The Community Partnership and the District of Columbia’s
Public Homeless Assistance System
Martha R. Burt
Sam Hall
June 2, 2008
Urban Institute
2100 M Street, N.W.
Washington, D.C. 20037
www.urban.org
The contents of this report are the views of the authors and do not necessarily reflect the views or
policies of the Urban Institute, its trustees, or funders.
The Community Partnership and the District of Columbia’s Public Homeless System
i
CONTENTS
Acknowledgments.......................................................................................................................... iv
Chapter 1: Introduction ............................................................................................................... 1
Overview of Homeless Assistance in the District........................................................................... 2
Methods and Data Sources.............................................................................................................. 6
Data from TCP............................................................................................................................ 6
Other Information ....................................................................................................................... 6
Chapter 2: TCP’s and DHS’s Responsibilities for Emergency Shelters.................................. 7
Highlights........................................................................................................................................ 7
Conditions at Emergency Shelter Facilities.................................................................................... 8
Problem Shelters ......................................................................................................................... 8
Compliance with the ADA ......................................................................................................... 9
Official Arrangements for Maintenance ....................................................................................... 10
Upkeep of Current Emergency Shelter Facilities ..................................................................... 10
Contract Language Regarding Maintenance Responsibilities .............................................. 11
Are Repairs “Emergency” or “Regular”? ............................................................................. 12
When Do Shelters Use Their Own Funds for Repairs? ........................................................ 12
Improvement in Maintenance Performance.......................................................................... 12
Maintenance Comparisons and Overall Problems with Upkeep .......................................... 13
Implications................................................................................................................................... 13
Chapter 3: Orchestrating the Continuum of Care .................................................................. 15
Highlights...................................................................................................................................... 15
The D.C. Initiative and Developing the Continuum of Care ........................................................ 16
Before the D.C. Initiative.......................................................................................................... 16
During the D.C. Initiative ......................................................................................................... 17
After the D.C. Initiative ............................................................................................................ 19
Organizing the Continuum of Care............................................................................................... 20
The Current Situation and the Issue of Leadership....................................................................... 22
THe ICH And District Leadership................................................................................................ 23
Managing Shelter Plus Care Resources ........................................................................................ 23
Training and Technical Assistance ............................................................................................... 24
Communications in General ......................................................................................................... 25
Implications................................................................................................................................... 26
The Community Partnership and the District of Columbia’s Public Homeless System
ii
Chapter 4: Contract and Grant Financial Management ........................................................ 27
Highlights...................................................................................................................................... 27
TCP’s Contracting Responsibilities.............................................................................................. 27
Contract Size............................................................................................................................. 29
“Low Barrier” Should Not Mean “No Services”...................................................................... 30
Competition in Grants and Contracts........................................................................................ 30
Transferring Contracts to DHS ..................................................................................................... 31
Implications................................................................................................................................... 33
Chapter 5: Quality Assurance and Monitoring Activities ...................................................... 35
Related to Health and Safety Issues.............................................................................................. 35
Highlights...................................................................................................................................... 35
Accountability for the Continuum of Care ................................................................................... 36
Current Practice of Program Monitoring .................................................................................. 37
Results of SMU Monitoring Reports .................................................................................... 40
Who Is Responsible?............................................................................................................. 41
Attending to Issues Raised by the Inspector General ............................................................... 42
Enforcement of Corrective Action Plans .............................................................................. 42
Contract Content that Cannot Be Met and Has No Expectations of Being Met................... 43
Responsibility for Staff Training .............................................................................................. 43
Responsibility for Security ....................................................................................................... 44
Implications................................................................................................................................... 45
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling 47
Highlights...................................................................................................................................... 47
Fair Hearings and Administrative Reviews .................................................................................. 48
Client Rights in Programs......................................................................................................... 49
Administrative Reviews and Fair Hearings for Clients Responding to Provider Actions........ 49
Emergency Actions............................................................................................................... 52
Alternatives to Negative Actions.............................................................................................. 53
Administrative Reviews and Fair Hearings Initiated by a Client Complaint ........................... 54
Private versus Public Shelters....................................................................................................... 54
A Final Issue—Individual and “Group” Rights............................................................................ 55
Implications................................................................................................................................... 56
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking ..................... 57
Highlights...................................................................................................................................... 57
The Homeless Management Information System......................................................................... 58
HMIS Coverage ........................................................................................................................ 58
HMIS Data Quality................................................................................................................... 59
The Community Partnership and the District of Columbia’s Public Homeless System
iii
The Annual Point-in-Time Count ................................................................................................. 60
Reporting Using HMIS ................................................................................................................. 61
Routine Reporting..................................................................................................................... 61
Summary and Performance Reporting...................................................................................... 62
Example of What ART Can Produce—Emergency Shelter Use by Single Adults .............. 63
Example of What ART Can Produce—Recognition that Goals Set Too Low ..................... 63
The “Closed” Nature of the District’s HMIS................................................................................ 64
Implications................................................................................................................................... 65
Chapter 8: Performance Standards and Client Outcomes ..................................................... 67
Highlights...................................................................................................................................... 67
Occupancy and Length of Stay ..................................................................................................... 68
Occupancy ................................................................................................................................ 68
Occupancy as a Measure of Program Performance .............................................................. 68
Occupancy as an Indicator of System Performance ............................................................. 69
Length of Stay—Chronicity...................................................................................................... 70
Challenges to Reducing Chronic Homelessness................................................................... 71
Client Outcomes............................................................................................................................ 71
Implications................................................................................................................................... 73
References..................................................................................................................................... 75
Appendix A: List of People Interviewed ...................................................................................... 77
Appendix B: List of Acronyms..................................................................................................... 79
The Community Partnership and the District of Columbia’s Public Homeless System
iv
ACKNOWLEDGMENTS
Many individuals have assisted us on this project; we would especially like to thank the
following:
•
Directors and staff at DHS who took the time out of their busy schedules to provide
insight and direction at all levels of this report; special thanks to Clarence Carter, Fred
Swan, Kate Jesberg, Sakina Thompson, and Deborah Carroll.
•
TCP Director Sue Marshall, and staff Cornell Chapelle, Darlene Mathews, Amy
McPherson, Michele Salters, Clarence Stewart, Tamura Upchurch, Xiawei Zheng,
Mathew Winters, and Tom Fredrickson, who provided us significant access to the many
resources at The Community Partnership.
•
Mayoral staff at both the Office of the City Administrator and the Office of the Deputy
Mayor for Planning and Economic Development, particularly Julie Hudman, Laura
Zeilinger, and Leslie Steen, who helped shape our work plan and the structure of our
research.
•
Council Member Tommy Wells and his staff, Adam Maier, Ram Uppuluri, and Yulondra
Barlow, who worked with us in defining our goals for this assessment.
•
District agency staff and directors, with special thanks to Michael Kelly at DCHD, Tori
Whitney at APRA, Stephen Baron at DMH, Leila Edmonds and Guyton Harvey at
DHCD, Susan Schaffer at PSA, Calvin Johnson at CSOSA, Brian Jordan at MPD, and
Michael Williams at FEMS for their willingness and energy in helping us identify the
breadth of homelessness in the District as it affects the work of city agencies..
•
Advocacy groups, especially Washington Legal Clinic for the Homeless staff Patricia
Mullahy Fugere, Mary Ann Luby, Amber Harding, Marcy J. Dunlap, and Andy Silver,
who informed many of the sections in this report.
•
The more than 60 District homeless shelter directors and providers whose insights, data,
and anecdotes were the basis of many of the sections presented here.
•
Katie Vinopal, research assistant at the Urban Institute, who built and conducted analysis
of the Program Monitoring Report database.
For a complete list of people who contributed to this assessment, see Appendix A.
We appreciate the significant contributions everyone made to our research and writing of this
report. The findings and views herein are solely those of the contractors and authors, who are
responsible for any errors or omissions.
1
Chapter 1: Introduction
Chapter 1
Introduction
Homelessness has been a continuing presence in the District of Columbia for almost three
decades. It only became a high priority issue for public action, however, when the administration
of Mayor Adrian Fenty assumed control of District government in January 2007. As a City
Council member, the Mayor had been instrumental in passing the Homeless Services Reform Act
of 2005 (HSRA); he quickly made clear that ending homelessness in the District would be
among the most important goals of his administration.
Toward this end, the Department of Human Services was authorized to contract with the Urban
Institute to conduct an assessment of the District’s homeless assistance system, with the
expectation that the results of such an assessment could help guide efforts to transform the
system to make it more effective at reducing and ultimately ending homelessness. This is the first
evaluation report, begun in July 2007. It addresses issues related to the Community Partnership
(TCP) identified in our initial survey of key informants, upon which we based our plan for this
evaluation.
TCP 1 is a nonprofit organization occupying an intermediate position between District
government agencies and homeless assistance providers. Since 1994 it has been the vehicle
through which District funding from the Department of Human Services (DHS) and the
Department of Housing and Community Development (DHCD) flows to homeless assistance
programs. It also manages the application process to and the flow of federal dollars from U.S.
Department of Housing and Urban Development (HUD) homeless funding programs to
programs in the District. A number of cities, including Miami/Dade County, Florida and
Columbus/Franklin County, Ohio, have found the use of similar intermediary nonprofits to be an
efficient and effective way to manage homeless resources.
We were charged to make a detailed assessment of TCP activities and accomplishments,
designed to answer the broad questions “What does District government get for the money it
pays TCP through the DHS contract 2 ?” and “What does TCP do for the District of Columbia
community?” This first report of the evaluation focuses on providing answers to those questions.
1
2
A list of acronyms is included as Appendix B.
The most recent contract between DHS and TCP was executed in FY 2005. FY 2006, the base year, was funded at
about $25 million. Originally the contract provided about $28 million for Option Year 1 (FY 2007) and about $30
million for Option Year 2 (FY 2008). Resources for Option Year 1 were augmented by $9.1 million in 2007,
acknowledging the realities of system growth and providing resources for some new programs. Of these funds, $4.7
million helped convert several hypothermia shelters into year-round shelters, opened new facilities, supported a new
contract for food services, and similar activities. About $1 million provided new youth shelter beds and services and
about $1 million went for system transformation including closing D.C. Village. The remainder covered
expenditures for some small projects. The contract for Option Year 2 (FY 2008) has been in place since early
December 2007 and incorporates resources to continue the activities and services that were new in Option Year 1.
Modifications to the Option Year 2 contract are likely, as new services are designed and implemented following
developments in the DHS Family Services Administration and the Interagency Council.
Chapter 1: Introduction
2
The second evaluation report, Transforming the District of Columbia’s Public Homeless System
(Burt and Hall 2008a), addresses the much larger issue of how the District could organize itself
to end homelessness. Many topics discussed in this first report are also pertinent to the second;
findings reported here are incorporated into the second report as they shed light on the discussion
of what it will take to end homelessness in the District of Columbia. In addition, a short
summary report, Major Recommendations: Summary Report of the Urban Institute’s Assessment
of the District of Columbia's Public Homeless Assistance System (Burt and Hall 2008b) presents
major recommendations and integrates findings from both reports.
Within the two broad questions for this report, we identified areas of specific concern and
discuss each in its own chapter.
ƒ
Chapter 2 focuses on issues surrounding the emergency shelters that receive DHS
funding through TCP, including their condition, upkeep, and allocation of
responsibilities.
ƒ
Chapter 3 describes the scope of TCP’s activities relating to managing the rest of the
District’s Continuum of Care, including the annual HUD funding application process,
training, communications, and leadership.
ƒ
Chapter 4 examines TCP’s contracts and grants financial management activities for all
programs for which at least some District or HUD funding flows through TCP.
ƒ
Chapter 5 discusses findings related to quality assurance and monitoring for all
residential programs within the Continuum of Care.
ƒ
Chapter 6 addresses the issues of client and provider rights and responsibilities under the
Homeless Services Reform Act (HSRA) and how these play out in practice.
ƒ
Chapter 7 describes TCP’s capacity to collect, analyze, and report data for policymaking.
ƒ
Chapter 8 discusses TCP’s approach to setting performance standards and selecting client
outcomes.
OVERVIEW OF HOMELESS ASSISTANCE IN THE DISTRICT
Before launching into detailed discussions of issues in the following chapters, we thought it
would be useful for the reader to have a brief overview of homeless assistance programs in the
District of Columbia. With so much recent focus on emergency shelters for families, conditions
at D.C. Village, and the push to close D.C. Village and transition families living there and in
other temporary accommodations to better situations, it is important not to lose sight of the fact
that the city’s homeless assistance network is far larger than emergency shelters serving families,
or even than all the emergency shelters for which the District both pays and supplies the
facilities. In addition, as many of the issues to be discussed in subsequent chapters have to do
with what TCP is able to get agencies and programs to do, it is important to understand just how
much of the city’s homeless assistance network actually comes within its purview and how much
remains outside.
3
Chapter 1: Introduction
We present information on the types of programs offered, the agencies offering them, the
populations being served (single adults or families), and the number of beds involved in the
various programs (counting an average of three beds for each family unit). To define for the
reader what TCP is able to influence and what it is not, and the scope of its authority and
responsibility, we assume that if a program or agency receives funding through TCP (the funding
having originated in DHS, HUD, or DHCD), it is within TCP’s responsibility and sphere of
influence—part of the Continuum of Care. All such programs must report to the Homeless
Management Information System (HMIS) managed by TCP; we take such reporting as the
indicator of whether a program or agency is inside or outside the Continuum of Care.
Figure 1.1: Homeless Assistance Agencies and Projects
Operating in the District of Columbia in January 2008, by
Participation in the Continuum of Care/HMIS
Number of agencies/programs
Total
200
180
160
140
120
100
80
60
40
20
0
In CoC/HMIS
Not in CoC/HMIS
180
105
75
71
41
30
Agencies
Projects
Source: Urban Institute Analysis of 2008 Housing Inventory Charts. Note: Agencies designated as “in CoC/HMIS” report to the HMIS for
at least one project, but may also have projects that do not report to HMIS.
We look first at homeless assistance agencies and projects operating in the District (table 1.1 and
figure 1.1), and then at bed counts (figure 1.2), observing in each instance who is being served
and through what type of project. Figure 1.1 gives the simplest overall picture, depicting the
number of agencies and projects offering homeless assistance services in the District that do and
do not participate in the Continuum of Care. In figure 1.1 we can see that more agencies are
participating in the Continuum of Care or contributing data on their clients to the HMIS than are
not doing so—40 versus 31 agencies. The participating agencies offer more projects than the
nonparticipating ones, though, so the number of projects in the Continuum of Care (101) is
greater than the number that do not participate (80).
4
Chapter 1: Introduction
Table 1.1: Homeless Assistance Agencies and Projects in the District of
Columbia Open and Operating on January 31, 2008, by Project Type and
Participation in HMIS/Continuum of Care
Totals
All agencies
In the CoC/HMIS (58%)
Not in the CoC/HMIS (42%)
All projects
In the CoC/HMIS (57%)
Not in the CoC/HMIS (43%)
For Familiesd
Agencies serving families
In the CoC/HMIS (58%)
Not in the CoC/HMIS (42%)
Projects serving families
In the CoC/HMIS (62%)
Not in the CoC/HMIS (38%)
For Single Adults
Agencies serving single adults
In the CoC/HMIS (55%)
Not in the CoC/HMIS (45%)
Projects serving single adults
In the CoC/HMIS (55%)
Not in the CoC/HMIS (45%)
Number of
Emergency
a
Shelters
Number of
Transitional
Housing
b
Projects
Number of
Permanent
Supportive
Housing
Projects c
Total
Number of
Agencies
or Projects
19
8
11
33
20
13
50
29
21
88
54
34
30
13
17
59
29
30
71
41
30
180
103
77
8
4
4
9
4
5
28
18
10
35
26
9
9
3
6
11
4
7
38
22
16
55
34
21
15
5
10
24
16
8
31
16
15
53
28
25
25
13
12
48
25
23
47
26
21
125
69
56
Source: Urban Institute analysis of Housing Inventory Chart data prepared for the January 24th Metropolitan Council of
Governments’ annual Point in Time Count of homeless people, reflecting the situation as of January 31st, 2008.
a
One program does both families and singles; it is counted as a family program because it primarily provides
family beds.
b
Two projects serve both families and single adults; they are counted with the projects serving single adults if
single adults predominate, or with projects serving families if families predominate.
c
Six projects serve both families and single adults; they are counted with the projects serving single adults if
single adults predominate, or with projects serving families if families predominate.
d
All domestic violence projects and projects reporting that they serve “mixed” populations are included in the “families”
category.
Table 1.1 shows the same comparisons broken out by two important factors—what type of
households the projects serve (rows), and what types of projects the agencies offer (columns).
The second and third panels provide the information for projects serving families and projects
serving single adults, respectively. Projects for victims of domestic violence are included with
family projects, and those for single youth are included with those serving single adults. All three
panels show the proportion of agencies or projects that do or do not participate in the Continuum
of Care (e.g., 58 percent of all agencies do, and 42 percent do not, participate). The first three
columns of table 1.1 show breakouts by whether the agency or project is offering emergency
5
Chapter 1: Introduction
shelter, transitional housing, or permanent supportive housing.3 The last column shows the total
number of agencies offering at least one project.
Figure 1.2: District of Columbia Homeless Assistance Beds in
January 2008, by Project Type, Population Served, and
Participation in the Continuum of Care
Families-in
Families-not
Singles-in
Singles-not
3000
2426
Number of beds
2500
2000
1500
1446
1373
1000
624 555
500
374
98
202
373
721
608
450
0
Emergency Shelter
Transitional Housing
Permanent Supportive
Housing
Source: Urban Institute analysis of 2008 Housing Inventory Charts. Note: Beds for youth are included in the figures for single adults.
Although some variation exists in the proportion of agencies and projects participating in the
Continuum of Care, overall the proportion remains in the range of half to two-thirds in each
category—lower for projects serving single adults, higher for those serving families. Permanent
supportive housing projects and the agencies that offer them are the least likely to be in the
Continuum of Care, while all the big emergency shelters do participate. Figure 1.2 shows the bed
coverage within the Continuum of Care for the three types of projects serving single adults and
families, 4 revealing that far higher proportions of beds are included than projects or agencies.
3
If an agency offers two or more types of programs, it is counted in each relevant column. An agency offering an
emergency shelter and a transitional housing program is counted once in the column for emergency shelters and
once in the column for transitional shelters. It is, however, counted only once in the “totals” column. Thus, the
number of agencies in the emergency shelter, transitional housing, and permanent supportive housing columns add
up to considerably more than the number of agencies shown in the “totals” column. As each program represents
only itself, however, the program numbers in the last three columns of each row do add up to the number in the
“total” column.
4
The data in figure 1.2 come from Urban Institute analysis of Housing Inventory Chart data prepared for the January 24th
Metropolitan Council of Governments’ annual Point in Time Count of homeless people and reflect the situation as of January
31st, 2008. For emergency shelter, we included only permanent year-round beds (i.e., not hypothermia or seasonal beds).
Chapter 1: Introduction
6
Still, having such a large proportion of permanent supportive housing beds remaining outside the
Continuum of Care is a point of potential difficulty if the District really wants to develop the
2,500 net new units of similar housing in the next few years and assure that the longest-term,
most disabled homeless people are able to get into it and thus end their homelessness.
METHODS AND DATA SOURCES
Urban Institute staff pursued a number of approaches to gathering the information that we
describe in this report, which came from a variety of sources. These include the following:
DATA FROM TCP
• Interviews with most TCP staff, from the Executive Director to the accountants
•
Examination of budget and contract documents between DHS and DHCD and TCP
•
Examination of budget and contract documents for more than 100 contracts between TCP
and homeless service providers
•
Examination of standard reports from the HMIS maintained by TCP
•
Direct analysis of hundreds of thousands of raw HMIS records on emergency shelter
users, supplied to us by TCP
•
Review of performance standards adopted by different contract agencies and comparisons
of the standards adopted to actual performance
•
Examination of common standards for provider behavior and working with clients
•
Examination of inspection and maintenance reports
OTHER INFORMATION
• Interviews with more than 100 public agency staff, staff of elected officials, and staff of
homeless assistance and advocacy agencies ranging from executive directors to case
workers
•
Analysis of Administrative Review and Fair Hearing records maintained by DHS,
covering the first six months of these processes under the HSRA
•
Analysis of monitoring reports covering the first six months of operations of DHS’
Shelter Monitoring Unit
•
Examination of critical documents including Homeless No More (the District’s ten-year
plan to end homelessness), the HSRA and its amendments, regulations, common
standards, earlier reports on the District’s homeless assistance system, the Inspector
General’s report on D.C. Village, among others.
Chapter 2: Responsibilities for Emergency Shelters
7
Chapter 2
TCP’s and DHS’s Responsibilities for Emergency Shelters
HIGHLIGHTS
ƒ
District-owned shelters remain in poor physical condition. Funds for district shelter
repairs are, for the most part, too low or too slow to provide needed comprehensive and longterm repairs.
ƒ
The number of clients per shelter is too high, and the toll is being taken on the structure
of the facilities. Many District emergency shelters have expanded and taken on more clients
than their capacity, often as a result of directives from District authorities. The number of
clients per facility needs to be reduced to slow building dilapidation.
ƒ
The District is not in compliance with laws requiring noncommunal settings for some
disabled people and families with children. In violation of District and federal law, all
emergency shelters for single adults owned by the district are communal.
ƒ
Both nonstructural and structural shelter repairs have shown significant signs of
improvement. This is especially true for short term noncomprehensive repairs, but
accountability and the overall demand for repairs has slowed response rates and overall
timeliness. TCP and now OPM have improved their timely response to maintenance needs,
but disagreements over what constitutes an emergency and over responsibility for shelters
such as Adam’s Place, which the District leases but does not own, have resulted in slower
response rates.
ƒ
Compared to private maintenance procedures, the District’s short term shelter upkeep
system gets relatively high marks from providers. Providers using privately owned
buildings mentioned problems of response time from private landlords that, in general, were
worse than the TCP and OPM performance for District-owned facilities being criticized by
everyone from clients to the Inspector General.
This chapter describes our findings with respect to several pressing questions about emergency
shelters. These include their physical condition, overcrowding, and (lack of) compliance with
requirements for accommodating people with disabilities, per the Americans with Disabilities
Act. We look at problem shelters, official arrangements for maintenance, and recent changes in
maintenance arrangements and performance. Finally, we look at the implications for the future.
Before describing conditions at the District’s emergency shelters, it is important for the reader to
understand the scope of TCP’s influence over homeless assistance agencies and projects in the
city and the District government’s involvement with the physical plant being used by the biggest
of the city’s emergency shelters. At least 2,100 year-round low barrier and temporary emergency
shelter beds, about two-thirds of all emergency shelter beds available, are located in 14 Districtowned facilities. The remaining shelters are privately owned. The District-owned facilities are
mostly old school buildings, but other structures such as nursing homes are in use as well.
Chapter 2: Responsibilities for Emergency Shelters
8
Numerous inspection reports, provider testimonies, and maintenance requests support the
conclusion that these are mostly in bad physical shape, with significant structural damage,
severely strained mechanical systems, and other problems. Most are also severely overcrowded.
These are the buildings for which TCP bears the brunt of criticism for the living conditions they
provide, although the District itself is largely responsible for redressing the most serious
problems. About 200 or so transitional housing beds are also located on these premises and face
the same issues. This chapter goes into some detail about the controversies surrounding these
buildings and who bears the responsibility for their condition.
CONDITIONS AT EMERGENCY SHELTER FACILITIES
The physical condition of many emergency shelters under the charge of the District remains
poor. Many (but not all) of the buildings are overcrowded, old, and unrenovated. One needs to
look no further than D.C. Village to see the potential costs of continuing to use run-down
schools, hospitals, and other District buildings not originally meant for shelter. Since December
2006 D.C. Village has absorbed $344,000 in repairs without showing much change, and is now
slated to be torn down. The use of these shelters is tantamount to the “warehousing” so many of
those we interviewed equated negatively with old systems. In the words of one OPM employee,
trying to fix these buildings “is like putting a bow-tie on a pig.”
PROBLEM SHELTERS
District-owned emergency shelters tended to get the worst ratings of physical plant and operating
conditions. The following are either District-owned facilities or are under control of the District:
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Franklin School
New York Avenue
Adam’s Place
801 East
D.C. Village
New Endeavors by Women
Emery
Gale (not open, under repair)
Blair
Madison
Federal City
John Young Open Door
La Casa
D.C. General (Harriet Tubman)
CCNV
All but New Endeavors, Blair, and Gale are functioning emergency shelters, and all, as we note
below, are TCP’s and the District’s responsibility to maintain, the former for minor and the latter
for major repairs. The situation of D.C. Village does not need further discussion; its faults have
been well documented and it has been closed. Other shelters, though, have shown similar
problems. As we discuss in chapter 5, John Young, New York Avenue, and Franklin School
have serious deficiencies that maintenance has not been able to fully address. Franklin’s
Chapter 2: Responsibilities for Emergency Shelters
9
monitoring report mentions rat infestations and structural damage, some of which is potentially
dangerous— i.e., loose railings, cracked walls, holes in the ceiling. At others, such as New
Endeavors (a transitional housing project), 18 of 38 rooms are inhabitable due to a badly leaking
roof, water damage, mold, and improper heating and cooling. The District bought this building in
its present condition and has yet to remedy its significant physical flaws. Almost every shelter
listed has a serious deficiency, from gaping holes a client once fell through (801 East) to cracked
ceilings, leaky pipes, unusable toilets, and soiled beds. In general, the monitoring reports
discussed in chapter 5 have adequately identified these deficiencies; more is needed in terms of
policies to reduce the strain on individual shelters and increase the resources available to fix
them.
For emergency shelters, the problem of natural deterioration of very old buildings is exacerbated
by the overflow occupancy conditions that many of the shelters experience on a regular basis. As
detailed in chapter 8, most of the shelters just named operate at or in excess of capacity much of
the time. For instance, New York Avenue was at 164 percent of its capacity on the night of
October 21, 2007, housing 328 men. Franklin School, on that same night, held 300 men (125
percent of its capacity), filling all 240 year-round beds and all 60 seasonal (hypothermia) beds,
which are generally cots set up for overflow guests. 801 East was at 111 percent, swelling to 333
clients. The District, TCP, and providers are literally pushing shelters to their bursting point and
it is, as one interviewee put it, “tearing these building apart from the inside.” Overcrowded living
situations create tension in shelters that takes shape in a number of ways. Of the Fair Hearings
cases filed to date, large emergency shelters top the list.
ƒ
CCNV has had 96 cases brought to either an Administrative Review or a Fair Hearing,
representing 33 percent of the cases thus far, and
ƒ
D.C. Village (now closed) comes in second, responsible for 12 percent of the cases. 5
Shelters are continually keeping seasonal hypothermia beds open year round. While this is a
problem of funding for major repairs and availability of shelter space, it is also an issue of
project size and the number of beds packed into District contracts and individual sites. Serious
changes must be put in place to reduce crowding and to staunch the further destruction of these
District buildings.
