant isn’t present. At least weekly, a captain will visit the unit as well as a
member of the top administration, such as a warden or associate warden. A mental
health clinician visits the unit at least weekly and is further required to conduct a
mental health review including a personal interview after every 30 days of
continuous placement in the SHU.
In a number of facilities assessed, the warden conducts rounds of the SHU with
members of his/her management staff. The assessment team observed this practice
at a number of facilities. Each member of the management team accompanies the
warden on the tour and observes each inmate’s cell, stopping to respond to
questions or concerns that the inmates may have.

152

Special management unit management
Administrative oversight of SMU programs is similar to what is found with oversight
of SHU programs. As noted earlier in this report, SMU programs at USP Lewisburg,
USP Florence, and USP Allenwood were assessed as part of this review. Generally,
each facility is managed by a warden and there are associate wardens assigned to
manage correctional services or custody operations, as well as associate wardens
responsible for programming. The special management unit custody and security is
overseen by the facility captain, or in some cases by a deputy captain. Lieutenants
are assigned, either directly to the units, as was the case at USP Allenwood, or
covering a geographical sector of the facility that includes multiple units, as was the
case at USP Lewisburg. Correctional officers are assigned to posts inside the housing
units supervising the inmate population on the various ranges.
Unit management personnel perform a prominent role in the SMU programs. Unit
managers, caseworkers, counselors and unit secretaries provide services to the
inmate population that includes classification, casework, disciplinary and
programming. Unit team members work cooperatively with correctional services staff
to manage each unit. It is important to note that correctional officers and lieutenants
report in a parallel chain of command to the facility captain/deputy captain. Unit
managers report to associate warden than the captain, although that is not always
the case. Regardless, there is a different chain of command for correctional services
staff and unit management staff.

ADX Florence unit management
The ADX Florence is managed directly by the complex warden, who is based
physically at the ADX. Three associate wardens, also based at the ADX, are
responsible for correctional services, programming, human resources, and a number
of other ancillary functions. ADX custody operations fall under the associate warden
for correctional services. A correctional captain manages security for the unit and all
custody personnel. This position is supported by a deputy, a position that was vacant
the time of the assessment. Each tour of duty is managed by a shift commander, who
is the operations lieutenant that oversees shift operations. There are additional
lieutenants that supervise “facility activities,” the SHU, the control unit, and the
special security unit. SHU, control unit, and special security unit lieutenant posts are
filled 16 hours per day on day and swing shift. An administrative lieutenant manages
staff scheduling and administrative duties.
Unit teams provide services to the inmate population and work cooperatively with
correctional services personnel. There are three unit teams that provide unit
management services e.g., inmate classification, minor discipline, casework. The SHU
does not have a separate unit team assigned to it and the facility unit teams follow

153

their inmates and continue to monitor them while they are housed in the SHU,
similar to the process described above for SHUs throughout the Bureau.

Restrictive housing management structure
observations
FINDING: The management structure of the Bureau facilities is staffed with
sufficient personnel to provide management and oversight of its restrictive
housing units.
Each complex is managed by a warden, who is further supported by subordinate level
wardens and associate wardens. Associate wardens supervise department heads of
the various departments throughout the correctional complex. These department
heads provide direct oversight of functional areas and provide support to the
segregated units depending on their function and responsibilities.
Department heads that have particular responsibility for the segregated units include
facility captains and unit managers. The captain provides oversight over lieutenants
who are the key figures in managing the segregated units, as first line supervisors of
the staff assigned there. The SHU and SMU lieutenants are the officers in charge of
operations and are tasked with the responsibility of ensuring that all aspects of
policy and procedure are complied with. These policies and procedures include
program statements, federal regulations, procedures and post orders.
The Bureau utilizes a bifurcated management structure in which the Correctional
Services Division oversees the uniformed staff and a unit team with a separate
command structure provides direct service to the inmate population.
Correctional services is responsible for most of the conditions of confinement an
inmate is subjected to, as well as ensuring discipline and order is maintained in the
units. As specified before, correctional services report to an associate warden, and in
a parallel chain of command, unit teams report to an associate warden as well,
providing casework, classification, minor discipline, and services to the inmate
population. In the SHUs, unit teams provide support; whereas, in the SMU the unit
teams provide direct service to the inmate population in the units they oversee.
Although a bifurcated chain of command is somewhat unusual, unit management is a
long-standing and successful management strategy that is used not only in the
Bureau, but also in correctional departments throughout the nation.
Correctional officers are the main providers of service and the staff that maintain
security, order, and discipline in the restrictive housing units. Officers self-select

154

their assignments to restrictive housing units on the basis of seniority: the exception
being the lieutenant who is assigned by the facility/complex warden121.
This is a practice mandated by the labor contract and is not the most ideal method of
staff assignment. Best practice dictates that staff assigned to restrictive housing
units are selected on the basis of their performance and competencies. Furthermore,
in many jurisdictions, they are rotated out of these units after a period of time
because of the high stress nature of the assignments and the potential for ‘burnout.’
This cannot be accomplished in a system where staff select their assignments.

Correctional operations staffing levels and
approach
This section of the report addresses the approach and methodology of staffing
restrictive housing units in the Bureau, which includes special housing units (SHU),
special management units (SMU), ADX, ADX general population and step-down units,
and special security unit. In assessing staffing in these units, the following
documents were reviewed to determine if staffing levels and approaches were
consistent with Bureau policy, statutes, and nationally recognized best practices, to
include conformance with American Correctional Association Standards.

Special housing unit staffing
The staffing of SHUs is consistent throughout the Bureau of Prisons facilities
assessed. With the exception of the female unit at USP Hazelton, each of the SHUs
was managed by the lieutenant that was designated as the SHU lieutenant,
responsible for management and oversight of the unit. In most facilities, particularly
the larger USPs, there was a SHU lieutenant on the day shift and a SHU lieutenant
assigned to swing shift. Typically, a lieutenant was present at the facility from 6 AM
to 10 PM. When the SHU lieutenant is not on duty, the operations lieutenant or shift
commander is responsible for managing SHU operations. The SHU lieutenant is
assigned by the warden to the post and may rotate out of the unit based on the
warden’s decision. The SHU lieutenants report to the facility captain and are part of

Work rosters are created in accordance with Article 18 of the collective bargaining
agreement called the Master Agreement. The Master Agreement provides that staff can bid on
certain posts by seniority, and preference requests are considered based on seniority.
Management is required to make reasonable efforts to honor the requests (i.e. not deny request
arbitrarily).

121

155

the Correctional Services Division of the facility, which consists of uniformed
correctional staff. In all units assessed, the facility captain reports to one of the
associate wardens, who reports directly to a warden.
The SHU lieutenant is responsible for supervising all staff assigned to the unit. The
lieutenant is responsible for all security related matters in the unit to include the
training of personnel; maintenance of inmate records; such as form BP-292, which is
a record of inmate activity; census counts; ensuring inmates receive services and
conditions of confinement are maintained at an acceptable level; movement
procedures; overseeing use of force incidents; management of emergencies; searches
of inmates; conducting supervision rounds; conducting segregation reviews as the
SRO; and other security functions.
The number of correctional officer positions assigned to the SHU varies from facility
to facility depending on the size of the unit. In all facilities assessed, the SHU #1 post
is designated as the officer in charge of floor operations. SHU #1 reports to the SHU
lieutenant or, in that person’s absence, to the operations lieutenant for the facility.
SHU #1 is normally the officer that carries the key to the inmate living areas, known
as “ranges”. SHU #1 oversees the activities and performance of other line officers
working in the unit, although he/she is not formally a supervisor. Specific duties may
include maintaining the equipment inventory; conducting census counts; maintaining
records, such as form BP-292, which tracks individual conditions of confinement;
ensuring that officer’s conduct and record 30 minute rounds, which consist of
wellness checks of inmates in their cells; ensuring meals are provided to inmates;
controlling all inmate movement, ensuring that the application of restraints meets
procedural requirements; conducting searches; and other security related duties.
Additional officers assigned to the unit, specifically SHU #2 through SHU #6, (the
number of officers assigned depends on the size of the unit), are range officers, who
perform most of their security duties on the ranges supervising inmates, controlling
their movements, conducting wellness checks, and providing them with meals and
services. In units with a secure control center, SHU #2 is the control room officer,
who manages access though electronically controlled doors and keeps unit records in
logbooks and a computerized record-keeping program.
In the larger USPs, a recreation officer is assigned to the day shift, and in some cases,
swing shift to supervise inmate recreation. Inmates are normally asked at the
beginning of the day shift they are interested in attending outdoor recreation. The
recreation is conducted in secure recreation cages located adjacent to the cellblocks.
The recreation officers are responsible for supervising the inmates while they
recreate. In the smaller units, the post is a day shift post, as the recreation can be
completed during the eight-hour tour of duty. In the larger units, recreation may
continue into second or swing shift. Our assessment also revealed the assignment of
a property officer in the USP SHUs. The property officer is responsible for

156

inventorying, distributing and maintaining inmate property, while the inmates are
housed in the SHUs.
SHU seniority bidding for post assignments. With the exception of the SHU lieutenant
all correctional officers working in the SHUs select their assignments on the basis of
a seniority bidding system. Already stated above, The Master Agreement between the
American Federation of Government Employees and the Bureau mandates that
officers are allowed to bid based on seniority for their shift and post assignments
each quarter of the year. This process ensures that the most senior bidder is awarded
the post. The only exception is that after one year the officer must bid to a different
shift, but not necessarily to a different assignment/post. There are some exceptions
to this rule, but as it relates to the SHU assignments, officers select their post and
shift assignment. As a result, seniority plays a prominent role in these assignments.
Program Statement 5500.14, Correctional Services Procedures Manual provides
further guidance as it relates to post assignments. Chapter 1, Section 101 requires
that wardens develop a quarterly assignment roster that is prepared at 13-week
intervals. This policy further requires that officers are subject to post rotation in
accordance with the contract.
SHU management and program staff. Staffing of the SHU consists of unit lieutenants
and correctional officers. There are typically no other staff assigned directly to work
in these units. Bureau facilities have an established practice of assigning staff from
the facility to make periodic visits to the unit to meet with inmates and provide
service. This includes case managers, medical and mental health staff, etc.
There is no separate unit team assigned and housed in the SHUs. However, the unit
team members continue to work with inmates that were residing in their units prior
to being placed in segregation/special housing. In practice, unit team members are
expected to make periodic rounds of the SHU (least once per day) to ensure that
members of a unit team make regular tours of the ranges. The presence of unit team
members is documented on the sign in sheet. Members of the assessment team
reviewed sign in sheets for a minimum of a three-month period at each of the sites
assessed. This review revealed that unit team members, including the unit manager,
made regular visits to the SHU in accordance with procedural requirements.
Similarly, other staff members are also required to conduct rounds and meet with
inmates providing service. A review of the sign in logs reveals that staff such as
chaplains, mental health clinicians, captains, associate wardens, the warden, and
other staff make periodic visits to the SHUs as well.

157

Special management unit staffing
The history and background of the inmates housed in the SMU dictates a need for
close supervision and a high level of security services.
The SMU programs have a bifurcated management structure. Correctional officers
working the unit are supervised by lieutenants that are specifically assigned to the
day shift or swing shift. When a lieutenant is not present in the unit, correctional
officers report to the operations lieutenant for the facility. All lieutenants report to
the facility captain and are members of the Correctional Services Division. The
facility captain reports to an associate warden at the facility. As is the case with SHU
staffing and chain of command, housing unit post SMU #1 is designated as the unit
officer in charge on each shift, and is considered second in command with regards to
security in the SMU.
Additional officers are assigned to each shift and provide correctional services,
security and supervision to the inmates in the SMUs. SMU officers have similar
functions and responsibilities as SHU officers and conduct census counts, maintain
records of inmate activity, conduct periodic wellness checks and rounds, conduct
searches, supervise inmate movement, and other security-related duties.
The SMU’s are also managed, in part, by the unit team. The unit team is comprised of
a unit manager, case managers, counselors, and a unit secretary. The unit team is
primarily involved in classification matters that determine the inmate’s SMU phase
status and subsequent placement. The unit manager also reports to an associate
warden. Typically, the unit manager and facility captain report to a different
associate warden. In this arrangement, it is expected that the unit manager and
captain work closely with one another on operations and procedural issues. As
evidence of this, Program Statement 5500.14, Correctional Services Procedures
Manual requires that the captain and unit manager jointly sign and approve
correctional officer post orders. This approach suggests that the two staff are
operating in a united fashion to avoid divisiveness in this bifurcated management
structure.
SMU seniority bidding for post assignments. SMU line staff bid each quarter for a post
assignment and are granted assignments on the basis of seniority. Therefore, a
rotation occurs four times per year. The most senior person bidding for a post is
awarded the post assignment. There is a four quarter limit on shift assignments to
the same shift. In the case of an officer with significant seniority, he can only bid a
particular shift four times in a row, and then must bid to a different shift for at least
one quarter, before returning to the previous shift. This rotation requirement
prevents an officer from bidding to the day shift more than four quarters in a row.
According to lieutenants interviewed, a staff member can be removed from special
management assignments for disciplinary reasons, although none of the lieutenants

158

could recall an instance where an officer was removed for disciplinary reasons. As
described above, shift rotation takes place on a quarterly basis affecting correctional
officers only and there is no selection process for line staff, only the shift bidding
process, where officers choose their assignments on the basis of seniority. The
lieutenant’s assignments, on the other hand, are selected by management and
seniority does not officially play a part in those assignments.
SMU program staffing. Staffing for two of the SMU programs is described in more
detail below. These two facilities were selected because they provide programming
for all four levels of the SMU, with USP Lewisburg housing mainly Level 1 and Level 2,
while USP Allenwood provides housing and programming for Level 3 and Level 4
inmates.
USP Lewisburg
USP Lewisburg is the largest SMU in the Bureau, whose program is designed to house
disruptive inmates in Level 1 and Two of the SMU program. Supervisory staffing
includes a captain and a subordinate deputy captain, as well as a number of
lieutenants assigned to manage the daily operations of the facility. Each shift has an
operations lieutenant that operates as the shift commander and an East and West
lieutenant who manage security operations in a sector of the facility. D and G units
house inmates who are disruptive, or who pose special management challenges. Each
of these units has a lieutenant present on the day shift. The east lieutenant is
responsible for B, C, and X units. The west lieutenant is responsible for E, F, J, and I
units. Officers are assigned to units based upon a quarterly seniority bid system as
discussed previously. Based upon a review of 30 days of rosters and an interview
with the captain, security staffing in the facility is sufficient and consistent with the
roster and master schedule.
In addition to the officers assigned to each unit there are also a number of
correctional program staff assigned. Three unit managers are assigned to the eleven
units in the facility (a fourth unit manager is assigned to the camp). In each unit
there is an assigned case manager and a counselor. Additional assigned program
staff provide reentry services, psychology services, religious services, and
educational services, and casework services to the inmate population. Staffing at this
facility is appropriate given the mission of the facility, and the size of the population
in the assessment team’s judgment.
USP Allenwood
USP Allenwood is a step-down facility from USP Lewisburg that houses those inmates
who have progressed to Level 3 and Level 4 of the SMU program. There are two SMU
housing units at USP Allenwood. Each of the two units have staff assigned and there
are two lieutenants that supervise the program five days per week . The lieutenants
are assigned to the Correctional Services Division of the facility. In addition, the unit

159

management team provide services to both housing units. Shift rosters were
examined for a 14-day period in the months of January and February 2014. SMU
staffing on each unit was typically four officers on day shift (not including a shared
property officer); three officers on afternoon shift; and one officer on the night shift.
On Saturday and Sunday, the day shift complement is often reduced to three officers,
as there is less activity to supervise on those days.
Of the 14 shift rosters examined, there were no instances when the staffing
complement was lower than described above. Occasionally, on third shift a second
officer is added to the SMU. This officer is also listed as a floater and can be utilized
in other areas of the facility based upon need.
There is a lieutenant assigned to unit 4A SMU and a lieutenant assigned to unit 4B
SMU on the day shift only. Supervision on afternoon and night shifts is provided by
the shift or operations lieutenant. One of the four officers on the day shift is
designated as the recreation officer, who manages and supervises inmate recreation.
A property officer is added to the complement and is shared between units 4A and
4B five days a week on the day shift.
A review of the rosters for the 14-day period reveals that staff are deployed
consistently and the units did not run short of personnel at any time. The lieutenants
queried on this issue also indicated that the posts are consistently manned per the
staffing plan. During the course of an interview with the union’s representative it was
confirmed that this was in fact the facility’s staffing practices.

Control unit staffing at Florence ADX
The Florence ADX is designated as a control unit within the Bureau. According to
Program Statement 5212.07, Control Unit Programs, placement in the control unit is
reserved for inmates who are unable to function in a less restrictive environment
without being a threat to others or to the orderly operation of the institution.
Inmates are referred to the control unit by facility wardens and approval for
placement is made by an assistant director within the Bureau, with input from a
regional director. Although inmates are frequently reviewed as described in other
sections of this report, length of stay in ADX Florence can be substantial.
The ADX is managed by a warden, who is based at the ADX. There is a correctional
complex warden, who is the warden of one of the three Florence facilities and is
responsible for all the facilities on the correctional complex. Three associate
wardens, based at the ADX, are responsible for correctional services, psychology
services, human resources and a number of other functions. ADX custody operations

160

fall under the associate warden for correctional services. A correctional captain
manages security for the unit and all custody personnel. This position is supported
by a deputy captain, a position that was vacant at the time of the assessment122.
Each tour of duty is managed by a shift commander, who is a lieutenant and who
oversees shift operations. Correctional officers working throughout the facility are
supervised by a lieutenant, who is either the shift supervisor or one of the area
lieutenants. There are area lieutenants assigned to the Activities position (two shifts,
16 hours, seven days), the SHU (two shifts, 16 hours, five days), and the special
security unit, and the control unit (two shifts, 16 hours, five days). There are
specialty lieutenant positions that include an administrative lieutenant, and other
five-day post lieutenants including EPO, Special Investigative Service (SIS), and SIA
lieutenants. All lieutenants report to the captain.
Unit teams provide services to the inmate population. There are three unit teams that
provide unit management services throughout the facility. These teams are
responsible for inmate classification, discipline, casework and to some extent
programming. Unit teams are assigned to geographic sectors of the facility and
service those inmates living within the sector cellblocks. The SHU within the ADX
does not have a separate unit team assigned to it, and the facility unit teams follow
their inmates and continue to monitor them while they are housed in SHU. The
assignment of unit teams to SHUs has been described above and there is no
difference as to how they are utilized at the ADX than was described for other SHUs.
Centralized management is practiced with regard to management of the SHU. The
lieutenant works with the operations lieutenant, who reports to the facility deputy
captain, who reports to the complex captain.
ADX seniority bidding for post assignments. As with all Bureau facilities assessed,
correctional officer personnel are assigned on the basis of seniority. Each quarter of
the calendar year, post bidding takes place and staff select their assignments, days
off and shift on the basis of seniority. This was confirmed with the administrative
lieutenant. Staff may continually bid posts, days off, and shifts they are interested in
each quarter. Staff must rotate to a different shift periodically. If they bid on sick and
annual relief positions, changing of the shifts as indicated above is not required.
Lieutenants are assigned by management to their positions. Seniority may be taken
into account for those assignments. Also, SIS personnel, tool room personnel, and
security officers are assigned by management to their posts.
Unit management staff are assigned by management to their positions. Unit
managers, unit counselors, and caseworkers are members of the unit teams. The unit
The Bureau reported that the position was filled in April, 2014 and assumed duties after the
completion of the project site visit.
122

161

teams manage classification and casework related functions, as well as providing
other services to the inmate population. The unit teams report to an associate
warden. In a parallel fashion, security staff report to their lieutenants and the facility
captain, who also reports to an associate warden.

Staffing observations
Program Statement 5500.14, Correctional Services Procedures Manual, dated October
19, 2012, outlines procedures relating to correctional services and staffing. Chapter 1
entitled management of correctional services describes methodology for computing
the correctional staff complement and the preparation requirements for the
quarterly and daily staff roster assignment. The policy further describes correctional
officer rotation requirements, referencing the Labor Contract Master Agreement. The
procedures include requirements relating to the preparation, use, and distribution of
post orders. Post orders are documents that describe in detail the requirements of
the job and the schedule of duties for a particular post. The procedures outlined a
format for post orders describing procedures and special instructions regarding a
particular post.
The Bureau has an organized and comprehensive process for determining staffing
levels based upon post assignment needs. This process is reviewed on a regular
basis. The application of post assignments is quite consistent in segregation units
throughout the agency. Post orders are comprehensive and detailed, outlining duties
and responsibilities of the officers assigned. Our analysis also determined that once
staffing levels are set utilizing a formula to compute the correctional complement
needed, there is little variation in staffing levels from what is prescribed in the daily
roster and master schedule. When there are vacancies, staff are normally reassigned
to work in the various restrictive housing units. If there are no staff available,
overtime expenditures are typically authorized. This analysis also is in agreement
with the Commission on Accreditation of the American Correctional Association that
routinely reviews Bureau facilities for accreditation. ACA has accredited all of the
Bureau facilities assessed. The accreditation process includes a review of staffing
methodology, consistent with staffing standards. Those staffing standards were
referenced earlier in this document.
When staff were questioned regarding removing a poor performing staff member
from the assignment, they believed that an individual could be removed for
disciplinary purposes, but there were no instances when staff could recall this taking
place.
Staffing of these SHUs was found to be consistent throughout the assessed facilities.
Post orders and job requirements are similar from one facility to another and focus
attention on uniformity and consistency in its facilities. In addition to the
consistency of staffing levels, the assessment team examined facility rosters

162

carefully to determine if staffing levels were maintained on a regular basis. In our
experience, due to budget difficulties in many jurisdictions, as well as correctional
officer vacancies, it is not unusual to see posts left vacant, from time to time, when
there are insufficient personnel available to man the posts. This review revealed that
there were few occasions identified where a post was left vacant in a SHU for any
reason.
FINDING: Each facility reviewed had sufficient staff to perform the functions of
managing the restrictive housing units.
FINDING: The presence of a correctional services team working alongside a unit
management team appears to be an effective management approach and provides
sufficient personnel to conduct the work as required by the Bureau.

