Content uploaded by Richard Lakeman
Author content
All content in this area was uploaded by Richard Lakeman on Apr 01, 2014
Content may be subject to copyright.
HOW HOMELESS SECTOR WORKERS
DEAL WITH THE DEATH OF SERVICE USERS:
A GROUNDED THEORY STUDY
RICHARD LAKEMAN
School of Nursing, Dublin City University, Dublin, Ireland and School of
Health and Human Sciences, Southern Cross University, Lismore, New South
Wales, Australia
Homeless sector workers often encounter the deaths of service users. A modified
grounded theory methodology project was used to explore how workers make sense
of, respond to, and cope with sudden death. In-depth interviews were undertaken
with 16 paid homeless sector workers who had experienced the death of someone
with whom they worked. Transcripts of interviews and field notes were analyzed
using the constant comparative method and a theory that described the positive
framing of death emerged. Dealing with death and trauma is not something that
most workers expect when they begin work but exposure to the death of a service user
heightens expectations that other service users may be harmed. How workers cope or
positively frame death depends on several interlinked processes. These include how
the death is encountered; how the worker and others mark the death; and the extent
that the vulnerability of self, peers, and service users is recognized and responded to.
Successfully framing death enables the worker to continue working in the sector
whilst maintaining enthusiasm for the work and compassion for service users.
Homeless populations around the world have been found to
have much higher rates of both morbidity and mortality relative
to those who are securely housed. Homeless people tend to have
high rates of alcohol and drug dependence and associated
Received 9 December 2009; accepted 21 September 2010.
Richard Lakeman is now a Senior Lecturer in the School of Health and Human
Sciences at Southern Cross University in Lismore, New South Wales, Australia.
This research project was part funded by a 410,000 grant from the Electric Supply
Board (ESB) Electric Aid Ireland fund and a 44,000 grant from The Homeless Agency
(Ireland). It represents phase one of a project to explore the needs of homeless sector
workers in relation to responding to sudden death.
Address correspondence to Richard Lakeman, School of Health and Human Sciences,
Southern Cross University, P.O. Box 157, Lismore, NSW 2480, Australia. E-mail: richard.
lakeman@scu.edu.au
Death Studies, 35: 925–948, 2011
Copyright #Taylor & Francis Group, LLC
ISSN: 0748-1187 print=1091-7683 online
DOI: 10.1080/07481187.2011.553328
925
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
problems, communicable diseases, psychiatric disorders, and
exposure to violence and trauma (Kushel, Evans, Perry, Robertson,
& Moss, 2003; Schanzer, Dominguez, Shrout, & Caton, 2007). Rates
of suicidal thoughts and suicide attempts have also been found to be
exceptionally high, with some surveys finding rates 10 times higher
than the general population (Desai, Liu-Mares, Dausey, &
Rosenheck, 2003; Fitzpatrick, Irwin, Lagory, & Ritchey, 2007;
Haw, Hawton, & Casey, 2006). Research has consistently found that
homeless or marginally housed individuals have a shorter life
expectancy and are more likely to die of homicide, suicide, trauma,
AIDS-related conditions, drug overdose, and other alcohol- and
drug-related problems relative to the domiciled population (Cheung
& Hwang, 2004; Haw et al., 2006; Hwang, Orav, O’Connell, Lebow,
& Brennan, 1997; Morrison, 2009; Nordentoft & Wandall-Holm,
2003). People who work in the homeless sector are likely to encoun-
ter individuals at extreme risk of death and to be regularly exposed
to the deaths of service users during the course of their work.
Exposure to death in the workplace has been found to be
highly traumatic for workers (Kinder & Cooper, 2009) and people
who help survivors may themselves be traumatized in the process
(Sabin-Farrell & Turpin, 2003). Some groups, such as emergency
personnel, are regularly exposed to mutilation, death, human suf-
fering, and pain. The impact of exposure to trauma and sudden
death on health professionals and emergency workers has been
explored particularly in relation to ‘‘emotional labor,’’ that is, deal-
ing with the strong emotions of self and others (Stayt, 2009), stress
and coping (Alexander & Klein, 2009), and posttraumatic stress
responses (Jonsson & Segesten, 2004). The experience of health
professionals who care for the seriously ill, dying, and bereaved
has been the subject of extensive study (see Papadatou, 2000),
but the experiences of welfare workers and non-health profes-
sionals have received little attention.
Homeless sector workers may respond to trauma and sudden
death of service users differently to other occupational groups
because of the nature of their work and the people whom they
serve. Workers are less likely to share clear professional identities
and role demarcations as people who work in hospitals or emerg-
ency services. They provide services to people who are often mar-
ginalized and excluded from mainstream society in situations that
are fraught with risk. Their relationships with service users are
926 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
likely to be different (sometimes more intense and sometimes
fleeting), and for some service users the homeless sector worker
may be the most significant person in their lives as relationships
with family, friends, or professionals may be tenuous or strained.
Some workers, such as those who work in street outreach or in
shelters in which drug or alcohol intoxication of service users is tol-
erated, are exposed to exceptionally high levels of risk and uncer-
tainty. How front-line homeless sector workers experience and
deal with the deaths of people whom they are charged with helping
is the topic of exploration.
Local Context
There are at least 250 agencies delivering services to homeless
people in Dublin. Some, such as emergency shelters, work exclus-
ively with people deemed to be homeless whereas others, such as
alcohol and drug treatment agencies, serve a broader population.