COMPLIANCE WITH THE ADA
The District-owned buildings used as emergency shelters are seemingly in violation of two key
aspects of District and federal laws requiring appropriate accommodations for people with
disabilities. First, some individuals with disabilities require noncommunal settings, 6 but District-
5
See chapter 6 for Fair Hearings data and a description of the process.
6
Amber W. Harding at the Washington Legal Clinic for the Homeless noted in her report titled “Americans with
Disabilities Act Compliance of the District of Columbia’s Emergency Shelter System” that four types of disabled
individuals could require noncommunal emergency shelter:
1.
Persons with physical health conditions that will worsen in communal settings such as those with immunesuppressed disorders (HIV/AIDS, lupus, some cancers), those who require a sterile environment due to
Chapter 2: Responsibilities for Emergency Shelters
10
owned emergency shelters cannot provide such settings, being stacked wall to wall and floor to
ceiling with triple-decker bunk beds. Second, D.C. General’s hypothermia provisions for
families violate D.C. law 7 stating that all shelters for families (households with children) must be
apartment-style housing.
With that said, ADA compliance has significantly improved at DHS, and TCP continues to run
trainings on ways of providing reasonable accommodation. Under Modification 15 of the TCP
DHS Contract, $495,000 was added for ADA-related repairs and maintenance, increasing the
total funds allotted for ADA compliance to $1,077,775. Steps have been taken to improve the
District’s shelter system for those with disabilities. More, however, is needed before the District
will be truly in compliance, including the creation of additional units of noncommunal
emergency shelter for single disabled people and accessible units for families.
OFFICIAL ARRANGEMENTS FOR MAINTENANCE
Maintenance problems in District-owned buildings, we found, could mostly be grouped into
three categories:
ƒ
ƒ
ƒ
Timeliness and response,
Accountability, and
Monitoring of repairs.
This chapter deals with timeliness and adequacy of response. Chapter 5 addresses issues of
accountability and monitoring.
UPKEEP OF CURRENT EMERGENCY SHELTER FACILITIES
In some circumstances, funds have been allocated for major overhauls of struggling facilities.
CCNV, for example, has recently started comprehensive repairs on both structural and
nonstructural problems, funded by Capital Improvements’ Homeless Shelter Pool. Major repairs
at Gale Shelter are also underway or planned, along with changes at Emery and Madison. The
recent surgery or wound care concerns, or others at high risk for infection and communicable diseases
(such as premature babies with health complications);
7
2.
Persons with conditions that may be contagious or can pose a threat to other residents, such as active
tuberculosis, who would otherwise be excluded from services if not served through a noncommunal
placement;
3.
Persons with mental health conditions that are exacerbated by a communal setting such as agoraphobia,
social anxiety disorder, or paranoid schizophrenia; and
4.
Persons with dietary, medicinal, or other accommodation needs that are difficult to meet in a communal
setting, such as the need for frequent or specific meals (where food storage is not possible), insulin
dependence, the need for TTY/TTD to receive phone calls, the need for the environment to remain
unchanged for a blind person, or the need for privacy for caretaking or nursing needs (p.3).
D.C. Code § 4-753.01(d).
Chapter 2: Responsibilities for Emergency Shelters
11
repairs will likely total over $9 million for Gale and $7.5 million for CCNV. Rarely, though, are
funds of this magnitude allocated to District-owned facilities and it is clear that in order to truly
fix some of the larger shelters, this much or more is needed for each. What is more, Capital
Improvements is notoriously slow moving and, by most accounts, will not be done any time
soon. For now, funds are generally provided to bring these buildings up to a reasonable state of
repair and functioning but not actually to fix them. It is unclear how much would ultimately be
needed if fixing were the goal. Before making such an investment the District would have to
consider whether an investment of this magnitude might better be made in a restructured system
of emergency shelter once the long-term shelter stayers are accommodated in PSH.
Contract Language Regarding Maintenance Responsibilities
Fortunately, short term maintenance procedures—noncomprehensive repairs—are beginning to
show signs of real improvement within the District. This is not to say there are no problems, but
some of the most serious underlying troubles have been addressed, according to the providers
that run these facilities. Here we outline the current maintenance procedures, commenting on
what works and what does not. Issues of accountability and specific problems with contract
language are addressed in chapter 5.
Maintenance responsibilities under the recently amended DHS-TCP contract are as follows: for
all District-owned shelters, the District (meaning OPM, which has fully taken over this function
from DHS’s Facilities Management Operations Division (FMOD)) is responsible for overall
structural maintenance of the shelters, including major facility upkeep and repairs such as:
ƒ
ƒ
ƒ
ƒ
Roofs,
Outside plumbing,
Exterior and foundation troubles, and
Major heating, ventilation, and cooling systems.
TCP then covers smaller, nonstructural upkeep such as:
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
Interior windows,
Interior doors and locks,
Ceiling tiles,
Damaged floor coverings,
Interior painting,
Interior plastering,
Dry walling,
Interior plumbing,
Fixtures,
Light bulbs,
Portioning,
Moving and hauling, and
Furniture. (C.3.1.1.260)
According to most providers that relied on TCP for small maintenance tasks, TCP has provided
prompt service for nonstructural emergency maintenance, and occasionally will even help the
District with its maintenance responsibilities to speed the process along. The latter happened
Chapter 2: Responsibilities for Emergency Shelters
12
with repairs to D.C. Village Cottages 4A, 4B, 5A, and 5B. In these circumstances, DHS’s
contract with TCP is amended to give TCP the resources and authority to undertake the repairs.
Are Repairs “Emergency” or “Regular”?
It must be noted that some shelter providers say they still encounter response issues when
dealing with TCP. These occur most when there are questions as to whether the repair is an
emergency. Some of these instances occurred when FMOD was still handling requests for major
repairs and TCP would refer emergency requests to FMOD. It happened that FMOD did not
agree that something was an emergency, and thus did not respond as quickly as the provider
desired. Now that FMOD’s responsibilities and funds have been transferred to OPM, delays of
this type should not occur, and providers say that response time has improved considerably.
There are times, though, when the requests for repairs inundate TCP’s maintenance system—too
many come in for TCP to be able to deal with everyone immediately. Such system overloads will
likely require more resources and perhaps more preventive maintenance. One implication is that
TCP should work out a triage system for incoming repairs. Depending on the severity of the
problem and the availability of resources, TCP should be able to rank priorities and work on the
top priorities first.
When Do Shelters Use Their Own Funds for Repairs?
A major problem for some shelters—most specifically Adam’s Place—is the ambiguity
surrounding who is responsible for upkeep. Even though the contractual language has largely
been cleared up in this area (see chapter 5), the majority of emergency shelters have some money
in their budget for repairs (all but Harriet Tubman and Franklin) and Adam’s Place is not
“District-owned,” in the sense that the District only holds the building’s lease. Some shelters had
as much as $25,000 allocated for maintenance and repairs. In interviews for this study, shelter
providers reported being confused regarding when they would use funds in their own contracts
and when TCP or OPM would be obligated to step in. TCP and the District must sit down with
the agencies that operate these shelters and clear up any misunderstanding of who is responsible
for what.
Improvement in Maintenance Performance
The District’s ability to provide structural upkeep is improving. Since January 2007, the Office
of Property Management (OPM) has increased its support to DHS/FMOD for shelter repair
responsibilities. Today, OPM has completely taken over the structural maintenance process and
is waiting for a Memorandum of Agreement between OPM and DHS to clear the necessary
administrative processes. The responsibilities and additional funds have already been transferred.
At the time of our interview, OPM’s Deputy Director of Portfolio Management took pride in his
agency’s ability to get things “fixed quicker.” To a large extent, our interviews give evidence
that this perception is true, with most of the providers finding the move to OPM refreshing.
Emergencies, they say, are now responded to within days and the entire process, from making
the call to when the problem is fixed, can be considered timely and efficient. As an example, one
now sees air conditioning vents traveling up the sides of Franklin School—a shelter previously
deemed “impossible” to air condition by some in FMOD. Nonemergency structural repairs have
also shown signs of improvement according to providers, but there is little evidence to draw from
as the transition is relatively recent.
Chapter 2: Responsibilities for Emergency Shelters
13
An equal number of interviewees had not noticed any difference after the switch to OPM, at least
not yet. One provider noted that OPM did all of its maintenance including minor interior repairs
that are generally the responsibility of TCP. Evidently, there is still some confusion on an
individual level as to who is responsible for what, but the contractual language has been cleared
up and both TCP and OPM have shown signs of real improvement.
Maintenance Comparisons and Overall Problems with Upkeep
All privately owned buildings used as emergency shelter except Park Road are responsible for
their own maintenance and upkeep. When comparing District maintenance procedures to those
offered by private landlords, the differences reported were minimal. If anything, OPM and TCP
received higher rankings than private landlords or maintenance contractors. Interviewees
mentioned the same problems for both FMOD and privately run maintenance, citing a lack of
response or slow movement on certain repairs as the biggest issue. As an example, at Park Road,
TCP holds the lease for the shelter, making it responsible for the shelter’s upkeep even though
the facility is not owned by the District. Park Road’s maintenance and property management is
contracted out to a company called Visions. Interviewees did not report significant
improvements in the quality of maintenance or client satisfaction; some interviewees found the
property management provided by Visions to be significantly slower than what other providers
reported for FMOD. For the most part, OPM and TCP were able to escape these criticisms.
IMPLICATIONS
The issues of timeliness and response were alleviated to a notable extent with the move to OPM.
We heard from many providers that TCP is able to move quickly for more minor repairs,
although there was a slowdown during the few months that it was preoccupied with closing D.C.
Village. Monitoring reports discussed in chapter 4 note delayed responses with DHS (FMOD),
OPM, and TCP, but interviewees consistently noted positive changes within the last year. In all,
things are moving forward in the area of facility maintenance.
The array of problems with the oversized, schoolhouse type shelters left as the legacy of previous
city administrations strongly suggests an overhaul of the entire system, shrinking the need for
emergency shelter through development of PSH for long-term shelter stayers and shifting the
remaining emergency system toward more manageable settings for homeless services, as we
discuss in the second report of this assessment. The new Mississippi Avenue Shelter, currently
finishing up its planning stage, is a step in the right direction with its 19 unit capacity. But the
planning process has already taken over two years and construction is unlikely to be completed
within the coming two years. For the short term, the District is faced with the constant and
expensive job of upkeep for the emergency shelter system. While the process itself is, by most
accounts, still underfunded and lacks any sort of preventive maintenance, improvements have
been made.
Other implications of our findings with respect to emergency shelter conditions are:
1. Any ambiguities over maintenance responsibility for shelters with maintenance funds in
their budgets need to be cleared up between TCP, DHS, and individual providers.
Chapter 2: Responsibilities for Emergency Shelters
14
2. It is unlikely that contracting out maintenance and repair responsibilities to private
companies would increase the quality of District-owned shelters.
3. Noncommunal emergency shelter units are needed if the District is to be in compliance
with federal and local ADA or other disability laws.
4. TCP might create a formal triage system to insure efficiency during times of increased
maintenance requests.
5. A reduction of clients in District shelters is needed to stop the dilapidation of facilities
and to raise the quality of homeless services.
6. More funds are needed to provide comprehensive repairs to many of the District-owned
emergency shelters. But care should be taken in committing massive funding because
plans to address the larger issue of system transformation and restructured emergency
shelters might change the priorities for how such funds should be used.
Chapter 3: Orchestrating the Continuum of Care
15
Chapter 3: Orchestrating the Continuum of Care
HIGHLIGHTS
ƒ
TCP oversaw the shift from a concentration of public funding at the emergency
shelter level to more transitional and permanent supportive housing (from an 80/20
percent distribution in 1994 to a 46/54 percent distribution today).
ƒ
TCP runs the annual Continuum of Care planning and application process that
brings tens of millions of federal dollars into the District every year to support more
than 40 transitional and permanent supportive housing projects.
ƒ
Serious leadership is needed in the Continuum of Care. TCP provided the leadership
and staffing to orchestrate the early planning for a CoC. But events since 1999 have made
it unlikely that by itself it could assume that role again.
ƒ
In 2007, TCP managed Shelter Plus Care resources that supported about 800 people
in housing with supportive services to help them stabilize and remain housed. Close
to 350 individuals and 450 parents and children in formerly homeless families benefit
from these resources.
ƒ
In general, training and technical assistance were given good marks, with the
majority of the interviewees finding them helpful and a few finding them, at worst,
tedious. Recently, at provider urging and DHS concurrence, TCP has added a number of
service oriented training sessions.
ƒ
TCP has made significant improvements on its communication practices, but still
falls short of complete success. TCP communicates effectively to providers with whom
it contracts, through a well organized email list-serve and “word of mouth” network. But,
perhaps because it does less of the whole-community organizing and planning than it
once did, it is less effective in branching out to the public, to service providers and
advocates outside the public homeless system, and to government agencies other than
DHS.
ƒ
TCP has skills that are valuable for the District and its homeless assistance
providers. TCP can be more flexible and more nimble than government agencies. It can
respond more quickly. It knows the homeless system intimately, and knows how to make
things happen. It knows government agencies intimately (both DHS and HUD), and can
get modifications and new ideas through those systems better than anyone else. Finally,
its focus is completely on homelessness, so it does not get pulled off of its work on
homeless issues to attend to five or six other issues on its plate.
This chapter covers many topics, all related to the structure and operations of the Continuum of
Care (CoC) managed by The Community Partnership (TCP). It begins with a brief historical
review to apprise the reader of how the current system got to be the way it is. It looks at the
scope of the current system and describes TCP’s various roles relating to its development,
focusing especially on projects other than emergency shelters (which were the topic of chapter 2)
Chapter 3: Orchestrating the Continuum of Care
16
and on the functions that help the homeless assistance network to work in some ways as a
system. 8 These functions include handling the annual application to the U.S. Department of
Housing and Urban Development (HUD) for funding to support many of the District’s
transitional and permanent supportive housing projects, developing training sessions on a wide
variety of topics and offering them to staff of provider agencies at every level, maintaining a
website, and publishing periodic reports on system functioning and homelessness in the District
to aid communications around homelessness and the homeless assistance network. TCP’s role in
maintaining a Homeless Management Information System (HMIS) is described in chapter 7, and
its data gathering and analysis activities pertinent to policy decision making is described in detail
in chapter 8.
THE D.C. INITIATIVE AND DEVELOPING THE CONTINUUM OF CARE
To put the scope of TCP’s current activities into context, we briefly review how TCP came to be
in the position it currently occupies, as intermediary between the Department of Human Services
(DHS) and homeless assistance providers. TCP changed from a small nonprofit advocacy
organization to the agency “in charge” of homelessness in the District in 1994, when HUD
selected it to receive a five-year, $20 million grant known as the D.C. Initiative.
BEFORE THE D.C. INITIATIVE
Before the D.C. Initiative began, city government funds (no less than $22 million annually) were
being devoted almost entirely to maintaining about 3,500 emergency units, including motel
rooms, for single men, single women, and families. 9 Virtually no case management was
available in city emergency units, whether these units served single men or women or homeless
families. Several arrangements, such as the Capital City Inn, that the city used to shelter
homeless families were so bad, so dangerous, and so hopeless, especially for children, that they
led to a major public scandal.
The nonprofit (private) community at that time offered almost as many emergency beds for
adults as were paid for by the city. Further, the private service community supplied more than 10
times the transitional shelter capacity offered by the District. The city paid for 52 percent of the
emergency beds available in the District, but only 7 percent of the transitional units. Permanent
supportive housing for disabled homeless people was even more skewed to providers outside the
public system.
In 1994, TCP was a small nonprofit advocacy organization. The year before, The Clinton
administration’s first HUD Secretary, Henry Cisneros, had decided he wanted to showcase the
District and support it to develop a Continuum of Care—the first jurisdiction in the country to
start on this endeavor, and to receive a large infusion of HUD funding to make it happen. A large
part of HUD’s Continuum of Care concept, and thus of the D.C. Initiative’s design, consisted of
8
We reserve to the second report for this assessment a detailed discussion of the ways that the current array of
programs and funders does not function as a system and what might be done to increase the level of organization
and goal orientation.
9
This section relies heavily on work first reported in Burt (1995), an evaluation of the D.C. Initiative’s first years.
Chapter 3: Orchestrating the Continuum of Care
17
shifting the emphasis of the public system away from emergency shelter and toward the
transitional and permanent supportive housing projects that HUD saw as more likely to help
people leave homelessness.
District government was in sufficient disarray at the time (it would shortly come under the aegis
of the Financial Control Board) that HUD did not want funding for the D.C. Initiative to be
under District government control. It insisted that a nonprofit agency be selected for the task,
modeling the concept on Columbus, Ohio’s successful Community Shelter Board. TCP was
selected to be that nonprofit, becoming the “agency in charge” of homelessness in the District in
1994.
DURING THE D.C. INITIATIVE
During the D.C. Initiative, TCP supplied the only leadership available in the District related to
homelessness. Shortly after the D.C. Initiative began, the Financial Control Board took over
government functions and perforce concerned itself with other matters, such as the city’s nearbankruptcy circumstances. After a year of negotiations among HUD, DHS, and TCP following
the official start of the D.C. Initiative, in May 1995 DHS formally transferred all authority and
funding for homeless shelter and services to TCP for both single adults and families. 10 A few
months later (November 1995), federal Shelter Plus Care and HOME dollars were transferred
from the District to TCP management. Thus it took more than a year after the District made a
commitment to the D.C. Initiative for TCP to achieve the level of authority and resources that
were part of the D.C. Initiative’s design.
TCP organized and ran the D.C. Initiative, starting the Continuum of Care process and beginning
the shift from a concentration of public funding at the emergency shelter level to more
transitional and permanent supportive housing (from an 80/20 percent distribution in 1994 to a
34/66 percent distribution today), all as part of the mandate that accompanied the D.C. Initiative.
It orchestrated community meetings to identify gaps and unmet needs. It established and
supported provider groups (for providers doing outreach or serving single men or families) and
built on a pre-existing group of providers that served single women; these groups met monthly
for a number of years and provided the venues for discussing service problems, bottlenecks, and
options for serving clients better. Several of these groups continue to meet, in one form or
another. In short, TCP generally carried on during the period when District government basically
abdicated responsibility for homelessness and the Financial Control Board was in charge.
At the end of the D.C. Initiative, the public system in the District looked a lot different than it
had in 1994. Figures 3.1 and 3.2 show that shift between 1994–95 and 2000–2002, as well as the
system structure as of January 1, 2008. 11
10
The funding that came with this agreement ($6.9 million) was significantly less than had been promised ($14.8
million), which in itself was only two-thirds of what the District had spent on emergency shelter annually when it
was in charge.
11
Figures for the 1994–1995 and 2000–2002 periods are taken from Burt (2002).
18
Chapter 3: Orchestrating the Continuum of Care
Figure 3.1: Public and Private
Commitments by Program Type: 1994-2008
Public (in HMIS)
Private (not in HMIS)
4000
N umber of beds
452
388
3000
300
928
1194
2000
3099
1000
1250
2946
820
1510
1896
2800
1548
1654
TH
PSH
1997
1329
600
195
0
ES
TH
PSH
1994-1995
ES
2000-2002
ES
TH
PSH
2008
Source: Burt (2002) for 1994-1995 and 2000-2002 data; Urban Institute analysis of Housing Inventory Chart and TCP data for January 31, 2008.
Note: “ES” = emergency shelter beds; “TH” = transitional housing program beds; “PSH” = permanent supportive housing program beds.
Figure 3.1 shows the number of beds offered through public and private investment before the
D.C. Initiative began (1994–1995), after it was over (2000–2002), and today. In each period, the
distribution is shown separately for emergency shelter (ES), transitional housing (TH), and
permanent supportive housing (PSH). In 1994–1995 there were 3,099 emergency shelter beds in
the public system, while private agencies offered only 452. The distribution was reversed for
permanent supportive housing beds, with the public system offering only 195 while the private
system offered 1,510. By the end of the D.C. Initiative, emergency shelter provision had not
changed much, but the District had almost tripled its number of transitional housing beds and
increased by eightfold the number of permanent supportive housing beds it offered. Growth in
transitional housing beds had stabilized at about the level achieved by 2002 until the System
Transformation Initiative accompanying closure of D.C. Village added 660 more transitional
beds to the system. Both the public and the private systems have continued to add permanent
supportive housing beds while the number of emergency shelter beds has shrunk slightly overall.
Figure 3.2 arrays the same information differently to answer a different question—how is public
and private investment distributed across project types and how has that distribution changed
over the years. As is obvious from figure 3.2, the District, unlike the private system, was and still
is heavily committed to emergency shelter beds. As already noted, however, thanks to increased
commitments to transitional and permanent supportive housing beds, the proportion devoted to
emergency shelter has declined from 80 to 34 percent between 1994–1995 and 2008.
19
Chapter 3: Orchestrating the Continuum of Care
Figure 3.2: Changes in the Types of Programs
Supported by Public vs. Private Investment:
1995-2008
ES
TH
PSH
Number of beds
6000
1654
1329
1548
1997
5000
4000
195
600
3000
1250
1510
2000
3099
1000
0
Public
2946
2800
820
1194
452
388
Private
1994-1995
Public
1896
Private
2000-2002
928
300
Public
Private
2008
Source: Burt (2002) for 1994-1995 and 2000-2002 data; Urban Institute analysis of Housing Inventory Chart and TCP data for January 31, 2008.
Note: “ES” = emergency shelter beds; “TH” = transitional housing program beds; “PSH” = permanent supportive housing program beds.
“Public” = in HMIS; “Private” = not in HMIS.
AFTER THE D.C. INITIATIVE
When the District emerged from the Control Board period and the Williams administration
began, it partially reasserted a government role. 12 Spurred on, in part, by advocates, the Williams
administration asserted that TCP “works for us, we tell you what to do, you do what’s in your
contract, you have no authority to do anything else.” This approach and subsequent actions
undermined the authority of TCP but did not relieve it of any responsibility, nor did DHS begin
to act in any way that could be described as providing leadership. TCP continued its role as
Continuum of Care coordinator, applicant on behalf of the community for HUD funds, developer
of new projects and approaches indicated as a need by the Continuum of Care planning process,
implementer of ideas such as the Community Care Grants, conductor of trainings, administrator
of the HMIS, and pass-through agency conveying District, HUD, and other funds to providers
through upwards of 140 contracts annually.
Nor did the Williams administration exert any consistent or fruitful leadership of its own related
to homelessness (see Burt 2002). Even its ten-year plan, Homeless No More, went nowhere for
the year after its release that remained of the Williams administration.
12
Material for this section was first reported in Burt (2002), in an assessment for the FannieMae Foundation.
Chapter 3: Orchestrating the Continuum of Care
20
It is now six years later and a new Mayoral administration is in charge. Addressing homelessness
is high on its agenda. The question for District is, who, or what, is finally going to provide the
centralized, organized, efficient, effective, and visionary leadership that is needed to greatly
reduce or end homelessness? The further question pertinent to this evaluation is what role TCP
will, or should, play in such a leadership structure, and which of its existing roles and
responsibilities should it continue to perform?
ORGANIZING THE CONTINUUM OF CARE
TCP runs the annual Continuum of Care planning and application process that brings tens of
millions of federal dollars into the District every year to support more than 40 transitional and
permanent supportive housing projects. This responsibility started with the D.C. Initiative even
before HUD began applying the Continuum of Care approach to funding to the rest of the
country in 1996 (Burt 1995, 2005).
Early on, an important goal of the D.C. Initiative was to change relationships among providers
and funders, shifting from adversarial to collaborative interactions to promote improvements in
the system. Most of the people involved in this process felt that relationships among providers,
and between providers and the funding authority (TCP), were significantly better than they had
been before (Burt 1995, 2005).
TCP began the process of creating a communitywide plan, convening four focus groups—one
each for providers serving single men, single women, families, and youth. Later on an outreach
group formed. Several of these groups still continue today in various forms. The groups were
charged with helping to plan the scope, structure, and nature of services needed for the
population they served. This was a critical experience in bringing providers together to decide on
goals for a homeless services system, and what the best system of care would be to achieve those
goals. The immediate result of these activities was an Operational Plan that guided activities
under the D.C. Initiative.
The process begun in those early days evolved into the annual planning and strategizing
activities that precede and are required by the annual application for HUD funding. Every spring,
TCP facilitates a process that includes an analysis of gaps and unmet needs in the system, a
performance assessment and priority ranking procedure for applications to be submitted for
funding, and a community meeting to discuss the needs and rankings, writing the common parts
of the application and assembling the individual project applications, and submitting to HUD.
In the first six or seven years of this process, the gaps analysis component involved widespread
participation of community members, providers, advocates, and others in several community
meetings and committee work. In those days, agencies were applying in three-year funding
cycles. The way HUD structured the CoC funding had the effect of allowing highly organized
communities that began participating in the Continuum of Care process early on, such as the
District of Columbia, to design and develop new projects every year, often receiving two or even
three times as much funding as their initial “share” (as determined by HUD) indicated they
would get. Planning had some real meaning, and some real payoff, at that time.
Chapter 3: Orchestrating the Continuum of Care
21
In recent years, however, more and more communities participate in the CoC funding cycle and
use “their” shares; no longer are any shares “left over” for redistribution. Most communities have
long since seen the vast bulk of their HUD funding going to renew grants for already-operating
projects rather than to create new projects. Funding cycles have shifted from three- to one-year
grants, to be able to fund the largest number of projects, and “planning” has become at best an
interesting exercise rather than an activity for which one can expect to see a payoff, at least as far
as the HUD application is concerned.
One of the consequences of this transition is that TCP has developed a faster approach to
developing the annual application. One could see this speed as “streamlining,” or as “truncating.”
For stakeholders interested in getting their HUD money with the least time investment, the new
process is “streamlined” and “smoother.” For stakeholders who want more thinking about where
the system is and should be going, and who want to be part of that thinking, the new process is
“truncated” and “limited.”
Still, almost everyone interviewed said they would rather be working on this process with TCP
than with District government agencies. The history of government efforts in the years since the
end of the D.C. Initiative to “get organized” on the homeless front is one of starts and stops,
promises and failure to follow through on those promises. In all of these activities, stakeholders
with many different interests in the homeless issue, from providers to advocates to business
people to homeless and formerly homeless people themselves came to meetings, thought about
what was needed, planned the future—often despite initial skepticism based on past experience
that in all probability their effort would not make a difference. In every instance their initial
skepticism was justified and government follow through was weak or absent.
Highlights from these years include the Deputy Mayor for Children, Youth, and Families’
process of developing and then failing to act on a strategic plan; organizing and then publishing a
10-year plan to end homelessness and then not implementing it, and creating an Interagency
Council on Homelessness (ICH) and not mobilizing it until a new administration took office.
