Australian Institute of Criminology

Skip to content

Deaths in prison custody

Key indicators of performance

The NDICP reports on the following three high-level indicators for deaths in prison custody:

  • trends in number of deaths in prisons and the proportion of total deaths involving Indigenous prisoners;
  • trends in the rate of death per 100 relevant adult prisoners; and
  • trends in causes and circumstances of deaths in prison custody.

2008–09

A total of 43 deaths occurred in prison custody in Australia in 2008–09. Across the jurisdictions:

  • New South Wales recorded 15 deaths;
  • Queensland recorded nine deaths;
  • Victoria recorded eight deaths;
  • South Australia recorded five deaths;
  • Western Australia recorded four deaths;
  • Tasmania recorded one death;
  • Australian Capital Territory recorded one death; and
  • no deaths were recorded in the Northern Territory (see Table 7).

2009–10

  • A total of 58 deaths occurred in prison custody in Australia in 2009–10. Across the jurisdictions:
New South Wales recorded 21 deaths;
  • Queensland recorded 10 deaths;
  • Victoria recorded eight deaths;
  • Western Australia recorded eight deaths;
  • South Australia recorded five deaths;
  • Tasmania recorded three deaths;
  • the Northern Territory recorded two deaths; and
  • Australian Capital Territory recorded one death (see Table 8).

2010–11

A total of 58 deaths occurred in prison custody in Australia in 2010–11. Across the jurisdictions:

  • New South Wales recorded 20 deaths;
  • Queensland recorded 11 deaths;
  • Victoria recorded 10 deaths;
  • the Northern Territory recorded six deaths;
  • Western Australia recorded five deaths;
  • South Australia recorded four deaths;
  • Tasmania recorded two deaths; and
  • no deaths were recorded in the Australian Capital Territory (see Table 9).
Table 7 Key indicators of performance—prison custody deaths, 2008–09
Prison
1 Indigenous 7 (16%)
Non-Indigenous 36 (84%)
2 Rate of death—Indigenous 0.10a
Rate of death—non-Indigenous 0.18a
3a Main cause—Indigenous Natural causes—(71%; n=5)
Main cause—non-Indigenous Natural causes—(58%; n=21)
3b Main circumstances—Indigenous Natural causes—(71%; n=5)
Main circumstances—non-Indigenous Natural causes—(58%; n=21)

a: Rate per 100 prisoners on an average day (SCRCSP 2010)

Source: AIC NDICP 2008–09 [computer file]

Table 8 Key indicators of performance—prison custody deaths, 2009–10
Prison
1 Indigenous 14 (24%)
Non-Indigenous 44 (76%)
2 Rate of death—Indigenous 0.19a
Rate of death—non-Indigenous 0.21a
3a Main cause—Indigenous Natural causes—(79%; n=11)
Main cause—non-Indigenous Natural causes—(61%; n=27)
3b Main circumstances—Indigenous Natural causes—(79%; n=11)
Main circumstances—non-Indigenous Natural causes—(61%; n=27)

a: Rate per 100 prisoners on an average day (SCRCSP 2011)

Source: AIC NDICP 2009–10 [computer file]

Table 9 Key indicators of performance—prison custody deaths, 2010–11
Prison
1 Indigenous 12 (21%)
Non-Indigenous 46 (79%)
2 Rate of death—Indigenous 0.16a
Rate of death—non-Indigenous 0.22a
3a Main cause—Indigenous Natural causes—(67%; n=8)
Main cause—non-Indigenous Natural causes—(65%; n=30)
3b Main circumstances—Indigenous Natural causes—(67%; n=8)
Main circumstances—non-Indigenous Natural causes—(65%; n=30)

a: Rate per 100 prisoners on an average day (SCRCSP 2012)

Source: AIC NDICP 2010–11 [computer file]

Important consideration about prison populations

Prison populations differ greatly across the states and territories, which has an impact on the number and distribution of deaths recorded (for trends in the prison population and the distribution between the states and territories, see Appendix E). In general terms, those jurisdictions with larger prison populations tend to have higher numbers of deaths in custody. The rates of death in prison relative to each jurisdiction’s prison population over the last three financial years are presented in Table 10.

It should also be noted that when a prison custody death occurs in a jurisdiction with a small number of prisoners, such as Tasmania or the Australian Capital Territory, the rate of death can appear to be very high as a function of the smaller population size. For example, although New South Wales has consistently recorded the most number of deaths in the last three years, it does not have the highest rate of death among all the jurisdictions. Consequently, when interpreting rates of death, it is important to account for the population size as well as the number of deaths.

Table 10 Prison custody deaths by Indigenous status, 2008–09 to 2010–11 (rate per 100 prisoners)a
2008–09
Indigenous Non-Indigenous Persons
n Rate n Rate Total n Rate
New South Wales 1 0.05 14 0.18 15 0.15
Victoria 0 8 0.20 8 0.19
Queensland 2 0.13 7 0.17 9 0.16
Western Australia 3 0.18 1 0.04 4 0.10
South Australia 1 0.23 4 0.27 5 0.26
Tasmania 0 1 0.22 1 0.19
Northern Territory 0 0 0
Australian Capital Territory 0 1 1.01 1 0.85
Australia 7 0.10 36 0.18 43 0.16
2009–10
Indigenous Non-Indigenous Persons
n Rate n Rate Total n Rate
New South Wales 4 0.18 17 0.22 21 0.20
Victoria 1 0.37 7 0.17 8 0.18
Queensland 4 0.24 6 0.15 10 0.18
Western Australia 4 0.21 4 0.14 8 0.17
South Australia 0 5 0.33 5 0.25
Tasmania 0 3 0.71 3 0.61
Northern Territory 1 0.11 1 0.51 2 0.19
Australian Capital Territory 0 1 0.63 1 0.53
Australia 14 0.19 44 0.21 58 0.20
2010–11
Indigenous Non-Indigenous Persons
n Rate n Rate Total n Rate
New South Wales 1 0.04 19 0.25 20 0.20
Victoria 1 0.35 9 0.21 10 0.22
Queensland 1 0.06 10 0.26 11 0.20
Western Australia 2 0.11 3 0.11 5 0.11
South Australia 1 0.21 3 0.20 4 0.20
Tasmania 1 1.69 1 0.25 2 0.42
Northern Territory 5 0.52 1 0.48 6 0.51
Australian Capital Territory 0 0 0
Australia 12 0.16 46 0.22 58 0.20

a: Rate per 100 relevant prisoners on an average day (SCRCSP 2012–10)

Source: AIC NDICP 2007–08 to 2010–11 [computer file]

Trends

A total of 1,393 deaths have been recorded in prison custody across Australia since 1 January 1980. The rate of death fluctuated considerably over almost two decades up to the late 1990s, reaching a peak of 0.44 deaths per 100 prisoners in 1997–98; however, there was an increase during this period (see Figure 4). Following the peak in the late 1990s, there was a decline in the rate of deaths in prison, with the rate in 2005–06 reaching its lowest point since the collection of this data began. Since this all-time low, the rate of death per 100 prisoners has increased slightly.

Figure 4 Prison custody deaths by year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E1

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Demographic characteristics

Indigenous status

Throughout the 2008–09 to 2010–11 reporting period, 33 of the 159 deaths (21%) in prison custody were of Indigenous prisoners (see Table 10). As has typically been the case in previous years, the rate of death among Indigenous prisoners was lower than the comparative rate for non-Indigenous prisoners across the last three years. The most recent statistics indicate that Indigenous Australians comprise just over a quarter of the national prison population (26%; SCRCSP 2012), whereas Indigenous prisoners represented 21 percent of prison deaths over the same period.

Across the three years, the results were as follows:

  • In 2008–09, there were 43 deaths in prison custody, seven of which involved an Indigenous prisoner (16%). The annual rate of death for Indigenous prisoners was 0.10 per 100 compared with 0.18 for non-Indigenous prisoners.
  • In 2009–10, there were 58 deaths in prison custody, 14 of which involved an Indigenous prisoner (24%). The annual rate of death for Indigenous prisoners was 0.19 per 100 compared with 0.21 for non-Indigenous prisoners.
  • In 2010–11, there were 58 deaths in prison custody, 12 of which involved an Indigenous prisoner (21%). The annual rate of death for Indigenous prisoners was 0.16 per 100 compared with 0.22 for non-Indigenous prisoners.

Trends in Indigenous deaths in prison custody

Despite a comparatively lower rate of death among Indigenous prisoners, it is worth noting that the actual number of Indigenous deaths increased in the two most recent years of collection, representing a departure from the apparent downward trend seen over the last decade (see Table 11 and Figure 5). The number of Indigenous prison custody deaths in 2009–10 was the highest on record—equal with totals in 1999–2000 and 2000–01. However, owing to a commensurate increase in the size of the Indigenous prisoner population, this increase in the number of deaths resulted in only a modest short-term increase in the rate of death since its lowest point in 2005–06. Compared with the long-term trend, the rate of Indigenous death in prison in recent years continues to remain much lower than historical trends.

Figure 5 Prison custody deaths by Indigenous status and year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E3

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Table 11 Prison custody deaths by Indigenous status and year, 1979–80 to 2011–12 (n)
Indigenous Non-Indigenous Total
1979–80 4 11 15
1980–81 2 27 29
1981–82 3 24 27
1982–83 4 21 25
1983–84 2 31 33
1984–85 5 22 27
1985–86 4 14 18
1986–87 2 35 37
1987–88 4 40 44
1988–89 5 32 37
1989–90 9 44 53
1990–91 5 30 35
1991–92 4 27 31
1992–93 5 38 43
1993–94 12 45 57
1994–95 12 43 55
1995–96 13 36 49
1996–97 11 52 63
1997–98 12 69 81
1998–99 10 54 64
1999–2000 14 50 64
2000–01 14 44 58
2001–02 6 47 53
2002–03 12 32 44
2003–04 6 33 39
2004–05 8 31 39
2005–06 3 25 28
2006–07 8 32 40
2007–08 6 40 46
2008–09 7 36 43
2009–10 14 44 58
2010–11 12 46 58
Total 238 1,155 1,393

Source: AIC NDICP 1979–80 to 2011–12 [computer file]

One of the questions that the NDICP endeavours to answer each year is whether Indigenous prisoners are more likely to die in custody than non-Indigenous prisoners. This question can be answered using the statistical approach known as the rate–ratio method. This method compares the rate of death among Indigenous prisoners with the equivalent rate among non-Indigenous prisoners. Where the ratio is above one, this indicates a point in time when Indigenous prisoners were more likely to die than non-Indigenous prisoners relative to their population size. Conversely, where the ratio is less than one, this is a sign of under-representation.

As shown in Figure 6, between 1981–82 and 2002–03, the rate–ratio fluctuated above and below parity, but since 2003–04, the ratio has been consistently under. That is, analysis of 30 years’ worth of data showed that there were 10 points in time when Indigenous persons were overrepresented among deaths in custody and 20 points when they were under-represented. Importantly, for the last eight years in a row, Indigenous prisoners were less likely to die in custody than their non-Indigenous counterparts and at no other point in time, where data has been collected, has this been the case.

Figure 6 Over-representation ratio of Indigenous deaths in custody by year, 1981–82 to 2010–11a

a: Denominator: see Table E3

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Sex

During the last three financial years of reporting (2008–09 to 2010–11), there were a total of 159 deaths in prison custody, of which 154 were males (97%) and five were females (3%; see Table 12). By jurisdiction, New South Wales, Victoria and Queensland recorded the highest number of deaths in prison. Due to the relatively small number of female prison custody deaths each year, small increases in recorded cases are likely to substantially affect the rate (see Figure 7).

For the three year reporting period, the results were as follows:

  • in 2008–09, there were 43 deaths, 41 of which were male prisoners and two were female. The annual rate of death for male prisoners was 0.16 per 100, while the rate of death for female prisoners was 0.10 per 100;
  • in 2009–10, there were 58 deaths, all of which were male prisoners. The annual rate of death for these male prisoners was 0.22 per 100; and
  • in 2010–11, there were 58 deaths, 55 of which were male prisoners and three were female. The annual rate of death for male prisoners was 0.21 per 100, while the rate of death for female prisoners was 0.15 per 100.

