Australian Institute of Criminology

Skip to content

Motor vehicle pursuit and shooting deaths

Deaths resulting from motor vehicle pursuits and shootings during police operations

Deaths occurring during motor vehicle pursuits or as a result of shootings during a police operation attract a great deal of media and public attention. They often take place in public areas, which means they can impact on other members of the community who are not directly involved (eg where a person being pursued by police in a vehicle may lose control of the car and collide with another car or bystanders). This section presents information on the characteristics and circumstances of deaths of alleged offenders that result from motor vehicle pursuits and police shootings.

Important consideration about motor vehicle pursuit deaths

Deaths that occur in situations where police officers are pursuing an alleged offender in a motor vehicle fall within the scope of the NDICP under point three of the definition handed down by the RCIADIC, in that they are

[a] death, wherever occurring, of a person who dies, or is fatally injured, in the process of police or prison officers attempting to detain that person (RCIADIC 1991: Rec 41 (c) iii).

The focus of the definition is on the person that police officers were attempting to detain. Deaths of persons who were innocent bystanders of a motor vehicle pursuit, such as passengers, pedestrians or other third parties, are outside the scope of the NDICP. These deaths are excluded from analysis and reporting because the deceased was never in custody and officers were not seeking to take them into custody. Moreover, had they survived the accident they would have been free to leave. Deaths of passengers in motor vehicle pursuits are only included if they were involved in the commission of an offence, such as a break and enter or motor vehicle theft, which led to the motor vehicle pursuit.

Another point of clarification is around the specific types of situations that are covered by the words ‘attempting to detain’. For the purposes of the NDICP, this has been interpreted to mean any situation where police officers are actively trying to place someone under arrest for breaking the law. This means situations where police officers have witnessed a person break the law, or have evidence to the effect that a crime has been committed and are in the process of trying to place that person under arrest.

There are occasionally deaths that occur in the course of police officers pursuing a motor vehicle where it is not immediately clear if the definition of a death in custody has been met. Some examples include:

Police officers are instructed by central command to terminate the pursuit because the risk of harm to public safety outweighs the benefits of capturing the alleged offender. Officers then slow down, pull to the side of the road and turn off the lights and siren. Nine minutes later they are advised over the radio that an ambulance has been called to attend a motor vehicle accident in their vicinity. It is then discovered that this is the same vehicle that was involved in the earlier pursuit.
Police officers receive a call late at night from a person reporting suspicious activity outside their house. Officers arrive and find a person sitting in their parked car. As they approach the vehicle to speak with the driver, the vehicle accelerates away. They return to their police car and begin heading in the same direction. All officers can see are the taillights of a vehicle several hundred metres ahead. As they continue on, officers come across the vehicle which has crashed into a tree and find the driver has sustained fatal injuries.

In making a determination about whether or not such cases fall within the scope of the definition, the NDICP research staff:

  • consult with the police service involved;
  • consult with the relevant state or deputy state coroner;
  • take into account the time that has elapsed between the pursuit and the accident; and
  • consider whether the deceased reasonably believed that police were actively pursuing them.

The decision to include or exclude cases is also guided by the intent of the Royal Commissioners in handing down the definition used by the NDICP, in that the spirit of the program is to make public the circumstances of any death where the actions of officers involved may have, in some way, contributed to the death.

In the first example above, such a death would be excluded because the pursuit was terminated, officers were no longer attempting to detain that person and sufficient time had elapsed between the termination of the pursuit and the accident such that the driver would reasonably believe that they had eluded police.

In the second example above, while officers were not seeking to detain that person for breaking the law (they were simply endeavouring to speak with the driver about their behaviour), such a case would be included because the deceased could reasonably believe that the police were in pursuit and there was a continuous chain of events that resulted in a fatal accident.