ƒ
A Strategic Plan for 2000 to 2004 was developed to provide guidance for the post-D.C.
Initiative system. The idea was that the District government was taking back control, and
that TCP was “no more than a contractor to DHS and needed to take its lead from DHS,”
to paraphrase the way several interviewers described what happened. TCP and the
Deputy Mayor’s office jointly held meetings and worked on the plan. The plan set up
committees to address a number of difficult issues around homelessness (e.g., client
rights, the legislative basis for the District’s expenditures on homeless assistance services
beyond emergency shelter, the fate of various 12-hour shelters). But the plan never
emerged as a document presenting a coherent overall strategy for the future of the
homeless assistance system. After an initial flurry of activity, meetings to develop the
plan stopped. A plan was eventually written which, along with various committees and
commitments, called for a coordinator of homeless-related activities across six
government agencies as well as a “homeless budget” to orchestrate and rationalize the
homeless-related activities of the same agencies. The “homeless budget” and coordinator
never happened.
Chapter 3: Orchestrating the Continuum of Care
22
ƒ
The 10-year plan, Homeless No More, was published in fall 2005, when the Williams
administration still had one more year in office. No significant implementation had begun
under the plan by the time that administration ended.
ƒ
The Homeless Services Reform Act, passed in November 2005, created an Interagency
Council on Homelessness. During the Williams administration this council was
essentially inactive.
During all these years, TCP still had responsibility for orchestrating the District’s homeless
assistance system, but with considerably undermined authority and with no government agency
assuming responsibility for and acting to provide system wide organization, leadership, or
accountability. Advocates and some providers often express frustration that TCP cannot “make
government agencies do what they should,” despite the fact that some of them were the same
people who challenged TCP’s right and authority to think and act independently and insisted that
it was a contract agency like any other.
THE CURRENT SITUATION AND THE ISSUE OF LEADERSHIP
For the reasons just described, the analysis and planning process that once preceded the annual
application for HUD funding, and that TCP led, has shrunk to the point that people interested felt
it no longer appears to fulfill the need for community participation, strategizing, thinking,
reflecting, assessing, and evaluating where the system as a whole has been and where it is and
should be going. When the first draft of this report was prepared in October 2007, the community
did not have a venue that provided the opportunity for broad, organized, and continued
contributions to the vision of what should be and how to get there. Nor did it have an obvious
leader whose primary job is homelessness and how to end it. Possible candidates for that
leadership role were the ICH, TCP, or some combination or outgrowth of the two—but there was
no doubt that serious leadership was needed, and that widespread community involvement had to
be part of the package.
TCP once provided the leadership and staffing to orchestrate the early planning for a Continuum
of Care. But events since 1999 have made it unlikely that by itself it could ever assume that role
again. TCP is, however, an excellent implementer, as its work on Shelter Plus Care, the HMIS,
financial management, and most recently its activities on behalf of closing D.C. Village attest. It
is also an excellent designer of projects to fill an identified service gap—examples include
Community Care Grants for families and Pathways to Housing for chronically homeless single
adults. It is also very good at organizing and sustaining community participation, even if it has
not been concentrating on doing that under its post-D.C. Initiative contracts with DHS. Finally,
its long experience working with District and federal government offices would be useful in any
organizing/implementing roles it is given.
TCP’s “in-between” status as the fiduciary arm of DHS (and DHCD) with respect to homeless
assistance coupled with its nonprofit status has been extremely useful for the District and can
continue to be useful in the future. TCP can be more flexible and more nimble than DHS or any
other government agency. It can respond more quickly. It knows the homeless system intimately,
and knows how to make things happen. It knows government agencies intimately (both DHS and
HUD), and can get modifications and new ideas through those systems better than anyone else.
Chapter 3: Orchestrating the Continuum of Care
23
These skills are valuable for the District and its homeless assistance providers. Finally, its focus
is completely on homelessness, so it does not get pulled off of its work on homeless issues to
attend to five or six other issues on its plate. All these TCP characteristics mean the District
should keep it around.
THE ICH AND DISTRICT LEADERSHIP
The ICH membership includes representatives of various stakeholder communities in addition to
representatives of government agencies and elected officials—that is, the people who will have
to be working together to make the changes needed to end homelessness. The ICH began to meet
in spring 2007; by early summer one of its committees had taken on the specific high-priority
task of closing D.C. Village (the final family moved out on October 26, 2007). The ICH then
created the PSH Work Group, a subcommittee of its Strategic Planning Committee, to develop
and implement a framework for creating the 2,500 units of permanent supportive housing called
for in Homeless No More. The PSH Work Group issued a draft plan for public comment in early
April 2008, and is proceeding with implementation steps. Other ICH committees are involved in
planning and implementing other changes that will contribute to ending homelessness. The ICH
appears to be well on its way to providing the broad ongoing forum for careful deliberation and
development of community-wide collaborative relationships that the District needs if it is going
to end homelessness.
TCP is part of the ICH and has been deeply involved in the ICH’s major efforts to date. While it
is no longer in the position to be “the leader” on homelessness as it was in the mid-1990s, it does
have highly developed implementation skills. These skills have been put to good use in ICH
efforts to date, and may be expected to continue to be helpful in bringing ICH ideas to fruition.
MANAGING SHELTER PLUS CARE RESOURCES
Shelter Plus Care is a federally funded program that provides permanent “housing plus”
supportive services for hard-to-house homeless individuals and families with a disability—
usually severe mental illness, a chronic substance abuse problem, or both (dual diagnosis), or
those living with HIV/AIDS or a physical disability. In 2007 TCP managed Shelter Plus Care
resources that supported about 800 people in housing with supportive services to help them
stabilize and remain housed. Close to 350 individuals and 450 parents and children in formerly
homeless families benefit from these resources.
To make the District’s Shelter Plus Care resources stretch as far as possible, TCP has worked
hard over the years to develop relationships with private landlords whose units house the vast
majority of households using Shelter Plus Care certificates. When new households apply for and
are accepted into the Shelter Plus Care program, their first task is to find a housing unit then TCP
has to inspect it to be sure it qualifies for the program (is within rent guidelines and meets quality
standards), after which the household may move into its new unit.
TCP has relationships with more than 200 landlords and is able to facilitate the process of
finding apartments that prospective tenants are able to afford. TCP also negotiates rents, and is
sometimes able to obtain reduced rents because landlords have come to trust that households
placed by TCP have the supports in place to avert housing crises. Landlords know that if they
Chapter 3: Orchestrating the Continuum of Care
24
call TCP when a tenant appears to be getting into trouble, TCP will help resolve the issues,
almost always succeeding in keeping the tenant in housing. Landlords appreciate this type of
support, as vacancies are costly to them and TCP’s assistance helps keep vacancies to a
minimum.
The strength of the relationship TCP has with landlords was tested recently when the Mayor’s
office decided to close D.C. Village with very little lead time to make arrangements to move
resident families to other housing. Appealing to its network of landlords, within a couple of
weeks TCP was able to identify more than 200 units of vacant and available housing; it took only
a few more weeks to get commitment of these units for the families moving out of the District’s
family emergency shelters. DHCD and DCHA also identified available units that were used in
the D.C. Village closure. TCP’s activities in the matter of closing D.C. Village, in collaboration
with DHS, DHCD, DCHA, staff of the Mayor’s office, and providers in the District’s family
emergency shelter system, is a good example of its ability to implement policy decisions. It
knows the community, is able to mobilize quickly, and can act with flexibility.
The Shelter Plus Care program provides another example of TCP’s administrative acumen.
Shelter Plus Care resources are usually described in terms of the number of housing units to be
subsidized. But in reality Shelter Plus Care comes to a community as a fixed amount of dollars.
Creative administrators can make those dollars stretch to cover more than the official number of
units authorized under the Shelter Plus Care grant, usually by negotiating lower rents with
landlords with whom the administrator has a long-standing relationship. In 2006, TCP did just
this, obtaining enough savings from negotiated lower rents to be able to house an additional 74
families (94 adults and 166 children) with significant disabilities.
TRAINING AND TECHNICAL ASSISTANCE
Service providers we interviewed lauded TCP’s trainings, meetings, and forum engagements
designed to increase skills and assure the proper functioning of various aspects of the system.
Communication about training was seen as clear, timely, and effective. Interviewees mentioned
prompt email notification and constant monitoring of shelter staff turnover so providers can be
notified of training when new employees come on board.
The most common complaint about TCP trainings was that they focused too much on
administrative topics such as HMIS. We did find that TCP provides a lot in terms of
administrative trainings, including seminars on data entry, data quality, and report functions for
the HMIS; what providers have to do to be in compliance with the HSRA and Americans with
Disabilities Act (ADA); how unusual incidents and suspensions, transfers, and expulsions must
be handled to be legal; accounting requirements for subcontract recipients; and similar issues.
Some trainings are “mandatory,” including all that relate to legal compliance.
Depending on the topic, training will be geared to provider staff at different levels. There is an
annual contractors meeting that all agency and project directors attend, trainings on HMIS data
entry that data entry clerks attend; trainings on client rights and responsibilities that caseworkers
attend, and so on. Recently, at provider urging and DHS concurrence, TCP has added a number
of service oriented training sessions, including:
Chapter 3: Orchestrating the Continuum of Care
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
25
Emergency preparedness, including “ongoing coordination of services operations and
safety”;
Customer service;
Transgender clients;
Housing discrimination;
Dealing with problem clients;
Hypothermia;
Conflict resolution;
Health, including tuberculosis;
Best practices;
Mental illness; and
Resources for family providers.
Providers also mentioned supportive technical services from TCP that they found extremely
helpful. These include assistance with contracts, funding applications, dealing with landlords,
budgeting, managing files, HMIS, HSRA and program rules, ADA requirements and how a
specific property can come into compliance, federal reporting, and generating statistics for their
own use.
In general, training and technical assistance were given good marks, with the majority of the
interviewees finding them helpful and a few finding them, at worst, tedious. If anything,
respondents desired more training, not less.
COMMUNICATIONS IN GENERAL
Unbeknownst to some, TCP activities and services are numerous and broad in scope. But
whereas contracting and other fiduciary responsibilities garnered TCP high marks, dissemination
of its work to the community has historically been its weakest point. Reports by the Enterprise
Foundation (1998) for the Financial Control Board and Bass and Howes (2000) for the Deputy
Mayor for Children, Youth, and Families both concluded that the community’s homeless
advocates, and some of its providers, felt TCP had too much control and made decisions without
adequate communication. These reports tempered their conclusions with documentation of the
many ways that TCP does structure communication and consultation, but still noted that
dissatisfaction persists.
Today, it seems that the Partnership has made significant improvements on this front, but still
falls short of complete success. In all, TCP communicates effectively to providers with whom it
contracts, through a well organized email list-serve and word-of-mouth network. But, perhaps
because it does less of the whole-community organizing and planning than it once did, it is less
effective in branching out to the public, to service providers and advocates outside the public
homeless system, and to government agencies other than DHS. Indeed, some of the private
shelters we interviewed were not able to tell us what TCP was. Hopefully, this assessment will
serve as a reference for community members to explain what TCP does for homeless services,
and ultimately clear up misconceptions about the Partnership’s roles and responsibilities.
Chapter 3: Orchestrating the Continuum of Care
26
IMPLICATIONS
1. TCP does many things well. Of the activities described in this chapter, these include
managing Shelter Plus Care resources, developing and conducting trainings, providing
technical assistance, and organizing the annual HUD application. Another chapter
describes its excellence in financial management. It is an excellent implementer of
programs and plans. It should be supported to continue doing these things—no other
entity would be likely to do them better, or as well.
2. Communications are still an issue on some fronts. Those less involved in the homeless
assistance system don’t know what TCP does and are frustrated by that. Others would
like to see more information forthcoming based on the HMIS—not just a single “Report
to the Community” once a year, but perhaps a discussion forum of what some data mean
and what they imply for where the community should be going.
3. TCP can serve the extremely useful role of implementer in partnership with the
Interagency Council.
4. Along with leadership the planning and implementation process needs to be broadened to
include many community stakeholders, and to be open to new ideas from new as well as
long-established actors in the homeless arena.
Chapter 4: Contract and Grant Financial Management
27
Chapter 4
Contract and Grant Financial Management
HIGHLIGHTS
ƒ
Responsibility for managing contracts and grants using District and federal monies
for the District’s public homeless system should remain under the charge of TCP.
Almost all CoC participants lauded TCP for its careful and prompt management of
financial services. Those who remember difficulties getting questions answered and
payment delays when contracts came through DHS or, even earlier, OESSS, were
especially interested in seeing this function remain with TCP.
ƒ
Procurement or budget policies need to be redesigned so that DHS as well as TCP
have clear mechanisms to assure accountability, while both still maintain flexibility
in a changing landscape. As the homeless population, best practices, and political will
shift over the funding cycle, DHS and TCP need to be able to develop new services and
projects and restructure old ones to insure efficacy. At the same time, DHS as well as
TCP needs to be party to mechanisms that assure provider accountability. Changes in
budgeting and TCP relations with the DHS Procurement Office will be needed.
ƒ
Contracts and RFPs need to be redesigned to insure proper homeless service
settings, maximum competition, and adequate support. Contracts and RFPs are, in
general, too large, forcing providers to take on more responsibilities than they can
adequately handle. Smaller providers choose not to compete for existing RFPs because
they know they cannot perform the services being requested for the money being offered,
and do not want to operate in the environment of the big shelters. As a consequence, TCP
relies too heavily on large providers that can attempt to handle the large number of beds
involved and that have the capacity to raise additional funds to supplement the TCP
contracts. At least one large provider has now declared its intention to get out of the mass
shelter business, which should force the issue of how to redesign the existing system.
This chapter covers TCP’s responsibilities with respect to issuing and monitoring contracts with
District homeless service providers and assuring prompt payment of invoices. It describes
contracting and payment processes, types of contracts, and options for increasing the
competitiveness of the RFP process. It also begins the discussion of the role that DHS currently
plays and might play in the future in contracting and assuring provider accountability—a
discussion that we continue in chapter 5.
TCP’S CONTRACTING RESPONSIBILITIES
TCP has responsibility for federal and District dollars for homeless services, administering:
•
•
•
90 subcontracts for DHS,
43 subcontracts for HUD, and
7 subcontracts for DHCD.
Chapter 4: Contract and Grant Financial Management
28
In addition, TCP is the administrative agency handling all of the rent subsidies for scattered-site
apartments that the District receives from HUD’s Shelter Plus Care program. As part of
managing this program, TCP’s accounting department deals with over 200 landlords and pays
rents for about 450 housing units within the Continuum, writing over 850 checks a month that
range from $300 to over $60,000.
For all practical purposes, TCP took over financial responsibility for the District’s public
homeless system in 1994, but it took DHS a year to sign over all the money for homeless
services. At that point, DHS retained no control over the money, as it was not a contract but a
transfer. The transfer lasted until 1999, when the D.C. Initiative ended, the Control Board turned
District government back to local control, and the District took over running its own affairs again
with the election of Anthony Williams as mayor. DHS then put the work that TCP had been
doing out for competitive bid and ultimately gave the contract to TCP—the only bidder. Since
then TCP has worked under a DHS grant or contract with strict controls over its responsibilities.
TCP-DHS CONTRACT FLEXIBILITY
Homeless services require flexibility in any Continuum of Care. Populations, political will, and
best practices are constantly changing the landscape of homeless assistance projects. When they
develop an idea, policymakers want something new to happen now. TCP has been the vehicle
through which such changes have been implemented with considerable speed and
responsiveness. Examples during FY07 include additional beds for homeless youth, converting
several spaces used as hypothermia shelters into year-round shelters, closing D.C. Village and
transferring all of its families to other accommodations, and contracting for food services. When
such situations occur, DHS and TCP negotiate a contract modification. If the money is already
available for these innovations, it takes TCP and its DHS contract officer at most 30 days to put a
contract modification in place and start to develop the new service or activity. If District agencies
need to find the money, however long that takes is added onto the 30 days, but it still takes 30
days or less to get started on a new service once the money is in hand.
Each time the TCP-DHS contract as a whole is negotiated, as DHS exercises its option on each
Option Year in the basic contract, the contract modifications created in the previous year come
up for discussion. If the modification covered a one-time activity there would be no need to
incorporate the modification into the next year’s contract. But if a modification has implications
for ongoing funding, as would be the case with needing to sustain support for the new youth beds
or the new family units, then the next contract must include the resources for those activities in
addition to support for the activities that were approved when the base contract was signed.
Many things can hamper this flexibility, including a lack of leadership, conflicts of interest, and
over reliance on a small group of providers, but DHS and TCP have been able to make it work
quite well over the years. One big bottleneck is the procurement procedure at DHS, which comes
into play once the program office has approved a contract and the accountants take over. There
have been times when this process has proved to be a yoke on the District’s, and therefore
TCP’s, ability to act in a timely manner. Interviewees noted that DHS’s financial office lacks a
sense of urgency for some of these changes and prolongs the process with its standard operating
procedures. These are the very procedures that homeless assistance providers experienced before
TCP assumed fiduciary responsibility for the public homeless system, and that providers are not
at all eager to experience again.
Chapter 4: Contract and Grant Financial Management
29
CONTRACT SIZE
Many providers and advocates interviewed for this evaluation expressed their frustrations with
large projects, mostly emergency shelters, in which the clients may number in the hundreds.
People express a desire for “smaller contracts,” but smaller contracts are not possible as long as
the system consists almost entirely of large warehousing shelters. Many respondents voiced
strong preferences for seeing the whole system change, eliminating the huge shelters where no
one can receive any serious attention to help end their homelessness and many languish from
year to year.
So the issue is not just “we need smaller contracts,” but “we need smaller projects.” Further,
these smaller projects need to be geared to specific issues to which providers can bring
specialized expertise, to help people gain and keep employment, overcome addictions, learn to
manage chronic illnesses, or whatever will contribute to their leaving homelessness and staying
housed.
District homeless service contracts are too large, because District shelters are too large. Yet
given the payment levels in District contracts for emergency services, agencies cannot run small
projects without going bankrupt. The current system structure and contract size discourage new
applicants and smaller more personalized services. Several providers said they had not bid on
specific services (e.g., taking over Spring Road, taking over cottages 4 and 5 at D.C. Village)
because they did not feel they could deliver a quality service for the price offered. It is also true
that the use of 200–500 person shelters has caused a serious degradation of service quality for
certain projects, not to mention liability issues. Below we discuss the way contracts contribute to
these negative characteristics and recommend possible changes in the contracts themselves.
CONTRACT FUNDING LEVEL
The issue of contract and grant size and the structure of the District’s emergency shelter system
must be taken up by both TCP and DHS, but ultimately by the Mayor and City Council. There
are few providers, whether of emergency, transitional, or permanent supportive housing, that felt
they could effectively meet their contractual obligations solely with funds awarded by the
District. By most accounts, additional fundraising is a necessity in fulfilling goals, with providers
needing to raise anywhere from 3 to 12 percent 13 over their TCP emergency shelter contracts to
provide the services they believe are needed, or in some cases just to come up to what they
consider the bare minimum.
The problem seems to be getting worse as funding amounts are not being adjusted sufficiently
for cost of living increases. In the end, some providers did not, or could not, supplement
contracts for all projects, in part or in whole. These interviewees noted that budget gaps were
closed with sub-competitive wages or low service levels. As shown in table 4.1, some shelters
reported not having any case managers. What is more, providers we interviewed who do
supplement find they are still not able to provide all the services their clients need.
13
There was at least one exception, with a provider needing to raise over 50 percent of their funds in order to
provide the level of services they deemed necessary,
Chapter 4: Contract and Grant Financial Management
30
Table 4.1 Low Caseworker Staff-to-Client Ratios in Emergency Shelters
Provider
Franklin Shelter
New York Avenue
Adams Place
Number of Beds
240
360
150
Number of Case
Managers (FTE)
1
3
0
Case Manager to
Client Ratio
1/240
1/120
0/150
801 East
D.C. Village
John Young
334
238
80
0
4.5
0.5
0/334
1/52
1/160
Harriet Tubman
CCNV
75
525
0.5
10
1/150
1/52
Source: Provider interviews and facility size data from Housing Inventory Charts.
Note: The number of beds in each shelter is artificially low here, as many emergency shelters are over capacity (see chapter 8).
“LOW BARRIER” SHOULD NOT MEAN “NO SERVICES”
Some maintain that operators of large low-barrier emergency shelters are not obligated to
provide more than a single case manager simply because clients are not required to participate
and most choose not to use case management. But this low service level is clearly not the
intention of the Homeless Services Reform Act, nor is it good practice. In fact, we disagree with
the notion that low-barrier shelters do not need as much staff time to work with clients—actually,
they need much more, because they should be working to engage residents in efforts to leave the
shelters rather than just ignoring everyone who does not seek out their services. Of the providers
we spoke to, almost all felt they could use more case management, and most put the optimum
ratio at around 15–20 clients per case manager. Indeed, the low case management level is more a
product of dwindling resources, unwieldy client loads, and numerous contract deliverables. The
many downsides of this dilution of services are discussed in more depth in our second report, but
we have seen that the results are significant and negative.
COMPETITION IN GRANTS AND CONTRACTS
Representatives of smaller organizations with more personalized service projects said that the
level of effort requested in District contracts and the perceived impossibility of complying
without significant outside resources had discouraged them from responding to some RFPs.
These respondents noted that as a result, TCP is overly reliant on the “Mini-Continuums”—
Catholic Charities, CCNV, the Coalition for the Homeless—for running the large emergency
shelters such as D.C. Village, Franklin School, Emery, Adam’s Place, New York Avenue, 801
East, and CCNV. This, they argue, leaves TCP inflexible and afraid to apply any effective
enforcement of contract and performance standards for fear of losing a high number of shelter
beds. One provider of a shelter with significant fundraising capabilities and no TCP contract
noted the amount of pressure on TCP staff:
…it is really hard to come down on big providers. They fear they might lose
them. And they might because the contracts are so awful for them. They want you
to do too much for too little. We can’t take more (clients) so we don’t.
Chapter 4: Contract and Grant Financial Management
31
To a certain extent, DHS and TCP have acknowledged the problem of unwieldy contracts and
service obligations. Currently, the Park Road shelter has multiple subcontractors, with
maintenance, case management, and security split among three different subcontractors. Plans
were under way in spring 2008 to fund case management for family emergency shelters
separately from shelter operations, to provide continuity of care to homeless and formerly
homeless families moving from one project to another and back into housing. Further, Catholic
Charities—one of the District’s biggest providers—has expressed interest in getting out of the
business of big shelters and into smaller, more personal service projects.
There is, however, still a general resistance to splitting services within shelters, especially
District-owned ones, or to moving from existing shelters for fear of losing any beds. More
thought is needed if the District is to succeed in its desire to transition away from the
overcrowded and under-served shelters it supports today. New beds must be lined up, new RFPs
need to be created, and the approach to emergency shelters will need to change from a system
that bases funding on numbers of beds to a system that consciously rewards quality of service.
The obvious way to end up needing fewer shelter beds, and thus to be able to restructure the
emergency shelter system, is to move the people who use the most shelter resources into
permanent supportive housing. Doing so could halve the demand for emergency shelter, as these
people use half the bed-nights available in the District every year.
TRANSFERRING CONTRACTS TO DHS
When this assessment began, bringing contractual obligations and financial responsibilities for
homeless services “in house,” or under the charge of DHS, was an attractive option to some City
Council members, Mayoral staff, and providers. During our initial interviews with people in the
Mayor’s office and City Council offices, people frequently posed questions such as “Why are we
going through TCP? Why don’t we contract directly with providers?”
There are two aspects to this question that might not be immediately apparent. One is the specific
contracting and payment function that we address in this chapter. The second is the issue of
accountability, and how DHS, whose money is being spent, can be sure that providers are doing
what they are supposed to be doing. We reserve discussion of this second issue to chapter 5.
By almost every account, TCP is praised for its careful and prompt financial management, as it
has been in past studies done for the Control Board and the Williams administration. Over the
last month, we found little evidence to believe a DHS takeover of fiduciary responsibilities
would lead to an increase of efficiency or even accountability. Rather, all evidence from
providers who have contracts with TCP indicates that TCP does very well in running contracts
and responding to invoices in an efficient and timely manner. “TCP is light-years better than
DHS” was a typical provider comment among those who remember the old days.
Recognition of TCP’s highly lauded performance in its fiduciary function should refocus the
policy choices being considered on something other than this function, which is best left where it
is. Two previous studies specifically addressed TCP’s ability to channel federal and District
money into District homeless services; both found the Partnership to be an effective
intermediary. Bass and Howes (2000) note in their report titled “Assessment of the Community
Partnership for the Prevention of Homelessness,” that:
Chapter 4: Contract and Grant Financial Management
32
Very few interviewees raised questions about TCP’s grant making process and
most felt the process is “fair.” Interviewees also praised TCP for doing a good job
of distributing funds in a timely fashion. One provider explained: “Their followthrough on grants is rigorous.” (p. 3)
Supporting these findings, an earlier report done by the Enterprise Foundation (1998) found in
interviews with providers that TCP excelled in its role, “channeling federal money with minimal
overhead costs, leading the difficult process toward standardized forms and reporting and
improved procedures, (and) overseeing the McKinney Continuum of Care and application filing”
(p. iii). Our evidence suggests that TCP has maintained this level of financial services and may
have gotten better.
Most providers we interviewed for this assessment, especially those who remember the pre-TCP
grant and contracting system through DHS and OESSS, greatly appreciated the current
timeliness and consistency of payments and the clarity of the funding deadlines and expectations
for applications. There was consensus that funds under TCP control took at most 30 days to be
transferred after submitting an invoice—with 15-day turnaround time if invoices were submitted
before the 15th of every month. One provider explained “We get paid on time every time.” Those
same providers noted, across the board, that the waiting periods for funds under District
contracting services, pre-1995, were unpredictable, occasionally taking as long as six months and
frequently surpassing 90 days. For many providers this gap in funding caused real hardship, as
they either had to take out loans to make payroll or cut staff and thereby reduce the quality or
scope of the homeless services they offered. On-time payments mean reliable services (and thus
good relationships with landlords and clients and a stable staff) and the ability to plan ahead with
confidence. Providers today place a high premium on consistency and timeliness in the funding
process. In those areas, TCP has a proven track record of success, especially relative to District
government agencies. It is clear that TCP is able to carry out its fiduciary role largely because it
is not burdened by the very slow-moving procurement processes within DHS, which many
described as “glacial.”