Figure 7 Prison custody deaths by sex and year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E2

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Table 12 Prison custody deaths by sex and year, 2008–09 to 2010–11 (rate per 100 prisoners)a
2008–09
Male Female Persons
n Rate n Rate Total n Rate
New South Wales 15 0.16 0 15 0.15
Victoria 8 0.20 0 8 0.19
Queensland 8 0.15 1 0.23 9 0.16
Western Australia 4 0.11 0 4 0.10
South Australia 5 0.28 0 5 0.26
Tasmania 1 0.21 0 1 0.19
Northern Territory 0 0 0
Australian Capital Territory 0 1 9.09 1 0.85
Australia 41 0.16 2 0.10 43 0.16
2009–10
Male Female Persons
n Rate n Rate Total n Rate
New South Wales 21 0.22 0 21 0.20
Victoria 8 0.19 0 8 0.18
Queensland 10 0.19 0 10 0.18
Western Australia 8 0.18 0 8 0.17
South Australia 5 0.27 0 5 0.25
Tasmania 3 0.67 0 3 0.61
Northern Territory 2 0.19 0 2 0.19
Australian Capital Territory 1 0.58 0 1 0.53
Australia 58 0.22 0 58 0.20
2010–11
Male Female Persons
n Rate n Rate Total n Rate
New South Wales 19 0.20 1 0.14 20 0.20
Victoria 10 0.23 0 10 0.22
Queensland 10 0.20 1 0.24 11 0.20
Western Australia 4 0.09 1 0.27 5 0.11
South Australia 4 0.22 0 4 0.20
Tasmania 2 0.45 0 2 0.42
Northern Territory 6 0.54 0 6 0.51
Australian Capital Territory 0 0 0
Australia 55 0.21 3 0.15 58 0.20

a: Denominator: see Table E2

Rate per 100 relevant prisoners on an average day (SCRCSP 2012)

Source: AIC NDICP 2007–08 to 2010–11 [computer file]

Trends

Since 1979–80, deaths of males in prison custody have consistently outnumbered female deaths each year (see Table 13). The proportion of deaths in prison custody between the sexes reflects the distribution of Australia’s prison population, where 93 percent of prisoners on an average day are male (SCRCSP 2012) and 96 percent of deaths in prison custody are of males. The rate of death for males and females in prison remained relatively low and stable from 1981–82 to 1992–93 (see Figure 7); however, there was a slight increase in the rate of male deaths from 1991–92 to 1997–98, after which the rate declined and has remained stable ever since.

Table 13 Prison custody deaths by sex and year, 1979–80 to 2010–11 (n)
Male Female Total
1979–80 15 0 15
1980–81 29 0 29
1981–82 23 4 27
1982–83 24 1 25
1983–84 32 1 33
1984–85 27 0 27
1985–86 16 2 18
1986–87 34 3 37
1987–88 43 1 44
1988–89 36 1 37
1989–90 51 2 53
1990–91 34 1 35
1991–92 30 1 31
1992–93 43 0 43
1993–94 51 6 57
1994–95 54 1 55
1995–96 49 0 49
1996–97 61 2 63
1997–98 78 3 81
1998–99 61 3 64
1999–2000 60 4 64
2000–01 56 2 58
2001–02 49 4 53
2002–03 41 3 44
2003–04 37 2 39
2004–05 35 4 39
2005–06 27 1 28
2006–07 39 1 40
2007–08 43 3 46
2008–09 41 2 43
2009–10 58 0 58
2010–11 55 3 58
Total 1,332 61 1,393

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Age

The median age at death in the period from 2008–09 to 2010–11 was between 42 and 52 years, with the median age at death for non-Indigenous prisoners being generally higher than for Indigenous prisoners (non-Indigenous 42 to 53.5 years cf Indigenous 33.5 to 50 years; see Table 14). Of interest is the fact that the median age at death among non-Indigenous prisoners increased by close to 12 years when comparing 2008–09 with 2009–10 (42 years cf 53.5 years). This rise was the direct result of more than double the number of deaths of prisoners 55 years and older compared with the previous 12 months. There was also a considerable decrease in the median age of death among Indigenous prisoners in 2010–11 compared with the previous year (33.5 years cf 50 years). This substantial drop in median age at death was due to a combination of fewer deaths of Indigenous prisoners aged 40 years and over (9 deaths cf 2 deaths), at the same time as the number of deaths of persons aged 39 years or less doubled (10 deaths cf 5 deaths) compared with the previous year.

Collectively, in the last three years of reporting, deaths among those aged 55 years and above were the most prevalent (34%; n=54), closely followed by prisoners aged 40 to 54 years (32%; n=50) and those aged 25 to 39 years (28%; n=44). The least number of deaths were recorded among prisoners aged less than 25 years (7%; n=11). Disaggregated by jurisdiction, New South Wales, Victoria, Queensland and Western Australia were the only states to record a death of a prisoner less than 25 years of age. Of these 11 deaths of prisoners under the age of 25 years, three were Indigenous persons. Conversely, the Australian Capital Territory was the only jurisdiction not to have recorded a death of a person 55 years or older.

Over the reporting period, the results were as follows:

  • in 2008–09, the median age of death among the overall prison population of Australia was 42 (Indigenous 35 years and non-Indigenous 42 years);
  • in 2009–10, the median age at death among the prison population was 52 years (Indigenous 50 years and non-Indigenous 53.5 years); and
  • in 2010–11, the median age at death among the prison population was 46.5 years (Indigenous 33.5 years and non-Indigenous 51 years).
Table 14 Prison custody deaths by age category and year, 2008–09 to 2010–11 (n)
2008–09
Less than 25 yrs 25–39 yrs 40–54 yrs 55+ yrs Total Median age at death (yrs) Mean age of prisoner populationa (yrs)
New South Wales 1 6 5 3 15 41 34.9
Victoria 0 1 4 3 8 49.5 37.5
Queensland 0 5 3 1 9 36 34.9
Western Australia 0 3 1 0 4 33.5 34.5
South Australia 0 1 2 2 5 53 36.2
Tasmania 0 0 0 1 1 64 35.0
Northern Territory 0 0 0 0 0 33.8
Australian Capital Territory 0 0 1 0 1 42 33.5
Totals
Indigenous 0 4 2 1 7 35 32.1
Non-Indigenous 1 12 14 9 36 42 36.4
All persons 1 16 16 10 43 42 35.3
2009–10
Less than 25 yrs 25–39 yrs 40–54 yrs 55+ yrs Total Median age at death (yrs) Mean age of prisoner populationa (yrs)
New South Wales 2 1 7 11 21 56 35.1
Victoria 0 3 1 4 8 54.5 37.2
Queensland 1 2 3 4 10 53 34.9
Western Australia 2 2 4 0 8 40 34.6
South Australia 0 0 1 4 5 62 37.0
Tasmania 0 1 2 0 3 46 35.1
Northern Territory 0 1 0 1 2 49.5 34.3
Australian Capital Territory 0 0 1 0 1 44 33.7
Totals
Indigenous 2 3 6 3 14 50 32.0
Non-Indigenous 3 7 13 21 44 53.5 36.6
All persons 5 10 19 24 58 52 35.4
2010–11
Less than 25 yrs 25–39 yrs 40–54 yrs 55+ yrs Total Median age at death Mean age of prisoner populationa
New South Wales 4 5 5 6 20 43 35.4
Victoria 1 1 3 5 10 57.5 37.4
Queensland 0 2 3 6 11 62 34.8
Western Australia 0 3 1 1 5 38 33.1
South Australia 0 2 1 1 4 40 37.3
Tasmania 0 1 0 1 2 50 34.9
Northern Territory 0 4 2 0 6 35.5 34.5
Australian Capital Territory 0 0 0 0 0 32.3
Totals
Indigenous 1 9 2 0 12 33.5 32.1
Non-Indigenous 4 9 13 20 46 51 36.8
All persons 5 18 15 20 58 46.5 35.6

a: ABS 2011b

Source: AIC NDICP 2007–08 to 2010–11 [computer file]

The ageing of the prisoner population in Australia

Available data from the ABS indicates that the prison population has been steadily ageing (ABS unpublished data; see Table 15). The ageing of Australia’s prisoner population was the focus of a paper produced by Baidawi et al. (2011), where it is observed that over the decade from 2000 to 2010, the number of prisoners aged over 50 years had increased by 37 percent, with those over 65 years increasing by 142 percent (Baidawi et al. 2011). It is concluded that the main driver for this ageing prisoner population is a combination of ‘prosecution and sentencing laws and practices—including mandatory minimum sentencing and reduced options for early release…’ (Aday et al. cited in Baidawi et al. 2011: 2).

Table 15 Median age of Australia’s prison population by legal status, Indigenous status, sex and year, 1999–2011 (years)
Sentenced Unsentenced
Indigenous Non-Indigenous Indigenous Non-Indigenous
Male Female Male Female Male Female Male Female
1999 28.0 28.9 31.4 30.5 26.4 25.8 28.8 28.2
2000 28.6 28.3 31.7 30.6 26.7 24.6 28.9 28.0
2001 29.2 28.0 32.0 31.5 27.2 28.2 29.2 27.9
2002 29.3 28.8 32.5 31.5 27.5 28.6 29.5 30.0
2003 30.0 29.4 33.0 32.5 27.2 28.3 30.3 31.0
2004 30.1 29.6 33.5 33.7 28.3 30.3 31.0 32.1
2005 30.5 31.0 33.9 34.4 28.1 30.1 31.7 33.4
2006 30.9 30.6 34.3 35.5 28.6 30.2 31.5 31.9
2007 31.4 30.4 34.7 35.8 27.9 27.1 31.6 31.4
2008 30.7 30.5 35.2 36.4 28.4 28.4 32.1 33.5
2009 31.1 32.1 35.1 36.2 28.5 29.8 32.2 33.6
2010 31.1 31.8 35.3 37.1 28.4 30.3 32.2 32.3
2011 30.9 32.3 35.6 37.1 29.0 29.8 31.9 34.0

a: ABS unpublished data

An examination of the deaths in custody data showed that deaths of older prisoners have been increasing in recent years to some of the highest levels ever recorded. The majority of the deaths of older prisoners were due to natural causes, characterised by serious illnesses such as ischaemic heart disease, cancer and respiratory illness. This emerging trend has significant implications for corrective services agencies and the provision of health care to its ageing prisoner population. A closer examination of cause of death among older prisoners is provided below.

Trends

Since 1979–80, close to one-fifth (19%; n=266) of deaths in prison have been of persons aged 25 years or less, 40 percent (n=551) were persons aged 25 to 39 years, 23 percent (n=314) were persons aged 40 to 54 years and 19 percent (n=262) were aged 55 years and above (see Table 16). For two decades from 1979–80 to 1999–2000, deaths of persons aged less than 25 years generally outnumbered those aged 55 years and over each year. However, since 2000–01, this trend has reversed noticeably, with the number of deaths of persons 55 years and over being consistently much higher than deaths among prisoners less than 25 years. In the last two financial years (2009–10 and 2010–11), deaths of persons aged 55 years and older have become the most frequent among the age categories. This trend appears indicative of the ageing prisoner population.

Between 1981–82 and 2010–11:

  • the death rate for prisoners aged 55 years and over fluctuated greatly but has been consistently higher than all other age groups, with a maximum of 3.0 deaths per 100 prisoners in 1990–91 to a minimum of 0.56 in 2008–09; and
  • the rate of death for those prisoners aged less than 55 years has been relatively low and constant over this period, with the exception of persons aged 40–54 years in 1986–87 and 1996–97 (0.5 deaths and 0.6 deaths per 100 respectively). The rate for all other age groups was consistently under 0.5 deaths per 100 relevant prisoners (see Figure 8).