Limitations of the data sources

Police and coronial records (the 2 main data sources for this section) do not always contain complete data on each motor vehicle pursuit or shooting incident. In particular, there is a substantial amount of missing information for three variables—time of incident, duration of pursuit and speeds reached during the pursuit. The NDICP data collection forms sent to police departments do not currently ask for information on these variables, which means information on these three variables can only be obtained from coronial records. Coronial findings are relied on to confirm information from police reports and to complete missing information. However, not all information is reported on, or reported consistently in coronial records across cases. This means that some variables will have only limited information available.

The length of time taken for some coronial findings to become available also means NDICP cases are often updated retrospectively. For the most recent years, some of the variables collected on pursuits and shootings may not yet be available. For example, in Table 78, which presents information on the average speeds reached and average duration of the pursuit, there are 25 cases where the speed reached in the pursuit is missing or unavailable and 39 cases where the length of pursuit is missing. Extensive efforts were made during 2011 to locate these missing data for historical cases. However, if the information is not contained in the coronial finding, then it is almost impossible to obtain.

Motor vehicle pursuit deaths

There were 31 motor vehicle pursuit deaths over the reporting period, which is high relative to historical totals (see Table 74 and Figure 16). Over the last three years, more motor vehicle pursuit deaths occurred in Western Australia (36%; n=11) and Queensland (23%; n=7) than in any other jurisdiction. New South Wales and Victoria recorded a very low number of pursuit fatalities over this period (n=2 and n=2 respectively).

Table 74 Deaths in motor vehicle pursuitsa by jurisdiction, 2008–09 to 2010–11 (n)
2008–09 2009–10 2010–11
New South Wales 1 1 0
Victoria 1 1 0
Queensland 4 1 2
Western Australia 5 2 4
South Australia 3 0 0
Tasmania 0 1 1
Northern Territory 1 2 0
Australian Capital Territory 0 1 0
Total 15 9 7

a: Deaths of innocent passengers and third parties dying in motor vehicle pursuits are outside the scope of the NDICP

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

Trends

Since the collection of these data began in 1989–90, there have been a total of 206 deaths of alleged offenders during motor vehicle pursuits (see Table 74). Using fourth-order polynomial regression, the long-term trend can best be described as curvilinear, in that there was a steady increase over the decade 1989–90 to 1999–2000, a peak in 2001–02 and then a moderate decline over last decade (see Figure 29). It is also important to note that the number recorded in 2008–09 (n=15) is the second highest annual total on record. However, following this second peak, the numbers recorded in the last two years have returned to the lower end of historical totals. Consequently, it is difficult to conclude that pursuit deaths are increasing or whether the number recorded in 2008–09 was simply an aberration from what has otherwise been quite a stable trend in recent years. For this reason, the AIC is currently working with police agencies to conduct some targeted research into motor vehicle pursuit fatalities in Australia. The findings from this research will be released in the coming months.

Since 1989–90, the number of deaths in motor vehicle pursuits for each jurisdiction has been:

  • 73 deaths (35%) in New South Wales;
  • 42 deaths (20%) in Western Australia;
  • 34 deaths (17%) in Victoria;
  • 24 deaths (12%) in Queensland;
  • 19 deaths (9%) in South Australia;
  • 8 deaths (4%) in the Northern Territory;
  • 4 deaths (2%) in Tasmania; and
  • 2 deaths (1%) in the Australian Capital Territory.

Figure 29 Motor vehicle pursuit deaths, 1989–90 to 2010–11 (n)

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

Demographic characteristics

Indigenous status

Over the last three financial years, there were nine deaths of Indigenous persons in motor vehicle pursuits and 22 deaths involving non-Indigenous persons (see Table 75). Since 1989–90, Indigenous persons have comprised 21 percent (n=44) of all deaths. The number of Indigenous deaths from pursuits has remained consistently low, ranging between zero and five deaths each year. More pronounced fluctuations can be seen when looking at deaths of non-Indigenous persons from pursuits, which have ranged from zero to 17 deaths each year.

