Download all data pages
PDF 11Mb
Data

Suicide and self-harm among people in contact with the justice system

On this page


People in custody have high rates of mental distress, suicidal behaviours and other risk factors for suicide (AIHW 2022a, 2023a; Marzano et al. 2016; Rose et al 2019). The National Mental Health and Suicide Prevention Agreement (the Agreement) identified suicide prevention priority populations (Australian Government 2022). One of these priority populations is ‘people who are (or have been previously) in contact with the criminal justice system.’ This release provides an overview of suicide and self-harm among people in contact with the criminal justice system.

This release brings together two separate data sources to explore suicide and self-harm among people in contact with the criminal justice system: 

In addition, the below page explores contact with the legal system more broadly, not necessarily for criminal reasons. For example, people who were involved in litigation or child custody disputes are included as well as people who had been to prison:

Deaths by suicide among adults in custody: Data from the National Deaths in Custody Program

This release presents data from the National Deaths in Custody Program (NDICP), managed by the Australian Institute of Criminology (AIC). The National Deaths in Custody Program collects data on deaths occurring in prison custody, police custody and custody-related operations since 1979.

In this release, ‘deaths by suicide’ includes deaths where the manner of death was determined to be intentionally self-inflicted. This includes deaths in prison custody, police custody and custody-related operations. The determination of the manner of death is made by a coroner through a coronial finding. Deaths classified as self-inflicted but with unintentional or unknown intent are included in ‘other deaths.’ See ‘What is a death in custody?’ below for more information on the inclusion criteria. 

Suicide among adults who died in custody 1989–90 to 2021–22

Caution needs to be taken when interpreting the results and visualisations below. Due to the small numbers of deaths by suicide in custody, the results can change considerably between years. Some data are also aggregated to 5-year groupings due to small numbers and to maintain confidentiality. The number and proportion of suicide deaths in recent years may be subject to more revisions. The revisions are subject to coronial processes, which can have some delays. 

This release does not present rates of death in police custody due to the lack of a reliable data source for the number of individuals placed into police custody each year or those who come into contact with police during custody-related operations (McAlister & Bricknell, 2023).

For more information on National Deaths in Custody Program methods, please see Deaths in custody in Australia 2022–23. 

The total number of deaths in custody has increased over time, rising from as low as 53 in 2005–06 and 59 and 61 in 1990–91 and 1991–92, to highs of 112 and 111 in 2018–19 and 2019–2020. This increase may be attributed to the growing incarcerated population since 1989–90 (ABS 2022a). 

In parallel, the number of suicide deaths in custody ranged between 23 and 44 deaths per year from 1989–90 and 2004–05. After this period, there was a slight decrease, with suicide deaths in custody consistently remaining lower, ranging from 7 to 23 each year. This downward trend is reflected in the proportion of suicide deaths relative to all deaths in custody since 1989–90. The data visualisation below illustrates this steady decrease, particularly noticeable from around 2004–05, with suicide deaths accounting for less than 30% (ranging from 7 to 28 deaths) of all deaths in custody since that period.

The interactive timeseries visualisation shows suicide deaths and other deaths in custody by year from 1989 to 2022. Percent (%) and number (n) can be chosen.

Suicide among adults who died in custody 1992–97 to 2017–22

Patterns of suicide deaths in custody by age have changed over time

During 1992–97, the proportion of deaths in custody that were due to suicide was highest for adults aged 18–24 years (48%). Since 1997–2002, the highest proportion of suicide deaths have been among people in custody aged 25–39 years (ranging from 12%–46%,14–100 deaths), except in 2017–22, when the highest proportion (16%, 6 deaths) was among people aged 18–24 years.

The number and proportion of deaths in custody from suicide have generally decreased across age groups. Note that the 2017–22 percentages are subject to revision and may change. They should be interpreted with caution. 

The interactive bar chart visualisation shows suicide deaths and other deaths in custody by age groups aggregated by 5-year time periods from 1992 to 2022. Data are shown by the following age groups; ‘18-24 years’, ‘25-39 years’, ‘20-54 years’, ‘55 years and over’, and ‘All age groups’. ‘Deaths by suicide in custody’ and ‘Other deaths in custody’ can be viewed as a percent (%) or a number (n). ‘All deaths in custody’ can be viewed as a number.