A greater proportion of Indigenous deaths in prison were of younger prisoners when compared with their non-Indigenous counterparts (see Figure 9). Of the Indigenous prisoner deaths, 24 percent (n=58) were aged less than 25 years and almost half were aged 25–39 years (48%, n=113), which is higher than the equivalent proportions among non-Indigenous prisoner deaths (18%; n=208 were less than 25 years and 38%; n=438 were aged 25–39 years). Conversely, a higher proportion on non-Indigenous prison custody deaths involved persons aged over 55 years (22%; n=249 for non-Indigenous prisoners cf 6%; n=13 for Indigenous prisoners)

Table 16 Prison custody deaths by age category and year, 1979–80 to 2010–11 (n)
Less than 25 yrs 25–39 yrs 40–54 yrs 55+ yrs Total
1979–80 4 4 5 2 15
1980–81 5 14 8 2 29
1981–82 8 10 7 2 27
1982–83 5 13 3 4 25
1983–84 14 12 4 3 33
1984–85 5 14 4 4 27
1985–86 3 9 1 5 18
1986–87 10 16 8 3 37
1987–88 15 20 5 4 44
1988–89 11 15 8 3 37
1989–90 16 23 8 6 53
1990–91 8 12 5 10 35
1991–92 7 12 8 4 31
1992–93 14 19 5 5 43
1993–94 12 27 10 8 57
1994–95 16 21 12 6 55
1995–96 12 21 11 5 49
1996–97 10 26 21 6 63
1997–98 26 30 16 9 81
1998–99 9 35 11 9 64
1999–2000 14 28 10 12 64
2000–01 7 32 6 13 58
2001–02 5 19 19 10 53
2002–03 6 16 13 9 44
2003–04 3 12 16 8 39
2004–05 4 16 11 8 39
2005–06 1 8 6 13 28
2006–07 3 9 11 17 40
2007–08 2 14 12 18 46
2008–09 1 16 16 10 43
2009–10 5 10 19 24 58
2010–11 5 18 15 20 58
Total 266 551 314 262 1,393

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Figure 8 Prison custody deaths by age category and year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E4

Rates calculated per 100 relevant prisoners (ABS-A 2011–1981)

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Figure 9 Prison custody deaths by age category and Indigenous status, 1979–80 to 2010–11 (%)

Circumstances surrounding deaths in prison custody

The NDICP collects information on both the cause and manner of each death. Cause of death information relates to the precipitating medical cause as reported by the custodial authority and confirmed by the coroner. The manner of death is different from the cause of death, as it refers to the circumstances in which the person died. In some cases, cause and manner of death will correspond. For example, where a person dies as a result of natural causes, their death will be recorded as natural causes for both cause and manner of death. In other cases, cause and manner of death will differ. For example, where a person dies from acute drug toxicity, cause of death will be recorded as drug-related and manner of death will be recorded as either intentionally self-administered or as an accidental overdose.

Cause of death

Of the 159 deaths in prison custody recorded over the last three financial years, 157 had information available about the cause of death (see Table 17). Of these 157 cases, the most common cause of death was attributed to a natural cause process (65%; n=102—Indigenous deaths 73%; n=24), followed by deaths resulting from hangings (24%; n=38). Of the hanging deaths, seven (18%) involved Indigenous prisoners. There were no Indigenous deaths resulting from a head injury or external/multiple trauma.

Of the prisoners who died from hanging (n=38) between 2008–09 and 2010–11:

  • other fittings in the cell, such as doors or ceiling vents were used in 16 deaths;
  • cell bars were used in six deaths;
  • shower fixtures were used in six deaths;
  • bed bunks were used in four deaths;
  • fittings outside of the cell were used in two deaths;
  • ‘other’ hanging point was used in one death; and
  • the hanging point for the remaining three cases is currently unknown.
Table 17 Prison custody deaths by cause of death and year, 2008–09 to 2010–11 (n)
2008–09
Hanging Natural causes Head injury External/multiple trauma Drugs Other/multiple causes Total
New South Wales 5 9 1 0 0 0 15
Victoria 1 5 0 0 1 1 8
Queensland 3 5 0 1 0 0 9
Western Australia 1 2 0 0 1 0 4
South Australia 1 4 0 0 0 0 5
Tasmania 0 1 0 0 0 0 1
Northern Territory 0 0 0 0 0 0 0
Australia Capital Territory 1 0 0 0 0 0 1
Totals
Indigenous 1 5 0 0 1 0 7
Non-Indigenous 11 21 1 1 1 1 36
All persons 12 26 1 1 2 1 43
2009–10
Hanging Natural causes Head injury External/multiple trauma Drugs Other/multiple causes Total
New South Wales 4 15 0 2 0 0 21
Victoria 2 4 0 1 0 1 8
Queensland 2 6 1 0 0 1 10
Western Australia 4 3 0 1 0 0 8
South Australia 0 5 0 0 0 0 5
Tasmania 0 3 0 0 0 0 3
Northern Territory 1 1 0 0 0 0 2
Australia Capital Territory 0 1 0 0 0 0 1
Totals
Indigenous 3 11 0 0 0 0 14
Non-Indigenous 10 27 1 4 0 2 44
All persons 13 38 1 4 0 2 58
2010–11
Hanging Natural causes Head injury External/multiple trauma Drugs Other/multiple causes Total
New South Wales 7 9 1 1 1 0 19
Victoria 1 8 1 0 0 0 10
Queensland 1 8 0 0 0 1 10
Western Australia 1 4 0 0 0 0 5
South Australia 2 2 0 0 0 0 4
Tasmania 0 2 0 0 0 0 2
Northern Territory 1 5 0 0 0 0 6
Australia Capital Territory 0 0 0 0 0 0 0
Totals
Indigenous 3 8 0 0 0 1 12
Non-Indigenous 10 30 2 1 1 0 44
All personsa 13 38 2 1 1 1 56

a: Cause of death in 2 cases is still awaiting determination through a coronial inquest. When this information becomes available, the NDICP database will be updated accordingly

Source: AIC NDICP 2007–08 to 2010–11 [computer file]

Trends

For more than two decades between 1 January 1980 and 30 June 2000, deaths in prison due to hanging were generally the most frequent cause of death each year. However, from 2000–01 onwards, the number of hanging deaths began to gradually decrease to a 26 year low of six deaths in 2005–06. At the same time as hanging deaths decreased, the numbers of deaths due to a natural cause have gradually increased since 2000–01 and have now become the most common cause of death in prison in recent years (see Table 18, Figures 10 and Figure 11). This finding is consistent with the Trends in Aboriginal and Torres Strait Islander Deaths in Custody and Incarceration Report (ATSIC 1997), implemented by the Australian Government, which found that since 1996–97, deaths due to natural causes were likely to surpass deaths due to hangings.

Since 1979–80, the annual numbers of deaths due to drug and/or acute alcohol toxicity have been consistently low, with a small increase in the late 1990s, followed by a gradual decline. Similarly, deaths due to external and/or multiple traumas occur infrequently and in some years were non-existent.

Over the last 32 years, just over two in every five (42%; n=584) prison deaths since 1979–80 having been due to natural causes, followed by 38 percent (n=525) that were due to hanging (see Table 18, Figures 10 and 11). Deaths due to drugs and/or alcohol were the third most common cause of death in prison (8%; n=117), followed by deaths due to external/multiple trauma (8%; n=111). Deaths as a result of all other causes occur infrequently.

Table 18 Prison custody deaths by cause of death and year, 1979–80 to 2010–11 (n)
Hanging Natural causes Head injury Gunshot External/multiple trauma Drugs/alcohol Other/multiple Total
1979–80 3 8 1 0 3 0 0 15
1980–81 10 13 1 1 4 0 0 29
1981–82 9 7 0 1 5 2 3 27
1982–83 8 12 0 0 2 3 0 25
1983–84 20 6 0 1 2 3 1 33
1984–85 8 12 0 0 4 2 1 27
1985–86 8 7 0 0 0 2 1 18
1986–87 13 15 2 0 3 4 0 37
1987–88 21 10 0 0 6 5 0 42
1988–89 15 11 1 0 2 5 2 36
1989–90 27 16 0 1 6 2 1 53
1990–91 12 17 1 0 2 1 2 35
1991–92 15 8 2 1 1 4 0 31
1992–93 19 11 1 1 4 5 2 43
1993–94 24 19 0 0 3 10 1 57
1994–95 27 15 0 0 7 6 0 55
1995–96 19 14 1 0 5 10 0 49
1996–97 23 28 1 0 4 7 0 63
1997–98 39 21 0 0 10 11 0 81
1998–99 26 18 0 0 8 12 0 64
1999–2000 29 22 1 0 6 6 0 64
2000–01 25 24 0 0 2 6 1 58
2001–02 21 24 0 1 4 2 1 53
2002–03 19 21 0 0 2 1 1 44
2003–04 14 19 1 0 3 2 0 39
2004–05 13 23 1 0 2 0 0 39
2005–06 6 18 1 0 3 0 0 28
2006–07 8 27 1 0 0 3 1 40
2007–08 6 36 0 0 2 0 2 46
2008–09 12 26 1 0 1 2 1 43
2009–10 13 38 1 0 4 0 2 58
2010–11 13 38 2 0 1 1 1 56
Totalª 525 584 20 7 111 117 24 1,388

a: 5 cases have been excluded due to missing information about cause of death

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Figure 10 Prison custody deaths by cause of death and Indigenous status, 1979–80 to 2010–11 (%)

Figure 11 Prison custody deaths by cause of death and year, 1979–80 to 2010–11 (%)

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Natural causes

Throughout the reporting period, 102 out of the 159 deaths (64%) in prison custody were due to natural causes. Of these, 96 deaths have information available from medical records or autopsy reports about the type of natural cause of death (see Table 19). Available data showed that just over one in three natural cause deaths were due to illnesses of the heart, such as ischaemic heart disease or coronary infarcts (34%; n=33). The second most common natural cause was cancer (28%; n=27). The number of prisoners dying from multiple/other causes, such as the combined effects of cirrhosis of the liver, heart disease and chronic obstructive airways disease in one case, are also worth noting (13%; n=12).

There were some minor differences in the proportion of deaths in each category when comparing Indigenous and non-Indigenous prisoners, in that:

  • illnesses of the heart were the most prevalent cause of death, with over one in three deaths falling in this category for both cohorts (33%; n=8 for Indigenous and 35%; n=25 for non-Indigenous);
  • cancer was more prevalent among non-Indigenous prisoners (32%; n=23 cf 17%; n=4); and
  • multiple/other causes were more prevalent among Indigenous prisoners (38%; n=9) than their Indigenous counterparts (21%; n=15).
Table 19 Natural cause deaths in prison custody by type of natural cause and year, 2008–09 to 2010–11 (n)
2008–09
Hearta Cancerb Strokec Respiratoryd Digestivee Conditions/diseasesf Multiple/otherg Total
New South Wales 3 2 1 0 2 0 1 9
Victoria 0 3 0 0 0 1 1 5
Queensland 2 1 1 0 0 1 0 5
Western Australia 1 0 1 0 0 0 0 2
South Australia 1 1 0 0 0 2 0 4
Tasmania 0 1 0 0 0 0 0 1
Northern Territory 0 0 0 0 0 0 0 0
Australia Capital Territory 0 0 0 0 0 0 0 0
Totals
Indigenous 2 1 1 0 0 1 0 5
Non-Indigenous 5 7 2 0 2 3 2 21
All persons 7 8 3 0 2 4 2 26
2009–10
Hearta Cancerb Strokec Respiratoryd Digestivee Conditions/diseasesf Multiple/otherg Total
New South Wales 10 4 0 0 0 0 1 15
Victoria 2 2 0 0 0 0 1 4
Queensland 1 1 0 0 0 2 2 6
Western Australia 1 0 0 1 0 0 1 3
South Australia 1 1 0 0 0 0 1 3
Tasmania 0 1 0 0 0 0 1 3
Northern Territory 0 0 0 0 0 0 1 1
Australia Capital Territory 1 0 0 0 0 0 0 1
Totals
Indigenous 4 2 0 1 0 1 3 11
Non-Indigenous 12 7 0 0 0 1 5 25
All personsh 16 9 0 1 0 2 8 36
2010–11
Hearta Cancerb Strokec Respiratoryd Digestivee Conditions/diseasesf Multiple/otherg Total
New South Wales 3 3 1 0 0 0 1 8
Victoria 3 2 0 2 0 0 1 8
Queensland 1 4 0 1 0 0 0 6
Western Australia 1 0 0 0 0 1 1 3
South Australia 0 1 0 1 0 0 0 2
Tasmania 0 0 0 0 0 0 2 2
Northern Territory 2 0 0 0 0 1 2 5
Australia Capital Territory 0 0 0 0 0 0 0 0
Totals
Indigenous 2 1 0 0 0 1 4 8
Non-Indigenous 8 9 1 4 0 1 3 26
All personsi 10 10 1 4 0 2 7 34

a: Includes ischaemic heart disease, coronary atherosclerosis, thrombosis and infarcts

b: Includes all forms of carcinoma

c: Includes cerebrovascular accidents and intracerebral haemorrhages

d: Includes asthma complications, pneumonia, pulmonary fibrosis and chronic obstructive airways disease

e: Includes esophagitis, gastritis and all other conditions effecting the digestive system

f: Includes diabetes, appendicitis, all forms of cirrhosis and any other form of non-infectious disease

g: Multiple/other causes includes persons dying from multiple natural causes, infectious diseases such as hepatitis or HIV/AIDS, epilepsy complications, cases not classifiable elsewhere and cases where the coroner is unable to determine the precise type of medical cause of death and therefore hands down a finding of ‘undetermined’

h: 2 cases have been excluded due to missing information about type of natural cause death

i: 4 cases have been excluded due to missing information about type of natural cause death

Trends

Overall, since 1979–80, deaths from natural causes have exceeded those due to all other causes, reaching an historic high of 38 deaths in each of the last two years of reporting (see Table 18). The frequency of natural deaths only began increasing from 1996–97 and has remained consistently high over the last 15 financial years. Most prominent is the increase in the number of heart-related deaths, particularly over the last decade, reaching a maximum of 17 deaths in 2007–08. Similarly, the number of cancer-related deaths also saw a considerable increase since the latter part of the 1990s, with the number recorded in 2010–11 being the highest on record. It is also worth mentioning that deaths caused by respiratory illnesses comprise almost one in every 10 (9%; n=53) of all natural cause deaths. The remaining categories of deaths have tended to fluctuate with no discernible trend. Given that natural cause deaths have become the most prevalent in prison custody each year, the NDICP will continue to monitor such deaths closely.

