Deaths in custody: a gender-specific analysis

In Australia and overseas there exists a paucity of research on deaths in custody where women are the primary focus of study; most of the literature in this area treats women as a subset of the primary focus - men. It has been suggested that this lack of information about women dying in custody is due to the relatively small number of female deaths when compared with the number of male deaths. Indeed, Scraton and Chadwick (1995, p. 89) comment that:

...the relatively small proportion of women in prison compared to men is one of the reasons for the apparent "invisibility" of women in research studies or official accounts.

In her review of the literature on suicides in prison, Liebling (1992) adds that the lack of information on women also stems from the fact that most studies use all-male samples or have a tendency not to consider women as a separate and distinct group.

Of the limited research on deaths in custody that examines gender differences, the focus has predominately been on suicide in prisons. One suggested explanation for this narrow focus is that suicide has consistently been found to be the leading cause of death in prison (Sattar 2001). Lloyd (1990) reviewed the international literature on suicide and self-injury in prison and found that of the 13 studies reviewed, only six provided any specific information on female inmate suicides and, of these, only three provided information about the broader gender composition in the penal population under study.

Most literature on suicide deaths in prison has found no significant gender-based differences in the rates of suicide in prison (Sattar 2001; Morrison 1996; Dooley 1994; Liebling 1992, 1994). Hanging has been found to be the most common cause of death for either gender in prison custody (Sattar 2001; Dalton 1999; Crighton & Towl 1997; Morrison 1996; Lester & Danto 1993; Liebling 1992; Green et al. 1992; Lloyd 1990). However, most of these studies had a very small number of female suicides in their sample group (between one and four females), thus making it difficult to generalise about the nature and circumstance of female prison deaths based on the results. For example, in a study of prison suicide and suicide attempts in England and Wales between 1985 and 1993 there were only six females included in the sample group (Liebling 1996).

The only article to focus specifically on female deaths in custody is by Kerley and Cunneen (1994). The authors examine differences in the use of prison and police custody between Indigenous and non-Indigenous women in Australia and examine the 11 Indigenous female deaths looked at by the Royal Commission into Aboriginal Deaths in Custody (RCIADIC), and five such deaths that occurred after that time.

The majority of these deaths occurred in police custody, where the women had been detained for relatively minor offences and had died of causes other than suicide. However, the paper does not examine differences between gender or Indigenous status and the findings are based on a relatively small sample size.

There are three main limitations with the existing literature on female deaths in custody:

• the focus has been exclusively on deaths in prison custody;¹
• the focus has been on suicide as the cause of death; and
• there have been small sample sizes.

Focusing only on deaths within the prison system excludes deaths that occur in police custody and custody-related police operations. These types of deaths account for a significant proportion of custodial deaths in Australia. The focus on suicide as the cause of death is just as limiting because around half of all custodial deaths are attributed to causes other than suicide.

What is a “Death in Custody”?

For the purposes of the NDICP a "death in custody" is defined as a death, wherever occurring, of a person:

• who is in prison custody or police custody or detention as a juvenile;
• whose death is caused or contributed to by traumatic injuries sustained, or by lack of proper care while in custody or detention;
• who is fatally injured in the process of police or prison officers attempting to detain that person; or
• who is fatally injured in the process of that person escaping or attempting to escape from prison custody or police custody or juvenile detention.

(Royal Commission into Aboriginal Deaths in Custody 1991, pp. 189-190)

The Present Study

Given the limited available research on female deaths in custody and, in particular, deaths other than suicides in prison, the present study aims to bridge this gap by examining female deaths in all forms of custody, with specific focus on the circumstances and characteristics of the 75 female deaths in custody that have occurred in Australia between 1980 and 2000.

The National Deaths in Custody Program (NDICP) holds over two decades of data on deaths in custody in Australia, enabling a more extensive examination of female deaths in all forms of custody than has previously been undertaken. The aim of this study, then, is specifically to examine women who have died in custody in order to draw out any patterns specific to women that may have been overlooked in the past. For example, what factors characterise the death of females in custody in Australia? Are there any differences between female Indigenous and non-Indigenous deaths in custody? What are the main gender differentials in deaths in custody in Australia?

The identification of gender differences in deaths in custody will assist in achieving a greater understanding of this occurrence, and how to better tailor policies to prevent both male and female deaths in custody. Furthermore, incarceration rates for women in Australian prisons have been increasing in recent times at a disproportionate rate (Cameron 2001). As a result of this increase it is critical for policy-makers to have reliable data for preventative policy development.