Services may be considered on a continuum (from low to high
threshold). Low-threshold services are those that have few if any con-
ditions associated with access and tend to be available to whoever
needs them. These include street outreach services, emergency
shelters, needle exchange programs, sexual health services, soup
kitchens, and homeless drop-in services. High-threshold services tend
to serve fewer people and have conditions relating to access (e.g.,
requiring a referral from another agency or requiring adherence to
rules such as abstaining from drugs or alcohol). These include resi-
dential alcohol and drug treatment facilities and longer term hous-
ing. Services may also involve varying degrees of intensity of
involvement with homeless people. Intensive forms of case man-
agement and residential drug treatment tend to involve more
intensive relationships with service users than drop-in services or
those that provide a discrete service such as methadone dispensing.
However, the strength of attachments between people in these
various services is not solely a function of the threshold or intensity
of the service provided (as shall be discussed later).
The Homeless Agency attempts to count the number of
people sleeping in public spaces in Dublin City on a regular basis.
Between 2007 and 2009 the minimum number counted (which is
likely to considerably underestimate the actual number) has varied
between 98 and 115 people on a given night (Homeless Agency,
Homeless Workers 927
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
2009). Many of these individuals are likely to episodically use one
of the 15 emergency accommodation facilities in the city. Most of
the 2,366 adults reported as homeless in Dublin in 2008 accessed
emergency accommodation or were in longer term transitional
accommodation (Homeless Agency, 2009).
Since 2006, the Homeless Agency has maintained a death
notification policy that requires homeless services to report the
death of a service user within the Dublin area. From January
to December 2007, 53 deaths were reported (41 men and 12
women) aged between 26 and 59. These individuals were residing
in homeless accommodation or were in recent contact with home-
less services before their death (Homeless Agency, 2008). This sug-
gests, and interviewees in this research confirmed, that workers in
any of the homeless services are likely to encounter the death of a
service user frequently. Detailed information on the causes of
death is not publically available but it appears that the most com-
mon causes of death are related to alcohol use in men and drug
overdose in women.
Methodology
The methodology used in this project was most closely aligned to
classical grounded theory as first outlined by Glaser and Strauss
(1967) and further elaborated by Glaser (1978, 1992, 1998,
2001). Glaser and Strauss argued for an increased emphasis on
generation or discovery of theory derived from data in the social
sciences rather than on the verification of ‘‘thought up’’ or logically
deduced theories. An assumption of grounded theory is that
human behavior is characterized by latent patterns or processes
that grounded theory seeks to make visible and explain. Concep-
tualization rather than description is at the heart of grounded
theory, which Glaser (2001) asserted transcends the descriptive
in ‘‘its ability to generate ‘wise’ propositions that explain behavior
in an area, especially its main concern, its ability to organize and
make meaningful many seemingly disparate incidences’’ (p. 25).
He goes further by asserting that grounded theory is concerned
with the generation of concepts that are abstract in terms of time,
place, and people. Corbin and Strauss (2008) take a softer line in
relation to abstraction in grounded theory, suggesting that concepts
devoid of context are ‘‘like jelly donuts devoid of jelly’’ (p. 306)
928 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
and that contextualization of concepts is necessary for readers to
fully understand the research. The present project errs toward con-
ceptual description rather than theory in places; so, as such it might
be more properly described as a modified grounded theory.
Several related techniques, characteristic of grounded theory,
were used in this project. A form of sampling called theoretical
sampling (Glaser, 1998) involved allowing the emerging theory to
guide where to go next, what data to collect or further review so
as to arrive at the main concern of the participants as quickly and
efficiently as possible. The constant comparative method of analysis
involved processes of both coding and theoretical sampling.
In-depth interviews with front-line homeless sector workers
were the primary source of data. Numerous memos were made
about the data, concepts, and properties (substantive codes) and
their relationships to capture, track, preserve, and develop concep-
tual ideas (Glaser, 1998, p. 180). The aim was to analyze and code to
generate an emergent set of categories (a higher level concept) and
their properties. Sampling or data collection and analysis involved
an iterative or near-constant process. From the outset, each indi-
cator was compared to others, assigned a conceptual code, and then
the indicator was compared to each concept. The data were revis-
ited many times in order to elaborate, saturate, and verify the
emerging categories (this is known as substantive coding).
Transcripts of interviews were subject to an initial, painstaking
line-by-line coding in keeping with Glaser’s (1978) recommenda-
tions. Subsequent coding was guided by theoretical sampling.
Early coding produced mostly in-vivo codes, that is, derived from
the language of participants (e.g., lighting a candle or paying
respects). As these incidents were compared they gave rise to more
conceptual in-vitro codes reflecting a higher level of conceptual
abstraction (e.g., marking death). As these emerged, the entire pool
of data was further reviewed for further examples of the concept
and properties. Framing death and moving on, the overarching
process and subordinate categories emerged from the data, rather
than being forced by applying preconceived theories or ideas.
Ethical Considerations
The project received ethical approval from both the Dublin City
University and James Cook University ethics committees. It was
Homeless Workers 929
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
recognized that discussing experiences associated with sudden
death might be traumatic for people; so, a free and confidential
consultation with a local counseling service was offered to intervie-
wees. To the author’s knowledge, no respondents availed them-
selves of this opportunity. The potential for retraumatization was
reduced by the sensitivity of the interviewer who had a back-
ground in psychotherapy as well as mental health work in the
homeless sector.
Recruitment and Interviewees
Homeless sector workers were recruited through an advertisement
in a sector-wide newsletter and via information sheets left at vari-
ous agencies. People then contacted the researcher directly for
further information. Approximately 40 people made contact via
e-mail or phone and discussed their experiences. Sixteen homeless
sector workers were interviewed in depth at a place of their choice
for between 40 and 120 min. Initially convenience and availability
dictated who would be interviewed, but, later, theoretical sampling
was used. The interviews commenced with the worker being
invited to share any thoughts or observations about sudden death
in the sector, and then usually one or more experiences were
explored in depth. In keeping with the grounded theory method-
ology, no schedule of questions was used; rather, emerging themes
in analysis guided choice of questions. The interviews were
recorded and transcribed and approximately 100 detailed memos
were made relating to interviews.