Hanging deaths

The regular collection of data on hanging points and materials used in hanging deaths began in 1989–90; before then, all cases have missing information for these two variables. Specifically, of the total number of hangings since 1979–80 (n=525), information about the hanging point is available in 374 cases (71%) and on the material used in 378 cases (72%). For those cases where information is available, the use of ‘other cell fittings’ inside cells (such as door handles and vents from the ceiling) have been the most common hanging point (39%; n=145; see Table 21). The second most common hanging point used by prisoners is cell bars (36%; n=133); the use of shower fixtures remains the third most common hanging point (13%; n=47). Other hanging points, such as bed bunks (9%; n=34), fittings outside of the cell (3%; n=11) and ‘other’ (1%; n=4) were rarely used by prisoners. Overall, since the RCIADIC in 1991, the range of materials used in hangings has not changed considerably. In an analysis of 30 hanging incidents, the Commissioners found that some of the most common possessions used included strips of sheeting or blanket, football socks and belts (RCIADIC 1991). Consistent with these previous trends, bed sheets were consistently the most commonly used material in hangings by prisoners from 1989–90 to 2010–11 (61%; n=232; see Table 22).

It was also noted by the Commissioners that it is not possible to remove certain material possessions from inmates; rather, it is necessary for cell architects to design out possible hanging points (RCIADIC 1991). As many of the items used by inmates in hangings are essential items, such as sheets, there are therefore additional challenges with regards to prevention planning.

The AIC is currently working with corrective services agencies to undertake some targeted research into the decline in hanging deaths in custody, including an examination of the significant investment made in redesigning cells and removing hanging points. This research will be released in the coming months.

Table 20 Natural cause deaths in prison custody by type of natural cause and year, 1979–80 to 2010–11 (n)
Hearta Cancerb Strokec Respiratoryd Digestivee Conditions/diseasesf Multiple/otherg Total
1979–80 3 2 1 0 0 0 1 7
1980–81 6 1 0 0 1 0 4 12
1981–82 3 0 0 2 0 0 1 6
1982–83 4 0 2 1 0 2 3 12
1983–84 3 0 0 1 0 0 1 5
1984–85 6 1 0 1 1 0 2 11
1985–86 6 0 0 0 0 0 0 6
1986–87 7 0 1 2 0 1 2 13
1987–88 3 0 1 1 1 3 0 9
1988–89 6 1 0 1 0 1 0 9
1989–90 6 2 1 2 0 1 2 14
1990–91 11 1 1 0 0 1 3 17
1991–92 2 1 1 3 1 0 0 8
1992–93 3 4 0 1 0 0 3 11
1993–94 10 3 0 3 0 0 3 19
1994–95 10 1 1 1 0 0 2 15
1995–96 8 0 1 1 0 2 2 14
1996–97 15 1 1 5 0 1 5 28
1997–98 10 4 0 3 0 1 3 21
1998–99 8 4 2 1 0 1 2 18
1999–2000 9 5 1 2 1 1 2 21
2000–01 14 4 0 4 0 1 1 24
2001–02 14 6 0 2 1 0 1 24
2002–03 10 2 2 3 0 0 4 21
2003–04 9 5 0 1 0 1 2 18
2004–05 7 6 3 2 1 2 2 23
2005–06 8 3 2 2 0 2 1 18
2006–07 12 7 0 0 0 2 6 27
2007–08 17 8 2 3 0 0 5 35
2008–09 7 8 3 0 2 4 2 26
2009–10 16 9 0 1 0 2 8 36
2010–11 10 10 1 4 0 2 7 34
Totalh 263 99 27 53 9 31 80 562

a: Includes ischaemic heart disease, coronary atherosclerosis, thrombosis and infarcts

b: Includes all forms of carcinoma

c: Includes cerebrovascular accidents and intracerebral haemorrhages

d: Includes asthma complications, pneumonia, pulmonary fibrosis and chronic obstructive airways disease

e: Includes esophagitis, gastritis and all other conditions effecting the digestive system

f: Includes diabetes, appendicitis, all forms of cirrhosis and any other form of non-infectious disease

g: Multiple/other causes includes persons dying from multiple natural causes, infectious diseases such as hepatitis or HIV/AIDS, epilepsy complications, cases not classifiable elsewhere and cases where the coroner is unable to determine the precise type of medical cause of death and therefore hands down a finding of ‘undetermined’

h: 22 cases have been excluded due to missing information about type of natural cause death

Table 21 Prison custody hanging deaths by hanging point used and year, 1979–80 to 2010–11 (n)
Cell bars Shower fixture Other cell fitting Fitting outside cell Bed bunk Other Missing Total
1979–80 0 0 0 0 0 0 3 3
1980–81 0 0 0 0 0 0 10 10
1981–82 0 0 0 0 0 0 9 9
1982–83 0 0 0 0 0 0 8 8
1983–84 0 0 0 0 0 0 20 20
1984–85 0 0 0 0 0 0 8 8
1985–86 0 0 0 0 0 0 8 8
1986–87 0 0 0 0 0 0 13 13
1987–88 0 0 0 0 0 0 21 21
1988–89 0 0 0 0 0 0 15 15
1989–90 8 0 1 0 0 0 18 27
1990–91 6 0 4 0 0 0 2 12
1991–92 4 1 9 0 0 0 1 15
1992–93 8 1 8 0 1 0 1 19
1993–94 11 4 7 1 1 0 0 24
1994–95 6 5 13 0 3 0 0 27
1995–96 9 0 5 2 1 0 2 19
1996–97 9 0 8 1 1 1 3 23
1997–98 5 10 20 0 3 0 1 39
1998–99 12 1 7 0 2 0 4 26
1999–2000 8 2 12 1 6 0 0 29
2000–01 8 3 8 1 4 0 1 25
2001–02 7 7 4 2 1 0 0 21
2002–03 6 3 9 0 0 1 0 19
2003–04 9 0 2 0 2 1 0 14
2004–05 6 1 5 0 1 0 0 13
2005–06 1 2 2 0 1 0 0 6
2006–07 1 1 3 0 3 0 0 8
2007–08 3 0 2 1 0 0 0 6
2008–09 3 2 4 0 2 1 0 12
2009–10 2 2 6 0 2 0 1 13
2010–11 1 2 6 2 0 0 2 13
Total 133 47 145 11 34 4 151 525

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Table 22 Prison custody hanging deaths by material used and year, 1979–80 to 2010–11 (n)
Sheets Shoelaces Belt Other clothing Rope/cord Other Missing Total
1979–80 0 0 0 0 0 0 3 3
1980–81 0 0 0 0 0 0 10 10
1981–82 0 0 0 0 0 0 9 9
1982–83 0 0 0 0 0 0 8 8
1983–84 0 0 0 0 0 0 20 20
1984–85 0 0 0 0 0 0 8 8
1985–86 0 0 0 0 0 0 8 8
1986–87 0 0 0 0 0 0 13 13
1987–88 0 0 0 0 0 0 21 21
1988–89 0 0 0 0 0 0 15 15
1989–90 6 0 0 1 0 2 18 27
1990–91 6 0 1 0 1 4 0 12
1991–92 8 0 1 1 4 1 0 15
1992–93 9 0 2 2 4 2 0 19
1993–94 17 1 1 4 0 1 0 24
1994–95 17 0 2 2 2 3 1 27
1995–96 13 0 1 1 2 1 1 19
1996–97 11 5 0 3 2 0 2 23
1997–98 22 4 1 1 6 1 4 39
1998–99 15 2 0 1 2 3 3 26
1999–2000 15 6 0 4 4 0 0 29
2000–01 16 2 0 2 1 3 1 25
2001–02 14 1 0 1 2 3 0 21
2002–03 12 4 0 1 1 1 0 19
2003–04 9 2 0 1 1 1 0 14
2004–05 9 0 0 2 1 1 0 13
2005–06 3 0 0 0 1 2 0 6
2006–07 4 0 0 4 0 0 0 8
2007–08 4 0 0 1 1 0 0 6
2008–09 8 0 0 0 3 1 0 12
2009–10 6 1 0 1 3 2 0 13
2010–11 8 0 0 1 1 1 2 13
Total 232 28 9 34 42 33 147 525

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Manner of death

Manner of death is closely related to cause of death and as a result, the findings may be similar. Manner of death is designed to capture the circumstances in which the person died and may be classed as either self-inflicted, due to natural causes, justifiable homicide, unlawful homicide, accidental or an other situation. Self-inflicted deaths include cases where the deceased intentionally killed themselves, or where there is insufficient information to determine the intent of the deceased (ie ‘Briginshaw’ test not met; see Briginshaw v Briginshaw HCA 34 (1938) 60 CLR 336 (30 June 1938)). For example, most hangings and deaths resulting from self-inflicted wounds are classified as self-inflicted deaths. Accidental deaths sometimes include deaths that result from acute substance toxicity, but only where the coroner finds the prisoner accidentally overdosed. It is important to note that some alcohol and drug-related deaths are classified as accidental until such time as the coroner determines whether the death was intentional or not; if deemed intentional, the case is reclassified as being a self-inflicted death. Finally, justifiable homicide refers to homicides that occur under circumstances authorised by law, for example, a prison officer acting in self-defence.

Between 2008–09 and 2010–11, there were a total of 157 deaths in prison custody where information was available about the manner of death (see Table 23). Of these, the most prevalent were natural causes (65%; n=102), followed by self-inflicted deaths (29%; n=45).

There were 33 Indigenous deaths over the period, the majority of which occurred as a result of natural causes (73%; n=24), with a further eight deaths being self-inflicted (24%). The only accidental deaths during this period occurred in 2008–09 and in 2010–11, of which one involved an Indigenous prisoner and three involved a non-Indigenous prisoner. All of the unlawful homicides (n=6) over the last three years took place in 2009–10 and in 2010–11, and all involved non-Indigenous prisoners.

For each specific year, the results were as follows:

  • in 2008–09, there were 43 deaths, of which 14 were self-inflicted (1 Indigenous and 13 non-Indigenous), 26 were natural causes (6 Indigenous and 20 non-Indigenous) and three were accidental (1 Indigenous and 2 non-Indigenous);
  • in 2009–10, there were 58 deaths, of which 16 were self-inflicted (3 Indigenous and 13 non-Indigenous), 38 were natural cause deaths (11 Indigenous and 27 non-Indigenous) and four were unlawful homicides (all non-Indigenous); and
  • in 2010–11, there were 56 deaths where information about cause of death is available. Of these, 15 were self-inflicted (4 Indigenous and 11 non-Indigenous), 38 were due to natural causes (8 Indigenous and 30 non-Indigenous), two were unlawful homicides (both non-Indigenous) and there was one accidental death of a non-Indigenous prisoner.