Female Custodial Deaths in Australia 1980–2000

General trends in female custodial deaths

Between 1980 and 2000, a total of 1,442 people died in all forms of custody in Australia. The majority of deaths were of male persons (n=1,367), with the death of females accounting for five per cent of all deaths (n=75). Of the 75 female deaths in custody during the 21-year period, 24 deaths involved Indigenous (Aboriginal and Torres Strait Islander) persons (32%). Of the 1,367 deaths of male persons in custody, 248 deaths were of Indigenous persons (18%) (see Table 1).

Over the 21 years, the number of female deaths in custody has remained relatively stable. In contrast, the number of male deaths in custody has increased from 41 deaths in 1980 to 88 deaths in 2000 (Figure 1).

Figure 1 : Australian deaths in custody [see attached PDF for graph]

Cause of death

Hanging was the leading cause of death for females who died in custody between 1980 and 2000 (32%) (Table 1). A reading of the case histories of the 75 female deaths in custody suggests that many of these women experienced adversity, such as problematic licit and illicit substance use, alcohol misuse, domestic violence, mental illness or self-harm in their personal histories. The involvement of drugs and alcohol featured predominately in the lives of these women even where drugs or alcohol were not directly involved in the cause of death (see case studies later in this paper). It is important to note that any deaths caused by chronic degenerative conditions resulting from long-term alcohol and/or drug abuse are categorised as deaths due to natural causes.

Manner of death

Self-inflicted deaths were the most common manner of death for both females and males in custody in Australia over the past two decades (36% for females and 44% for males). In the cases where manner of death was established, 50 per cent of Indigenous female deaths were due to natural causes, while the leading manner of death for non-Indigenous women was self-inflicted death (43%) (Table 1). Thirty-eight per cent of Indigenous males died in custody due to natural causes as opposed to 26 per cent of non-Indigenous males (Table 1). The leading manner of death for non-Indigenous males over the 21 years was self-inflicted death, accounting for 45 per cent of all deaths.

Case Study One

The deceased was an Indigenous woman who died at the age of 41 as a result of legal intervention. She had a history of psychiatric problems and alcohol abuse, with numerous voluntary and involuntary admissions into psychiatric institutions. There was also evidence that she had been physically and sexually abused as a child. On the day of her death, she attended her local welfare centre intoxicated and carrying a hatchet. The deceased began destroying property and the police were called. When the police arrived, they repeatedly told the deceased to drop the hatchet but she refused. She then raised the hatchet and started running towards an officer. The officer fired five fatal shots in response. The toxicology report indicated that the deceased had a blood alcohol level of 0.3. There was also evidence that she had taken antidepressants and other drugs prior to her death (case no. 9470).

Case Study Two

The deceased was a non-Indigenous woman, 20 years of age. She had been sentenced to a six-month fixed prison sentence relating to charges of break and enter and breach of parole. The deceased had been in custody for four days when, during night shift, staff discovered her suspended by a piece of clothing which was attached to a window bar. Resuscitation was attempted but the woman had already passed away. The deceased had made an earlier attempt at suicide. The coroner found that the prisoner had wilfully hanged herself (case no. 9968).

Most serious offence leading to custody

The NDICP collects information on the most serious offence leading to the final period of custody. At the time of their death, most women had been incarcerated for offences related to "good order" (such as public drunkenness, prostitution and disorderly conduct) (37%). Males were most likely to be in custody for violent offences (52%). The second most common category of most serious offence for which women were in custody before their deaths was both property crime (23%) and violent offences (23%) (Table 1).

Good order offences were the most common "most serious offence" leading to custody for both Indigenous and non-Indigenous women. Fifty-four per cent of Indigenous women were in custody for good order offences as their most serious offence at the time of death compared with 28 per cent of non-Indigenous women (Table 1).

Case Study Three

The deceased was an 18-year-old Indigenous woman. She was charged with offences relating to stealing and a variety of other good order offences, and was placed on remand in prison for a short period. The woman became ill while in prison but was misdiagnosed and therefore incorrectly treated. Directly after her release she collapsed and was transferred to a hospital where she later died. The death resulted from complications arising from the misdiagnosed illness that arose in prison and, as such, the death was recorded in the NDICP database as a death in custody due to natural causes (case no. 9847).

Custodial authority ²

In terms of which custodial authority was in charge of the individual at the time of death, a number of interesting differences emerge based on the gender (and Indigenous status) of those who died. The analysis reveals that the majority of female deaths occurred in police custody or custody-related police operations (51%), followed by prison custody (48%) and juvenile detention (1%). In comparison, the majority of male deaths occurred in prison custody (64%), followed by police custody or custody-related police operations (35%) (Table 1). One of the main factors contributing to this gender variation is that most Indigenous women over the 21-year period died in police custody (79%, or 19 out of 24), with very few deaths of Indigenous women in prison custody (n=4). This is in contrast to the deaths of Indigenous males in custody where just over half died in prison custody (56%).

