Homelessness can profoundly affect a person’s mental and physical health, their education and employment opportunities, and their ability to participate fully in social and community life. As such, homelessness is recognised as an important focus area for understanding the welfare of veterans, though until recently there has been limited evidence on homelessness among Australian veterans.

The Department of Veterans’ Affairs (DVA) commissioned the Australian Institute of Health and Welfare (AIHW) to investigate the use of homelessness services by clients reported in the AIHW’s Specialist Homelessness Services Collection (SHSC) who have served for at least 1 day in the Australian Defence Force (ADF) since 2001, described here as the contemporary ex-serving ADF population.

The results presented in this report are based on data generated by linking information from the Department of Defence staff and payroll management system—the Personnel Management Key Solution (PMKeyS)—with the AIHW’s SHSC. The cohort is defined by the matches between the 2 data sources, using a Statistical Linkage Key (SLK) methodology (see Technical Notes for more information). This means that the scope of the homelessness examined in this report is defined by the SHSC, and the scope of ex-serving ADF is defined by the PMKeyS. 

Defining homelessness

The SHSC contains information on individuals who use specialist homelessness services in Australia from a SHS agency (an agency providing SHS and receiving funding under the former National Affordable Housing Agreement or the National Partnership Agreement on Homelessness).

This means that the SHSC does not include information about all people in Australia who may need homelessness services, or all those who are homeless, only those who accessed SHS. This approach is different from other reported data on homelessness such as the Australian Bureau of Statistics (ABS) Census of Population and Housing, which measures homelessness based on an individual’s living arrangement, irrespective of whether they have used homelessness services (ABS 2016). Additionally, the ABS Census measures homelessness based on the living arrangement of the individual only on Census night every 5 years, whereas the SHSC contains data for the length of time clients received specialist homelessness services. SHS agencies also provide services to people who are at risk of becoming homeless, so not all SHS clients are experiencing homelessness. As a result, estimates of homelessness from the Census or other sources differ from those from the SHSC.

Homelessness and risk of homelessness

SHS agencies provide a range of services to assist those who are experiencing homelessness or who are at risk of homelessness. Clients are considered to be homeless if they are: living in non-conventional accommodation; sleeping rough; or living in short-term or emergency accommodation. Clients are considered to be at risk of homelessness if they are currently living in accommodation but are: at risk of losing their accommodation; or experiencing 1 or more factors that can contribute to homelessness. A small proportion of clients may also have their housing situation recorded as ‘not stated’ at the start of support if their housing situation is unknown.

Defining the ADF population

This report covers SHS use by ‘contemporary ex-serving ADF members’, which is defined as ADF members who had at least 1 day of service on or after 1 January 2001 (when the PMKeyS was introduced) and who discharged after this date. All data about ex-serving ADF members presented in this report refer to this cohort, including references to the broader contemporary ex-serving ADF population. 

Comparison with ADF indicator

An ADF indicator was introduced into the SHSC in July 2017. The ADF indicator records whether the SHS client self-reports as a current or former ADF member, either on a full or part-time basis. This indicator was introduced to provide a better understanding of the extent to which veterans may need support from SHS. 

The SHSC ADF indicator identified 1,295 current or former ADF members who had used SHS in 2017–18 (AIHW 2019). In comparison, between 2011–12 and 2016–17, analysis of linked data found 1,215 unique SHS clients who were ex-serving ADF members with service since 2001. Accounting for clients who used SHS in multiple years, the sum of total clients in each year across the 6-year period was 1,867 clients, or an average of 311 clients per year.

Important factors to consider (including more detail about the differences between the ADF indicator and the linked data used in this report) are provided in Box 1. 

Box 1: Factors to consider when reading this report

The scope of this report is restricted to data about use of homelessness services collected as part of the SHSC. The SHSC contains information on people seeking services from agencies that receive funding to provide SHS under the (former) National Affordable Housing Agreement or the National Partnership Agreement on Homelessness. This report does not include information on ex-serving ADF members who may have experienced homelessness but did not seek support from a SHS agency.

The scope of this report is restricted to ex-serving ADF SHS clients with at least 1 day of service since 1 January 2001, who discharged after 1 January 2001 and before 1 July 2017. As a result, this report cannot be used to estimate the use of homelessness services by ex-serving members who left the ADF before 1 January 2001.

