Caution: If at any point you feel worried about harming yourself while viewing the information on this website – or if you think someone else may be in danger – please stop reading and seek help.

Help and support

Suicide among people receiving specialist homelessness services: A last year of life study

Key findings

Among people who received Specialist Homelessness Services (SHS) between 2012–13 and 2021–22 and in their last year of life:

there were around 1,600 suicide deaths, representing over 1 in 10 (13%) of all SHS client deaths over the decade
the crude suicide rate was 115.1 per 100,000 population
around a third of SHS clients who died by suicide received support within their last month of life
people aged 25–44 represented over half (55%) of all suicide deaths
over half (55%) who died by suicide were unemployed
nearly 1 in 5 (19%) who died by suicide experienced rough sleeping, and over 1 in 3 (38%) experienced ‘other homelessness’ (temporary or unstable accommodation).

Suicide risk among people experiencing or at risk of homelessness

Homelessness and housing insecurity are associated with an increased risk of suicidal thoughts and behaviours (Brackertz 2020). Numerous studies have consistently shown that people experiencing homelessness face significantly higher rates of suicidal ideation and attempts compared to the general population (Ayano et al. 2019). For example, a Queensland study found that between 1990 and 2009, the suicide rate among homeless individuals was 27.6 per 100,000 – almost double that of those who were not homeless (Arnautovska 2014).

A report prepared by the Australian Housing and Urban Research Institute highlights three main risk factors linking housing instability and suicide: prolonged financial stress caused by high housing costs, insecurity from eviction or homelessness, and the long-term impact of adverse childhood events on mental health. Additionally, factors such as unemployment, undiagnosed mental illness, and legal issues were strongly associated with suicide risk among homeless individuals (Brackertz 2020).

The National Mental Health and Suicide Prevention Agreement (Australian Government 2022) identifies people experiencing or at risk of homelessness as a priority population for suicide prevention. Thus, data on suicide and the circumstances surrounding death among this population is important for informing targeted prevention activities (Henkind et al. 2023).

Data sources and methods

This article draws on data from the NACS linked dataset, focusing on individuals who died by suicide and received Specialist Homelessness Services (SHS) support in their last year of life. The NACS linked dataset includes:

National Deaths Index (NDI)
Alcohol and Other Drug Treatment Services National Minimum Data Set (AODTS-NMDS)
Commonwealth funded health services (MBS and PBS)
Specialist Homelessness Services Collection (SHSC).

For more information about the linkage process, and details of the dataset years included in the NACS, see the technical notes.

Specialist Homelessness Services in Australia

SHS play a pivotal role in supporting people who are homeless or at risk of homelessness. These agencies provide a wide range of services, including general support, crisis accommodation, and targeted assistance for specific groups such as young people or those affected by domestic and family violence. It is important to note that while SHS engagement data reflects the reach of these services, it does not fully capture the underlying prevalence of homelessness or housing instability in Australia. For more information see Homelessness services.

How were suicides among SHS clients counted?

Deaths by suicide were identified using the National Deaths Index (NDI), linked to the Specialist Homelessness Services Collection within the NACS linked dataset. An individual was counted where their last SHS support period concluded within 12 months of their death (including those who died during their support period). For simplicity, the term ‘SHS clients’ is used throughout the report, and the period 2012–2022 is referred to as the ‘study period,’ with data also analysed by financial year.

Please refer to the data sources and technical notes pages for more detailed information on data linkage and coverage for these data.

See People receiving specialist homelessness services support in last year of life for a detailed report all-cause mortality among people who accessed SHS in their last year of life.

Suicide among people receiving SHS support in last year of life

Around 12,500 people received SHS support in their last year of life over the study period – around 7,800 males and around 4,700 females. Among these people, suicide was the second leading cause of death, following accidental poisoning (AIHW 2024a).

Among people receiving SHS support within their last year of life, between 2012–13 and 2021–22:

There were around 1,600 suicide deaths, representing over 1 in 10 (13%) of all deaths over the decade.
The crude suicide rate was 2.3 times as high for males (173 per 100,000 people) than for females (75 per 100,000). However, the proportion of suicides relative to all deaths was similar for both sexes at around 13%.
Overall, the crude suicide rate and number of suicide deaths increased for females. A similar, though smaller, increase was observed among males.

