Characteristics of ex-serving Australian Defence Force members hospitalised for suicidality and intentional self-harm

Identifying intentional self-harm in Emergency Department data

To complement analysis of admitted patient care, AIHW have examined the use of Emergency Department (ED) services relating to non-fatal suicide behaviour such as intentional self-harm (ISH) or suicidal ideation. Most people experiencing suicidal ideation or who self-harm do not go on to end their lives—but these behaviours are a strong risk factor for suicide (Arias et al. 2015 and Crandall et al. 2006). Therefore, understanding of ISH in ED hospital settings is key to suicide prevention.

Presentations relating to ISH and/or suicidal ideation cannot be easily identified in the current national ED data collection. The challenges to identification include limitations in ED presentation diagnosis coding systems, variability between diagnosis coders, and variability between coding sets available in ED systems across jurisdictions. See Technical notes for more information.

Given the known challenges and limitations of identifying suicide and self-harm related presentations within ED data, this analysis further aimed to enhance identification by supplementing standard diagnoses codes with additional codes related to self-harm (Sara and Wu 2023). To account for jurisdictional differences in coding systems, the analysis integrated codes from two of the main coding classifications used in ED data collection systems (International Statistical Classification of Diseases – ICD – and Systemised Nomenclature of Medicine - SNOMED), applying mapping developed by the Independent Hospital and Aged Care Pricing Authority (IHACPA). Supplementary codes were also used to identify the associated comorbid risk factors (Zhang et al 2022). See Technical notes for more information. 

ISH is often defined as deliberately injuring or hurting oneself, with or without the intention of dying. Currently, it is not possible to determine suicidal intent from all intentional self-harm in ED data. The following analysis therefore includes both suicide attempts and non-suicidal self-harming behaviours.

Ex-serving ADF members’ index presentation to emergency departments

Between 2013 and 2020, 117,300 ex-serving ADF members attended the ED for any reason, including 98,900 males and 18,400 females. 

Most ex-serving members that presented to ED were between the ages of 35–54, for both males (53.7%) and females (64.7%).

The age at index presentation was broadly reflective of the age structure of the ex-ADF population during this period, and similarly this was also observed for the total Australian population (See Supplementary table S.ED.1). However, it should be noted that the study ex-serving ADF population have a different age structure compared to the total Australian population due to the absence of older veterans with service prior to 1985, fewer ex-serving in the 17–24 age group and fewer ex-serving females aged 65+ years reflecting historical differences in female ADF recruitment.

Of the ex-serving ADF members who had presented to ED, 4,400 (3.7%) had an ED presentation for ISH or suicidal ideation, from here on referred to as presentations for self-harm or suicidal behaviour, which was above the 2.8% for all Australians. There were 3,600 (3.7%) ex-serving males and 770 (4.1%) ex-serving females who presented for self-harm or suicidal behaviour.

Most index presentations for self-harm or suicidal behaviour from ex-serving ADF members were among those aged 35–54 years (62.7% for males and 68.9% for females). For the total Australian population, the majority of those who presented for self-harm or suicidal behaviour were younger, aged 17–34 (51.8% for males and 56.3% for females). This difference is in part due to the relatively low proportion of ex-serving ADF members at younger ages among the ex-serving ADF population. See Supplementary table S.ED.2.

Presentations related to self-harm or suicidal behaviour

Proportions of ED presentations by male ex-serving ADF members for self-harm or suicidal behaviour were compared with all Australian males of the same age group as shown in Table 6 below. Presentations to ED by ex-serving ADF males of all age groups were 1.2 to 1.6 times more likely related to self-harm or suicidal behaviour compared to all Australian males of the same age group. This was similarly observed for the female ex-serving ADF population except for those aged 65+ years.

Source: AIHW analysis of linked Defence Historical Personnel data–PMKeyS–DVA client–NDI–MCD–NHMD-NNAPEDCD–MBS–PBS–RPBS data (2010–2020) and AIHW NHDH (2019-2020)

Notes:

1. ED presentations for self-harm or suicidal behaviour refer to ED presentations for intentional self-harm and/or suicidal ideation.
2. Bolding indicates statistical significance was found for the relative difference between the proportions.
3. Includes public hospital data from New South Wales, Victoria, Queensland, South Australia, Tasmania and the Australian Capital Territory. Additional data for ex-serving members was provided by the Department of Veterans' Affairs (DVA) for DVA-funded emergency department care in public hospitals in all states and territories, including Western Australia and the Northern Territory.