The ages of interviewees ranged from the late 20 s to mid-50 s
and their experience working in the homeless sector ranged from 2
to 30þyears. A small number of interviewees identified a pro-
fessional affiliation such as psychiatry or social work. The majority
did not have a professional affiliation, although most had com-
pleted some university level education in areas such as addiction
studies. Some were employed in low-threshold outreach capacities
focusing on particular target groups such as those with HIV, sex
workers, people sleeping in public spaces, or with alcohol prob-
lems. Others worked in supported accommodation including
emergency shelters, high-support hostels for individuals or home-
less families, and residential alcohol- and drug-treatment facilities
involved in providing intensive therapy. All had experience
930 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
directly providing some kind of service to people who had subse-
quently died. Some had attended the funeral or had discovered the
body of a service user on the day they were interviewed.
Results
The products of a grounded theory study are conceptual and ought
to be of a higher level of abstraction than description. Neverthe-
less, in presenting the findings, some quotes from interviewees
are used to illustrate concepts. The main concern of participants
was to positively frame the death of service users and move on.
This and associated processes are illustrated in Figure 1. Some
participants were able to reflect on a completed process of framing
death, whereas others were immersed in the process. The straight
line in Figure 1 reflects work as usual, although homeless sector
work takes place against a backdrop of expectation that harm
may befall service users. Encounters with death always demand
a response and a deviation from the usual rhythms and responsibil-
ities of work. The processes of responding to death, marking death,
and recognizing and responding to vulnerability are interlinked
processes that bring people more closely in contact with the
deceased (or the death encounter) and may be considered compo-
nents of the core concern of workers to positively frame death.
These are not necessarily linear processes (following one after
FIGURE 1 The process involved in positively framing the death of a service
user.
Homeless Workers 931
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
another) but may occur simultaneously, have greater salience at
particular points in time, and be returned to. Positively framing
the life and death of the service user is both an end product of
the aforementioned processes and a process in its own right. If suc-
cessfully resolved, the worker is able to carry on their work, revisit-
ing the framed picture at their leisure without being preoccupied or
overly distressed by death.
Expecting the Unexpected
Even before workers directly encounter the death, they hold
expectations about their work and the likelihood of harm befalling
service users. The concept ‘‘expecting the unexpected’’ describes
the paradox that homeless sector workers understand that service
users are often at high risk of death, and when it occurs, it is almost
always accompanied by shock. This realization of the vulnerability
of service users to harm can come to workers quite abruptly.
Rarely do workers commence employment expecting to encounter
death, and seldom is this possibility discussed when people began
employment or during their orientation to the workplace. Some
people described a moment in which they realized that the people
whom they worked with could not possibly sustain the levels of
drinking, drug taking, or risky behavior without some serious
consequences.
People who are homeless for long periods of time inevitably
experience many losses. In long-term accommodation or care
environments, some workers noted being initially surprised at
how frequently and casually service users spoke of the deaths of
friends, family, and acquaintances. In drug treatment environ-
ments these losses are not so often acknowledged, or there may
be superficiality to discussions in which the impact of death and
loss on oneself is not spoken of. Sudden death is the ‘‘ghost in
the room’’ in many discussions with homeless people. The fre-
quency of death and loss among some groups such as intravenous
drug users can be very high. One experienced worker (Interviewee
6) recalled working with someone close to her own age and dis-
cussing a photograph of children taken in the 1950s. All of the
acquaintances and family members had died in this photo leaving
the service user the lone survivor. All had died from alcohol, drug
use, or violent assault. In contrast to service users who had lost the
932 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
majority of their friends in some deprived areas, she noted that she
had lost one personal friend in her life.
Exposure to the stories of service users and other staff and a
recognition of the vulnerability of service users in a sense primes
workers to expect and deal with death. However, when confronted
by sudden death, people noted that it often was not who they
expected or when. This illustrates the paradox of expecting the
unexpected. Encountering death of a service user can in turn lead
to a heightened sense of expectancy in the future, and if death is
not worked through adequately or framed positively, a sense of
futility may ensue.
Encountering Death
How death is encountered makes a substantial difference to how it
can be dealt with. Whether the person witnessed or is present at
death, how the body is discovered, and in what circumstances
are dimensions of encountering death. In residential facilities, a ter-
minal illness may be diagnosed, or the person may become acutely
ill and is admitted to hospital. The homeless sector worker may
observe the service user’s deterioration and accompany the person
to consultations with health professionals. The person may die in
the hospital. To some extent, this expectancy of death attenuates
the emotional impact of the person’s death. Dying and death is
medically contained, and the worker is in a similar position to
those who may work in palliative care or aged care. The emotional
labor of caring for the person is or ideally should be acknowledged
by colleagues through gestures of support. The degree to which
dying or suffering is witnessed is a property of encountering death.
Sudden, unexpected death, in contrast, is associated with
more ambivalent and sometimes conflicted responses. Some kinds
of death, such as those that occurred as a result of overdose, sui-
cide, and homicide are particularly shocking. Homeless sector
workers rarely see coroners’ reports, and the actual cause of death
(even whether the death is classified as suicide) often remains
unclear for workers. People noted that opiate overdoses often
occurred when people seemed to be doing well and perhaps had
been abstinent for some time. The death of someone who appears
to be doing well or improving, or the death of a child, can be parti-
cularly difficult. As one worker who had encountered many deaths
Homeless Workers 933
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
of young people commented, ‘‘sometimes there is no dignity in
dying’’ (Interviewee 7).
The directness of the encounter with death varies as well. For
some, the news can be broken in passing by service users or other
workers face-to-face, over the phone, or formally through staff
meetings or briefing sessions. Rarely, workers may be called at
home by staff or emergency workers to assist in identification of
the body. Sometimes, the worker may not hear of the death for
some time, and there may be no opportunity to see the deceased.