Time between last homelessness service and death

Around a third of SHS clients received support within their last month of life

Among those who died by suicide and received SHS support in their last year of life, a large proportion (37%) had their final interaction with these services within a month of their death, with almost 1 in 4 (24%) having an open support period the same day they died. Another third had their last support period end 2–5 months before death (31%), while a similar proportion (32%) last received support 6–12 months before dying by suicide.

Overall, proportions for males and females were very similar. Compared with males, a higher proportion of female SHS clients were receiving support or had their support end on the day of their death. Males, however, had a higher proportion of suicides occurring within 1–30 days (not including the day of death) of their last SHS interaction compared to females. It is important to note, however, that SHS data do not indicate whether the SHS agency had direct contact with the individual on the day they died. In some cases, the SHS agency may have closed the support period after learning of the individual’s death.

The close proximity between engagement with SHS and suicide may indicate a period of acute crisis, highlighting SHS as potential intervention points for suicide prevention. For instance, a study in the United States found that people who had experienced homelessness in the past year were more likely to report attempting suicide in that past year, compared to those who were homeless prior to the past year or were never homeless (Bommersbach et al. 2020). Furthermore, analysis of the NACS dataset shows that individuals who received SHS support closer to their date of death tended to have longer durations of support (for example, a higher number of support days), compared to those whose services occurred further from their death. This finding highlights the potential relationship between service use intensity and death by suicide.

Considering the above, future analysis will focus on teasing apart the various components of service use intensity, developing more robust measures to better define people with complex circumstances and how their experience of SHS support might relate to time of death.

Age at death

Younger age groups accounted for more suicide deaths

Among people who received SHS support in their last year of life, younger age groups (less than 45 years) had a higher proportion of suicide deaths compared to other causes of death than older age groups. This aligns with broader patterns in the leading underlying causes of death, where chronic diseases become more prominent for those aged 45 and over (AIHW 2024b).

Over the study period, among those who received SHS support in their last year of life:

The 25–34 and 35–44 age groups had the highest number of suicide deaths (440 and 435, respectively). These age groups combined represented over half (55%) of all suicide deaths.
Around 2 in 5 (39%) of all deaths from any cause among SHS clients aged 0–24 years were due to suicide, the highest proportion of all age groups.

It is important to note that the age profile of people who died after/while receiving SHS support partly reflects the demographics of the SHS client population, which is younger than the broader Australian population. In 2022–23, the median age of SHS clients was approximately 30 years (derived from AIHW 2025), compared to around 38 years for the Australian general population in the same year (ABS 2024).

Labour force status among SHS clients who died by suicide

Measuring labour force status

Labour force data is collected from SHS clients aged 15 years and older. In the NACS dataset, a client's labour force status is recorded at 3 different time points: the week before presentation, at presentation and at the end of the support period. Examining labour force status across these time points provides insights into how individuals may transition between different statuses (such as, from unemployment to employment).

Labour force status is categorised as follows:

Unemployed: individuals not employed who have actively sought full-time or part-time work during the four weeks prior to the end of the reference week and are currently available to work, or those waiting to start a new job within the next four weeks but are available to work if a job were offered.
Employed: individuals who worked at least one hour for pay or were temporarily absent from work for less than four weeks prior to the end of the reference week.
Not in the labour force: individuals engaged in unpaid household duties or other voluntary work, those retired, voluntarily inactive, or permanently unable to work during the reference week.

Over half of SHS clients who died by suicide were unemployed

Labour force status, such as unemployment or not being in the labour force, is strongly associated with an increased risk of suicide. According to a previous AIHW analysis (2021), unemployment was linked to a 75% higher risk of suicide compared to being employed, while not being in the labour force was associated with an 80% higher risk, even after adjusting for other risk factors and demographics (see Regression risk models for selected census variables). In 2023–24, over half (51%) of SHS clients were unemployed, and nearly one-third (32%) were not in the labour force, according to the latest annual report (AIHW 2025).

