Palliative Care for Homeless Individuals:
Emerging Models
MAYA BERKOWITZ

JUNE 2016
Faculty Advisor: Chris Hurley
Site Supervisor: Ed Dwyer-O’Connor

 Table of Contents
Introduction .................................................................... 2
Making the Case for Emerging Models ............................... 3
Exploring the Research and Field of Practice ...................... 3
Defining Palliative Care and Hospice ............................................... 3
Choosing a Mobile Model ............................................................... 3
Care Needs for Homeless Individuals............................................... 4
Range of End of Life Care Services for Homeless Individuals .............. 5
Examples of End of Life Care Models Internationally ........................6

Common MPCH Program Components .............................. 7
Patient Referral Process................................................................. 7
Program Staffing Models ...............................................................8
Program Size ................................................................................9
Patient Eligibility Criteria ...............................................................9
End of Life Planning Conversations ............................................... 11
Geographic and Site-based Limitations ......................................... 12
Smaller Programs Benefit from Larger Infrastructure ..................... 13
Training The Next Generation ...................................................... 14
Access to Low-Intervention End of Life Beds.................................. 14

Program Funding ...........................................................16
Conclusion ..................................................................... 17
Contributions .................................................................18
Sources and Interviews ...................................................19

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

1

 Introduction
This report was developed as a partnership between The University of
Washington School of Public Health’s Community-Oriented Public Health
Practice Program and the University of Washington Medical Center (UWMC)/
Harborview Medical Center’s (HMC) Mobile Palliative Care program for
Homeless Individuals (MPCH). The goal of the report is to document the need
for emerging MPCH programs and identify critical program components for
successful implementation and sustainability. With HMC’s MPCH venture in
the initial pilot phase there is an interest and benefit to learning more about
what others are doing. In serving this purpose this report aims to:


Further document the need for flexible palliative care services that meet
the specific needs of homeless individuals; and



Compile and articulate the experiences and learnings of providers and
researchers from around the world about the process of building and
maintaining end of life care programs for homeless individuals.

populations. Health care providers currently working in the field that were
interviewed for this project are in the process of developing new models of care
to better serve marginalized communities to provide comprehensive,
accessible and affordable end of life care.

In January of 2014, HMC began a pilot MPCH program based out of the Pioneer
Square Clinic. Public Health – Seattle & King County (PHSKC) funded this
program through grant dollars from the Health Resources and Services
Administration (HRSA). The goal of the program is to address the end of life
care and pain management needs of disenfranchised and hard to reach
homeless individuals in the Seattle area, many of who have complex care
needs. The program aims to support patients at the end of their life with
holistic care and to give dying individuals more control of their care and death,
with the goal of preventing unnecessary emergency room visits and long inpatient hospital admissions. As HMC evaluates the early results of their pilot,
they are interested in understanding how other MPCH programs provide
similar services in order to strengthen and expand their current program.
Most MPCH programs internationally have begun in the last 3-5 years. This
expansion follows the growing acceptance of hospice care and general
palliative care programs in the last 20 years throughout the developed world.
Growth and acceptance of the field of palliative care in healthcare has paved
the way for specialized programs such as MPCH, tailored to specific

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

2

 Making the Case for Emerging Models
Older homeless individuals often need support managing chronic illnesses and
pain. While there are a wide variety of homeless health programs that have
been implemented nationally and internationally very few programs are
designated to providing palliative care. MPCH has a unique focus on palliative
care and providing services at a time and place in people’s lives where they
have increased care needs. To meet these needs, MPCH programs help
homeless individuals manage pain and connect with other health care and
social services. Initial findings from the Seattle-based MPHC indicate that the
program is also reducing emergency room and in-patient hospital stays,
improving quality of life and reducing the cost of care.1
One of the goals of MPCH is to provide palliative care to a medically and
socially complex population, reducing the use of unnecessary emergency
department and in-patient hospital care. The Seattle-based program has data
demonstrating 25% reduction in days spent in the hospital for individuals after
six month enrollment in the MPCH program (1,611 total days 6 months prior,
1,208 total days 6 months post). More impressively, emergency room visits
dropped by more than 50% after program implementation (344 visits 6 months
prior, 158 visits 6 months post). Reducing the need for emergency
interventions and hospitalizations both reduces costs and increases patient
comfort and quality of life. MPCH is a model that shows promise in both being
able to address the wide range of patients needs at the end of life, while also
reducing the cost of care.

spiritual.”2 This definition highlights the multifaceted approach that palliative
care professionals take to address suffering. Palliative care is a growing field of
medicine that addresses the wide ranging needs of seriously ill patients and
their families. Despite an increase of palliative care services, a recent study of
hospital data from developed countries estimated 69-82% people would
benefit from palliative care who do not currently have the option.3
Comparatively, hospice services have been more institutionally established.
Hospice services primarily funded by Medicare part A provide care for patients
who are terminally ill with a life expectancy of six months or less. People
admitted to hospice care are no longer seeking curative treatment. Hospice is
provided at home or in care facilities and offers round the clock care for
patients.4 Research has found that patients who use hospice live an average of
29 days longer, have lower health care related costs and report higher
satisfaction with their care.5,6,7 Similar findings have been also been
demonstrated among patients who use palliative care compared to those who
do not.8 Palliative care services, however are available to anyone seeking
symptom management and improved quality of life and are not restricted to
people in the last six months of life like hospice services are.