Table 75 Motor vehicle pursuit deaths by Indigenous status and year, 1989–90 to 2010–11 (n)
Indigenous Non-Indigenous Total
1989–90 3 2 5
1990–91 1 0 1
1991–92 0 2 2
1992–93 2 8 10
1993–94 0 5 5
1994–95 0 7 7
1995–96 5 4 9
1996–97 5 6 11
1997–98 0 11 11
1998–99 2 3 5
1999–2000 2 10 12
2000–01 2 10 12
2001–02 2 17 19
2002–03 1 14 15
2003–04 2 10 12
2004–05 3 8 11
2005–06 4 6 10
2006–07 0 9 9
2007–08 1 8 9
2008–09 2 13 13
2009–10 3 6 9
2010–11 4 3 7
Total 44 162 206

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

Sex

The overwhelming majority (94%) of persons dying in motor vehicle pursuits have been male (n=193; see Table 76). Only one female has died during a police motor vehicle pursuit in the last three financial years and there have been a total of 13 female deaths since 1989–90.

Table 76 Motor vehicle pursuit deaths by sex, 1989–90 to 2010–11 (n)
Male Female Total
1989–90 4 1 5
1990–91 1 0 1
1991–92 2 0 2
1992–93 9 1 10
1993–94 4 1 5
1994–95 7 0 7
1995–96 9 0 9
1996–97 11 0 11
1997–98 11 0 11
1998–99 3 2 5
1999–2000 12 0 12
2000–01 12 0 12
2001–02 18 1 19
2002–03 13 2 15
2003–04 12 0 12
2004–05 10 1 11
2005–06 9 1 10
2006–07 8 1 9
2007–08 8 1 9
2008–09 15 0 15
2009–10 8 1 9
2010–11 7 0 7
Total 196 13 206

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

Age

Of greatest concern when examining deaths in motor vehicle pursuits is the relatively young age of those dying (see Table 77). Over the last 22 years, persons aged 24 years or less represent almost two out of every three deaths (60%; n=124). Moreover, persons aged 15–19 years account for one in three (33%; n=68) pursuit deaths, while all persons aged 30 years and over represent just one in four (25%; n=51) deaths occurring in motor vehicle pursuits.

One possible interpretation of the relatively young age profile of persons dying in motor vehicle pursuits is that young inexperienced drivers might be at an increased risk of becoming a fatality during a pursuit. This is particularly important given that the legal age of obtaining a driver’s licence (specifically, a learner’spermit) is 16 years in all jurisdictions (except the Australian Capital Territory, where the legal driving age is 15 years and 9 months).

Of the 124 deaths of persons aged 24 years or less in motor vehicle pursuits, one-quarter (25%; n=31) were aged 16 years or less. The concern here is that one in four deaths of young people in motor vehicle pursuits are either of persons ineligible for a driving licence because they are too young, or they have only a few months experience operating a motor vehicle. Since 1989–90, the following numbers of deaths were recorded across this young age bracket:

  • 16 deaths of 19 year olds;
  • 22 deaths of 18 year olds;
  • 13 deaths of 17 year olds;
  • 12 deaths of 16 year olds;
  • 4 deaths of 15 year olds;
  • 9 deaths of 14 year olds; and
  • 6 deaths of children aged 13 years or less.
Table 77 Motor vehicle pursuit deaths by age in years at time of death, 1989–90 to 2010–11 (n)
Less than 15 15–19 20–24 25–29 30+ Total
1989–90 1 3 0 0 1 5
1990–91 0 1 0 0 0 1
1991–92 0 0 1 0 1 2
1992–93 2 4 2 1 1 10
1993–94 0 0 2 3 0 5
1994–95 0 3 4 0 0 7
1995–96 1 5 2 1 0 9
1996–97 2 6 0 1 2 11
1997–98 0 2 2 2 5 11
1998–99 1 3 0 0 1 5
1999–2000 1 3 2 3 3 12
2000–01 1 5 4 0 2 12
2001–02 2 9 3 2 3 19
2002–03 0 4 3 1 7 15
2003–04 1 2 1 3 5 12
2004–05 0 5 2 2 2 11
2005–06 0 2 2 4 2 10
2006–07 0 1 2 1 5 9
2007–08 1 4 0 1 3 9
2008–09 0 2 5 4 4 15
2009–10 1 2 3 1 2 9
2010–11 0 2 2 1 2 7
Total 14 68 42 31 51 206