Among those who received SHS support in their last year of life and died by suicide during the study period (2012–13 to 2021–22), at the time of their SHS presentation:

Over 1 in 2 (55%) reported being unemployed, with a higher proportion of males (59%) compared to females (48%) reported unemployment.
The 15–24 age group had the highest proportion of unemployment for both males (71%) and females (53%). Youth unemployment is a leading risk factor for homelessness, driven by factors such as family conflict, limited income opportunities, and a lack of affordable accommodation (Roche & Barker 2017).
Unemployment decreased as age increased, while the proportion of those ‘not in the labour force’ rose with age.
Employment was relatively uncommon. The highest proportion of people reporting employment was among females aged 15–24 (12%). The lowest proportions were among females aged 25–34 and males aged 45 and over (both 5.5%).

Experience or risk of homelessness in last year of life

Measuring homelessness status

All SHS clients are either homeless or at risk of homelessness. For each SHS support period in the NACS data, a client's housing situation is recorded at 3 different time points: the week before presentation, at presentation and at the end of support. Using these variables, a client’s housing status was derived using a cascading hierarchy into following categories:

Rough sleeping: any experience of rough sleeping such as no shelter, improvised dwelling (such as, park, motor vehicle, tent or improvised building/dwelling), or renting/living rent-free in a caravan park.
Other homelessness: experiences of homelessness other than rough sleeping, such as couch surfing or staying in short-term or temporary accommodation.
Not homeless: housing situations where clients were at risk of homelessness but not classified as homeless.
Not stated: housing data were missing or unknown.

It is important to note that some individuals may have experienced periods of homelessness outside the times when they were receiving support from SHS agencies. As a result, the data presented here may underestimate the number of SHS clients who experienced homelessness in the year prior to suicide.

For more information, please see the data sources page.

People who are rough sleeping are the most vulnerable and visible of people experiencing homelessness. They face considerable challenges, including fear of violence and theft. SHS services provide a range of information and support to assist people experiencing rough sleeping into more suitable accommodation (AIHW 2024d). For a detailed analysis of characteristics and service use patterns of rough sleeping SHS clients, see Specialist homelessness services client pathways: Rough sleeping clients in 2016–18.

Nearly 1 in 5 SHS clients who died by suicide experienced rough sleeping

During the study period, among those who received SHS support in their last year of life and died by suicide:

Nearly 1 in 5 (19%) experienced rough sleeping. This was slightly lower than those who died from any cause of death (24%) (AIHW 2024a). The proportion of SHS clients experiencing rough sleeping was similar across age groups, with those aged 35–44 having the highest proportion (23%).
Rough sleeping was more common among males, with almost 1 in 4 males (24%) experiencing rough sleeping compared to 1 in 10 females (10%).
Over one-third (38%) of individuals were experiencing ‘other homelessness’, which includes those in short-term temporary accommodation or couch surfing.
Males aged 35–44 had the highest proportion of experiencing rough sleeping (31%).
The proportion of people in the ‘not homeless’ category increased steadily with age, while the proportion of those classified as ‘other homeless’ generally decreased with age. This pattern may reflect the underlying housing situation of clients, where being housed is more common among older people receiving SHS services compared to younger age groups (AIHW 2024d).
Females were more likely than males to be in the ‘not homeless’ category (44% compared to 30%).

Future work

This descriptive study is the first analysis exploring specialist service use prior to death by suicide. Future analysis on SHS clients may include:

risk of death among SHS clients who are unemployed and/or have a change in employment status over their support period
interaction of SHS clients with Alcohol and Other Drug Treatment Services
use of mental health services and treatment among SHS clients.

Future research will expand to include other AIHW-funded NACS datasets, including those focused on people receiving drug and/or alcohol treatment. Additionally, further analysis will explore a broader range of risk factors and patterns of health service use, leveraging data from the MBS and PBS datasets.