More direct encounters include discovering the body or being with
the service user when the person dies. Discovering the body can be
particularly difficult especially when there are signs of trauma or
body fluids. Encountering the broken body is a more visceral,
shocking experience, and some respondents spoke of experiencing
intrusive recollections of the dead body.
Responding to Death
The initial reactions to encountering death vary as a function of the
personal vulnerability of the worker, the relationship that the
worker had with the service user, the worker’s role in the agency,
how the death was encountered, and how recently the death
occurred. Workers need to respond according to the procedures
laid out, and death also provokes a psychological response. Almost
everyone spoke of being shocked or surprised and going into
‘‘autopilot’’ when discovering a body or first hearing the news of
death. On encountering death, people took solace from following
procedures (where they are clearly laid out). Minimally, this
involved verifying whether the information received was true, such
as in the case of a service user who was reported dead by another
service user. When a worker discovers a body, procedures
included securing the scene and then notifying police, ambulance,
and the line manager. The police or ambulance personnel typically
directed how the body needed to be handled from that point for-
ward. Once the death had been officially confirmed, there is then
a need to notify other staff, other service users as appropriate, and
mobilize support for the other service users.
A common psychological response was to feel guilty and to
project blame onto agencies that failed to intervene to protect ser-
vice users. Workers generally appreciated that they could not and
934 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
did not intervene to save the person from harm or death but that in
many cases more could have been done to help or protect them.
The person was seen as responsible for the choices and actions,
but rarely did respondents speak of being angry with the deceased.
More often, people discussed being angry with services that were
perceived to have failed the person who died in some way, such
as failing to provide shelter, not trying hard enough to engage with
the person, or not marking the death in a respectful way.
People described experiencing the full gamut of emotions
around the death of service users, from little emotional response
to being overwhelmed by grief. Strong emotional responses typi-
cally occurred days or weeks after encountering death when the
initial shock or surprise subsided. Emotional responses were typi-
cally triggered by encountering an event that reminded them of
the deceased or was particularly stressful. Workers may carry parti-
cular hopes or aspirations for service users that are shattered when
the service user dies. For example, one worker with a lengthy his-
tory working in the sector and who stated that he was rarely
emotionally affected by the work described having a professional
relationship with a service user over several years:
The social worker rang me up and said, ‘‘Oh, have you heard about [the
person]?’’ And I said, ‘‘No.’’ And, ‘‘Well he died last night in [the hospital].’’
And my immediate response was, thank goodness he didn’t die alone,
thank goodness he was surrounded by people who cared about him and
he was in a place where he was getting care . . . The next day my colleague
rang me on the way into work and he said, ‘‘Have you seen the
paper?’’ ...And there was a big article about [the person] who had died
in a hospital in Dublin and died in the waiting room on his own. Now that
kind of really upset me . . . the one thing I never wanted to happen for him
was that he would die alone. (Interviewee 2)
In order to carry on with necessary work, and perhaps also to
prevent one from being overwhelmed with emotion, several
people spoke of putting their emotions, ambiguous or disturbing
images, or thoughts and memories ‘‘in boxes’’ or ‘‘pushing them
back.’’ Although initially functional or helpful in maintaining equi-
librium, most ultimately found that it was not helpful in the longer
term. There was a need to experience rather than suppress
thoughts and feelings. One person (Interviewee 9) stated she felt
numbed as a consequence of attempting to put so many thoughts
and feelings in boxes, and this was interspersed by emotions and
Homeless Workers 935
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
thoughts flooding unbidden through flashbacks (akin to post-
traumatic stress responses).
Marking Death
Marking death serves a therapeutic function in that it helps people
to experience emotions and is part of properly framing death.
Marking death is undertaken in highly personal and private ways
as well as collectively through formal mechanisms such as mem-
orial services and funerals.
How death is marked or memorialized, collectively, is in part
culturally determined. In Irish society, a ‘‘good turnout’’ at a
funeral is recognized as a sign of respect for the deceased. Some
workers (most notably those aligned to one of the professions)
described making a conscious decision not to attend the funerals
of service users. Others stated they do attend funerals or their
decision to attend would be conditional on how well they knew
the service user or their personal feelings towards the person. At
least for some people, marking death with other people is an
important means of closure.
People were aggrieved if they perceived that death was not
marked respectfully, for example, when the death of a service user
is announced at a team meeting and immediately followed by dis-
cussion of more mundane matters. Some people also felt disre-
spected if they were not given the choice to attend the funeral.
However, attendance at funerals sometimes poses problems for
homeless sector workers who may have had a highly sensitive role
in relation to the deceased or if they were aware of behavior of the
deceased that was highly shameful or stigmatizing. Out of respect
for the family and the deceased, sometimes knowledge cannot be
shared or processed at the funeral and needs to be dealt with
through other means. For example, one worker spoke of attending
a funeral and making herself known to the mother of the deceased:
I felt really uncomfortable afterwards because I knew that he was up in the
[park] selling himself and his mother didn’t know that, I found that really
difficult to have that information and this mother didn’t know that about
her son . . . I kind of struggled with that a little bit. (Interviewee 4)
Funerals can also sometimes be difficult when only a small
number of people turn up or the attendees are only from the
936 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
homeless sector. The extent to which people have lived margina-
lized lives is brought home to the worker, which can itself be trau-
matizing rather than consoling. There may also be complications
around the circumstances of a death that preclude the possibility
of attendance at funerals by both other service users and workers.