Risk factors related to self-harm or suicidal behaviour

Analysis was undertaken to identify key risk factors associated with ED presentations for self-harm or suicidal behaviour. Risk factors evaluated include mental and behavioural disorders, substance use disorders (excluding alcohol), alcohol use disorder and related conditions and social issues (Arias et al. 2015 and Crandall et al. 2006). These were analysed for ED presentations during the study period, including presentations prior to the initial ED presentation for self-harm or suicidal behaviour (index presentation).

The proportion of ED presentations for each risk factor was calculated by dividing the presentations for a risk factor by the total number of presentations for persons with an ED presentation with a self-harm or suicidal behaviour. Presentations for risk factors could have occurred before, during or after the ED presentation for self-harm or suicidal behaviour.

Of ED presentations by the self-harm or suicidal behaviour cohort that related to mental and behavioural disorders (Table 7), presentations among:

• ex-serving ADF males aged 65 years and over were 1.3 times more likely to be for mental and behavioural disorders compared to all Australian males aged 65 years and over (8.2% vs 6.2%)
• ex-serving ADF females aged 25–34 were 1.3 times more likely to be for mental and behavioural disorders compared to all Australian females aged 25–34 (16.4% vs 12.9%).

Source: AIHW analysis of linked Defence Historical Personnel data–PMKeyS–DVA client–NDI–MCD–NHMD-NNAPEDCD–MBS–PBS–RPBS data (2010–2020) and AIHW NHDH (2019-2020)

Notes:

n.p. Not available for publication but included in totals where applicable, unless otherwise indicated. In this case this is a result of low numbers being potentially identifying.

1. ED presentations for self-harm or suicidal behaviour refer to ED presentations for intentional self-harm and/or suicidal ideation.
2. The self-harm or suicidal behaviour cohort is defined as those who have presented to an ED that includes a diagnosis of self-harm or suicidal behaviour at any point during the analysis period of 1 July 2013 to 30 June 2020.
3. Proportions of presentations is a percentage of all presentations among those who have had a self-harm or suicidal behaviour ED presentation.
4. Bolding indicates statistical significance was found for the relative difference between the proportions.
5. Includes public hospital data from New South Wales, Victoria, Queensland, South Australia, Tasmania and the Australian Capital Territory. Additional data for ex-serving members was provided by the Department of Veterans' Affairs (DVA) for DVA-funded emergency department care in public hospitals in all states and territories, including Western Australia and the Northern Territory.

Of ED presentations by the self-harm or suicidal behaviour cohort that related to substance use disorders, not including alcohol (Table 8), presentations among:

• ex-serving ADF males in the age groups of 25–34, 35–44, and 45–54, were approximately half as likely to be for substance use disorders compared to all Australian males of the same age group (2.4% vs 4.3% for those aged 35–44).
• ex-serving ADF males aged 65 and over were 3 times more likely to be for substance use disorders compared to all Australian males aged 65 and over (0.6% vs 0.2%).

Source: AIHW analysis of linked Defence Historical Personnel data–PMKeyS–DVA client–NDI–MCD–NHMD-NNAPEDCD–MBS–PBS–RPBS data (2010–2020) and AIHW NHDH (2019-2020)

Notes:

n.p. Not available for publication but included in totals where applicable, unless otherwise indicated. In this case this is a result of low numbers being potentially identifying.

1. ED presentations for self-harm or suicidal behaviour refer to ED presentations for intentional self-harm and/or suicidal ideation.
2. The self-harm or suicidal behaviour cohort is defined as those who have presented to an ED that includes a diagnosis of self-harm or suicidal behaviour at any point during the analysis period of 1 July 2013 to 30 June 2020.
3. Proportions of presentations is a percentage of all presentations among those who have had a self-harm or suicidal behaviour ED presentation.
4. Bolding indicates statistical significance was found for the relative difference between the proportions.
5. Includes public hospital data from New South Wales, Victoria, Queensland, South Australia, Tasmania and the Australian Capital Territory. Additional data for ex-serving members was provided by the Department of Veterans' Affairs (DVA) for DVA-funded emergency department care in public hospitals in all states and territories, including Western Australia and the Northern Territory.