Staff training, curriculum, and approach
This section addresses the approach, policy requirements and training provided to
staff working in segregation units in the Bureau of Prisons, which includes SHUs and
SMUs. There are a number of training mandates that the Bureau adheres to, and
nationally accepted practices that will be reviewed as part of this analysis. The
following documents were reviewed as part of the assessment of Bureau restrictive
housing unit staff training. In addition, nationally recognized best practices including
American Correctional Association Standards were also examined.
Program Statement 3906.20, Employee Development Manual, dated January
24, 2007.
Bureau of Prisons ART curriculum and lesson plans.
Bureau of Prisons quarterly SHU training curriculum and PowerPoint
presentations.
ACA Standard 4 – 4075 - The training plan is developed, and updated based
on an annual assessment that identifies current job related training needs.
ACA Standard 4 – 4084 – 1- Written policy, procedure, and practice provide
that all correctional officers receive at least 40 hours of annual training. This
training shall include at a minimum the following areas:
o

Standards of conduct/ethics

Security/safety/fire/medical/emergency procedures

163

Supervision of offenders including training on sexual abuse and
assault; and use of force

Additional topics shall be included based upon the needs assessment
of staff and institution requirements

ACA Standard 4-4090 - All security custody personnel are trained in
approved methods of self-defense and the use of force as a last resort to
control inmates.
ACA Standard 4-4092 - All personnel authorized to use chemical agents
receive thorough training in their use and in the treatment of individuals
exposed to a chemical agent.
ACA Standard 4-4220 - All institution personnel are trained in the
implementation of written emergency plans. Work stoppage and
riot/disturbance plans are communicated only to appropriate supervisory or
other personnel directly involved in the implementation of those plans.
ACA Standard 4-4373 - There is a written suicide prevention plan that is
approved by the health authority and reviewed by the facility or program
administrator. The plan includes staff and offender critical incident
debriefing that covers the management of suicidal incidents, suicide watch,
assaults, prolonged threats, and death of an offender or staff member. It
ensures a review of critical incidents by administration, security, and health
services. All staff with responsibility for offender supervision are trained on
an annual basis in the implementation of the program. Training should
include but not being limited to:
o

identifying the warning signs and symptoms of impending suicidal
behavior;

understanding the demographic and cultural parameters of suicidal
behavior, including incidence and variations in precipitating factors;

responding to suicidal and depressed offenders;

communication between correctional and healthcare personnel;

referral procedures;

housing observation and suicide watch level procedures; and

follow-up monitoring of offenders who make a suicide attempt.

164

Bureau of Prisons employee development manual
Program Statement 3906.20 establishes comprehensive training procedures, specify
training management responsibilities, and provide employees with access to training.
This policy and manual requires that the agency develop an annual training plan that
addresses the needs of the employees that is approved by a Training Committee with
employee representation. This requirement also conforms to ACA standards. The
training plan is the basis for the annual refresher training that is conducted each
year and is discussed below. New employees are required to attend a three-week
Introduction to Correctional Techniques program, which is considered basic training
for new correctional employees and is conducted at the training academy in Glynco,
GA. There is another component that is completed at the assigned institution.

Annual refresher training
All staff that work in Bureau facilities are required to participate in an annual 40hour in-service training program, commonly known as annual refresher training
(ART). ART training is conducted and provided during the first quarter of the
calendar year. The ART program is designed to comply with ACA standard 4-4084-1
relating to recommended annual training. The assessment team met with facility
administrators, human resource representatives, training staff, and supervisory staff
to determine the content and quality of the training programs provided. Each facility
develops a curriculum that contains Bureau required core curriculum and adds
content that is applicable to local training needs. The following curriculum is
presented as an example of the training content for a typical Bureau facility.
The curriculum offerings represent a comprehensive in-service training program that
is offered in Bureau facilities with some variation depending on the site. The CNA
team met with administrative, human resources, and training staff during the facility
assessments. At each site, a minimum of 15 randomly selected training records were
reviewed of staff assigned to the facility and a maximum of 25 records, mainly those
staff assigned to work in restrictive housing units, such as the SHU, SMU, and control
unit. The records revealed that ART training is a high priority in the Bureau and is
conducted in the first quarter of the calendar year. With few exceptions, staff records
reviewed revealed that staff completed the required training in 2013.
FINDING: The Bureau’s commitment to staff training is outstanding and consistent
with best practices in corrections.

165

Specialized training
The assessment team examined a number of high risk and high liability areas where
specialized training is offered and also recommended by the ACA to determine if
Bureau training covers these important topics. Training areas chosen for review
included the following:
Suicide prevention. Each of the facilities assessed provided a training program on
suicide prevention and recognition of the signs of suicidal behavior. Most often this
training was conducted as part of Annual Refresher Training, but also was a topic of
training in all of the SHU training programs offered quarterly in facilities that
operate SHUs. PS 5324.08 states that supplemental training will be conducted
approximately 6 months after the conclusion of institution ART. A review of selected
training records indicated that the vast majority of Bureau staff receive the annual
training and are familiarized (annually) in the recognition of the signs of suicidal
behavior.
Administration of oleo capsicum chemical agent. Oleo capsicum (OC) is a chemical
irritant dispensed from aerosol containers or via projectiles to disable an inmate who
is threatening or acting violent in a correctional institution. The administration of
any chemical agent, such as OC is considered a use of force. There are specific
manufacturer’s guidelines for the use of the chemical agent and staff need to be
trained regarding the proper method to deploy, the amount to deploy, the type of
reaction an inmate may have to the agent, and methods for decontamination after
the use of the chemical agent. Bureau policy requires that those individuals
authorized to administer chemical agents must be trained annually and also receive a
quarterly refresher training. Our findings were that annual OC training is conducted
at all facilities, and in a number of facilities, quarterly refresher training is mandated.
Use of force. Use of force training is provided to guide staff on under what
circumstances the use force may be used, levels of force allowed, and reporting
procedures following use of force incidents. Each of the facilities assessed provided
instruction on use of force as part of the Annual Refresher Training, as well as SHU
specific training.
Emergency plan training. Staff are regularly trained on an annual basis regarding
emergency plans and the management of emergencies. This topic is included in the
Annual Refresher Training program. Additional training in the Incident Command
System is also provided as part of the ART training. Management of emergencies is
also a training topic in a number of SHU training programs reviewed.
Self defense. This training is provided to enhance officer safety by instructing them
on self-protection against violent attacks. Self-defense training is a topic of
instruction in ART at all facilities subject to our assessment.

166

Special housing unit specific training
The Bureau requires that facilities that have SHUs provide a specialized curriculum
of instruction for staff that work in SHUs. The requirement calls for an eight-hour
program to be delivered on a quarterly basis for staff assigned to the units123. Beyond
the requirement that training must be provided, no guidance is provided with regard
to content or curriculum. Training content is determined locally by the facility
warden and his/her staff. The assessment team inquired about this training at each
facility in order to determine the content of training being provided, the length of the
program, the training delivery methods, who was required to complete the training,
as well as compliance with Bureau requirements and mandates. The following is a
summary of that analysis.
Training methodology/hours of instruction. Ten facilities were visited and assessed as
part of this process. Eight of the ten facilities have a training program for staff that
are assigned to the SHU that is an eight-hour program. USP Victorville, USP Hazelton,
USP Florence, USP Allenwood, USP Coleman, USP Terre Haute, USP Lewisburg, and the
Florence ADX all provide an eight-hour program. The delivery method and
attendance requirements do vary from facility to facility.
For example, at USP Victorville the warden has mandated that all staff at the facility
regardless of their assignment participate in the quarterly training. Records reviewed
indicated that over 300 staff have been provided with this training during the past
year. The program is an eight-hour course of instruction provided in a classroom
setting.
Similarly, an eight-hour classroom session is provided quarterly at USP Hazelton, USP
Florence, USP Terre Haute, USP Coleman, USP Lewisburg, and the ADX for staff
assigned to the SHU as a regular assignment or on a temporary basis. An alternative
method of delivering the required training is provided at USP Allenwood, where no
classroom instruction is provided, but the SHU lieutenant has created a training
binder with written content that staff assigned to the SHU must review each quarter
of the year. The lieutenant indicated he verifies that staff review the content by
questioning them on various aspects of the training and has the staff sign a training
log that indicates they have reviewed the material.

The Bureau reported that in November 2014, a standardized SHU Training lesson plan and
slide show were completed. SHU staff will be required to participate in this training four hours
per quarter. Additionally, a CENTRA training for captains regarding the presentation of this
training is being routed for approval.

123

167

At USP Hazelton, the quarterly SHU training is offered only three quarters of the
year. During the first quarter of the calendar year, because Annual Refresher
Training is taking place, there is no SHU training. The captain explained that at this
facility there is no requirement for SHU training during the first quarter, and ART
participation satisfies the SHU training requirement for that quarter.
The SHU training at USP Tucson and FCI Butner consists of a four-hour block of
training provided on a quarterly basis for SHU staff as it is the policy at those
facilities that SHU training consists of a four-hour program.
Level of staff participation. At USP Victorville, over 300 staff received the required
training in the past year. At a number of the facilities, records indicated that the
majority of staff assigned to work in the SHU received the SHU training. There were
instances where staff who were assigned to SHU per the roster did not receive the
training when it was offered during the quarter. This was especially true for those
staff assigned as sick and annual leave replacements for staff who are regularly
assigned SHU employees. The only facility where 100 percent of the staff assigned to
the SHU participated in the training was at USP Allenwood, where the training was
not conducted in a classroom setting.
At USP Victorville, the strategy of training all staff ensures that staff assigned to
work in the SHU either on a permanent or temporary assignment are trained. In each
of the other facilities, there were staff assigned to work in the SHU at various times
that did not complete the training program. The frequency of the training (quarterly)
and the significant number of staff who may work in the SHU one or two days a
week, or in some cases, less frequently makes it challenging to achieve 100 percent
compliance with the training requirement. This issue is further exacerbated by the
fact that staff bid their posts on the basis of seniority quarterly, which can result in
high turnover in the SHUs every three months, creating a new cohort of personnel
that need to be trained four times per year.

Special housing unit training curriculum
Based upon staff interviews, there is no standardized curriculum for the SHU
training. It was determined that the SHU training content varies from facility to
facility. However, there appear to be similarities in the training content provided with
respect to inmate management and security procedures. Examples of the training
content found in the various training programs are noted below.
USP Hazelton: Hunger Strike Management; SHU Strategies; Prison Rape
Elimination Act; and Suicide Prevention.
USP Allenwood: Use of Batons; Searches; Supervision of ; Cell rotation (21
days); Meal Service; Completing 292 Forms; Inmate Movement; Levels of

168

Supervision; Food Slot Operations; Food Trays; Recreation; Privileges; Cell
Assignments; and Use of Force.
FCI Butner: Application of Restraints; Suicide Risk Assessment and
Prevention; Conducting Rounds; Searches; and Emergency Response.
USP Coleman: Application of Restraints; Use of Batons; Inmate Discipline;
Sanitation; Searches; Razor Procedure; Key Control; Cell Rotation(21 days);
Completing 292 Forms; Food Service; Suicide Prevention; Inmate Supervision;
Recreation; Privileges; Escorts; Programs; and Unit Evacuation.
USP Terre Haute: Use of Force; Suicide Prevention; Prison Rape Elimination
Act; SHU Operations (includes Escorts, Law Library, Haircuts, Recreation,
Safety and Security).
ADX: Use of Force; Interpersonal Communications; Conducting Rounds: Meal
Service; Searches; Psychology/Suicide Prevention; Admissions and Releases;
Escorts; and Searches.
In the six examples provided, close attention is paid to unit operating procedures,
proper record-keeping (BP-292 Forms, which note if inmates accepted meals,
participated in recreation etc.), inmate escort and supervision practices, use of force,
suicide prevention and recognition of signs of suicidal behavior, privileges and
security procedures.

Staff training observations
The Bureau is heavily invested in both in-service training and preparing staff to carry
out their duties efficiently and effectively. On an annual basis, a training plan is
developed consistent with the requirements of Employee Development Manual, which
is also in accordance with ACA standards previously referenced. From the training
plan, a national curriculum is developed for the coming year’s Annual Refresher
Training. The training curriculum includes all topics recommended by the ACA,
including those high-risk areas that are addressed in the standards, where training is
recommended. The Bureau also adds content to address identified training needs.
During the first quarter of each calendar year, staff are scheduled to participate in
ART training at each site and, as noted, the vast majority of personnel complete the
training during the first quarter of the calendar year. This includes all staff assigned
to the facility, such as correctional officers, supervisors, unit management,
professional and support staff. The amount of training provided and the number of
staff receiving ART training is impressive and, as previously stated, is a best practice.
Many other correctional agencies are unable to provide this level of training, either
for budgetary reasons or logistical reasons that can not be overcome. The Bureau is

169

seems to be deeply committed to the annual training and makes it a high priority
mandate that all facility leaders must and do adhere to.
FINDING: The training curriculum used by Bureau facilities is consistent with best
practices, providing a range of topics that meet industry standards and ACA
standards.
High liability training areas, such as the use of force (to include OC training),
providing first aid/medical attention, security procedures, emergency procedures,
sexual misconduct, inmate mental health issues and suicide prevention are training
topics addressed.
The SHU training is also provided at each of the sites the assessment team reviewed.
There are inconsistencies on how the SHU training is conducted, as some sites train
four hours per quarter while others eight hours per quarter. The majority of the
facilities provide classroom instruction and one facility requires staff to read a
binder that contains training content outside of the classroom. Inconsistency was
also found with the mandatory nature of the training, as the majority of the facilities
were unable to train 100 percent of those staff assigned to work in the SHUs. USP
Victorville, however, exceeded Bureau standards by training all staff, regardless of
whether they were assigned to a SHU post or not.
It is difficult to train all staff members that work in the unit on a quarterly basis,
especially those who are temporarily assigned to fill a post to backfill for a regular
SHU officer that is absent on a particular day. The quarterly rotation of staff
resulting in turnover also contributes to the problem of ensuring SHU staff receive
the training.
FINDING: SHU training is not consistent throughout the Bureau in terms of
delivery, content, hours of instruction, schedule, and mandatory attendance.

Security systems and practices
This section of the report addresses Bureau security systems and practices as
described in policy and procedures, as well as observed during the assessment
process at facilities reviewed. Team observations contributed to the summary and
findings, as well as numerous documents that were also reviewed during the
assessment process. The report does not document any practices or procedures that
were observed by the project team that if reported would compromise the safety and
security of the institutions. The following documents were reviewed during the
assessment process and contributed to this section of the report.
Statutory requirements

170

Bureau program statements relating to control units #5212.07, dated
2/20/2001; special housing units #5290.10, dated 7/29/2011; special
management units #P5217.01, dated 11/19/2008, and use of force and
application of restraints #P5566.06, dated 11/30/2005
Correctional officer post orders124
Bureau performance review reports
American Correctional Association standards125
Bureau Operations Memorandum 004-2013 (5500) Oleoresin Capsicum
(OC) Aerosol Spray Pilot Program
Bureau Memorandum, Conducting 30 minute Checks, authored by
Assistant Director for the Correctional Programs Division, April 26, 2012
On-site observations of operational practices
Staff interviews
Inmate interviews

Physical plant characteristics
The physical plants of the Bureau segregated units are typically cellblock style units
with a controlled entry point from a sally port126 leading from a main corridor into
Post Orders are detailed descriptions of the duties and responsibilities of a correctional
officer. They include the schedule of daily events and activities, as well as procedural
requirements that an officer must comply with. Post Orders often include excerpts from
policies and procedures that an officer must be familiar with. It is required that an officer
review their post orders on a periodic basis, usually when they assume a new post, and sign
and date the post order signature sheet.

124

The American Correctional Association (ACA) is a nationally recognized private, nonprofit
organization whose mission includes the development and promotion of effective standards
for the care, custody, training, and treatment of offenders. As part of its accreditation process,
a visiting committee of ACA auditors audits the correctional facility against standards and
expected practices documentation and evaluates the quality of life or conditions of
confinement. An acceptable quality of life rating is necessary for a facility to be eligible for
accreditation. The quality of life in a facility includes cell size and time inmates spend outside
the cells, adequacy of medical services, offender programs, recreation, food service, sanitation,
use of segregation, crowding, and reported and/or documentation of incidents of violence.

125

126
A sally port is a secure, controlled entryway in a correctional facility, which normally
includes the use of two doors to control access to an area. Sally port doors are not opened at

171

the unit. The majority of the units viewed are linear style, with cells located on either
side of the cellblock corridor. The cellblock area is protected by a “range” gate that
leads from the cellblock common area, usually near the control room. Access
through the range gate is carefully controlled by an officer, who controls entry into
the cellblock area. At the opposite end of the corridor, there is typically an
emergency exit as a secondary means of egress for fire evacuation purposes. All cells
viewed were of solid steel construction with a tamper resistant security glazing
vision panel to allow visibility inside the cell.
All cells in SHU and SMU housing are occupied by two inmates, except security cells
used to monitor inmates that require close monitoring due to security or mental
health issues. All the cells at the ADX have one bunk and are occupied by one inmate.
Each cell door is also equipped with a small panel that can be unlocked by the
correctional officer to deliver food (also known as the “food slot”) that serves as a
protected method of handcuffing inmates before the cell door is opened. The cells
themselves are equipped with sleeping bunks, which are secured 12 inches off the
floor, one on top of another; a toilet/sink combination unit; one or two shelves; a
writing surface; one or two seats, property storage units; and, in many cases with the
exception of the older facilities, a shower unit inside the cell. In the older units,
showers are located at one end of the range and inmates must be escorted from their
cells to the showers, one at a time. Most of the cellblocks viewed have cameras
installed at either end of the range to record inmate and staff movement, with
footage digitally recorded.
Each unit has at least one cell that is of a modified design to house disruptive
inmates and/or suicidal inmates that need to be controlled more closely. These cells
often have a secure bunk constructed of concrete or steel and have the capability of
securing inmates to the bunk with restraints if they are violent or a significant threat
to security. These cells were positioned in the unit in such a way to allow for close
staff observation of the inmate. Suicide Watch cells are not ordinarily contained in
the physical plan of the restrictive housing unit. Rather, per policy they are generally
located in the Health Services area or another similarly private location. A policy
waiver is required in order for a Suicide Watch cell to be placed in SHU.
Inmate recreation is conducted in secure recreation areas constructed of chain-link
or expanded metal. Inmates recreate either one at a time or in small groups. Entry
into the recreation areas is through a locked gate, equipped with a handcuff port to
allow an officer to secure the inmate in handcuffs before opening the gate.

the same time so the control officer can control entry and prevent someone from breeching the
inner door.

172

The common area is located between the unit entrance door and the cellblock area. It
typically consists of a control room or officer station, holding cells, interview rooms,
staff offices, multipurpose rooms, storage areas and a room where inmates can
access computerized law library materials. The multipurpose rooms are used for
interviews, barbering services, and programming, where appropriate.

Patrol requirements
In April 2012, the assistant director for the Correctional Programs Division issued a
memorandum, which requires correctional staff that work in SHUs to observe all
inmates confined in continuous lockdown status, such as administrative detention or
disciplinary segregation every 30 minutes on an irregular schedule, but with rounds
that are not to take place less than 40 minutes apart. These observations are
documented in a logbook maintained for that purpose.
The purpose of these observations is to ensure that inmate activity is carefully
supervised and the health/mental health of each inmate is monitored. This practice
was observed at all facilities assessed. The Post Orders of SHU staff included a
description of this requirement and it is a mandated practice throughout the
assessed facilities. The 30-minute rounds are documented by the officer conducting
the rounds and the logbook containing this information is reviewed and signed off
by a supervisor at the end of the shift.
The April 2012 memorandum and the practice of observing inmates in SHU every 30
minutes complies with the ACA standard. The standard also provides guidance in the
management of violent, mentally disordered inmates, or suicidal inmates. Bureau
officials at each site assessed have an established practice of removing suicidal
inmates and placing them in a cell specially designed to provide for direct
observation of inmates that are at risk. Bureau either assigns an inmate companion127
or staff member to directly observe the inmate in crisis and keep an ongoing log of
the inmate’s activity and movements.
Bureau policy128 and post orders also require that a lieutenant, typically the SHU/SMU
lieutenant if on duty, must visit the unit and conduct a tour of the cellblock area
once per shift. This practice ensures that supervisory staff are conducting inmate
wellness checks and verifying that line staff are following approved procedures.
Inmate companions are trained to provide direct observation of inmates housed in suicide
resistant cells. The companions receive special training from the Bureau staff and also keep a
log of the inmate's activity while being observed on suicide watch status.

127

Program Statement 5290.10, Special Housing Units, dated August 1, 2011, Section 12 (m) Staff Monitoring.
128

173

Patrol requirements observations. The assessment team found a meticulous level of
compliance with the monitoring and supervisor tour requirement as 30 minute logs
are typically up-to-date and entries are initialed by the recording officer and
subsequently reviewed by a supervising officer, who also signs the document.
Lieutenant rounds also take place each shift, as required. We would estimate a better
than 95 percent compliance rate with these log entries, and found few cases where
staff failed to make an entry or write their initials. There certainly were cases where
noncompliance was detected, but these were anomalies. The log entry sheets are
archived and maintained for future reference.
FINDING: The observation of inmates in special housing and those that are being
monitored as having suicidal tendencies in specially designed cells under direct
supervision is consistent with best practices and in compliance with ACA
standards.

Inmate movement practices
All inmates residing in SHUs, regardless of the reason for that placement, are
transported in restraints whenever they leave their cells. Post orders require that an
inmate be handcuffed from behind through the food slot before the cell door is
opened to remove the inmate. In cases where there are two inmates residing in a cell,
both inmates must be handcuffed before the cell door is opened. In these cases, two
officers are present when the cell door is opened. The inmate may be transported by
a single officer, once the cell door is closed and the second inmate is secured. Special
circumstances may require a second officer be present at all times during an escort.
For example, the handcuffing procedure at USP Lewisburg requires that the inmate
be restrained from behind utilizing hand restraints, with palms up and thumbs out.
The handcuffs are double locked, which helps prevent the inmate from tampering
with the handcuffs. Inmates are also scanned with a hand-held metal detector and
pat searched prior to any movement taking place. The double locking of handcuffs is
not a universal practice throughout the Bureau for internal inmate movement, except
when the inmate is being transported outside of the unit. A device, known as the
“black box” (a cover for the handcuff key slot that prevents an inmate from
tampering with the key slot) is often used for transportation outside of the facility,
or if the inmate poses a substantial risk to security. Internal movement of inmates
mostly takes place when the inmate is being moved to recreation, unit showers, law
library, or for staff interviews.
Movement within the unit is recorded when the inmate is transported to the law
library, recreation, and to receive a haircut. Movement out of cell for interviews for
example, is not recorded. Movement outside of the unit is noted in the unit log.

