Abstract
Objective
To examine if there is a relationship between introducing a Homeless Health Outreach Team (HHOT) and reduced acute Mental Health service usage.
Methods
Electronic Medical Record (EMR) data were collected on a group of clients of the Tweed Byron HHOT 6 months before and after establishing the service. The data were evaluated for demographics, and differences in Emergency Department (ED) presentations, mental health admissions, length of stay and community mental health engagement.
Results
The introduction of the team coincided with a significant reduction in ED presentations and an increase in community mental health engagement. There was an overall reduction in bed days but an increase in mean length of stay for those admitted post-intervention.
Conclusions
The establishment of the HHOT coincided with reduced acute mental health service usage via ED and inpatient Mental Health Units (MHU). There is scope for expansion of such a service as well as exploration of costings analysis. A long-term focus on ‘housing first’ and outreach approaches to homeless service provision could improve individual and service provision outcomes.
Homeless persons are among the most disenfranchised groups in our community, and numbers across Australia continue to rise. In the Northern New South Wales (NNSW) area of Tweed and Byron Bay, added complexities such as stretched rural service provision across large regions, limited temporary and long-term accommodation providers, short-term letting reducing long-term rental stock, and recent flooding disasters have only served to exacerbate the issue. While ‘rough sleeping’ is an omnipresent and increasingly visible occurrence in places like Byron Bay, 1 it is generally accepted that homelessness also includes housing that falls below minimal community standards, such as unstable, unsafe or insecure accommodation, for example, ‘Couchsurfing’ with family or friends, or long-term boarding houses. 2
International data suggests ‘housing first’ approaches, such as those adopted in Finland, 3 and assertive outreach are two evidenced based models that can improve long-term health and social outcomes. Health services delivered under primary care, early intervention community models rather than through crisis management such as the hospital ED and police and ambulance have been shown to reduce overall government costs. 4 A Homeless Health Outreach Team (HHOT) was recently established in NNSW in the Tweed and Byron areas to improve engagement with local service providers, reduce the reliance on emergency services and improve the quality of life in this disenfranchised community.
Characteristics unique to the area
NNSW is unique in several ways: The temperate climate provides relief from the colder winters of southern capitals, and a traditionally inclusive community, born out of an initial wave of migrants in the 1960s and 1970s, has been open-minded to alternative lifestyles. Many travellers and those living on the fringe are attracted to Byron’s rural and relaxed lifestyle or stay on while moving between Gold Coast and Brisbane to the north and Sydney to the south.
However, according to the Byron Shire Mayor, the area’s reputation as a holiday destination and subsequent proliferation of private short-term rental accommodation (STRA) has ‘gutted’ the local community. 5 Comparison rates of private rental advertisements in the area equate to one STRA for every ten residents, four times the rate in Central Sydney. 6 As these homes are permanently removed from local housing stock, accommodation squeezing, increased demand and overcrowding eventually flow downhill to those already on the verge of homelessness.
As such, NNSW has some of the highest rates of homelessness in the state, if not Australia-wide. Census data from 2016 revealed an increase in homelessness of 44% between 2011 and 2016 in Tweed/Byron. 7 However, when explored in greater detail, this data also reveals the severity of the issue, with minimal to no supported or temporary accommodation, boarding houses, nor refuges, resulting in an overrepresentation of rough sleeping. 21% of NSW’s rough sleepers reside in NNSW despite only representing less than 4% of the state’s total population. Similarly, rough sleepers represent almost a quarter of the NNSW homeless population, compared to less than 5% in Greater Sydney.8,9
Homeless persons in Australia are less likely to access health services, are over-represented in morbidity and mortality statistics and have a life expectancy up to 30 years shorter than the general population. 10 Mental Health concerns are identified as a leading cause of homelessness, resulting in a reliance on already stretched EDs rather than accessing mainstream community services. 11 Locally, unpublished data revealed that in the financial year 2019, of the 1384 mental health presentations to Tweed and Byron EDs, 10% were recorded by staff as homeless, with a strong possibility of under-reporting due to lack of recognition. A further snapshot of mental health referrals from Byron Bay ED to Byron Bay Mental Health Acute Care Service in October 2018 revealed that 38% of referrals were homeless. 12
Establishment of HHOT
A small HHOT was established in late 2020, funded by a NSW Health support grant to assist vulnerable people in the wake of the COVID-19 pandemic. The team includes a psychiatry trainee, two social workers, two nurses and one part-time psychiatrist. The model aims to support people experiencing (or at risk of) homelessness to actively engage in healthcare and mainstream service provision, to improve their health and well-being, with a primary focus on Mental Health and Drug and Alcohol care. Clinicians provide street-based outreach in partnership with local Specialist Homelessness Services (SHS) and regular clinics at local Community Managed Organisations (CMO). Collaboration with clients, local service providers and communities to build and strengthen client support networks and assist clients onto a pathway out of homelessness through the development of a person-centred care plan is at the core of the HHOT model of care.
Method
This study measured the effect on unplanned service usage, such as ED presentations and mental health admissions, by homeless people engaged with HHOT before and after introducing the HHOT. A retrospective pre/post-design collected data over 6 months before an individual engaged with HHOT and compared it with service use data of the same persons in the 6 months post-HHOT engagement. The 6 months pre-period was between July and December 2020, and the 6 months post-period from January to June 2021. Participants’ demographic data, such as age, sex and the diagnosis, were also collected.