Case Study Four

The deceased was a non-Indigenous woman in her late twenties. She died in police custody of a drug overdose. Earlier that day ambulance officers had attended to the deceased on two occasions after she had collapsed in a public park and a street. On both occasions she refused their assistance. Shortly after, she was arrested for drunkenness in a public place and disorderly behaviour and was transferred to a watch-house. After being charged, the deceased was placed in a cell and checked regularly by staff. An attempt was made to awake the deceased and it was at that time her death was discovered. Later, illicit substances were located in the deceased's handbag. The coroner ruled that the cause of death was mixed drug toxicity (case no. 9477).

Prison custody

Since the early 1980s Australia's prison population has more than doubled from a total of 9,826 in 1982 to 22,458 in 2001. The female prison population has increased four-fold from a total of 335 in 1982 to 1,498 in 2001 (ABS 2002). Despite this sharp increase in the number of females imprisoned in Australia, there has not been a subsequent increase in the number of female prison deaths.

In the decade between 1980 and 1989 there were 14 female deaths in Australian prisons. This figure increased to 20 deaths in the following 10-year period (1990 to 1999) - an average of two deaths per year. There were two additional deaths in 2000. The total number of females who have died in Australian prisons between 1980 and 2000 is 36, or around four per cent of all prison deaths during this time. The ratio of male prison deaths to female prison deaths generally reflects the ratio of male to female prisoners in Australia. According to the 2001 prison census, women accounted for seven per cent of the total prison population and men accounted for 93 per cent (ABS 2002).

Twenty-five per cent of women in prison custody died as a result of an accident. This includes deaths that result from the acute effects of drugs or alcohol. While deaths attributed to drugs and alcohol make up less than 17 per cent of female deaths in prison custody, as noted earlier, alcohol- and drug-related deaths may also be classified as natural or accidental deaths.

Approximately 45 per cent of female prison deaths involved unsentenced (or remand) prisoners, whereas only 30 per cent of male deaths in prison were of remand prisoners. Almost half of all non- Indigenous female deaths in prison during 1980-2000 were unsentenced prisoners (47%). The proportion of deaths among unsentenced prisoners is high considering that in Australia unsentenced prisoners comprise 24 per cent of all female prisoners (ABS 2002). These figures suggest for female remand prisoners than sentenced prisoners.

Case Study Five

The deceased was an Indigenous woman aged 40 years. She had a long history of alcohol abuse and was suffering from severe generalised chronic alcoholic disease. Police were called when the woman was found intoxicated in a public place. They placed her in a police vehicle and were attempting to transfer her to a sobering-up shelter. Upon arrival it was discovered that the deceased was unconscious, having suffered a pulmonary embolism during transit. Despite resuscitation attempts, she could not be revived. The coroner found that she died of natural causes, with the condition of chronic alcoholism found to have been a significant condition contributing to the death (case no. 9353).

Police custody and custody-related police operations

In terms of cause and manner of death in police custody there are some notable differences between men and women over the 21-year period under review. The leading cause of death for women in police custody was natural causes (32%). In contrast, men who died in police custody were more likely to have died from hanging (23%) or multiple trauma injuries (23%). In terms of manner of death, natural causes was most common for women who died in police custody between 1980 and 2000 (34%). In contrast, self-inflicted death was the most common manner of death for men in police custody (37%).

Of those cases where most serious offence was known, 45 per cent of the women and 34 per cent of men who died in police custody were in custody for offences against good order. Although caution should be exercised due to the relatively small number of deaths, it is interesting to note that a higher proportion of Indigenous women were in police custody at the time of their death because of good order offences than non-Indigenous women (57 per cent and 35 per cent respectively).

The analysis of 21 years of NDICP data has highlighted not only gender differences in custodial deaths in Australia, but also differences that exist within the female population on the basis of Indigenous status. The findings in this paper raise several important and interesting questions. These include:

• Why have female deaths in prison custody remained relatively stable despite substantial increases in the number of female prisoners?
• Why have proportionately more women than men died in police custody?

With relation to the stable female prison death rate, one suggestion is that there have been improvements in the correctional management of female prisoners across Australia that have resulted in the continued low numbers of deaths per year despite the rapid increase in incarceration numbers. Although this may be a feasible explanation, there have been no comprehensive evaluations of such approaches or programs within the Australian context, therefore no evidence is available to either support or refute this claim.