The proportion of suicide among First Nations adults who died in custody decreased over time

From 1992–97 until 1997–2002 around a third (33% and 37% in 1992–97 and 1997–2002, 25 and 31 deaths respectively) of deaths in custody among First Nations people were by suicide. In comparison, the proportion of deaths by suicide among non-Indigenous Australians in custody was slightly higher, with around 2 out of every 5 deaths of non-Indigenous Australians (ranging from 39% to 40%, or 152 and 142 deaths, respectively) being from suicide over the same period. 

From 2002–07, the proportions of deaths by suicide among all deaths in custody decreased for both First Nations people (23% to 6.8%, or 19 to 6 deaths) and non-Indigenous Australians (37% to 11%, or 103 to 44). From 2012–17, the proportion of deaths by suicide among First Nations people (15%, 13 deaths) more than halved compared with 1992-97 (33%, 25 deaths), before declining further to 6.8% (6 deaths) in 2017–2022. Overall, the numbers of all deaths in custody, including non-suicide deaths, increased over time, which may be due to the increased numbers of people in custody.

The interactive bar chart visualisation shows suicide deaths and other deaths in custody by Indigenous status aggregated by 5-year time periods from 1992 to 2022. ‘Deaths by suicide in custody’ and ‘Other deaths in custody’ can be viewed as a percent (%) or a number (n). ‘All deaths in custody’ can be viewed as a number.

Suicide deaths in custody declined in men over time

Around 1 in 5 (22%, 86) men who died in custody died by suicide in 2012–17, a decrease of almost half since 1997–2002 (38%, 170 deaths). Please note that data for the most recent years, 2017–22, are not presented ('n.p.') due to the small number of reported deaths by gender.

It is important to approach the findings regarding suicide among women who died in custody with caution. Due to small numbers (ranging between below 5 and 13), the proportion of suicide deaths among women who died in custody can vary significantly over time and cannot be directly compared to those among men who died in custody. 

During 2012–17, less than one-third of women (29%, 5 deaths) who died in custody died by suicide. The numbers of women who died by suicide in custody are too small to identify any meaningful trends over time.

The interactive bar chart visualisation shows suicide deaths and other deaths in custody by sex aggregated by 5-year time periods from 1992 to 2022. ‘Percentage of deaths in custody (%)’ and ‘Number of deaths in custody (n)’ can be selected. ‘Sex’ (‘Women’, ‘Men’, ‘Persons’) can also be selected.

Suicide among adults who died in prison custody 1982–87 to 2017–22

Note that the previous section on suicide deaths in custody included data from 1992–97 to 2017–22 on prison and police custody. The section below includes data about sentencing and people in prison custody, which are analysed from 1982–87.  

For further information on data availability, please see Deaths in custody in Australia from the Australian Institute of Criminology. 

Suicide remained higher among those who were unsentenced and died in prison custody since 1982–87, compared with those who were sentenced

Being unsentenced or 'on remand' means awaiting trial or sentencing and is associated with high rates of suicide and suicidality (Zhong et al. 2021). Remand can be an emotionally tumultuous experience for many people (Sarre et al. 2016).

For the period 2017–22, around 1 in 5 (22%, 25 deaths) deaths among people who were unsentenced and died in prison custody were due to suicide. In comparison, during the same period, the proportion of deaths by suicide among people who were sentenced and died in prison custody was considerably lower (6.9%, 18 deaths).

The interactive bar chart visualisation shows suicide deaths and other deaths in prison custody by sentence status and manner of death aggregated by 5-year time periods from 1982 to 2022. ‘Deaths by suicide in custody’ and ‘Other deaths in custody’ can be viewed as a percent (%) or a number (n). ‘All deaths in custody’ can be viewed as a number.

History of self-harm among adults in prisons: Data from the National Prisoner Health Data Collection 2022

This section reports on history of self-harm among people in prison using data from the 2022 6th National Prisoner Health Data Collection (NPHDC). The numbers may underestimate the true prevalence of self-harm among people in prison due to the reliance on self-reporting from people surveyed. Yet these data can provide insight into the wellbeing of people in contact with the justice system.