In one instance, a person’s identity was unable to be confirmed,
and no funeral arrangements could be made. A worker stated,
So eventually I just organized [the memorial service] myself, and it was
great, it was nice. It wasn’t just for him, it was for all people but his name
was mentioned and the residents loved it, it was very important for
them . . . it gave some closure to it because up to then there was nothing,
he died and that was it. (Interviewee 5)
In some agencies, workers arranged their own collective ways
of marking and memorializing death outside of the traditional
funeral such as convening a special group, planting a tree, or hav-
ing a sing-along. People also spoke of small things that they person-
ally did to mark death, for example, lighting a candle or spending a
few moments in quiet reflection at some point during the working
day. In Dublin, there are more communal means of memoralizing
or remembering those whom have died. For example, some work-
ers attend annual church memorial services dedicated to homeless
people or victims of drug-related deaths. A publication entitled the
Brass Munkie written by and for drug users routinely includes a
remembrance page and stories about people who have died. This
is read by both service users and workers.
Recognizing the Vulnerability of Service Users, Self, and Colleagues
Workers were acutely aware of the vulnerability of other service
users in relation to the death of a friend or acquaintance, and this
sensitivity is heightened immediately postbereavement. Many ser-
vice users use self-destructive methods of coping and events, such
as the death of a friend or acquaintance leave some people vulner-
able to exacerbation of drinking, illicit drug use, or reckless beha-
vior. Some respondents cited examples of dramatic breakdowns in
the cohesion of therapeutic groups when a member of the group
died. Workers took great care with how they broke the news of
death to service users and facilitated opportunities for them to
express their thoughts and feelings.
Homeless Workers 937
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
In well-functioning work teams the emotional vulnerabilities
of colleagues are recognized. People may be particularly vulner-
able because of past experiences, such as the suicide of a family
member, miscarriage, personal bereavement, the accumulation of
stress, or personal identification with a service user. All respon-
dents described at least one relationship with someone who had
died that they recalled was characterized by special bonds. These
relationships could all be characterized as ‘‘professional,’’ but the
worker had a particularly warm relationship, held high hopes
for, or worked particularly well with the service user. Astute and
sensitive colleagues would recognize these bonds.
The deaths of service users can be wounding in different ways
for different people. One worker commented that she found it
easier to deal with the death of a friend or personal acquaintance
because she ‘‘knew where to park it’’ (Interviewee 8) and it is
clearer how one can and ought to respond. Encountering death
can confront service users and workers alike with their own mor-
tality. For the worker, the death of a service user may undermine
professional confidence and sense of effectiveness, and people may
be vulnerable to posttraumatic stress symptoms and professional
burnout.
Responding to Vulnerability
People appreciated it when their vulnerability was recognized by
others through a kind word, an inquiry as to their well-being, per-
mission to leave work early, or an invitation to talk about how they
were feeling. Some spoke of ringing up or making a point of con-
tacting a worker whom they knew would be affected by the death
of a particular service user. Some workers had established semifor-
mal networks with people from other agencies who worked with
particular service users. The frequency of meeting would be
increased after a service user died, and these would take on a more
supportive function.
A number of formal processes were available to people to
assist them make sense of, learn from, and deal with the emotional
impact and trauma of death. Most respondents had refused the
offer of formal debriefing or counseling, at least initially, and wor-
ried that they might be perceived as ‘‘weak’’ or ‘‘unprofessional’’ if
they immediately availed. Those who had experienced formal
938 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
debriefing, personal counseling, or supervision from someone not
in a line-management position reported it as being immensely
valuable to express and process ambiguous, irrational, and painful
thoughts and feelings about the deceased person:
At the debriefing . . . you can say what you want and you can allow yourself
to be completely irrational and be angry about it, that is fine. It is important
to vent, like everyone needs to vent, but you have to have a safe place to do
it and I don’t think internally is a safe place regardless of how brilliant the
organization is, it is not safe. (Interviewee 5)
Some workers reported that similar needs were met through
informal peer networks, although most of these workers
acknowledged they felt lucky and such networks were not avail-
able to all.
Strategies to care for one’s self were described by one as ‘‘criti-
cal’’ to enduring the traumas associated with front-line homeless
sector work and particularly the deaths of service users. Some
more experienced workers spoke about maintaining clear bound-
aries in their working relationships and between their working
and home lives. Maintaining compassion, hope, and a friendly,
optimistic, and professional demeanor is essential to effective work
in the homeless sector, but it is helpful to make a clear distinction
between being friendly and being friends and to make a demar-
cation between home life and professional life. Fostering a positive
personal=home life helps in coping with the traumas of working
life. People spoke of exercise, such as going to the gym, kick box-
ing, or walking, and various activities with loved ones as helping to
‘‘switch off’’ from work.
Personal counseling was of value to many people to help dis-
tinguish between the professional and personal problems, and
supervision to prevent encroachment of work-related issues into
people’s personal lives. This is an ongoing struggle for many
people who make a personal investment in helping others and
often spend lengthy periods of time with people and come to know
them well. Framing the death of a service user as a professional loss
rather than a personal one enabled the worker to more easily con-
tinue in her or his role. If failings or blame is to be apportioned,
framing the failure as a service=system or professional failing
rather than a personal one at least enables the worker to begin
to address service problems or weaknesses.
Homeless Workers 939
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
The importance of acknowledging successes (however small)
in the job was highlighted as a self-care strategy, particularly useful
to stave off a potential sense of futility and to maintain enthusiasm.
What kind of compensates for the work that I do . . . are those joys that
come along when you see somebody just finally getting to the point of
being drug free and who starts making good decisions about their future
. . . or families that are reunited because someone has got stable on
methadone. (Interviewee 6)
Others had combined activities in their work that gave more
immediate rewards with front-line outreach work. It may be
unreasonable to expect most people to continually engage in
low-threshold outreach work over a long period of time without
some personal cost.