Choosing a Mobile Model

Exploring the Research and Field of Practice

The rigidity and structure of traditional hospice care is a poor fit with the needs
and circumstances of homeless individuals. In addition, mainstream hospice
providers are not trained and prepared to effectively identify and serve
individuals without safe and reliable shelter. Examples of hospice rules and
structures that can be incompatible with the needs of homeless individuals
include:

Defining Palliative Care and Hospice



Life expectancy of 6 months or less



Care provided in a facility or home setting



Facilities that provide hospice services do not participate in harmreduction practices



Hospice providers who have not been trained in cultural competency for
homeless individuals.

Palliative care prioritizes quality of life and symptom management for patients.
The World Health Organization (WHO) defines palliative care as “an approach
that improves the quality of life of patients and their families facing the
problem associated with life-threatening illness, through the prevention and
relief of suffering by means of early identification and impeccable assessment
and treatment of pain and other problems, physical, psychosocial and

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

3

 One study finds that “many people never make it into our health care
institutions at all, despite having very real needs. Barriers of culture,
economics, and other circumstances create gaps between a growing number of
citizens and the health care industry.”9 Over the last 20 years, mobile health
care and street outreach programs have developed as one solution for
providing care to those who can’t or won’t access more traditional models of
care.10 HMC and others have identified a need for palliative health care services
for homeless individuals who are unable or unwilling to utilize traditional
palliative care or hospice services. This has led to the development of a mobile
program that can flexibly go where individuals live including on the streets, in
shelters, supportive housing or elsewhere.

Care Needs for Homeless Individuals
Poor mental and physical health can be both a cause and effect of
homelessness. Individuals experiencing homelessness are at greater risk of
many acute and chronic illnesses as well as physical disabilities, substance
abuse and mental health problems.11 Homeless individuals have shortened life
expectancies, between 42-54 years, and often experience health problems
similar to traditional geriatric patients.12 “Homeless persons aged 50-65
frequently fall between the cracks of governmental safety nets: while not
technically old enough to qualify for Medicare, their physical health, assaulted
by poor nutrition and severe living conditions, may resemble that of a 70-yearold.”13
Data from the Seattle-based MPCH program indicates that they are serving
homeless individuals with health conditions that are representative of aging
homeless population at large. Some initial data from 100 of the Seattle-based
program participants indicates that 57% have one or more mental health
diagnosis, 80% have a chemical dependency diagnosis and the average total
number diagnoses is 6. Program participants have a wide range of complex
medical and behavioral health needs as well as requiring pain management
support. HMC’s MPCH program and others like it, serving a particularly
vulnerable population with complicated health problems, can provide
individuals with the medical support they need to have agency in their end of
life planning and care.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

4

 Range of End of Life Care Services for Homeless Individuals
There are very few palliative care services for homeless individuals world-wide.
Beyond formal palliative care programs, end of life care for homeless
individuals is supported in a variety of different programs and settings. There
are mobile care programs that serve homeless individuals and end of life care
programs through hospice and medical respite, however there are few that
specifically combine a mobile model and palliative care services. End of life
services for homeless individuals are provided in many ways, some of the
models discussed in this report include: aging in place, supportive housing,
hospice, and MPCH Programs.

Aging in place services are provided in many shelter and low-income senior
housing programs both in the U.S. and internationally. UK-based St. Mungo’s
is an example of an aging in place model. St. Mungo’s, in partnership with
Marie Curie Cancer Care, developed a resource guide for supporting homeless
individuals in a shelter system at the end of their life. The guide provides a
framework for addressing access barriers palliative care for homeless
individuals, and gives shelter staff information on how to care for and have end
of life conversations with people who are dying in the shelter. Their guide is
designed to give shelter staff and homeless individuals support in end of life
decision making and connections to health care providers and other resources
that help keep people out of hospitals and in shelters while honoring end of life
wishes.14

Supportive housing is another setting for palliative care. Supportive
housing models are used internationally to provide permanent and temporary
low-income housing for homeless individuals experiencing multiple comorbidities. These housing models typically combine access to housing with
other vital services including mental health, clinic services, social services and
case-management, and more. This model aims to provide stable housing for
some of the most ill and vulnerable houseless individuals while also removing
barriers to access care and social services. This unique combination of stable
housing and services makes them a useful setting for palliative and end of life
care. Research addressing supportive housing models have found that they
improve homeless individuals’ health and well-being while also reducing the

Palliative Care for Homeless Individuals: Emerging Models

cost of care.15 There are several organizations in the Seattle area providing
supportive housing services to homeless individuals. Many of the referrals to
the Seattle-based MPCH program come from healthcare providers at local
supportive housing sites.16

Medical respite programs can be a source of palliative care post hospital
discharge for homeless individuals. The medical respite model offers homeless
individuals an alternative to going back to living on the streets after being
discharged from the hospital. Respite facilities provide homeless individuals
with medical care, pain management and social services while recovering after
a hospitalization. Data released by the National Health Care for the Homeless
Council identified 73 respite programs in the U.S. that provide care to homeless
individuals with an average stay of 42 days.17 Research indicates that homeless
individuals who are placed in respite programs have 50% fewer hospital
readmissions within 90 days of being released from the hospital.18 Respite
provides a necessary and cost effective transition from the hospital for
homeless individuals. Seattle’s medical respite program, the Edward Thomas
House, is operated by HMC.

