‘I just hope they take it seriously’: homeless men talk about their health care

People experiencing homelessness in Australia often have complex health and social issues, including the trimorbidity of mental illness, substance use disorders and physical health conditions,¹ as reported both overseas^2,3 and in Australia.^4–6 These issues are compounded by deep social exclusion,⁷ and these men are at risk of delayed,^8,9 costly^9,10 and ineffective use of health services, including non-attendance at appointments and the frequent use of ambulance and acute hospital services when primary care may be more appropriate.^1,11–13 Providing care for this group is widely acknowledged to be demanding and difficult.^1,2,11 Evidence-based components of effective health care for homeless people include specialised community-based services offering outreach to clients and in-reach into hospital-based services,^2,9 service user and peer involvement^2,14 and personalised case workers who empower and advocate for clients and coordinate their care.^2,9,13 Coordinated, collaborative and integrated health and social care has been described as the holy grail of homeless health care.⁷ However, fragmentation and duplication of the various services and agencies involved appear to be persistent problems in many health systems.^11,15,16

Approximately 45% of 116 427 Australians experiencing homelessness on census night 2016 were adult men, and another 39 026 were in other marginal housing.¹⁷ However, the voices of homeless men remain under-represented in the Australian healthcare literature, despite two recent interview studies of homeless people.^18,19 We have been unable to identify any Australian focus group studies. The aim of the present study was to explore the insider perspectives²⁰ of homeless men in order to identify potential improvements in the provision of their health care.

Both authors are academic clinicians who have worked in general practice with homeless patients. The authors adopted a constructivist realism orientation,²¹ acknowledging that participant accounts serve as social signals and are constructed,²² but also illuminate their personal, social and physical worlds. Focus groups rather than individual interviews were chosen to explore the perspectives of homeless men in order to reduce the effect of social distance between the facilitator and participants, and to gain an insight into group talk and dynamics.²³

Focus groups were undertaken after lunch at an 80-bed Brisbane homeless men’s hostel on days that were convenient for the facilitator and hostel nurse. A recruitment flyer was displayed at the hostel medical centre reception on the day of a group, headed ‘Are you interested in talking to researchers about your experience of healthcare systems and services?’ Participants who indicated their interest to the hostel nurse were given written participant information and invited to attend. Clients having lunch in the hostel on the day of a group were also approached individually face to face and invited by the hostel nurse to attend, and given participant information if interested. Sampling was driven by convenience and was non-purposive. Potential participants who appeared to be intoxicated, agitated, aggressive, over-sedated, unusually socially withdrawn or acutely psychotic at the time of recruitment were not approached. No record was kept of clients who were approached but refused to participate, although the hostel nurse reported that recruitment was not difficult, and the gift voucher appeared to encourage participation. (Participants’ time was compensated by a A$50 gift voucher, distributed to each participant at the beginning of the group.) No participants who consented to participate subsequently withdrew.

Participants were informed that participation would not affect in any way the health or social care offered or provided within or outside the hostel. Written information and facilitator assurance were provided regarding confidentiality, deidentification, freedom to withdraw and absence of pressure to share any sensitive or personal information. Three participants with low literacy levels were assisted by the facilitator with written participant information, consent and demographic forms.

There were four participants in each focus group. Clients were invited to talk about their experiences of healthcare services and offer suggestions for improvement, but were advised that the health care provided at the hostel was not the focus of the research. This did not appear to cause any difficulties, and these services were not discussed specifically in any groups.

All focus groups were facilitated by one of the authors (NS), an academic GP with previous experience of focus group facilitation and >10 years clinical experience of mental health and substance use issues. NS had also provided part-time general practice services to the hostel at the time of the focus groups, and approximately two-thirds of the participants had used these services. The hostel nurse was present for some time during all focus groups, but left to attend to other hostel matters as needed. Two groups were interrupted by a call for medical attention elsewhere in the hostel, resulting in the facilitator absenting herself from the group and the participants dispersing. No attempt was made to reconvene these groups.

The focus group guide is given in Box 1. The facilitator took brief handwritten notes during the groups and the sessions were audio recorded, professionally transcribed and coded using ATLAS.ti software (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany). Transcripts and findings were not returned or discussed further with participants.