Positively Framing the Death of a Service User
Positively framing the death of a service user is both an ongoing
cognitive process of sense making, and an outcome of the inter-
linked processes of responding to death, marking death, and recog-
nizing and responding to vulnerabilities. A positive frame enables
people to continue working in their roles or with vulnerable popu-
lations with enthusiasm, optimism, and compassion. Positively
framing the death of a service user entails both a general way of
viewing oneself in relation to service users generally and to the
deceased person specifically.
Workers who reported obtaining great satisfaction and
enthusiasm for their work described particular ways of framing
their work and their relationship to deceased service users. They
tended to articulate an awareness and acceptance of the limitations
of their power and role in relation to the deceased service user and
service users generally (e.g., ‘‘You can only do as much as you
can’’ [Interviewee 1]) and focus their attention on what assistance
they can provide to others (e.g., ‘‘[T]here is nothing more I can
do for that person now but the living still need looking after . . . It
is an acceptance, this is part of life’’ [Interviewee 2]). These workers
recognized that at best they could influence people toward making
good decisions.
Positively framing the death of service users can involve
acknowledging and holding on to a memory of work that has been
940 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
done well or contributions made by workers or services to the per-
son’s quality of life. One worker described how a former homeless
man and heavy drinker was housed in transitional accommodation:
He got the best service that we could possibly give him here, we kind
of went above and beyond a lot of times . . . he was happy . . . he had his
own apartment ...he had his own pictures up in his own apartment.
(Interviewee 3)
The traumatic and seemingly senseless deaths of some service
users pose challenges to positively framing the person’s death,
making sense of their life and homeless sector work. When con-
fronted by particularly traumatic experiences, a positive frame
may involve a renewed commitment to justice, advocacy, or work
in general so as to give meaning and significance to a person’s
death:
There was one fellow who was shot dead.. Yeah I think something hap-
pened inside me when that happened because I really did get affected by
that, definitely something kind of . . . It made me question, ‘‘Jesus, is this
the job for me and is this what I want to do?’’ But in a way it has kind of
pushed me harder to try even better or to try and stick with people and
see how they are doing and things like that. (Interviewee 4)
Confronting death can challenge people’s capacity to carry
hope or continue to invest in others. The importance of maintain-
ing hope for others was repeatedly highlighted:
I think it is really important that people always keep the vision, no matter
the circumstances, how awful it is for homeless people, or how awful it is for
the drug users or the sex workers, that there is always a vision ahead that
things can get better or things can change. (Interviewee 6)
When a service user dies, vulnerabilities of both self and
others are laid bare. If these vulnerabilities not adequately
attended to, or the experience processed, people may remain
trapped in a cycle in which they continue to encounter the death
through intrusive recollections and memories, or they remain
hypervigilant and anxious that harm may befall other service users.
In a sense the experience is poorly contained. In contrast, a suc-
cessful response or outcome is a positive frame in which an image
of a deceased person and the homeless sector worker is captured,
contained, and preserved much like a picture frame on a
Homeless Workers 941
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
mantelpiece or wall assists in preserving and sustaining a memory.
The frame enables the person to be remembered and revisited
rather than details being suppressed or strong emotions spilling
out unbidden. Many factors impinge on the process of framing the
life and death of homeless service users, or to process the death in
the way that people might when working in other capacities or
with other populations. These processes enable people to keep work-
ing effectively in their work despite confronting trauma and death.
Discussion
How to establish the credibility of a grounded theory study has
been well described in the literature (Charmaz, 2006; Corbin &
Strauss, 2008; Glaser & Strauss, 1967). Glaser and Strauss (1967)
suggested that a grounded theory should fit the phenomena, be
understandable, and be sufficiently abstract to apply to a variety
of contexts. The credibility of a grounded theory depends on
‘‘grab and fit,’’ that is, whether the concepts make sense and fit
with the data and the real world experience of the social actors
involved.
These findings were presented to a diverse selection of home-
less sector workers undertaking a course of study on case manage-
ment in the homeless sector, and the expressed consensus was that
the theory appeared to fit with their experiences and made sense.
Diligence in the application of open coding, comparative analysis,
memoing and theoretical sampling go some way to ensure that the
theory generated fits the data.
The grounded theory in this study accords with findings relat-
ing to how other groups respond to death, but there are subtle dif-
ferences relating to the role of worker and service user. The work
of the homeless sector worker intersects with other social care and
health workers. However, homeless people have been found to
have particular concerns around dying, death, and end of life.
These include fears of dying anonymously, alone, and without
remembrance and a generally positive attitude toward developing
advance directives or ‘‘living wills’’ (Song, Bartels, et al., 2007;
Song, Ratner, et al., 2007). Although homeless people can
experience a ‘‘good death,’’ that is, an awareness, acceptance,
and preparation for death (Beverley, Charles, & Margaret, 1995)
by all those concerned, death is more likely to be unexpected
942 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
and traumatic. In research involving other groups, the mode of
death has been found to make a difference to how people cope
and grieve. Reed (1998) conducted a survey of survivors of sudden
loss from suicide or accident and concluded that the mode of death
and the availability of social support are important determinants of
grief symptomology. Similarly, aspects of the way death was
encountered in this study affected emotional response and how
the death was ultimately framed.
The impact of caring for dying people by health professionals
has been explored (Redinbaugh et al., 2003), and the process has
been conceptualized as ‘‘grieving’’ (Papadatou, 2000). Although
respondents in this study sometimes used the term grief (for want
of a better word) to describe some feelings, the processes described
were different from and broader than conventional understandings
of grief and loss. Grief is understood to be triggered by losses such
as personal bonds, valued relationships, the nonrealization of
professional goals, realization of one’s own mortality, and the
emergence of past unaddressed losses (Papadatou, 2000). Workers
do grieve in this sense, but how they do so is affected by the mar-
ginalized social position of the homeless person and indeed them-
selves as workers in the sector. Traditional rituals such as funerals
help many people accept the reality of loss (Worden, 2001), but the
homeless sector worker often cannot engage in such rituals in the
same way as family members and others, rendering other ways
of dealing with the experience of death more important.