Of ED presentations by the self-harm or suicidal behaviour cohort that related to alcohol use disorders (Table 9), presentations among:

• ex-serving ADF males aged 45–54 years were less likely (0.9 times) to be for alcohol use disorders compared to all Australian males aged 45–54 years (6.5% vs 7.4%).
• ex-serving ADF males aged 65 and over were almost 2 times more likely to be for alcohol use disorders compared to all Australian males aged 65 and over (5.3% vs 2.8%).
• ex-serving ADF females aged 23–34 years were approximately a third less likely to be for alcohol use disorders compared to all Australian females aged 23–34 years (0.6% vs 2.1%).
• ex-serving ADF females aged 45–54 years were half as likely to be for alcohol use disorders compared to all Australian females aged 45–54 years (2.6% vs 4.7%).

For complete tables of ED presentations for these risk factors by age group, refer to Supplementary Tables S.ED.4–S.ED.6.

Source: AIHW analysis of linked Defence Historical Personnel data–PMKeyS–DVA client–NDI–MCD–NHMD-NNAPEDCD–MBS–PBS–RPBS data (2010–2020) and AIHW NHDH (2019-2020)

Notes:

n.p. Not available for publication but included in totals where applicable, unless otherwise indicated. In this case this is a result of low numbers being potentially identifying.

1. ED presentations for self-harm or suicidal behaviour refer to ED presentations for intentional self-harm and/or suicidal ideation.
2. The self-harm or suicidal behaviour cohort is defined as those who have presented to an ED that includes a diagnosis of self-harm or suicidal behaviour at any point during the analysis period of 1 July 2013 to 30 June 2020.
3. Proportions of presentations is a percentage of all presentations among those who have had a self-harm or suicidal behaviour ED presentation.
4. Bolding indicates statistical significance was found for the relative difference between the proportions.
5. Includes public hospital data from New South Wales, Victoria, Queensland, South Australia, Tasmania and the Australian Capital Territory. Additional data for ex-serving members was provided by the Department of Veterans' Affairs (DVA) for DVA-funded emergency department care in public hospitals in all states and territories, including Western Australia and the Northern Territory.

The number of ED self-harm or suicidal behaviour cohort patients who have had more than one presentation for these risk factors were further investigated. Among ex-serving ADF males who had a presentation for self-harm or suicidal behaviour:

• 20.8% had at least 2 ED presentations for mental and behavioural disorders
• 2.9% had at least 2 ED presentations for substance use disorder, not including alcohol
• 6.4% had at least 2 ED presentations for alcohol use disorder
• 0.5% had at least 2 ED presentations for social issues.

The equivalent proportions for ex-serving ADF females are similar to ex-serving ADF males. See Supplementary table S.ED.7 for further information and proportions for the ex-serving ADF female and the total Australian self-harm cohorts. Note that these data are not analysed by age group due to small numbers. Therefore, any differences between the ex-serving and total Australian populations may be related to the underlying population age structure. Age differences can be partially informed by findings presented above and Supplementary tables S.ED.4-S.ED.6.

The number of patients who presented to ED who have had risk factor presentations in the lead up to the index self-harm or suicidal behaviour ED presentation were also further investigated. In the 30 days prior to the index presentation for self-harm or suicidal behaviour, most ex-serving males did not present to ED. However, among ex-serving ADF males who presented in that period:

• 6.4% presented to an ED for mental and behavioural disorders
• 0.7% presented to an ED for substance use disorder, not including alcohol
• 1.3% presented to an ED for alcohol use disorder
• 0.2% presented to an ED for social issues.

The equivalent proportions of presentations in the 30 days prior for ex-serving ADF females are not discussed here due to low numbers. For complete breakdown by time period up to 1 year prior to the index self-harm or suicidal behaviour presentation, and available proportions for the ex-serving female and Australian self-harm cohorts, refer to Supplementary table S.ED.8. Comparisons to the total Australian population should be made with caution due to the differences in age.