Hospice services for homeless individuals are often located in shelters or
freestanding facilities. These services, like more traditional hospice models, are
specifically for people in the last six months of their life and who are no longer
seeking curative treatment. This model can be difficult to accept for some
homeless individuals who are ill and preparing for death much earlier in their
life than many. With the support of Medicaid and Medicare reimbursement,
hospice services for homeless individuals exist in most major U.S. cities and
furthered with the recent expansion of Medicaid in many states. Regardless of
the possibility for reimbursement, the current capacity and structure of these
services falls substantially short of meeting the growing need. MPHC programs
expand upon and offer an alternative to the existing structure of hospice
services for homeless individuals who require more culturally appropriate and
flexibly-delivered palliative care

MAYA BERKOWITZ

5

 Examples of End of Life Care Models Internationally
One example of hospice care for homeless individuals from Canada is the
Ottawa Mission. The Mission in partnership with Ottawa Inner City Health,
provides a 14 bed acute hospice program for homeless individuals as well along
with a new seven bed chronic palliative care unit. The acute unit provides a
higher level of medical care and patient management, while the chronic
palliative care unit provides less intensive medical intervention with lower
staffing levels. Both units provide as needed 24-hour care for patients using
nurses and volunteers.19
There are very few MPCH programs in operation world-wide, all of them
outside of the United States. A total of four mobile palliative care programs
were identified as currently operating or in development through internet
research and interviews with experts in the field of homelessness and end of
life care. Only the Toronto-based Palliative Education and Care for the
Homeless (PEACH) program was identified as comparable to the current
Seattle-based MPCH program. The PEACH program provides “quality early,
integrated and safe palliative care for the homeless and vulnerably housed
populations with a focus on prioritizing dignity for all.”20 Individuals in Calgary,
CA and Queensland, AU are in the process of establishing MPCH programs.
Programs and institutions in the developed world are trying a range of
homeless programs and pilot projects, however few specifically focus on the
need for palliative care.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

6

 Common MPCH Program Components

Patient Referral Process

This section identifies the most commonly observed structures and practices of
existing MPCH programs and other end of life programs for homeless
individuals based on research and interviews with providers. In the process of
interviewing providers, similar experiences, challenges and learnings emerged.
Many of these programs are currently focused on building sustainable
resources and capacity. Dr. Dosani, who runs Toronto’s PEACH program
mentioned above is clear about where to start when developing a new MPCH
program.

Most of the programs researched for this report used a wide referral network
including housing/shelter, social service and health care providers. Many
referrals fall outside of the traditional medical referral path, encompassing the
interdisciplinary care model referenced by Dr. Dosani.

“At the beginning it is about referrals, it’s about connecting to
home care, it’s about coverage, it’s about creating
interdisciplinary teams that wrap-around patients and having
right cases.”
– Naheed Dosani21
These program components identified by Dosani will be expanded upon below
to add depth and clarity to their importance.

“We accept referrals from a grassroots network, we get
referrals from social workers, nurses, doctors, hospitals. Our
number one referral source is from social workers who work in
shelters. From there we meet with the patient, talk to their
primary healthcare provider, have them sign a waiver to release
their medical records and from there we see if they are an
appropriate case.”
– Naheed Dosani21
The Seattle-based program receives most of its referrals from nurses based at
supportive housing sites. These nurses work closely with residents to identify
individuals’ health needs, goals and options for care.
“It took us awhile to centralize our referral system and get the
word out to partner organizations and others serving this
population but as more people have worked with use and are
more familiar with our services we are seeing an increase in
referrals.”
– Colette Conlisk16
The PEACH program identified the centrality of developing relationships with
hospital discharge planners who are responsible for helping patients identify
services to meet their needs after hospitalization. This helps address the need
for wrap-around care described by Dr. Dosani.
“We connect with discharge planners at high yield sites. We
have good working relationships with all of those discharge
planners…it is an important place to get referrals.”
– Naheed Dosani21
Knowing where older homeless individuals seek care and connecting with
health care and social service providers in those places is key to being able to

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

7

 successfully identify appropriate patients who would benefit from care they
can receive in these programs.

Program Staffing Models
The MPCH programs studied have broadly similar staffing models. The PEACH
program in Toronto has .2 FTE physician and a .6 FTE nurse. The Seattle-based
team has a nurse practitioner funded at .5FTE and a nurse funded at .5FTE.
“The PEACH team is not the nurse, me and the patient. These
patients come in with care teams that already exist often or a
case worker. So the PEACH team for each patient is actually
different. Each patient has a unique team wrapped around them
that are funded from different sources. The key players are our
team, publically funded home care, and others.”
– Naheed Dosani21
Current MPCH program providers struggle to piece together wrap-around
services and care teams for most patients. Providers uniformly feel that their
staffing models do not adequately allow for the extensive care coordination
necessary to meet patient needs, as well as fulfill the administrative tasks
related to referral, intake and reporting.
“while we also always thought she [program nurse] would do
frontline work, in reality she does 5% frontline work and 95%
coordination. We would like to get a social worker, but we are a
physician funded organization. We were really lucky we were
even able to get a nurse. What we do instead is work with the
social workers who are available through other programs
instead. We are used to working barebones.”
– Naheed Dosani21
One staffing challenge is that MPCH programs are run through hospital/
medical-based organizations that are based in the provider/ nurse model.
However, these emerging specialty programs may benefit from
interdisciplinary support from social workers or others with non-medical
training.
“So much of my job is coordination and social work. I’m a nurse
by training. There are days when I feel ill-equipped to be a grief
counselor. I have nursing skills, and I occasionally get to use
them, but sometimes I wonder if there is another set of skills
that are more necessary to meet the needs of our patients and

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

8

 for the success of the program.”
– Colette Conlisk16
Most programs studied acknowledged the need for additional administrative
resources as well as more comprehensive mental health care, social work and
grief counseling to provide optimal support. Lack of these support and training
resources are challenges to the development of viable programs.