National Prisoner Health Data Collection: Data from the NPHDC were collected over 2 weeks and 4 surveys in 2022. Of these surveys, 2 included data on self-harm. The first was among people entering prison across Australia (‘entrants forms’). The entrants form surveyed around 370 people. The second surveyed people leaving prison (‘dischargees forms’). Dischargees were those leaving prison during the survey period or within 4 weeks after the survey period. The dischargees form surveyed around 430 people. 

Eligible people in prison aged 18 years and over were invited to participate in the voluntary survey. All states and territories were included, except for Victoria which did not participate. Both people who were sentenced and unsentenced were eligible for the surveys. For more information on the survey sampling and methods please see The health of people in Australia’s prisons 2022 report.

For the purpose of this section, self-harm refers to a person who intentionally inflicted physical harm or injury to their own body with or without suicidal intent (AIHW 2023b).

History of self-harm among adults entering prison was highest among women and young adults

Around 2 in 5 (42%, 25) women and 1 in 6 (17%, 54) men surveyed entering prison reported a history of self-harm.

Generally, a history of self-harm decreased by age among those entering prison. More than one-quarter (29%, 12) of people entering prison aged 18–24 years reported a history of self-harm. People entering prison aged 25–34 years had a similar percentage (28%, 39). However, 14% (15) of people aged 35–44 years and 13% (8) of people aged 45–54 years had a history of self-harm. History of self-harm for people aged 55 years and over cannot be reported on due to low numbers (fewer than 5). 

More than 1 in 7 (15%, 28) First Nations people entering prison reported a history of self-harm. More than one-quarter (27%, 50) of non-Indigenous Australians entering prison reported a history of self-harm.

These proportions are lower than those found in other health surveys of people in prison. For example, a survey of adults on remand and sentenced in prison from the Australian Capital Territory (ACT) found around one-third (31%) of people held in ACT prisons had attempted suicide (Butler et al. 2018). Further, a study of First Nations men held both on remand and sentenced in Victoria found more than half (55%) had a history of attempted suicide (Shepherd et al. 2018). Because these two examples are specific to suicide attempts, we would expect that self-harm (which includes suicide attempts and injuries without suicidal intent) would be higher due to the broader inclusion criteria. The smaller proportions from the NPHDC may reflect differences in survey designs, for example the questions used. However, the sampling methods were similar. That is, no specific sampling criteria beyond voluntary participation in the surveys and being held in prison. For more information, please see the original, Health of People in Australia's Prisons 2022 report.


The interactive bar chart visualisation shows history of self-harm and no history of self-harm reported by people in prisons by sex, age group, First Nations status and all entrants. 'History of self-harm’ can be viewed as a percent (%) or a number (n).

Recent thoughts of self-harm among adults entering prison are highest among women

More than 1 in 10 (15%, 57) people entering prison indicated they had thoughts of self-harm in the 12 months before the survey. Around 1 in 3 (30%, 18) women and 1 in 10 (13%, 39) men entering prison reported recent thoughts (in the past 12 months) of self-harm. 

Almost 1 in 6 (17%, 32) non-Indigenous Australians entering prison indicated they had recent thoughts of self-harm. About 1 in 8 (13%, 24) First Nations people entering prison reported recent thoughts of self-harm. 

Among age groups, people surveyed aged 25–34 years entering prison had the highest proportion (20%, 28) of recent thoughts of self-harm. Recent thoughts of self-harm among other age groups ranged between 10% (6) in people aged 45–54, and 15% (16) in people aged 18–24. 

The interactive bar chart visualisation shows 12-month history of thoughts self-harm and no thoughts of self-harm reported by people in prisons by sex, age group, First Nations status and all entrants. ‘Thoughts of self-harm’ can be viewed as a percent (%) or a number (n).

 

Over one-quarter of women entering prison were identified as at risk of self-harm or suicide

Following the entrants survey, researchers asked prison staff whether the participant was identified as currently at risk of suicide or self-harm (excluding at 4 prisons in New South Wales where researchers administered surveys). 