Trends

Since 1979–80, 43 percent (n=593) of all deaths in prison custody were self-inflicted, while a further 42 percent (n=584) were due to natural causes (see Table 24 and Figure 12). Over this period:

  • self-inflicted deaths and deaths due to natural causes have consistently been the two most frequent circumstances in which prisoners die each year;
  • deaths classified as an unlawful homicide, such as a prisoner fatally assaulting another prisoner, were more frequent between 1992–93 and 1999–2000, but have since decreased, with the exception of four cases recorded in 2009–10 and two in 2010–11; and
  • natural cause deaths have been increasing markedly in recent years, with the numbers of these deaths in 2009–10 and 2010–11 being the highest recorded.

As a proportion of all deaths recorded since 1979–80, natural causes have been increasing since 1997–98, while accidental and self-inflicted deaths, and those due to unlawful homicide, have decreased since 1999–2000 (see Figure 13).

Despite the decrease in self-inflicted deaths in prison in recent years, the overall number of self-inflicted deaths is slightly higher than for all other circumstances of death. According to a recent study by Fazel et al. (2011), which analysed 861 separate incidents of suicides in a prison setting from across 12 countries, the results showed that prisoners in general have a higher risk towards self-inflicted deaths. In making this conclusion, the authors recommend that more efforts should be focused towards including this group in national suicide prevention strategies.

Figure 12 Prison custody deaths by manner of death and Indigenous status, 1979–80 to 2010–11 (%)

Table 23 Prison custody deaths by manner of death and year, 2008–09 to 2010–11 (n)
2008–09
Self–inflicted Natural causes Unlawful homicide Accident Total
New South Wales 5 9 0 1 15
Victoria 2 5 0 1 8
Queensland 4 5 0 0 9
Western Australia 1 2 0 1 4
South Australia 1 4 0 0 5
Tasmania 0 1 0 0 1
Northern Territory 0 0 0 0 0
Australian Capital Territory 1 0 0 0 1
Totals
Indigenous 1 5 0 1 7
Non-Indigenous 13 21 0 2 36
All persons 14 26 0 3 43
2009–10
Self–inflicted Natural causes Unlawful homicide Accident Total
New South Wales 4 15 2 0 21
Victoria 3 4 1 0 8
Queensland 3 6 1 0 10
Western Australia 5 3 0 0 8
South Australia 0 5 0 0 5
Tasmania 0 3 0 0 3
Northern Territory 1 1 0 0 2
Australian Capital Territory 0 1 0 0 1
Totals
Indigenous 3 11 0 0 14
Non-Indigenous 13 27 4 0 44
All persons 16 38 4 0 58
2010–11
Self–inflicted Natural causes Unlawful homicide Accident Total
New South Wales 8 9 1 1 19
Victoria 1 8 1 0 10
Queensland 2 8 0 0 10
Western Australia 1 4 0 0 5
South Australia 2 2 0 0 4
Tasmania 0 2 0 0 2
Northern Territory 1 5 0 0 6
Australian Capital Territory 0 0 0 0 0
Totals
Indigenous 4 8 0 0 12
Non-Indigenous 11 30 2 1 44
All personsa 15 38 2 1 56

a: Manner of death is currently under investigation for 2 non-Indigenous prisoners

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Table 24 Prison custody deaths by manner of death and year, 1979–80 to 2010–11 (n)
Self–inflicted Natural causes Justifiable homicide Unlawful homicide Accident Other Total
1979–80 6 8 0 0 0 0 14
1980–81 10 13 1 2 1 2 29
1981–82 12 7 0 1 4 3 27
1982–83 11 12 0 0 2 0 25
1983–84 22 6 0 0 2 3 33
1984–85 10 12 0 2 2 1 27
1985–86 9 7 0 0 2 0 18
1986–87 15 15 0 2 3 2 37
1987–88 21 10 0 1 9 1 42
1988–89 16 11 0 2 6 1 36
1989–90 27 16 0 2 5 3 53
1990–91 15 17 0 1 2 0 35
1991–92 15 8 1 1 5 1 31
1992–93 24 11 1 3 4 0 43
1993–94 28 19 0 4 5 1 57
1994–95 27 15 0 6 7 0 55
1995–96 25 14 0 4 6 0 49
1996–97 24 28 0 4 7 0 63
1997–98 42 21 0 8 9 1 81
1998–99 27 18 0 8 11 0 64
1999–2000 32 22 1 3 5 1 64
2000–01 27 24 0 1 6 0 58
2001–02 24 24 1 4 0 0 54
2002–03 21 21 0 1 1 0 43
2003–04 17 19 0 2 1 0 39
2004–05 16 23 0 0 0 0 39
2005–06 9 18 0 1 0 0 28
2006–07 8 27 0 1 4 0 40
2007–08 8 36 0 0 1 1 46
2008–09 14 26 0 0 3 0 43
2009–10 16 38 0 4 0 0 58
2010–11 15 38 0 2 1 0 56
Totalª 593 584 5 70 114 21 1,387

a: 6 cases have been excluded due to missing data about manner of death

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Figure 13 Prison custody deaths by manner of death and year, 1979–80 to 2010–11 (%)

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Mental Illness

There is a growing body of evidence that suggests that mental illness, both diagnosed and undiagnosed, is more prevalent in the criminal justice system than in the general community (AIHW 2011c, 2010; Beyond Bars 2007; Ogloff et al. 2007; Senate Committee on Mental Health 2006). In 2003, the NSW Department of Corrective Services conducted an inmate health survey, which found that

54 percent of women and 39 percent of men in prison have at some point in their lives been diagnosed by a medical doctor as having a ‘psychiatric problem’ (Butler & Milner cited in Beyond Bars 2007: 1).

This finding is also reflected in more recent research. Using data from the National Prisoner Health Census 2010 (which excluded New South Wales and Victoria), the Australian Institute of Health and Welfare (AIHW) observed that 31 percent (n=192) of the 610 prison entrants surveyed indicated that they had been told they had a mental illness at some stage in their life, with 16 percent (n=97) indicating that they were currently on mental health medication (AIHW 2011b).

During the process of reviewing more than 1,100 historical cases as part of the AIC’s review of the NDICP in 2011, information was extracted from coronial findings regarding whether the deceased had a history of mental illness and/or was suffering a mental illness when they died. Findings relating to the prevalence of mental illness and the type of disorders reported among people who died in prison custody between 1989–90 and 2010–11 are presented in Tables 25 and 26. Information is only presented for those cases where it was clearly stated in the coronial finding that the deceased had a mental illness and the specific condition was recorded, and then coded into one of the following 11 categories:

  • adjustment disorders;
  • anxiety disorders;
  • mood disorders;
  • personality disorders;
  • psychotic disorders;
  • sexual disorders;
  • somatoform disorders;
  • substance disorder;
  • developmental disorders;
  • intellectual disability; or
  • other/unknown/not stated.

These 11 categories were derived from those developed by the World Health Organisation’s International Statistical Classification of Diseases and Related Health Problems, 10th edition. It should be noted that there is a great deal of missing information in relation to these variables. For all deaths in prison custody between 1989–90 and 2010–11 (n=1,101), there were 299 cases (27%) where the coronial finding indicated that the deceased had a mental illness. For those persons who died in prison custody where the available information did not indicate the presence of a mental health issue, the lack of available information should not be taken as confirmation that the person had no mental health issues. It is possible that individuals had mental health issues but that these issues were not diagnosed and/or were not recorded in the available information. For this reason, these data should be treated as a conservative estimate of the prevalence of mental illness among those persons who died in prison custody. Finally, the assignment of mental illness, disorders or conditions to the 11 categories was made by NDICP research staff based on often limited available information and should also be treated with caution.

Table 25 shows that the prevalence of mental illness each year ranged between seven and 62 percent for Indigenous deaths and between eight and 48 percent for non-Indigenous deaths. Proportions were calculated by dividing the number of persons with a mental illness by the total number of deaths in the cohort each year (ie Indigenous deaths of prisoners with a mental illness divided by the number of Indigenous deaths in prison custody each year). Overall, more than one-quarter (29% for Indigenous and 27% for non-Indigenous) of persons who died in prison custody were reported as suffering from some form of mental illness before their death.

Mood disorders, predominantly depression, were the most prevalent among non-Indigenous persons (42%; n=100), followed by psychotic disorders such as schizophrenia (20%; n=47) and other/unknown disorders (13%; n=32; see Table 26). For Indigenous prisoner deaths, psychotic disorders were the most prevalent (33%; n=19), closely followed by mood disorders (29%; n=17) and other/unknown mental illnesses (21%; n=12). These data provide a conservative estimate of the prevalence of mental illness, yet mental illness appeared to be a factor in just over one in every four deaths in prison custody. The AIC will continue to work closely with corrective services agencies to monitor this issue.

Table 25 Deaths in prison custody of persons with a mental illness by Indigenous status and year (as a proportion of all deaths), 1989–90 to 2010–11 (n)
Indigenous Non-Indigenous Total
n % n % n %
1989–90 1 11 4 9 5 9
1990–91 2 40 9 30 11 31
1991–92 2 50 7 26 9 29
1992–93 3 60 11 29 14 33
1993–94 1 8 11 24 12 21
1994–95 6 50 14 33 20 36
1995–96 2 15 9 26 11 23
1996–97 2 18 10 19 12 19
1997–98 2 17 16 23 18 22
1998–99 4 40 8 15 12 18
1999–2000 5 36 13 26 18 28
2000–01 7 50 13 29 20 34
2001–02 2 33 14 29 16 30
2002–03 5 42 12 38 17 39
2003–04 1 17 16 48 17 44
2004–05 4 50 14 45 18 46
2005–06 0 0 7 28 7 25
2006–07 5 63 11 34 16 40
2007–08 0 0 15 38 15 33
2008–09 1 14 16 44 17 40
2009–10 1 7 7 16 8 14
2010–11 2 17 4 9 6 10
Total 58 29 241 27 299 27

Source: AIC NDICP 1989–90 to 2010–11 [computer file]

Most serious offence/charge

  • The NDICP collects information on the most serious offence or charge for which the person is held in prison. The offences are grouped into six categories from most to least serious; that is, violent offences, theft-related, drug-related, traffic, good order and other. For the purposes of the NDICP, these groups are defined as:
  • violent offence—includes homicide, assault, sex offences, other offences against the person and robbery;
  • theft-related offence—includes break and enter, other theft, property damage and fraud;
  • drug-related offence—includes possessing, using, dealing, trafficking and manufacturing/growing drugs;
  • traffic offence—includes road traffic, driving and license offences;
  • good order offence—includes justice procedure offences, breaches of sentences (including fine default), public drunkenness, protective custody for intoxication in jurisdictions where public drunkenness is not an offence and other offences against good order (eg prostitution, betting and gambling, disorderly conduct, vagrancy and offensive behaviour); and
  • other—includes other offences not elsewhere classified or where the most serious offence is unknown.

Over half (59%; n=93) of the 158 deaths in prison custody between 2008–09 and 2010–11 (where offending was known) involved persons who were detained for a violence-related offence (see Table 27). This was followed by those with theft (15%; n=24), ‘other’ (14%; n=22), drug (7%; n=11), traffic (3%; n=5) and good order offences (2%; n=3). Of the 33 deaths involving Indigenous prisoners, the most common offence was violent (58%; n=19), with ‘other’ offences, such as breaching a domestic violence order or weapons offences, being the second most prevalent (18%; n=6). Throughout the reporting period, there were only three deaths of prisoners who had a most serious offence/charge of a good order-related offence and all were non-Indigenous.