174

Special movement circumstances. In cases where staff safety dictates that inmates
need more supervision than described above, additional precautions can be taken.
These include the following:
Two-man hold: The two-man hold is one method utilized by the Bureau.
The two-man hold requires that two staff be present the entire time the
inmate is out of his cell or secure area. This requires one staff member will
remain in direct physical contact with the inmate at all times and both
staff members are directly responsible for the control of the inmate.
Three-man hold: Again, in cases where staff safety is a concern, the
supervising officer may require that three staff be present the entire time
the inmate is out of his cell or secure area. In a three-man hold situation,
two staff members remain in direct physical contact with the inmate at all
times, and all three staff are directly responsible for the control of the
inmate.
Lieutenant hold: The lieutenant hold technique calls for a lieutenant to be
present the entire time the inmate is out of the cell and the lieutenant is
required to provide supervisory oversight during the inmate movement.
Other hold: In circumstances where an additional security is needed, staff
may be required to wear protective gear and additional personnel may be
dispatched to supervise inmate movement.
Program Statement P5566.06 Use of Force and Application of Restraints, dated
November 30, 2005 also allows for “…staff to place an inmate temporarily in
restraints to prevent an inmate from hurting self, staff, or others, and/or to prevent
serious property damage. When the temporary application of restraints is
determined necessary, and after staff have gained control of the inmate, the warden
or designee is to be notified immediately on whether use of restraints should
continue.” The policy states that restraints should only be used when other effective
means of control had failed, or are impractical, and that restraints can remain on the
inmate until self-control is regained.
Special management and control unit movement practices. Inmates residing in SMU
programs at Level 1 and Level 2 are subject to restrictive movement as described
above. As inmates are moved to Level 3 and Level 4, they earn additional privileges
and are allowed to participate in small group congregate activity unrestrained.
However, movement outside of the unit to medical and other locations, are
conducted while the inmate is in restraints.
Inmates housed in the ADX are subject to restraint conditions consistent with the
conditions noted above for SHU inmates, except two officer escorts are the norm.
Movement outside the unit requires utilization of the Martin chain, handcuffs

175

attached to the chain, with a black box covering handcuffs. Additional protective
equipment is provided at the ADX for officer safety and to better control the inmate
population, such as rapid rotation batons in defensive position.
Progressive and ambulatory restraints. Program Statement 5566.06, Use of Force and
Application of Restraints allows for the placement of inmates in restrictive and secure
restraints to control behavior and protect staff and property. Use of ambulatory
restraints is authorized when the inmate is acting aggressively and/or lacks selfcontrol. This is allowed until the inmate establishes a pattern of self-control and is
no longer acting out. The policy refers to this as a “pattern of nondisruptive behavior
over a period of time.”
The placement of inmates in four-point restraints is authorized per policy when
placement is “the only means available to obtain and maintain control over an
inmate.”
Each of these procedures requires regular observation of the inmate, examinations by
medical personnel, regular reporting of activity, and review as to whether the
placement in restraints is still necessary. The assessment team did not have an
opportunity to review this process in great detail; however, our use of force analysis
did include cases where inmates were placed in ambulatory and/or four-point
restraints.
ACA Standard 4-4190 regarding the use of restraints reads: “Written policy,
procedure, and practice provide that instruments of restraint, such as handcuffs,
irons, and straitjackets, are never applied as punishment and are applied only with
the approval of the warden/superintendent or designee.”
ACA Standard 4-4191 regarding restraints reads: “Written policy, procedure, and
practice provide that the unit is placed in a four/five point restraint (arms, head and
legs secured), advanced approval must be obtained from the warden/superintendent
or designee. Subsequently, the health authority or designee must be notified to
assess the inmate’s medical and mental health condition, and to advise whether, on
the basis of serious danger to self or others, the inmate should be placed in a
medical/mental health unit for emergency and involuntary treatment with sedation
and/or other medical management, as appropriate. If the offender is not transferred
to a medical/mental health unit and is restrained in a four/five point position, the
following minimum procedures will be followed:
Direct visual observation by staff must be continuous prior to obtaining
approval from the health authority or designee;
Subsequent visual observation must be made at least every 15 minutes;
and,

176

Restraint procedures are in accordance with guidelines endorsed by the
designated health authority.”
Inmate movement practices/observations. Assessment of off-site transportation was
beyond the scope of this review. However, the team did examine internal inmate
movement, the use of restraints to safely facilitate inmate movement, and post
orders that guide and direct officers in the performance of their duties. In visiting
the selected facilities, there was only one instance observed where a staff member
was observed opening a cell door without having two officers present, in apparent
violation of operational procedures. This was at the female facility at USP Hazelton
and staffing levels in that SHU may have contributed to this violation, as there were
few officers available to conduct the movements. Otherwise, our observations found
that Bureau staff and supervisors carry out their duties consistent with the post
orders and in compliance with movement procedures as described above. Across the
board, post orders are well drafted documents that include comprehensive content
that act as a guide to ensure staff perform in accordance with procedures.
Procedures and policy for the use of restraints describe the use of restraints as a
method to control inmates and not to administer punishment. Bureau policy is in
compliance with ACA standard 4-4190, which states restraints shall not be used as
punishment. In the Use of Force section of this report, inmates interviewed at one
facility claimed that the application of ambulatory restraints is managed improperly
as the restraints are at times applied tightly, and cause minor injury to wrists. This
assessment was unable to verify or confirm this alleged practice, but it should be
reviewed to determine if there is any validity to the claim.
A review of Program Statement P5566.06, Use of Force and Application of Restraints
describes procedures for the use of four-point restraints. This policy is in compliance
with ACA standard 4-4191. However, the assessment team did not examine in detail
situations where four-point restraints were utilized as part of this study as no cases
where four-point restraints were applied were taking place during the site visits. The
policy does require that inmates be checked every 15 minutes and that qualified
health personnel evaluate the inmate and ensure that the placement is appropriate.
Subsequently, a lieutenant is charged with reviewing the placement in four-point
restraints to determine if a period of calming effect has occurred, which will allow
staff to remove the inmate from four-point restraints. Extensive reporting
requirements are also outlined in the policy.

177

Impact of gang issues and separation
practices
One of the most difficult security issues facing the Bureau staff is the management
of gang activity of the various gangs in the system. As is the case in many state
correctional facilities, this issue plagues the general population of the Bureau’s USPs,
which carries over to its segregated units. This assessment revealed that it is
common practice for inmates in the USPs to associate themselves with a particular
gang or what is termed “cars.” A car is a group of inmates who develop a close
association as protection against other established inmate groups, such as gangs.
The prevalence of inmate grouping and gangs places pressure on administrators to
carefully place inmates in housing units throughout the agency, as placement of a
gang member in a cellblock housing a high concentration of rival gang members can
result in violence.
In the case of the SHUs, administrators and staff exercise caution when making bed
placements, as certain gang members cannot be housed with one another or violence
will likely ensue. Given that all of the SHUs are double celled, an evaluation of each
inmate’s record and associations has to be conducted by SHU and unit management
staff prior to cell assignment.
The SHU situation is complicated by the fact that a high percentage of inmates
residing in SHUs have requested protection from other inmates, often due to gang
related issues. Many inmates in the system have what is termed as “separatees.” A
separatee is an inmate, who due to safety and security concerns cannot be placed in
housing units with other specific individuals. As noted in chapter 3, the number of
inmates that have separatee issues is significant and impacts inmate management in
the agency. For example, the warden at USP Lewisburg advised the assessment team
that of 748 inmates at that facility, 334 had separatee issues.129 Some inmates had as
many as 14 separatees. Keeping in mind that these inmates cannot congregate with
one another, bed management in the general population and in restrictive housing
units is very complicated.
In the SMU at USP Allenwood, where inmates in Level 3Three and 4 Level 4 are
allowed to congregate, the lieutenants and unit management staff have to carefully
place inmates according to their gang or group designation. For example, on one unit
of the Allenwood SMU, one range can only house Caucasian inmates, Southern
California Mexican Mafia, or African-American inmates. On another range, Southern

129

Interview briefing with Warden on January 21, 2014.

178

California Mexican Mafia, Caucasian protective custody and African-American
inmates are housed. On yet another range, Texas Mexican Mafia, inmates from the
District of Columbia, and nonaffiliated African-American inmates are housed; and on
Range 4 Latin King, NETA, Northeast region inmates, and Caucasian Cars are housed.
Staff indicated that the mixing of rival groups in the same housing unit can affect the
security of the unit. This unit allows congregate group activity and only one range at
a time is allowed out for recreation. The mixing of ranges is such a concern that the
keys to the nonrecreating range are kept outside the unit to ensure that the
nonrecreating range doesn’t mix with the range that is out on recreation.130
The issue of gang separation and taking a cautious approach with inmate housing to
prevent violence, is less of a concern at the ADX because most of the recreation is in
individualized areas and there is no double bunking of inmates. It does, however,
become a concern when congregate activity takes place in the step-down unit, where
limited congregate activity is allowed.
Special Investigative Service. The Bureau assigns personnel at each USP and FCI the
team visited to investigate criminal activity and gather intelligence to aid in the safe
operation of the correctional facilities. The SIS investigates separatee issues and
validates inmate protective custody requests. One of their most critical
responsibilities is gathering gang intelligence to assist facility supervisors in
managing the inmate population. SIS personnel also investigate and validate inmate
gang membership and keep ongoing records of gang membership, as well as
determining who the gang leaders are. The SIS performs a valuable security function
to aid in the management of the inmate population that has a heavy concentration of
gang and car membership, all of which complicates inmate management and
placement practices.
The gang issue in the USPs that was assessed impacts the management of the SHU as
it creates separatee issues and likely increases the number of inmates requesting
protective custody. This, in turn, results in these inmates being housed in restrictive
housing. This was confirmed through interviews with staff and inmates who
indicated that many of the segregated inmates were unable to live in general
population because of the influence, harassment, intimidation, and threats of
violence from members of gangs and cars in the facilities.
FINDING: Difficulties in the management of the gang problem exacerbates the
protective custody problem, thus causing a high incidence of inmates leaving
general population by their own request and being placed in segregation units,
after being threatened in general population.
130

Interview with Lieutenants on March 4, 2014.

179

Post orders
Post Orders are detailed descriptions of the duties and responsibilities of a
correctional officer. They include the schedule of daily events and activities, as well
as procedural requirements that an officer must adhere to. Post Orders often include
excerpts from policies and procedures that an officer must be familiar with to assist
them in performing their duties. It is required that an officer review their post orders
on a periodic basis, usually when they assume a new post, and sign and date the post
order signature sheet. ACA Standard 4-4178 requires that there be written orders for
every correctional officer post and that the orders should be reviewed annually and
updated if necessary. ACA standard 4-4179 requires that policy, procedure, and
practice provide for personnel to read the appropriate post order each time they
assume a new post and sign and date the post order.
Post orders were provided to the assessment team at each site for all posts in SHU,
SMU, and at the various programs at the ADX. The post orders, as a general rule,
include numerous procedures and instructions on important topics, which assist
correctional officers in performing their jobs.
The content of post orders, which are relatively consistent from one facility to
another, reflects that a high priority is placed on utilizing post orders to provide
procedural and job performance guidance to correctional staff. The post orders
reviewed address the majority of the areas of security procedure that a correctional
officer needs to be familiar with in the performance of their duties.
Compliance with ACA standards is noted and each of the facilities visited had passed
an accreditation audit within the past three years, and the above-referenced Post
Order ACA standards were also complied with.
FINDING: Bureau post orders are extremely comprehensive documents that meet
national standard requirements and are considered a best practice.
The Bureau utilizes post orders as working documents to guide the job performance
of correctional staff.

Use of force and critical incidents
The focus of this section of the report addresses the project team’s findings
regarding the use of force and critical incidents within the Bureau’s restrictive
housing units. An assessment of the Bureau policies, operational practices and
compliance levels were examined throughout the review process.

180

According to federal regulations and associated program statements, the Bureau of
Prisons can authorize staff to use force only as a last alternative after all other
reasonable efforts to resolve a situation have failed. When authorized, staff must use
only that amount of force necessary to gain control of the inmate, to protect and
ensure the safety of inmates, staff, and others, to prevent serious property damage
and to ensure institution security and good order. Based on our review there was no
supported evidence presented that policy or routine staff practice included operating
outside the federal regulations.
Program statements and facility supplemental statements (referred to as Institutional
Supplements) related to the use of force in restrictive housing have all been
developed in line with the requirements identified in the federal regulations. In
addition to the regulatory requirements the Bureau program statements and facility
supplemental statements also identify policies, procedures and objectives regarding
when the use of force can be authorized, the type of force to be used, required
documentation, the review process and staff training.
For the most part, the Bureau’s program statements serve as core documents
personnel use to reference the appropriate procedures to follow when applying
restraints, when being involved in a critical incident, or when involved in a use of
force incident. At each facility visited, the project team reviewed multiple Use of
Force incident reports, associated documents, inmate files, staff training files and
videos of incidents that occurred. At several of the facilities team members were also
present to observe after action reviews conducted by the facility management team.
The After Action Review team by Bureau policy consists of the warden, associate
warden, captain and health services administrator or designee. The team is
responsible in part for reviewing all use of force incidents at the facility.
FINDING: The Bureau has established a comprehensive and extensive program
statement that clearly identifies the step-by-step requirements associated with
managing potential use of force and critical incidents consistent with nationally
recognized standards.
The Bureau’s Program Statement 5566.06 requires that a report be prepared by staff
on all incidents involving the use of force, chemical agents, progressive restraints,
and nonlethal weaponry. The report must describe the incident, establish the identity
of all individuals involved and include appropriate mental health/medical reports.
These reports must be submitted to the warden or his/her designee no later than the
end of the tour of duty. The program statement further requires that use of force
incidents be videotaped if time and circumstances permit. All calculated use of force
incidents should be videotaped.
All the required documentation that has been prepared is reviewed for completeness
and maintained by the facility captain or deputy captain and becomes available for
subsequent reviews including the required after action review. By policy, the after

181

action review team consists of the warden, associate warden, captain and health
services administrator or designee. The warden must provide the appropriate
documentation to the regional director within two days after the inmate has been
released from restraints.
Based upon a review of documentation and information provided, the assessment
team found that generally appropriate procedures consistent with policy are
routinely being followed. In cases where deficiencies are noted, corrective action is
applied. In the project team’s review of documentation and video footage, there were
deficiencies noted. Some of the deficiencies included: equipment failure, (video
camera battery goes out), lack of proper staff identification, inappropriate
application of restraints, and submittal of incomplete documentation. These were all
considered deficiencies that required corrective action – however, these instances
were considered the exception to practice rather than the norm. In each case,
corrective action steps were initiated by management personnel. An overall review of
these reports did not reveal any consistent issues with meeting the reporting
requirements.

Types of force
The Bureau’s applicable Program Statement (P5566.06) and associated federal
regulation (28 CFR 552.21) identify the primary types of force: Immediate Use of
Force; Calculated Use of Force and/or Application of Restraints; Use of Force Team
Technique. The following descriptions as provided in the federal regulations and
program statement identify when each type of force may be considered appropriate.
Immediate use of force. Staff may immediately use force and/or apply restraints
when the behavior constitutes an immediate, serious threat to the inmate, staff,
others, property, or to institution security and good order.
Calculated use of force and/or application of restraint. This occurs in situations
where an inmate is in an area that can be isolated (e.g., a locked cell, a range,
recreation yard) and where there is no immediate, direct threat to the inmate or
others. When there is time for the calculated use of force or application of restraints,
staff must first determine if the situation can be resolved without resorting to force.
Use of force team technique. If use of force is determined to be necessary, and other
means of gaining control of an inmate are deemed inappropriate or ineffective, then
the Use of Force Team Technique shall be used to control the inmate and to apply
soft restraints, to include ambulatory leg restraints. The Use of Force Team
Technique ordinarily involves trained staff, clothed in protective gear, who enter the
inmate’s area in tandem, each with a coordinated responsibility for helping achieve
immediate control of the inmate.

182

Any exception to this rule is prohibited, except where the facts and circumstances
known to the staff member would warrant a person using sound correctional
judgment to reasonably believe other action is necessary (as a last resort) to prevent
serious physical injury, or serious property damage which would immediately
endanger the safety of staff, inmates, or others.
A review of the use of force incidents revealed that the majority of incidents involved
the application of physical restraints (use of hand restraints). The policy on the
application of restraints does not restrict the use of restraints in situations requiring
precautionary restraints, particularly in the movement or transfer of inmates (e.g.,
the use of handcuffs in moving inmates to and from a cell in detention, escorting an
inmate to an SHU, pending investigation). Staff routinely place inmates in restraints
when moving the inmate to and from a cell/recreation/shower in most segregation
units. This is a recognized practice that is consistent with national standards, with
the exception of applying restraints for verified protective custody inmates. Most
agencies have separate internal movement procedures for inmates verified as
requiring protection which does not require the application of physical restraints
each time the inmate is moved out of their cell. Inmates assigned to the SMU
program in Level 3 and 4 or for inmates in the Transitional and Pre-Transfer steps of
the ADX–general population and Step-Down Program are routinely moved outside the
cell however within the unit without restraints.
In addition to the established routine movement procedures, personnel are also
authorized by policy to apply physical restraints necessary to gain control of an
inmate who appears to be dangerous for one of the following reasons:
Assaulted another individual.
Destroyed government property.
Attempted suicide.
Inflicted injury upon self.
Becomes violent or displays signs of imminent violence.
This is consistent with national standards. In reviewing incident reports the use of
ambulatory restraints was normally the result of staff attempting to control
escalating inmate disruptive behavior or to prevent inmate self-injury. It was noted
that a number of force incidents applied by staff were related to either resolving
conflicts between cell mates, inmates refusing to submit to being restrained before
movement or during a recreation incident. Overall, based on observation, video and
document review, staff appeared to routinely follow established policy.
Use of chemical agents. In some circumstances, the use of chemical agents may be
required. By policy the use of chemical agents is used only after approval and a

183

review of the inmate’s medical file, unless such a delay would endanger the safety of
the inmate, other inmates, staff and the community or result in severe property
damage or escape. Only staff that have been trained in confrontation avoidance, use
of force team technique and use of chemical agents are authorized to apply chemical
agents.
A review of incident reports revealed that the use of force incidents resulting in
chemical agents being used primarily fell into one of three categories:
1.

To gain control of inmates fighting or participating in assaults on the
recreation yard. As many as six inmates can be placed in one recreation yard
depending on the size of the yard. A pre-screening is normally conducted by
staff prior to placement on a specific yard however the recreation period
appeared to be a time when many of the reported assaults/fighting took
place. A review of videotapes of incidents related to recreation revealed staff
normally responded in a timely manner, ordered inmates to cease, followed
protocol and if necessary administered chemical agent from hand-held
canisters and/or a pepper ball launcher to gain control over the situation.

A second category in which the use of force and the application of chemical
agents appeared to occur was when two cellmates were fighting in the cell
and ceased to stop after several direct orders. A pre-screening is conducted
prior to housing two inmates together, however staff report that occasionally
fights between cell mates occur and they fail to stop after several direct
orders.

A third type of use of force case that often involves the use of chemical
agents occurs when an inmate is unwilling to have restraints applied in order
to be removed from the cell/shower/recreation yard, and staff must respond
to gain control over the inmate(s). If conflict avoidance techniques do not
work, then a five-man team is assembled wearing protective gear, the inmate
is provided repeated direct orders to comply with the application of
restraints and the inmate continues to refuse. Chemical agent is occasionally
applied to gain control over the inmate’s extremities in order to apply the
restraints. If the chemical agent is unsuccessful and the inmate remains
noncompliant, this may result in the five-member team entering the area and
gaining control over the inmate applying force to contain the inmate(s). Based
on a review of documentation and available video operational practices
appear to be consistent with policy.

Overall, the requirement for staff to follow the step-by-step procedures seems to be a
part of routine practices and the application of confrontational avoidance techniques
is an established part of the use of force protocol.

184

Chemical agent use (oleoresin capsicum). In 2013, Bureau Director Charles E. Samuels,
Jr. issued Operations Memorandum # 004-2013 (5500), authorizing the use of
oleoresin capsicum (OC), an aerosol chemical agent pepper spray for use by
authorized personnel to incapacitate or disable disruptive, assaultive, or armed
inmates posing a threat to the safety of others or to institution security and good
order.
The OC product is designed primarily for an immediate use of force or a calculated
use of force, where an inmate needs to be brought under control to avoid injuries to
others. OC is particularly effective in disabling assaultive inmates without causing
significant or long-term injury to the inmate. This less than lethal technology has
become widely used in corrections to prevent violence, control disruptive inmates,
and avoid staff injury.
The Bureau memorandum directs that only trained staff shall use OC and detailed
reporting requirements follow the use of OC, consistent with the Use of Force and
Application of Restraints policy referenced above.
The Bureau memorandum also identifies which staff members occupying certain
posts are authorized to carry the OC aerosol dispenser and that Post Orders be
revised providing specific instructions and direction for the use of OC. It is also
required that instructions be consistent with the Use of Force and Application of
Restraints policy. To ensure consistent application in compliance with the
memorandum, post orders are required to be submitted to the regional correctional
services administrator for signature and approval.
The memorandum and policy outlines requirements for decontamination of the
inmate and area where OC has been applied. Typically, fresh air and rinsing with
water will reduce the effects of the chemical agent. Policy dictates that
decontamination take place approximately 15 minutes after application of OC.
Additional procedures are outlined in the document and in post orders at each
facility where the use of OC is authorized.
OC training is emphasized in the Bureau memorandum, which requires that staff be
thoroughly trained in the use of OC, decontamination procedures, and reporting
procedures. Each officer authorized to carry OC must receive a four-hour block of
training, as well as a quarterly re-familiarization training. The facility captain is
responsible to maintain records of training and personnel authorized to carry OC.
ACA Standard 4-4199, reads: “Written policy and procedure govern the availability,
control, and use of chemical agents, electrical disablers, and related security devices
and specify the level of authority required for their access and use. Chemical agents
and electrical disablers are used only with the authorization of the
warden/superintendent or designee.” A review of the policy documents and practices
observed found overall compliance with ACA standards.