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

Circumstances surrounding motor vehicle pursuit deaths

Speed and duration of pursuit

Of the 206 motor vehicle pursuit deaths in Australia between 1989–90 and 2010–11, information about the speeds reached was available in 178 cases and duration of the pursuit was available for 163 cases. Since 1989–90, the top speeds reached during pursuits have been high across all the jurisdictions. The maximum speeds reached during motor vehicle pursuits ranged from 154km/h in the Northern Territory to 220km/h in Western Australia and Victoria, with pursuit times ranging from a matter of seconds to more than an hour (see Table 78). Where the duration of pursuits was known (n=163), the average duration of fatal pursuits was five minutes and 33 seconds.

Unfortunately, at the time of writing, information was available on both the speeds reached and duration of the pursuit for only 24 of the 31 pursuit deaths that occurred in the last three financial years (2008–09 to 2010–11). Over this period, the average speed was 131.8km/h and the average duration of the pursuit was three minutes and three seconds.

It should be noted that in most pursuit situations, once speeds reach in excess of 140km/h, it is recommended that the pursuit be scaled back or terminated. Police officers are acutely aware of the risks associated with high-speed pursuits and as part of the pursuit response undertake regular risk assessments in conjunction with the local area commanders. Advice received from police agencies indicated that in almost every situation where the pursuit reaches speeds in excess of 140km/h, the decision is made to terminate the pursuit and this is in line with the Australian Federal Police National Guidelines for terminating pursuits (AFP 2007). Obviously, as a vehicle increases in speed, the reaction time afforded to the driver in responding to changes in driving conditions is reduced. Therefore, the higher the speed reached in a pursuit the greater the likelihood that an accident will occur.

Figure 30 shows that while the average top speeds reached in pursuits were quite high, they declined over the decade 1989–90 to 1999–2000. Since then, the average pursuit speeds have remained fairly stable. Information about the speeds reached in the seven fatal motor vehicle pursuits in 2010–11 is currently only available in five cases. The missing data will be captured once the coronial findings become available.

Table 78 Speed and duration of motor vehicle pursuits, 1989–90 to 2010–11a
  Average speed (km/h) Top speed (km/h) Average pursuit time (mins: secs) Pursuit time (mins: secs) Cases (n)
Min Max
New South Wales 131.7 210 5:39 0:05 50:00 64
Victoria 141.3 220 6:43 0:04 81:00 31
Queensland 127.3 180 5:29 0:09 22:00 19
Western Australia 152.7 220 4:47 0:24 15:00 36
South Australia 123.3 190 5:39 0:15 32:00 16
Tasmania 95.0 160 2:45 2:00 4:00 4
Northern Territory 117.2 154 4:42 1:30 7:00 6
Australian Capital Territory 158.5 167 4:30 1:00 8:00 2
Australia 135.4 220 5:33 0:04 81:00 178

a: Averages were calculated using the total number of cases for which top speed and length of pursuit data were available

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

Figure 30 Average top speed reached in motor vehicle pursuit deaths, 1989–90 to 2010–11 (kilometres per hour)

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

Time of incident

More motor vehicle pursuit deaths occur in the early hours of the morning between midnight and 4 am (36%; n=71) than at any other time during the day (see Table 79). The second most common time for a pursuit death to occur is between the hours of 8 pm and midnight (20%; n=40).