Program Size
Program capacity for mobile-specific programs varies to a degree and is
generally larger than similar palliative programs that are site-based with a
limited and fixed number of beds. The PEACH program in Toronto has cared
for over 65 homeless individuals since the program began in July of 2014.
“At any time we have between 18-25 patients on our caseload.
We see about 40-45 cases a year. Our goal was to do about one
new intake visit a week our first year. We did 47, which we felt
pretty good about.”
– Naheed Dosani21
The Seattle-based program has a larger caseload, with about 30-45 patients
receiving care at any given time. Providers in the program see about 18-20
patients a week as well as providing case-management via phone. Most
patients are seen at least every other week unless they need to be seen or
followed-up with more frequently. In contrast, site-based programs are limited
by the number of beds they have. For example, the hospice program at the
Ottawa Mission has 14 beds for hospice-enrolled individuals and recently
added another seven beds for individuals who need more chronic, long-term
care with an extended life-expectancy allowing them to serve 21 patients at a
time, half the capacity of the mobile program in Seattle and at a higher
program cost per patient per day.
“We identified that some people are on the palliative care path,
but that it may be a year or two before they are ready to die.
Having the flexibility of both acute [hospice] beds and chronic
palliative care allows us to be more flexible.”
– Marg Smeaton22

Patient Eligibility Criteria
Programs have a limited capacity, and therefore must be intentional about
who they accept. For any program serving homeless individuals it is important
that the program first define homelessness. Definitions of homelessness vary
considerably and MPCH programs must be clear about their criteria as it relates
to admission. Some programs may define homeless as those without shelter,

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

9

 while other program definitions may also include people living in shelter,
transitional housing, unstably housed, etc.

for some individuals that need palliative care.”
– Marg Smeaton22

Another important patient eligibility criterion is disease status and prognosis.
MPCH Program participants must have a life limiting illness for which they
need care managing their pain and quality of life. The Seattle-based program
accepts individuals into the program with a life expectancy of 6 months or less;
the PEACH program in Toronto does not have a specific life-expectancy
requirement.
Note: Many programs contacted both have life-expectancy requirements and
also make exceptions for patients who have a longer life-expectancy on a caseby-case basis. These exceptions are often made for homeless individuals who
are the least likely to receive other kinds of care and who are struggling to
manage pain on their own with their primary care provider.
Naheed Dosani uses the Mississauga Halton Regional Hospice Palliative Care
Early Identification & Prognostic Indicator Guide to identify whether patients
qualify for the program.
“We have to be cognizant that in our program we have a low
threshold for admission because homelessness cuts the lifespan
by 50%, so we view homelessness as a terminal diagnosis.
Homelessness in and of itself is life limiting and could be a
palliative diagnosis, so when you add another diagnosis it is a
slam-dunk. Especially if their goals are not treatment oriented.
For those seeking quality of life treatment [from the PEACH
program] this document makes a world of difference.”
– Naheed Dosani21
Most programs have limited mental health service capacity, which limits
admission to patients with minor and/or well managed mental health needs,
effectively limiting access to palliative programs for many homeless
individuals.
“A significant challenge we face is access to enough mental
health care. We have a psychiatrist who comes to the hospice
once a month and there is a social worker on staff at the
mission, but it is never enough and it limits our capacity to care

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

10

 End of Life Planning Conversations
Advanced care planning (ACP) is an important element of end of life care for
everyone, not just homeless individuals. ACP allows people to decide who they
want to make decisions for them when they are incapacitated, the kinds of life
saving treatment they would like to receive, what will happen to their body and
possessions after death, and any other decisions about their care or death they
want to express.23 In general individuals who complete ACP are more likely to
receive appropriate care and less likely to receive unnecessary or undesired
care. In one study, among the general population 90% of those who completed
ACP died in their desired location compared to 50% of people who did not
complete ACP. ACP has also been shown to improve patients symptom and
pain management and increase patient comfort and happiness at the end of
life while also reducing healthcare costs.24 John Song, an Internist working at
the University of Minnesota Medical Center conducts research about ACP
specifically with homeless individuals.

decisions for them at the end of their life when they die, and what should
happen to their belongings. It also gives homeless individuals an opportunity to
decide if they want to reach out to people who they have been estranged from.
Song found that it is important to tailor ACP to homeless populations.
“Our new patient intake is done by a nurse and includes end of
life planning. They talk about what kind of funeral they want,
what kind of care they want at the end of their life, if they
want to contact anyone from their biological family and who of
their street family they want with them. Most people who
receive palliative care here sign a do not resuscitate (DNR), but
some people choose to get transferred to the hospital”
– Marg Smeaton22