Fewer than 1 in 15 (6.3%, 16) people entering prison were identified by prison staff as at risk of self-harm or suicide. Over one-quarter (28%, 8) of women entering prison were identified by staff as at risk of self-harm or suicide. Among men surveyed entering prison, 3.6% (8) were identified by staff as at risk of self-harm or suicide.

The proportion of First Nations people entering prison identified at risk of self-harm or suicide by staff was 5.4% (8). Among non-Indigenous Australians entering prison, 6.8% (7) were identified at risk for self-harm or suicide by staff.

Age groups are not presented due to small numbers in each category.

The interactive bar chart visualisation shows identified as at risk of self-harm and not identified as at risk of self-harm reported by people in prisons by sex, First Nations status and all entrants. ‘Identified at risk’ can be viewed as a percent (%) or a number (n).

Suicide among people with legal system contact and associated risk factors: Data from the National Mortality Database

As part of the National Suicide and Self-harm Monitoring Project, the AIHW commissioned the Australian Bureau of Statistics (ABS) to code psychosocial risk factors (‘Z-codes’) among cases of suicide. Data coding began in 2017 and is available for all deaths by suicide since 2017. For this analysis, specific Z-codes were used to flag in the National Mortality Database (NMD) whether a person who died by suicide had contact with the legal system. Note that this includes people who had contact with the legal system for both criminal (e.g. imprisonment) and non-criminal reasons (e.g. child custody or support proceedings). Please see the technical notes for further details. 

1 in 10 suicide deaths had legal system contact

The data visualisation below shows the proportion of people who died by suicide and had contact with the legal system. 

Between 2017 and 2022:

  • Overall, 1 in 10 (10%, 2,015) people who died by suicide had any contact with the legal system.
  • More than 1 in 6 (17%, 210) First Nations people who died by suicide had any contact with the legal system.
  • Over 1 in 6 (15%, 521) people aged 35–44 years who died by suicide had previous contact with the legal system. This was the highest proportion among all age groups.
  • The proportion of men who died by suicide and had contact with the legal system was 12% (1,798), compared with 4.7% (226) of women.

The interactive bar chart visualisation shows mentions of contact with the legal system among people who died by suicide by age group, First Nations status, and sex. “Contact with the legal system” can be viewed as a percent (%).

Psychosocial risk factors among those who had legal system contact and died by suicide

The National Suicide and Self-harm Monitoring Project continues to work with the ABS to identify and code psychosocial, mental and behavioural risk factors mentioned in cases of deaths referred to a coroner, including deaths by suicide. To explore the most prevalent psychosocial risk factors among people who died by suicide in the general population, please visit Psychosocial Risk Factors and Deaths by Suicide. For an overview of inclusion and exclusion criteria for psychosocial risk factors in the NMD, please refer to Table 1 in the technical notes.

‘Problems related to legal circumstances’, and ‘personal history of self-harm’ were the most common psychosocial risk factors among most age groups

Among those who had contact with the legal system, 'Problems related to legal circumstances' was the leading psychosocial risk factor for suicide between 2017 and 2022, being mentioned in approximately 80% (1,609) of these cases. The most common non-legal risk factor was a 'personal history of self-harm,' mentioned in 24% (473) of the suicide cases involving individuals with legal system contact.

The ranks and prevalence of the most frequently mentioned psychosocial risk factors by age group are presented below.

Among people who died by suicide and had contact with the legal system:

  • Generally, the proportions of psychosocial risk factors mentioned were higher than those without legal system contact across all age groups. For instance, 25% (113) of 25–34-year-olds with legal system contact had 'Disruption of family by separation and divorce' mentioned in their file, making it the fourth most mentioned psychosocial risk factor for that age group. In contrast, 'Disruption of family by separation and divorce' was mentioned in fewer than 18% (563) of cases for 25–34-year-olds among those without legal system contact.
  • ‘Problems related to other legal circumstances’ was the most common across all age groups, ranging from 77% (352) among 25–34-year-olds up to 82% (334) among people aged 55 years and over.
  • ‘Personal history of self-harm’ was the second most common across all age groups except those aged 35–44 years. For this age group, ‘Personal history of self-harm’ was the fourth most common risk factor (22%, 113), following ‘Disruption to family by separation or divorce (ranked second, 29%, 153) and ‘Problems in relationship with spouse or partner’ (ranked third, 25%, 129).