185

The assessment team had occasion to review post orders at the facilities visited and
found that post order special instructions are contained in those post orders that are
considered “OC Carry” posts. Post order special instructions are thorough and
consistent with the Operations Memorandum issued by Director Samuels. These
procedures outline the allowed use of OC, which posts may use or carry OC, the
purpose for using the product, reporting procedures, decontamination procedures
and training requirements.
Spot checks were conducted at each facility and initial training was taking place and
quarterly refresher training was also occurring at most sites. For example at the ADX,
the captain delegates the quarterly refresher training to a lieutenant at that facility.
The lieutenant ensures that OC authorized staff review a PowerPoint presentation
and sign a log, which indicates they had completed the training. That document is
forwarded to the regional office for review by regional staff. Each site handles the
quarterly training differently, however the assessment team’s spot checks revealed
that some form of quarterly training was taking place at all sites.
The application of OC chemical agent constitutes a use of force when used on an
inmate. Use of force cases are analyzed as a separate section of this report. With
regard to the procedures and training concerning the administration of the OC
program, the assessment team found overall compliance with the policy. We did view
cases where OC was dispensed on inmates either through the use of the aerosol
spray or from a pepper ball launcher. The video recordings of the use of force cases
we reviewed did not reveal any abuse with respect to OC use.
Calculated use of force. This type of force occurs in situations where an inmate is in
an area that can be isolated (e.g., a locked cell, a range) and there is no immediate,
direct threat to the inmate or others. When there is time for the calculated use of
force, staff must first determine if the situation can be resolved without resorting to
force.
All of the calculated use of force incidents reviewed by the project team were due to
inmates refusing to obey orders. Overall in most instances the inmate complied with
the verbal order of the team leader without further resistance; in these cases the
reported use of force consisted of putting an inmate in restraints and removing him
from the cell/recreation yard/shower.
In another few instances where inmate(s) continued to refuse the orders, the team
had to enter the cell and force was used. A review of documentation and videos
revealed that occasionally inmates would repeatedly refuse to “cuff up” after one of
them had broken a sprinkler head in the cell; chemical agent was authorized but not
deployed and the inmates were subdued and seen by medical personal for
examination of any injuries. The videos were reviewed and the level of force used
appeared appropriate.

186

The project team reviewed several videos at each facility to observe operational
practices and to ensure there were efforts made by staff to de-escalate and achieve
compliance before any force was used. In most cases staff followed standard
procedures. In cases where procedures were not followed facility management
personnel, the after action review team or regional team ensured corrective action
would be applied.

Use of force observations
The code of federal regulations clearly identifies the requirements as to when force
can be applied. The regulations authorize the Bureau to use force only as a last
alternative after all other reasonable efforts have failed. Programs Statements, facility
supplemental statements and staff training are aligned with the federal regulations
and are used by staff to clearly identify policy, procedures and expected practices.
Based on the project team’s review there was no indication that there is an ongoing
practice by management to operate outside the federal regulations or program
statements.
The project team did review documentation from isolated incidents where an
individual staff member may not have followed policy and corrective action was
taken.
The program statement includes the requirements for an extensive review process
including several layers of management review to ensure incidents involving the use
of force are examined by multiple staff. These requirements appeared to be met on a
regular basis.
The project team reviewed over one hundred incident reports, policies were
examined, videos were assessed and both staff and inmates were interviewed. As
noted above, one concern was expressed by some inmates at USP Lewisburg or
inmates who had previously been at USP Lewisburg. Several inmates referenced the
operational practice of some staff to apply restraints in a manner inconsistent with
policy. This practice was described as applying the restraints excessively tight often
times after a previous incident.

Use of force at USP Lewisburg
As indicated above, our review of uses of force and critical incidents at each facility
showed few cases that were inconsistent with policy. However, as noted before, there
was a concern shared by a significant percentage of the interviewed inmates,
especially those housed at USP Allenwood who had previously been assigned to SMU
Levels 1 and 2 at USP Lewisburg.

187

Several inmates reported that some of the staff at USP Lewisburg have a tendency to
apply physical restraints tighter than necessary or tighter than normal. Inmates
shared that this was often the practice by some staff either after an earlier
challenging interaction with an inmate, previous experiences an inmate may have had
with staff, a combination of the overall nature of events occurring throughout the
day or for no apparent reason. In addition, several inmates at USP Allenwood
reported that staff at USP Lewisburg used verbal and physical intimidation
techniques against inmates routinely. No supportive evidence was presented to
indicate this was the routine staff practice, however the high number of reported
incidents were mentioned during inmate interviews suggests the need for further
investigation by the Bureau.
The project team did not observe this practice directly, However, based on (1) the
number of individuals that had expressed this concern and (2) the fact that many of
the inmates were not currently at USP Lewisburg however had previously been there
and shared this information was an issue that deserves mention and further review
and evaluation by the Bureau.

Conditions of confinement
The focus of this section of the report addresses the findings and recommendations
related to the housing, services and activity levels provided to individuals housed in
the three main types of restrictive housing units being analyzed: SHUs, SMUs, and
ADX (with general population, step-down, control, and special security units). This
section addresses the conditions of confinement rather than placement and
procedural issues which will be discussed in another section as well as an
assessment of educational programs.
To determine these conditions we examined several key factors:
Regulatory requirements131
Bureau program statements
Facility supplemental statements
Bureau program review reports

131

Title 28 CFR Part 541.

188

American Correctional Association standards and most recent facility
reports
Housing unit activity schedules
Manual and electronic systems designed to document conditions of
confinement
On-site observations of operational practices;
Staff interviews
Inmate interviews
According to federal regulations and Bureau program statements, all three of the
restrictive housing units are designed to help ensure the safety, security, and orderly
operation of the Bureau facilities. The conditions of confinement are generally
considered more restrictive than those found in the general population. Specific
conditions of confinement for each type of segregation vary based on the unit and
the inmates’ classification level within each unit.
Program statements and facility supplemental statements related to conditions of
confinement in segregation units have all been developed in-line with the statutory
requirements identified in the federal regulations. In addition to the statutory
requirements, the program statements and facility supplemental statements also
identify policies, procedures and objectives regarding a specific area, including
conditions of confinement.
These program statements for the most part serve as the core documents staff use to
reference the minimum required conditions of confinement for individuals housed in
each of the forms of restrictive housing units. To be in compliance with the program
statements, housing unit activity schedules have been developed, staff post
responsibilities have been established, staffing patterns have been designed, staff
training implemented and personnel and performance oversight is currently
provided.
The review team found that personnel were generally familiar with the program
statements and the condition of confinement requirements. Activity schedules were
in place, appropriately trained personnel were assigned to adhere to these
requirements and oversight was being made. A general review is described in more
detail below. Applicable federal regulations and Bureau program statements are
referenced as well as the associated American Correctional Association standards.
These references have all been italicized. In addition to the references the project
team’s general assessments are provided.

189

The conditions of the living quarters are the result of an ongoing physical plant
maintenance program that evaluates and addresses issues related to maintaining
nationally accepted environmental conditions for living quarters. Overall, the
preventive and ongoing maintenance program appears to be a priority at the facilities
visited. Most facilities maintained a high level of sanitation. Some of the facilities
housing units were air conditioned while others were not. At USP Lewisburg, the
most recent ACA accreditation review completed in 2011 indicated there was
inadequate lighting (low) in the SMU cells.132 This issue had since been corrected as
reported in documentation provided and based on observation of the units. In
October 2011, the Bureau Program Review Division reported that the sanitation
condition of the SHU at USP Florence was below average and corrective action was
required.133 Again, based on supported documentation and on-site observations, an
ongoing practice has been established to maintain appropriate sanitation conditions
at USP Florence on a regular basis134.
FINDING: Overall the sanitation and physical plant maintenance programs in
Bureau facilities are considered consistent with nationally recognized best
practices.
Cell occupancy.
Living quarters ordinarily house inmates according to the design of the unit. The
number of inmates assigned to a cell should not exceed the number for which the
space was designated. Conditions may be altered by the warden as long as it meets
applicable standards. (28 CFR 541.31, P5217.01, P5270.10, ACA 4-4134, 4-4140, 44141.)
Inmates are generally housed based on their status and level within each restrictive
housing unit. At the ADX, all inmates were housed in single cells in the same housing
unit as other inmates assigned to the same level of the program. All cells are
designed for one inmate and include one bed, toilet, wash basin and a shower. The
size of the cell is consistent with nationally recognized standards.

Commission on Accreditation for Corrections Standards Compliance Intensive
Reaccreditation Process Audit Federal Bureau of Prisons Lewisburg, Pennsylvania, August
2011.
132

Bureau Program Review Division Memorandum dated October 27, 2011, Subject:
Correctional Services Program Review FCC Florence.
133

A Correctional Services program review conducted in October 2014 after the completion of
the site visits of this project, the Bureau Program Review Division reported the condition of
both SHU units were at a “high level of sanitation in common areas and cells.”

134

190

The design of the SMU program is to have inmates housed according to their level
within the program. Inmates assigned to Level 1 are either housed alone or with one
other Level 1 inmate, inmates in Level 2 may be housed with one other Level 2 inmate
and inmates assigned to Level 3 and four are designed to be housed with an inmate
in the same level. At USP Florence and USP Allenwood, this policy appeared to be in
practice. At USP Lewisburg, the operational practice observed was that inmates in
Level 1 and Two may be housed together. In addition, at USP Lewisburg, the program
was designed to house only inmates in Level 1 and Two at the time of the site visit.
However, staff reported that – in part because of limited space at other sites
including USP Allenwood – some Level 3 and 4 inmates were being housed at USP
Lewisburg on the same ranges as Level 1 and 2 inmates. There were approximately
100 inmates in Level 3 or 4 housed at USP Lewisburg during the project team’s site
visit. The number of inmates assigned to the cells in SMU’s did not exceed the
number of beds in the cell. Most cells contained two beds. USP Lewisburg did have a
few four-person cells containing four beds and the size of the cells were consistent
with nationally recognized housing standards.
FINDING: The size and furnishings provided in the cells in the SMUs were
consistent with nationally accepted practices.
In a SHU, inmates are assigned to either administrative detention or disciplinary
segregation status. Administrative detention status inmates are generally not housed
in the same cell as inmates assigned to disciplinary segregation status, however they
are assigned to the same housing unit. Administrative detention is a nonpunitive
status and disciplinary segregation is a punitive status.135 Some facilities house
disciplinary segregation inmates on a separate range (housing floor) from
administrative segregation inmates, while others have both types of inmates on the
same range however in different cells.
The average cell in a SHU contains two beds and the size of the cell is approximately
85 square feet with approximately 35 square feet of unencumbered space. The exact
cell sizes vary to the physical plant configuration even within the same facility. In
each SHU and SMU there are also a few cells designed to house one inmate. Some of

28 CFR 541.22 (a) Administrative detention is an administrative status which removes
the inmate from the general population when necessary to ensure the safety, security and
orderly operation of correctional facilities, or to protect the public. Administrative
detention status is nonpunitive, and can occur for a variety of reasons. (b) Disciplinary
segregation is a punitive status imposed only by a disciplinary hearing officer (DHO) as a
sanction for committing a prohibited act(s).
135

191

these cells are normally used on a temporary basis to more closely observe an
inmate.
FINDING: The size and furnishings provided in the SHUs is consistent with
nationally accepted practices when providing housing for the designed number of
inmates. The fact that most cells in the SHUs contained showers exceeded
national standards. However in one facility, the number of inmates housed in a
cell exceeded the design capacity.
At USP Victorville, the SHU consists of 120 cells of which 118 are designed for two
inmates and two cells are designed for one inmate. Based on the overall SHU
population level, facility staff reported they routinely found it necessary to house
three inmates in a cell designed for two (contains two beds). This housing practice
was confirmed by the warden, administrative personnel, housing unit staff and
inmates. The third inmate is provided a mattress accompanied by appropriate
bedding and is required to sleep on the floor of the cell.
The frequency in which this practice occurs was reported as routine when the
population level exceeds or comes close to capacity. The length of time an individual
is housed on the floor was not being tracked and staff interviewed were not aware of
the average length of time an inmate is housed in a cell with two other inmates. The
warden reported this most often occurs when issues arise with the transportation
systems ability to move inmates. On those occasions inmates get backlogged
awaiting movement.
On the second day of the on-site review, there were 35 USP inmates classified for
SHU placement housed in the SHU at FCI Victorville I and 42 USP inmates housed at
FCI Victorville II. There were no inmates being tripled-celled in the SHU at USP
Victorville at the time of the on-site review.
It was reported by supervisory staff that five days prior to the review team’s arrival, a
number of inmates were transferred from the USP Victorville SHU to either the FCI I
SHU or FCI II SHU. The warden reported there was no written policy or written
procedure in place that identified the protocol to follow to determine when three
individuals can be housed in a two-person cell, who can be assigned three to a cell,
how long they can be assigned or when the FCI I or FCI II special housing units can be
used to house USP SHU inmates. Once three individuals are assigned to a cell
containing two beds and one individual is required to sleep on a mattress on the
floor, unencumbered space is reduced and nationally recognized housing standards
are no longer maintained.
Clothing. Inmates should receive adequate institution clothing, including footwear.
Inmates have opportunities to exchange clothing or have it washed. (28 CFR 541,
P5580.08, P5217.01, P5270.10, ACA 4-4263.)

192

Based on observation, interviews and policies reviewed appropriate clothing,
footwear and the opportunity to exchange the clothing and/or have it washed is
available on a regular basis. Additional limited clothing items may be purchased
from the commissary for inmates in specific levels within each restrictive housing
unit.
Bedding. Inmates should receive a mattress, blankets, a pillow, and linens for
sleeping. Inmates have necessary opportunities to exchange linens. (28 CFR 541,
P5217.01, P5270.10, ACA 4-4263).
Based on site observation, interviews and policies reviewed inmates generally receive
a mattress with a built-in pillow, sheets, blanket and a towel. The sheets can be
exchanged at least once per week.
Food. Inmates receive nutritionally adequate meals and may be required to eat all
meals in their living quarters. (28 CFR 541, P5217.01, P5270.10, ACA 4-4264, 4-4316,
4-4318, 4-4319, 4-4320).
The Bureau provides a standard menu that applies to most facilities and is approved
by a dietician and nutritionist. Inmates may receive medical and religious diets when
considered appropriate and may request heart healthy meals. Substitutions may be
made in segregation units when meat or poultry items containing bones are being
served to the general population. Inmates receive three meals per day and all meals
are consumed in the cell. The timing of the meals at some of the facilities reflected
the breakfast meal was occasionally being served approximately 14 or more hours
after the last meal of the previous day. Although this feeding schedule was not
observed at every facility the scheduling and time of delivery of meals should be
closely monitored to avoid extensive periods with no food services.
Personal hygiene. Inmates should have access to a wash basin and toilet. Inmates
receive necessary personal hygiene items. (28 CFR 541, P5270.10, P5217.01 ACA 44261). Inmates should have the opportunity to shower and shave at least three times
per week (P5270.10, P5217.01, ACA 4-4262). Inmates should have access to necessary
hair care services (28 CFR 541.21, P5270.10, P5217.01; ACA 4-4263).
All cells contained a wash basin and toilet. Most of the cells in the SHU and control
unit contained a shower within the cell. Those facilities where a shower was not
located within the cell, staff provided the inmate with access to a shower at least
three times per week. The cells located in the SMUs at USP Lewisburg, USP Florence,
and USP Allenwood did not contain a shower, however multiple showers were located
within the housing unit. Documentation over a period of several months was
reviewed and reflected showers and the opportunity to shave were being provided in
a manner consistent with federal regulations. Hair care services are also available at
least once per month. Personal hygiene items are available to the inmate in the
housing unit as well as through the commissary.

193

Exercise. SHU and SMU inmates have the opportunity to exercise outside their
individual quarters for five hours per week, ordinarily in one-hour periods on
different days. The warden may deny these exercise periods for up to one week at a
time if it is determined that an inmate’s recreation itself jeopardizes the safety,
security, or orderly operation of the institution. However, recreation conditions
specified here may not otherwise be limited, even as part of a disciplinary sanction
imposed under P5217.01, P5370.11; ACA 4-4270. 28 CFR 541, P5270.10.
Exercise and Recreation The frequency of exercise outside the cell varies based on
the type of segregation/program assigned and the inmates’ level within the program.
SHU exercise and recreation. Inmates assigned to a SHU are generally provided the
opportunity to exercise outside their cells five hours per week, ordinarily in one-hour
periods five days per week. One facility scheduled recreation four days per week and
provided inmates between one and one quarter and one and one half hours of
recreation per day. The five hours provided is consistent with the minimum
requirements cited in the federal regulations. Inmates in administrative detention
status are generally not placed in the same recreation cage as an inmate in
disciplinary segregation status; however, inmates in both statuses are provided the
same amount of recreation time per week. Outdoor recreation is provided in
individual recreation cages located adjacent to the unit. Each individual recreation
cage may contain between one and four inmates based on the size of the cage and
the number and type of inmates requesting access. No indoor recreation is provided.
The procedure for gaining access to recreation was found to be similar at most SHUs.
Inmates are asked by an officer on each scheduled recreation day whether they are
interested in recreation for that day. The request is generally made either before the
breakfast meal or after the meal is served. No response or a negative response is
considered a refusal and documented in the electronic file as the same. Recreation is
generally provided first thing in the morning and at some facilities is completed by
0900. At other facilities, recreation continues until approximately 1600. The
recreation scheduled varied at several facilities and was normally based on the
number of inmates requesting recreation for that day.
The Program Review Division reported in October 2011 that at USP Florence the
facility failed to properly complete documentation identifying recreation, meals and
unit officer signatures in the SHU136. The project team reviewed both manual and
automated reports for the most recent four months and found that the
documentation indicated that inmate access to meals, recreation and unit officer
signatures were being provided on a regular basis.
The Bureau subsequently report that at the FCC Florence Correctional Services Program
Review conducted in October 2014, revealed these issues were corrected.

136

194

Reports and documentation reviewed at several facilities reflected that a very small
percentage of inmates were utilizing the recreation cages. Inmates interviewed
expressed concerns that recreation requests occur only once per day, at
approximately 0500 and that if the inmates are not standing at the cell door at that
time they were not allowed to participate in recreation on that day. The requirement
to stand at the door was not found to be a required practice at all facilities.
Project team members observed the practice at several facilities which consisted of
the officer going to each cell and requesting whether the inmate wanted to go to
recreation that day or not. Standing at the cell door was not a required procedure
during the project team’s observation. Staff reported that on average, 50 percent of
the inmates in administrative detention status do not participate in recreation on any
given day. A significant number of inmates in administrative detention status
reported to team members that they chose not to participate in recreation because
they feared being assaulted. Inmates in administrative detention status are not
placed in the same recreation cage as an inmate in disciplinary segregation status,
however in some facilities recreation may take place at the same time.
SMU exercise and recreation. Inmates assigned to the SMU program receive the
opportunity to exercise outside their cells at least five hours per week, ordinarily in
one-hour periods five days per week. SMU Level 1 and 2 inmates receive one hour of
outdoor recreation, five days a week. SMU Level 3 and 4 inmates may receive up to 15
hours of indoor / outdoor recreation per week depending upon where they are
housed. There is no difference in the amount of recreation time offered to Level 1
and 2 inmates or between Level 3 and 4 inmates, except for where they are housed.
Inmates in Level 3 and 4 housed at USP Lewisburg were not provided indoor
recreation. As a result, they were limited to approximately five hours of outdoor
recreation per week based on weather conditions. Overall, however, practices were
consistent with minimum federal regulations.
ADX. exercise and recreation Inmates assigned to the ADX program receive the
opportunity to exercise outside their cells at least seven hours per week. This
practice is consistent with federal regulations. ADX General Population inmates
receive two hours per day out of cell five days a week for a total of ten hours. Control
inmates receive 1.5 hours a day four days per week and one hour one day for a total
of seven hours per week out of cell. The ADX Step-down inmates normally receive 1.5
hours inside and 1.5 hours outside cell seven days a week for a total of 21 hours per
week.
Personal property. Inmates may have reasonable amounts of personal property.
Personal property may be limited for reasons of fire safety, sanitation, or available
space (28 CFR 553, 28 CFR 541, P5270.10, P5217.01; ACA 4-4261, 4-4265, 4-4292).
In administrative detention status an inmate may ordinarily be allowed a reasonable
amount of personal property and reasonable access to the commissary. In

195

disciplinary segregation status an inmate’s personal property will be impounded,
with the exception of limited reading/writing materials, and religious articles. The
warden may modify the quantity and type of personal property allowed. Personal
property may be limited or withheld for reasons of security, fire safety, or
housekeeping. Unauthorized use of any authorized item may result in the restriction
of the item. Also, commissary privileges may be limited.
SHU personal property. Although administrative detention status is a nonpunitive
status and disciplinary segregation is a punitive status, there are very few differences
in conditions of confinement between the two groups. However, personal property is
one area where there are a few minor differences. The two notable differences in
personal property were inmates assigned to administrative detention were allowed to
purchase a portable radio with ear buds and a limited amount of food items and
coffee from the commissary. Inmates in disciplinary segregation were normally not
allowed to purchase those same items.
Other noted differences between the two SHU statuses outside of personal property
was that administrative detention inmates are to have access to educational
programming while inmates in disciplinary segregation may have their programming
suspended. In addition at two facilities, inmates in administrative detention received
a two hour visit per week while inmates in disciplinary segregation received a one
hour visit per week. No other significant differences in conditions were noted
between the two statuses.
With respect to personal property, administrative detention status inmates according
to the SHU program statement are ordinarily allowed to have the following: Bible,
Koran, or other scriptures (1), books, paperback (5), eyeglasses, prescription (2), legal
material (see policy on inmate legal activities), magazine (3), mail (10), newspaper (1),
personal hygiene items (1 of each type) (no dental floss or razors*), photo album (25
photos), authorized religious medals/headgear (e.g., kufi), shoes, shower shoes,
snack foods without aluminum foil wrappers (5 individual packs), soft drinks,
powdered (1 container), stationery/stamps (20 each), wedding band (1), radio with
ear plugs (1) and a watch (1).
FINDING: There is very little difference between the personal property of inmates
assigned to administrative detention and disciplinary segregation with the
exception of noted commissary items. This is problematic given that at some
facilities the majority of administrative detention inmates are on protective
custody status.
SMU personal property. Inmates are limited in the amount of personal property that
may be possessed, purchased and maintained in their assigned cell. As participants
progress through the program, more property privileges become available. Separate
commissary lists are available and generally grouped into lists for inmates in Level 1
and two and those inmates in Level 3 and 4.