“We know that advanced care directives improve care in the
general populations. I feel that they are even more important
for homeless individuals. They can often end up in the hospital
without anyone to advocate for them on their behalf, and have
[healthcare] interventions that they don’t want. Also we found
that usually next of kin is not who they want making healthcare
decisions for them. An important part of an advanced directive
for homeless individuals is the power of attorney. We found that
these conversations are most productive when they happen with
a trusted shelter worker or case-manager. Often healthcare
providers do not have the skills, desire or time to guide these
conversations.”
– John Song25
In a randomized trial Song found that when offered, 44% of homeless
individuals participated in creating an advanced directive. 59% of individuals
who received guidance from a counselor completed an advanced directive
compared to 30% who were given materials to complete on their own.26 Some
important components to include in end of life planning conversations
specifically for homeless individuals include what they want to happen to their
body when they die, who they want to be notified, who they want making

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

11

 Geographic and Site-based Limitations
All programs studied have some geographical referral limits based on funding
requirements, population demographics, travel limitations or a combination of
these factors. MPCH programs identified for this project all use a home-visit
model seeing individuals in transitional housing, a shelter or on the street
instead of using a van/mobile setting for patient visits.
“Ontario is broken up into 14 different districts known as Local
Health Integration Networks (LHIN). Inter City Health Associates
where I work is only funded to support patients who are
homeless from Toronto Central LHIN, therefore I am only funded
to see people who are homeless and dying from Toronto Central
LHIN. There are actually barriers for patients from different
LHINs. This is especially challenging with a transient patient
population.”
– Naheed Dosani21

“A huge part of my day can be eaten up just trying to get from
patient to patient. This is the only way to see many of our
patients, but it means we do need to think about where
someone lives when we take them on. Also if we had more time
and space in clinic it may allow us to see more patients without
having to go to them as often.”
– Colette Conlisk16
However, there are also benefits to the mobile model in terms of access for the
intended population. Naheed Dosani from the PEACH program in Toronto
disagrees with Conlisk’s proposal to extend clinic hours. He doesn’t endorse or
offer clinic-based office hours in his program.
“There is no point in doing that [hosting clinic hours]: we go to
where people are. The work we are doing requires you to think
differently. Having a physical space in my opinion is absolutely
pointless.”
– Naheed Dosani21

Urban downtown cores are “home” to the majority of homeless individuals or
at least where they spend much of their day. It is also where many social
services are consolidated. This centralization may make it easier to have a
specific geographical boundary for a program.
“Ottawa Inner City Health is an umbrella organization that has
nine different programs housed in different shelters mostly
within the downtown core of the city. Most of our programs are
within a 12 block radius, serving the homeless population in the
downtown. It is important to focus the work where your
population is, that is why providing care in shelter settings has
been successful for us. We go to them, not the other way
around.”
– Marg Smeaton22
The Seattle-based program sees homeless individuals in the downtown Seattle
core, specifically in Pioneer Square as well as a few patients who live in Ballard,
a neighborhood across town. They offer a very limited number of weekly clinic
drop-in hours but provide most of their visits outside of the traditional clinic
setting. This can have pros and cons, since the mobile model requires staff to
spend a significant amount of time traveling around the city.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

12

 Smaller Programs Benefit from Larger Infrastructure
MPCH programs are bolstered by larger healthcare organizations and
institutional partnerships. This relationship provides necessary infrastructure
and support that would otherwise be unavailable due to program size and
financial limitations. Supports provided include:
1. Access to electronic medical records and other shared information
systems.
“Ottawa Inner City Health is an umbrella for nine different bedbased programs and clinics in downtown Ottawa. We share one
health record system and providers rotate in and out of the
different programs. Every Thursday the providers get together
to discuss patients across the sites. This allows all the providers
to understand what is going on with all of the patients, keeping
our system flexible. If a patient needs to move from a program
with less care to one with more we can make that transition
pretty seamless for individuals.”
– Marg Smeaton22
2. Back-up healthcare providers, specifically those who can assist with
providing prescriptions and stepping in for providers while sick or on
vacation.

get a prescription. It is very hard to make a system like this
sustainable. That’s why he [ARNP] always has his phone with
him, even on vacation.
– Colette Conlisk16
3. Access to larger care networks and organizational partners that can help
provide wrap-around services, refer patients, and strategic partnerships.
“Find a place that refers at high volumes. We have two sites
that refer quite frequently, to the point were I visit every week.
One is the largest men’s shelter in Toronto and they have a
managed alcohol program within it. We get 1-2 referrals a week
from that program because people are suffering from end-stage
liver disease. You want early referrals from program partners. It
is important for people to know what you do and that you are
good at it.”
– Naheed Dosani21

“We found that just having medical care wasn’t enough. Social
service providers are your partner when it comes to talking to
homeless individuals about end of life decisions. It is important
that they are working with someone they trust”
– John Song25

“Inner City Health Associates (ICHA) is a group of 60 doctors
doing a variety of work. One of the doctors has a kind of
palliative background and covers for me when I am away. If
push comes to shove the primary care lead in our organization
can prescribe or even our medical director. Coverage for me is
easy, but coverage for our nurse is harder. We have no funded
coverage for her [the nurse].”
– Naheed Dosani21
The Seattle-based program lacks the kind of institutional support and access to
providers who are trained and comfortable filling in for providers in this
program when necessary.
“There is very little institutional support for this program. If
either of us is out it means that referrals aren’t getting
processed. If the ARNP isn’t available, it means that no one can

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

13

 Training The Next Generation

Access to Low-Intervention End of Life Beds

Palliative care services in hospitals and out-patient settings are growing rapidly
in the U.S. and other developed nations. However, in the U.S. palliative care
has only been a board-certified medical specialty since 2008 resulting in fewer
than needed trained palliative care providers.27 There are even fewer providers
who have the interest and competencies to work with homeless individuals.
The success of small palliative care programs serving homeless individuals
requires that the next generation of medical providers receive specialized
training. Today’s medical residents, nursing externs and social work students
can gain end of life care skills, and the culturally competency needed to work
with homeless individuals and bolster the modest staffing levels of these small
programs.