Among those who died by suicide with no legal system contact:

  • The most common risk factor across age groups was ‘Personal history of self-harm’, except for those aged 55 years and over. Among people aged 55 years, ‘Limitations of activities due to disability’ was the most common psychosocial risk factor mentioned (20%, 1,103) followed by ‘Personal history of self-harm’ (ranked second, 18%, 984) and ‘Disappearance and death of family member’ (ranked third, 11%, 638).
  • 'Limitations of activities due to disability' was the most common risk factor for those aged 55 years. In contrast, it was the tenth most common psychosocial risk factor for the same age group with legal system contact, mentioned in 2.9% (12) of cases when 'legal system contact psychosocial risk factors' are excluded from the visualisation.

The interactive bar chart visualisation shows psychosocial risk factors among people who died by suicide who had contact with the legal system compared to people who did not have contact by age group. “Psychosocial risk factors” can be viewed as a percent (%).

Psychosocial risk factors for suicide differed by legal system contact and sex

Among people who died by suicide and had contact with the legal system:

  • Both men and women had ‘Problems related to other legal circumstances’ as their most frequently mentioned psychosocial risk factor. Women had a slightly higher prevalence (84%, 189) compared to men (79%, 1,420).

Among those who died by suicide with no legal system contact:

  • 'Limitation of activities due to disability' was the sixth most frequent psychosocial risk factor among men (7.4%, 955) and the fifth among women (8.0%, 367) without legal system contact. This factor did not appear in the top 10 psychosocial risk factors for men or women with legal system contact.
  • 'Disappearance and death of a family member' was the third most common risk factor (11%, 508) among women but was the third most common (7.5%, 17) among women with legal system contact.

The interactive bar chart visualisation shows psychosocial risk factors among people who died by suicide who had contact with the legal system compared to people who did not have contact by sex. “Psychosocial risk factors” can be viewed as a percent (%).

Psychosocial risk factors by Indigenous status

Among First Nations people who died by suicide:

  • 'Personal history of self-harm' was the third most common psychosocial risk factor among those with legal system contact (25%, 53), but it was the most common among those with no such contact (26%, 269).

Among non-Indigenous Australians who died by suicide:

  • The most frequently mentioned psychosocial risk factor among those with legal system contact was ‘Problems related to other legal circumstances’ (81%, 1,445), followed by ‘Disruption of family by separation and divorce’ (24%, 422), and ‘Personal history of self-harm’ (23%, 416).
  • 'Disappearance and death of a family member' was the ninth most common risk factor among those who had legal system contact (6.4%, 114) and the fourth most common among with no contact (9.0%, 1,449).

The interactive bar chart visualisation shows psychosocial risk factors among people who died by suicide who had contact with the legal system compared to people who did not have contact by First Nations status. “Psychosocial risk factors” can be viewed as a percent (%).

Mental and behavioural disorders among those who had legal system contact and died by suicide

Substance-related mental and behavioural disorders were more common among those with legal system contact

Among those who died by suicide, 'Mood [affective] disorders' were the leading mental and behavioural disorders recorded between 2017 and 2022. 'Mood [affective] disorders' were mentioned in 38% (757) of cases that had legal system contact, and 44% (7,643) of cases without such contact. This was followed by ‘Alcohol disorders’ (25%, 511) in those with contact and ‘Anxiety disorders’ (20%, 3,480) among those without contact.

Ranks and prevalence of most frequently mentioned mental and behavioural disorders by age group are presented below. People aged over 55 years generally had the lowest proportions of mental and behavioural disorders mentioned in their cases among both people who did and did not have contact with the legal system.

Among people who had contact with the legal system:

  • ‘Mood [affective] disorders’ were consistently the most frequently mental and behavioural disorder across all age groups, ranging from range 35% (74) among people aged under 25 years and those 55 years and over (141), to 41% (187) among those aged 25–34 years.

Comparisons with those with no legal system contact:

  • Substance-related mental and behavioural disorders tended to be ranked higher and mentioned more frequently among those who had contact with the legal system across all age groups.