The transcript data were initially inductively coded by one of the authors (NS) using descriptive and in vivo coding.²⁴ Both authors discussed, compared and contrasted focus group transcripts and/or verbatim excerpts when considering categories and themes to ensure that these were grounded in the data. Categories were derived inductively. The coding tree for the theme dismissive care, for example, included the categories ‘hope they take it seriously’, ‘turned away’, ‘bung it on’, ‘we know what we need’, ‘not believed’, ‘touching up’ and ‘classed’. The authors reached a consensus on conceptual categories and key themes.²⁴ Discussion of participant experiences of general practice and out-patient services was particularly encouraged in the final two groups, after it was noted during preliminary analysis that most discussion was centred on emergency hospital presentations. No new themes emerged in the fifth group.

The study had full ethics approval from the University of Queensland Human Research Ethics Committee (2017001368). Gatekeeper approval was obtained from the coordinator of the Ozcare Men’s Homeless Hostel. No unanticipated ethical issues were experienced during the focus groups.

Five focus groups were undertaken between October 2017 and September 2018 with a total of 20 male clients, 17 of whom provided demographic information. Client ages ranged from 28 to 58 years (mean age 43 years). Mean group duration was 38 min (range 19–53 min). Further participant characteristics are provided in Table 1.

Table 1.  Participant characteristics (n = 17)
ED, emergency department; GPs, general practitioners

The facilitator’s impression of focus group participants was that their engagement was sincere and generous, with most participants appearing relatively relaxed and confident. Care was taken not to pressure any participants to contribute, although the facilitator offered more reticent participants the opportunity to talk by asking whether they would like to add anything. All participants appeared to respect each other’s input, and often agreed verbally or non-verbally with the current speaker.

Participants in all groups expressed gratitude for health care provided in both hospital and community sectors for life-threatening physical illness and trauma:

And at the moment I’ve got prostate cancer. I’m there every day having radiotherapy and that, and they’re all perfect…I’m lucky to be here in Australia really. (FG1P4)

So I’m still here, they got me through, so I’ve got nothing bad to say. (FG4P1)

When I was asleep, she’d just go and rub that foot for me, I think, and I would just wake up, say ‘Thank you’, and that, and I’d fall asleep. Sometimes I’d wake up and just talk to her and that. (FG5P1, speaking of a nurse at the time of a prolonged hospital admission in his adolescence)

However, negative experiences with health care were more commonly reported. These are presented below under the following four core themes: dismissive care, care fragmentation, inconsistent medical management of pain and inadequate acknowledgement of psychological distress. These four themes all relate to the overarching concept of difficulty securing an effective ‘ticket of entry’ to health care (see Fig. 1).

Fig. 1.  The vicious cycle of efforts to obtain health care of patients with trimorbidity and social exclusion.

Several participants in all groups reported experiences of healthcare providers discounting their need for help:

And I was classed as, like, basically, you’re probably off your face and you don’t need help. (FG1P1)

Some participants reported reacting to these experiences by disengaging with services:

P1: I gave them their chance when I needed them to speak to me.

P3: Yeah.

P1: So why the hell should I cringe now and speak to them? They turned me away, I didn’t turn them away. (FG1)

Waiting times of several hours in emergency departments (EDs) were commonly reported, and often experienced as discriminatory, although a few participants acknowledged workload pressures, including the need to prioritise critically ill patients and a lack of in-patient beds:

I do understand, like, sometimes – sometimes they’re busy…that’s just the luck of the draw, you know, you never know what you’re getting when you walk into an emergency room. (FG1P1)

In one group, caring and compassion were explicitly discussed:

If something serious happens and they want to take care of the most serious issues first. That’s – that, I totally agree with that. But, um, I think a lot of people walk into that – those, sort of, situations and at the end of the day they just want someone to care about them…whether you walk in with a belly ache or you – you’ve got a life changing injury, you still want some level of compassion. (FG5P2)

Participants in two groups reported physical assault by ED security staff:

P3: the next minute they’ve just tackled me to the ground and, sort of, given me a bit of a, you know…

P1: Touching you up.