Table 26 Deaths in prison custody of persons with a known mental illness by type of disorder, Indigenous status and year, 1989–90 to 2010–11 (n)
Adjustment Anxiety Mood Personality Psychotic Sexual Somatoform Substance Developmental Intellectual disability Other/unknown Total
I N I N I N I N I N I N I N I N I N I N I N I N
1989–90 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 2 1 4
1990–91 0 0 0 1 1 4 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 2 2 9
1991–92 0 0 0 0 1 4 0 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 2 7
1992–93 0 1 0 0 1 1 1 2 1 1 0 0 0 0 0 0 0 0 0 0 0 6 3 11
1993–94 0 1 0 0 0 4 0 2 1 3 0 0 0 0 0 0 0 0 0 1 0 0 1 11
1994–95 0 0 0 0 3 6 0 2 2 2 0 0 0 0 0 0 0 0 0 0 1 4 6 14
1995–96 0 0 0 0 0 4 0 0 1 2 0 0 0 0 0 0 0 0 0 0 1 3 2 9
1996–97 0 0 0 1 0 2 0 3 0 1 0 0 0 0 0 0 0 0 0 1 2 2 2 10
1997–98 0 0 0 0 2 9 0 3 0 1 0 0 0 0 0 1 0 0 0 0 0 2 2 16
1998–99 0 0 0 0 0 7 0 0 2 1 0 0 0 0 0 0 1 0 0 0 1 0 4 8
1999–2000 0 0 0 0 0 5 0 1 2 4 0 0 0 0 1 0 0 1 0 0 2 2 5 13
2000–01 0 0 0 0 3 4 0 1 3 1 0 0 0 0 0 5 0 0 0 0 1 2 7 13
2001–02 0 0 0 0 1 7 0 1 1 3 0 0 0 0 0 2 0 0 0 0 0 1 2 14
2002–03 0 1 0 0 1 4 0 2 3 2 0 0 0 0 1 1 0 0 0 0 0 2 5 12
2003–04 0 0 0 3 0 7 0 1 0 3 0 0 0 0 1 2 0 0 0 0 0 0 1 16
2004–05 0 0 0 0 1 5 0 1 0 6 1 0 0 0 1 1 0 0 0 0 1 1 4 14
2005–06 0 0 0 0 0 5 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 7
2006–07 0 0 0 0 2 2 1 0 2 3 0 0 0 0 0 5 0 0 0 0 0 1 5 11
2007–08 0 0 0 2 0 4 0 0 0 4 0 0 0 1 0 2 0 0 0 1 0 1 0 15
2008–09 0 0 0 1 0 8 0 0 0 5 1 0 0 0 0 2 0 0 0 0 0 0 1 16
2009–10 0 0 0 0 0 5 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 1 7
2010–11 0 0 0 0 1 2 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 2 4
Total 0 3 0 8 17 100 2 22 19 47 2 1 0 1 5 23 1 1 0 3 12 32 58 241

Note: I=Indigenous; N=non-Indigenous

Source: AIC NDICP 1989–90 to 2010–11 [computer file]

Trends

Analysis of data collected by the NDICP showed that since 1979–80, just over half (53%; n=722) of the deaths in prison custody involved persons who were imprisoned for a violent offence (see Table 28 and Figure 14). Further, more than one-quarter (28%; n=378) of all deaths were of prisoners who had been imprisoned for theft-related offences. Violent offences have consistently been the most serious offence committed by the deceased prior to incarceration each year since 1993–94.

Overall, the number of deaths of prisoners who have violent offences as their most serious offence has increased considerably, reaching an historical high in 2009–10 (n=39). Conversely, the number of deaths of prisoners with theft-related offences as their most serious offence has remained relatively low over the last decade. This trend is directly linked to changes in the proportion of prisoners detained for violent and theft-related offences over the last 30 years. Data from the ABS showed that the proportion of prisoners detained for violent offences has increased from 29 percent in 1981–82 to 46 percent in 2010–11 (ABS-A 2011–1981). At the same time, the proportion of those prisoners held on theft-related offences has dropped from 46 percent to 27 percent over the same period (ABS-A 2011–1981).

The rates of death since 1981–82 have generally fluctuated between 0.10 and 0.50 deaths per 100 prisoners (with some minor exceptions) for the four most common types of offence category—violent, theft-related, drug-related and good order offences (see Figure 15). The trends for specific offences show that:

  • rates of violent offence deaths have been relatively constant over the years;
  • rates of deaths among theft-related offenders fluctuated until the late 1990s, but have since declined; and
  • rates of drug-related offence deaths showed some fluctuation until 1997 but have tended to decrease since then.

Figure 14 Prison custody deaths by most serious offence and Indigenous status, 1979–80 to 2010–11 (%)

Table 27 Prison custody deaths by most serious offence and year, 2008–09 to 2010–11 (n)
2008–09
Violent Theft-related Drug-related Good order offence Traffic Other Total
New South Wales 5 5 2 1 0 2 15
Victoria 6 2 0 0 0 0 8
Queensland 7 0 1 0 0 1 9
Western Australia 2 1 0 0 0 1 4
South Australia 5 0 0 0 0 0 5
Tasmania 1 0 0 0 0 0 1
Northern Territory 0 0 0 0 0 0 0
Australian Capital Territory 0 0 0 0 0 1 1
Totals
Indigenous 3 1 2 0 0 1 7
Non-Indigenous 23 7 1 1 0 4 36
All persons 26 8 3 1 0 5 43
2009–10
Violent Theft-related Drug-related Good order offence Traffic Other Total
New South Wales 13 2 1 0 2 3 21
Victoria 6 1 0 0 1 0 8
Queensland 7 2 0 0 0 1 10
Western Australia 5 0 0 0 2 1 8
South Australia 4 0 1 0 0 0 5
Tasmania 3 0 0 0 0 0 3
Northern Territory 0 0 1 0 0 1 2
Australian Capital Territory 1 0 0 0 0 0 1
Totals
Indigenous 9 2 1 0 1 1 14
Non-Indigenous 30 3 2 0 4 5 44
All persons 39 5 3 0 5 6 58
2010–11
Violent Theft-related Drug-related Good order offence Traffic Other Total
New South Wales 10 5 2 2 0 1 20
Victoria 5 2 2 0 0 0 9
Queensland 4 2 0 0 0 5 11
Western Australia 3 1 1 0 0 0 5
South Australia 1 1 0 0 0 2 4
Tasmania 0 0 0 0 0 2 2
Northern Territory 5 0 0 0 0 1 6
Australian Capital Territory 0 0 0 0 0 0 0
Totals
Indigenous 7 1 0 0 0 4 12
Non-Indigenous 21 10 5 2 0 7 45
All personsa 28 11 5 2 0 11 57

a: 1 non-Indigenous case was excluded due to missing data about the offence committed

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Table 28 Prison custody deaths by most serious offence and year, 1979–80 to 2010–11 (n)
Violent Theft-related Drug-related Traffic Good order Other Total
1979–80 8 3 1 1 1 1 15
1980–81 10 7 3 2 5 1 28
1981–82 9 11 3 0 3 0 26
1982–83 11 6 1 2 3 1 24
1983–84 14 13 0 0 6 0 33
1984–85 13 9 2 2 0 0 26
1985–86 8 6 1 0 1 0 16
1986–87 14 10 2 1 5 0 32
1987–88 17 18 2 1 3 2 43
1988–89 22 10 1 1 1 1 36
1989–90 28 16 3 0 2 2 51
1990–91 19 10 2 0 1 3 35
1991–92 16 8 2 0 4 1 31
1992–93 18 19 3 0 2 1 43
1993–94 25 22 2 2 5 1 57
1994–95 28 18 4 2 2 1 55
1995–96 21 20 5 2 0 1 49
1996–97 29 18 7 2 4 3 63
1997–98 38 24 11 1 6 0 80
1998–99 27 26 5 1 5 0 64
1999–2000 33 24 2 2 1 2 64
2000–01 33 9 7 2 54 3 58
2001–02 28 11 6 2 4 2 53
2002–03 32 7 0 2 2 1 44
2003–04 30 2 3 1 1 2 39
2004–05 23 9 3 3 0 1 39
2005–06 22 4 1 0 1 0 28
2006–07 22 9 3 2 4 0 40
2007–08 31 5 4 3 3 0 46
2008–09 26 8 3 0 1 5 43
2009–10 39 5 3 5 0 6 58
2010–11 28 11 5 0 2 11 57
Totalª 722 378 100 42 82 52 1,376

a: 17 cases have been excluded due to missing data about offending

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Figure 15 Prison custody deaths by most serious offence and year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E5

Rate per 100 relevant prisoners at 30 June each year (ABS-A 2011–1981)

Note: Cases for which traffic and other offences were recorded as the most serious offence have been excluded due to small numbers

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Location of death

The two most common locations for a death to occur in prison were cells (40%; n=63) followed closely by public hospitals (39%; n=62; see Table 29). However, when comparing the location of deaths across the three years, there was variation in the number of deaths between prison cells and public hospitals, in that in both 2008–09 and 2009–10 there were more deaths in cells, while in 2010–11 more deaths occurred in a public hospital.

The Northern Territory was the only jurisdiction not to have a death in a prison cell. Of the 33 deaths involving Indigenous prisoners, just over half (55%; n=18) died in a public hospital, while more than one in three (36%; n=12) died in a prison cell. The remaining three (9%) Indigenous prisoner deaths occurred in a prison hospital.

Deaths in a custodial setting, such as exercise yard or common area, have been the least common over the last three years, with only five such deaths occurring.

For each year in the reporting period, the results were as follows:

  • of the 43 deaths in 2008–09, 18 (42%) occurred in a prison cell, 17 (40%) occurred in a public hospital, seven (16%) in a prison hospital and one (2%) death in a custodial setting;
  • of the 58 deaths in 2009–10, there were 24 (41%) in a prison cell, 18 (31%) in a public hospital, 11 (19%) in a prison hospital, four deaths in a custodial setting and one (2%) in ‘other’ (which includes ambulances); and
  • in 2010–11, there were 58 deaths, of which 27 (47%) occurred in a public hospital, 21 (36%) in a prison cell and 10 (17%) in a prison hospital.
Table 29 Prison custody deaths by location and year, 2008–09 to 2010–11 (n)
2008–09
Public hospital Prison hospital Cell Custodial setting Other Total
New South Wales 6 3 6 0 0 15
Victoria 1 4 2 1 0 8
Queensland 3 0 6 0 0 9
Western Australia 2 0 2 0 0 4
South Australia 3 0 2 0 0 5
Tasmania 1 0 0 0 0 1
Northern Territory 0 0 0 0 0 0
Australian Capital Territory 1 0 0 0 0 1
Totals
Indigenous 4 0 3 0 0 7
Non-Indigenous 13 7 15 1 0 36
All persons 17 7 18 1 0 43
2009–10
Public hospital Prison hospital Cell Custodial setting Other Total
New South Wales 3 8 9 1 0 21
Victoria 2 2 2 1 1 8
Queensland 4 1 3 2 0 10
Western Australia 1 0 7 0 0 8
South Australia 4 0 1 0 0 5
Tasmania 2 0 1 0 0 3
Northern Territory 2 0 0 0 0 2
Australian Capital Territory 0 0 1 0 0 1
Totals
Indigenous 7 2 5 0 0 14
Non-Indigenous 11 9 19 4 1 44
All persons 18 11 24 4 1 58
2010–11
Public hospital Prison hospital Cell Custodial setting Other Total
New South Wales 6 3 11 0 0 20
Victoria 3 5 2 0 0 10
Queensland 5 2 4 0 0 11
Western Australia 3 0 2 0 0 5
South Australia 2 0 2 0 0 4
Tasmania 2 0 0 0 0 2
Northern Territory 6 0 0 0 0 6
Australian Capital Territory 0 0 0 0 0 0
Totals
Indigenous 7 1 4 0 0 12
Non-Indigenous 20 9 17 0 0 46
All persons 27 10 21 0 0 58

Source: AIC NDICP 2008–09 to 2010–11 [computer file]

Trends

Since 1979–80, over half of all prison deaths (55%; n=725) have occurred in a cell (see Table 30). The proportion of deaths occurring in cells is the same for both Indigenous (55%; n=125) and non-Indigenous (55%; n=601) prisoners. In trend terms, the number of prisoner deaths in cells started to rise in the late 1980s, reaching an historical high of 53 deaths in 1997–98, but have gradually decreased since then. Despite being the second most common location of death (30%; n=394), the number of prisoners dying in a public hospital each year has remained generally consistent over the last three decades.