The homeless service user typically inhabits the margins of
society and for the most part is invisible except for the occasional
intrusion into our collective consciousness through newspaper
headlines or requests for spare change. The homeless sector
workers (even those with conventional professional identities) are
also to a large extent invisible and are relatively powerless to
address or control the circumstances that lead people to be at risk
of death. The homeless sector worker must sometimes confront the
death of the service user that is colored by violence, gross injustice,
and death with little dignity in which the worst fears of individuals
are sometimes realized. It is hardly surprising that the experience
of dealing with such issues is sometimes marginalized and dealing
with the impact sometimes difficult.
Homeless workers also tend to have different relationships
with service users that vary in attachment, intimacy, and role
Homeless Workers 943
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
expectations to those of other professional groups. Nevertheless,
some of the concepts derived from this grounded theory have res-
onance with research undertaken with physicians, nurses, and
others. Physicians who have found particular patient deaths to
be emotionally powerful (and these have been categorized into
good, overtreated, or shocking=unexpected death) have reported
changes in their clinical behavior and career paths (Jackson et al.,
2005). Medical doctors who cared for patients for longer periods of
time reported stronger emotional reactions and were more vulner-
able to feelings of loss (Redinbaugh et al., 2003). Those results are
consistent with the present grounded theory, which suggests that
the nature of the death encounter as well as the worker’s identifi-
cation with the deceased impacts on the response to the death
and the process of framing death.
O’Hara, Harper, Chartrand, and Johnston (1996) examined
the impact of death on a small cohort of nurses working in
long-term hospital wards and found that they coped by sharing
feelings, framing death positively, and framing their work as con-
tributing to a good death. Those who were more affected by death
tended to ‘‘take their stress home’’ and to consider their relation-
ships with patients as more personal. Ekedahl and Wengstro
¨m
(2006) concluded that being careful about maintaining professional
boundaries was an important strategy in coping with the deaths of
patients by 15 nurses who worked in cancer care. Although home-
less workers may have more difficulty in framing their work as
contributing to a good death, they can frame their work with the
person and others in positive terms. Making a ‘‘boundary demar-
cation’’ is similar to both the self-care strategies described by
homeless sector workers (such as fostering a positive personal life)
and the positive framing of their relationship with the deceased,
which in some instances involved seeing the relationship as
professional rather than personal.
There are some limitations inherent in this grounded theory.
A grounded theory methodology may faithfully account for what
is going on in a social field, but the resulting theory may not
address what ought to be happening. For example, this theory
holds that encountering death provokes a personal emotional
response and demands a procedural response (as a worker). This
theory falls short of determining what the optimal initial emotional
response is (a concern that some participants had) or what the best
944 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
kind of policies or procedures are. Similarly, marking death in
some way collectively and personally is helpful in terms of posi-
tively framing death. However, the theory does not reveal how
one ought to mark death. The level of abstraction of this theory
does not describe how the various subprocesses (expecting,
encountering, responding, and marking death) contributes to posi-
tively framing death. It is possible, for example, that psychological
theories relating to attribution may provide a different and useful
account of the process. There are also limits to the extent that each
concept can be explored in one publication. Further, elucidation
of the attributes of each concept would assist in illuminating the
theoretical links and operationalizing concepts for future research
and verification with different groups.
Research with other groups and utilizing different methodolo-
gies has highlighted concepts that are similar to those induced in
this present study. Exploration of the phenomenology of nursing
students encountering death for the first time revealed how part
of the process involved ‘‘integrating patient death into the realm
of professional practice’’ or reconstructing a view of work that
incorporated patient death (Kelly, 1998). This has similarities to
the task of framing discovered in this study in which both the
deceased individual is remembered and the relationship with the
worker and the worker’s relationship with their work is also
reflected upon. For the experienced respondents in this study,
most had managed to construct a view of work that incorporates
the possibility of encountering death and trauma.
How workers can be supported in positively framing death
and their work can be inferred from the theory as follows:
.It is likely to be helpful to raise the possibility of encountering
death or trauma of service users during workplace induction.
.Clear policy and procedures relating to what to do when death is
encountered ought to be in place. The worker needs to consider,
and make known to others, how and when they would like to be
notified if a service user is involved in a traumatic incident.
.Workers ought to be encouraged to mark the death in a personal
way and ought to be offered the opportunity to participate in
collective activities to memorialize the death.
.Workers ought to receive some training around maintaining
personal and professional boundaries.
Homeless Workers 945
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
.Clinical supervision should be available to frontline staff in dif-
ficult roles, and one-to-one professional psychological debriefing
should be available to everyone following traumatic incidents.
.Individual workers and organizations need to be able to review
events leading up to death and trauma, to learn from experience
and prevent harm to others.
For some, positively framing death becomes easier with time.
Rather than suppressing thoughts, feelings, and images, the main
concern of homeless sector workers confronting sudden death is
framing the experience as intensely human, acknowledging the tra-
gedy and indignity of death, but also preserving hope and opti-
mism that they can make a difference to those who remain.
References
Alexander, D. A., & Klein, S. (2009). First responders after disasters: A review of
stress reactions, at-risk, vulnerability, and resilience factors. Prehospital &
Disaster Medicine,24(2), 87–94.
Beverley, M., Charles, W., & Margaret, C. (1995). Threats to the good death: The
cultural context of stress and coping among hospice nurses. Sociology of Health
& Illness,17, 222–241.
Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative
analysis. London, UK: Sage.