Table 79 Motor vehicle pursuit deaths by time of incident, 1989–90 to 2010–11 (n)
12.00–4.00 am 4.01–8.00 am 8.01 am–12.00 pm 12.01–4.00 pm 4.01–8.00 pm 8.01–11.59 pm Total
1989–90 4 0 0 0 1 0 5
1990–91 1 0 0 0 0 0 1
1991–92 1 1 0 0 0 0 2
1992–93 1 3 0 2 0 4 10
1993–94 1 2 1 0 0 1 5
1994–95 1 1 0 2 0 2 6
1995–96 1 3 1 1 0 3 9
1996–97 3 3 0 2 1 2 11
1997–98 3 1 1 0 1 3 9
1998–99 3 0 1 0 0 1 5
1999–2000 6 1 0 0 3 2 12
2000–01 3 2 0 2 2 3 12
2001–02 8 0 1 4 1 5 19
2002–03 4 3 1 3 3 1 15
2003–04 4 2 2 0 2 0 10
2004–05 6 1 0 0 0 4 11
2005–06 2 1 3 1 1 1 9
2006–07 4 0 2 0 1 1 8
2007–08 5 1 0 1 2 0 9
2008–09 6 2 0 0 2 5 15
2009–10 3 0 0 2 1 1 7
2010–11 1 0 1 2 1 21 6
Totala 71 27 14 22 22 40 196

a: 10 cases have been excluded due to missing time of incident data

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

Cause of death

Deaths in motor vehicle pursuits almost always occur as a result of external/multiple injuries (74%; n=152) or head injuries (20%; n=41) sustained by the deceased when their vehicle crashes into another vehicle, or fixture on the side of the road, such as a tree, telegraph pole or sign (see Table 80). Occasionally, a motor vehicle pursuit will end in a standoff between police officers and the alleged offender, with death resulting from a gunshot—either self-inflicted or by police (4%; n=9). Over the 22 years for which data are available, there have been four deaths (2%) where the cause of death was from something else, such as a drug overdose or drowning.

Table 80 Motor vehicle pursuit deaths by cause of death, 1989–90 to 2010–11 (n)
External/multiple trauma Head injury Gunshot Other/drugs Total
1989–90 4 1 0 0 5
1990–91 1 0 0 0 1
1991–92 1 1 0 0 2
1992–93 7 3 0 0 10
1993–94 3 0 1 1 5
1994–95 4 2 1 0 7
1995–96 9 0 0 0 9
1996–97 6 3 2 0 11
1997–98 9 0 2 0 11
1998–99 5 0 0 0 5
1999–2000 9 2 1 0 12
2000–01 6 4 1 1 12
2001–02 12 7 0 0 19
2002–03 11 3 1 0 15
2003–04 11 0 0 1 12
2004–05 8 3 0 0 11
2005–06 3 6 0 1 10
2006–07 9 0 0 0 9
2007–08 8 1 0 0 9
2008–09 13 2 0 0 15
2009–10 6 3 0 0 9
2010–11 7 0 0 0 7
Total 152 41 9 4 206

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

Manner of death

Due to the fact that most deaths in motor vehicle pursuits occur as a result of a vehicle accident, the manner of death in most cases is accidental (87%; n=179). Since 2006–07, every death in a motor vehicle pursuit (n=48) except one has resulted from an accident (see Table 81). All other circumstances of death in motor vehicle pursuits occur very infrequently.

Table 81 Motor vehicle pursuit deaths by manner of death, 1989–90 to 2010–11 (n)
Accident Self-inflicted Unlawful homicide Justifiable homicide Other Total
1989–90 4 0 1 0 0 5
1990–91 1 0 0 0 0 1
1991–92 1 1 0 0 0 2
1992–93 10 0 0 0 0 10
1993–94 4 1 0 0 0 5
1994–95 6 1 0 0 0 7
1995–96 9 0 0 0 0 9
1996–97 8 2 1 0 0 11
1997–98 8 1 1 1 0 11
1998–99 5 0 0 0 0 5
1999–2000 11 0 0 1 0 12
2000–01 11 0 1 0 0 12
2001–02 16 2 1 0 0 19
2002–03 11 4 0 0 0 15
2003–04 11 1 0 0 0 12
2004–05 8 0 3 0 0 11
2005–06 7 0 2 0 1 10
2006–07 9 0 0 0 0 9
2007–08 8 1 0 0 0 9
2008–09 15 0 0 0 0 15
2009–10 9 0 0 0 0 9
2010–11 7 0 0 0 0 7
Total 179 14 10 2 1 206

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

Most serious offence

The data presented in this section captures the most serious offence committed by those alleged offenders who died in a motor vehicle pursuit. In the event that the deceased committed multiple offences, all offenders are recorded in the NDICP database but only the most serious is analysed and reported. Offence seriousness is ranked in the order in which they appear in Table 82, from left to right; categories that are derived from the National Offence Index (ABS 2009b).