P3: …punch in the ribs and – and so forth. (FG1)

In two groups participants reported that inaccurate, outdated or prejudicial information in medical records (e.g. previous drug use or confrontational incidents) appeared to compromise their current care. Participants in all groups reported avoiding medical care unless their situation was critical:

I don’t really trust hospitals or doctors unless I’m really dying. (FG4P4)

In all groups some participants described fragmented, even chaotic, care:

Sometimes people ask me, so where are you being seen and who’s seeing you, and I can’t remember because there’s so many different people involved. Um, I haven’t got a clue for most of them, I don’t know which hospital treats me for my legs and which one treats me for my arms, and – don’t remember. (FG3P3)

One participant discussed missing follow-up appointments despite prior intentions of attending, because of falling deeply asleep after many days of insomnia. Participants in two groups reported positive experiences of case workers who were ‘a single point of contact’, ‘someone who can vouch for you’ and/or someone to wake them for, and transport them to, appointments.

Participants tended to talk more about ED than out-patient or general practice experiences, but several commented that general practice experiences improve once they ‘get to know the doctor’:

So you walk in there and you automatically feel comfortable and relaxed and, like you’re going to get help and you’re going back to the same doctor. (FG5P2)

However, participant transience made this continuity difficult to achieve:

P2: Every time you see another doctor you got to start from the beginning again…

P1: They start from scratch again, that’s right. So you end up running around and chasing your tail in circles.

P2: Even though you get – say, okay I’ll sign the paperwork to release my documents to you, you’re still chasing your tail, because nine times out of 10 them documents don’t turn up. (FG4)

Although most participants agreed that comprehensive medical records should be readily available, some were suspicious of electronic information:

Throw all the technology out the window and, you know, sit there and talk to somebody. (FG5P2)

Participants in four groups described inconsistent prescribing for pain. One participant described ‘walking through an imaginary door’ and having his painkillers confiscated during his mental health admission from the ED:

And they asked me, ‘Oh, are you on medication?’ I said, ‘Oh, yeah, I’ve – you guys have just written me out a prescription next door for oxycodone.’ And they said, ‘Oh we can’t give you that, that’s a narcotic.’ And I thought, well, that’s great because I’m in pain and you can’t give me anything, but I’m glad I got a joint in me pocket. (FG4P1)

General practitioners (GPs) were also reported to have inconsistent approaches to pain management, ranging from liberal prescribing on patient request to a blanket refusal to prescribe ‘anything strong’. Several participants appeared to be bemused by this inconsistency. A few other participants commented that GPs could be worried about previous addiction issues relapsing, and that it could be difficult for GPs to know whether to trust patients. Participants reported turning to the ambulance service or ‘mates’ to obtain pain relief that was not forthcoming elsewhere:

We’ll grasp at straws, you know, when the pain is so bad. (FG3P3)

Several participants recommended marijuana for pain relief.

Participants in four groups described difficulty communicating with mental health services, sometimes presenting only as a last resort, or involuntarily, and harbouring misgivings and resentment based on previous negative experiences:

It’s just like they didn’t believe a word I was saying. Um, I really thought I needed to be admitted, I didn’t feel safe…there’s been times where I wanted to go in there, like I knew I needed some sort of help, but I was a bit reluctant because they might do the same thing, just not worry about it…Like, my father took his own life and, um, yeah, he didn’t really get the help he needed. (FG3P2)

Several participants reported ‘clamming up’, having to ‘swallow your pride a little bit as a man’ and other difficulties in communicating their distress, including having to ‘prove’ that they were very unwell:

P4: Don’t make people think that they have to go to ED and prove that they’re suffering from mental health, like, you know, they have to, like, act mental…

P2: Yeah. But that’s what you have to do…and how do you prove that you’re losing your mind?

P4: Then some guy completely fried out of his brain comes in and because he’s acting like an idiot…

P2: He’s in.

P4: …oh, yeah, get him in…and then you think, well, so what, so now do I have to, like, bung it on a bit? (FG1)

Participants were sometimes bewildered by experiences they perceived as rejection:

It’s just the hospitals, they won’t touch me. Especially mental health won’t touch me. I don’t know why, they won’t touch me. You go there for help and they just – they just send me home, say nothing’s wrong with me. (FG4P4)

One participant described being persistent and presenting to several different hospitals to secure admission:

It gets to that point, you know, like, where you can’t function in life…it becomes, I need to see a doctor and I just hope they take it seriously, and if they don’t, I just go to the next hospital…something has to be done. (FG1P4)

Another participant reported that medication prescribed in an acute care mental health setting was unavailable in the community:

…they give you access to medications that you can’t get on a regular basis and that you can’t get in your day to day life. And then they go, ‘Well, look…he’s fine, he’s good to go.’ But then you get out you can’t get that same medication and, like, the symptoms start up again. (FG1P4)

Here we present some insider perspectives on the Australian healthcare system from one of its most vulnerable patient groups. New findings include participants’ difficulty identifying effective ‘tickets of entry’ to compassionate and integrated health care and consistent medical management, for pain and psychological distress in particular. Participants seemed unclear how to ‘prove’ that they ‘needed help’. They appeared bewildered when their presentations with pain and/or overwhelming feelings of psychological distress appeared to be dismissed or discounted, and medication denied. There appears to be an almost ‘cross-cultural’ mismatch between mental health services’ responses in these situations and participant expectations of assistance. People with severe and persistent mental illness are known to experience the escalation of their symptoms as overwhelming,²⁵ and peers from a similar ‘cultural’ background may assist with the effective communication of anxiety, agitation and despair. It is important to avoid repeated negative experiences for both patients and providers triggering vicious cycles of confrontation, disengagement and/or dismissive behaviours, entrenching mutual distrust, avoidance and stigma (see Fig. 1).

The strengths of this study include the naturalistic data, the particularly rich and compelling participant voices provided by the focus groups²³ and the focus on health care delivered external to the service hosting the research. However, recruitment was confined to adult men in one inner city hostel in Queensland. Different or additional key themes may be found with female or younger participants, or with other marginalised groups. The facilitator’s role as one of the hostel’s visiting GPs may have limited discussion of general practice experiences.

The international literature also includes reports by homeless people and injecting drug users of rude, dismissive and/or unwelcoming health services, experiences of discrimination and stigma^11,12 and perceptions of being judged, labelled and rejected.^12,16 However, the participants in the present study were also grateful for health care received for life-threatening physical trauma and illness, and acknowledged pressures and constraints on healthcare systems, as well as difficulties for healthcare providers caring for substance-using clients. Nevertheless, accounts of apparently dismissive and stigmatising health care were compelling, most saliently in EDs, but also in primary care. Reports of potentially avoidable physical violence suggest a need to improve training, support and monitoring of security staff in showing compassion and de-escalating agitation.

It is also likely that healthcare systems and providers are challenged conceptually, clinically and administratively in managing clients with trimorbidity adequately. Clinical challenges include, for example, the influence of substance use on mental health symptoms and signs, and the interweaving of pain, anxiety and psychological distress with substance withdrawal, emotional and physical trauma, mental and physical illness, social isolation and exclusion, and problems with addiction. Other clinical challenges include the risk of medication that offers short-term symptom relief exacerbating or reactivating existing substance use disorders. There is a lack of conceptual clarity about how best to operationalise ‘harm minimisation’ approaches to patients with substance use disorders in remission or recovery, and use opiate, benzodiazepine (and, to some extent, antipsychotic and other psychoactive) medication for patients with trimorbidity. This is contributing to management perceived by our participants to be dismissive, inadequate, irresponsible or even capricious. This may be a particular problem as patients transition through healthcare system boundaries, such as from in-patient to community care. Collaborative efforts across the sector are needed to develop consistent approaches, and avoid both under- and overtreatment of pain, anxiety and psychotic symptoms, including self-medication with substances resulting in drug overdose deaths.^26,27

Our findings also support ongoing efforts to address the challenge of fragmentation of care and services. Patient registration should be considered in order to support culturally sensitive general practices and hospitals to build trusting relationships over time in order to defragment care. Inaccurate or non-pertinent information in medical records, including those of people in recovery or remission from injecting substance use disorders, should be removed or unflagged so as not to further entrench stigma. Improved electronic medical record systems have the potential to address the major challenges that patient transience and instability pose for effective transfer, handover and follow-up of care, provided that they do not result in further discrimination.

There is work to be done across the sector (including general practice, emergency care, mental health and public health) to collaboratively develop compassionate, consistent and integrated approaches to marginalised patients with trimorbidity. It is important that healthcare professionals are alert to the risks of further traumatising and stigmatising vulnerable people while attempting to provide, or discharge patients from, health care.²⁸ Partnerships with people with lived experience of trimorbidity, including men experiencing homelessness, may enable the identification of better solutions. This study presents the voices and perspectives of homeless men, with the aim of contributing to this important and difficult work.

No competing interests to declare.