Download data tables

Supplementary tables

Homelessness and suicide – Suicide among people using SHS support in last year of life

Download data

Amendments Expand

16 July 2025

In the section, Time between last homelessness service and death, 1st paragraph, the proportion of people who died by suicide and received an SHS service within a month of their death was incorrectly rounded to 38%. The correct proportion is 37%.
In the section, Labour force status among SHS clients who died by suicide, the 4th dot point incorrectly stated that ‘in all other age groups, a higher proportion of males reported being employed compared to females’. This dot point has been changed to: ‘The highest proportion of people reporting employment was among females aged 15–24 (12%). The lowest proportions were among females aged 25–34 and males aged 45 and over (both 5.5%).’
In the section, Experience or risk of homelessness, 2 sentences were removed because the cited statistics did not appear in the published versions of the source material:
- 1st paragraph:
  ‘Among SHS clients who died and accessed an SHS service in their last year of life during the study period, over 1 in 10 (11%) of people experiencing rough sleeping died by suicide (AIHW, 2024a).’
- 2nd dot point:
  ‘This is consistent with findings from a SHS report, which found 16.4 of male clients per 100,00 population experienced rough sleeping, compared to 11.9 for female clients – a rate almost 1.4 times as high (AIHW 2024d).’
In supplementary table NACS.5, the youngest age group was incorrectly labelled as 15–24. The correct label is 0–24.
Other non-substantive text edits were made to improve accuracy.

References Expand

Arnautovska U, Sveticic J and De Leo D (2014) 'What differentiates homeless persons who died by suicide from other suicides in Australia? A comparative analysis using a unique mortality register', Social Psychiatry and Psychiatric Epidemiology, 49(4): 583–9, https://doi.org/10.1007/s00127-013-0774-z

Australian Bureau of Statistics (ABS) (June 2024), National, state and territory population, ABS, Australian Government, accessed 11 February 2025.

Australian Government (2022) National Mental Health and Suicide Prevention Agreement, Federal Financial Relations, Australian Government, accessed 3 March 2023.

Australian Institute of Health and Welfare (AIHW) (2024a) People receiving specialist homelessness services support in the last year of life, AIHW, Australian Government, accessed 18 December 2024.

AIHW (2024b) Deaths in Australia, AIHW, Australian Government, accessed 18 December 2024.

AIHW (2024c) Specialist Homelessness Services: trends in clients rough sleeping, AIHW, Australian Government, accessed 10 January 2025.

AIHW (2024d) Specialist Homelessness Services Collection data cubes 2011–12 to 2023–24, AIHW, Australian Government, accessed 15 January 2025.

AIHW (2025). Specialist homelessness services annual report, accessed 11 June 2025.

Ayano G, Tsegay L, Abraha M and Yohannes K (2019) ‘Suicidal Ideation and Attempt among Homeless People: a Systematic Review and Meta-Analysis’, Psychiatr Q, 90(4):829–842, doi: 10.1007/s11126-019-09667-8. PMID: 31463733.

Bommersbach TJ, Stefanovics EA, Rhee TG, Tsai J and Rosenheck RA (2020) Suicide attempts and homelessness: timing of attempts among recently homeless, past homeless, and never homeless adults, Psychiatric Services, 71(12):1225-31, https://doi.org/10.1176/appi.ps.202000073

Brackertz N (2020) The role of housing insecurity and homelessness in suicidal behaviour and effective interventions to reduce suicidal thoughts and behaviours: a review of the evidence, Evidence Check prepared by Australian Housing and Urban Research Institute Limited for the National Suicide Prevention Adviser and the National Suicide Prevention Taskforce, commissioned through the Suicide Prevention Research Fund, managed by Suicide Prevention Australia, accessed 24 December 2024.

Hail-Jares K, Vichta-Ohlsen R and Nash C (2021) Safer inside? Comparing the experiences and risks faced by young people who couch-surf and sleep rough, Journal of Youth Studies, 16;24(3):305-22, https://doi.org/10.1080/13676261.2020.1727425.

Roche S and Barker J (2017) Youth homelessness and its relationship with family conflict: Models for policy and practice, Melbourne: Institute of Child Protection Studies, Australian Catholic University, and Melbourne City Mission, accessed 2 January 2024.