The interactive bar chart visualisation shows mental and behavioural disorders among people who died by suicide who had contact with the legal system compared to people who did not have contact by age group. “Mental and behavioural disorders” can be viewed as a percent (%).

Mental disorders in women and substance disorders in men were more common among those with legal system contact than those without contact

Among people who died by suicide and had contact with the legal system:

  • Proportions of mental and behavioural disorders tended to be higher among women, particularly for non-substance-related disorders. For example, ‘Mood [affective] disorders’ were ranked first among both men and women with the legal system contact, but the proportion was higher among women (48%,108) compared to men (36%, 648). 
  • Women had higher proportions of ‘Anxiety disorders’ (32%, 79), ‘Other substance disorders’ (17%, 38), ‘Other mental and behavioural disorders’ (15%, 33), ‘Personality disorders’ (8.8%, 7), and ‘Opioid disorders’ (5.8%, 13), compared to women with no contact. The prevalence of 'Mood [affective] disorders' was similar for women with and without legal system contact (48%, 109 vs. 49%, 2,257). Men had a similar pattern in these risk factors, though the proportions were lower than for women.
  • Men had higher proportions of 'Alcohol disorders' (26%, 464 compared to 21%, 47 of women), 'Stimulant disorders' (14%, 251 compared to 12%, 27 of women), which ranked fourth among men and sixth among women, and 'Cannabinoid disorders' which ranked sixth among men (10%, 175) and ninth among women (5.3%, 12). However, women had higher proportions of 'Other substance disorders' (17%, 38 compared to 13%, 233 of men), which ranked fourth among women and fifth among men, and 'Opioid disorders' (5.8%, 13 compared to 4.0%, 71 of men), which ranked eighth among women and ninth among men.

Among those who died by suicide with no legal system contact:

  • 'Mood [affective] disorders', 'Alcohol disorders', and 'Anxiety disorders' remained the top three risk factors in men for both those with and without legal system contact. While the proportion of 'Anxiety disorders' was the same among men with and without contact (17%, 301 and 2,220, respectively), the proportions of other mental and behavioural risk factors varied between the two groups. For instance, 36% (648) of men with contact had 'Mood [affective] disorders' mentioned compared to 42% (5,386) of men without contact. Around a quarter (26%, 464) of men with contact had 'Alcohol disorders' compared to 20% (2,633) of men without contact.

Substance-related mental and behavioural disorders tended to be higher in both ranks and proportions among men and women with legal system contact compared to those without. For example, the proportions of 'Stimulant disorders' among men and women with legal system contact were more than double those of men and women without contact (14%, 251 of men and 12%, 27 of women with contact compared to 6.0%, 773 of men and 4.4%, 200 of women without contact).

The interactive bar chart visualisation shows mental and behavioural disorders among people who died by suicide who had contact with the legal system compared to people who did not have contact by sex. “Mental and behavioural disorders” can be viewed as a percent (%).

Mental and behavioural disorder prevalence varies by legal system contact among First Nations people 

Among First Nations people who died by suicide:

  • The three most common mental and behavioural disorders among those with legal system contact were 'Mood [affective] disorders' (28%, 59), 'Alcohol disorders' (27%, 56), and 'Stimulant disorders' (22%, 46). Among those without legal system contact, the three most common disorders were 'Alcohol disorders' (32%, 337), 'Mood [affective] disorders' (29%, 307), and 'Cannabinoid disorders' (15%, 153).

Among non-Indigenous Australians who died by suicide:

  • All substance-related mental and behavioural disorders among those with legal system contact dropped in rank among those without contact, while all other mental and behavioural disorders either increased in rank or stayed the same.
  • The most mentioned mental and behavioural disorders among those with legal system contact were 'Mood [affective] disorders' (39%, 692), Alcohol disorders' (25%, 446), and 'Anxiety disorders' (19%, 345). Among those without legal system contact, the most common disorders were 'Mood [affective] disorders' (45%, 7,210), 'Anxiety disorders' (20%, 3,288), and 'Alcohol disorders' (18%, 2,978).

 

The interactive bar chart visualisation shows mental and behavioural disorders among people who died by suicide who had contact with the legal system compared to people who did not have contact by First Nations status. “Mental and behavioural disorders” can be viewed as a percent (%).

Previous page Next page