For those cases where information is available about the most serious offence since 1989–90, it is interesting to note that 44 percent (n=88) involved traffic offences, such as driving under the influence or dangerous operation of a motor vehicle, followed by theft-related offences (usually the theft of the motor vehicle being pursued; 44%; n=87). Motor vehicle pursuits precipitated by violent, drug-related, good order and other offences occur very infrequently, with the combined total of these categories representing just 12 percent (n=24) of all offences that result in a motor vehicle pursuit death.

Table 82 Motor vehicle pursuit deaths by most serious offence, 1989–90 to 2010–11 (n)
Violent Theft-related Drug-related Traffic Good order Other Total
1989–90 1 3 0 1 0 0 5
1990–91 0 1 0 0 0 0 1
1991–92 0 0 0 2 0 0 2
1992–93 0 6 1 3 0 0 10
1993–94 1 1 0 3 0 0 5
1994–95 1 2 2 2 0 0 7
1995–96 0 5 0 4 0 0 9
1996–97 2 8 0 1 0 0 11
1997–98 2 4 0 3 0 0 9
1998–99 2 2 0 1 0 0 5
1999–2000 0 8 1 2 1 0 12
2000–01 0 7 0 5 0 0 12
2001–02 0 12 0 7 0 0 19
2002–03 2 6 0 7 0 0 15
2003–04 0 5 0 6 0 0 11
2004–05 0 6 0 3 1 0 10
2005–06 0 3 3 3 1 0 10
2006–07 2 0 0 6 0 0 8
2007–08 0 2 0 6 0 0 8
2008–09 0 2 0 12 0 1 15
2009–10 0 1 0 7 0 0 8
2010–11 0 3 0 4 0 0 7
Totala 13 87 7 88 3 1 199

a: 7 cases have been excluded due to missing data about most serious offence

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

Box 2: Motor vehicle thefts in Australia

Established in 1999 as a national taskforce to bring about reductions in vehicle thefts, the National Motor Vehicle Theft Reduction Council (NMVTRC) also collects statistics on all motor vehicle thefts in Australia, disaggregating data between those that are profit-motivated and those that are for short-term purposes.

In the most recent NMVTRC factsheet published in October 2011, it can be seen that since July 2006, there has been a steady year-on-year decline in the number of passenger and light commercial vehicle thefts for short-term purposes, while profit-motivated thefts over the same period have shown greater fluctuation (NMVTRC 2011).

Moreover, in a paper produced by Weatherburn et al. (2009) from the NSW Bureau of Crime Statistics and Research, the point is made that a relatively small group of offenders commit a disproportionately large number of motor vehicle thefts and that these ‘high rate’ offenders tend to have a history of prior offending. In particular, the paper finds that

the vast majority of high rate offenders (…63 percent for MVT) were convicted at least once during the two-year observation period…and nearly 27 percent…were convicted two or more times (Weatherburn et al. 2009: 5).

Of all the pursuit deaths where the offence was known (n=201), almost two in five (38%; n=76) involved the theft of a motor vehicle. It can therefore be argued that by continuing to reduce the number of motor vehicle thefts each year, the NMVTRC is also reducing the number of opportunities for offenders to die in motor vehicle theft-related pursuits.

Location of death

As would be expected due to the nature of motor vehicle pursuits, the majority occur in public places (79%; n=163), followed by public hospitals after an accident (19%; n=40; see Table 83).