In-patient end of life care are among the most intensive and expensive medical
services, accounting for one out of every four Medicaid dollars spent.29
Individuals hospitalized at the end of their life are also more likely to receive
unwanted care interventions. Compounding the unnecessary costs associated
with inpatient end of life care, most patients do not choose the hospital as a
place to die.30 However, when homeless individuals are in their last days of life
they can often end up in the hospital because they don’t have an alternative.
MPCH programs studied identified a growing need for low-cost, lowintervention beds for homeless individuals, who may dislike the structure and
regulation of a hospital or hospice setting. In some communities, medical
respite, skilled nursing facilities (SNF) and supportive housing are used as
alternatives to hospital-based palliative care for homeless individuals.

“I always have a resident on. We are booked for an elective for
something like two years. So there is always someone with an
M.D. who can liaise and do basic care. At least a senior resident.
These residents are critical to the on going success of this
program.”
– Naheed Dosani21

“In Oakland, CA at Highland Hospital they are managing to get
people into SNF beds and then getting hospice layered on top of
it; like an outside hospice. It is mostly people who are dying at
Highland who they are managing to get out into these other
situations which is a creative solution. But it feels very catch as
catch can. There isn’t a system.”
– Margot Kushel31
It is often hard to get homeless individuals into a SNF and once they are there
staff are not formally trained and/or equipped to provide palliative care or
specifically meet the needs of homeless individuals. There is a need to develop
a model of care for homeless individuals that provides a safe place to receive
culturally competent care in their last days of life.
“This is where being a mainstream palliative care physician is
helpful. We have developed relationships with mainstream
palliative care facilities. Do we face resistance from these
facilities? All the time. There are also opportunities to get
people into hospice, long-term care, etc. We advocate for
admission. We have places and we consider them partners.
Some of these partnerships have been more successful than
others. We never discharge a patient, even when they get
admitted to a facility, we never sign-off.”
– Naheed Dosani21

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

14

 The Seattle-based MPCH program is developing a partnership with a
supportive housing provider who is opening a facility near HMC in the next few
years. This partnership will provide four beds for MPCH program patients in the
last days or weeks of their life who do not have safe shelter. This partnership
will both provide a place for people to die on their own terms, outside of a
hospital at lower cost.
“We are excited about this partnership. As it stands now many
of our patients end up dying in the hospital because it is really
hard to find beds for them. Most facilities are either illequipped to address an individual’s medical needs, or they are
not use to working with homeless populations. Having a safe
place for people to die, while getting the care they need in an
appropriate setting is really important and valuable”
– Colette Conlisk16

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

15

 Program Funding
The Seattle-based program is funded for 2014-2016 with HRSA funds locally
administered by PHSKC. The budget for the program is approximately
$100,000 a year. The PEACH program in Toronto has a similarly sized budget.
Up to this point MPCH programs have successfully operated with small
budgets using a traditional medical model including a provider and a registered
nurse (RN). In contrast, traditional palliative care programs include social
workers, chaplains and other support staff to meet a holistic range of patient
needs. As MPCH programs are currently structured and funded program
components such as grief counseling, program coordination, specialty
referrals, and social work support consume a significant amount of MPCH
nursing staff hours.
MPCH programs may benefit from including a social worker in their care
model, especially because on average homeless individuals have more social
work needs than the general palliative care patient population. Due to current
budget constraints these programs are also not able to provide the level of
mental health care, grief counseling, social services, and case-management
that would benefit many. As MCPH models develop it is critical to consider the
social work and mental health resources that many program participants will
need, as well as program administrative demands when building and working
to expand MPCH program budgets.
Many of the programs outside of the U.S. are funded primarily by their national
healthcare systems. In Ottawa the mission’s palliative care providers, nurses
and other healthcare staff are paid for by the government. All equipment, food
and other costs are paid for through private donations to the mission.
“The provincial government reimburses the hospice $42 a day
per bed per hospice bed that is filled. The cost per day for a
traditional ward bed in Canada is something like $3,000. The
government knows it is better to help support our shelter
hospice program where each bed costs about $220 a night. The
work we do doesn’t just save money, it also improves people’s
quality of life. It works better for everyone.”
– Marg Smeaton22

Palliative Care for Homeless Individuals: Emerging Models

In the U.S. programs are funded through public grants, individual donations
and reimbursement from Medicaid and Medicare. The Seattle-based MPCH is
primarily funded through public grants with minimal reimbursement for
services. Federal dollars are available to reimburse for some palliative care
services and housing costs. Supportive housing services are not yet recognized
and paid for by health systems, making it hard to develop and sustain these
programs.
“Where we have really fallen down is out-patient palliative
care. I think in some ways this has to do with our completely
siloed payment structures, our lack of coordination of EHRs, the
in-patient out-patient divide, but at the end of the day I don’t
know what needs to change to address these [financial]
challenges.”
– Margot Kushel31
Margot Kushel and other U.S. based providers and researchers interviewed for
this project acknowledged that in the US, collaboration between community
partners offers some of the most optimistic opportunities for palliative care
programming for homeless individuals including:


Partnerships of medical, housing and social service organizations working
together to fund elements of a collective team working together.