Being the “gatekeeper for the District’s homeless dollars” is a complicated and intensive process,
one that requires bureaucratic savvy; established, clear, organized procedures; and a certain
amount of flexibility (Bass and Howes 2000, 3). Our conclusion from the present assessment is
that TCP is well suited for this task. This is especially true with regard to HUD contracts and the
complex SuperNOFA process. An intermediary that is intimately familiar with what HUD wants,
what it will or will not accept, and how to negotiate on behalf of providers is a very valuable
asset, and one that has helped District providers bring in millions of federal dollars. Few
interviewees—even among those holding otherwise unfavorable views of TCP—thought that
transferring the contracting function to DHS would improve the process of administering the
District’s homeless-related grants and contracts. Most of those with historical experience
strongly opposed such a shift.
Those that did see DHS as a viable option for a fiduciary tended to think that procedures under
DHS would be less bureaucratic (i.e., would require fewer steps, paperwork, assurances, and so
on) than TCP’s procedures. These also tended to be providers who have never had to deal with
the District contracting procedures. It is unlikely that transferring responsibility to DHS would
eliminate any bureaucracy at all; in fact, the level of bureaucratic procedure would almost
Chapter 4: Contract and Grant Financial Management
33
certainly become more complicated, take more time, and be less satisfactory than the current
arrangements through TCP. TCP has increased the number of required forms and paperwork, but
this has more to do with the increased demand for documentation from HUD than with an
inherent love of forms and procedures.
If the District is still looking to transfer responsibilities, it should not ignore the fact that TCP
was the only organization to respond to the District’s RFP the only other time it was
competitively offered, or that DHS had significant problems producing another RFP for the
position in 2005. The answers to the problems in the contracting system likely lie elsewhere.
This is not to say, however, that no changes are needed in the types of subcontracts issued or the
clarity of contractual language regarding responsibility. TCP is not always in full compliance
with its own prime contract from DHS. Also, in the past TCP has not always altered the language
of its subcontracts with providers when its own contract changes in ways that have implications
for provider obligations. And DHS needs a way to be directly involved in assuring provider
compliance with contract conditions. There most certainly are issues within the contract system,
which we address in chapter 5 (chapter 2 already discussed contract issues involving
maintenance of buildings used for emergency shelter).
IMPLICATIONS
1. Keep contracts and grants financial management for the public homeless system with
TCP.
2. Develop a mechanism whereby DHS as well as TCP has oversight authority for the
subcontracts issued by TCP to service providers. This might take the form of a clause in
subcontracts naming both agencies as contract enforcers.
3. In conjunction with moving long-term shelter stayers into newly developed PSH and thus
freeing up emergency shelter space, begin a serious, disciplined, in-depth discussion of
what the District’s emergency shelter system would ideally be like, and what it will take
to move to that system.
Chapter 4: Contract and Grant Financial Management
34
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
35
Chapter 5
Quality Assurance and Monitoring Activities
Related to Health and Safety Issues
HIGHLIGHTS
ƒ
The Community Partnership (TCP) is often blamed for a lack of services even if it is
not contractually responsible for providing those services. In general, misconceptions
abound concerning the scope and limits of TCP’s contractual responsibilities. It is
important to understand who is responsible for what, so District officials can adequately
judge TCP on what it is contracted to do. If more is wanted from TCP, funding should be
allocated for additional activities.
ƒ
Accountability problems related to monitoring District funded projects have been
cleared up informally, but sections C.3.1.1.8–9, C.3.1.1.20, and C.3.1.1.38 of the TCP
DHS Contract should be amended to reflect those agreements. TCP no longer
monitors DHS funded projects because of specific HSRA language requiring District
overview and a perceived conflict of interest. The contract language on monitoring,
however, still needs to be amended to reflect this change.
ƒ
Program monitoring protocols are now being carried out by DHS, per the HSRA.
They seem to be thorough in their coverage. There is, however, some evidence to support
the claim that monitors need more training to understand individual program goals,
approaches, and clients.
ƒ
Language in subcontracts to providers must be changed to reflect informal
understandings on Certificates of Occupancy and Living Wage payments. Many
providers currently face the awkward situation of being asked to sign a contract
containing deliverables they view as impossible, the most outstanding of which are the
requirements to pay a living wage and obtain a Certificate of Occupancy. Contracts
should be modified to reflect that the District rather than TCP or providers is responsible
for the overcrowded conditions of emergency shelters. If living wages are to be paid,
more funding must be allocated in order to ensure compliance.
ƒ
Language should be included in subcontracts giving DHS the authority to conduct
oversight and accountability activities vis-à-vis subcontract holders, whether jointly
with TCP or on its own. DHS is accountable to the Mayor and City Council for how its
money is used and the results that are achieved. As the source of funding for TCP’s
subcontracts, DHS needs to have direct authority to monitor contracts and take steps to
assure compliance when necessary.
ƒ
TCP is in compliance with its contractual obligations for training, but funding levels
for training are not adequate to meet what many in District government expect TCP
training to offer. TCP training focuses on issues of compliance with legal data collection
requirements as its contract stipulates. For TCP to provide more training on client
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
36
services and a wide variety of general best practices, more funds will need to be allocated
specifically for that purpose. There is also a question as to whether TCP is the best
channel for training focused on clinical issues. TCP could certainly organize such
training and arrange for people with the relevant substantive expertise as presenters.
ƒ
Responsibility for security services seems to be established for District shelters and
Park Road; section C.3.9.7 and H.8.A of the TCP DHS contract should be amended
to reflect this agreement. While security responsibilities are clear to those actors
involved, unclear language regarding who is responsible for District owned family
shelters needs to be fixed in the TCP-DHS contracts.
This chapter addresses issues of provider performance in compliance with DHS and HUD
expectations. It looks at the current system for assuring program quality through monitoring and
other mechanisms.
ACCOUNTABILITY FOR THE CONTINUUM OF CARE
Much of the ire directed at TCP that we heard during data collection came from District Council
members, Mayoral staff, and providers who were collectively frustrated with the inability to
pinpoint responsibility for serious breakdowns in the Continuum such as the deplorable
conditions at D.C. Village. They wanted to know whose fault it was that some D.C. shelters and
projects remain in abysmal conditions.
All told, people are frustrated at the amount of “finger pointing” that goes on when it comes to
assigning responsibility for problems in the city’s homeless assistance system. And there has
been a fair amount of passing the buck. But it seems, whether rightly or wrongly, that the bulk of
the accusations coming from the community (especially from local elected officials and their
staff) over these contract ambiguities is directed at TCP. One staff member of the City Council
pointedly blamed TCP for not fixing essentially everything that went wrong at D.C. Village,
asking “If it isn’t TCP, what are they here for?”
Several chapters of this report attempt to sort out the various aspects of responsibility and
accountability that get confused when people are frustrated. Chapter 4 looked at the ways that
TCP handles the financial arrangements of contracts with providers. Chapter 2 looked at the
physical conditions of publicly funded emergency shelters in the District, showing that the worst
shelters are those in buildings the District owns, and for which the District is and always has
been responsible for major structural repairs and maintenance. It described the difficulties that
providers and TCP have had over the years in getting the District (specifically, DHS’s Office of
Facilities Management) to fulfill these obligations, and the turnaround that has occurred since the
Office of Property Management assumed responsibility and since sufficient resources have been
allocated for renovation and major systems replacement. Several chapters have examined the
conditions of overcrowding at these same emergency shelters and the ways that it is a direct
consequence of District policies amounting to a “right to shelter” despite the fact that the HSRA
explicitly says that such a right does not exist.
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
37
This chapter takes up another aspect of assuring that the projects offering services to homeless
people in the District are doing what they are expected to do—procedures for determining
whether homeless assistance providers are adhering to the terms of their subcontracts with TCP.
Another issue is whether, when DHS changes TCP’s own contract responsibilities in ways that
are intended to be passed through to providers, the provider subcontracts are modified to reflect
these changes. A final issue is whether provider subcontracts contain language asking for things
that clearly cannot be done, or not done with available resources. The chapter ends with a
discussion of how the various ways of assuring contract compliance and service quality might be
brought together to form a coherent whole.
The content of provider subcontracts—the specification of what providers are obliged to do, or
provide, if they sign the contract—is the starting point for monitoring activities. If something is
in the contract, a monitor ought to be able to observe whether it happens. If contract language is
vague, it may be impossible for a monitor to say whether a particular provider is or is not in
compliance. If an expectation for performance is in the mind of a funder but that expectation
never got into the contract, it is unfair to monitor a provider’s delivery of that performance that
was never explicit or agreed to.
The Inspector General’s Report of Inspection (2006) on D.C. Village Emergency Shelter for
Homeless Families is a good example of intensive monitoring of every aspect of contract
language, covering both TCP’s contract with DHS and the Coalition for the Homeless’ contract
with TCP to run much of D.C. Village for homeless families. The report found many instances in
which reality did not correspond to what contract language demanded. In response, TCP has
attended to many of the points of noncompliance; however, contract language remains unamended in some key areas. As a result, a number of local stakeholders are still confused as to
who is responsible for maintenance, for performance monitoring, for training, for advocacy, and
for the litany of other tasks and roles that make an effective homeless services system.
This is, in part, a reflection of the difficulties that TCP has in communicating the scope and
limits of its roles and powers to District government officials and the broader public. Many
people expect TCP to deliver much more than it is allowed, let alone obligated, to do. We
discussed communication problems in Chapter 3. Here we address tangible issues of
accountability—those that are more than just misconceptions and miscommunication—to help
readers understand who is responsible for what. Specifically, this chapter describes the current
system of program monitoring with respect to staffing, compliance with rules and regulations,
maintenance of a “livable” environment, and case management practice. We detail the changes
that have been made and offer recommendations for issues that are still outstanding.
CURRENT PRACTICE OF PROGRAM MONITORING
Today, TCP is responsible for monitoring DHCD and HUD funded grants but no longer
performs direct facility and contract monitoring for DHS funded grants. The change stemmed
from specific HSRA language requiring District overview and a perceived conflict of interest
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
38
brought up by the Washington Legal Clinic for the Homeless. 14 Now, TCP only monitors DHS
funded providers through review of Performance Monitoring reports submitted with HMIS data;
direct site visits are only conducted if a project receives HUD or DHCD funds. If a project
receives both DHS and HUD funding, and is thus subject to monitoring from both DHS and
TCP, every effort is made to conduct the site visits jointly to limit the burden on the provider of
having to pull together a great deal of material for each visit.
DHS opened its Shelter Monitoring Unit (SMU) in March 2007 and placed it in the Family
Services Agency (FSA). The SMU is now in charge of all site visits done on an annual basis and
all Monitoring Site-Visit/Program Reports. These reports address the following areas specified in
the Common Standards and Additional Standards that govern each publicly funded homeless
assistance project:
Staffing
ƒ
Check personnel files for necessary documentation (resume, reference checks, job
description, criminal background check, TB test, drug and alcohol test, annual
evaluations, confirmation of training, time sheets). Monitoring report contains specific
yes/no questions for each.
ƒ
Conduct interviews with staff at every level, focusing especially on staff quality, training,
and supervision.
Building Maintenance
ƒ
Monitor performs a walk through and observes state of building, also interviews staff and
residents and asks about problems (sometimes there are official fire safety evaluations by
the fire department). Providers noted that some evaluations were thorough and timely,
while others seemed cursory. 15
ƒ
SMU specifically fills out a fire safety evaluation, focusing on posting evacuation plans,
conducting fire drills, and maintaining working smoke detectors and fire extinguishers
(fire safety evaluations, however, are not always included in the monitoring reports).
Rules and Regulations
ƒ
Obtained through visible evidence of compliance (posting in plain sight), and through
talking with staff and residents.
14
It is not clear to us why it is a conflict of interest for TCP to be both funder and monitor of compliance, but it is
not a conflict of interest for DHS, the ultimate funder, to monitor its own contract. In most communities for most
services, funders monitor their own contracts; whether for homeless assistance or any other activities.
15
Providers further offered reasons for these differences. A facility with no complaints that looked clean and well
run might receive a fairly cursory walkthrough, but facilities known to have many structural, maintenance, and
cleanliness problems received much more intensive scrutiny.
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
ƒ
39
Focus on having necessary documentation (certificate of occupancy, program/universal
rules, human rights notice, reasonable accommodation policy, “Are files in a locked
cabinet?”, “Are complaints recorded?”, etc.), and on communication/treatment of
residents (“Are residents aware of program rules?”, “Is there discrimination occurring?”,
“Is there a formal way for residents to express grievances?”, “Are clients aware of
services they are eligible for?”, and, importantly, “Is there documentation that all of this
is happening?”).
“Livability” or Quality of Basic Service
ƒ
Through observation and interviews with residents, monitor answers yes or no questions
about whether residents’ basic needs are being met (clean beds, food, toilets, cool water,
AC/heat, hot showers, phones, storage, private space, play area if they house children,
escrow accounts, etc.).
Case Management
ƒ
Through interviews, monitor determines if case management staff is qualified and doing
an adequate job and if follow-ups for residents are provided. Only covers the work that
case managers do with clients who will accept case management. Does not calculate
staff/client ratios or ask whether the project has enough case managers to make anything
happen.
Each report includes a summary narrative using information from the above categories and
comments from staff and residents, noting successes, deficiencies, and corrective actions needed.
The monitoring procedure itself, as noted, includes on-site facility inspections, interviews with
between 10 and 20 percent of staff and clients per project, and random client record checks,
including the notice of rights and grievance procedures, case management files, case worker
notes, and progress within a project. The initial review often includes two to three site visits,
although some providers mentioned that the visits were short and could have been done with
fewer, slightly longer visits. If a deficiency is noted, providers are required to submit a
Corrective Action Plan within a certain time frame, ranging from as short as 24 hours for safety
issues to as long as 30 days (e.g. for incomplete files). Follow-ups on Corrective Action Plans
are encouraged and were frequent according to some providers and monitoring staff, as reported
below, although at least one provider and one monitor noted they “do not happen every time.” 16
When follow-ups are done, or if the SMU is alerted to a deficiency either through complaints or
through its own monitoring, monitors will often conduct unannounced site visits to the shelter in
question. In the end, if a project remains deficient, TCP will ultimately sanction the provider
16
It is likely that follow-ups are inconsistent because of time constraints. Since May, 2007, the SMU has been
operating with only three staff members because one of the monitors is ill. What is more, with the closure of D.C.
Village in fall 2007, the focus of the monitoring unit turned to inspection of the new units to which families moved,
which created a backlog in completing annual monitoring reports for existing shelters.
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
40
until the deficiencies have been cleared. Everyone interviewed about monitoring said that such
sanctions were “rare,” and we were able to find only two specific examples.
While it is too soon to see the true effects of the transfer of responsibility for monitoring from
TCP to the DHS/FSA Shelter Monitoring Unit, we have seen evidence that the unit is adequately
staffed and the checklists are thorough. Indeed, all four staff members of the shelter monitoring
unit are licensed case workers and each went through three days of training with DHS lawyers
and social workers before assuming their monitoring duties. 17
There were, however, a few interviewees who complained that the monitors did not understand
the projects they looked at, and that some monitors only completed short, cursory observations in
order to fill out their forms. These clients felt TCP did a more thorough job and that the SMU
staff would provide more useful information if they spent more time understanding the
individual services they were monitoring. The latter point is undoubtedly true. The following
results are mixed but speak to the efficacy of the current monitoring system. While the reports do
address some of the main concerns, there are still gaps that need to be filled.
Results of SMU Monitoring Reports 18
Of the monitoring reports we received from SMU, 73 percent (27 of 37) included follow-up
letters from the providers. Seventeen of these follow-up letters (63 percent) indicated that all
deficiencies have been taken care of, are in the processes of being taken care of, or give a logical
reason why the seeming “deficiencies” are actually necessary or unavoidable. Five follow-up
letters indicated that the provider was mostly in compliance, but cited things such as lack of
funding and age of building structure as barriers to full compliance. Some also mentioned that
work orders had been submitted to DHS and they were awaiting a response.
The remaining five shelters with follow-ups had slightly more serious problems. For example,
the Blair Transitional Rehabilitation Program follow-up letter indicated that facility managers
had reported several needed repairs to DHS and were waiting for a response. Catholic Charities’
Hermano Pedro Women’s Shelter wrote that they were not in compliance with posting “human
rights” and “people with disabilities” notices because DHS did not provide them and they were
requesting more information from TCP on how to obtain materials and what needed to be posted.
John Young Center Emergency Program reported that many of the listed deficiencies were
ongoing problems that had been brought to the attention of TCP and the Office of Property
Management (OPM) several times but still had not been fully resolved. The Harriet Tubman
Women’s Low Barrier Shelter indicated a similar problem: several of their building deficiencies
had been discussed with OPM and TCP with no response for long periods of time. Some work
had been started but some of the problems had not yet been addressed. Finally, the Franklin
Shelter program’s follow-up letter addresses only the fire safety concerns, which they adequately
17
These monitoring trainings earn nine Continuing Education Credits toward the monitors’ social worker
certification renewal.
18
Reports analyzed for this assessment were as of September 2007. Much monitoring activity has occurred since
that time, as summarized in the SMU’s report of April 10, 2008 to the Interagency Council on Homelessness,
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
41
fixed. Other building maintenance concerns for Franklin, of which there are many according to
the monitoring report, are not addressed.
As is evident, there are gaps in follow-up and ultimately in following through. Ten of the reports
we examined did not have any evidence of a follow-up at the time we reviewed the files. Further,
many of the follow-ups did not address all the deficiencies noted in the initial monitoring report.
For instance, Franklin was cited for the presence of rats, cracked walls, improper heating, soiled
mattresses, among a number of other things, but the follow-up letter simply addressed issues
having to do with fire safety. This is part of a larger issue, mentioned in chapter 2, having to do
with shelter size and type, but it also touches on issues of enforcement and the capacity of the
SMU. Fortunately, there is strong agreement among facility operators that once OPM assumed
responsibility for major repairs they have seen a great improvement in response time for many of
these maintenance issues. Problems with communication still exist, however, and will not go
away as long as there are lasting structural problems that require large amounts of funding for
comprehensive repairs, possible in conjunction with a redesign of the entire emergency shelter
system.
Who Is Responsible?
Issues of who is responsible for maintenance in District-owned buildings have largely been
cleared up, as we discussed in chapter 2. The IG report on D.C. Village documented several
instances of conflicting contract language on just who was responsible for upkeep in District
owned facilities, and difficulties with defining what was a “minor” repair (TCP’s responsibility)
and a “major” repair (OPM’s responsibility). Today, a contract modification has specific
language outlining TCP’s maintenance responsibilities and the apparent contradiction in Section
C.3.1.1.27 has been eliminated. More, however, may need to be defined to clarify who is
responsible for repairs when shelters have some funds for repairs in their own budget or shelter
buildings are under lease to the District but not directly District-owned—for example, Adam’s
Place.
Some interviewees mentioned that repairs were done “incompetently” and without monitoring;
these concerns were also addressed in a contract modification. Section C.3.1.1.26 B and C now
deal directly with assigning responsibility for monitoring repair work:
The Contractor shall conduct an initial walk through of the site where the repairs
are to be completed with the OFM 19 or the DHS Capital Manager prior to a
nonstructural repair of substantial size (C.3.1.1.26, B).
The Contractor shall conduct a final inspection with OFM or the DHS Capital
Manager to assure that all scope of work requirements are adhered to (C.3.1.1.26,
C).
For large structural repairs contracted out by the District, more needs to be done in terms of
oversight and specific contract language. A number of interviewees brought up faulty repairs or
19
The Office of Facilities Management (OFM) has since been dissolved. All responsibilities previously held by
OFM have been transferred to OPM.
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
42
structural work. For instance, wheel chair lifts or air conditioning systems would be installed but
never work. This may not, however, continue to be an issue as OPM has generally gotten good
marks on its repairs. Providers also reported that for the maintenance issues under its
responsibility, TCP was also prompt in responding and “got the job done.” This is not to say that
the buildings being used as shelters don’t need a lot of help—they do. But when minor repairs
have been needed, TCP is seen as handling them quickly and competently.
ATTENDING TO ISSUES RAISED BY THE INSPECTOR GENERAL
Problems with language within the monitoring reports discussed above have largely been cleared
up. For instance, the Inspector General’s report on D.C. Village notes on p.40:
DHS monitoring reports fail to address key safety issues such as criminal
background checks and substance abuse screenings for employees, and the
security of residents’ prescription medications…. The reports do not address
contract requirements that apply to DCV employees: training, tuberculosis testing,
and the maintenance of detailed personnel files.
With the exception of checking for the security of residents’ prescription medications—not an
insignificant exception—SMU monitoring reports now address all of the other deficiencies
mentioned above. The Inspector General report did recommend, however, that monitoring should
be increased from an annual to a quarterly check—this, along with prescription drug security,
warrants renewed attention. It is likely, though, that such an increase in monitoring would
require significantly more resources if it were to be applied across the board. An alternative is to
triage programs, making more frequent visits to those that have the most problems and retaining
annual visits for projects that have few issues and resolve them quickly.
In terms of accountability, we recommend that the DHS contract with TCP be amended to reflect
the new monitoring roles. As mentioned, until the contract language is clear, there will continue
to be finger pointing for Continuum shortfalls with blame likely falling unfairly on TCP.
Specifically, the following sections dealing with TCP’s obligation to provide monitoring should
be amended to explain the current agreement:
•
•
•
Sections C.3.1.1.8–9,
C.3.1.1.20, and
C.3.1.1.38
This will go a long way toward helping District government staff identify where and how to
improve the system.
Enforcement of Corrective Action Plans
It must be noted that providers did not always take the threat of enforcement for deficiencies
seriously. Many believed them to be “just threats.” A few respondents openly mocked the
system, with one provider noting, “…everyone is supposed to have TB tests. I haven’t had my
staff get them in years and every time I get a bad mark on it. It doesn’t matter!” But this
sentiment was not widespread and it is not necessarily a reflection of the current monitoring
procedures. TCP is authorized to dock a project up to 10 percent of its subcontract funding if
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
43
continued noncompliance is bad enough. To date, however, we have found only two
circumstances where sanctions occurred because of provider deficiencies.
Contract Content that Cannot Be Met and Has No Expectations of Being Met
There are serious enforcement issues, mentioned in chapter 2, involved with getting Certificates
of Occupancy (CO) for shelter facilities and compliance with a Living Wage of $11.75 an
hour—both of which are required under District law. Many providers, especially those using
District-owned shelters, cannot pass inspection for COs because they actually house far more
people every night than any CO would allow. So the District, which is responsible for
establishing and enforcing COs, has given itself waivers for all of its facilities used as shelters.
Yet contract language still requires a CO of providers. In addition, it appears that all providers
with more than 50 employees must comply with the District’s Living Wage statute, but the
amount in their contract does not allow them to increase salaries. By all accounts the root cause
of noncompliance with CO and Living Wage rules may be attributed to lawmakers who enact
requirements but then fail to fund the dramatic cost increases these mandates entail. Living Wage
ordinances were not accompanied by fund increases for wages and structural maintenance for
shelters, as mentioned, remains woefully underfunded in most cases.
While not as prevalent, interviewees also mentioned similar circumstances where they were told
certain measures would not be enforced. Examples include:
ƒ
Subcontractors are supposed to be penalized if a family stays in an emergency shelter for
more than 6 months, but at least one interviewee was told “don’t worry, it won’t be
enforced.”
ƒ
Some interviewees said they ignored requirements for drug testing and FBI background
checks for incoming staffers because of uncompensated costs and the belief that there
would be no enforcement.
Providers we interviewed fell into two categories. Some expressed hesitancy to sign their
contracts and some signed under the explicitly stated understanding that these issues would never
be enforced. Both circumstances put providers in unfair and untenable situations, especially with
regard to accountability. As we saw at D.C. Village, providers can be blamed, sometimes in a
very public way, for not adhering to their contract language. Currently, the District has exempted
District-owned facilities from CO requirements and TCP from Living Wage ordinances.
Contracts, however, do not reflect these policies. It seems clear to us that all provider contracts
should have specific language that codifies what they are responsible for in terms of wages, COs,
and any other issues for which they find themselves in a double bind. Informal understandings
about government intentions not to enforce a contract provision are destructive to the system and
are unfair to providers. This issue has been brought up with both TCP and DHS and both parties
agree that changes need to be made.
RESPONSIBILITY FOR STAFF TRAINING
Training offered by TCP largely relates to making sure that provider staff are able to do things
such as entering their client data correctly and reporting it to the HMIS, meeting all the
requirements to apply for HUD funding, having adequate procedures for accounting and grants
management, being in compliance with the requirements of the HSRA and related regulations,
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
44
understanding and offering reasonable accommodations to people with disabilities, appropriate
responses and data submission during hypothermia season, and the like. TCP has mostly not
been responsible for providing training on clinical or treatment issues, but it has sometimes
offered sessions on new techniques or programmatic approaches that might be useful in District
projects.
Responsibility for staff training and general staff competency is to some extent still
misunderstood. 20 In our initial interviews with City Council member staff and providers, some
mentioned frustration that TCP did not provide training for staff, especially in regard to best
practices in client services.
TCP’s performance in this regard is confused by a history of unstable and minimal funding for
training. In 2005, to meet budgetary shortfalls, DHS eliminated $275,000 for training that had
been in the TCP contract. TCP then responded by dropping all training except for hypothermia
and HMIS. 21 During that time, TCP was still contractually obligated to provide training but did
not receive the resources to do so. Thus “who was at fault?” is less than clear. Today, the
contract has been amended to provide $60,000 for training on:
•
•
•
•
Common standards and other HSRA-related matters,
Best practices,
Cardio-pulmonary resuscitation, and
Emergency first aid (C.3.1.1.34).
TCP has since taken up these training and more, and most providers we interviewed expressed
positive reactions to the sessions. There is still, however, an issue with under-funding: $60,000
falls far short of the original $275,000 required to provide training, and more is demanded of
TCP than the list of training topics. DHS and District officials should assess what is reasonable
to expect for a given level of funds for training, and recognize all the activities for which TCP is
responsible that rely on it carrying out the technical training that gives the ever-changing
provider staff the skills to do what they need to do, before making requests for specific types of
training such as “working with clients” and “dealing with unusual incidents” that are not within
TCP’s specific areas of expertise. As of now, TCP is operating well in excess of its contractual
obligations for training. If more training is needed, on new topics or more sessions for existing
topics, funds need to be allocated directly for that purpose. It is likely that specific modifications
of contract language will also need to be developed.
RESPONSIBILITY FOR SECURITY
Since 2004, Hawk One Security has provided security for all District-owned buildings under
contract to OPM. This includes the buildings used as shelters except for La Casa, which has
historically contracted for its own security. TCP contracts for security services for one site, Park
20
21
Trainings available through TCP were outlined in chapter 3.
Funds for hypothermia training are part of general hypothermia funds and therefore remained in the budget, as did
HMIS training which is funded through HUD.
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
45
Road, to King Protective Services, and provides security cameras for other subcontractors that
serve families with children. Contract modifications on security responsibilities recommended in
the Inspector General’s Report on D.C. Village have not been carried out, but since D.C. Village
has been closed there is no need to pursue them with respect to D.C. Village.