The second question stems from the finding that of the 75 female deaths in custody between 1980 and 2000, approximately half occurred in police custody while the majority of male deaths occurred in prison custody. These findings are consistent across the two decades 1980-89 and 1990-99, with approximately half of all female deaths occurring in police custody in both periods (46 per cent and 55 per cent respectively).

The answer lies in examining the differences between Indigenous and non-Indigenous women and the custodial authority in charge of the individual at the time of death. The majority of non-Indigenous women died in prison custody, while most Indigenous women died in police custody (again, consistent across the decades 1980-89 and 1990-99). The disproportionate number of Indigenous women who have died in police custody is the predominate reason why police custody deaths account for the majority of all female custodial deaths.

It has been suggested that this is due to issues specific to Indigenous females, their use of public space, and the use of police custody for minor offences. In their examination of custodial deaths of Indigenous women, Kerley and Cunneen (1994) noted similar issues as those highlighted in the present study. They argue that the custodial system administrated by the police was a "major factor in the deaths of Aboriginal and Torres Strait Islander women" (p. 547). They suggest that there was potential for non custodial alternatives to have been used in many cases that they examined - for instance, greater use of non-custodial sanctions for minor public order offences, with the use of custody and imprisonment as a last resort.

More recently, Gardiner and Takagaki (2002) have examined some of these issues in detail within the Victorian context. They found similar patterns to Kerley and Cunneen (1994). The most relevant to this discussion was that Indigenous women continued to be over-represented in police statistics in comparison to non-Indigenous women for minor public order offences.

Statistics indicate a continued high level of police intervention in the daily life of Indigenous females and a disproportionate tendency on the part of police to process Indigenous offenders by way of arrest rather than caution. (2002, p. 318)

The findings in the present study may lend support to these arguments. Although a large proportion of both Indigenous and non-Indigenous women were in custody at the time of their death as a result of good order offences (over half of which relate directly to public intoxication), the likelihood of detention for good order offences was greater for Indigenous women. One out of two Indigenous women and 28 per cent of all non-Indigenous women who died in custody were detained for such offences.

The Final Report of the Royal Commission into Aboriginal Deaths in Custody (1991) noted the high incidence of good-order offences in the criminal histories of the women whose deaths it investigated. Similarly, in 1995, the National Police Custody Survey also found that Indigenous people were held in custody at higher rates than other Australians and that detention for public drunkenness was a serious problem among Indigenous women in particular (Carcach & McDonald 1997).

Hence, the issue of the criminalisation of public drunkenness and other minor offences, and the impact these offences have had on the involvement of Indigenous persons within the criminal justice system is not a new concern.

One policy option that could be considered in light of the findings involves the broader use of strategies to divert persons away from detainment in police custody for less serious "good order" offences. For example some jurisdictions that have decriminalised public drunkenness have also established sobering-up centres that are used as an alternative to police cells as a place to hold and care for intoxicated individuals.

However, addressing the issue of public drunkenness is complex, especially in jurisdictions where it remains an offence.³ In these jurisdictions police may have no other alternative than to place an intoxicated person into custody as the law dictates. The option to divert these persons away from custodial authorities and criminal sanction is not available. Until public drunkenness is decriminalised in all jurisdictions in Australia, and the underlying issues of alcohol involvement in Indigenous communities are adequately addressed, the patterns identified in this paper will continue.

This study has identified the existence of gender differentials among persons who died in custody in Australia. Most importantly, the study has found that women do not die in the same circumstances as their male counterparts. This finding has significant implications in terms of policy formation and development. The study suggests the need for prevention policies to be gender-specific, and to be mindful of the differences that exist between male and female deaths and between Indigenous and non-Indigenous female deaths in custody. This will ensure that the specific gender-based issues will be adequately addressed.

Notes

1. In addition to prison custody deaths, Morrison (1996) also examined deaths in "institutional" police custody in Australia. However, the primary focus was again on suicide as the cause of death. In Morrison's study (1996) no differences were found in terms of the suicide rate in police institutional custody between males and females.

2. Deaths in prison custody include deaths that occur in prisons or juvenile detention facilities. This also includes the deaths that occur during transfer to or from prison or juvenile detention centres, or in medical facilities following transfer from adult and juvenile detention centres (Royal Commission into Aboriginal Deaths in Custody 1991, pp. 189-190).

   Deaths in police custody are divided into two main categories:

   Category 1:

   1. Deaths in institutional settings (for example, police stations or lockups, police vehicles, during transfer to or from such an institution, or in hospitals following transfer from an institution).
   2. Other deaths in police operations where officers were in close contact with the deceased. This would include most raids and shootings by police. However, it would not include most sieges where a perimeter was established around a premise but officers did not have such close contact with the person to be able to significantly influence or control the person's behaviour.