Table 30 Prison custody deaths by location, 1979–80 to 2010–11 (n)
Public hospital Prison hospital Cell Custodial setting Private property Public place Other custodial settinga Total
1979–80 7 0 5 0 0 0 0 12
1980–81 6 0 14 0 0 0 0 20
1981–82 9 0 5 0 0 0 0 14
1982–83 8 0 13 0 0 0 0 21
1983–84 6 0 23 0 0 0 0 29
1984–85 10 0 10 0 0 0 0 20
1985–86 4 0 9 0 0 0 0 13
1986–87 14 0 15 0 0 0 0 29
1987–88 8 0 26 0 0 0 0 34
1988–89 11 0 20 0 0 0 0 31
1989–90 10 0 31 0 0 0 0 41
1990–91 14 0 21 0 0 0 0 35
1991–92 9 2 18 1 0 1 0 31
1992–93 12 3 26 2 0 0 0 43
1993–94 11 10 34 2 0 0 0 57
1994–95 11 7 33 3 0 0 1 55
1995–96 9 1 34 4 0 0 1 49
1996–97 20 5 32 4 0 1 1 63
1997–98 17 7 53 4 0 0 0 81
1998–99 9 5 41 8 0 0 1 64
1999–2000 14 7 38 5 0 0 0 64
2000–01 18 1 35 2 0 2 0 58
2001–02 18 1 27 6 1 0 0 53
2002–03 13 5 21 4 0 1 0 44
2003–04 10 4 22 2 1 0 0 39
2004–05 17 5 16 1 0 0 0 39
2005–06 7 3 14 4 0 0 0 28
2006–07 14 10 14 2 0 0 0 40
2007–08 16 12 12 5 0 0 1 46
2008–09 17 7 18 1 0 0 0 43
2009–10 18 11 24 4 0 0 1 58
2010–11 27 10 21 0 0 0 0 58
Totalb 394 116 725 65 2 5 6 1,312

a: Other custodial setting includes 1 death, which occurred in a secure psychiatric hospital

b: 81 cases have been excluded due to missing data

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Legal status of prisoners who died in custody

Of the 159 deaths over the reporting period, 108 (68%) were of sentenced prisoners and 51 (32%) were of unsentenced prisoners (see Table 31). In each year of the reporting period, the rate of death for sentenced prisoners was lower than for unsentenced.

For each year in the reporting period, the results were as follows:

  • 25 (58%) of the 43 deaths in 2008–09 were of sentenced prisoners and 18 (42%) were of unsentenced prisoners. A much greater proportion of sentenced Indigenous prisoners died than the equivalent sentenced non-Indigenous prisoners (86%; n=6 cf 53%; n=19);
  • there were 58 deaths in prison custody in 2009–10, of which 42 (72%) were sentenced prisoners and 16 (28%) were unsentenced. Similar to the previous year, although not as pronounced, a greater proportion of sentenced Indigenous prisoners died when compared with non-Indigenous prisoners (79%; n=11 cf 71%; n=31); and
  • 41 (71%) of the 58 deaths in 2010–11 were sentenced and 17 (29%) were unsentenced. In line with the previous year, a greater proportion of sentenced Indigenous prisoners died than non-Indigenous prisoners (75%; n=9 cf 70%; n=32).
  • Table 31 Prison custody deaths by legal status and year, 2008–09 to 2010–11 (rate per 100 prisoners)a
    2008–09
    Sentenced Unsentenced Persons
    n Rate n Rate Total n Rate
    New South Wales 8 0.11 7 0.27 15 0.15
    Victoria 5 0.15 3 0.35 8 0.19
    Queensland 4 0.09 5 0.42 9 0.16
    Western Australia 3 0.09 1 0.13 4 0.10
    South Australia 4 0.31 1 0.15 5 0.26
    Tasmania 1 0.24 0 1 0.19
    Northern Territory 0 0 0
    Australian Capital Territory 0 1 1.04 1 0.88
    Australia 25 0.12 18 0.28 43 0.16
    2009–10
    Sentenced Unsentenced Persons
    n Rate n Rate Total n Rate
    New South Wales 16 0.21 5 0.18 21 0.20
    Victoria 6 0.17 2 0.23 8 0.18
    Queensland 7 0.16 3 0.26 10 0.18
    Western Australia 4 0.10 4 0.53 8 0.17
    South Australia 5 0.39 0 5 0.25
    Tasmania 3 0.78 0 3 0.61
    Northern Territory 1 0.12 1 0.41 2 0.19
    Australian Capital Territory 0 1 1.28 1 0.53
    Australia 42 0.19 16 0.24 58 0.20
    2010–11
    Sentenced Unsentenced Persons
    n Rate n Rate Total n Rate
    New South Wales 14 0.19 6 0.22 20 0.20
    Victoria 7 0.19 3 0.35 10 0.22
    Queensland 8 0.18 3 0.26 11 0.20
    Western Australia 4 0.10 1 0.12 5 0.11
    South Australia 2 0.15 2 0.30 4 0.20
    Tasmania 2 0.52 0 2 0.42
    Northern Territory 4 0.45 2 0.73 6 0.52
    Australian Capital Territory 0 0 0
    Australia 41 0.19 17 0.26 58 0.20

    a: Denominator: see Table E6

    Rate per 100 relevant prisoners (ABS 2012)

    Source: AIC NDICP 2008–09 to 2010–11 [computer file]

Trends

Of the total number of persons who died in prison custody since 1979–80, over two-thirds (68%; n=942) were sentenced. There has been a notable rise in the number of prison custody deaths of sentenced prisoners since 1992–93. The increase among this cohort is closely associated with increasing numbers of overall prisoners in custody, along with increasing numbers of deaths from natural causes. Analysis showed that more than eight out of every 10 natural cause deaths was of a sentenced prisoner (85%; n=492). Also of note was the fact that similar proportions of sentenced Indigenous and non-Indigenous prisoners died from natural causes (86%; n=108 for Indigenous cf 85%; n=384 for non-Indigenous).

Overall, 72 percent of Indigenous prisoners (n=170) and 68 percent of non-Indigenous prisoners (n=772) who died in custody since 1979–80 were sentenced at the time of their death. The remaining 15 deaths had missing information about legal status at time of death. The overall proportion of sentenced prisoners who died each year fluctuated between 60 and 80 percent with some minor exceptions (see Table 32).

Table 32 Prison custody deaths by legal status and year, 1979–80 to 2010–11 (n)
Sentenced Unsentenced Total
1979–80 11 4 15
1980–81 20 8 28
1981–82 18 8 26
1982–83 19 6 25
1983–84 20 13 33
1984–85 23 3 26
1985–86 9 6 15
1986–87 25 9 34
1987–88 29 14 43
1988–89 26 9 35
1989–90 35 15 50
1990–91 22 13 35
1991–92 21 10 31
1992–93 29 14 43
1993–94 38 19 57
1994–95 38 17 55
1995–96 36 13 49
1996–97 46 17 63
1997–98 48 33 81
1998–99 45 19 64
1999–2000 41 23 64
2000–01 42 16 58
2001–02 37 16 53
2002–03 25 19 44
2003–04 26 13 39
2004–05 21 18 39
2005–06 22 6 28
2006–07 29 11 40
2007–08 33 13 46
2008–09 25 18 43
2009–10 42 16 58
2010–11 41 17 58
Totala 942 436 1,378

a: 15 cases have been excluded due to missing data

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Analysis showed that the rate of death among sentenced prisoners has remained quite stable over the last 30 years and has never increased above 0.30 deaths per 100 sentenced prisoners (see Figure 16). Considerable fluctuations can be seen in the rate of death of unsentenced prisoners, particularly over the period from 1981–82 to 1998–99. However, since the year 2000, there has been a marked decline in the rate of death, to levels similar to that seen among sentenced prisoners. This decline in the rate is partly driven by a decline in the proportion of self-inflicted deaths among unsentenced prisoners. Over the decade from 1989–90 to 1999–2000, almost three-quarters of all deaths of unsentenced prisoners were self-inflicted (72%; n=139), whereas over the decade from 2000–01 to 2010–11, the proportion of such deaths has dropped to 64 percent (n=105). Another factor contributing to the decline is a reduction in deaths of unsentenced prisoners from drug and/or alcohol toxicity between the two decades (9%; n=17 in the decade 1989–90 to 1999–2000 cf 3%; n=5 in the decade from 2000–01 to 2010–11).

Figure 16 Prison custody deaths by legal status and year, 1981–82 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E6

Rate per 100 relevant prisoners (ABS-A 1998–1981 and ABS-B 2011–1999)

Source: AIC NDICP 1981–82 to 2010–11 [computer file]

Type of prison in which deaths occurred: Government versus private prisons

When interpreting the data presented in this section, it is important to note that:

  • The prison system in each jurisdiction is designed and managed as a whole to securely and safely accommodate prisoners with a range of risks and needs. Prison systems are planned to meet a range of security requirements and the diverse needs of the total prisoner population.
  • The function and security level of each prison is determined centrally by the corrective services agency, taking into account the overall requirements of the system and is not uniformly allocated between the public and private operators.
  • Depending on their function in the overall system, some prisons will house more ‘at risk’ prisoners than others, including seriously and terminally ill prisoners accommodated in a medical or hospital unit, which may have an impact on the numbers of deaths that occur at a particular prison or type of prison.
  • Those prisons that are privately managed have no ability to select which prisoners will be placed in their prisons. Prisoner classification and placement is the responsibility of the corrective services agency centrally.
  • Both public and private prisons are subject to the same standards, requirements, internal and external accountability mechanisms, and scrutiny in managing prisons.

The first private prison was established in Australia in January 1990, being a 240 bed medium-security institution for convicted offenders in Borallon, Queensland (Harding 1997). Over the following 11 years, a further seven private prisons came into operation in Australia. In 2009, a ninth prison, Parklea in New South Wales, was privatised and a tenth was opened in Southern Queensland in January 2012.

Information about private prisons across Australia is presented in Table 33, including the year in which they became private, their prisoner capacity and security level.

Available data from the Steering Committee for the Review of Commonwealth/State Service Provision indicated that since 1998–99, the proportion of Australia’s prison population held in privately managed prisons ranged from 15 to 19 percent each year (SCRCSP 2012). It is also important to remember that rates of death are likely to fluctuate more dramatically for private prison deaths owing to the comparatively small number of prisoners and prisoner deaths recorded.

During the last three years, the distribution of deaths between prisons has largely reflected the distribution of the prison population. Given that the majority of prisoners are housed in government prisons, it is not surprising that the majority of deaths occurred in these prisons (81%; n=129), while the remainder took place in private prisons (19%; n=30; see Table 34). For example, in 2010–11, just under one in five prisoners in Australia was held in a private prison (19%; SCRCSP 2012) and one in five (21%; n=12) prison custody deaths occurred in a private prison.

  • of the 43 deaths in prison custody in 2008–09, 11 occurred in privately run prisons. The annual rate of death for prisoners held in private prisons was 0.25 per 100, compared with 0.14 for prisoners in government-run facilities;
  • in 2009–10, there were 58 deaths in prison custody, with seven (12%) in a private prison. The rate of death for prisoners in private prisons was 0.14 per 100, compared with 0.21 for prisoners in government prisons; and
  • in 2010–10, there were 58 deaths in prison custody, with 12 occurring (21%) in a private prison. The annual rate of death for prisoners in private prisons was 0.22 per 100, compared with 0.20 for prisoners in government prisons.
Table 33 Private prisons in Australia
Privately managed since Prison State Prisoner capacity Security level
1990–2012 Borallon CC Queensland 492 High and medium
1992 Arthur Gorrie CC Queensland 908 Medium and maximum
1993 Junee CC New South Wales 790 Minimum and medium
1995 Mount Gambier Prison South Australia 172 Low and medium
1996–2000 Deer Park Metropolitan Women’s CC Victoria 300 Maximum
1997 Port Phillip Prison Victoria 804 Maximum
1997 Fulham CC Victoria 820 Minimum and medium
2001 Acacia Prison Western Australia 995 Medium
2009 Parklea CC New South Wales 828 Minimum and maximum
2012 Southern Queensland CC Queensland 300 High and low

Note: CC=correctional centre

Table 34 Prison custody deaths by type of prison and year, 2008–09 to 2010–11 (rate per 100 prisoners)ª
2008–09
Private Government All prisons
n Rate n Rate Total n Rate
New South Wales 1 0.13 14 0.15 15 0.15
Victoria 5 0.34 3 0.11 8 0.19
Queensland 4 0.32 5 0.11 9 0.16
Western Australia 0 4 0.13 4 0.10
South Australia 1 0.68 4 0.22 5 0.26
Tasmania 0 1 0.19 1 0.19
Northern Territory 0 0 0 0.00
Australian Capital Territory 0 1 0.85 1 0.85
Total 11 0.25 32 0.14 43 0.16
2009–10
Private Government All prisons
n Rate n Rate Total n Rate
New South Wales 1 0.08 20 0.22 21 0.20
Victoria 3 0.20 5 0.17 8 0.18
Queensland 2 0.15 8 0.19 10 0.18
Western Australia 1 0.10 7 0.18 8 0.17
South Australia 0 5 0.28 5 0.25
Tasmania 0 3 0.61 3 0.61
Northern Territory 0 2 0.19 2 0.19
Australian Capital Territory 0 1 0.53 1 0.53
Total 7 0.14 51 0.21 58 0.20
2010–11
Private Government All prisons
n Rate n Rate Total n Rate
New South Wales 4 0.27 16 0.19 20 0.20
Victoria 5 0.33 5 0.16 10 0.22
Queensland 3 0.22 8 0.19 11 0.20
Western Australia 0 5 0.14 5 0.11
South Australia 0 4 0.22 4 0.20
Tasmania 0 2 0.42 2 0.42
Northern Territory 0 6 0.51 6 0.51
Australian Capital Territory 0 0 0
Total 12 0.22 46 0.20 58 0.20

a: Denominator: see Table E10

Rate per 100 relevant prisoners (SCRCSP 2012)

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Indigenous status

After disaggregating the data by Indigenous status and type of prison, it can be seen that a smaller proportion of deaths in private prisons involve Indigenous prisoners when compared with government prisons (11%; n=14 cf 18%; n=244, see Table 35).