Table 83 Motor vehicle pursuit deaths by location, 1989–90 to 2010–11 (n)
Public hospital Private property Public place Other Total
1989–90 0 0 5 0 5
1990–91 0 0 1 0 1
1991–92 0 0 2 0 2
1992–93 3 0 7 0 10
1993–94 1 0 3 1 5
1994–95 0 0 7 0 7
1995–96 5 1 3 0 9
1996–97 4 0 7 0 11
1997–98 4 0 7 0 11
1998–99 2 0 3 0 5
1999–2000 2 1 9 0 12
2000–01 2 0 10 0 12
2001–02 2 0 17 0 19
2002–03 2 0 13 0 15
2003–04 0 0 12 0 12
2004–05 0 0 11 0 11
2005–06 2 0 8 0 10
2006–07 1 0 8 0 9
2007–08 4 0 5 0 9
2008–09 3 0 12 0 15
2009–10 3 0 6 0 9
2010/11 0 0 7 0 7
Total 40 2 163 1 206

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

Shooting deaths

Deaths resulting from shootings can be classified as either a Category 1 or 2 death, depending on the circumstances surrounding the shooting. Shooting deaths during police operations include situations where a person dies due to being shot by police, government official or security guard, as well as those situations where a person dies due to self-inflicted gunshot wounds in the presence of police. The decision to utilise a firearm in a police operation is not one that is made lightly and as can be seen from the available data, is a decision that is made very infrequently.

Reporting period 2008–09 to 2010–11

Over the reporting period, there were 14 persons shot by police and 13 persons who shot themselves in the presence of police (see Table 84). Deaths across the three years were distributed as follows:

  • In 2008–09, there were 10 deaths resulting from a gunshot, five involving persons shot by police and five where the deceased shot themselves in the presence of police (see Table 84). Available data showed that no Indigenous persons were shot by police; one Indigenous gunshot death was self-inflicted.
  • In 2009–10, the number of shooting deaths dropped slightly from the previous 12 months (n=10 cf n=8). There were three persons shot by police in 2009–10 and five deaths resulting from a self-inflicted gunshot. As with the previous year, no Indigenous persons were shot by police; one Indigenous gunshot death was self-inflicted.
  • In 2010–11, there was a small increase in the number of persons shot by police, while at the same time, there was a decline in self-inflicted gunshot deaths, resulting in a total number of shooting deaths at levels in line with recorded totals in recent years. One Indigenous person was shot by police in 2010–11, with all other shooting deaths involving non-Indigenous persons.
Table 84 Police shooting deaths, 2008–09 to 2010–11 (n)
2008–09
Shot by police Shot self Total
New South Wales 2 1 3
Victoria 1 1 2
Queensland 0 1 1
Western Australia 0 1 1
South Australia 2 0 2
Tasmania 0 0 0
Northern Territory 0 1 1
Australian Capital Territory 0 0 0
Totals
Indigenous 0 1 1
Non-Indigenous 5 4 9
All persons 5 5 10
2009–10
Shot by police Shot self Total
New South Wales 3 1 4
Victoria 0 1 1
Queensland 0 2 2
Western Australia 0 0 0
South Australia 0 1 1
Tasmania 0 0 0
Northern Territory 0 0 0
Australian Capital Territory 0 0 0
Totals
Indigenous 0 1 1
Non-Indigenous 3 4 7
All persons 3 5 8
2010–11
Shot by police Shot self Total
New South Wales 2 1 3
Victoria 1 0 1
Queensland 0 0 0
Western Australia 0 1 1
South Australia 1 1 2
Tasmania 1 0 1
Northern Territory 0 0 0
Australian Capital Territory 1 0 1
Totals
Indigenous 1 0 1
Non-Indigenous 5 3 8
All persons 6 3 9

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

Trends

Over the last 22 years, a total of 199 shooting deaths have been recorded in police custody-related operations (see Table 85). The vast majority of all shooting deaths (94%; n=186) have involved non-Indigenous persons, with just over half (52%; n=97) of these being persons being shot by police.

More than two in five shooting deaths (46%; n=92) were persons shooting themselves in the presence of police. The relatively large proportion of self-inflicted shooting deaths in police custody-related operations suggests that mental health issues are playing a greater role in the prevalence of these deaths. Analysis of the data on persons shooting themselves showed that 44 percent (n=40) were persons suffering from some form of mental illness at the time of death. Mental health data for persons shot by police is presented below.