The issue still remains for other family shelters, however. Specifically, the IG called for
clarifications to section C.3.9.7 22 and section H.8.A 23 , where there is apparent ambiguity as to
who would provide security for families with children living in District owned shelters. Our
interviewees expressed little to no confusion, though. Both DHS and TCP staff agree that District
government is responsible for all security services at District owned shelters. While this does
leave open a small chance that more finger-pointing will occur if a security problem arises, we
do not believe that it is likely. OPM staff meet monthly with Hawk One and there is little
question as to who is responsible.
IMPLICATIONS
1. Sections C.3.1.1.8–9, C.3.1.1.20, and C.3.1.1.38 of DHS’s contract with TCP should be
amended to reflect agreements on monitoring roles for DHS-funded homeless service
projects.
2. Our assessment suggests that all provider contracts need specific language that codifies
what they are responsible for in terms of wages, COs, and any other issues for which they
find themselves in a double bind.
3. DHS needs a mechanism through which it can hold homeless service providers
accountable, since the agency itself will be held accountable by elected officials for
service provider performance. Subcontracts between TCP and homeless service providers
should therefore contain a clause giving DHS supervisory/monitoring/quality assurance
authority equal to TCP’s. In addition, DHS might want to institute one or more activities
that bring its staff into direct contact with providers and provide the opportunity to
establish expectations and discuss their implications. Regular (e.g., every other month)
meetings with subgroups of providers offering similar services to similar populations
might be a place to start.
4. If the District and other Continuum actors decide that more training is needed, additional
funds will need to be allocated to TCP specifically for that purpose.
22
“The Contractor shall provide 24-hour security services or electronic surveillance systems for the safety of
families and children who reside in shelters.”
23
“The District is responsible for… the provision of energy, communication, building rental, security, maintenance
and building and equipment repair services for the Contractors who use District of Columbia owned facilities.”
Chapter 5: Quality Assurance and Monitoring Activities Related to Health and Safety Issues
46
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
47
Chapter 6
Rights and Expectations of Clients and Providers; Rule
Setting and Handling
HIGHLIGHTS
ƒ
The Administrative Review and Fair Hearings Process must be shortened for more
timely decisions. Providers see the Administrative Review and Fair Hearings process as
too long and weighted toward individual clients. As a result, some providers shy away
from terminating, transferring, denying entry, and suspending clients even under
circumstances that warrant these actions. This lack of action, they believe, comes at the
expense of other project users.
ƒ
Providers in small projects, or those offering transitional or permanent supportive
housing, have a number of ways to induce clients to behave in ways that do not
jeopardize their own or other people’s welfare. Some providers were able to
encourage good behavior using privileges rather than suspensions, transfers, terminations,
or denials of service.
ƒ
Emergency actions are still a viable and effective option for providers. The
disincentives associated with denials, transfers, suspensions, and terminations are not
applicable to emergency actions, which are generally prompt and effective.
ƒ
TCP must provide more training on correctly reporting change in services and
unusual incidents. We did not find the Administrative Review and Fair Hearings process
to be biased against providers. Rather, there is a need for more meticulous reporting of
incidents; filing and reporting errors made by providers are responsible for many of the
rulings against providers.
ƒ
The internal grievance procedures available at many projects may be an adequate
tool and appropriate alternative for dealing with client complaints. Providers we
interviewed felt that their clients resolve most of their issues with the project through
these internal procedures, which may account for the relatively few client complaints that
come through the Administrative Review and Fair Hearings process.
ƒ
Public emergency shelters operate under very different conditions than private
emergency shelters; unless conditions are the same, it is not appropriate to compare
the two. Private shelters can and do turn people away. They can and do set a maximum
length of stay, and enforce it. They can and do ask disruptive people to leave, without
having to go through an administrative hearing. They can and do limit the people they
serve to the level of resources they believe are appropriate to cover the services needed.
For these and other reasons their premises will look cleaner and in better repair, they will
be less crowded, and they will be able to offer each resident more individual attention.
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
48
Since the District revoked its “Right to Shelter” in the 1980s, many people we interviewed
believe that advocates—most notably the Washington Legal Clinic for the Homeless (WLCH)—
have pushed to create a legal basis that in effect reinstates the right. In 2005, the Homeless
Services Reform Act (HSRA) outlined standards for homeless services in the District and the
procedures to resolve disputes within the system. The Act sets up an unprecedented legal
framework for homeless services and client and provider rights.
Some see HSRA as the culmination of advocacy efforts to reinstate a right to shelter, while
others regard HSRA as balancing the playing field between the powers of providers and their
clients. This section addresses the effects of the new system of rules and standards as seen by
providers and advocates. It also reports the results of analyzing records of 286 Fair Hearings and
Administrative Reviews currently kept on file by Sakina Thompson, the Assistant Attorney
General assigned to work with the Department of Human Services. 24
FAIR HEARINGS AND ADMINISTRATIVE REVIEWS
The Fair Hearings and Administrative Review process works as follows. HSRA requires that any
homeless services agency receiving public monies must provide a client with “adequate notice”
of a change in their services, such as a termination, 25 suspension, transfer, or denial of eligibility,
within “appropriate time-frames.” Notices for various actions differ and most are clearly laid out
in HSRA. 26 If a client disagrees with this action, he or she has the right to ask for an
Administrative Review or Fair Hearing. Clients also have a right to request a Fair Hearing for
24
These records are an unofficial compilation of information originally intended for the personal use of the Assistant
Attorney General.
25
Emergency and non-emergency Terminations, however, warrant a brief clarification. Non-emergency terminations
are issued when the following occur:
•
•
•
•
•
•
•
•
The client possesses a weapon,
The client possesses or sells narcotics on the premise,
The client assaults anyone on the premise,
The client endangers the safety of others on the premise,
The client intentionally destroys provider property,
The client steals property of anyone on the premise,
The client fails to accept two or more offers of appropriate permanent or supportive housing, or
The client “knowingly engages in repeated violations of program rules.” (HSRA, Sec. 22)
Emergency terminations differ only in that a client must represent an “imminent threat to the health and safety” of
others, which is defined under HSRA as an “act or credible threat of violence” (Sec. 24). Therefore, many of the
termination charges above can be deemed an Emergency, as they deal with violent or dangerous situations. This
becomes important when discussing options for providers and, specifically, the use of emergency actions as a viable
option.
26
Time frames for notices generally state a provider must give written and oral notification of a change in services
15 days prior to action. The only two real exceptions to this are for emergency situations and suspensions of
supportive services that are for no longer than 10 days.
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
49
violations of client rights or provider standards. In either case, the provider is “the respondent”
and the client is “the petitioner.” Clients must be informed of these rights and many other rights
when they enter a shelter facility, in writing. As part of that notification and shelter intake
procedure, clients are informed of their legal right to appeal a provider’s action through a Fair
Hearing and Administrative Review process. Administrative Reviews are designed to ascertain
the legal validity of a request for a Fair Hearing and must be completed within 15 days of a Fair
Hearing request unless there is an extension for good cause. Ultimately, the goal of
Administrative Reviews is to find an informal resolution to a problem. If no resolution can be
found, the Office of Administrative Hearings (OAH) will grant a Fair Hearing, to be completed
within 15 days of the initial request. There a final decision will be made.
CLIENT RIGHTS IN PROJECTS
Along with the procedures for changing client services, HSRA lays out an array of client rights,
provider rights, and common standards for all shelter types; program rules; and the locus of
authority over contracts and rulemaking. Many believe the new legal structure has created a right
to shelter, if not a sense of entitlement. One District staff member noted, “D.C. has a culture of
entitlement in housing. They won’t kick people out, but even if they do, it takes them a number
of months to get it processed. The legal structure acts as if they have the right to shelter.” While
HSRA has specific language under Sec. 28 explaining that no part of the law shall “create an
entitlement…to services” except during severe weather conditions, most providers agreed that in
essence clients felt entitled. They saw the Administrative Review and Fair Hearings process as
so cumbersome and weighted on the side of the client, in large part because the client usually has
legal representation, that they usually decline to pursue terminations. As a result, they
complained of clients staying put and refusing to take steps toward self-sufficiency. An
emergency shelter provider put it clearly:
Sometimes, it is true, we need to ask someone to leave but we know we can’t ask
them to leave…because we will not get into Administrative Review….because
we’ll lose… Why bother? This is common knowledge.
This sentiment led most providers to shy away from terminating, transferring, denying, or
suspending a client’s shelter stay when there were no violent actions. Some providers were just
not willing to go through the process because it meant exacerbating the relationship with the
client and subsequently having to wait for long periods of time for the entire process to finish, all
the while still having to deal with the client. As a result, providers felt they lost the power to
enforce certain standards of behavior.
ADMINISTRATIVE REVIEWS AND FAIR HEARINGS FOR CLIENTS RESPONDING TO PROVIDER
ACTIONS
The most common way that clients and providers arrive at an Administrative Review or Fair
Hearing is for the provider to take action to terminate, suspend, or transfer a client and for the
client to bring a complaint that the provider’s action is wrong. These cases represent 92 percent
of the Fair Hearings and Administrative Reviews on file. Provider reports of lengthy processes
are supported by the data in table 6.1. In short, the process can be extremely slow and a sizeable
portion of the providers drop their notice of termination, transfer, denial of eligibility,
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
50
suspension, or general defense from client accusations.27 On average, the FH process—from
submission of a request for hearing to the final decision—takes 36 days. This average is
artificially low because the analysis includes emergency actions which tend to take much less
time. The average length of the fair hearing process for non-emergency actions—i.e. non-violent
unusual incidents—was almost 48 days long. This means a client who is terminated, suspended,
transferred, or receives any other kind of punitive action will likely be able to stay in a shelter for
an average of 48 days before a decision is made. Indeed, HSRA Section 26 (d) reads, “any client
who requests a fair hearing within 15 days of receipt of written notice of a suspension or
termination of shelter supportive housing shall continue to receive shelter or supportive housing
pending a final decision from the fair hearing process.” For 20 percent of providers who go
through the FH process, the wait, the chances of being denied, and the knowledge that they will
have to continue to work with the client while the process plays itself out are too much and they
drop the action against the client (see table 6.2).
Table 6.1 Average Number of Days in the Administrative Review and Fair Hearings Process
Average number of days
until an Administrative
Review
Nonemergency
action
Emergency
action
Average number of days
until review by the Office of
Administrative Hearings
Average number of
days until a final
decision
23
36
48
7
34
15
Source: Urban Institute analysis of an unofficial compilation of information originally intended for the personal use of the
Assistant Attorney General.
Note: Emergency actions rarely go through the Office of Administrative Hearings (n = 24) and for a large proportion of those
that do, the Office of Administrative Hearings has no effect on the final decision. Therefore, the number of days until a review
by the Office of Administrative Hearing is misleading and, as a result, the number of days until a final decision is skewed
upward.
Advocates for client rights, on the other hand, consistently looked at the difficulty providers have
with terminating a client as a good thing, believing that if providers are paid to deal with
homelessness, they should provide that care even when a client is particularly difficult. One
interviewee explained, “It is a lot easier to scream noncompliance than it is to work with
someone no matter what.” They believe that only if clients are a danger to themselves or others
should providers be able to halt services. These interviewees tended to view terminations,
transfers, and suspensions as too often abused, and did not believe that the system was biased in
favor of clients. More than one provider, however, took offense at the notion that providers
terminated, rejected, suspended, or transferred individuals for anything less than legitimate
reasons. The process, they said, was too difficult for anyone to use it lightly.
Further, the advocacy position assumes that the only thing providers should be asking of clients
is peaceful coexistence. This is essentially a “right to shelter” argument and does not allow for a
27
In one extreme, a suspension took 386 days to settle between the client and provider, but that was due to
negotiations between client and provider that occurred behind the scenes, not because the OAH process would have
taken that long.
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
51
provider to have any legitimate expectations that a client should be working to help him or
herself.
As shown in table 6.2, claims that there is no apparent bias in favor of clients are supported in
the outcomes of most Administrative Reviews or OAH hearings. First, 32 percent of all FH on
record are dismissed—often because (1) the client failed to appear, (2) the OAH found it had no
jurisdiction, or (3) the reviewers found no wrongdoing by the provider. In all three situations, the
provider’s decision to terminate, transfer, or suspend stood. In another 15 percent of cases, the
client formally withdrew his or her request for a hearing, thus ultimately letting the provider’s
actions go unchallenged. Add another 20 percent for final decisions that “upheld” the provider’s
actions and you get a provider friendly decision 67 percent of the time. This certainly does not
seem biased in favor of the client. It must again be noted, however, that providers said they
generally made a conscious effort to avoid terminations that would likely lead to Fair Hearings,
so it is logical to assume the cases that do come to Administrative Review and OAH are ones for
which the providers see no other option and are confident in their position. But data do suggest
that the Fair Hearings system is not biased against providers.
Table 6.2 Conclusions in the Administrative Review and Fair Hearings Process (n=217)
Percent
Percent of
Percent of
Percent of
Percent of cases
Percentage of
settled in
provider
provider
cases
withdrawn by the
cases withdrawn
Administrati
actions
actions
dismissed
client
by the provider
ve Review
denied
upheld
2
32
13
20
15
20
Source: Urban Institute analysis of an unofficial compilation of information originally intended for the personal use of the
Assistant Attorney General.
Note: (1) The proportion settled is likely low, as a portion of the client and provider withdrawals almost certainly represent a
settlement. The records we received did not have enough detail to let us separate settlements from cases that were withdrawn.
(2) In 2 percent of the cases, respondent and petitioner withdrawals were not mutually exclusive—that is, both the client and the
provider withdrew their claim.
With that said, the level of dismissals was heavily influenced by a client’s failure to appear, and
thus says little about the system’s intrinsic bias.
•
28 percent of the cases decided up to July 13, 2007 noted that the client failed to appear
at the Administrative Review or at the OAH hearing.
A client’s “failure to appear” or “failure to prosecute” ended in a provider-friendly decision in all
but one case. More research is needed to adequately explain why clients are not showing up, but
this nonetheless reduces confidence that the system’s actions are unbiased. Indeed, we cannot
say that the providers would have won these cases had the client shown up; if fact, it is likely
they would not have. Still, there is value in the above conclusions, as they describe how the
system currently works.
As a side note, client advocate claims that some providers overuse punitive measures can be
checked by analyzing denials—a decision that turns down a provider’s action. As shown in table
6.2, around 13 percent of Fair Hearings result in the denial of a provider’s action. These
numbers, though, are confused by the fact that 42 percent of the “denial” decisions mentioned an
“incorrect filing” by the provider. This means that the provider action might have been
legitimate, but the provider filled out the paperwork incorrectly or did not follow HSRA rules of
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
52
notice. TCP holds staff trainings for dealing with and documenting unusual incidents, but one
interviewee mentioned that paperwork still created a serious challenge to its ability to terminate a
client.
In all, we found no evidence to support the claim that providers were overusing punitive
measures; it is more likely an issue of insufficient training around handling unusual incidents—
largely the result of high turnover among staff and the amount of paperwork needed in the Fair
Hearing process. In general, providers noted that those who were methodical about correct filing
and overall procedure were more successful. Continuing training on proper procedure is
advisable as the best way to safeguard both client and provider rights.
Both advocate and provider views are theoretically valid and have some support from the data.
On the one hand, some providers felt that if they could do nothing to discourage clients’ behavior
that negatively affected themselves or others, motivating clients to make changes in their lives
might be harder. On the other hand, the argument goes, if providers are paid by the District to
deal with homeless individuals, should they not deal with them? The data suggest that the
resolution lies somewhere in the middle. The Fair Hearing process provides a useful tool to
insure fair treatment but changes might be considered to increase efficiency and highlight
alternative measures to encourage client progress.
Emergency Actions
There are encouraging statistics. Emergency terminations, transfers, and suspensions, for
example, move through the system in dramatically shorter time periods, as shown in table 6.1.
•
Currently, about 30 percent of all Administrative Review and OAH Hearings deal with
emergency actions.
On average, the process for emergencies took 15 days, from start to finish. These numbers,
however, are artificially high due to a few emergency terminations that were denied after much
debate and an OAH hearing—lasting over 90 days because of appeals and delays.
•
The majority of emergency actions (63 percent) were completed in five or fewer days,
with half of all emergency actions (50 percent) completed in two or fewer days.
Moreover, the process could take a week, but the client is out of the shelter immediately. This
happens when the DHS or the Administrative Review panel upholds an emergency termination,
but an OAH hearing does not give its final decision until a few days later. Clients do not have the
right to remain in a shelter for a prolonged period after DHS or the Administrative Review
makes a decision. HSRA section 26 (d), mentioned above, reads, “The right to continuation of
shelter or supportive housing pending appeal shall not apply in the case of emergency suspension
or termination pursuant to section 24.”
Therefore, another way of looking at the numbers is the average number of days it takes for an
Administrative Review or for DHS to rule on an emergency action: 7 days. Within our
interviews, we found no evidence of providers of emergency shelters unwilling to terminate a
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
53
violent client because of the length of the Emergency Fair Hearing process. Ultimately, there is a
viable and swift option for providers when a client’s behavior is dangerous. 28
With that said, non-emergency Fair Hearings do indeed take a long time. OAH and the
Administrative Review panels need to reduce the amount and length of intermediary steps, so, if
anything, they adhere to HSRA law that says both Administrative Review and OAH hearings
must be completed in 15 days of their respective requests. We do not have data on whether those
hearings that went longer supplied a “show of good cause as to why the deadline cannot be met,”
as is stated in Sec.27 of HSRA, but it is likely that many fell outside the legal time limit.
•
Fair Hearings had an average of at least one intermediary step for each request, where
appeals, scheduling, requests for notices, etc. prolonged the process.
As of now, the length of the Fair Hearing process seems to provide unwanted disincentives to
punitive measures and impede a shelter’s ability to discourage non-violent bad behavior or lack
of compliance.
ALTERNATIVES TO NEGATIVE ACTIONS
Denials, terminations, suspensions, and transfers are not the only viable options available to a
provider. Some shelters were able to provide effective disincentives without entering into the
Fair Hearing by taking away certain freedoms or privileges. Specifically, one provider set aside
the most uncomfortable beds and living situations (within client rights) for those who refused to
adhere to the shelter rules. Other options exist, but effective solutions take talented and creative
staff and case workers. Specific and effective options mentioned by providers include,
ƒ
ƒ
ƒ
ƒ
ƒ
Increase privileges to clients who show good behavior,
Decrease privileges to clients who show bad behavior,
Set aside more desirable beds for good behavior,
Use other clients as mentors, and
Establish more frequent and intensive positive interactions with staff.
Providers consistently mentioned a carrot and stick approach for effective dealings with clients.
In practice, to create a disincentive for bad behavior, privileges must be given so that there is a
potential to take them away. In other words, if a shelter did not have a set of privileges, clients
may not see any reason to follow through with rules or case management. Privileges included:
ƒ
ƒ
ƒ
ƒ
ƒ
28
Flexibility with curfews,
Flexibility with intake times,
General flexibility with shelter rules,
Choice of beds,
Holding beds (for emergency shelters),
The provisions of HSRA appear not to apply to certain types of PSH where the lease is in the tenant’s name; in
that case the situation will end in the Landlord Tenant Court, not in a Fair Hearing.
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
ƒ
ƒ
ƒ
ƒ
54
Increased privacy,
General “relaxation of program rules,”
Forgiving minor violations in program rules, and
Amicable relationships with staff.
ADMINISTRATIVE REVIEWS AND FAIR HEARINGS INITIATED BY A CLIENT COMPLAINT
HSRA also gives clients the option of an Administrative Review and Fair Hearing if they feel
their rights have been violated for reasons other than an attempted denial, termination,
suspension, or transfer. Of the FH cases we have on file, only about 8 percent (22 cases) were
brought because of alleged mistreatment and only one of those cases has so far been upheld—
that is, the OAH ruled in favor of the client. This could be seen in at least three ways.
First, optimistically, it could mean that clients are not finding cause to bring complaints to
Administrative Review or OAH hearings. In support of that theory, some providers mentioned
internal dispute mechanisms that dealt with client complaints so the issue could be resolved
amicably. Most providers mentioned that these types of disputes were solved at the program
level.
Second, the results could mean clients are not willing to go through the Fair Hearing process for
the same reasons as providers: it is too long and cumbersome. This is perhaps supported by the
high percentage of these cases that are dismissed due to a client’s failure to appear for the
Administrative Review or OAH.
And third, it could mean clients are unaware of their rights. One interviewee, a former homeless
client, mentioned general misunderstandings among homeless people about their rights, due to an
inability to read or to general apathy. In short, he states “They want the bed, they don’t care
about reading rights at that point.”
All three points are valid and warrant further investigation. In our experience, however, we
believe the numbers of complaints are lower than might be expected because of the internal
procedures of providers. Most shelter projects mentioned structures to deal with client
complaints, often with outside reviews and meetings with Program Directors, Executive
Directors, and Case Managers. Providers noted their success with internal dispute mechanisms,
and generally used them as a way to avoid the Fair Hearing process.
PRIVATE VERSUS PUBLIC SHELTERS
Public shelter systems lack some of the effective tools used in private homeless services. When
interviewed for this assessment, some City Council staffers and others pointed to the “appalling”
conditions and poor results at District shelters and the “immaculate” facilities and good results of
private shelters. From this they concluded that since TCP was being paid to “run” the public
shelters, TCP should have been able to make them work like the private shelters and was clearly
failing to do so.
This view, while holding some merit in its recognition of the dreadful conditions at many
shelters in the public system, fails to acknowledge the very different levels of flexibility
available to private and public homeless services. It also fails to take into account the
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
55
circumstances of District ownership of the worst public shelters, and years-long District failures
to provide adequate upkeep and attend to major maintenance issues.
Further, HSRA and strong client advocacy have created conditions that essentially require
providers in the public system to take everyone who comes, regardless of the overcrowding that
results. Nor are they allowed to select or exclude people, and are often charged with the task of
dealing with the most difficult and chronically homeless people. The private service providers
we interviewed confessed that they were not able to take the chronically homeless people who
make up the majority of public shelter users because they were “not equipped to deal with them.”
For most, perfect participation rates in case management plans and overall good behavior were a
result of being able to be selective at intake and to immediately terminate services when rules
were not followed—an ultimate disincentive. As one provider at a church shelter put it, “We do
demand good behavior and sometimes we ban people from coming. We have banned people for
life.” Representatives of other private shelters noted that they have a fairly specific idea of what
clients would fit in their project and tended to gear their intake away from those they “weren’t
right for.”
A FINAL ISSUE—INDIVIDUAL AND “GROUP” RIGHTS
Several providers mentioned that in situations with the potential to go to Administrative Review
and Fair Hearing, the client often has excellent legal representation from WLCH to argue for his
or her rights, but the provider has none. TCP’s contract with DHS specifies that it is to retain
legal counsel to represent the provider. TCP does have legal representation on retainer, but it is
not available full time and is not from someone who is steeped in laws and regulations pertinent
to homelessness.
The situations offered by providers that prompt these comments are ones that do not meet the
criterion of “emergency”—no one’s life is in danger, but the ability of other clients to continue
their own progress is threatened. The most common situation is one in which a major program
focus is recovery from addictions, and one person consistently refuses to stop using. Several
providers noted that their other clients sometimes also start to relapse, and certainly start
complaining about the violation of program rules that makes it harder for them to maintain their
own sobriety. Situations in which a mother consistently refuses to keep her children under
control were also cited.
Providers faced with these circumstances will try everything within their own power to negotiate,
revise treatment plans, work with the client, often over extended periods of time, documenting
all such chances and opportunities. If the provider does decide to terminate following these
efforts, and the client opts to protest and apply for Administrative Review, the process gets
dragged out even longer. During all this time the client still retains the right to remain in the
program, and continues the behavior that causes problems not only for the provider but also for
other clients. The average times shown in table 6.1 are just the tip of the iceberg for providers,
given the efforts they have made before deciding to terminate to work with the client.
DHS staff have urged TCP to hire a staff attorney and offered the resources to cover the cost. To
date TCP has resisted, but several providers mentioned effects on their decisions respecting
clients of worrying about legal representation and the difficulties with obtaining a lawyer to take
Chapter 6: Rights and Expectations of Clients and Providers; Rule Setting and Handling
56
their side; some have paid for their own representation. Based on this assessment, we strongly
support the idea of TCP having an attorney on staff full time. Such an attorney could become
intimately familiar with the homeless assistance network and homeless-related laws and
regulations in the District. He or she could come to understand the case to be made in many
instances, which is not just “for the provider” but is also “for the rest of the clients.” A good part
of the representation would be to bolster the client responsibilities aspects of the HSRA as a
balance for the client rights side, as the Act intended. Further, such an attorney could work with
providers to assure that they are making every possible good faith effort to comply with HSRA
and other legal provisions. By doing so, the attorney could also affect the number of cases that
get to the Administrative Review level but are dropped for lack of adequate documentation,
failure to follow procedures, and so on.
IMPLICATIONS
1. Shortening the Administrative Review and Fair Hearings process would help alleviate
unintended consequences in shelters. Strategies should be developed for doing this.
2. More training sessions for correctly reporting unusual incidents would likely help the
efficacy of the Fair Hearings process. Additional funds should be allocated specifically
for this purpose.
3. Unless conditions are the same, basing policy decisions on comparisons with private
shelters is inappropriate. Future policy decisions should be cognizant of the differing
funding and flexibility situations of public and private shelters.
4. Adding an attorney to TCP’s staff would help balance provider and client powers in the
Fair Hearings process and increase providers’ ability to comply with HSRA law. TCP
should reconsider its resistance to hiring a staff attorney.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
57
CHAPTER 7
Capacity to Collect, Analyze, and Report Data for
Policymaking
HIGHLIGHTS
ƒ
Most permanent year-round emergency shelter beds/units are covered by the HMIS
(92 and 79 percent for single adults and families, respectively). Hypothermia beds
are also covered.
ƒ
Coverage is also high for units in transitional housing projects and permanent
supportive housing projects serving families (79 and 62 percent, respectively)
ƒ
Only 53 percent of transitional housing beds serving single adults report to the
HMIS—a proportion that drops to 30 percent for permanent supportive housing
beds.
ƒ
HMIS data quality is high with respect to completeness and accuracy, for those
beds/units being reported and for those fields required of everyone. One obvious
issue is incomplete coverage, which occurs among projects that do not get their money
through TCP and thus cannot be required to participate. A larger issue is that almost
nothing is required in low barrier shelters, which comprise about 90 percent of
emergency shelter beds for single adults. Thus information is missing for sizeable
proportions of emergency shelter users even on such basic things as age and race.
ƒ
TCP orchestrates the annual one-day point-in-time count of homeless people for the
District, as part of the count that nine local jurisdictions have conducted jointly since
2001 under the aegis of the Metropolitan Washington Council of Governments. In 2007,
the count took place on January 25; for the first time, TCP used the HMIS to supply the
data for people staying in projects within the Continuum of Care. In 2008 it happened on
January 24.