Medical records of 43 persons who were clients of the HHOT (n = 43) were accessed from the hospital medical records system, which covers presentations to all hospitals within NNSW. Data was analysed via the statistical package SPSS, and a Wilcoxon Matched Pairs Signed Ranked test was used to determine if there was a difference between before and after service use for all 43 clients. Different individuals were hospitalised in the pre–post periods, and a two-tailed T-test using summary data was used to analyse the change in mean length of stay.
Ethics approval
The Northern NSW Local Health District Research Office Ethics Committee reviewed and approved the project. Subsequently, approval was given to access medical records and collect de-identified data of individuals. (Number QA394).
Results
Demographics
There were 31 males and 12 females aged between 26 and 65, with a mean age of 45 years (SD 10.7). Chronic psychotic illnesses predominated, as over half the sample had a diagnosis of schizophrenia (n = 18), schizo-affective disorder (5), delusional disorder (3), psychosis not otherwise specified (2) or drug-induced psychosis (2). The remaining individuals had a primary or secondary mental health or drug and alcohol diagnosis. Primary or comorbid substance use disorders were identified in over two-thirds (29, 67%), and over half of these (16, 37%) used crystal methamphetamine regularly. Over half the population (26, 60%) were sleeping rough at the commencement of engagement with HHOT, with five sleeping in their cars. The remainder were residing in unstable, unsuitable, or temporary accommodation.
Service engagement and usage
A statistically significant relationship (p < .001) was found between the introduction of HHOT and reduced presentations to ED for mental health reasons by this group. Presentations dropped from 50 in the 6 months before only 20 in the 6 months post. Unsurprisingly, the cohort’s community mental health engagement increased significantly, as previously isolated individuals were now engaging with the HHOT or linked in with local community mental health teams (p < .001).
Pre- and post-intervention comparison
Discussion
The study demonstrated that a service such as HHOT can engage homeless individuals almost three times as much as traditional community mental health services and reduce ED presentations.
This study also highlights the high prevalence of substance use disorders and chronic psychotic illnesses among the homeless population in NNSW engaged with HHOT, which appears consistent with other homeless research in larger cities such as Sydney and may perpetuate homelessness. 13
This study also demonstrates the introduction of HHOT coincided with a reduced number, yet on average longer admissions to local MHUs. Our team initially hypothesised engagement with HHOT would lead to a reduction in bed days and a decrease in the average length of stay as acute inpatient teams were more likely to discharge homeless individuals sooner, knowing HHOT would engage in assertive follow-up. However, the opposite occurred, the mean length of stay increased, and short admissions of less than a week almost disappeared (1, 12%, compared to 8 50% in the pre-data set). These results suggest that introducing a HHOT allows for community-based crisis engagement and intervention that previously would have occurred in the hospital, limiting admissions to those acutely unwell. Anecdotally, the threshold for admission while engaged with HHOT was typically high or with a view to longer rehabilitation admission.
Based on the cost of non-metropolitan mental health beds in NSW ($ 4801 at the time of writing), 14 a saving of $268,856 was made on bed usage during the 6-month post-intervention period. Additionally, reductions in total ED presentations by this cohort represented a saving of $20,400. The implication of the potential savings and benefit to homeless service users are relevant even if some findings did not reach statistical significance.
Given the limitations in social and affordable housing throughout NSW, this study focused on service engagement rather than stable accommodation as a primary end goal. There is the future scope for research to look at pathways out of homelessness as service users engage with HHOT and are linked with housing providers and SHSs.
Limitations
Unfortunately, this small study had multiple limitations. The most significant of which is somewhat inherent in the population group and its itinerancy. Due to EMR limitations, we had no way of counting hospitals presentations outside the Local Health District (LHD); this may be particularly important given the Tweed Shire borders with Queensland. The impact of the COVID-19 pandemic may have also been a confounding factor, as total ED presentations varied secondary to lockdown measures and covid trends. However, local data shows total mental health ED presentations varied by less than 2% in the pre–post periods. 15 A more extensive data set or xtended examination period may have revealed more statistically significant results. It is essential to acknowledge the limitations of this pre/post study design (particularly given the potential natural seasonal variations of only a 6-month pre/post period) and other confounding variables, such as that described above that may account for these outcomes.
Conclusion and implications
Long-term overseas data suggests that assertive outreach and a ‘housing first’ approach to homelessness are two of the most effective evidence-based approaches to improving long-term outcomes in this marginalised population group.16,17 With repeated governmental failures in providing good affordable and supported housing options, specialised assertive outreach, such as that offered by HHOTs throughout Australia, can at least attempt to actively engage these individuals and divert service provision away from acute services. This may lead to improved outcomes individually and cost-savings for service providers.
Footnotes
Disclosure
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethics Approval
The project was reviewed and approved by the Northern NSW Local Health District Research Office under a non-Human Research Ethics Committee pathway for low-risk projects and deemed to be consistent with the principles and values of the National Statement on Ethical Conduct in Human Research (2007). The approval number is QA394. All clients had agreed to engage with HHOT voluntarily and patient information was de-identified.