Negotiating with Medicaid managed care plans to develop coverage
options for early palliative care and housing outside of the hospital system
for homeless individuals on Medicaid.

As palliative care providers and homelessness advocates continue to identify
the breadth and depth of care necessary, the need for services continue to
grow among aging homeless individuals. Emerging models such as MPCH can
serve as a platform for further program development and an opportunity to
build an interconnected network of programs and services to address the
palliative care needs of homeless individuals.

MAYA BERKOWITZ

16

 Conclusion
I had the opportunity to speak to many people in the field of palliative care,
homeless advocacy, homeless health care and researchers for this project. I
walked away from these conversations knowing more about what is happening
in the field of end of life care for homeless individuals and personally excited
about the work that people are doing across the country and internationally.
Experiences of today’s MPCH providers point us toward the needs of the future
and can be summarized by the following recommendations:


Train culturally competent medical and mental health providers to address
the palliative care needs of homeless individuals.



Develop flexible program models, such as MPCH that address the specific
needs and limitations of homeless individuals.



Create sustainable funding models that support emerging program
models.



Integrate MPCH programs with larger health care systems, providing
structure, technical and staffing support and a referral base.



Collaborate with palliative care providers and professionals working in
homelessness services to build effective, holistic, and flexible care models
that can be tailored to different communities.



Develop unified definitions and criteria, systematic delivery systems, and
opportunities to shared resources to best serve homeless individuals.

As the aging population expands and homelessness continues to grow
throughout the world, there will be an increased need for palliative care
services for homeless individuals. Current providers have documented the ways
in which standard palliative care models do not meet the needs of homeless
individuals. Homeless individuals need flexible program admission criteria,
harm-reduction prescribing practices, and access to care outside of clinic and
hospital settings.
“This is going to be a continued problem as this population
continues to age. I think that we need to start talking about it
and exploring funding options... we need to come together and

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

17

 start imagining what a comprehensive, large-scale, integrated
solution looks like.”
– Margot Kushel31
The MPCH programs and other innovative services are designing solutions to
address the need for palliative care for homeless individuals. While current
programs are a good first step in identifying and caring for individuals who are
better served outside the traditional healthcare system, they could be more
effective with added institutional support allowing them to better address the
complexity and scope of palliative care needs.
An important next step to expanding and sharing this emerging model of care
includes palliative care providers and homeless advocates continuing
conversations about the palliative care needs of homeless individuals. What
emerges from those collaborations and early pilot projects will serve to
leverage financial support for the palliative care needs of homeless individuals.

A special thanks to all of those who spoke to me about this project, which
includes:
Angela Plunkett
Anne Kinderman
Ashley Mollison
Ed Dwyer O’Connor
Heather Harris
John Song
Marg Smeaton
Margot Kushel
Naheed Dosani
Niamh Brophy
Simon Colgan
Tony Owen Boxwell

In the last 50 years there have been great strides in understanding and
addressing the economic, social and cultural causes and implications of
homelessness. This work has been important to understanding the impact of
homelessness in differing populations including veterans, teens, HIV+
individuals, substance abuse issues and people with mental illness. The aim of
this project is to build on this legacy of addressing the needs of homeless
individuals by mobilizing programs and organizations nationally and
internationally to collaborate to build a strategic response to the need for
palliative care.

Contributions
This project was possible thanks to the support and investment from
Harborview Medical Center, specifically those working in the MPCH program
including Ed Dwyer-O’Connor and Colette Conlisk. A special thank you to the
University of Washington Community-Oriented Public Health Practice and
especially my faculty advisor, Chris Hurley. This project benefitted from the
support and investment of many, both in the fields of palliative care and
homelessness.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

18

 Sources and Interviews

content/uploads/2011/04/PrematureMortalityFinal.pdf. Accessed
November 28, 2015.

1.

Cagle JG. Weathering the Storm: Palliative Care and Elderly Homeless
Persons. J. Hous. Elderly 2009;23(1-2):29-46.
doi:10.1080/02763890802664588.

12.

Hearth Inc. Who are the Elderly Homeless? Facts 2010. Available at:
http://www.hearth-home.org/homeless-elders/. Accessed November 28,
2015.

2.

World Health Organization. Definition of Palliative Care. Available at:
http://www.who.int/cancer/palliative/definition/en/. Accessed October 25,
2015.

13.

3.

Murtagh FEM, Bausewein C, Verne J, Groeneveld EI, Kaloki YE, Higginson
IJ. How many people need palliative care? A study developing and
comparing methods for population-based estimates. Palliat. Med.
2014;28(1):49-58. doi:10.1177/0269216313489367.

National Coalition for the Homeless. Homelessness Among Elderly Persons.
Washington D.C.; 2009. Available at:
http://www.nationalhomeless.org/factsheets/Elderly.pdf. Accessed
November 28, 2015.