196

USP Allenwood, which does not house Level 1 or 2 inmates, primarily has one list and
the difference in approved property between Level 3 and 4 inmates is that inmates in
Level 4 may purchase one pair of sweat pants and one sweatshirt and inmates in
Level 3 cannot. Other than the clothing mentioned there were no significant
differences.
At USP Lewisburg the approved property list for SMU inmates is the same for each
level,137 although a separate commissary list is provided for inmates in Level 1 and 2
compared to Level 3 and 4. The primary difference between inmates in Level 3 and 4
and those in Level 1 and Two at USP Lewisburg is inmates in Level 3 and 4 have
access to additional clothing and snack items. Inmates placed on commissary
restriction will be able to purchase stamps and certain hygiene items (as specified by
staff) based on their level.
At USP Florence, two separate commissary lists are provided, one for inmates in SHU
and Level 1 and 2 of the SMU and one for inmates in Level 3 and 4 of SMU. The
primary differences between the two lists are Level 3 and 4 inmates have access to
more food items, clothing, miscellaneous items and additional options in each
category.
ADX personal property. Property limitations including limitations on legal materials
were found to be appropriate in volume and type to ensure safety and good order of
the ADX. Personal property was being stored appropriately in the space provided. All
general population, step-down, control unit, special security unit, and SHU inmates,
with the exception of those inmates assigned to disciplinary segregation, are issued
televisions. With the exception of inmates assigned to the SHU, each television has
leisure viewing channels. Separate commissary lists are provided for each
classification within the ADX and appeared appropriate based on the classification
and security nature of the facility. Primary differences in the various commissary
lists included the number of options available, access to additional clothing and food
and snack items. Overall, access to property and commissary items is consistent with
national standards.
Correspondence and telephone use. Inmates may correspond with persons in the
community and use the telephone in accordance with 28 CFR 540.17 and Program
Statements P5217.01, P5264.08 and P5270.10. Special mail and unmonitored
attorney telephone calls are handled in accordance with 28 CFR 540.17. 28 CFR
540.17; ACA 4-4266, 4-4271, 4-4272. 28 CFR 540.16(b) states, the warden shall
permit an inmate in segregation to have full correspondence privileges unless placed
on restricted general correspondence under 540.15. 28 CFR. 540.15(a) states the

137

USP Lewisburg SMU Handbook, page 51, Appendix B.

197

warden may place an inmate on restricted general correspondence based on
misconduct or as a matter of classification. The warden shall permit an inmate who
has not been restricted from telephone use as a result of a specific institution
disciplinary sanction to make at least one telephone call each month.
28 CFR 540.101(d) Procedures states ordinarily an inmate who has sufficient funds is
allowed at least three minutes for a telephone call. 28 CFR 540.103 states the warden
may not apply frequency limitations on inmate telephone calls to attorneys when the
inmate demonstrates that communication with attorneys by correspondence, visiting
or normal telephone use is not adequate. 28 CFR 540.105(b), Expenses of inmate
telephone use, states that the warden shall provide at least one collect call each
month for an inmate who is without funds. 28 CFR 540.17; P5217.01; ACA 4-4266, 44271, 4-4272, Program Statement P5264.08 states a local institution supplement is
required and must include, in part, the following: a maximum length of telephone
calls of 15 minutes.
The frequency and length of time of a social telephone call as cited in the Bureau’s
program statements is at least one telephone call per month for no more than 15
minutes, unless the inmate is placed on telephone restriction as a result of a
disciplinary sanction.
SHU correspondence and telephone use. In the SHU’s this is the general practice for
both administrative detention and disciplinary segregation inmates. There is no
difference in access to a telephone for inmates assigned to administrative detention
or disciplinary segregation. The minimum frequency is provided and the maximum
duration is set by the program statement. All other correspondence, including mail
services are the same as those services provided to inmates assigned to general
population, with the exception of email access.
SMU correspondence and telephone use In the SMU program opportunities to send
and receive written correspondence and make telephone calls are subject to
monitoring and analysis for intelligence purposes. Legal correspondence and calls
are not subject to monitoring and analysis. Unless an inmate is on telephone
restriction status, inmates assigned to Level 1 of the SMU are allowed to make two
telephone calls per month. Inmates assigned to Level 2 are allowed to make up to
four telephone calls per month. Telephone calls may last up to 15 minutes. Level 3
inmates can access telephones from the common area of their housing unit and may
make up to 150 minutes of calls per month. Level 4 inmates may make up to 300
minutes of calls per month and participate in the electronic mail program (email).
There is a program at USP Florence for inmates who have lengthy telephone
restrictions of months or years to earn back telephone privileges by compliance with
certain items. This was not found to be an incentive found in all facilities.
ADX correspondence and telephone use In the ADX General Population, Control
and SHU inmates schedule legal calls through their assigned counselor. General

198

Population inmates can make (2) 15 minute telephone calls per month, Control
inmates can make (1) 15 minute telephone call per month unless on disciplinary
status then (1) 15 minute call every 90 days, SHU inmates receive (1) 15 minute
personal telephone call per month on administrative detention status and (1) 15
minute telephone call if on disciplinary status. Control step-down unit inmates can
receive (3) 15 minute calls per month. Incentive telephone calls can be earned. All
inmate personal telephone calls must be live monitored by ADX staff.
All incoming and outgoing ADX inmate correspondence is inspected by Special
Investigation Service technicians. Incoming and outgoing social mail is reviewed by
SIS technicians and must be processed within 36 hours. Legal mail is picked up and
delivered by the unit counselor which must be processed within 24 hours of receipt. .
Visiting. Inmates may receive visitors in accordance with 28 CFR 540.40. 28 CFR
540.43, Frequency of visits and number of visitors, P5217.01, P5270.10 and P5267.08.
The warden shall allow each inmate a minimum of four hours visiting time per
month. The warden may limit the length or frequency of visits only to avoid chronic
overcrowding. Inmates may be provided noncontact visits, through the use of
videoconferencing or other technology. (P5217.01, P5270.10, P5267.08, ACA 4-4267.)
Title 28 CFR 541.17 states: “Ordinarily, an inmate in administrative detention or
disciplinary segregation status may receive visits in accord with the same rules and
regulations that apply to general population inmates, providing such visits do not
pose a threat to the security or orderly operation of the institution. In such cases, the
warden may authorize special visiting procedures to preclude such a threat.”
SHU visitation. Generally an inmate retains their visiting privileges while in
administrative detention or disciplinary segregation status. However, visiting may be
restricted or disallowed when an inmate (while in administrative detention or
disciplinary segregation) is charged with or was found guilty of a prohibited act
related to visiting guidelines, or has acted in a way that would reasonably indicate a
threat to security or order in the Visiting Room.
A minimum of four hours per month of visitation is required by statute and the
program statement unless the length of time creates chronic overcrowding. At one
facility, FCI Butner II, inmates assigned to the SHU received contact visits, however at
all other SHUs, only noncontact visits were provided. The frequency and length of
visit for both administrative detention and disciplinary segregation inmates at most
facilities were generally the same, one visit per week for two hours. At USP Coleman I
and FCI Butner II inmates in disciplinary segregation status received a one-hour visit.
SHU inmates generally receive noncontact visits while general population inmates
receive contact visits.
At USP Florence, the SHU inmates receive visits through video from a secure room
within the SHU. Video visits at USP Florence may last two hours and there is a limit of
five visits per month for SHU inmates. Few SHU inmates actually receive visits based

199

on information obtained during interviews. Legal visits are arranged upon request
and take place in the general visiting room area either in a glass noncontact visiting
booth or in private room designated for attorneys, depending on the circumstances.
SMU visitation. At USP Lewisburg and USP Florence, social visits are provided
primarily through video technology. The inmates utilize the video visiting room
located in their assigned housing unit, and their visitors utilize the video-visiting
units located in the Visiting Room. Regular visitation for SMU inmates in Levels One,
Two, and Three is only available to immediate family members, and the relationship
must be verified. At USP Lewisburg, inmates in Level 3 are permitted noncontact
visits in the visiting room and inmates in Level 4 are permitted contact visits in the
visiting room and are not limited to immediate family members. SMU inmates are
limited to a maximum of five social visits per month. This does not include legal
visits. SMU inmates must submit a request, in writing at least one week in advance of
the expected visit. As the availability of video equipment may be limited, visits may
be limited to two hours per inmate at USP Florence and one hour visits at USP
Lewisburg, although more time may be allotted based on availability of visiting
booths.
At USP Allenwood, legal visits are allowed in a manner consistent with national
practices. Social visits are noncontact visits and are ordinarily scheduled from 0830
to 1500 hours, Tuesday through Thursday. The social visits for these inmates are
limited to immediate family members only. Level 3 inmates are allowed two one-hour
visits per month and Level 4 inmates are allowed four one-hour visits per month. The
frequency and duration allowed for each visit is reportedly limited due to the
number of noncontact rooms available. There are four private noncontact visitation
rooms. Staff at USP Allenwood reported the number of visits for SMU inmates was
quite low because of several factors including the distances that most visitors have
to travel in most instances.
The applicable federal regulation states: “The Warden shall allow each inmate a
minimum of four hours visiting time per month. The Warden may limit the length or
frequency of visits only to avoid chronic overcrowding.”138
FINDING: The frequency of visitation allowed for inmates in Level 3 at USP
Allenwood is inconsistent with Bureau practices for the same level of inmate at
other facilities and not representative of national best practices.
At USP Florence, the frequency for Level 3 inmates was five visits per month for up
to two hours per visit through the use of video technology. At USP Lewisburg, Level 3
inmates were allowed up to five one-hour noncontact social visits per month. When
138

28 CFR 540.43, Frequency of visits and number of visitors.

200

comparing USP Lewisburg visitation policies with USP Allenwood visitation policies,
inmates in the same level are allowed a difference of three hours in visits per month.
RECOMMENDATION 7.1: A further review of the frequency and duration of visits
should be conducted at USP Allenwood for Level 3 inmates. Serious consideration
should be given to providing the allowance for additional time for those inmates
in Level 3.
Although the warden may limit the length of time or frequency of visits to avoid
chronic overcrowding, there was no evidence presented to indicate the established
frequency and duration for Level 3 inmates as identified in the facility policy should
be less than four hours per month. In addition to the hours provided, there is an
issue of consistency in the conditions of confinement for each level. Currently, an
inmate housed at USP Allenwood in Level 3 is allowed up to two hours of social visits
per month while the same inmate housed at USP Florence would receive ten hours of
social visits per month.
Establishing a reference to “more time may be allowed based on availability” in the
policy, expanding the number of days or hours visitation is allowed for Level 3
inmates, use of video technology and/or increasing the number of noncontact rooms
should all be considered as viable options to provide more consistent visitation
practices.
ADX Visitation. The legal visits for inmates confined at the ADX are scheduled
through the housing unit counselor. ADX inmates can receive five general visits per
month with each visit lasting as long as seven hours. Inmates are dressed in different
clothing for visits; SHU-Orange Jumpsuits, Control-Yellow Jumpsuits, and General
Population-White Jumpsuits. Hand restraints are removed from General Population
and SHU inmates after the inmate is secured in the visitation booth with leg irons
remaining in place. Control inmates remain in full restraints even after placement in
the visitation booth. The ADX has ten noncontact visitation booths for general visits
and four noncontact visitation booths for legal visits.
Legal activity. SHU inmates must submit a request to use the electronic law library
located in a room in SHU. The equipment is available seven days per week. A review
of the logbooks reflects that this equipment is frequently used. SMU Level 1 and 2
inmates submit a request to use the electronic law library that is located in their
respective housing units. At USP Florence, there are two enclosures in each unit that
contain the equipment and a printer. They were observed in frequent use during our
review. SMU Level 3 and 4 inmates have open access to this equipment that is located
in the common area of their units, which they can frequently access. That equipment
was also observed in use during our review. Upon approval of the unit manager, an
inmate may request another inmate to assist in legal matters / use of the equipment.
This was also observed in action with Level 1 and Two SMU inmates. At USP
Allenwood, the records reflected a minimum of one hour access and frequently more

201

than that. SMU inmates have a dedicated electronic law library in the common
recreation area that may be used during recreation. An additional terminal is located
in a room adjacent to the common area when a SMU inmate needs privacy or for
noise considerations. The project team noted consistent access to the electronic law
library throughout site visits.
Access to medical care. Inmates in restrictive housing units have limited capability to
access and communicate to staff because of the nature of secure confinement.
Nationally accepted best practice dictates that inmates are afforded daily access to
medical care, or in more emergent circumstances immediately. Correctional
institutions should have in place a method to allow for unimpeded access to medical
care. This is particularly difficult to carry out in restrictive housing units, where the
inmates are mainly locked in their cells 23 hours per day, seven days per week. This
section of the report examines access to healthcare in the Bureau facilities assessed.
Title 28 Code of Federal Regulations, section 541.32 (a) addresses the provision of
medical and mental health care in the SHU. The regulation specifies that a health
services staff member will visit each inmate daily to provide necessary medical care,
exclusive of emergency care, which is provided on an as needed basis.
The assessment team examined SHU sign-in records and logbooks, as well as making
direct observations at each of the facilities assessed to determine the frequency that
medical staff signed into the unit to make rounds of the cellblock ranges and also
administer medication. The following findings are the result of that analysis.
USP Tucson. Medical staff visit the unit two times per day. During the
morning visit, healthcare staff observe each inmate in their cells and respond
to inquiries. Also the medical staff receive formal written medical requests
from the inmate. Evening medical rounds focus more on administering
medication.
USP Victorville. Records were reviewed for a 60 day period and the following
was revealed. The average number of medical staff visits to the unit was two
per day. In five of the 60 days medical staff signed into the unit one time. A
medical staff member visited the unit at least once per day for all 60 days.
USP Hazelton. At this facility, sign in records revealed that once per day a
physician’s assistant visited the unit to visually observe the inmates.
USP Florence. Assessment team members observed medical staff traveling
cell to cell to check on the inmates and make visual observations. Records
revealed that medical staff entered the unit multiple times per day.
USP Terre Haute. Sign in logs revealed that a registered nurse enters the unit
a minimum of once per day and a medical staff person visits the unit
multiple times each day.

202

ADX Florence. SHU records revealed the presence of medical staff two times
per day to conduct rounds and administer medication.
USP Allenwood. In the SHU medical staff signed into the unit two times per
day to conduct rounds and administer medication.
FCI Butner. 90 days of sign in logs were reviewed and it was determined that
medical staff visited the SHU multiple times per day administering
medications, observing inmates, and collecting medical slips.
USP Coleman. 62 days of sign in logs were reviewed revealing the presence of
medical staff on a daily basis in the unit.
The ACA standards that address the provision of medical care in restrictive housing
units call for, at a minimum, daily visits by qualified healthcare personnel and
unimpeded access to prescribed medication.139
SMU healthcare access requirements are outlined in Program Statement P5217.01,
Special Management Units, dated 11/19/2008. Section 5 (a) 15 regarding medical care
reads: “…a health services staff member visits inmates daily to provide necessary
medical care. Emergency medical care is always available either at the institution or
from the community.”
Three SMUs were visited during this assessment. At USP Florence, SMU sign in
records and logs revealed that health care staff visited the unit one time per day,
consistent with the requirements of the program statement.
The largest of the SMU programs is at USP Lewisburg. Logbooks, sign in sheets and
observations revealed that health care staff enter the units at least one time per day
to conduct their rounds. A spot check of sign in documents was conducted covering
a six-week period to make this determination.
At USP Allenwood, which houses SMU inmates in Level 3 and Level 4, records
revealed that health care staff visited the two SMU cellblocks, two times per day.

4-4258 Written policy, procedure, and practice provide that inmates in segregation
receive daily visits from the senior correctional supervisor in charge, daily visits from a
qualified healthcare official (unless medical attention is needed more frequently), and
visits from members of the program staff upon request. 4-4261 Written policy, procedure,
and practice provide that all inmates in segregation provide prescribed medication,
including that is not degrading, and access to basic personal items for use in their cells
unless there is imminent danger that an inmate or any other inmate(s) will destroy an
item or induce self-injury.
139

203

ADX control unit healthcare personnel make rounds of the various control unit
housing areas once per shift or two times per day. Section 541.67, of Title 28 Code of
Federal Regulations requires that “…a medical staff member see the inmate daily and
regularly record medical and behavioral impressions.”
Bureau regulations and program statements provide clear direction with respect to
inmate access to healthcare in segregated units. Inmates in these units have limited
mobility and a limited capacity to communicate with staff. This has been recognized
by the corrections profession and nationally accepted correctional standards, such as
the ACA standards require that inmates be seen daily in the living unit and also have
unimpeded access to their prescribed medications.
Bureau regulations call for a minimum of a daily visit by a qualified healthcare
professional to these units. Our assessment process included a review of sign in or
attendance records of health care staff entering the various segregated units. Our
findings are outlined in the above paragraphs. These findings are supported by
interviews with staff, interviews with inmates, and documentary evidence that
revealed that health care staff meet and often exceed the minimum requirement of
attending to inmates in segregated units.
This examination did not include a review of medical records or the quality of
medical care that is provided, as this was beyond the scope of this analysis. A
number of the inmates interviewed did express concern about the quality of the care
they were receiving; however, there were few complaints that medical staff were
unavailable to speak with inmates during medical rounds.
FINDING: The presence of health care staff in the segregated units is ongoing at
least daily, and often more frequently, consistent with Bureau policy and ACA
standards.
Health care staff administer medications to inmates and this is an ongoing process
that takes place at different times during the day. In addition, the medical staff are
required each day to visually observe all inmates, regardless whether or not they are
receiving medication, as a wellness check to determine if they may need healthcare
services. There were no examples that we detected where a medical staff person
didn’t make a visit to a segregated housing unit at least once per day.

Adequacy and allocation of program space
The focus of this section of the report addresses the findings and recommendations
related to program space available in the three main segregation units being
analyzed: SHUs, SMUs, and ADX. The reported lack of space has been cited at several
facilities by staff as one of the primary reasons why programming is provided for

204

most inmates individually in their cell using a self-study format. This section focuses
more on the potential use of existing space available to provide small group
programming for appropriate inmates rather than on specific programming
opportunities, which are addressed in another section of the report.
According to federal regulations and applicable Bureau program statements, all three
of the restrictive housing units are designed to help ensure the safety, security, and
orderly operation of the Bureau facilities. Title 28 Code of Federal Regulations,
section 544.81 Education states: “the warden shall ensure that an inmate with the
need, capacity and sufficient time to serve has the opportunity to: Complete an Adult
Literacy program leading to a General Educational Development (GED) certificate
and/or high school diploma.” The required access to program activities is based in
part on the inmates’ status and the type of restrictive housing unit in which they are
housed.

Statutory and Bureau requirements
Special housing unit programs. In administrative detention status, an inmate will
have access to programming activities to the extent safety, security, orderly
operation of a correctional facility, or public safety are not jeopardized. In
disciplinary segregation status, participation in programming activities, e.g.,
educational programs may be suspended. (28 CFR 541.31, 28 CFR 544.81, P5270.10).
Inmates assigned to an SHU are either in administrative detention status
(nonpunitive) or disciplinary segregation status (punitive). Based on federal
regulations, inmates in administrative segregation status are allowed to have
program access as long as it does not jeopardize safety. Inmates in disciplinary
segregation status may have their program activities suspended.
Educational and religious program activities are provided on an individual basis
normally while the inmate is in the cell and the staff member is outside the door.
There is no small group programming provided with the exception of recreation
where between one and four inmates may be placed in a single recreation yard. In
most facilities, educational services were available however the services were
considered limited and provided in a self-study format while the inmate remained in
the cell.
At FCI Butner II, inmates enrolled in GED or ESL classes prior to being in the SHU
were allowed to remain in those classes for 60 days and provided with appropriate
instructional materials to continue their studies. Inmates in the SHU more than 60
days are dropped from those courses and placed on a waiting list to restart the
courses upon their release from the SHU. A new education program with a series of
classes on various topics was reported to have started in the SHU beginning February
17, 2014. The inmates are provided self-study materials to be completed in-cell.

205

Twenty-six courses are offered in six categories: Math (7 classes); Science (3); Reading
(4); Social Studies (4); Writing (4); and Miscellaneous classes (4). The significant
number of enrollees in the first week is a strong indication of the desire of SHU
inmates to participate in programming. All of the inmates interviewed expressed a
desire for more program offerings while in the SHU. An OASIS computer based
course was being offered to inmates in the general population and could be offered
to inmates in the SHU to complete in the SHU Law Library if approved. Most other
facilities offered comparatively limited opportunities.
Special management unit programs. SMUs consist of four program levels and access
to program activities generally increases as the inmate advances through the
program. Progression through Level 1 is based upon the inmate’s compliance with
behavioral expectations as established by staff. Progression through Level 2 and
Three is based upon the inmate demonstrating the “potential for” or positive
“community” interaction. Progression from Level 4 to a general population facility is
“based upon the inmate’s ability to function in a general population setting with
inmates of various group affiliations.” P5217.01.
The project team found that a staff member representing the education department
was assigned to each of the SMU programs and based on documentation reviewed
frequently visited the units. Inmates in Level 1 and Two normally are involved in
individual self-study programming which takes place in the cell. Staff primarily visit
the inmate in front of their cell and communicate through the cell door. Inmates
without a verified high school diploma or GED certificate are required to participate
in a literacy program with the goal of improving their knowledge and skills through
academic activities. At all three SMU programs daily schedules reflect educational
personnel are scheduled to make rounds in the housing unit on a regular basis.
The opportunity for additional inmate interaction increases when the inmate reaches
Level 3 and 4. Most programming for Level 3 and 4 inmates is individualized selfstudy however during the inmates’ out-of-cell time they are allowed to seek
assistance from the staff member. There is no formal group educational
programming offered in the SMU program.
ADX programs. The federal regulations for the control unit program state that the
warden shall assign a member of the education staff to the unit on at least a parttime basis to assist in developing an educational program to fulfill each inmate’s
academic needs. The education staff member is ordinarily a member of the unit
team140.

140

28 CFR 541.46, 28 CFR 544.81.

206

The ADX consists of several different program levels ranging from the general
population, intermediate step of the step-down program, SHU, control unit, and
special security unit. In each phase of the program the level of access to program
activities changes. As an inmate progresses through the different program levels
access to program activities increases. A staff member representing the education
department is assigned to the each program. For those inmates in the control unit,
the majority of programming is conducted in cell and on an individual self-study
basis. Programming is also offered on the closed circuit TV within the inmate’s cell.
For those inmates in the ADX general population units, the majority of the
programming is conducted in-cell on an individual self-study basis and programming
is offered on the closed circuit TV within the inmate’s cell. The ADX recently
constructed five “therapeutic enclosures” in the gymnasium where five inmates in
separate enclosures can participate in a group program. At the time of the on-site
review, the psychology department was using the therapeutic enclosures to conduct
a reentry preparation program. One 90-minute “group” is meeting once a week.
Generally, inmates assigned to all of the ADX phases are provided self-study
materials to be completed in their cell.
Program statements and facility supplemental statements related to program access
in restrictive housing units are aligned with the regulatory requirements identified in
the federal regulations. The review team found that personnel were generally familiar
with the program statements and the program access requirements. Daily activity
schedules were in place, appropriately trained personnel were assigned to adhere to
these requirements and oversight was provided. The assessment team observed that
almost all programming outside of recreation was provided on an individual basis
while the inmate remained in their cell.