ƒ
TCP produces many routine reports from the HMIS, geared either to occupancy or
to HUD requirements.
ƒ
TCP’s ability to conduct complex analytic inquiries has been hampered in the past
by two problems: software inadequacies and the closed nature of the HMIS (no data
sharing across projects).
ƒ
The software problems are on their way to being solved, but the closed system
structure remains a serious challenge to producing timely analysis for policy
purposes.
This chapter describes the homeless management information system (HMIS) in use in the
District, which TCP administers. It describes TCP’s reporting activities and the types of data
that the HMIS can routinely provide. It also discusses the grave difficulties posed by the
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
58
“closed” nature of the HMIS, as well as software problems that impede TCP’s ability to
produce data relevant to policy decisions in a timely fashion.
THE HOMELESS MANAGEMENT INFORMATION SYSTEM
TCP maintains a homeless management information system to collect and integrate data from
homeless assistance projects. All projects receiving funding through TCP, whether the funding
comes from DHS or HUD, are required by contract to report to the HMIS.
HMIS COVERAGE
HMIS coverage for the various types of homeless assistance projects in the District serving
single adults and families in January 2008 is shown in table 7.1. All DHS and HUD contracts
that TCP manages are required to participate in HMIS, including virtually all emergency shelter
beds/units and high proportions
Table 7.1: Proportion of Beds/Units Included in HMIS
of transitional and permanent
Coverage, by Project Type and Population
supportive housing projects
(percentages)
serving families. However, HUD
Project type
Single adults Families
grants obtained before the
Permanent year-round
Continuum of Care started in
emergency shelter beds/units
92
79
1996 are different. Funding for
Transitional
housing
beds/units
53
79
transitional and permanent
Permanent supportive housing
supportive housing projects first
beds/units
30
62
became available from HUD in
Source: Urban Institute analysis of Housing Inventory Chart data as of January
1988. For eight years, from 1988
31, 2008. Note: For projects serving both single adults and families, beds for
single adults are counted with similar beds in projects limited to single adults,
through 1995, it was awarded
and
beds for families are counted with similar beds in projects serving only
annually through a national
families. Projects serving victims of domestic violence are only covered to the
competition. Starting in 1996, all extent allowed by law.
new applications were awarded
only through Continuum of Care and SuperNOFA applications that covered whole communities
rather than single providers.
District homeless assistance providers had been quite successful in their proposal-writing during
those early years of national competitions, and the resulting grants remain outside the framework
of TCP activities. Until the 2007 renewal cycle, HUD did not require projects funded through
that earlier mechanism to participate in HMIS, and many chose not to do so. This accounts for
the considerably lower rates of HMIS coverage for transitional and permanent supportive
housing beds/units for single adults that appear in table 7.1. 29 Family projects were later in
developing and thus were mostly funded through the Continuum of Care process; they also often
have funding through TCP from DHS as well as from HUD. The consequence is that most of
them must report information about their clients to the HMIS.
29
Percentages calculated by Urban Institute project staff for this report from TCP information collected for the 2008
Point in Time count organized by the Metropolitan Council of Governments. The underlying data were accurate as
of January 31, 2008.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
59
HMIS coverage of transitional and permanent supportive housing projects will increase in
coming years because in the past year HUD began requiring HMIS participation of the projects
receiving direct funding (i.e., the funding does not go through TCP) as a condition of continuing
to receive funds. Data analyses of 2007 information recently completed by TCP contain
information on these projects, which we include in the second report for this evaluation (Burt and
Hall 2008a). A final addition to HMIS coverage is CCNV, which for the first time in 2007
reported to HMIS the characteristics of occupants of the 525 beds being used at CCNV at that
time.
HMIS DATA QUALITY
HMISs being used by Continuums of Care around the country are a response to HUD’s
requirements for data reporting, and as such are set up to provide the data that HUD specifies.
Some do more, of course, but to understand what information the District’s HMIS contains, one
must know first what HUD requires, and for what types of projects.
Requirements are simplest for emergency shelter projects, which are only asked to submit the 16
data elements for each person. HUD describes these as “Universal Data Elements”:
1. Name
2. Social Security number
3. Date of birth
4. Race and ethnicity
5. Gender
6. Veteran status
7. Residence prior to program
8. Zip code of last permanent address
9. Month and year person left last permanent address
10. Program entry date
11. Program exit date
12. Unique person identification number
13. Program identification number
14. Program event number (is this the 1st, 2nd, 3rd, etc. time the person has used the program?
For 12-hour emergency shelters where a bed is not guaranteed every night, each night a
person uses shelter is treated as another event)
15. Unique household identification number
16. Children questions (do you have children with you? If yes, items 1–15 are asked for each
one.)
TCP wins national awards for the coverage, completeness, and accuracy of its data on emergency
shelter users. It is considered a “best practice” community. However, we discovered significant
issues even with these minimal data elements. The problem seems to be that intake workers at
low-barrier shelters do not feel they should require shelter users to give them the information.
Since low barrier shelters account for about 90 percent of all emergency shelter beds for single
adults in the District, the result is a lot of missing data. For example, age is missing for over half
and race is missing for over a third of shelter users. Whether this problem stems from poor
training of intake workers, a misunderstanding of what may be required of people using low
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
60
barrier shelters, or some other reason, it is not a necessary feature of low barrier projects and
should be corrected.
In addition to the universal data elements, transitional and permanent supportive housing projects
receiving HUD funding must provide more extensive data on each resident. Required fields are
known as “Program Level Data Elements.” These are intended to show not just status at entry
but also status at exit and changes along the way, and the HMIS is set up to record a person’s
status on these variables at different times during his or her participation in each project. They
include the following:
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Income and sources
Non-cash benefits
Physical disability
Developmental disability
General health status
Pregnancy status
HIV/AIDS status
Behavioral health status (mental illness and substance abuse)
Domestic violence
Education
Employment
Veterans
Services received
Destination (varieties of permanent housing, or other venues)
Follow-up after program exit
Children’s education
Other children questions
For TCP, the issue for these data elements is not accuracy and quality, which are adequate, but
coverage. With a significant proportion of beds/units not covered within HMIS, TCP cannot
provide a complete picture of homeless assistance and its outcomes in the District, as it can to a
much greater extent for emergency shelter.
THE ANNUAL POINT-IN-TIME COUNT
For the past eight years, the nine jurisdictions in the Metropolitan Washington Council of
Governments have conducted a one-day count of homeless people using shelters or living in
transitional or permanent supportive housing projects, or living on the streets or in places not
meant for habitation. TCP has organized this massive undertaking every year for the District.
This count only shows who is homeless on a single day (in 2007, that day was January 25), 30 but
this snapshot provides a lot of information, especially because it permits comparisons from year
30
Data from the 2008 count were published too late to be included in this report.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
61
to year that may be used to indicate progress toward specific goals. It shows, for instance, that 11
percent more formerly homeless disabled people are living in permanent supportive housing in
2007 than were doing so in 2006, an increase that reaches 21 percent when compared to the 2005
count. The count thus provides one measure of the success of determined efforts to increase the
supply of permanent supportive housing and to use it for ending the homelessness of long-time
street dwellers. It is also one of the few sources of information about the characteristics of street
homeless people who do not use shelters.
For the first time in 2006, TCP used data from the HMIS to count and describe the parts of the
homeless population on January 25 who were staying in emergency shelters or living in
transitional or permanent supportive housing projects that report to HMIS (information about
people staying in non-HMIS projects was collected from records of the projects themselves or
from interviews with residents).
REPORTING USING HMIS
Given the array of information being collected in HMIS, it is easy to see that the system should
be able to answer a wide variety of policy-related questions, at least for the projects that report to
it. HMIS in the District uses ServicePoint, a very complicated relational database developed by a
private vendor specifically to serve the data needs of homeless assistance projects and
community-wide networks. 31 Although it contains lots of data and is projected to produce some
standard reports, particularly those geared to HUD requirements, it has not been very easy for
communities around the country to use the data it contains to answer the types of policy
questions that arise every day. In particular, it is quite difficult to answer questions of the form
“How many As are also Bs, compared to the number of Cs who are Bs?” Questions of this type
might include “How many single women are affected by serious mental illness, compared to
single men, or compared to mothers in homeless families?” Or “What are the characteristics of
people who leave shelter quickly and never come back, compared to those who stay for more
than a year?”
We first discuss some of the reports that are routinely produced and relatively easy to do,
because the system was programmed from the start to generate them. Thereafter we describe the
challenges that have affected TCP’s ability to answer specific, not-previously-programmed data
questions quickly.
ROUTINE REPORTING
TCP regularly produces a variety of reports; some of them are as follows:
•
31
The Daily Census Report shows hour by hour the number of people occupying each
emergency shelter receiving District funds.
ServicePoint is one of 12–15 products developed by different vendors to meet the demand for HMIS databases. It
is probably the software package used by the most communities, with about one-third of the Continuums of Care in
the country running their HMIS using ServicePoint.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
62
•
For every facility in the public system, the nightly bed-list report shows each person and
which bed in that facility the person occupies.
•
The Clients Served report can be compiled for each provider, for any set of dates desired.
It shows the number of people who were already in the facility on the start date, the new
people who entered during the reporting period, and the total number of people served
during the reporting period. For each adult and child, it also shows age and gender.
•
During hypothermia season, Dashboard Reports show nightly use and availability of
hypothermia beds.
•
Entry and Exit Reports show the number of people who came in and out of facilities in
the public sector, and the Universal Data Elements (see above). For transitional and
permanent supportive housing projects they report the Program Data Elements (see
above), at both entry and exit (for those exiting during the reporting period). This is
essentially the same information, in the same format, gathered for annual reporting to
HUD, but for projects with DHS support, for reporting to DHS.
•
Up through 2005, the Washington Cumulative Monthly Report provided basic
demographic data on everyone using the system each month, as well as entries and exits
and cumulative numbers of clients served.
•
In 2006, TCP dropped the Cumulative Monthly Report and began to use the HMIS to
produce quarterly reports with the same information, which it keeps for its own and DHS
use and also sends to providers. The quarterly, semi-annual, and annual reports back to
providers cover both HUD-funded projects and projects that only have DHS funding,
thereby providing for DHS the same information that is routinely reported to HUD.
•
For HUD-funded projects, TCP is now using the HMIS to produce Annual Performance
Reports (APRs), which must be submitted to HUD with each year’s funding request.
Prior to this year, the projects with HUD funding produced these reports themselves and
sent them to TCP. With the continuing maturation of the HMIS, TCP has been able to
take over this reporting.
Projects and agencies retain for their own use the data they send to the HMIS. TCP has supplied
them all with report writing software and conducted trainings on using the data for their own
purposes. Some use their HMIS data for various purposes; others do not. Perhaps the most
common provider uses of the data are to generate descriptions of their clients and outcomes for
to go into proposals for new funding and to provide existing funders with reports of what they
have accomplished. Some providers also use HMIS for case management, and others have
systems of their own for this purpose that they run in parallel.
SUMMARY AND PERFORMANCE REPORTING
In chapter 8, Performance Standards and Client Outcomes, we discuss the actual outcomes that
District homeless assistance providers have set for themselves, working with TCP. In this
chapter we are concerned with the capacity of the HMIS to measure those outcomes and report
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
63
on achievement, as well as how the information may be used to reflect on ways the overall
system could be improved.
The District’s HMIS is only now poised to come into its own with respect to tracking
performance, and barriers still exist. The biggest problems have to do with being able to
summarize anything across agencies. For instance, a year ago, learning whether a person using
Shelter A also used Shelter B was almost a matter of hand counting. The problem lay in part with
available software, which was not adequate to conduct even this seemingly simple type of
analysis.
In November 2006, TCP acquired the Advanced Reporting Tool (ART) that the HMIS software
vendor had recently developed at the insistence of some very big customers. ART is specifically
designed to make it easier to answer the types of questions described above—“How many As are
Bs?” Getting ART up and running has taken close to 10 months, but the tool is finally beginning
to fulfill expectations for being able to produce policy-relevant reports. TCP has bought ART
licenses for all agencies using the HMIS and conducted training for agency administrators in the
use of ART. But in all likelihood few if any providers have begun yet to use this very flexible
tool, as TCP itself is just beginning to get it to work well with the HMIS database.
Example of What ART Can Produce—Emergency Shelter Use by Single Adults
•
During the first six months of 2006, 7,500 individuals (unduplicated) used the 1,885
emergency shelter beds available to single adults from providers that participate in the
HMIS. This means that approximately four people used each bed in the course of just six
months.
•
Most single adults leave emergency shelters after a very short time—a fact that will
undoubtedly surprise many readers of this report. 44 percent of single women and 51
percent of single men using emergency shelter during the first six months of 2006 spent
only one to seven days in shelter during that entire period.
•
Only 24 percent of single women and 17 percent of single men spent more than 60 days
in a shelter during the first six months of 2006.
Example of What ART Can Produce—Recognition that Goals Set Too Low
•
Providers of emergency shelter typically have a hard time appreciating the number of
different people who pass through their facilities over time. They only get to know the
ones who stay a long time, and thus are not very good at setting goals for the number of
people they will serve overall.
•
Analysis using ART of the first six months of 2007 shows several emergency shelter
providers reaching or slightly exceeding their annual goal within six months; others had
served more than three times their annual goal in just six months, and one had served
even more people in six months.
•
Further, annual goals for the number of people who would be employed were also being
met or exceeded with a good part of the year still to go—projects were already over goal
by between 7 and 27 percent nine months into the year.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
64
This and similar information from ART reports and analyses are the basis for developing a true
performance-based system in the District. TCP anticipates sharing these results with providers,
getting their views and interpretations, and then holding discussions to reach a mutual
understanding of what level of performance is fair to expect and what can be done to improve
performance.
THE “CLOSED” NATURE OF THE DISTRICT’S HMIS
Having just described the difficulties that lack of appropriate software has posed for getting
useful information about the District’s homeless assistance system, it must also be said that there
is another very large part of the problem. Difficulties with data retrieval and analysis will likely
continue to exist unless the District changes the very nature of its HMIS.
The District’s HMIS is a “closed system.” This means that data from Project A can only be seen
by Project A, data from Project B can only be seen by Project B, and so on. Even projects run by
the same agency are not set up to be able to look across to each other to see whether they are
serving the same people. (One large provider has just asked TCP to “open” its own projects to
each other, and this will happen going forward. But that is only one agency.)
In theory, as the HMIS administrator TCP can piece together the ways that a single person uses
multiple agencies and projects, but the process is laborious even with ART. It was extremely
painstaking before, and was not undertaken except to answer the very limited question of “how
many unduplicated people did we serve last year?”
Many communities across the country operate with closed systems similar to the District’s, but
many use systems that are open to a greater or lesser extent. Each community that shares data has
worked diligently to assure appropriate safeguards for privacy and confidentiality, and many
different levels of “openness” have evolved, often within the same system, based on “need to
know” and each client’s willingness to have his or her data shared. Some of the District’s
immediate neighbors share data (e.g., the Continuums of Care in Prince George’s County and
Southern Maryland) and providers experienced in those systems as well as the District’s HMIS
find the open systems not only more user-friendly for them but also more conducive to getting
homeless people the services they need.
There are good historical reasons why the District developed a closed system. The city was one
of the first to attempt an HMIS, in the days more than 10 years ago when such things did not
exist outside of three or four cities in the country and approaches were just being haltingly
developed. No one knew what risks to privacy and confidentiality would hide in software and
hardware glitches, and some providers and advocates also questioned what the data would be
used for (they feared it would be used to deny services to homeless people). Nor were relations
among District homeless assistance providers and District government agencies ones of trust and
openness. After considerable discussion about the type of system that would be acceptable to
District providers—because without their acceptance there would be no system at all—the
community opted for a completely closed system. We will not belabor this point further in this
report. We will, however, revisit it extensively in the next report.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
65
IMPLICATIONS
1. Many of the goals the Mayor is setting for system transformation will be extremely
difficult to accomplish without also transforming the HMIS into a system that is more
open and therefore more capable of being responsive to policymakers’ questions and
guiding system development.
2. It is time to reopen a discussion of system “openness.” The District’s HMIS will not be
truly useful until the system is at least somewhat more open than it currently is.
3. More consistent and complete collection of universal data elements should be required at
all shelters, including low barrier shelters.
4. Better and more usable software would improve TCP’s ability to respond to data
requests.
5. Now that data reflecting actual performance are starting to be readily available on at least
a quarterly basis, it is time for all participants in the system—public agencies and local
elected officials, TCP, providers, advocates, and other stakeholders—to take several steps
back and ask what the system is trying to accomplish.
6. Once the overall goals are set, the next steps are to determine:
a. What type and level of accomplishment they should expect of the system;
b. How the various types of projects currently being offered or others that might be
developed would contribute toward those accomplishments, and thereafter how
the system should be reconfigured;
c. How progress should be measured, including what part in measuring progress the
HMIS should play and what parts will need to be documented by other means.
7. The result should be a set of goals, a structure for implementing the goals, a system of
measurement to assess progress toward achieving the goals, and routine reporting to the
public about progress being made. Examples from other communities will be described in
the next report.
8. It would be highly desirable to bring more transitional and permanent supportive housing
projects for single adults into the HMIS. Strategies should be developed for doing this.
Chapter 7: Capacity to Collect, Analyze, and Report Data for Policymaking
66
Chapter 8: Performance Standards and Client Outcomes
67
Chapter 8
Performance Standards and Client Outcomes
HIGHLIGHTS
•
Occupancy is generally above 80 percent in homeless assistance projects receiving
District funds. Emergency shelters for single adults tend to be at capacity (100 percent
occupancy) or to exceed maximum capacity, sometimes by quite a lot. Family shelters
tend to have 10 to 15 percent vacant capacity, largely because they are smaller so one
empty unit makes a bigger percentage difference.
•
Markedly low occupancy rates may indicate any or all of several things: program
entry criteria are too selective (i.e., few people meet them), the program is not attractive
to homeless people, the program is undergoing internal upheavals, or the overall
community system is very inefficient (i.e., it cannot refer needy homeless households to
available resources in a timely manner).
•
Emergency shelter occupancy that is consistently over capacity may also indicate
several things: not enough capacity for the type of program, or inadequate system design
that leaves people in emergency shelter who should be able to move out to any of several
types of permanent housing.
•
At least 2,200 single adults, or 67 percent, of the single adults living on the streets or
sleeping in shelters in the District on a given day are chronically homeless. However,
over the course of a year these people comprise a much smaller percentage of single
adults experiencing homelessness in the District—maybe no more than 10 percent. A
highly targeted, multi-year, multi-agency collaborative effort will be needed to end their
homelessness. Other communities are undertaking similar goals with very promising
results; the District has taken the first steps toward realizing this goal.
•
The HMIS managed by TCP is the only consistent source of information about the
District’s homeless assistance system. Its coverage of emergency shelter beds is very
high, and is reasonably high also for projects of all types serving families. Transitional
and permanent supportive housing projects serving single adults are significantly less
well covered, largely because they do not receive funding through TCP and thus exercise
the option to refuse to participate in the data system.
•
The first wave of (relatively primitive) performance indicators is just becoming
available from the HMIS. It indicates that performance standards are consistently being
set far below the level of actual performance. Now that data are available to reflect actual
performance, this will allow participants in the system to take several steps back and ask
themselves what the system is trying to accomplish, how their activities do or might
contribute, and what level of accomplishment they should expect of themselves.
•
TCP’s capacity to report routinely on program and system performance is in its
infancy. An Advanced Reporting Tool (ART) recently developed by the HMIS software
Chapter 8: Performance Standards and Client Outcomes
68
manufacturer is starting to increase the flexibility of the system and should serve the
District in good stead in the future.
Most homeless assistance systems make some effort to document their accomplishments, and the
District’s is no exception. In this chapter we focus on “performance” in two senses. The first is
occupancy levels—the extent to which projects receiving public funding are operating close to
capacity. Related to occupancy levels is length of stay or number of days used per week, month,
or year. The second is client outcomes—increasing the proportion of people using homeless
assistance projects who reach desired goals such as permanent housing, increased income,
employment, education and job training, and gaining control over mental health or substance
abuse problems if they exist. Other types of “performance” such as maintaining clean and safe
premises were discussed in previous chapters.
OCCUPANCY AND LENGTH OF STAY
OCCUPANCY
Occupancy rates may be used to asses a project’s performance, or a system’s performance. We
focus mostly on occupancy as a measure of program performance, but also comment on it briefly
as a measure of system performance.
Occupancy as a Measure of Program Performance
There are two basic ways to fund homeless assistance projects—on a per diem basis, or through
grants and contracts. Cities funding emergency homeless assistance on a per diem basis (e.g.,
New York and Philadelphia) only pay a provider if the provider gives a homeless person a night
of shelter. The city does not waste money if a shelter only fills 50 percent of its beds, although of
course having half the beds empty is a waste of opportunities for people who need shelter beds.
However, if the payment mechanism is a contract or grant that does not depend on a nightly head
count for reimbursement, low occupancy is a waste of money as well as a waste of shelter
resources. District government contracts for homeless assistance projects operate in this manner
rather than on a per diem reimbursement basis. Therefore, occupancy rates are important
indicators of efficiency—or inefficiency and wasted resources, if occupancy rates are low. There
are other implications if occupancy rates exceed capacity, which we discuss below.
TCP routinely measures the occupancy rate of each homeless project that receives District (or
HUD) funds and is able to produce daily reports on occupancy. Further, during hypothermia
season, in addition to the daily reports issued by TCP based on its HMIS, the United Planning
Organization tracks real-time information on the location of available beds and is able to direct
police and outreach workers to the right places that have the capacity to receive people being
transported to shelter for the night.
Consistently low occupancy rates at shelters and other residential projects trigger concern and
discussions aimed at discovering the reason(s) and taking steps to change them. In addition to
this internal use of occupancy information supplied through TCP’s HMIS, a project’s occupancy
rate is one of three criteria TCP uses to rank transitional and permanent supportive housing
projects during the annual process of applying for continuing HUD funding. Usually the
Chapter 8: Performance Standards and Client Outcomes
69
occupancy rate is not low enough to reduce a project’s ranking below what would be funded, but
it has happened. 32
Occupancy levels exceeding capacity, however, are not only inefficient but potentially dangerous
to residents and facilities. Occupancy at District emergency shelters tends to exceed 80 percent,
and often exceeds 100 percent. Overflow conditions are most likely to occur at the emergency
shelters serving single men and women, with D.C. Village being the notable exception for
families (the District closed D.C. Village, but there is still a need to beware of creating another
situation in which, because of “right to shelter” practices coupled with inadequate resources,
families are sheltered in greatly overcrowded conditions). Apartment-style family shelters tend to
operate with occupancy levels in the 80 to 90 percent range. As examples, on the night of
October 21, 2007, occupancy rates at selected emergency shelters all exceeded 80 percent, as
shown in table 8.1.
Table 8.1 One-Day Occupancy Levels for Selected Facilities, October 21, 2007
Project
801 East
Adams Place Shelter
Franklin School
La Casa
New York Avenue HAC
Open Door 12-hour
Harriet Tubman/D.C. General Cafeteria
John Young/Federal City
Madison
Meridian Hill/Hermano Pedro
Girard Street
Park Road
Spring Road
Valley Place
Population
served
Single men
Single men
Single men
Single men
Single men
Single women
Single women
Single women
Single women
Single women
Families
Families
Families
Families
Year-round Occupancy
capacity
rate (%)
300
111
150
100
240
125
60
100
200
164
108
93
75
127
80
116
39
100
20
90
20
85
45
89
28
82
18
89
Source: Dashboard Report based on HMIS data, supplied by TCP.
Note: Occupancy rates greater than 100 percent mean that overflow capacity is being used.
Occupancy as an Indicator of System Performance
An occupancy rate greater than 100 indicates that the facility sheltered more people on the night
in question than it had regular beds to accommodate. The overflow slept in seasonal/overflow
areas, which often consist of cots or even mats on the floor of a large open space. In the District,
the facilities affected by these overflow conditions tend to be District-owned buildings used as
emergency shelters, for which the District has exempted itself from Certificate of Occupancy
32
In one instance of a program that was not refunded, occupancy was at 40 percent for significant periods due to
major problems at the program, including several staff changes at the top of the agency and considerable
disorganization.
Chapter 8: Performance Standards and Client Outcomes
70
requirements that would limit the number of occupants to a safe number. The buildings are old—
often 50, 60, or more years old—and have many physical problems. Overcrowding exacerbates
the physical fragility of the buildings and leads to even greater problems with maintenance than
would normally be the case. Many of these issues have been detailed earlier in this report.
Therefore one measure of performance that the District could enforce on itself would be to set
safe occupancy levels and allow restrictions on shelter entry to adhere to those occupancy limits.
With its current approach of “shelter anyone who asks, and make no demands on them,” limiting
occupancy to safe levels would seem at first glance to create a need for more emergency shelter
capacity. However, as noted in the second evaluation report for this assessment, with a changed
approach it is likely that the need for emergency shelter would be cut by half. That approach
would include imposing expectations that shelter users need to contribute to their own exit from
homelessness, casework and other resources to help them do so, permanent supportive housing
for many with long-standing disabilities, and a triage/priority structure that assures access for
new housing capacity to long-term shelter stayers as well as long-term street dwellers. We
describe this approach at length in the second report for this evaluation.
LENGTH OF STAY—CHRONICITY
The District does not use length of stay in emergency shelters as a performance criterion. Other
jurisdictions do so, however, and pursue policies designed to keep emergency shelter stays as
short as possible, and certainly to end shelter use by long-term shelter stayers. Because
promoting shorter lengths of stay has not been a priority in the District, TCP has not set any
performance standards with respect to it, except for permanent supportive housing where the
goal is to get people to stay longer, not to limit their stay.
Beginning with the 2006–7 contract
year, TCP started gathering
information that might be used to
begin a dialogue about patterns of
shelter use, including the significant
proportion of shelter users who have
lived in the shelter system for years.
Annual performance plans for
outreach programs, low barrier
shelters, and temporary (emergency)
shelters asked providers to estimate
the number of different
(unduplicated) people they serve in a
year, and also to estimate the number
of different chronically homeless
people they will serve in a year.
Numbers from the District’s
January 2008 Point in Time Count
•
•
•
~4,200 currently homeless single adults,
including those in transitional housing programs
(up ~500 from 2007)
~2,200 were chronically homeless (up ~400
from 2007)
About 84 percent of chronically homeless single
adults were in shelter on the night of the count;
the rest were unsheltered
Bottom line: These 2,200 single chronically
homeless adults should be the target population
for new permanent supportive housing units.