   Category 2: Other deaths during custody related police operations. This would cover situations where officers did not have such close contact with the person to be able to significantly influence or control the person's behaviour. It would include most sieges, as described above, and most cases where officers were attempting to detain a person - for example, pursuits. (This definition of a "death in police custody" is based on a 1994 resolution of the Australasian Police Ministers' Council.)

3. Most jurisdictions in Australia have now decriminalised public drunkenness, although it is still a criminal offence in Queensland, Tasmania and Victoria. However, a parliamentary inquiry into public drunkenness in Victoria was completed in 2001 that recommended the decriminalisation of the relevant offences in that state (Drugs and Crime Prevention Committee 2001).

References

• Australian Bureau of Statistics (ABS) 2002, Prisoners in Australia, 2001, cat. no. 4517.0, prepared for the Corrective Services Ministers' Council by the National Corrective Services Statistics Unit, Australian Bureau of Statistics, Melbourne.
• Cameron, M. 2001, Women prisoners and correctional programs, Trends and Issues in Crime and Criminal Justice, no. 194, Australian Institute of Criminology, Canberra.
• Carcach, C. & McDonald, D. 1997, National Police Custody Survey, Research and Public Policy Series, no. 9, Australian Institute of Criminology, Canberra.
• Crighton, D. & Towl, G. 1997, "Self-inflicted deaths in prison in England and Wales: An analysis of the data for 1988-90 and 1994-95", in G. Towl (ed.), Suicide and Self-Injury in Prisons, The British Psychological Society, Leicester, pp. 12-20.
• Dalton, V. 1999, Suicide in prison 1980 to 1998: National overview, Trends and Issues in Crime and Criminal Justice, no. 126, Australian Institute of Criminology, Canberra.
• Dooley, E. 1994, "Unnatural deaths in prison: Is there a future?" in A. Liebling & T. Ward (eds), Deaths in Custody: International Perspectives, Whiting & Birch, London, pp. 28-35.
• Drugs and Crime Prevention Committee 2001, Inquiry into Public Drunkenness - Final Report, Parliament of Victoria, Melbourne.
• Gardiner, G. & Takagaki, T. 2002, "Indigenous women and the police in Victoria: Patterns of offending and victimisation in the 1990s", Current Issues in Criminal Justice, vol. 13, no. 3, pp. 301-21.
• Green, C., Andre, G., Kendall, K., Looman, T. & Polvi, N. 1992, "A study of 133 suicides among Canadian federal prisoners", Forum on Corrections Research, vol. 4, no. 3, pp. 17-19.
• Kerley, K. & Cunneen, C. 1994, "Deaths in custody in Australia: The untold story of Aboriginal and Torres Strait Islander women", The Canadian Journal of Women and the Law, vol. 7, no. 1, pp. 531-51.
• Lester, D. & Danto, B. 1993, Suicide Behind Bars: Prediction and Prevention, Charles Press, Philadelphia.
• Liebling, A. 1992, Suicide in Prison, Routledge, London.
• --- 1994, "Suicide amongst women prisoners", The Howard Journal of Criminal Justice, vol. 33, no. 1, pp.1-9.
• --- 1996, "Prison suicide: What progress research?" in A. Liebling (ed.), Deaths in Custody: Caring for People at Risk, Whiting & Birch, London, pp. 41-54.
• Lloyd, C. 1990, Suicide and Self-Injury in Prison: A Literature Review, Home Office Research Study no. 115, Home Office Research and Planning Unit, London.
• Morrison, S. 1996, "Custodial suicide in Australia: A comparative study of different populations", Medicine, Science and Law, vol. 36, no. 2, pp. 167-77.
• Royal Commission into Aboriginal Deaths in Custody 1991 (Commissioner Elliott Johnston), National Report, Volume 1, Australian Government Publishing Service, Canberra.
• Sattar, G. 2001, Rates and Causes of Death Among Prisoners and Offenders Under Community Supervision, Home Office Research Study no. 231, Home Office Research, Development and Statistics Directorate, London.
• Scraton, P. & Chadwick, K. 1995, "Speaking ill of the dead: Institutionalised responses to deaths in custody", in E. Sbarbaro & R. Keller (eds.), Prison Crisis: Critical Readings, Harrow & Heston, New York, pp. 73-95.

Lisa Collins is a Research Assistant and Jenny Mouzos is a Research Analyst at the Australian Institute of Criminology