Table 35 Prison custody deaths by type of prison, Indigenous status and year, 1979–80 to 2010–11 (n)
Private Government Total
Indigenous Non-Indigenous Indigenous Non-Indigenous Indigenous Non-Indigenous
1979–80 0 0 4 11 4 11
1980–81 0 0 2 27 2 27
1981–82 0 0 3 24 3 24
1982–83 0 0 4 21 4 21
1983–84 0 0 2 31 2 31
1984–85 0 0 5 22 5 22
1985–86 0 0 4 14 4 14
1986–87 0 0 2 35 2 35
1987–88 0 0 4 40 4 40
1988–89 0 0 5 32 5 32
1989–90 0 0 9 44 9 44
1990–91 0 0 5 30 5 30
1991–92 0 0 4 27 4 27
1992–93 1 2 4 36 5 38
1993–94 1 4 11 41 12 45
1994–95 1 1 11 42 12 43
1995–96 1 4 12 31 13 35
1996–97 0 4 11 48 11 52
1997–98 0 10 12 59 12 69
1998–99 2 8 8 47 10 55
1999–2000 1 4 13 47 14 51
2000–01 1 9 13 36 14 45
2001–02 1 6 5 42 6 48
2002–03 0 5 12 27 12 32
2003–04 0 4 6 29 6 33
2004–05 1 5 7 26 8 31
2005–06 0 5 3 21 3 26
2006–07 0 4 8 28 8 32
2007–08 0 18 6 22 6 40
2008–09 0 11 7 25 7 36
2009–10 3 4 11 40 14 44
2010–11 1 11 11 35 12 46
Total 14 119 224 1,040 238 1,159

Source: AIC NDICP 1979–80 to 2010–11 [computer file]

Trends

Of the 133 total deaths in private prisons, 68 deaths (51%) were attributed to natural causes, while 42 were due to hanging (32%). Of the remaining 23 deaths, 12 resulted from acute substance toxicity (9%), nine were due to external/multiple trauma (7%) and one was from a head injury (1%); the cause of death is still being investigated in one case.

Of particular interest when comparing deaths in private and government-run prisons is the age at death, in that a greater proportion of deaths in private prisons were of prisoners aged 40 years or older (58%; n=77 cf 40%; n=499). Also of note is the proportion of deaths resulting from self-inflicted injuries, which was slightly lower for privately managed prisons (37%; n=49 cf 43%; n=544).

The rates of death have been decreasing over time for both types of facilities; however, the rate of death in private prisons spiked dramatically between 2007–08 and 2008–09 (see Figure 17). On closer examination of the data, it was revealed that this spike in the rate of death in private prisons was the result of a rise in the number of natural cause deaths among non-Indigenous prisoners (14 of the 18 deaths in 2007–08 were natural cause deaths of non-Indigenous prisoners). Overall, of the 68 deaths due to natural causes in private prisons, almost one-third (32%; n=22) occurred between 2007–08 and 2008–09.

In the early years of prison privatisation in Australia, there was some concern that the quality of care and range of services available in private prisons might be less than that in government-run facilities, based on the rationale that private contractors would be solely concerned with profit maximisation. For example, it was observed at the time that ‘incarceration policies may be driven by a private sector lobby intent on maximising imprisonment levels and thus the opportunity for profitable participation’ (Harding 1992: 3).

In a follow-up piece six years later, Harding found that with an effective regulatory regime ‘for ensuring that the private operators do what is expected of them’ and effective accountability ‘for ensuring that what is expected of them is clearly, rigorously and equitably spelt out in the management contract’ (Harding 1998: 5), the potential benefits of privatisation outweigh the drawbacks. For a useful discussion of the arguments for and against private prisons, see Roth (2004).

One of the measures that can be used to indicate whether the quality of care and range of services available in private prisons is comparable with that in the government sector is the rate of death. In a report by Biles and Dalton (2001), rates of death in private prisons were compared with those in government-run institutions over the decade from 1990 to 2000. It was found that

the rate of death from all causes in private prisons was 2.77 per 1,000 prisoners per year (CI 2.51–3.03) compared with 3.28 per 1,000 prisoners per year in public prisons (CI 3.19–3.37) (Biles & Dalton 2001: 298).

That is, the rate of death was generally lower in private prisons than in government-run institutions over that decade. Looking now at the period from 1998–99 to 2010–11, it can be seen that this trend has largely continued, with the rate of death in private prisons remaining generally lower than that in government prisons, with the exception of 2007–08 and 2008–09 (see Figure 17) where there was a dramatic spike in the rate of death and after which both the number and rate of death returned to levels consistent with historical trends.

Figure 17 Prison custody deaths by type of prison and year, 1998–99 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E10

Rate per 100 relevant prisoners (SCRCSP 2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002, 2001, 2000, 1999)

Source: AIC NDICP 1998–99 to 2010–11 [computer file]

Looking more closely at the causes of death between government and privately run prisons provides some insight into the factors underpinning the anomalous spike for privately run prisons in 2007–08 and 2008–09. Data presented in Figure 18 shows the rate of death for those prisoners dying from natural causes, while data in Figure 19 shows the rate of death for all other causes of death (ie unnatural deaths), disaggregated by prison type. It is evident that while the rate at which prisoners have died from unnatural causes has remained consistent between both government and privately run prisons, the observed spike in the overall rate of private prison deaths seen in 2007–08 and 2008–09 was driven by a short-term increase in natural cause deaths.

Figure 18 Natural cause deaths in prison by type of prison and year, 1998–99 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E10

Rate per 100 relevant prisoners (SCRCSP 2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002, 2001, 2000, 1999)

Source: AIC NDICP 1998–99 to 2010–11 [computer file]

Figure 19 Unnatural cause deaths in prison by type of prison and year, 1998–99 to 2010–11 (rate per 100 prisoners)a

a: Denominator: see Table E10

Rate per 100 relevant prisoners (SCRCSP 2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002, 2001, 2000, 1999)

Source: AIC NDICP 1998–99 to 2010–11 [computer file]

Box 1 Prisoner health in Australia

There have been a number of studies conducted in Australia over the last decade that examined the health of Australia’s prisoners (eg see AIHW 2010; Belcher et al. 2006; Deloitte 2003; Indig et al. 2010).

The most notable study is the Australian Institute of Health and Welfare’s National Prisoner Health Census, which was a snapshot of prisoner health over a two week period in October and November 2010 (AIHW 2011c). In total, data were collected from ‘610 prison entrants, almost 6,000 prisoners in custody who visited a clinic, and more than 5,500 prisoners who were taking prescribed medication’ (AIHW 2011c: viii). The main findings from that survey were:

  • 31 percent of prison entrants reported at some time being told they had a mental illness;
  • 21 percent of prison entrants reported a history of self harm;
  • 35 percent of prison entrants tested positive to hepatitis C antibody;
  • 26 percent of prison entrants reported having a current chronic condition (asthma 12%, arthritis 8%, cardiovascular disease 5%, diabetes 4% and cancer <1%); and
  • 66 percent of prison entrants reported illicit use of drugs in the 12 months prior to their current incarceration (AIHW 2011c).

Corrective services agencies in some jurisdictions also track the health of their prisoners. For example, Justice Health and Forensic Mental Health Network administered health surveys to NSW prisoners in 1996, 2001 and 2009. Using a stratified random sample of all inmates from 30 adult prisons, a total of 996 prisoners participated in the most recent survey (Indig et al. 2010).

The key findings from that survey were:

  • 56 percent of respondents were overweight or obese;
  • 20 percent had been told by a doctor that they had a heart problem;
  • 56 percent of men and 35 percent of women had a history of a head injury resulting in unconsciousness;
  • 28 percent of men and 45 percent of women tested positive to hepatitis C antibody;
  • 44 percent of respondents indicated daily/near daily use of drugs in the year before prison; and
  • 49 percent of respondents indicated they had been assessed or treated by a doctor or psychiatrist for a mental health problem (Indig et al. 2010).

Summary of deaths in prison custody

Since 1978–80, there have been a total of 1,393 deaths in prison custody across Australia, of which 238 (17%) were Indigenous prisoners and 1,155 were non-Indigenous (83%). Of these 1,393 deaths, males (96%; n=1,332) have consistently outnumbered females (4%; n=61). In terms of age, two in five deaths (40%; n=551) were among those prisoners aged between 25 and 39 years, followed by those aged 40 to 54 years (23%; n=314), those less than 25 years (19%; n=266) and 55 years and above (19%; n=262). Despite having more deaths overall, deaths of persons 25 to 39 years have been in decline over the last decade, while the frequency of deaths involving those between 40 to 54 years and 55 years and older has been increasing.

The most common cause of death is natural causes (42%; n=584), closely followed by those who died from hangings (38%; n=525). Deaths involving Indigenous persons have followed this overall trend, in that most are due to natural causes (53%; n=127) and hangings (35%; n=84). The most common type of natural cause deaths were heart problems (such as heart attacks and heart disease; 47%; n=263). The overall number of heart-related deaths has also gradually increased, most likely because of the increasing number of deaths of older sentenced prisoners in Australia.

Over the 32 years for which data are available, slightly more prisoners have died from self-inflicted injuries, such as hanging or incise wounds, than in any other situation (43%; n=593). However, due to recent rises in natural cause deaths, the total number of such deaths (42%; n=583) is now very similar to those that are self-inflicted. Unlike the overall trend, Indigenous deaths in prison custody are more likely to occur as a result of a natural cause (53%; n=127) rather than being self-inflicted (38%; n=90). With regards to hanging deaths, prisoners use ‘other cell fittings’ (39%; n=145), followed by cell bars (36%; n=133) as the preferred hanging point, with bed sheets (61%; n=232) being the most common material used.

Looking at most serious offences leading to custody, it was shown that over half (53%; n=722) of all prisoner deaths were persons with violent offending histories, followed by those prisoners who had committed theft-related offences (28%; n=378). Similarly, far more violent Indigenous offenders died in custody (58%; n=137) than all other types, with the next most common being offenders with theft-related offences (22%; n=53).

Data from the most recent reporting period (2008–09 to 2010–11) revealed that the rate of death was generally higher among those who were unsentenced (0.24–0.28 per 100) compared with those who were sentenced (0.12–0.19 per 100). Finally, with regards to deaths in private and government-run prisons, the rate of death has dropped for private prisons, after a considerable spike in 2007–08 and 2008–09. The rate is again lower in private prisons than those run by the government; this is influenced by the nature and demographics of the prisoners housed in private and government prisons.

Table 36 Prison custody deaths by type of prison and year, 1979–80 to 2010–11 (n)
Private Government Total
1979–80 0 15 15
1980–81 0 29 29
1981–82 0 27 27
1982–83 0 25 25
1983–84 0 33 33
1984–85 0 27 27
1985–86 0 18 18
1986–87 0 37 37
1987–88 0 44 44
1988–89 0 37 37
1989–90 0 53 53
1990–91 0 35 35
1991–92 0 31 31
1992–93 3 40 43
1993–94 5 52 57
1994–95 2 53 55
1995–96 5 44 49
1996–97 4 59 63
1997–98 10 71 81
1998–99 10 54 64
1999–2000 5 59 64
2000–01 10 48 58
2001–02 7 46 53
2002–03 5 39 44
2003–04 4 35 39
2004–05 6 33 39
2005–06 5 23 28
2006–07 4 36 40
2007–08 18 28 46
2008–09 11 32 43
2009–10 7 51 58
2010–11 12 46 58
Total 133 1,260 1,393

Source: AIC NDICP 1979–80 to 2010–11 [computer file]