Existing potential program space
The current practice observed by the project team reflected access to program
activities for most of the inmates assigned to a segregation unit is provided
individually while the inmate is in the cell and the staff member is outside the cell
door. The federal regulations do not address how program activities are to be
provided or the frequency in which they are to be provided. Outside of a few select
cases or during recreation and the day room time provided for inmates assigned to
an ADX transitional unit, or Level 3 and 4 of the SMU program, almost no formal
group program activity is being provided. As indicated in other sections of the
report, there had been recent initiatives including the building of multiple individual
enclosures to provide some limited “group” programming.
Overall, access to program activities appears to be consistent with meeting the
minimum requirements cited in the federal regulations. Most programming consists
primarily of cell-side services. Staff interviewed during the review process reported

207

that individual in-cell programming was preferred primarily due to safety concerns
and the lack of available small group space.
FINDING: Based on an assessment of the physical plants and on-site reviews, there
does appear to be an opportunity to provide properly scheduled small group
programming for appropriate inmates in existing space at the facilities.
Every facility the project team visited had space within the unit to provide small
group programming for three to six inmates at a time. Based on site observations
and staff interviews, all of the potential group programming space was not
consistently being used during the day.
In the other facilities visited limited small group programming space located
primarily on the front end of the range or unit that was being used primarily for
interviews, hearings, storage or screening and could also be considered available
programming space during select times. The project team identified limited space
available to provide small group programming for appropriately screened inmates at
all of the facilities visited.
The project team is not advocating group programming be provided to all inmates in
the segregation units however some may be considered appropriate. For example,
inmates who are in administrative detention status (nonpunitive) for the sole reason
they have been verified by staff as requiring protection or inmates who have
advanced in their assigned program to a higher level (SMU Level 3 or 4, ADX stepdown, transitional and pre-transfer unit) structured small group programming
should be considered an option as space appears to be available.
FINDING: There are a limited number of potential small group programming
spaces available at each facility.
The location of the potential space was all in an area in the housing unit that was
consistent with maintaining a secure environment for both the inmate and staff.
Though the size of each space varied, the rooms appeared to be sufficient to
accommodate multiple individuals for short periods of time. Currently, most
programming is offered individually while the inmate remains in the cell.
The project team is not advocating all individuals assigned to a restrictive housing
unit be provided access to small group programming, however some inmates with
specific classifications should be considered when proper security precautions are
applied. For example, those inmates who have been assigned to administrative
detention status (nonpunitive) solely for protection and have been verified by Bureau
staff as requiring protection, inmates in Level 3 and 4 of the SMU program, (the
program statement states frequent group counseling, they currently recreate in a
group setting and participate in day room activities as a group), and ADX Step-Down,

208

Transitional and Pre-Transfer inmates who currently recreate and share day room
activities at the same time should all be considered for small group programming.
RECOMMENDATION 7.2: Use of existing small group space should be seriously
considered for inmates who have been properly screened.

Observations on conditions of confinement
FINDING: The general conditions of confinement were found to be consistent with
national regulations and standards. Establishing, maintaining and monitoring the
conditions of confinement appeared to be a routine part of the daily operations of
managing the Bureau facilities visited. Policies were in place, staff were familiar
with the requirements, and post orders and job descriptions had been established
to enforcing compliance in this area.
The additional use of external monitoring and assessment through the Bureau’s ACA
efforts was considered a valuable management tool that provided quality feedback
regarding general conditions of confinement. The project team found that areas of
deficiency that were cited by the Program Review Division (PRD) and/or ACA were
addressed and/or seriously considered by facility management staff. Every facility
visited had received accreditation status by the ACA.
However, some areas of concern identified during this assessment were either not
present at the time of the PRD or ACA review or had not been identified by the PRD
and ACA reviews as concerns.
Triple-celling. The practice of housing three individuals in a cell that is
approximately 85 square feet in size and contains two beds should be
eliminated. This practice is inconsistent with nationally accepted standards
and creates an environment that threatens the safety of both inmates and
staff. The absence of an established written policy that identifies proper
procedures to follow before placing three individuals in a cell with two
beds, including monitoring practices, inmate screening, time limits and
authorization is inconsistent with national best practices.
Low levels of participation in out of cell recreation. Many inmates do not
take part in recreation due to access only at early hours (0500 wake up
calls for recreation), fear of being assaulted, passing room inspection, and
lack of equipment.
Lack of consistency in the conditions of confinement provided for individuals
in the same level. For example, and inmate in Level 3 housed at USP
Allenwood is allowed up to 150 minutes in social telephone calls per
month, two one-hour noncontact social visits per month and indoor and

209

outdoor recreation totaling approximately fifteen hours per week. The
same inmate if housed at USP Lewisburg would be allowed twice the
number of social telephone call minutes per month (up to 300 minutes),
more than twice the amount of social visits per month (five one-hour social
visits) and significantly fewer recreation hours per week (five hours of
outdoor recreation). The differences in frequency, type, and the duration of
conditions of confinement for an inmate classified at the same level in the
same program while housed at a different facility presents concerns
regarding the integrity and design of the level system.
Similarity in conditions of confinement for inmates in administrative
detention and disciplinary segregation status. Administrative detention is
identified as a nonpunitive status while disciplinary segregation is
identified as a punitive status. However, the inmates are assigned to the
same housing unit, inmate movement procedures including application of
restraints are the same, the frequency of recreation is the same, telephone
access is the same and visits at all but two facilities visited were the same.
With the exception of minor differences in personal property allowed and
in-cell programming opportunities the day-to-day conditions of
confinement were not much different. Considering one status is
nonpunitive and includes in part individuals that are in administrative
detention status strictly as a result of being verified as requiring protection
serious consideration should be placed on reevaluating the day-to-day
condition of confinement authorized for individuals in a nonpunitive
status.

Program reviews
The Program Review Division (PRD) is directly responsible for overseeing Bureauwide performance reviews that are designed in part to examine facility operational
compliance levels with applicable laws, rules, regulations, and policies. In addition,
the division examines the adequacy of controls, the efficiency of operations, and
effectiveness in achieving the desired program results. PRD is also responsible for
ensuring an analysis of specific program performance patterns is completed and
serves as the Bureau’s liaison with most external audit agencies.
To provide personnel with consistent direction and guidance a comprehensive
program statement141 has been established that identifies the purpose, requirements
Program Statement 1210.23, Management Control and Program Review Manual, dated
8/21/2002.

141

210

and procedures associated with meeting review responsibilities. As a result, local
operational reviews are required to be conducted by facility personnel at least once
every 10 to 14 months. These operational reviews are commonly considered facility
self-audits. In addition, a comprehensive program review managed by central office
is conducted and completed by external Bureau personnel at least once every three
years.
Throughout the review process the project team examined the operational and
monitoring practices and found that both program and operational reviews were
normally being completed on a scheduled basis. Facility personnel were familiar with
the process, documentation was available and dedicated personnel were assigned to
each facility that were held accountable for monitoring local compliance levels.

Program review observations
Policy and review schedules have been established that are consistent with federal
requirements and have been designed to ensure that both the operational and
program reviews are conducted in a manner that best addresses the requirements of
the Bureau. The division normally schedules the external performance reviews based
on specific areas being analyzed and the location of the facilities. For example, either
a correctional services or facilities management program performance review is
scheduled for a prison complex and will include an examination of several subject
areas to be reviewed within the complex. In most facilities specific guidelines have
been established to ensure certain policies and procedures are reviewed.
At the Federal Correctional Complex Tucson a facilities management review was
conducted on February 12-14, 2013, which included the Federal Correctional
Institution, USP, and the satellite work camp. Guidelines were established by PRD
that identified specific policies to be reviewed including policies impacting both
general population and SHUs at each of the facilities. The scope of coverage and
format used during the Tucson review was typical of what the project team observed
in evaluating other PRD reports.
A typical performance review normally does not focus strictly on a particular
housing type or specific subject matter, but analyzes and provides a report that
identifies the overall complex performance findings and trends. Program reviews are
not designed to examine each requirement identified in every policy in every housing
type. However, program review guidelines have been established to ensure specific
protective policies, conditions and procedures that are reviewed, include general
population, SHUs and SMUs. Additionally, to provide further observation of SHU
operations, each institution’s program review (all disciplines) includes the reviewerin-charge conducting a comprehensive tour of the SHU, promptly reporting findings
as appropriate, and including the findings in the final report.

211

For example, inmate access to recreation may be examined complex-wide and by
policy a sample number of general population, SHU and SMU files will be required to
be reviewed. If there are specific concerns identified in this area those concerns are
generally noted in a subsection entitled “general comments” or “deficiencies” in the
PRD report. Bureau policy states that the reviews include an examination of a select
percent or number of general population, SHU and/or SMU files. What is excluded in
the review is the same level of examination of conditions for the ADX. The Program
Review Division at the time of the project team’s assessment did not have specific
monitoring requirements for ADX policies as was the case for the SHUs and SMUs.
Operational reviews and performance reviews were being conducted at the ADX as
evidenced by documents provided, however specific guidelines ensuring select areas
were reviewed had not been established.
FINDING: Specific conditions of confinement and protective policy compliance
levels are monitored locally by facility personnel and in some areas by central
office personnel that are not assigned to the Program Review Division.
The lack of these specific guidelines for the ADX creates the potential for possible
omissions in examining areas that require ongoing review.
RECOMMENDATION 7.3: Guidelines that identify specific conditions of
confinement and protection policies consistent with applicable federal regulations
and national standards should be developed and included as part of the Program
Review Division performance review process142.

Reporting
In addition to a review of the specific policies and conditions examined by the
Program Review Division, an assessment of the performance reports was completed.
Program Review Reports from 2011 to 2013 were reviewed as well as Program
Statement 1210.03, Management Control and Program Review Manual. Each report
identified the dates the review was completed, overall rating, response requirements,
if applicable, reviewer assurance statement, background information, general
comments (concerns, if any) and deficiencies.
In examining the documents from 2011 to 2013, there were several program review
reports that identified concerns in the restrictive housing units that required a
response from facility personnel. Some of the concerns noted in the reports included
such areas as: inadequate cell search procedures; consistent documentation of staff
The Bureau has reported that ADX-specific program review guidelines were developed and
became effective June 13, 2014.

142

212

rounds; consistent documentation of inmate access to services; proper application of
restraints; and below average sanitation conditions. The positive aspect of the
reviews was that the Program Review Division was identifying concerns during their
review and requiring facility personnel to provide a written response and corrective
action plan for each.
A few of the concerns cited in the 2011 – 2013 reports were consistent with the
review team’s observations in 2014. Policies were in place that required activities and
service delivery to be documented when provided, however observations reflected
there were occasions when incomplete documentation was being provided. The
frequency of the omissions was few. The addition of the new automated software
program designed to document inmate service delivery was in use at several of the
facilities. The system allowed routine activities to be recorded electronically such as
meals delivered, recreation, and shower access. This procedural upgrade was viewed
by staff and the project team as a significant improvement over the manual system.
Staff were familiar with the automated system and procedures were in place to
ensure select service activities and conditions were documented and inputted into
the system.

Operational review observations
The Program Review Division (PRD) is an essential part of the Bureau tasked with the
responsibility of overseeing Bureau-wide performance reviews that are designed in
part to identify facility operational compliance levels with applicable laws, rules,
regulations, and policies. Policies and procedures have been developed, qualified
staff are assigned and site reviews are conducted on a regular basis consistent with
applicable regulations.
FINDING: Both program and operational reviews served a vital role in the overall
positive operations of the Bureau’s facilities.
Appropriate adjustments and revisions were being implemented to refine the system
and Bureau personnel reported additional modifications are being considered. In line
with the approach of continued growth and refinement, the project team has
identified a few recommendations in this area:
RECOMMENDATION 7.4: Establish a separate program performance review for the
USP Florence ADX that includes in part a comprehensive evaluation of specific
policies and procedures that are unique to the ADX.
The current review includes three facilities: Federal Correctional Institution; USP
Florence High, and USP Florence ADX with a separate satellite work camp. The
benefit of providing a separate review with specific guidelines focused on the unique

213

operations of the ADX can only assist in meeting the established mission of the
Bureau.
Specific guidelines to be reviewed have not been established for the ADX and should
be considered. These guidelines should reflect federal regulations, Bureau policies
and national standards. Based on the uniqueness of the program, including the
ADX’s operations and conditions of confinement a separate review may provide for a
more detailed assessment of the facilities overall compliance levels with established
policies and regulations.
RECOMMENDATION 7.5: Enhance the external oversight of the local operational
reviews that are being conducted no less than on an annual basis.
External program performance reviews appear to be identifying a number of issues
that should be addressed and corrected as a result of the local operational reviews
that are conducted by the facility. Written responses are required and provided on
concerns identified in the operational review, however reported corrective action may
not always resolve the deficiency. Exploring the expanded use of announced or
unannounced follow-ups and or requiring repeated assurances over a period of time
regarding the effectiveness of the correction action should be considered.
RECOMMENDATION 7.6: Reassess the performance review rating system.
A number of program reviews resulted in an overall rating that may not have been
consistent with the findings or definitions identified in the applicable program
statement. Based on the project team’s observations and review of documents, the
team did not identify any facilities that warranted an overall rating of “deficient” or
“at-risk,” and none of the reports indicated the same. Re-evaluating the overall
performance rating designation and ensuring an accurate rating is provided essential
to solidifying the integrity of the review process.

214

Chapter 8: Summary of conclusions
The following summarizes the major findings and recommendations that were
presented in the previous chapters of this report.

Extent of segregation
The Bureau experienced an increase in the use of restrictive housing that began in
2009 by rapidly implementing its existing SMU program. Data indicates that the ADX
population has remained quite stable over the same time period.
Only a limited amount of information is known about the SHU population in terms of
historic trends as the Bureau did not retain data on the number and type of inmates
placed in the SHU status - protective custody, investigation and disciplinary
segregation. More recent data indicates that the size of the SHU population has been
steadily declining since 2011. The Bureau was able to provide detailed SHU
population statistics beginning in February 2013. At that time, the count was 10,262
in over 100 facilities and has steadily declined since then reaching 8,939 by June
2014. This is a significant reduction from the self-reported count of the SHU
population of over 13,000 in 2011.
The decline can be attributed to several factors as reported by staff at all levels and
confirmed and observed by the CNA project team. The process of conducting weekly
reviews of SHU placements has been in place for some time and is a key factor in the
decline. These reviews are held in a formalized setting with the entire management
team of the facility participating in a review of the status of each inmate in SHU. The
review includes an assessment and evaluation of all placement options including
release from SHU. Staff noted that in the last three years the process has been reemphasized as a means to ensure that inmates are managed in the least restrictive
setting given their risk to the system.
As of November 2013, approximately 5 percent of the entire Bureau’s inmate
population was being housed in one of these three restrictive housing populations
with the vast majority in the SHU status. From a state prison population perspective,
this number is neither high nor low, but more in the middle.

215

However, this review indicates that there are excellent opportunities to significantly
lower the SHU and SMU populations by adopting the recommendations outlined in
this report. Specifically, we estimate that the current SMU and SHU populations can
be lowered significantly by adopting the following approaches:
As of 2014, there were 4,252 inmates in some form of investigation status. This is
partially a result of the lack of a policy requirement that investigation be limited to a
specified maximum time period, absence a rationale for continuing the inmate on
investigation beyond the specified time period. Presently, investigations can linger
for months without resolution:
Related to the investigation time period, once the investigation is
completed and the case heard by the DHO, the sanction issued is not
retroactive to the original date of placement in the SHU. As a result, if the
investigation continues for 180 day and the inmate is found guilty of the
infraction and issued disciplinary time of 30 days, the sanction and
disposition is in reality 210 days as it is not retroactive to the original
placement date. Giving the option for credit for time served will reduce the
bed days in segregation significantly given the size of the investigation
population.
It was reported that in November 2013, there were 1,437 protection cases
in special housing with an additional 172 housed in special housing but
pending placement back into the general population. As noted in the
report, many states have removed protection cases from administrative
segregation and created specialized housing that is secure but replicates
the conditions of confinement and programming of a general population
unit. Not all of those in protection status would qualify for such placement
as those whose protection is related to their own behavior could be
excluded as they represent an ongoing risk to others. This option would,
however, remove a significant number of offenders from the SHU. The
Bureau is already utilizing this option to a small degree with the
establishment of the RHU in Oakdale, Louisiana.
The Bureau should continue to utilize a process of careful review of all
referrals for placement to the SMU program. As noted in this report a
recent snapshot of the review process indicates that 22 percent of the
referrals initiated at the institutional level have been denied placement.
This recent level of denials are an increase over prior measured levels and
will assist in controlling any future growth as the SMU beds are reserved
for only those who require such placement.
It is recommended that the time period for completion of the SMU
program be reduced from the present 18-24 months to 12 months and also
the four levels be compressed to three levels by combining Level 3 and

216

Level 4 and allowing more differentiation between the condition of
confinement between the levels.
Implementation of these recommendations will significantly reduce the present
restrictive housing population, while ensuring that mechanisms are still in place to
maintain the safety and security of the institutions.

Conditions of confinement
In general, we found the housing units, cells, and furnishings provided in the cells to
be adequate to accommodate either single or double celling of the ADX, SMU and
SHU populations. Lighting, ventilation, access to showers and recreation areas were
also nonproblematic. Some concern is noted with recreation, as a significant portion
of the restrictive population decline to use scheduled recreation periods. Interviews
with inmates with few exceptions validated our on-site observations. The quality of
food provided to inmates was rarely cited as an issue by inmates.
Virtually all of the inmates in SMU and SHU are double-celled. Inmates do have
regular access to showers and recreation but the vast majority of their time is spent
in their cells with their cellmates (with the exception of the Level 4 SMU inmates).
Security staff, case managers, medical and mental health staff make regular visits to
the housing units. However, direct out-of-cell contact with staff is very limited. When
such contacts occur, they are brief and in front of the cells doors.
Establishing, maintaining and monitoring the conditions of confinement appeared to
be a routine part of the daily operations of managing the Bureau facilities visited.
Policies were in place, staff are familiar with the requirements and post and job
descriptions had been established to help in enforcing compliance.
A typical week for the inmate in restrictive housing consists of a few hours of
recreation, limited visits, and no phone calls (primarily due to the restriction issued
by the DHO as part of the disciplinary hearing process), and no participation in out
of cell treatment programs. Based on the information obtained by the prisoner
interviews, although inmates are afforded opportunities for out of cell activities,
portions of the SHU and SMU populations never leave their cells each week.
Administrative detention is identified as a nonpunitive status while disciplinary
segregation is identified as a punitive status. However, the inmates are assigned to
the same housing unit, inmate movement procedures including application of
restraints are the same, the frequency of recreation is the same, telephone access is
the same and visits at all but two facilities visited were the same. With the exception
of minor differences in personal property allowed, number and duration of phone
calls, and in-cell programming opportunities the day-to-day conditions of

217

confinement were basically equivalent. Considering one status is “nonpunitive” and
includes individuals who are in administrative detention status for protection,
serious consideration should be placed on reevaluating the day-to-day condition of
confinement authorized for individuals in a nonpunitive status.
The Bureau needs to increase the amount of out-of-cell activities and continue its
ongoing efforts to reduce the overall size of the restrictive housing population.

Impact of the “separatee” status
The SHU situation is complicated by the fact that a high percentage of inmates
residing in SHUs have requested protection from other inmates, often due to gang
related issues. Many inmates in the system have what are termed as “separatees.” A
separatee is an inmate who for security and safety concerns cannot be housed with
specific inmates in housing units where they may congregate with one another. The
number of inmates that have separatee issues is significant and impacts inmate
management in the agency. For example, the warden at USP Lewisburg advised the
assessment team that of 748 inmates at that facility, 334 of them had separatee
issues. Some individual inmates had as many as 14 separatees. Keeping in mind that
these inmates can’t congregate with one another, bed management in the general
population and segregation units is very complicated.
In the SMU at USP Allenwood, where inmates in Level 3 and Level 4 are allowed to
congregate, the lieutenants and unit management staff have to carefully place
inmates according to their gang or group designation.
The gang issue in the USPs that was assessed impacts the management of the SHU as
it creates separatee issues and likely increases the number of inmates requesting
protective custody, resulting in them being housed in restrictive housing. This
observation was confirmed through interviews with staff and inmates who indicated
that many of the segregated inmates were unable to live in general population
because of the influence, harassment, intimidation, and threats of violence from
members of gangs and “cars” (Prison based gangs in the facilities).
Difficulties in the management of the gang problem exacerbate the protective
custody problem, thus causing a high incidence of inmates leaving general
population by their own request and being placed in segregation units after being
threatened in general population. There is no simple solution for this issue but its
impact is significant throughout the Bureau. Through its investigation units and its
gang management unit the Bureau is aggressively attacking the problem. The wide
spread presence of the gang factions, especially in the United State Penitentiaries,
significantly impact the presence and size of the restrictive housing units.

218

Investigation, protective custody, and
pending transfer inmates in SHU
The SHU population needs to be understood in terms of the major statuses that one
can be assigned. Those assigned for disciplinary sanctions purposes are inmates who
have been found guilty by a well-functioning disciplinary process for serious
violations of prison rules and conduct. However, the much larger number of inmates
held administrative detention for investigation, protective custody, and pending
Bureau actions under restrictive conditions of confinement for extensive periods of
time deserves further discussion and scrutiny.
The largest number of inmates being held in SHU are either there for investigation or
protective custody. Although the Bureau states in its policy that both statuses are
“nonpunitive,” this is clearly not the case when examining the units from an
operational and program standpoint. Inmates in protective custody and investigative
status are housed in SHU experience the same living conditions as those placed in
what is an explicitly punitive environment. As of 2014, there were 4,252 inmates in
some form of investigation status and another 1,361 inmates in some form or
protective custody. Further another 1,802 inmates in the punitive SHUs awaiting
some action by the BOP to either have a hearing for discipline, or transfer to ADX or
SMU status or some other location. Thus, over 80 percent of the SHU inmates have
not been found guilty of disciplinary conduct.
As noted previously, investigations should be completed in a more timely manner. It
is suggested that a limit of 45 days be established, with provisions for an extension
of this time limit when circumstances exist that require additional time.
Also noted previously, there needs to be a clear difference in the conditions of
confinement from those in punitive segregation, those in investigation or protection
status, and those simply awaiting an administrative action of some kind.