14.

Kennedy P, Sarafi C, Greenish W. Homelessness and End of Life Care.
London, UK; 2013. Available at:
https://www.mariecurie.org.uk/globalassets/media/documents/commissio
ning-our-services/current-partnerships/homeless_report.pdf. Accessed
April 2, 2016.

15.

Abt Associates, Advocates for Human Potential Inc. Building Your ProgramPermanent Supportive Housing.; 2010. Available at:
http://store.samhsa.gov/shin/content/SMA10-4510/SMA10-4510-06BuildingYourProgram-PSH.pdf. Accessed April 2, 2016.

16.

Interview with Colette Conlisk, RN. 2015.

17.

National Health Care for the Homeless Council. 2015 Medical Respite
Program Directory: Descriptions of Medical Respite Programs in the United
States.; 2015. Available at: https://www.nhchc.org/wpcontent/uploads/2011/10/2015-medical-respite-program-directory.pdf.
Accessed April 3, 2016.

4.

Davis MP, Gutgsell T, Gamier P. What is the difference between palliative
care and hospice care? Cleve. Clin. J. Med. 2015;82(9):569-71.
doi:10.3949/ccjm.82a.14145.

5.

Connor SR, Pyenson B, Fitch K, Spence C, Iwasaki K. Comparing Hospice
and Nonhospice Patient Survival Among Patients Who Die Within a ThreeYear Window. J. Pain Symptom Manage. 2007;33(3):238-246.

6.

Pyenson B, Connor S, Fitch K, Kinzbrunner B. Medicare cost in matched
hospice and non-hospice cohorts. J. Pain Symptom Manage.
2004;28(3):200-210.

7.

Miller S, Rynd T. Quality of Life at the End of Life: The Public Health
Perspective. Public Heal. Aging 2005;Summer:41-47.

8.

Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients
with metastatic non-small-cell lung cancer. N. Engl. J. Med.
2010;363(8):733-42. doi:10.1056/NEJMoa1000678.

18.

Kertesz SG, Posner MA, O’Connell JJ, et al. Post-hospital medical respite
care and hospital readmission of homeless persons. J. Prev. Interv.
Community 2009;37(2):129-42. doi:10.1080/10852350902735734.

9.

Withers J. Street Medicine: An Example of Reality-based Health Care. J.
Health Care Poor Underserved 22(1):1-4. Available at:
https://muse.jhu.edu/article/414326. Accessed May 7, 2016.

19.

Podymow T, Turnbull J, Coyle D. Shelter-based palliative care for the
homeless terminally ill. Palliat. Med. 2006;20(2):81-86.
doi:10.1191/0269216306pm1103oa.

10.

Post P. Mobile Health Care for Homeless People: Using Vehicles to Extend
Care.; 2007. Available at: https://www.nhchc.org/wpcontent/uploads/2012/02/mobilehealth.pdf. Accessed May 7, 2016.

20.

Associates ICH. PEACH. Programs 2014. Available at: http://www.ichatoronto.ca/programs/peach. Accessed November 29, 2015.

11.

O’Connell J. Premature Mortality in Homeless Populations: A Review of the
Literature. Nashville; 2005. Available at:
http://santabarbarastreetmedicine.org/wordpress/wp-

21.

Interview with Naheed Dosani, M.D. 2016.

22.

Interview with Marg Smeaton, Manager of Health Services. 2016.

23.

HCH Clinicians. HCH Clinicians Can Help Homeless People Die with Dignity.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

19

 Heal. Hands 2004;8(1). Available at: http://www.nhchc.org/wpcontent/uploads/2011/09/endoflife.pdf. Accessed May 7, 2016.
24.

Scott IA, Mitchell GK, Reymond EJ, Daly MP. Difficult but necessary
conversations--the case for advance care planning. Med. J. Aust.
2013;199(10):662-6. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/24237095. Accessed March 18, 2016.

25.

Song J, Wall MM, Ratner ER, Bartels DM, Ulvestad N, Gelberg L. Engaging
homeless persons in end of life preparations. J. Gen. Intern. Med.
2008;23(12):2031-6; quiz 2037-45. doi:10.1007/s11606-008-0771-1.

26.

Interview with John Song, M.D. 2016.

27.

Certification for Hospice and Palliative Medicine Specialists. Am. Acad.
Hosp. Palliat. Med. 2015. Available at:
http://aahpm.org/career/certification. Accessed April 11, 2016.

28.

Punke H. Workforce Shortage Threatens Future of Palliative Care. Becker’s
Hospital Review. http://www.beckershospitalreview.com/hospitalphysician-relationships/workforce-shortage-threatens-future-of-palliativecare.html. Published November 5, 2013. Accessed April 11, 2016.

29.

Wang P. Cutting the High Cost of End-of-Life Care. Money 2012. Available
at: http://time.com/money/2793643/cutting-the-high-cost-of-end-of-lifecare/. Accessed May 8, 2016.

30.

The Dartmouth Atlas of Health Care. End-of-Life Care. Available at:
http://www.dartmouthatlas.org/keyissues/issue.aspx?con=2944. Accessed
May 8, 2016.

31.

Interview with Margot Kushel, M.D. 2016.

Palliative Care for Homeless Individuals: Emerging Models

MAYA BERKOWITZ

20