The analysis in this report links the PMKeyS and SHSC data using a SLK. The SLK is a combination of letters and numbers taken from an individual’s personal information, which is used to match individuals across data sets. While the SLK methodology is a robust approach to match individuals between data sources, it may miss some matches or make incorrect matches that may not have occurred if the full identifiable information was used.

The scope of this report is restricted to SHS use that occurred between 1 July 2011 and 30 June 2017. As the contemporary ex-serving ADF population consists of ex-serving ADF members who discharged on or after 1 January 2001, the maximum length of time between discharge and SHS use is 16 years (that is, a person who discharged in 2001 and used SHS for the first time in 2017). As such, the analysis population is limited to those who used SHS within 16 years of their discharge. Any demographic or service characteristics associated with SHS use more than 16 years after discharge will not be captured in this report. In addition, demographic or service characteristics associated with SHS use many years after discharge may be under-represented.

The scope of this report is restricted to SHS use that occurred after an ADF member was discharged. Data for SHS use that occurs prior to the financial year in which an ex-serving ADF member was discharged are not included in this report.

Data on homelessness support provided before 1 July 2011 are not available, which means the total days of support may be underestimated for clients already receiving accommodation or other homelessness support on 1 July 2011. Similarly, some clients may have continued to receive SHS support beyond 30 June 2017, but this information is not within the scope of this report. The data for 2016–17 was the most recent financial year of data available at the time of data linkage for this report.

In this report, counts of clients are reported for each individual financial year, and across the 6-year reporting period (2011–12 to 2016–17). For some analyses, the total number of clients for each individual financial year will sum to greater than the total number of unique clients in the 6-year period as clients may receive several services over multiple years and can be counted in the total for each financial year in which they received SHS support. 

Ex-serving ADF SHS client counts are unweighted and under-estimate the number of clients. Unlike national SHS client data (such as that presented in the Specialist homelessness services annual report 2017–18), no imputation strategy has been applied to correct for missing or incomplete data.

Two different extracts of Australian SHS client data were sourced for the analysis for this report. The first is data from the Specialist homelessness services annual report 2017–18. These data have several key methodological differences compared with the ex-serving ADF SHS client data, such as the inclusion of clients under the age of 16; and the inclusion of clients in outcome data only if all of their support periods have closed by the end of the year. Because of this, data from this extract are included in the report to provide context to the ex-serving ADF results by showing patterns of SHS use in the Australian population, rather than for direct comparison. 

The second is the unweighted proportion of Australians aged 17 and over who had used SHS from 2011–12 to 2016–17. These data are only used to compare the level of SHS use among ex-serving ADF members with the level among the general Australian population. These data are unweighted (no imputation strategy has been applied to correct for missing or incomplete data), so these counts under-estimate the number of SHS clients. These data also represent a unique count of clients across 6 financial years, so the counts will not be comparable to data published for individual financial years. These data were sourced from a customised data request and are not published elsewhere.

There are differences in the methodology of how the analysis for the current report identifies ADF members, compared with the ADF indicator:

  • The ADF indicator includes veterans who discharged before 1 January 2001 and applies to currently serving or former ADF members. 
  • The ADF indicator is self-reported and will only capture ADF members who identify themselves as such to the SHS agency. This report identifies ex-serving ADF SHS clients by linking the SHSC data with the Department of Defence’s PMKeyS data. 
  • The ADF indicator is only asked of clients aged 18 and over, whereas this report contains ex-serving ADF members aged 17. 

It was not possible to analyse a number of other potentially important social, demographic and psychological factors that may contribute to homelessness, due to a number of factors including the specific administrative information available in the PMKeyS data.

The results may indicate some groups of ex-serving ADF SHS clients that may be at higher risk of homelessness by examining characteristics that may be associated with homelessness. However, it is not possible to attribute a causal relationship between these characteristics and homelessness. 

Some of the results in this report are based on small numbers. Caution should be taken when interpreting these findings, as results based on small numbers can be sensitive to small changes in client numbers over time.


ABS (Australian Bureau of Statistics) 2016. Census of Population and Housing: Estimating Homelessness, 2016. ABS cat. no. 2049.0. Canberra: ABS.

AIHW (Australian Institute of Health and Welfare) 2019. Specialist homelessness services annual report 2017–18. Cat. no. HOU 299. Canberra: AIHW.