Inmates awaiting administrative actions
Somewhat related to inmates under investigation status are those awaiting some
final action by the Bureau’s central office. This population is excessive and needs to
be reduced. The reasons for delays in transferring inmates from SHU either to
SMU/ADX or back to the general population is complicated by lack of space and the
number of separates. Little can be done about the number of separatees but, as
noted later, much can be done about the lack of bed capacity.

219

SHU inmates who have pending transfer actions should be transferred out of SHU
within ten working days once their status for placement in SHU has expired.

Protective custody inmates in special
housing units
Housing protective custody inmates in a punitive setting clearly contradicts best
correctional practices. States that have been found to do so have suffered from ongoing litigation and court ordered reforms. Many states have developed specialized
protective custody housing units where inmates are provided the same level of basic
privileges and access to rehabilitative programs and even work assignments as
general population inmates. The one exception is that they are assigned to specially
designated housing units within a prison or a specialized facility.
It was reported that in November 2013 there were 1,437 protection cases in special
housing with an additional 172 housed in special housing but pending placement
back into the general population. In October 2013, the Bureau issued a memorandum
that provided field staff with procedures and criteria for placement of inmates in the
reintegration housing unit (RHU) located at the Federal Correctional Complex at
Oakdale, Louisiana.
The target population for the RHU consists of male inmates who “consistently refuse
to enter general population at multiple locations” and those who have been
designated through the classification process as protective custody.
The question then becomes ‘what is the appropriate manner to manage inmates who
have verified protection needs?’ such inmates are presently housed in administrative
detention, which is identified by Bureau policy as a nonpunitive status. However, the
inmates who are verified protection cases are assigned to the same housing unit as
punitive segregation inmates, inmate movement procedures including application of
restraints are the same, the frequency of recreation is the same, telephone access is
the same and visits at all but two facilities reviewed were the same. With the
exception of minor differences in personal property allowed and in-cell programming
opportunities, the day-to-day conditions of confinement were not much different.
Considering one status is nonpunitive and includes in part individuals that are in
administrative detention status strictly as a result of being verified as requiring
protection, serious consideration should be placed on reevaluating the day-to-day
condition of confinement authorized for individuals who have been verified as
needing protection.
This is a complex issue within the Bureau due to extensive presence of security
threat group members even in the SHU. Many of those have verified protection as a

220

direct result of their prior involvement with a security threat group. But many are
also victims or potential victims who need protection. These protection needs should
be provided but in a more normalized setting than what is presently provided in the
SHU.
In numerous state departments of correction, the conditions of confinement for
protection cases have been altered to parallel that provided to other general
population inmates. In these instances inmates are housed in SHU or similar units
while the claim for protection is being investigated. Once the need for protection is
verified they are moved to a separate unit that provides conditions of confinement
that are similar if not identical to that provided to general population inmates. These
units operate separate from other general population units and afford the inmate the
ability to function in a normalized prison environment while ensuring they are
protected. Kentucky has done this successfully at the Eddyville facility while Ohio
also operates units that replicate general population conditions for protective
custody inmates. It should be noted that the Bureau is moving in this same direction
with the establishment of the RHU in Oakdale, Louisiana.
The Bureau should to establish nonpunitive protective custody housing units that
have equivalent levels of programs and privileges as general population inmates.

The special management unit program
The Bureau has developed a much-needed step-down program for inmates removed
from the general population. The criteria for such removals are objective and
reasonable (only inmates who have committed serious acts of violence or pose a
threat to the safety and welfare of the other 95 percent of the inmate general
population).
The conditions of confinement for SMU inmates are more restrictive than for general
population inmates. SMU inmates are expected to complete the four-level SMU
program in 18 to 24 months, at which time they may be re-designated (transferred)
to an appropriate facility. At the time of the initiation of this review, SMU programs
were functioning at USP Lewisburg, USP Allenwood, and USP Florence. During the
course of the review the Bureau began the phasing out of the SMU at USP Florence
through the gradual transfer of inmates to USP Lewisburg.
Documentation provided by the DSCC indicated that a total of 5,435 inmates had
been submitted for SMU placement from January 2009 to June 6, 2014. Of the
inmates referred, 1,057 (19.5 percent) had been denied placement by the DSCC. This
does not include denials/rejections that occurred at the regional offices as those
records were not readily available.

221

A more recent picture of the validity of the referral and review process was obtained
by reviewing the number of referrals by month that have been submitted since
January 2013 through February 2014 and the number of these by month that have
been rejected by either the DSCC or the regional director. Table 21 summarizes these
decisions.
This more recent snapshot of the review process indicates that 22 percent of the
referrals have been denied placement.
What is uncertain is the need for four levels and overall time to complete all four
levels before an inmate can be returned to the general population. Currently, the
total minimal amount of time required to complete all four levels is approximately
18 months. We say “approximately,” because in the Bureau the levels are designated
by policy as six to eight months and two to four months in each level.
As noted in chapter 2, most of the reviewed state prison step down programs allow
for only three phases of confinement, which tend to last 90 days each. Many have
implemented a program that has a minimum stay exceeding 12 months. Nor is there
any research basis for extending the minimum amount of time in segregation for
fully compliant inmates beyond nine months or for that matter any length of time.
It was also discovered that SMU inmates who have completed a particular phase of
the program are not promptly transferred to the next program phase location. This
was a problem at USP Lewisburg for inmates who had completed Level 2 and were
scheduled to advance to Level 3. These backlogs in inmates awaiting transfer to the
next program level negate the intent of the program design and decrease the
motivation to change behavior.
Finally, relative to program content, much of the program elements of the status
(with the exception of Level 4) consist of in-cell written assignments that are not
testing the prisoner’s ability to return to the general population. In order for the
program to be more effective, the Bureau should examine whether to develop a range
of structured out-of-cell contacts and activities beginning at Level 2 and expanding at
Level 3/4.
As a result of this review, it is recommended that the Bureau modify the current SMU
program to have a stay of 12 months and only three levels and increase the amount
of structured out of cell activities for SMU inmates especially for those in Level 2.

Mental health services
In general, each of the restrictive housing units is providing mental health services to
the inmates assigned to the units. Mental health staff routinely visit the housing
units as required by Bureau policies. There are sufficient nonpsychiatric mental

222

health staff assigned to the units to provide sufficient mental health services.
Monthly mental health records are being updated on a regular basis as required by
Bureau policy. Inmates in special housing being monitored for having suicidal
tendencies in specially designed cells under direct supervision is consistent with best
correctional practices.
However, our review of the mental services found a number of areas that the Bureau
needs to significantly improve upon. The issues raised can be separated into three
separate subcategories:
1. Proper mental health diagnosis
2. Effective treatment
3. Sufficient psychiatric staffing
Relative to diagnosis, based on the Bureau’s four level mental health rating system,
only about 10 percent of the segregated populations were assigned to Care Levels 2,
suggesting minimal need for mental health care treatment beyond limited
psychotropic medications. However, our review of randomly selected Care Level 1
cases by two experienced psychiatrists found a number of inmates exhibiting
significant mental health symptoms which suggest their care level should be
increased to Care Level 2 or higher. A contributing factor to this issue is that very
few of the monthly mental health assessments occur in private settings on a face-toface basis.
The majority (between 68 and 72 percent) of restrictive housing inmates have been
classified by the Bureau as healthy or only having simple chronic medical care, and
not requiring any specialized medical treatment (see table 11). These proportions are
virtually identical to the other Bureau inmates (also 72 percent). Of those requiring
medical care, the level of care is mostly at the lowest threshold (between 26 and 30
percent at “Care Level 2 –Stable, Chronic Care”).
Somewhat surprisingly, relative to mental health care level, the proportions in need
of care or treatment are even lower with the vast majority assigned to Care Level 1.
The ADX has the highest proportion of Mental Health Care Levels 2 and 3 but they
only represent ten percent of the entire ADX population. These proportions are
comparable to the mental health care levels of the nonrestrictive populations.
Based on the assessment of the project psychiatrists the review further indicated
that the treatment being offered by the Bureau was insufficient or inappropriate in
over half of the cases. In particular, there was little evidence of structured out of cell
treatment services occurring for those in need of services beyond medication.
Much of the “treatment” being provided to the inmates takes the form of psychiatric
medications that can only be prescribed by a psychiatrist or a psychiatric nurse

223

practitioner whom is familiar with the patient’s symptoms and prior mental health
history. There is a significant lack of coordination between the psychology staff who
cannot prescribe these medications and psychiatric staff who are prescribing the
psychotropic medications. A significant number of the inmates which the review
found had been misdiagnosed and/or were receiving inadequate treatment should be
transferred from the SMU or SHU to a comprehensive mental health program
directed by a psychiatrist. It is acknowledged that mental health clinicians frequently
disagree on diagnoses and that doctoral level psychologists are most certainly
capable of directing a mental health program.
The shortage of psychiatric staff in Bureau facilities leads to numerous problems in
both proper diagnosis and treatment, particularly for the seriously mentally ill
inmates at the large segregation units. For example, at most facilities, the prescribing
physician was not a psychiatrist, which further added to the problems of
coordination with the psychology staff.
As a result the following recommendations are submitted:
All inmates should be seen in a private setting for a comprehensive mental
health evaluation prior to placement in any segregated setting.
A complete re-evaluation of the mental health record should be performed by
psychology and psychiatry staff every 30 days which would require a face to
face interview in a private setting.
Additional full time psychiatric staff are needed at the major restrictive
housing units (in particular USP Lewisburg).
A complete review of all inmates assigned to ADX, SMU and SHU should be
completed by the Bureau to identify all inmates who should be transferred to
a secure mental health program similar to the one being developed at the USP
Atlanta and USP Allenwood.

Crowding and segregation
As has been noted earlier, the ability of the Bureau to place inmates who are being
returned to general population from restrictive housing and housing inmates within
the SMU and SHU’s which are largely double-celled is often hampered by the number
of separatees that each inmate may have. As noted in chapter 3, virtually all of the
ADX and SMU, and 2/3rds of the SHU inmates have at least one other inmate they
must be separated from. Even within the nonsegregated population, 40 percent have
separation orders. Many of these separation orders are related to conflicting gang
affiliations.

224

In general, inmates with separation orders cannot be transferred to general
population in facilities where separatees are currently located. This in turn produces
delays in the transfer of inmates from segregation status back to the general
population.
A final complication is the level of crowding that permeates the Bureau’s facilities. As
of 2013, the Bureau was operating at 137 percent of its rated capacity and even
higher at its high (154 percent) and medium (144 percent) security facilities. Only
two options exist to reduce these exceedingly high crowding levels – add more and
expensive high security general population beds or reduce the current Bureau
population.
These two factors (high numbers of separatees and prison crowding) contribute to
the number of inmates in SHU who are awaiting transfer back to the general
population.

Due process
Sanctions issued by the DHO were found to be consistently lower than the allowed
level for each of the severity levels. For example, offenses in the Greatest Severity
Level carry a maximum time in disciplinary segregation of 12 months. Of all the case
files reviewed, not one included a sanction of 12 months in restrictive housing. For
100 level offenses such as a weapon, the typical sanction included 30 – 45 days
disciplinary segregation and in many cases was 15 – 20 days segregation. At USP
Victorville, the typical sanction for an assault or a weapon offense was 30 or 60 days
disciplinary segregation. At USP Tucson, a typical sanction of possession of a weapon
was 45 days while an assault resulted in 60 – 90 days segregation. At USP Coleman, a
weapons offense resulted in 30 days disciplinary segregation.
It was apparent from interviews and case file reviews that the DHO’s have made
extensive use of the loss of access to basic privileges (visits, telephone, and
commissary) for extended periods of time as a disciplinary sanction. The use of these
sanctions is an outgrowth of the objective to find alternative sanctions to the
placement of the offender in disciplinary segregation.
The Bureau DHO’s use the restriction of privileges such as visiting, commissary, and
telephone extensively as a sanction for offenses within all severity levels. Almost
every sanction issued by the DHO at each of the facilities reviewed included a
restriction of one or all of the above privileges. At USP Coleman, weapons offenses
typically resulted in a sanction of 41 days loss of Good Conduct Time, 15 days
placement in disciplinary segregation, 180 days restriction on visits, commissary and
telephones. Similar sanctions with some variance in the amount of segregation time
were found in virtually all the institutions reviewed. In the cases reviewed, it was a

225

normal practice for the DHO to use the restriction of 180 days loss of Commissary,
Telephone, and Visits.
The extensive use of restriction of privileges resulted in the accumulation of loss of
privileges over an extended period. It was not unusual to find inmates who had lost
visiting privileges for in excess of one or two years. During one interview, an inmate
reported that he had lost visiting and phone access for a period of seven years. This
was confirmed through review of disciplinary records provided by the DHO. File
reviews confirmed that similarly lengthy periods of restrictive privileges was
common place in the system.
Sanctions issued by the DHO are effective the date of the hearing and are not made
retroactive to the date of the incident report or the date of referral by the UDC to the
DHO. There is no time limit in policy that governs how quickly the DHO must hold a
hearing after the investigation/UDC process is complete. Scheduling of hearings is at
the discretion of the DHO.
The lack of time parameters for completion of disciplinary hearings results in
substantial variation among facilities in the amount of time served in restrictive
housing for similar offenses, and can result in disproportionately long sanctions.
During our review, we found institutions that had established internal informal
requirements that the hearing be completed within 14 days of receipt by the DHO
(USP Coleman). Another facility had established a guideline of 20 days. In the course
of this review, instances of hearings held more than 30 days after the incident date
were not unusual. Longer delays occurred when cases were continued awaiting the
results of drug tests, when cases where referred for further investigation, and when
cases were continued for further information. The result was that, in many cases, a
sanction of 30 days segregation in actuality became a sanction of 90 days
segregation or longer since the time served in segregation was not credited to the
sanction or made retroactive to the date of original confinement in segregation.
By comparison most state systems have specific time requirements for conducting
the hearing or issuing a continuance based on need for additional information
(investigation, availability of witnesses, etc.). Ohio policy requires that inmates
charged with a rule violation must be scheduled for a hearing as soon as practicable
but no later than seven days, excluding weekends and holidays, after the alleged
violation is reported, unless the hearing is prevented by exceptional circumstances,
unavoidable delays, or reasonable postponements. The exceptional circumstances,
unavoidable delays, or reasonable postponements must be documented.

226

Reentry
There is no formal Bureau-wide reentry preparedness program specific to restrictive
housing. Each facility visited seemed to have their own unique programming that
they tried to offer within the confines of the restrictive housing unit. Most staff
interviewed did not appear to recognize the need for programming beyond self-help
packet programs and no facility was able to provide data on the number of inmates
actually being released from restrictive housing. The mindset that it is okay and
preferable to discharge from a SHU needs to change.
The issue of inmates releasing from restrictive housing with little or no reentry
preparation is significant. The magnitude of the issue is not fully known since no
data was available on the frequency of this practice and limited research on the
effectiveness of reentry programs. One staffer reported that they do not need to
track that information since their goal is to minimize the time an inmate spends in
restrictive housing. While the goal of shortening the time in restrictive housing is
correct and will help this situation, it ignores the reality that inmates are still
releasing from restrictive housing.
Facilities do not provide step-down planning to transition an inmate from restrictive
housing to general population and subsequently to their eventual release. The
prevailing practice is to keep inmates in restrictive housing until such time as they
discharge to an RRC or directly to the community.
With the exception of assigned completion of self-study activities related to reentry,
inmates transition from often extended stays in restrictive housing to an abrupt
release to general population or the community without any meaningful step down
programming. Many of the staff interviewed found this completely acceptable and
indicated that it was preferable to release inmates from the SHU, SMU or ADX rather
than first transitioning to general population due to the risk of violence to the
general population. While this may be a sound decision for institutional security, it
can hardly be in the interests of the communities where these inmates are being
released. Inmates spending extended periods of time in close confinement with little
social interaction or skill-building programming are seriously unprepared for reentry
and re-socialization.
As can be seen from these examples, inmates in Bureau restrictive housing have very
limited access to reentry programming. Services are generally limited to providing
basic information on identification and benefit issues, and referrals to community
programs and services. This stands in stark contrast to range and depth of reentry
programming provided to Bureau general population inmates. It is the restrictive
housing population that is in the most need of programs and poses the greatest
potential risk in their transition back to the community.

227

Information system needs
The report has noted the difficulties in tracking the number and movement of
inmates within the restrictive housing units. For the SMU population, it was not
possible to track which level of the SMU program the inmate was assigned to and
how long he had been assigned to that level. The lack of such data was even more
pronounced for the much larger SHU population. If the Bureau is to develop more
effective forms of restrictive housing intervention, it will need to significantly
enhance its information data system and its capabilities to effectively analyze trends
within this population.

Summary of findings and recommendations
Statistical analysis of restrictive housing
FINDING: There are over 2,000 inmates in restrictive housing who will be released
within a year, which suggests the need for reentry services. The differences in
sentence lengths, time served, and time left to serve are especially pronounced
among the ADX and SMU prisoners.
FINDING: The majority of inmates in restrictive housing do not require specialized
medical treatment and are virtually identical to the other Bureau inmates. Somewhat
surprisingly, relative to mental health care level, the proportions reported in need of
care or treatment are lower than and comparable to the mental health care levels of
the inmates in general population housing.
FINDING: SMU and ADX inmates have lengthy disciplinary records that include
histories of repeated institutional violence and other types of serious misconduct. It
appears that these rates of misconduct significantly decline while in restrictive
housing status and after release from restrictive housing.
FINDING: The vast majority of ADX and SMU inmates released in 2011 were not
returned to restrictive housing status, although most incurred another disciplinary
report within two years. The average number of disciplinary reports declined sharply
after released from either ADX or SMU.
FINDING: The Bureau’s information system cannot directly provide the most basic
data on the number of inmates admitted and released from restrictive housing, the
current restrictive housing population, and the status of inmates assigned to
restrictive housing.

228

RECOMMENDATION: The Bureau information system needs to be modified so it will
directly measure the movement of restrictive housing placements and removals as
well as the basis for and current status of each placement.
FINDING: The use of four SMU levels that last at least 18 months is not consistent
with other, state programs, which typically require only two or three levels and a
minimum period of confinement ranging from 9 to 12 months.
FINDING: There was no strong or consistent relationship between time in SMU and
prison recidivism rates, although inmates with the longest placement in SMU had
higher rates of return to restrictive housing.
RECOMMENDATION: The current required minimum length of stay for SMU inmates
should be reduced from 18 months to 12 months and the number of levels
consolidated from four to three.

Due process
FINDING: The DHOs and others involved in the disciplinary process were well versed
in their duties. All appropriate notices and procedures were followed, and inmates
responded respectfully to the process and the decision.
FINDING: Bureau disciplinary processes and procedures provide substantial and
redundant assurances for due process compliance.
FINDING: The lack of time parameters for completion of disciplinary hearings results
in substantial variation between facilities in the amount of time served in restrictive
housing for similar offenses, and can result in disproportionately long sanctions.
FINDING: Sanctions issued by the DHO become effective on the date of the hearing
and are not made retroactive to the date of the incident report or the date of referral
by the UDC to the DHO.
RECOMMENDATION 4.1: Establish reasonable time frames in which the hearing
must be scheduled, while permitting reasonable continuances when waiting for
investigation reports, drug tests, etc.
RECOMMENDATION 4.2: Establish by policy that a sanction of restrictive housing
time should be issued retroactive to the date of the original admission, providing
credit for time served.
RECOMMENDATION 4.3: Establish a system for monitoring patterns and trends in
the use of disciplinary sanctions among Bureau facilities.
FINDING: A disproportionate number of inmates are being housed in the SHUs based
on protection claims.

229

FINDING: The application of the same security and operational restrictions to the
protective custody population as those applied in administrative segregation is
contrary to national accepted practices.
RECOMMENDATION 4.4: Expand housing alternatives for inmates in verified
protective custody status to provide levels of programs and privileges equivalent
to those provided for the general population.
FINDING: The conduct of weekly reviews of SHU placements in a formalized setting
with the entire management team of the facility is an exemplary practice that ensures
ongoing review of the status of inmates in SHU and evaluation of placement options.
RECOMMENDATION 4.5: Establish a policy standard requiring private, face-to-face
interviews for the restrictive housing review.
FINDING: The SRO reviews at some of the facilities reviewed appeared perfunctory
and lacked substance in contact and purpose.
FINDING: The review of randomly selected inmate records found omissions in the
maintenance and content of inmate records documenting the placement rationale in
the SHU.
RECOMMENDATION 4.6: Develop and deploy an electronic inmate record system
to document SHU placement decisions.
FINDING: The requirements that are contained in the policy and procedures that
govern the placement and review of inmates housed in the SHU are consistent with
national standards and afford inmates in these units due process in relation to their
placement in the units.
FINDING: The Bureau has established policies and procedures that afford due
process protections to inmates in the referral and assignment to SMU.
FINDING: The significant level of SMU placement request denials indicates that the
review process provides a valid and independent assessment beyond the institution
level of the necessity of the SMU placement.
FINDING: Scheduled SMU conditions reviews were in some cases not conducted in a
private setting, consistent with professional practices, and were not reflected in the
official inmate records in a timely manner.
FINDING: Current backlogs in inmates awaiting transfer to the next program level
negate the intent of the program design and decrease the motivation to change
behavior. Further it is inconsistent with the program’s objectives to hold graduates
of Level 2 in a unit that operates with that level’s restrictions rather than receiving
the benefits of advancement to Level 3.

230

FINDING: SMU operational practices at the facilities fail to meet the standards set by
the Bureau’s program statement.
FINDING: There is a lack of consistency in the conditions of confinement for an
inmate classified at the same level in the same program but housed at a different
facility. This presents concerns regarding the integrity and design of the level
system.
RECOMMENDATION 4.7: Reexamine the SMU levels as they currently operate,
their corresponding conditions of confinement, the length of time at each level,
and their compliance with the SMU program statement. The program should be
consolidated from four levels to three and the minimum length of time to
complete the program adjusted accordingly.

