Australian Institute of Criminology

Skip to content


This report marks the twentieth anniversary of the Royal Commission into Aboriginal Deaths in Custody (RCIADIC) and thus, 20 years of monitoring deaths in custody by the Australian Institute of Criminology (AIC). Through the National Deaths in Custody Program (NDICP), the AIC has been monitoring the extent and nature of all deaths that have occurred in prison, juvenile justice and police custody since 1992, with data also collected retrospectively back to 1 January 1980. The purpose of this program is to collect information about deaths in custody, analyse the circumstances of these deaths and report findings regularly to the Australian Government. In handing down their recommendations, the Commissioners emphasised the need for ongoing monitoring of all deaths in custody by the AIC (Recommendations 41 and 46). For this reason, one of the key functions of the NDICP is to serve as a performance and accountability measure for custodial authorities, in that it brings to light details about all deaths occurring in the custody of such authorities. All NDICP reports are publically available.

Overall, the NDICP has collected and analysed data on the following cases that occurred between 1 January 1980 and 30 June 2011:

  • 2,319 total deaths in custody across Australia (449 Indigenous deaths; 19%);
  • 1,393 deaths in prison custody (238 Indigenous deaths; 17%);
  • 903 deaths in police and police custody-related operations (203 Indigenous deaths; 23%);
  • 18 deaths in juvenile justice custody (8 Indigenous deaths; 44%); and
  • five deaths in other/Australian Government custody (all non-Indigenous deaths).

The data collected in this program are supplied directly to the AIC by police agencies, corrective services departments and juvenile justice agencies in each jurisdiction. Without their ongoing support and commitment to monitoring deaths in custody, the NDICP could not function. The AIC is grateful for all the time and effort spent by these agencies in providing data and reviewing publications prior to release. Data provided by custodial authorities is supplemented with information obtained through coronial findings, as well as toxicology and autopsy reports. In this way, the NDICP is also greatly assisted by the hard work of State and Deputy State Coroners, police investigators and all their support staff, who together ensure that every death in custody undergoes a thorough investigation. Monitoring deaths in custody in Australia would also not be possible without the contribution from all of these people.

Throughout 2011, the AIC undertook a comprehensive review of the NDICP, which included a focus on data quality, clarifying definitions, improvements to data collection and validation processes, as well as the development of new data on the prevalence of drugs and/or alcohol, and mental illness among those persons dying in custody. To comply with Australian Government reporting practice, the NDICP has moved to reporting on a financial year basis. For a full discussion of the review, see Appendix A of this report.

Analysis of data captured by the NDICP over the last 32 years demonstrates that significant improvements have been made to prevent deaths in some areas, but that work should continue in order to reduce other forms of deaths in custody.

First, it is of concern to see that the proportion of Indigenous prisoners has almost doubled over the 20 years since the RCAIDC. In 1991, when the final report was handed down by the RCIADIC, Indigenous people represented one in seven people in prison (14%; ABS 1998) and one in seven deaths in prison custody (14%; n=5). In 2011, Indigenous people represented just over one in four people in prison (26%; SCRCSP 2012) and one in five deaths (21%; n=12). Therefore, the number of Indigenous people in prison appears to have increased at a faster rate than the number of deaths of Indigenous prisoners.

Since 1979–80, there have been 238 deaths of Indigenous persons in prison custody, representing 17 percent of all deaths to occur in this setting. While the number of Indigenous deaths in recent years is high, the total is lower than would be expected based on the proportion of the prison population that is Indigenous. Available data indicates that since 2001, between one in five and one in four prisoners (20% to 26%) in Australia is an Indigenous person (SCRCSP 2012), while over the same period less than one in every five (19%; n=92) deaths in prison custody was of an Indigenous person. Over the last eight years, the rate of death has been consistently lower among Indigenous prisoners than their non-Indigenous counterparts.

It can be concluded that the headline finding of the RCIADIC that Indigenous persons were no more likely to die in prison custody than non-Indigenous persons remains true today. At the heart of the problem is the over-representation of Indigenous persons at every stage of the criminal justice system. Any efforts to reduce the number of Indigenous deaths in custody must therefore incorporate a focus on reducing the number of Indigenous people who end up in prison.

A second point of concern is the relative age profile of Indigenous deaths in custody when compared with their non-Indigenous counterparts. Almost half of all Indigenous deaths (48%; n=113) in prison custody were of persons aged 25–39 years, compared with less than two in five (38%; n=438) for the equivalent non-Indigenous cohort. For deaths in police custody and custody-related operations, almost two in five (38%; n=78) Indigenous deaths were of young persons under the age of 25 years, compared with just over one in four (27%; n=187) for their non-Indigenous counterparts. Apart from dying at relatively younger ages than non-Indigenous persons, a greater proportion of Indigenous deaths are due to natural causes. For example, 53 percent (n=127) of Indigenous deaths in prison custody were from serious health problems, such as heart disease, respiratory illness and cancer, whereas the equivalent proportion for non-Indigenous prisoners was 40 percent (n=457). These findings are likely to be a product of the recognised poorer health outcomes and lower life expectancy among Indigenous Australians.

Yet significant improvements have been made in some areas. Over the last decade, there has been a considerable decline in self-inflicted deaths in custody, such as hangings, most particularly among Indigenous prisoners. Between 1989–90 and 1999–2000, there was an annual average of 4.4 Indigenous deaths by hanging in prison; however, from 2000–01 to 2010–11, the annual average dropped to 2.5 hanging deaths per year. That is, during the 1990s, almost half (45%; n=48) of all Indigenous deaths in prison were due to hanging, while since 2000, this proportion has dropped to under one-third (29%; n=28). A marked drop in the proportion of hanging deaths was also found among non-Indigenous prisoners when comparing the 1990s with the decade since 2000. Analysis of the data showed that throughout the 1990s, more than two in five deaths (43%; n=212) of non-Indigenous prisoners were due to hanging; however, since 2000, this proportion has reduced to less than one in three (30%; n=122).

There has also been a marked decline in the number of hanging deaths occurring in police institutional settings, such as cells, watchhouses and police vans, with the number of such deaths recorded in recent years being among the lowest annual totals on record. There has only been one Indigenous death by hanging in a police institutional setting since 2005–06. This overall decline in hanging deaths in custody in recent years is most likely associated with the significant investment made by police and corrective services agencies in redesigning cells, removing hanging points, improved screening for signs of self-harm and better observation regimes for prisoners deemed at risk.

Whereas throughout the 1980s and 1990s, hanging deaths in prison were the most prevalent types of death each year, since 2003–04, deaths resulting from natural causes have now become the most common cause of deaths in custody. This emerging trend has implications for the provision of health services in custody settings.

With regards to the circumstances in which Indigenous persons died in police custody and custody-related operations since 1989–90, Indigenous persons were more likely to die in accidental circumstances (47%; n=66); the most common manner of death being motor vehicle pursuit-related accidents. Whereas for non-Indigenous persons, 39 percent (n=215) of police custody deaths were accidental. However, the number of self-inflicted deaths for Indigenous and non-Indigenous persons remains an area for further intervention, with 18 percent (n=26) of Indigenous deaths and 34 percent (n=192) of non-Indigenous deaths in police custody resulting from self-inflicted injuries.

Overall, despite significant increases in Indigenous persons coming into custody, concerted efforts have reduced the rate of self-inflicted injuries and deaths. The challenge that remains is to continue to reduce self-harm matters, while orienting health facilities to cater for the needs of an aging prison population and the concomitant rise in serious illness and disease.

Adam Tomison