In fall 2007, HMIS data were analyzed to provide accurate information about total people served,
turnover/ lengths of stay, and chronicity—i.e., meeting the HUD definition of a chronically
homeless person, which is a person with a disability who has been continuously homeless living
in emergency shelters or on the streets for at least a year, or who has had four episodes of
homelessness in three years. We also undertook our own length-of-stay analysis. Our second
Chapter 8: Performance Standards and Client Outcomes
71
report describes the patterns that people follow in using emergency shelters. Since people do not
necessarily stay in these shelters every night, as beds are not guaranteed from night to night,
several measures to assess chronicity in relation to shelter stay were used.
If the District is serious about changing the nature of its emergency shelter system by reducing
the number of large shelters and restricting them to truly emergency situations, then the people
that stay in shelter the longest should be high priority for assisting into permanent supportive
housing. Once the patterns are known and information about what types of people follow which
patterns is available, the District will be in a better position to design a more effective system
that meets the Mayor’s stated commitment to reduce dependency on emergency shelters and
change the nature of the shelters that remain. However, even now we know certain facts about
chronically homeless individuals, whether shelters users or street dwellers that identify the
magnitude of the task, as shown in the box at right.
Challenges to Reducing Chronic Homelessness
At this time, three circumstances would need to change for the District to be able to carry out the
major transformation of its emergency shelter system envisioned by the Mayor. The first two are:
(1) more permanent supportive housing would need to be available, whether created by building
new buildings, rehabilitating existing buildings or separate units, or subsidizing rents in
apartments throughout town; and (2) a system would have to be established for prioritizing who
should get the next available unit of permanent supportive housing.
The third circumstance is the District’s laws and regulations that prevent length of stay
restrictions on emergency shelter use. At present, although District laws do not provide an
absolute right to shelter, they come very close. During hypothermia season, which lasts five
months, homeless individuals do have a right to shelter, of the most minimal kind, offering only
the opportunity not to freeze to death outside. During other times of the year there has been little
focus on getting the large majority of people who basically live in the shelters out of them and
into more appropriate settings, especially permanent supportive housing.
Even if shelter providers had the resources to concentrate really good case planning, resource
linkages, and management on far more emergency shelter stayers than is now the case, under
current legal conditions they would have no leverage to induce people who have lived in the
shelter system for years to work toward leaving shelter. Combined with inadequate resources for
helping people move out of shelter, this legal and regulatory situation makes it almost impossible
to manage the current system at the point of entry. Our second report discusses at greater length
these circumstances that create obstacles to progress and puts them in the context of what other
communities have done to move their systems away from excessive and unproductive use of
emergency shelter.
CLIENT OUTCOMES
At the most basic level, investors in homeless assistance projects want to know that their money
is being used to help homeless people leave homelessness and be able to stay housed. After that,
different stakeholders will have different interests in client outcomes, depending on their goals
and their points of investment in homeless assistance and efforts to end homelessness.
Chapter 8: Performance Standards and Client Outcomes
72
Homeless assistance agencies and their individual projects may have their own detailed ways of
telling themselves how they are doing, but the only consistent information about the District’s
homeless assistance system as a whole comes from and through the Community Partnership, and
then only for agencies that participate in its Homeless Management Information System (HMIS).
Every agency receiving funds through TCP, whether these funds originate with DHS or HUD,
must complete an annual performance plan describing its expectations for the people it will serve
during the coming year. Temporary shelters, transitional housing projects, and permanent
supportive housing projects must also specify performance criteria related to certain outcomes,
as follows:
Outreach programs: number expected to serve annually and the number of chronically homeless
people expected to serve annually.
Low barrier and severe weather shelters: number expected to serve annually, number of
chronically homeless people expected to serve annually, and occupancy level.
Temporary shelters: number expected to serve annually, occupancy level, percentage with
positive housing destinations (which include a transitional or permanent supportive
housing project, housing with family or friends, or permanent housing in the community),
number who will gain income while in the project, and one self-sufficiency indicator
(participate in substance abuse treatment or mental health treatment or education/training
or working or looking for work).
Transitional housing projects: number expected to serve annually, occupancy level, percentage
with positive housing destinations (which include a transitional or permanent supportive
housing project, housing with family or friends, or permanent housing in the community),
number who will gain income while in the project, and one self-sufficiency indicator
(participate in substance abuse treatment or mental health treatment or education/training
or working or looking for work).
Permanent supportive housing projects: number expected to serve annually, occupancy level,
percentage remaining stably housed at least six months, number who will gain income
while in the project, and one self-sufficiency indicator (participate in substance abuse
treatment or mental health treatment or education/training or working or looking for
work).
Note that for the first two categories, outreach programs and low barrier/hypothermia shelters, no
standard is set except for head counts—that is, no mention is made of what the contacts with the
people served are supposed to accomplish.
For the remaining categories, TCP (and HUD) asks for reports of performance related to
destination for those leaving shelter (looking for the “good” destinations of movement to
permanent housing), increased income, and increased employment. Local providers also wanted
to be able to report on their clients’ progress on at least one measure of increasing selfsufficiency, hence participation in mental health or substance abuse treatment, education, and
training were added to the annual performance plan, and providers may select one of these as an
indicator of their performance.
Chapter 8: Performance Standards and Client Outcomes
73
For transitional and permanent supportive housing projects funded through TCP, most of which
participate in the annual SuperNOFA application to HUD for funding from its Supportive
Housing Program, TCP uses performance on three indicators—occupancy, rate of moving to
permanent housing or staying in permanent supportive housing, and income increases—to create
a ranking of projects in terms of performance. This ranking is used, in turn, to prioritize projects
on the list for HUD funding.
HUD sets performance levels for itself on three indicators, and averages across all transitional
and permanent supportive housing projects it funds throughout the country to calculate its own
performance level. HMIS data should be analyzed to assess how District projects do on these
HUD performance criteria. 33 For federal fiscal year 2008, these standards are as follows:
•
For transitional housing projects, at least 61.5 percent of people who exit will move to
permanent housing situations;
•
For permanent supportive housing projects, at least 71 percent of tenants will remain in
the housing at least six months;
•
For transitional and permanent supportive housing projects combined, at least 18 percent
of those exiting the projects will be employed.
It should be noted here that the ability to generate and use project performance information to
improve systemwide performance or move toward outcomes-based contracting is in its infancy.
The entire process needs considerable work, from the selection of performance criteria to the
selection of performance targets to the quality of data being recorded and reported and upon
which performance levels would be calculated to the proportion of local projects that participate
in the reporting system. In November 2006, TCP acquired an Advanced Reporting Tool (ART)
newly released by the HMIS software vendor, and has been working since then to get the bugs
out and make it function smoothly with the HMIS data configurations on this system. By next
year ART should begin to make the process of assessing project and system performance
considerably easier and more direct.
It takes several years and a good deal of collaborative work to develop a good performance
monitoring system. TCP has begun that work, but much remains to be done. To understand what
a mature performance monitoring system would look like, we attach as Appendix A to this report
a summary from Columbus, Ohio’s Community Shelter Board showing outcome types, expected
performance levels for different projects, and summary reports across the entire system revealing
how the whole system is doing.
IMPLICATIONS
1. A much greater appreciation is needed of the fact that a great many people use
emergency shelters for exactly what they are meant to do—provide a brief respite from a
33
The only programs for which TCP has the capacity to report are those that contribute their data to the HMIS. See
table 1.1 for HMIS coverage of different types of programs for single adults and families as of January 2008.
Chapter 8: Performance Standards and Client Outcomes
74
housing crisis. With almost half of shelter users spending less than a week in shelter, far
fewer shelter beds would be needed if these people, or those who only need one or two
months of shelter, were the only ones to use the shelters. This would be the case if
District were able to move the relatively few long-term shelter stayers into permanent
supportive housing.
2. Performance goals could be set a good deal higher than current practice. Appendix C
provides an example of the very high expectations that one community sets for its
homeless assistance projects. And those higher goals are being achieved. There is no
reason the District cannot do likewise.
3. To date, local elected officials and District government agencies have not been part of
establishing performance expectations for homeless assistance projects in the HMIS. Nor
has there been a process of having providers of similar services to similar populations
meet to determine a set of standards and expectations they believe to be realistic for
themselves, as has happened in some other communities and as is common in health care
and other service sectors. The process has gone on only between TCP and individual
providers. It would be highly desirable for the process of goal- and standard-setting to be
more open and collaborative, to have clear links to the overall expectations of ending
homelessness in the District, and ultimately perhaps to feed into performance-based
contracting.
References
75
REFERENCES
Bass and Howes, Inc. 2000. “Assessment of the Community Partnership for the Prevention of
Homelessness.” Washington, DC: Report to the Deputy Mayor for Children, Youth, and
Families, November 8, 2000.
Burt, M.R. 1995. “Assessment of the D.C. Initiative’s First Year. Prepared for the Community
Partnership for the Prevention of Homelessness.” Washington, DC: Urban Institute.
———. 2002. “Homelessness and the Homeless Assistance Network in the District of Columbia
and the Role of the Fannie Mae Foundation’s Help-the-Homeless Program. Washington,
DC: Urban Institute.
Burt, M.R. and S. Hall. 2008a. Transforming the District of Columbia’s Public Homeless
Assistance System. Washington, DC: The Urban Institute. Available at www.urban.org.
———. 2008b. Major Recommendations: Summary Report of the Urban Institute’s Assessment
of the District of Columbia's Public Homeless Assistance System. Washington, DC: The
Urban Institute. Available at www.urban.org.
Enterprise Foundation. 1998. “A Report on the Existing Conditions and Opportunities for the
District of Columbia’s Homeless Housing and Service Systems.” Columbia, MD:
Enterprise Foundation.
References
76
Appendix A: List of People Interviewed
77
APPENDIX A: LIST OF PEOPLE INTERVIEWED
D.C. Government
•
Office of the Mayor—Julie Hudman, Laura Zeilinger, Janice Ferebee, Leslie Steen,
Melissa Hook, and Oscar Rodriguez
•
Office of Councilmember Tommy Wells—Adam Maier, Ram Uppuluri, and Yolundra
Barlow
•
Service agencies:
• Department of Human Services (DHS)—Clarence Carter, Kate Jesberg, Fred
Swan, Sakina Thompson, Ricardo Lyles, Jean Wright, Deborah Carroll, Susie King,
Lisa Franklin-Kelly, and George Shepard
• Department of Mental Health (DMH)—Stephen Baron, Barbara Bazron, Michele
May, and Eric Strassman
• Department of Health’s Addiction Prevention and Recovery Agency
(DOH/APRA)—Tori Whitney
•
Housing agencies:
• D.C. Housing Authority (DCHA)—Michael Kelly and Adrianne Todman
• Department of Community Development (DHCD)—Leila Edmonds and Guyton
Harvey
•
Public safety agencies:
• Fire and Emergency Medical Services (FEMS)—Michael Williams, John Dudte,
Mytonia Newman, and Patricia White
• Metropolitan Police Department (MPD)—Brian Jordan
• Department of Corrections (DOC)—Devon Brown, Reena Chakraborty, and
Henry Lesansky
• Court Services and Offender Supervision Agency (CSOSA)—Calvin Johnson,
Claire Johnson, and Joyce McGinnis
Pretrial Services Agency (PSA)—Susan Schaffer and Virgin Kennedy
•
Other agencies:
• Office of Property Management (OPM—Rick Gersten, Regina Payton, and
Spencer Davis
Homeless Assistance Providers and Advocates
• Anchor Mental Health—Peggy Lawrence
• Calvary Women’s Services—Kristine Thompson
• Catholic Charities—Chapman Todd
• Center for the Study of Social Policy—Jim Gibson
• Central Union Mission—David Treadwell
Appendix A: List of People Interviewed
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
78
Coalition for Nonprofit and Economic Development—Robert Pohlman
Coalition for the Homeless—Michael Ferrell and Omega Butler
Community Connections—Helen Bergman
Community Council for the Homeless at Friendship Place—Jean-Michel Giraud
Community of Hope—Kelly Sweeney McShane
Covenant House Washington—Nicole Lee
D.C. Central Kitchen—Robert Egger
D.C. Parent Training and Information Center—Danielle Greene
Downtown Business Improvement District—Chet Grey
Edgewood Brookland Family Support Collaborative—Louvenia Williams
Families Forward—Ruby Gregory and Joi Buford
House of Ruth—Crystal Nichols
Latin American Youth Center—Steve Chaplain
Latino Transitional Housing Partnership—Jarrod Elwell
Local Initiatives Support Corporation—Oramenta Newsome
My Sister’s Place—Karen Fletcher
My Sister’s Place—Nichelle Mitchem and Inga James
N Street Village—Schroeder Stribling
National Alliance to End Homelessness—Richard Hooks Wayman
New Endeavors for Women—Wanda Steptoe
Pathways to Housing—Linda Kaufman
Street Sense—Laura Thompson and Jesse Smith
Urban Living Institute—Urla Barrow
Washington Legal Clinic for the Homeless—Patricia Mullahy-Fugere, Mary Ann
Luby, Scott McNeilly, Amber Harding, Marcy Dunlap, and Andy Silver
Wesley Seminary—Ann Michel
Women Empowered Against Violence—Heather Powers
Informal conversations with case managers and shelter managers at:
- Franklin School
- New York Avenue
- Adam’s Place
- 801 East
- New Endeavors by Women
- Blair
- Madison
- D.C. General (Harriet Tubman and the hypothermia shelter)
The Community Partnership—Sue Marshall, Cornell Chapelle, Tom Fredericksen, Darlene
Mathews, Amy McPherson, Michele Salters, Clarence Stewart, Tamura Upchurch, Mathew
Winters, and Xiaowei Zheng
Appendix B: List of Acronyms
79
APPENDIX B: LIST OF ACRONYMS
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Advanced Reporting Tool (ART)
Administrative Review (AR)
Americans with Disabilities Act (ADA)
Annual Homeless Assessment Report (AHAR)
Certificates of Occupancy (CO)
Community for Creative Non-Violence (CCNV)
Comprehensive Psychiatric Evaluation Program (CPEP)
Continuum of Care (CoC)
Court Services and Offender Supervision Agency (CSOSA)
Criminal Justice Coordinating Council (CJCC)
D.C. Housing Authority (DCHA)
Department of Housing and Community Development (DHCD)
Department of Corrections (DOC)
Department of Health, Addiction Prevention and Recovery Agency (DOH/APRA)
Department of Housing and Urban Development (HUD)
Department of Human Services (DHS)
Department of Mental Health (DMH)
Emergency Rental Assistance Program (ERAP)
Emergency Shelter (ES)
Facilities Management Operations Division (FMOD)
Fiscal Year (FY)
Fire and Emergency Medical Services (FEMS)
Head of Household (HoH)
Homeless Management Information System (HMIS)
Homeless Services Reform Act (HSRA)
Housing Assistance Center (HAC)
Inspector General (IG)
Interagency Council on Homelessness (ICH)
Length of Stay (LOS)
Metropolitan Police Department (MPD)
Office of Administrative Hearings (OAH)
Office of Facilities Management (OFM)
Office of Property Management (OPM)
Permanent Supportive Housing (PSH)
Police Services Area 101 (PSA101)
Pretrial Services Agency (PSA)
Request for Proposal (RFP)
Shelter Monitoring Unit (SMU)
Shelter Plus Care (S+C)
Substance Abuse Treatment and Mental Health Service Integration Taskforce
(SATMHSI)
Transitional Housing (TH)
Appendix B: List of Acronyms
•
•
Virginia William’s Family Resource Center (VWFRC)
Washington Legal Clinic For the Homeless (WLCH)
80
APPENDIX C
Most of a Presentation by
Tom Albanese, Community Shelter Board, Columbus Ohio
“HMIS as a Tool to Measure Performance of Programs
Across the Continuum of Care”
Showing How the Community Shelter Board Structures Its
Approach to Performance Measurement
Presented at the 2006 National HMIS Conference
September 18-19, 2006
Denver Colorado
FY2006 Program Outcomes Plan (POP)
& Service Description Form
Instructions
I. Overview
A. Purpose
The Program Outcomes Plan (POP) establishes individual program performance goals for all CSB funded
programs. Performance goals must be consistent with CSB Governance Policies, Performance
Standards, and Program Certification Standards. Agencies must submit a proposed POP for each
program applying for funding in FY2007.
B. POP Monitoring:
For purposes of contract compliance, program outcomes will be evaluated for compliance with CSB
performance standards on a semi-annual basis. Success in achieving performance standards during the
current contract period, along with other factors, will form the basis for funding decisions in the next
contract period.
II. Instructions
1.
2.
3.
4.
Review the 2006-07 CSB Definitions and Performance Standards below. A description of outcome
evaluation methodologies can be found at www.csb.org.
Complete one proposed Program Outcomes Plan for each program receiving funding. The final
Program Outcomes Plan must match the agreed upon POP per each agency’s final funding award.
Section I: Program Outcomes Chart: complete for both semi-annual and annual goals. Provide an
explanation for any variance from CSB Performance Standards (see below). In some instances (e.g.
housing retention and recidivism) annual performance targets will be the same as semi-annual
periods.
Section II: Program Services Description: Complete only the applicable items based on the program
type.
III. Definitions:
1.
2.
3.
4.
Access to CSB Direct Client Assistance: (Outreach, Resource Specialists, Direct Housing and
CSB Transition Program.) The percent of distinct households receiving either CSB Transition
Program direct client assistance (DCA) or CSB FHC DCA during the period, as a percentage of
distinct household exits.
Average CSB Direct Client Assistance (DCA) Amount per Household: (Direct Housing and CSB
Transition Program.) The average amount of total CSB direct client assistance received per
household during the period. This average is calculated by dividing the total assistance for all the
households by the total distinct number of households receiving assistance.
Average Length of Stay: the number of days clients are enrolled in or receive services from the
program (from entry to exit). For Direct Housing, average length of stay is measured from the point
of enrollment in the FHC program to the exit date from the YWCA Family Center .
Change in Income from Entry to Exit: (Tier II Emergency Shelter) the average increase in total
household income amount from entry to exit. The percentage change in income is calculated by
determining the difference in total household income amount for all sources at entry from the total
household income amount at exit for all sources and dividing by the total household income amount
for all sources at entry. Change in Income considers only the total number of distinct households
that exit (i.e. latest exit for clients with multiple stays during period). Income sources may include
employment, benefits, or other sources.
2006 National HMIS Conference - Denver, Colorado
2
5.
6.
7.
8.
9.
10.
11.
12.
13.
Clients/Households Served: the number of distinct individuals and families served by the program
(including new and carry-over). For Permanent Supportive Housing, clients served must meet
Rebuilding Lives eligibility criteria. Note that clients served equals households served for Permanent
Supportive Housing.
Detox Exits: (Tier I Inebriate Shelter) the number of households served that exit to an inpatient drug
or alcohol treatment facility. The percentage of detox exits is derived by dividing the number of detox
exits by the number of total housing outcomes (i.e., the number that exits the program). Detox
outcomes represent the number of distinct households that exit (i.e., latest exit for clients with
multiple stays during period).
Housing Retention: (Permanent Supportive Housing) the percent of clients who maintain their
housing, whether or not as part of the Permanent Supportive Housing program, and do not return to
emergency shelter within two weeks to three months of exit from the program.
Housing Stability: (Permanent Supportive Housing) the average length of time, measured in
months, that clients reside in the Permanent Supportive Housing project.
Movement: the percentage of exited clients who have any contact with another shelter within 7 days
of exit from the program. Movement is not applicable to Tier I family shelter and Resource
Specialists.
New Households Served: (Direct Housing) Number of distinct households that entered the program
during the period.
Program Occupancy Rate: (Tier II Emergency Shelter and Permanent Supportive Housing) the
average daily occupancy as a percentage of the program capacity.
Recidivism: the percentage of exited clients with a successful housing outcome (permanent or
transitional) who have any shelter contact within two weeks to three months of a successful housing
outcome, expressed as a percentage of total distinct clients with an exit to permanent or transitional.
Successful Housing Outcome: the number of households served that exit to permanent or
transitional housing. For Outreach, this includes Tier I shelters. For Tier I family shelter, this
includes Tier II family shelters. For Direct Housing, this excludes transitional housing. Note: for
PSH, deceased clients are not included in the count of exited clients. The percentage of successful
housing outcomes is derived by dividing the number of successful housing outcomes by the number
of total housing outcomes (i.e., the number that exits the program). For all programs excluding PSH,
housing outcomes represent the number of distinct households that exit (i.e., latest exit for clients
with multiple stays during period): for PSH, clients who remain in the program are considered
successful housing outcomes. Refer to Table 1 below for a complete list of housing outcomes.
Table 1: Successful Housing Outcomes
The following table identifies various destinations, including successful housing and shelter outcomes, as
identified in the CSB HMIS Data Definitions for 2005-06. Housing outcomes correspond to ServicePoint
pick list choices for ‘destination’ and are used to determine shelter and/or housing outcomes in the 2006
Program Evaluation.
ServicePoint Destination
CSB Definition
Client
Control
of
Housing?
CSB Evaluation Element
(1)
Permanent Housing: Rental
house/apartment (no
subsidy)
Privately owned, market rent housing (not
subsidized)
Permanent Housing: Public
Housing
Housing owned and subsidized by CMHA
Permanent: Section 8
Housing owned by a private landlord or
partner agency and subsidized through a
CMHA Section 8 Voucher or through Section
8 project-based subsidy
2006 National HMIS Conference - Denver, Colorado
Yes
Successful Housing
Outcome
Yes
Successful Housing
Outcome
Yes
Successful Housing
Outcome
3
ServicePoint Destination
CSB Definition
Client
Control
of
Housing?
CSB Evaluation Element
(1)
Permanent: Shelter Plus
Care
Housing owned by a private landlord or
partner agency and subsidized through the
Shelter Plus Care program administered by
CMHA
Yes
Successful Housing
Outcome
Permanent: HOME
subsidized
house/apartment
The YMCA Permanent Supportive Housing
(PSH) program at 40 West Long Street or
SE Scattered Site Permanent Supportive
Housing
Yes
Successful Housing
Outcome
Permanent: Other
subsidized
house/apartment
Housing owned by a private landlord or
partner agency that has an on-going subsidy
through HUD 202 or 811 program, tax
credits, or other sources, including HUD and
CSB
Yes
Successful Housing
Outcome
Permanent: Home
ownership
Housing that is owned by the client
Yes
Successful Housing
Outcome
Permanent: Moved in with
Family/Friends
DO NOT USE
N/A
N/A-Not a Housing Outcome
Transitional: Transitional
housing for homeless
Transitional (i.e. New Horizons)
Varies
Successful Housing
Outcome (except for Family
Housing Collaborative)
Transitional: Moved in with
Family/Friends
Temporary housing with family or friends
No
Unsuccessful Housing
Outcome
Institution: Psychiatric
hospital
Temporary/indefinite residence in a
psychiatric hospital for the treatment of
severe mental illness
No
Unsuccessful Housing
Outcome
Institution: Inpatient
alcohol/drug facility
Temporary/indefinite residence in an
inpatient facility for treatment of alcohol
and/or drug addiction
No
Unsuccessful Housing
Outcome
Institution: Jail/prison
Incarceration in local, state or federal prison
No
Unsuccessful Housing
Outcome
Emergency Shelter
Emergency Shelter (all including Tier II
shelters)
No
Unsuccessful Housing
Outcome (except: YWCAFamily Center = Successful
Housing Outcome; Outreach
= Successful Housing
Outcome)
N/A
N/A-Not a Housing Outcome
No
Unsuccessful Housing
Outcome
No
Unsuccessful Housing
Outcome
N/A
Unsuccessful Housing
Outcome
Other: Other Supportive
Housing
DO NOT USE
Other: Places not meant for
habitation (street)
Street, condemned buildings, etc.
Other
Hotel, other
Unknown
(1) Client’s are determined to be in control of their housing if the lease/mortgage is in their name or if they otherwise
have a written agreement that gives them a right to reside in their housing, such as a roommate agreement.
2006 National HMIS Conference - Denver, Colorado
4
14. Successful Income Outcome: (Resource Specialist) the number of households served that do not
have income at intake and that have income at exit. The percentage of successful income outcomes
is calculated by dividing the number of successful income outcomes by the number of total income
outcomes. Income outcomes considers only the total number of distinct households that exit (i.e.
latest exit for clients with multiple stays during period). Income sources may include employment,
benefits, or other sources.
15. Successful Permanent Housing Outcome: (Permanent Supportive Housing) the number of
households served that remain in the Permanent Supportive Housing program or exit the program
for other permanent housing and the percentage this represents of total distinct households served.
16. Turnover Rate: (Permanent Supportive Housing) the rate at which units become vacant relative to
the number of units occupied. Turnover rate is calculated by dividing the total units becoming vacant
during a semiannual period by the number of units occupied during the same period.
IV. 2006-07 CSB Performance Standards
Homelessness Prevention
Ends
Measurement
Annual Goal
Housing Stabilized
Successful Housing Outcomes
Do not enter shelter
Recidivism
Resources and services to maintain
housing
Pass program certification
Provide access to and coordination
with community resources to prevent
homelessness.
Efficient use of a pool of community
resources
CSB percent of total program budget
CSB funds will only be a portion of
total program budget
95% will maintain or obtain
permanent or transitional housing
<5% will return to shelter
Outreach Specialist
Ends
Measurement
Annual Goal
Basic human needs met in secure,
decent environment
Pass program certification
Provide access to secure, decent
shelter and housing.
Housing Stabilized
Successful housing outcomes
At least 60% successful housing
outcomes
Do not return to shelter
Recidivism
Access to resources/services to
maintain housing
Access to CSB Direct Client
Assistance
At least 25% will receive CSB DCA
Efficient use of a pool of community
resources
CSB percent of total budget
CSB funds will only be a portion of
total program budget
<10% will return to shelter
Tier I/Emergency Shelter
Ends
Measurement
Access to resources to avoid shelter
admission and stabilize housing
-Pass program certification
-Detox Exits (Inebriate Shelter Only)
2006 National HMIS Conference - Denver, Colorado
Annual Goal
-Provide access to and coordination
with community resources to avoid
shelter admission and stabilize
housing.
-At least 10% of inebriate shelter
exits will enter a detoxification
program
5
Ends
Measurement
Annual Goal
Basic needs met in secure, decent
environment
Pass program certification
Provide secure, decent shelter.
Temporary, short-term stay
Average Length of Stay
Average stay not to exceed:
•
30 days for men’s shelter
•
28 days for women’s shelter
•
20 days for family shelter
•
12 days for inebriate shelter
Move to positive, stable housing
Successful Housing Outcomes
Obtain permanent or transitional
housing:
•
At least 15% for men’s shelter
•
At least 24% for women’s
shelter
•
At least 70% for family shelter
(includes Tier II shelter)
•
At least 5% for inebriate shelter
Do not re-enter the emergency
shelter system
Recidivism
<10% will return to shelter (men’s,
women’s, family shelter)
<20% will return to shelter (inebriate)
Do not move inappropriately to other
emergency shelter
Movement
<20% will move to other shelter
Ongoing engagement with the
neighborhood
Pass program certification
Actively participate in community
engagement process.