Mental health
FINDING: Based on the review of the inmate mental health records and the inmate
interviews, the reviewers disagreed with the Bureau diagnosis in nearly two thirds of
the cases and found in over half of the cases that the treatment being offered was
insufficient or inappropriate.
RECOMMENDATION 5.1: All inmates should be seen in a private setting for a
comprehensive mental health evaluation prior to placement in any segregated
setting.
FINDING: The lack of ongoing assessments can lead to the absence of a proper
mental health status evaluation.
RECOMMENDATION 5.2: A complete reevaluation of the mental health record
should be performed by psychology and psychiatry staff every 30 days. Included
in this review should be a face-to-face interview by a member of the mental health
team in a private setting, and the results of this interview should be included in
the reevaluation record.
FINDING: Many inmates in restrictive housing demonstrated significant
symptomatology compatible with the presence of an SMI, which was as yet
undetected by the psychology staff.
RECOMMENDATION 5.3: A vigorous quality improvement program should be
established for the provision of mental health.
FINDING: Very few of the monthly mental health assessments occur in private
settings on a face-to-face basis.

231

FINDING: The shortage of psychiatric staff in Bureau facilities leads to numerous
problems in both diagnosis and treatment, particularly for seriously mentally ill
inmates.
Recommendation 5.4: Given the level of disagreement in the assessment and
treatment plan formulation, the Bureau should conduct an inter-reliability test for
its mental health staff to better determine the accuracy of the diagnosis and
treatment plan process.
RECOMMENDATION 5.5: Psychiatrists need to be more actively involved in
diagnosis and treatment.
FINDING: Overall, most restrictive housing units had no mental health programing
and especially no out-of-cell programming for any inmates with or without mental
illness.
RECOMMENDATION 5.6: A program of regular out-of-cell mental health treatment
needs to be implemented.
FINDING: Almost all facilities reported a lack of mental health staff required to
provide treatment services.
RECOMMENDATION 5.7: The Bureau should complete a clinical staffing needs
analysis.
FINDING: Our review identified inmates in restrictive housing whose mental
conditions should have precluded them from assignment to these units.
RECOMMENDATION 5.8: A protocol needs to be established that identifies those
inmates with serious mental illness who should be excluded from SHU, SMU, or
ADX housing.
FINDING: The assessment team encountered no cases in which an inmate’s serious
mental illness was due to prolonged placement in the ADX.
RECOMMENDATION 5.9: Any inmates who are found to be decompensating from
the effects of restrictive housing should be transferred to a mental health unit for
treatment and observation.
RECOMMENDATION 5.10: Inmates with SMI who are not excluded from restrictive
housing should start participating in a treatment program.
RECOMMENDATION 5.11: Inmates should not be housed in a SHU for protective
custody but rather should be in sheltered general population housing.

232

Reentry
FINDING: No data were available at any of the facilities visited that identified the
number of inmates scheduled to be released directly to the community from
restrictive housing within the next 180 days.
RECOMMENDATION 6.1: On a monthly basis, track and monitor the numbers of
inmates who are scheduled to be released within 180 days and are being released
directly from restrictive housing at each facility.
FINDING: Facilities do not provide step-down planning to transition an inmate from
restrictive housing to general population and subsequently to their eventual release.
The prevailing practice is to keep inmates in restrictive housing until they are
discharged to an RRC or directly to the community.
RECOMMENDATION 6.2: Establish a policy whereby only under extraordinary
circumstances would an inmate be discharged directly from a SHU, SMU, or ADX.
RECOMMENDATION 6.3: Develop a step-down program with increasing
incentives, more out-of-cell opportunities, and increasing opportunities for
congregate programming.
RECOMMENDATION 6.4: Ensure that when inmates complete psychology self-help
programs—for example on anger management, coping, or drug and alcohol
abuse—completion is documented so that case managers and counselors are
aware of it during reentry planning.
RECOMMENDATION 6.5: Develop and provide coordinated, comprehensive,
targeted, specialized cognitive reentry programming specifically designed for
inmates in restrictive housing.
RECOMMENDATION 6.6: Provide programming that identifies and addresses most
significant areas of need for high-risk inmates in order to assist them in
successfully reintegrating into the community.
RECOMMENDATION 6.7: Educate staff about the need for inmates in restrictive
housing to receive formal reentry programming if being released from restrictive
housing.
RECOMMENDATION 6.8: Establish and maintain a culture among all BOP staff,
employees, and contractors that recognizes the need for meaningful reentry
programs for all inmates in the Bureau of Prisons, including those in restrictive
housing, beginning at new officer and staff training and continuing in every
annual in-service training.

233

RECOMMENDATION 6.9: Review the practice of keeping inmates at the ADX until
halfway house release or release directly to the community.

Conditions of confinement
FINDING: The management structure of the Bureau facilities is staffed with sufficient
personnel to provide management and oversight of its segregated units.
FINDING: Each facility reviewed had sufficient staff to manage the segregation units.
FINDING: The presence of a correctional services team working alongside a unit
management team appears to be an effective management approach and provides
sufficient personnel to manage a difficult-to-manage inmate population.
FINDING: The Bureau’s commitment to staff training is outstanding and consistent
with best practices in corrections.
FINDING: The training curriculum used by Bureau facilities is consistent with best
practices, providing a range of topics that meet industry and ACA standards.
FINDING: SHU training is not consistent throughout the Bureau in terms of delivery,
content, hours of instruction, schedule, or mandatory attendance.
FINDING: The observation of inmates in special housing and those that are being
monitored as having suicidal tendencies in specially designed cells under direct
supervision is consistent with best practices and in compliance with ACA standards.
FINDING: Difficulties in the management of the gang problem exacerbates the
protective custody problem, thus causing a high incidence of inmates leaving general
population at their own request and being placed in segregation units after being
threatened in general population.
FINDING: Bureau post orders are extremely comprehensive documents that meet
national standards and requirements and are considered a best practice.
FINDING: The Bureau has established a comprehensive program statement that
clearly identifies the step-by-step requirements associated with managing potential
use of force and critical incidents consistent with nationally recognized standards.
FINDING: Overall the sanitation and physical plant maintenance programs in Bureau
facilities are considered consistent with nationally recognized best practices.
FINDING: The size and furnishings provided in the cells in the SMUs were consistent
with nationally accepted practices.

234

FINDING: The size and furnishings provided in the SHUs were consistent with
nationally accepted practices when providing housing for the designed number of
inmates. The fact that most cells in the SHUs contained showers exceeded national
standards. However, in one facility, the number of inmates housed in a cell exceeded
the design capacity.
FINDING: There is very little difference between the personal property of inmates
assigned to administrative segregation and disciplinary segregation, with the
exception of noted commissary items. This is problematic given the fact that at some
facilities the majority of administrative segregation inmates are on protective
custody status.
FINDING: The frequency of visitation allowed for inmates in Level 3 at USP Allenwood
is inconsistent with Bureau practices for the same level of inmate at other facilities
and not representative of national best practices.
RECOMMENDATION 7.1: A further review of the frequency and duration of visits
should be conducted at USP Allenwood for Level 3 inmates. Serious consideration
should be given to allowing additional time for inmates in Level 3.
FINDING: Health care staff are present in the segregated units at least daily, and
often more frequently, consistent with Bureau policy and ACA standards.
FINDING: A limited number of potential small-group programming spaces are
available at each facility.
RECOMMENDATION 7.2: Use of existing small-group space should be considered
for inmates who have been properly screened.
FINDING: The general conditions of confinement were found to be consistent with
national regulations and standards. Establishing, maintaining, and monitoring the
conditions of confinement appeared to be a routine part of the daily operations of
managing the Bureau facilities visited. Policies were in place, staff were familiar with
the requirements, and post and job descriptions had been established to enforce
compliance in this area.
RECOMMENDATION 7.3: Guidelines that identify specific conditions of
confinement and protection policies consistent with applicable federal regulations
and national standards should be developed and included as part of the PRD
performance review process.
RECOMMENDATION 7.4: Establish a PRD review for the ADX that is separate from
the rest of the Florence Complex.
FINDING: Both program and operational reviews served a vital role in the overall
effective operation of the Bureau’s facilities.

235

RECOMMENDATION 7.5: Establish a separate program performance review for the
USP Florence ADX that includes a comprehensive evaluation of policies and
procedures that are unique to the ADX.
RECOMMENDATION 7.6: Enhance the external oversight of the local operational
reviews that are being conducted on at least an annual basis.
RECOMMENDATION 7.7: Reassess the performance review rating system.

236

Appendix: Research team bios
Project Directors Ken McGinnis and Karl Becker and Program Manager Tammy Felix
were responsible for overseeing all work on this project. Mr. McGinnis was primarily
responsible for making decisions on the direction of research activities and
operational analysis, while Ms. Felix and Mr. Becker were responsible for monitoring
progress on the project work plan and assuring the quality and timeliness of all
products. Below we provide a list of the core team members, their role in the
development of this report, and a brief summary of their background and
experience.
Ken McGinnis, Project Director (CGI). Mr. McGinnis directed all project analysis and
the development and delivery of the final report to the Bureau. Mr. McGinnis worked
in, managed, assessed, and developed plans for the use of restricted housing units
over the course of his 30-year career in corrections. He has served as the Director of
the Michigan Department of Corrections and the Illinois Department of Corrections.
He also has worked as a warden in maximum, medium, and minimum security
institutions and received a number of national awards for his contributions to the
field of corrections, including the Walter A. Dunbar Award in recognition for his
contributions to the development of professional standards for correctional facility
operations. Mr. McGinnis has conducted operational reviews of correctional systems
and facilities across the country. He is a recognized expert in security and
management issues and has directed operational assessments for the Arizona,
Colorado, Florida, Kentucky, Louisiana, Massachusetts, Mississippi, North Dakota,
Oklahoma, Texas, Virginia, and District of Columbia correctional systems.
Dr. James Austin, Research Team Lead (JFA). Dr. Austin has over 30 years of
experience in criminal justice planning and research. He directed several DOJ-funded
research and evaluation programs. He has also assisted numerous state and local
correctional agencies design, implement and validate prison and jail classification
and risk assessment systems. With regard segregation, Dr. Austin was assigned to
the Illinois Department of Correction in one of the nation’s first specialized
segregation units in the early 1970s. More recently he has conducted comprehensive
assessment of the administrative segregation units in the states of Ohio, Mississippi,
Colorado, Illinois New Mexico, Oklahoma, and Maryland. He has conducted this work
through the JFA Institute (Ohio, Mississippi, Colorado), as the lead consultant for

237

Vera Institute (Illinois, Maryland, and New Mexico), and in support of CNA
(Oklahoma).
Karl Becker, Project Director (CGI). Mr. Becker was responsible for project logistics,
work plan implementation, and quality assurance. He provided additional oversight
of the development of project findings and the project report. Mr. Becker brings
experience in managing more than 50 major consulting engagements that have
included performance reviews, organizational assessments, and program evaluations
for correctional systems. Mr. Becker has more than 25 years of experience working
with federal, state, and local criminal justice agencies. He specializes in management,
health care, and financial administration of correctional systems and has extensive
expertise in institutional and departmental staffing assessments. Mr. Becker directed
program services, planning, and finance for 12 years as the Deputy Director of
Administration and Planning for the Illinois Department of Corrections. Over the last
12 years as a consultant specializing in correctional system management, Mr. Becker
has worked with senior correctional officials, governor’s office staff and legislative
appropriations staff on correctional system performance reviews and operational
assessments in Colorado, Florida, Illinois, Indiana, Kentucky, Massachusetts,
Oklahoma, and Virginia.
Larry Fields, Operations Reviewer (CNA). Larry Fields has over 40 years of
experience in the field of correctional administration and leadership. He has
demonstrated skills in facility management, community corrections and probation
and parole. His expertise has been further enhanced by a background that includes
work with juvenile offenders, psychiatric patients, and county jail operational
reviews. In addition to serving as Director of the Oklahoma Department of
Corrections, Mr. Fields has also served as Deputy Director of Institutions, Regional
Director, Warden, Deputy Warden, and Community Corrections Superintendent. His
government experience has been supplemented by his extensive experience
providing consultant services in correctional management and administration issues
nationally for the last 16 years.
Michael Lane, Operations Reviewer (CNA). Michael Lane has 35 years’ experience in
law enforcement and corrections. As Inspector General for the Illinois State Police,
Mr. Lane directed sworn and civilian staff responsible for financial, compliance, and
management audits of all Illinois State Police divisions, as well as ensuring
compliance with standards established by the Commission on Accreditation for Law
Enforcement Agencies. Mr. Lane was Director of the Illinois Department of
Corrections from 1981 to 1990, the longest tenure of any Corrections Director in the
history of the state. He administered a budget of more than $528 million and
managed a staff of 11,000 who supervised 38,000 adults and juveniles in prison or
on parole. Mr. Lane supervised a massive $500 million expansion program to meet an
adult inmate population that grew from 12,500 to 24,700. He established written,
standard policies and procedures for the state correctional system and implemented

238

a highly effective auditing compliance system. Mr. Lane’s career includes experience
managing all adult prisons in Illinois as an Assistant Director, as Warden of the
largest maximum security prison in the state, the Menard Correctional Center, and as
regional administrator of adult parole.
Mike Maloney, Operations Reviewer (CNA). Mr. Maloney completed a 30-year career
with the Massachusetts Department of Corrections in 2004. Since beginning his
career in 1974, he served as a Social Worker, Director of Classification Deputy
Superintendent, Superintendent, Deputy Commissioner, and Commissioner from
1997 to 2004. Throughout his career he has had operational responsibility over
segregation units and high-security facilities. While a Deputy Superintendent in the
largest medium security institution in Massachusetts, he was responsible for the
daily operation of a 100-man segregation unit, which included a 30-man protective
custody unit. As the Superintendent of Walpole Prison, the maximum security facility
in Massachusetts, he was responsible for all aspects of the operation of the
Department Segregation Unit, a department unit designed to hold the most violent
and disruptive inmates in the Massachusetts system and also supported the design
of the high-security unit, modeled after the Control Unit at U.S. Penitentiary Marion,
that replaced the DSU. Mr. Maloney was directly responsible for overseeing the
process to develop policy and procedures for the unit, to which inmates were
assigned based on a departmental disciplinary hearing. Mr. Maloney was the
Commissioner’s designee to approve or deny placement in the unit, and to act upon
appeals. As Deputy Commissioner, part of Mr. Maloney’s responsibility was to ensure
that all superintendents with segregation units within their facilities adhered to
departmental policy regarding the operation of segregation units.
Mary Marcial, Reentry Reviewer (CNA). Mary Marcial has 25 years of service with
the Connecticut Department of Correction. She began her career as a Correctional
Counselor in Addiction Services and in Classification and Case Management. In 1992
she was appointed as Warden of the state’s DWI unit and later that year was tasked
with activating a new institution, which included developing the facility’s policies,
procedures, post orders, staffing plans, and program mission. In 1995 she was
selected to be the first female Warden of the state’s reception and high-risk offender
special management unit, which at that time also served as the state’s Super-Max
facility. While there, she initiated a High Bond unit for the state’s pretrial offenders
and established the state’s Chronic Discipline Unit, developing policies, procedures,
and post orders for those high-risk offender units. In March 2003, Ms. Marcial was
selected to serve on the Commissioner’s executive team as Director of Programs and
Treatment. In that capacity she was responsible for numerous divisions including
Reentry Services, Education, Religious Services, Victim Services, Volunteer and
Recreation Services, Program Review and Development, Correctional Enterprises,
Health and Addiction Services, and Offender Classification and Population
Management. She currently works with ASCA’s Reentry Committee managing a

239

reentry information-sharing grant and has recently conducted an assessment of the
Virginia Department of Correction’s Reentry and community diversion programs.
Robert May, Reentry Team Lead (CNA). Mr. May is Assistant Director of Program
and Technology Services with the IJIS Institute where he oversees the Institute’s work
in Corrections/Reentry, Justice to Health, Gang Information Sharing, and Statewide
Automated Victims Information and Notification System Technology Assistance and
Procurement Reform. Mr. May most recently was a principal with the Criminal Justice
Institute where he also served as the Associate Director of the Association of State
Correctional Administrators (ASCA). He has over 37 years of experience in the fields
of criminal justice, law enforcement, substance abuse treatment programming,
correctional health care, alternatives to incarceration, and substance abuse program
design and implementation. Mr. May serves as ASCA's representative on the Second
Chance National Reentry Resource Center (NRRC) Steering Committee and as vice
chair of the NRRC’s Pre-Release Planning and Post-Release Supervision
Subcommittee. He has experience in treatment programming and alternatives to
incarceration and has conducted alternatives assessments and reentry work for state
correctional systems and counties. Mr. May directed ASCA’s Reentry Information
Sharing projects funded by the Bureau of Justice Assistance (BJA) with pilot sites in
Maryland, Rhode Island, and Hampden County, MA Sheriff’s Department. The main
goals of these projects were to leverage corrections information for reentry purposes
and enable effective sharing of information among corrections and various social
service agencies and service providers to improve reentry of offenders back into the
community.
Jon Ozmint, Due Process Team Lead (CNA). Mr. Ozmint is an attorney who began
his legal career in the U.S. Navy Judge Advocate General’s Corps, where he served in
various capacities in the criminal justice system. Mr. Ozmint later served as
prosecutor in the Tenth Circuit Solicitors Office, where he served as General Counsel
to the South Carolina State Department of Labor, Licensing and Regulation until
1994, when the Attorney General appointed him Deputy Attorney General and Chief
Prosecutor for the Statewide Grand Jury. In 2003 Mr. Ozmint was appointed Director
of the South Carolina Department of Corrections, where he served until 2011. Mr.
Ozmint has served as chairman of the Staff Safety Committee for the American
Correctional Association and of the Legal, Legislative and Policy Committee of the
Association of State Correctional Administrators (ASCA). His experience as a
practicing attorney, member of the Judge Advocate General’s Corp, and former
director of corrections provides him with unique skills, abilities and knowledge that
will enable CNA to conduct a thorough and comprehensive review of the due process
issues within the special housing units of the Bureau.
Tom Roth, SHU Operations Lead (MGT). Tom Roth has more than 27 years of
experience in the field of corrections. He has been involved in virtually every facet of
correctional management including central office leadership, facility management,

240

accreditation, training, and education. Mr. Roth has served as Deputy Chief of
Administration for the Illinois Department of Corrections, warden of multiple
correctional institutions, and accreditation auditor for the American Correctional
Association for nine years. He led the creation and implementation of an agency-wide
strategic plan for the Illinois Department of Corrections.
Emmitt Sparkman, Operations Reviewer (CNA). Emmitt Sparkman, the current
Deputy Commissioner of Institutions for the Mississippi Department of Corrections,
brings extensive experience in the analysis of restricted housing as a tool in
correctional system management. Mr. Sparkman has overseen all of the state’s
correctional institutions for the last ten years and has also served as Warden and
Deputy Warden in a career that spans more than 35 years in the operation of state
correctional facilities. He has been a key figure in the development of capacity plans
for the state of Mississippi that have managed the growth in the overall inmate
population without new prison construction. More recently, he has participated in a
study of the use of administrative segregation practices in the Colorado and
Oklahoma correctional systems.
Dr. Roberta Stellman, Mental Health Reviewer (CNA). Dr. Stellman is a boardcertified psychiatrist and Distinguished Fellow of the American Psychiatric
Association. She provided clinical psychiatric services to the inmate population of
the New Mexico Department of Corrections from 1983 to 2006. She has since served
as a consultant in correctional mental health treatment for the state of Delaware,
served as the federal court monitor for an agreement on the delivery of correctional
mental health services between the state of Delaware and the Justice Department,
and participated in a comprehensive outside review of mental health services in the
Massachusetts Department of Correction.
Dr. Pablo Stewart, Mental Health/Medical Service Team Lead (CNA). Dr. Stewart
was the Senior Attending Psychiatrist for 4 years at the Forensic Unit of San
Francisco General Hospital, with administrative and clinical responsibility for a 12bed, maximum-security psychiatric ward. He was appointed as the Psychiatric Expert
for the U.S. Federal Court in the Gates v. Deukmejian case and Madrid v. Gomez,
which addressed the quality and availability of psychiatric care provided to the
inmates at Pelican Bay State Prison. He has worked extensively on correctional mental
health issues as a consultant for the United States Department of Justice, Civil Rights
Division in both juvenile and adult systems.
George Vose, ADX/SMU Operations Lead (MGT). Mr. Vose has more than 30 years of
experience in the corrections field. He has served in a number of positions within the
Massachusetts Department of Corrections, including the director, deputy director,
and as superintendent of two facilities. He also served as the Commissioner of the
Rhode Island Department of Corrections. Through these positions, he has been
responsible for the development and allocation of programs and resources, as well as
determination of agency objectives, goals, and internal organizational structure. He

241

also served as first-in-command regarding correctional issues affecting public safety
and was responsible for the safety and security of all state correctional facilities. Mr.
Vose has provided consulting services and technical expertise to numerous state
departments of correction including Arizona, Maryland, New Mexico, and Wyoming,
as well as local government correctional systems in Bristol County, New Jersey, and
the City of Philadelphia. He served as the Project Manager for the NIC project,
“Assessing Prison Culture,” working with the consultant team and NIC to develop a
protocol for assessing prison culture and to administer the protocol in correctional
facilities throughout the U.S. He also was a consultant on a project for the Maryland
Department of Corrections to conduct an organizational analysis of Central Office
functions and staffing, and as a consultant on an assessment of the Vermont
Department of Corrections organizational structure for the NIC.
Tammy Felix, Program Manager (CNA). Ms. Felix was responsible for the overall
delivery of the project and served as the single point of contact for project
coordination and technical direction with the Bureau. She also provided project
updates and coordinated/ensured the delivery of the final report to the Bureau. Ms.
Felix has over 14 years’ experience performing analytical and research support work
for a variety of safety and security projects focusing on emergency management,
homeland security, and law enforcement issues. As the manager of a project
assessing the New York City Police Department’s (NYPD) implementation of vertical
patrol tactics in New York City Housing Authority facilities, Ms. Felix provided
oversight and analytic support for the review of the NYPD’s Housing Bureau Policies,
procedures, rules, training and practice, transcripts, review of other Expert Reports,
analysis of the conformity of practice with the policies and whether these policies
reflect professional standards within policing. She has also supported site visits in
Colorado and Massachusetts state prisons.

242

The CNA Corporation
This report was written by CNA Corporation’s Safety and Security (SAS)
division.
SAS’s work helps improve decision-making during crisis operations and
fosters innovative answers to challenges in the areas of first response;
emergency management; public health and agriculture; homeland
security; risk-management policy development and operations; and
response and recovery capabilities at a national level.

IRM-2014-U-009563

CNA Corporation is a not-for-profit research organization
that serves the public interest by providing
in-depth analysis and result-oriented solutions
to help government leaders choose
the best course of action
in setting policy and managing operations.

Nobody gets closer—
to the people, to the data, to the problem.

www.cna.org

703-824-2000

3003 Washington Boulevard, Arlington, VA 22201