Cheung, A. M., & Hwang, S. W. (2004). Risk of death among homeless women: A
cohort study and review of the literature. Canadian Medical Association Journal,
170, 1243–1247.
Corbin, J., & Strauss, A. (2008). Basics of qualitative research (3rd ed.). Thousand
Oaks, CA: Sage.
Desai, R. A., Liu-Mares, W., Dausey, D. J., & Rosenheck, R. A. (2003). Suicidal
ideation and suicide attempts in a sample of homeless people with mental
illness. Journal of Nervous & Mental Disease,191, 365–371.
Ekedahl, M., & Wengstro
¨m, Y. (2006). Nurses in cancer care: Coping strategies
when encountering existential issues. European Journal of Oncology Nursing,
10, 128–139.
Fitzpatrick, K. M., Irwin, J., Lagory, M., & Ritchey, F. (2007). Just thinking about
it: Social capital and suicide ideation among homeless persons. Journal of
Health Psychology,12, 750–760.
Glaser, B. G. (1978). Advances in the methodology of grounded theory: Theoretical sensi-
tivity. Mill Valley, CA: Sociology Press.
Glaser, B. G. (1992). Emergence vs. forcing: Basics of grounded theory analysis. Mill
Valley, CA: Sociology Press.
Glaser, B. G. (1998). Doing grounded theory: Issues and discussions. Mill Valley, CA:
Sociology Press.
946 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
Glaser, B. G. (2001). The grounded theory perspective: Conceptualisation contrasted with
description. Mill Valley, CA: Sociology Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. London, UK: Aldine Transaction.
Haw, C., Hawton, K., & Casey, D. (2006). Deliberate self-harm patients of no fixed
abode: A study of characteristics and subsequent deaths in patients presenting
to a general hospital. Social Psychiatry & Psychiatric Epidemiology,41, 918–925.
Homeless Agency. (2008). Findings from death policy notifications in 2007. Dublin,
Ireland: Author.
Homeless Agency. (2009). Rough sleeper count [Press release]. Update: The
Newsletter of the Homeless Agency,62.
Hwang, S. W., Orav, E. J., O’Connell, J. J., Lebow, J. M., & Brennan, T. A. (1997).
Causes of death in homeless adults in Boston. Annals of Internal Medicine,126,
625–628.
Jackson, V. A., Sullivan, A. M., Gadmer, N. M., Seltzer, D., Mitchell, A. M.,
Lakoma, M. D., ... Block, S. D. (2005). ‘‘It was haunting . . . ’’: Physicians’
descriptions of emotionally powerful patient deaths. Academic Medicine,80,
648–656.
Jonsson, A., & Segesten, K. (2004). Daily stress and concept of self in Swedish
ambulance personnel. Prehospital & Disaster Medicine,19, 226–234.
Kelly, C. T. (1998). The lived experience of female student nurses when encountering
patient death for the first time (Unpublished doctoral dissertation). Adelphi
University, Garden City, NY.
Kinder, A., & Cooper, C. L. (2009). The costs of suicide and sudden death within
an organization. Death Studies,33, 411–419.
Kushel, M. B., Evans, J. L., Perry, S., Robertson, M. J., & Moss, A. R. (2003). No
door to lock: Victimization among homeless and marginally housed persons.
Archives of Internal Medicine,163, 2492–2499.
Morrison, D. S. (2009). Homelessness as an independent risk factor for mortality:
Results from a retrospective cohort study. International Journal of Epidemiology,
38, 877–883.
Nordentoft, M., & Wandall-Holm, N. (2003). 10 year follow up study of mortality
among users of hostels for homeless people in Copenhagen. British Medical
Journal,327, 81–83.
O’Hara, P. A., Harper, D. W., Chartrand, L. D., & Johnston, S. F. (1996). Patient
death in a long-term care hospital: A study of the effect on nursing staff.
Journal of Gerontological Nursing,22(8), 27–35.
Papadatou, D. (2000). A proposed model of health professionals’ grieving process.
Omega: The Journal of Death & Dying,41, 59–77.
Redinbaugh, E. M., Sullivan, A. M., Block, S. D., Gadmer, N. M., Lakoma, M.,
Mitchell, A. M., . . . Arnold, R. M. (2003). Doctors’ emotional reactions to
recent death of a patient: Cross sectional study of hospital doctors. British
Medical Journal,327, 185–191.
Reed, M. D. (1998). Predicting grief symptomatology among the suddenly
bereaved. Suicide and Life-Threatening Behavior,28, 285–301.
Sabin-Farrell, R., & Turpin, G. (2003). Vicarious traumatization: Implications for
the mental health of health workers? Clinical Psychology Review,23, 449–480.
Homeless Workers 947
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
Schanzer, B., Dominguez, B., Shrout, P. E., & Caton, C. L. M. (2007). Homeless-
ness, health status, and health care use. American Journal of Public Health,97,
464–469.
Song, J., Bartels, D., Ratner, E., Alderton, L., Hudson, B., & Ahluwalia, J. (2007).
Dying on the streets: Homeless persons’ concerns and desires about end of
life care. Journal of General Internal Medicine,22, 435–441.
Song, J., Ratner, E., Bartels, D., Alderton, L., Hudson, B., & Ahluwalia, J. (2007).
Experiences with and attitudes toward death and dying among homeless
persons. Journal of General Internal Medicine,22, 427–434.
Stayt, L. C. (2009). Death, empathy and self preservation: The emotional labour
of caring for families of the critically ill in adult intensive care. Journal of
Clinical Nursing,18, 1267–1275.
Worden, J. W. (2001). Grief counseling and grief therapy: A handbook for the mental
health practitioner (3rd ed.). New York, NY: Routledge.
948 R. Lakeman
Downloaded by [Southern Cross University], [Richard Lakeman] at 15:06 25 October 2011