Efficient use of a pool of community
resources
CSB percent of total budget
CSB funds will only be a portion of
total shelter budget
Tier II Shelter
Ends
Measurement
Annual Goal
Basic needs met in a noncongregate environment
-Pass program certification
Programs will provide private or
shared living units.
Increased income
Change in income from entry to exit
Increase by at least 30%
Temporary, short-term stay
Average Length of Stay
Average stay not to exceed 80 days
Move to positive, stable housing
Successful Housing Outcomes
At least 70% will obtain permanent or
transitional housing
Do not re-enter the emergency
shelter system
Recidivism
<8% will return to shelter
Do not move inappropriately to other
emergency shelter
Movement
<10% will move to other shelter
Ongoing engagement with the
neighborhood
Pass program certification
Actively participate in community
engagement process.
Efficient use of a pool of community
resources
-Program Occupancy Rate
-CSB percent of total budget
-At least 95% occupancy rate
-CSB funds will only be a portion of
total shelter budget
2006 National HMIS Conference - Denver, Colorado
6
Resource Specialist
Ends
Basic needs met in a noncongregate environment
Measurement
Pass program certification
Successful Housing Outcomes
Housing Stabilized
Successful Housing Outcomes
Do not re-enter the emergency
shelter system
Recidivism
Access to resources/services to
maintain housing
-Successful Income Outcomes
-Access to CSB Direct Client
Assistance
Annual Goal
-Assist clients in accessing an
independent apartment
-See Housing Stabilized
-At least 45% for adult shelters
-At least 70% for family shelters
<5% will return to shelter
-At least 40% of clients with no
income at entry will obtain income by
exit.
-Agency to set own goal for FY06
Direct Housing
Ends
Measurement
Basic needs met in a noncongregate environment
-Pass program certification
-Independent apartment
Housing Stabilized
Successful Housing Outcomes
Temporary, short-term stay in shelter
Average Length of Stay
Do not re-enter the emergency
shelter system
Recidivism
Access to resources/services to
maintain housing
Access to CSB Direct Client
Assistance
Annual Goal
-All programs will pass certification
-Programs will assist clients in
accessing an independent apartment
At least 90%
Average stay not to exceed 20 days
<5% will return to shelter
At least 90% will receive CSB DCA
Permanent Supportive Housing
Ends
Measurement
Annual Goal
Basic needs met in a noncongregate environment
Pass program certification
Increased income
Change in income
Housing stabilized
-Housing Stability
-Successful Permanent Housing
Outcomes
-At least 12 months housing stability
-At least 90% successful permanent
housing outcomes
Do not re-enter the emergency
shelter system
Housing Retention
At least 90% will maintain permanent
housing
On-going engagement with the
neighborhood
Pass program certification
Actively participate in community
engagement process.
Efficient use of a pool of community
resources
-CSB percent of total budget
-Program Occupancy Rate
-CSB provides funding as member of
Rebuilding Lives Funder
Collaborative
-CSB funds will only be a portion of
total program budget
-At least 95% occupancy rate
-Housing is long-term (no time limit)
2006 National HMIS Conference - Denver, Colorado
At least 45% will increase their
income 6 months after entry
7
Tier 1 Adult Shelter
Program Performance Indicators
Agency:
Program:
HMIS Program Data Source:
XYZ Agency
Men's Shelter
XYZ Men's Shelter
CSB Performance Indicators
FY2005
Quarter 1
Quarter 2
Semi-Annual
Quarter 3
Quarter 4
Semi-Annual
Annual
7/1/04-9/30/04
10/1/04-12/31/04
7/1/04-12/31/04
1/1/05-3/31/05
4/1/05-6/30/05
1/1/05-6/30/05
7/1/04-6/30/05
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
POP Goal
Actual
Variance
# Households Served
450
481
31
450
606
156
700
803
103
450
468
18
450
496
46
700
736
36
1400
1253
-147
Average Length of Stay (Days)
30
38
8
30
22
-8
30
30
0
30
29
-1
30
27
-3
30
-30
30
# Successful Housing Outcomes
61
50
-11
61
61
0
127
101
-26
61
34
-27
61
63
2
127
90
-37
254
165
-89
% Successful Housing Outcomes
20%
13%
-7%
20%
22%
2%
20%
15%
-5%
20%
11%
-9%
20%
18%
-2%
20%
15%
-5%
20%
15%
-5%
% Recidivism
<10%
<10%
<10%
9%
<10%
<10%
<10%
<10%
% Movement
<20%
<20%
<20%
22%
<20%
<20%
<20%
<20%
-30
Note:
FY2005 quarter projections assumes 200 duplicated (carry-over & recidivist) households between quarters.
Successful housing outcomes based on total households served minus active households at end of period (145) to determine total exits.
SA2 & Annual Households Served and Successful Housing Outcomes numbers derived from the Columbus Report.
CSB Performance Indicators
Tier 1 Shelter
FY2006
Quarter 1
Quarter 2
Semi-Annual
Quarter 3
Quarter 4
Semi-Annual
Annual
7/1/05-9/30/05
10/1/05-12/31/05
7/1/05-12/31/05
1/1/06-3/31/06
4/1/06-6/30/06
1/1/06-6/30/06
7/1/05-6/30/06
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
POP Goal
# Households Served
450
449
-1
450
316
-134
700
631
-69
450
356
-94
450
700
1200
Average Length of Stay (Days)
30
29
-1
30
42
12
30
41
11
30
37
7
30
30
30
# Successful Housing Outcomes
61
64
3
61
27
-34
111
102
-9
61
40
-21
64
111
211
% Successful Housing Outcomes
20%
20%
0%
20%
14%
-6%
20%
21%
1%
20%
19%
-1%
20%
20%
20%
% Recidivism
10%
NM
NM
10%
NM
NM
10%
14%
NM
10%
NM
NM
10%
10%
10%
% Movement
20%
NM
NM
20%
NM
NM
20%
NM
NM
20%
NM
NM
20%
20%
20%
Actual
Variance
Note:
FY2006 assumes 200 duplicated (carry-over & recidivist) households between semi-annual periods and 200 between quarters.
Successful housing outcomes based on total households served minus active households at end of period (145) to determine total exits.
2006 National HMIS Conference - Denver, Colorado
8
CSB Performance Indicators
Tier 1 Shelter
FY2007
Projected
Quarter 1
Quarter 2
Semi-Annual
Quarter 3
Quarter 4
Semi-Annual
Annual
7/1/06-9/30/06
10/1/06-12/31/06
7/1/06-12/31/06
1/1/07-3/31/07
4/1/07-6/30/07
1/1/07-6/30/07
7/1/07-6/30/08
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
POP Goal
# Households Served
450
450
700
450
450
700
1200
Average Length of Stay (Days)
30
30
30
30
30
30
30
# Successful Housing Outcomes
61
61
111
61
64
100
190
% Successful Housing Outcomes
20%
20%
20%
20%
20%
18%
18%
% Recidivism
10%
10%
10%
10%
10%
10%
10%
% Movement
20%
20%
20%
20%
20%
20%
20%
Actual
Variance
Note:
FY2007 assumes 200 duplicated (carry-over & recidivist) households between semi-annual periods and 200 between quarters.
Successful housing outcomes based on total households served minus active households at end of period (145) to determine total exits.
CSB Performance Indicators
Tier 1 Shelter
FY2008
Projected
Quarter 1
Quarter 2
Semi-Annual
Quarter 3
Quarter 4
Semi-Annual
Annual
7/1/07-9/30/07
10/1/07-12/31/07
7/1/07-12/31/07
1/1/08-3/31/08
4/1/08-6/30/08
1/1/08-6/30/08
7/1/07-6/30/08
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
Projected
Actual
Variance
Projected
Actual
Variance
POP Goal
Actual
Variance
POP Goal
# Households Served
450
450
700
450
450
700
1200
Average Length of Stay (Days)
30
30
30
30
30
30
30
# Successful Housing Outcomes
61
61
111
61
64
100
190
% Successful Housing Outcomes
20%
20%
20%
20%
20%
18%
18%
% Recidivism
10%
10%
10%
10%
10%
10%
10%
% Movement
20%
20%
20%
20%
20%
20%
20%
Actual
Variance
Note:
FY2008 assumes 200 duplicated (carry-over & recidivist) households between semi-annual periods and 200 between quarters.
Successful housing outcomes based on total households served minus active households at end of period (145) to determine total exits.
2006 National HMIS Conference - Denver, Colorado
9
CommunityShelterBoard
Systemand ProgramlndicatorReport
FY06Quarter3: 1/1/06-3/31/06
Updated3/15/06
Summary
of programsand systemshavemet
the firstquarterthat 1OO"/"
This reportrepresents
dataqualitystandardsat initialSPIRgeneration.Thus,this reportcontainsdatafor all
systemsand programs.
GontinuallyAchievingPrograms& Systems:
./ All Permanent
Housingprograms
Supportive
,/ Tier 1 & Tier2 FamilyShelters
/ YWCAResourceSpecialist
,/ GladdenCommunityHousePrevention
./ CSBTransition
,/ HomelessFamiliesFoundation
ResourceSpecialist
/ Maryhaven
Engagement
Center
,/ VOA Men'sShelter
,/ SalvationArmyFamilyHousingCollaborative
lmprovedPrograms
./ Friendsof the HomelessRebecca's
Place
Programs& Systemsof Concern
,/ FaithMissionResourceSpecialists
./ Friendsof the HomelessResource
Specialists
,/ Friendsof the HomelessMen'sShelter
{ LSS--Faith
Mission-FaithMissionon 8th
,/ LSS--Faith
Mission-Nancy'sPlace
/ Maryhaven
Outreach
{ The men'ssystem
,/ The AdultResourceSpecialist
system
to staffat bothFaithMissionand
technicalassistance
CSBstaffhas providedextensive
hasbeenprovided
Friendsof the Homelessto improvetheirprograms.Thisassistance
levels.
program
direct
care
and
level
as
well
as
supervisory
at senior
conlirmations\SPIRsummary 5-22-06.doc
2005-2006\05-22-06\Monitoring
S:\Advmacy-Leadership\Board\Meetings
2006 National HMIS Conference - Denver, Colorado
10
EE-€
o
o
i. g8 $
L
be
E
E- i
(/)E
ewoclno
uawoAsltlcv
E ea tb
'. }E
U'
(D
Eg E
E E'6
<jao
o
o
=
E
E
at
GIJ.
=
o
=
E
o
o
a
awoclna
luaweLalqcv
U'
o
o
o
o
U,
ft) pnlctr
-7
be be
sAl C\I
f.-
o-< iil,f
rf)
tiw
:+1:
!:!r.!
be bS be
be
lr) $
o
9) poa
o
E 9
c >
bS
o
c\,1
(o7)prycy
E
-7
i
(\l
F.-
+l
'fl
7
(o
(o
O)
@
(o
rr)
ft) pos
@
@
o
f\
rr)
O)
@
7
}L
dq
be
-(D
srr)
be
rr)
.f
bS
}L
z
@
rifr,'i
be
o
f.-
iii
@
s
rr)
(\l
f.-
<\t
@
O
r-
=f
(o
O)
@
6
o
o
c\l
(o
rr)
s
(\l
=
o
E
EEE
ewocnc
tuawa^elqcv
s€
lenlpv
C ) >
(E
9'
-7
+l
-7
\t
s(\J
|r) (o
o
@
(\.l
O
c\,1
cf)
i
b(/,
*i:at+l
(o
iini\t
{
z
{
z
=
3
6
E
b,
z. z.
o
i=
po5
(f)
(\l
O
@
z. z.
;
.9
F
cl
c
rg
cl
pyog
fl pawatqqcvowocnC
z. z. z. z.
=
o
o
o
-7
{ {
z. z.
E
@
E
pnipl
@
(o
s
(f)
o
r()
CD
rr)
.tr
(\l
(o
-oo
z. z.
@
=
2
-c
.s
z
E
rr)
, trgcedeS f-CD
at,
E
E
E
6 =
(/|0A,
$
Al
o)
o
r,
o
r.)
o
(o
q,
@
cf)
c0
rr)
(o
(\,1
@
o
c!
6
o
o
'6
z. z.
@
O)
t*
o
(o
.9
6
-7
Pnipv
@
'tL
z.
=
o
F
i
t a
E
Q
!
=
Y
@
X E ;
" € t E
'
;
=
g
.
<
t
(l)
G
(E
s
o
o
(D
F
g
o
o
(D
F(
E
o
c
(D
c
(l)
E
o
=
.>
E
<6
LL
2006 National HMIS Conference - Denver, Colorado
(D
(5
-o
(D
c
E
(lt
LL
io
=
E
(5
LL
(g
o <l)
E
E
o E E
o
e5
o
a) a) a)
t
6
@
9
E
I
E
h
=
^
4
o
<
=
-
o
e
.
o
.
o
e
=
-
o.
.!P
-c o
c
(g
i6
o
ct
o
>
i
=
o
E
x
o
:
i = t s E + E
e E E d . C
O
O
=
\
>
\
=l q c g E -
(t,
o
s .9
o
J
:o.
)
F
c
;
(D
-o
(g
q
Z
@
:
o
an
o
!
d
d
=
i
=
=
Y
x
@
)
=
; ' . =
. h E
c
d
E
6
*
,
6
;
o
*
c
D
-
, E . E O b
. ! a E : E
E = f i 6 -
+ E E - sE
8 s - E i
11
5
L
E
E-€
Fes
@
o
N
o
o
a
.L
tfiF
o = -
aE,;
=sE
EE3
ctt
'6
E
9
cll
E
o
g6
o
7
7
c n
e 6 t
r E
= o
o
awoqna
uawaLalqcv
E ( )
o =
(,
5
a
$) pnpy
&)pos
=
.a^
(g(, ,
c t =
(st11uow
71,
PocloNewulenv
3
(o
t\
E
7
ED=
E=
5
# pn|pt
o
C'
o
(\l
(y,90)poClo Uewulenl
.E
a
o
o
o
.!
g
o
o,
o/oien4cl
#pn|''v
bS
('o
o)
@
r()
tt,
!t
EE
6 =
2E
-
Pn|pv
Atcedea
f(f)
(o
-7
@
o
(o
'tt
.o'
Y
o
'-cD
5
o
G
tg
o
o
o
o
E
o
o.
o,
U)
_!P
o(g
E
c)
F
.e
s
(J
q)
z.
aq)
E
o
.!P
()
.E,
6
ot
E
o
(5
o
c
(D
.s)
o
N
o
oo(U
o
c
E
o
o
E
o
o)
.g
E
;
(D
o)
E
E E
o E
ss ()o= oo= o()
o (1 a)
E
(-l
o
e
o-
I
J
E
o
E
o
U)
o
(t
N
L
2006 National HMIS Conference - Denver, Colorado
12
EE€
€ s-$
@
o
EE=
awoqnc
luawa^alq?v
(,.5-
'. >it ref i ia
U'
o
F SE
o
E
o
E O
E >
=
o
E
.=
at,
E
o
E
EEg
Gll.
(fi prucj
-7
-7
'tL
7
bA bQ
be dQ
r() (f)
o) s
be ;e
F-
@
;e
rr) s
rr)
ss
be
o
C\I
PApos
-7
bA ba
ro
<\
C!
c\
c?)
'7
;a be
o
s
F-
o
s<\
;e ;e be
o
F-
o
t-
=
6
o
o
q,
o
<)
=
at
o
awoqnc
luawa^alqcv
&) prycv
&)pos
E
EEE
q 8
g'e
$6
i
-7
-H.
'tL 7
cY)
F-
c!
.if
v
tr
@
CD
i;i,il 'tL
-7
-7
@
o
c!
cr)
CD st
c\t
C\
ro
-7
TT 'tL
awoqnc
uewa^alqcv
o)
c!
pnpl
leo!
ag
ct
=
()
(,
o
e len4cl
.9
z,
, filcede!
@
c\l
c.)
t(\l
c\
s
l--
-7
CV)
=
=
co
cv)
z. z. z. z.
o)
@
o
<\
cr)
c\l
cr)
<\
c!
@
s
:<
o
ewocnc
luowoLalqcv
rr)
o)
F-f
]
z. z.
ro
o
7
rr)
O
C\I
<\l
z.
z.
ro
o
rr)
c\t
rr)
cv)
c\
<f)
rr) ocr) o
o)
$
<\l
-f
F*
.if
(o
CE
$
c!
rr)
'tl
'tl.
.fL
l'ct)
ecueuel
(\ 6\t
\t
\t
g
(f)
-7
=
Ieryrl
IEOE
<\l
$
o
(o
s
<\l
rr)
O
r()
(f)
c\l
l
(D
(o
o
E
(E
E)
o
o-
|rt
o
z,
IJ,I
c,
E.
UI
=
UJ
U'
(n
-a
@
o
c
.o
@
.a
-c
-(E .=
(6
LL
LL
E
o
I
o
-c
o
.D
E
c
(l)
LL
(g
o
o
(D
-o
o
E.
o-
_c
<t)
(o
c
o
c
'6
c!
(o
c\l
.rf
o
rr)
rr)
s
7
F\
o)
rr)
-7
@
@
'tl
r()
z.
=e
c)
-g
o-U'
-@
c
o
o
@
s
ro
s
<\t
-c
U)
G
O
C!
O
C\I
-<D
o
E
LU
F
J
UI
T
C\I
@
c\
c\,l
c\l
<N
-f
a
rf)
s3 N(o
:
r
s
(o
6
CD
a^
o
o
o
o
Fc!
f.r
!t(l)
o
tt
:
:r
() (o7og
5)7uew eaelqzv ewoqnC
o
;e
rr)
f.-
f
o
(E
-<
Lr)
t-
-c
U)
_o
o
o
2006 National HMIS Conference - Denver, Colorado
z.
I
=
(D
q)
.o
at)
.U'
o
= EU'
-_
o
'=
(d
LL
LL
c
o
C)
c
(D
o
(D
q)
eE
-o)
LL
o
o
o
o)
=
-c
(g U)
E
E
LL
o
C)
=
E
(o
LL
(J
E
o
o)
(t'
0)
c
LU
c
(D
(6
b
(g
g
4
a
=
6
E
o
.F
(E
E
LL
U'
E
(D
-(D
E
4
@
t/,
c)
o
e i +
+ s e Q
_g
.q)
Y
E
(|)
s
g
.E
()
(l)
, b
o
o
+
B
5
E
P
R
,
9
-
:
E
a
o
:
Y
(g
E
o)
E .9
o) -c
()
.q)
(5
(5
o
E
o
()
o
5
o)
E
o
=
.s2
o(6
o
c
-i6
o
o)
(D
E
o
O
o o o o
=
o
o
o
F
=
a
'-
E
o
A
E
+
F
=
d
>
=
F
!
X
!
>
=
X
,=
o
:
o
l
d
S <+ R
:
o
O
13
r
=
*
a - q5
g
P
9
t
'
|
^i e
b
E
E
a
G
!
<
e
i
t s
L
EES
@
o
6
r. 9tb 9$
EEi
3E;
= E C )
; Sg l E
x
=
E9 6
6 a o
E9
G
II.
E
g
to
o
ewoqnc
luawo^elqcv
'tt
7
'Tl
;9
bS bS
(o
-f
z
at,
q,
E
o
o
(o7)pnlcy
o
ct
'6
o
=
PtJpog
@
cf)
t-
C\,1
bS
rO
o)
be
;9 be
be
lf)
o lr) o
* t\ s t\
.}t
7
'tt
ft)prycv
lr)
(o
c.)
o
o
(o
fu)pos
@
@
s(o
s<D
ewoctna
luawe^elqcv
z.
tt,
u,
(l,
o
C'
=
@
U'
E
()
(|)
U'
awoclno
wewe^elqcv
(\l
E
q)
c
{
z.
.9
g
z. z.
q
acuelEA
o
\i.
o
N
o
(o
o
E
(5
a
(o
@
o)
E'
o
o
o
=
o
E
pnlpv
po3
o
<D
t\
o,
(fJ
@
r
@
E
o
a
(o
o
+
(\t
|f)
*(o
l
o
o
.
c r o
-7
@
s
(f)
'fL
E €
z.
E *
= c )
=
6
E
E
( l ) q
E
5
* d
>
o
o
6
f
-o
c
o
(E
(t
o
J
o
o
o
()
t!
oU'
lrl
E
G
Er
o
o.
o
E
=
o
U'
u,
NE
-ut
U'
_g
(D
E
o
-r-
o)
o
U'
E
c
o
LL
C'
E
g*
U'
o
()
o '=
lJ(t,
o
U)
6
'6
E
(E
o
lL
U' U)
g
U'
o
6
E
o
-
(t
o
-c
f
J
2006 National HMIS Conference - Denver, Colorado
-c
-o
tr
(|)
\
(1)
c
(D
()
.E
(6
LL
()
=
>
(l)
E
o
eS
o)
H ' o
= . c t
=
e
E
C
L
O
O
F
E
o
F
- = o
o t o
o .9
-c
o
(J
.9 <5
. o o
r o P
F
g
.E
s()
(U
.-oo
c) o(5
E
E
-c
o
(5
o)
E
o
<J
E
o o
c o,
(D
o
E E
o o
o
o
= 3 =
o c) c) c)
=
o . =
>
o
H E
o
E
o
s
=
@
o
F ; 8
" q
i 6
14
EF
E
e8S
o
q
L
fiEF
"l
o = F
aE
'
=
-
L
= E e
= F E
= E E
5e6
Oero
E9
IE IL
lnun olqcvawoqnc z.
o
o
E
o
o
(o7)pryc1
=
o
=
o
o)
$nwa olqcvewoqnc z
E
=
at
o
o
C'
o
=
U'
&)p'l,lctr
h)pos
ro
(f)
(o
(o
e (st4uow71)pog to uawupul
.g.s
2'E E
= ( E o
ia;=
&)Pqct
FE
ro
U'
a,
=o
s
o
U'
=
o
-
E
G
ctl
o
o-
F
ul
F
E
o
c
4
=
att
z.
z
z_
ct
<\l
|r)
(D
o
rr)
o)
+
cr)
$l
pol
o)
@
(\I
O)
$
o)
@
dS
o
o,
-7
(\l
(\l
C\I
bs
@
CD
@
<\l
o,
Al
o,
tt
<t,
<\l
bR b<
<7'
t-
o
(r)
t.
(o
<\.l
-7
-7
->
(o
s!
O)
(o
<7)
(\,1
ro
s
$
ro
F.
{
ro
(\l
(D
o)
o)
(l,
U)
U)
-c
.9)
C\T
cn
c
o
=
o
o)
c
@
o
<\l
<\l
K)
CD
E
(E
at
o
6
o
c
o
2
(!
z. z.
o,
c
@
=
o
T
e
o
( f
=
(o)
E
E
o
( f
(O
(o
sro
O)
CD
F(0
(o
-7
ro
O)
CD
(\l
(\l
@
F.
as
o
'
s|r)
s.< B
E
B- ; 2
6
E
. 5
G
8 EE
s 8i
iE € > ;
E
p
E
E
CD
@
(o
z.
z.
{
ES
=
(o
--7
-7
(o
F\
@
CD
t-.
(o
\l
;s
rr)
t-
@
z.
(,
@
N
o
ro
lr)
t-
o
F.
s!
6
F.
*F-
ro
@
(o
(9
t-
ct)
ro
(o
CD
@
(o
E
v
E
A
=
P
' E EE
=
A
e
E 9 5
F
F gE
g
;' ;
: - ; 8
E E E
B d F
€
6 .
I
g
E
E }
(o
CD
E
>
s 3 E
; €
z.
(0
<o
\t
E
H
E 3 F
t-
->
@
(o
o)
o)
(f,
ro
@
cf)
O)
O)
o
ro
o
--7
->
*
t
5 * E
E A
r
E
: BE
gH
E E :
E
+a " ,E
s ? E E F a
E -
E e ;
E ;
E gg E+ E
3 g EE $ *
-
Y
-
€ E
E
,
H
E
=
H H H
E * e= E = 5 =
H # S 5 S ES
" ; - 3F F - E * F
->
o
t|.
z.
(,
o
(l)
- I >
'= '= .E
=
=
E
E
q
c
(6
z. z () IL
: x il *
o
o
o
o
= =
= E
.=
(l)
o)
o
o
o
z. z.
o
sro ;s
s ;s
;s
o
o,
t\
o
bS
c')
(|)
E
o
r'-
-7
(o
o
E
o
-7
6
C\l
(o
(o
-
<\t
<\,1 (0
o
(t
E
c
6l
(\l
C!
f.-
@
(t,
E
=
i\
(o
(')
Al
cr)
rrt
ro
o
o
af
=
F-
r)
<\t
co
(D
E
E
@
t.
<\l
o)
(7)
(f)
(o
o
.(f
bs
o
z
co
<\l
CD
-7
a
<t,
q,
.c
o
o
ct)
o
(t)
(f)
(o
;9
<\I
as
;s
as
;e
as
o
o
o
o
o
(.)
(f)
z
CD
7
-7
o
TU
v
o
bS
ro
-7
*(\l
x
o
;s
;s
ro tr)
F(a
o
(5
o
o
bs
s()
o
o)
sz
as
pni,'l
CD
@
@
acueueA
c
lr)
o
d
*
-7
z
O)
tuaun^alq?vawoqnc
=
c
c,
z
a
=
o
o)
(f)
Apede(
(E
bS
o
o,
F-
(APncv ;s ;9
CD
€ ;s
o) CD o,
(7'
=
(,
FO)
-7
ft)Pucv(o
!t
o
z. z.
-7
-7
-7
(o/o96)
Ieoep tluewuleul 7
'6
z
-7
sro s ;s ;s (os s s
9Jpor o;e
CD
C'
c
at
o
g
E
@
E
o
E
o
H
>r
.E
3
E
q)
c
@
5
o
(E
(|)
G
'6
(J
(d
z.
ct
c
'6
g,
o
.=
@
5
o
2006 National HMIS Conference - Denver, Colorado
E
E
o
a)
E
o
o
E
g
C)
(d
E
c)
o
6
o
<t)
o
(g
(1)
-c
ct
c
o
J
a
o
=
o
s
E
= ()
=
o
(o) z. (/)
.o
d
N
3
=c
E
@
q)
z
I
o)
.F
6
E
E
-t
(/)
()o
:
(l)
o
= E oc oo) E
(l)
o) .=
p (t, U)
U)
v
o
- - '= '- '=
= = =
E
E
(o)
o
c
o
o)
o
s
q)
F
(l)
6
c)
F
F
O
TL
o
o)
=@ U)
(' =
z. CD
c
U) = .=
==
o () -o
= cro,
t
8-
E
s
g
E
E
.9
€()
sq)
E
e
R
o
o
o
E
6
o
d
s
g
(!
t
o
e
o
E
I
G
o
o
o
e
@
a
N
c
c)
u
15