Indigenous persons represent only a very small proportion of persons involved in shootings (7%; n=13; see Table 85). Of all persons shot by police since 1989–90 (n=105), Indigenous persons represent just one in 13 such deaths (8%; n=8; see Figure 31).

Long-term trends show that the number of persons shot by police peaked twice, once in 1993–94 (9 deaths) and again in 1999–2000 (11 deaths). In all other years, the number of persons shot by police has fluctuated between two and seven deaths each year. It can also be seen in Figure 31 that the number of police shootings of Indigenous persons are low and occur infrequently.

Table 85 All shooting deaths in police custody and custody-related operations by Indigenous status and year, 1989–90 to 2010–11 (n)
Shot by police Shot self Shot by other official Total
Indigenous Non–Indig Indigenous Non–Indig Indigenous Non–Indig Indigenous Non–Indig Total
1989–90 1 1 0 0 0 0 1 1 2
1990–91 0 5 0 3 0 0 0 8 8
1991–92 0 4 0 2 0 0 0 6 6
1992–93 0 4 0 5 0 1 0 10 10
1993–94 1 8 0 7 0 0 1 15 16
1994–95 2 4 0 5 0 0 2 9 11
1995–96 0 4 0 7 0 0 0 11 11
1996–97 0 7 1 7 0 0 1 14 15
1997–98 0 5 0 2 0 0 0 7 7
1998–99 1 1 0 0 0 0 1 1 2
1999–2000 0 11 0 4 0 0 0 15 15
2000–01 1 2 0 4 0 0 1 6 7
2001–02 0 2 1 3 0 0 1 5 6
2002–03 1 4 0 5 0 0 1 9 10
2003–04 0 7 0 5 0 0 0 12 12
2004–05 0 6 0 5 0 0 0 11 11
2005–06 0 3 0 3 0 0 0 6 6
2006–07 0 3 0 4 0 0 0 7 7
2007–08 0 3 1 5 0 0 1 8 9
2008–09 0 5 1 4 0 0 1 9 10
2009–10 0 3 1 4 0 0 1 7 8
2010–11 1 5 0 3 0 0 1 8 9
Totala 8 97 5 87 0 1 13 185 198

a: 1 case has been excluded due to missing data

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

Figure 31 Persons shot by police by Indigenous status, 1989–90 to 2010–11 (n)

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

Demographic characteristics

Indigenous status

Table 86 presents information on persons shot by police as a proportion of all deaths in police custody each year. Over the last 22 years for which these data have been collected, 15 percent of all deaths in police custody have been persons shot by police. It is important to note that almost three times the proportion of non-Indigenous police custody deaths have been persons shot by police compared with Indigenous persons (17% cf 6%).

Table 86 Persons shot by police as a percentage of all deaths in police custody, 1989–90 to 2010–11
Indigenous persons shot by police Non-Indigenous persons shot by police Total persons shot by police All deaths in police custody
n % n % Total n % Total n
1989–90 1 11 1 5 2 7 29
1990–91 0 5 22 5 19 26
1991–92 0 4 20 4 16 25
1992–93 0 4 12 4 11 38
1993–94 1 33 8 29 9 29 31
1994–95 2 67 4 15 6 20 30
1995–96 0 4 16 4 13 31
1996–97 0 7 27 7 21 34
1997–98 0 5 22 5 18 28
1998–99 1 14 1 7 2 10 21
1999–2000 0 11 34 11 31 36
2000–01 1 14 2 8 3 9 33
2001–02 0 2 6 2 5 42
2002–03 1 9 4 13 5 12 41
2003–04 0 7 22 7 17 42
2004–05 0 6 24 6 17 36
2005–06 0 3 19 3 13 23
2006–07 0 3 11 3 10 31
2007–08 0 3 10 3 9 34
2008–09 0 5 17 5 14 37
2009–10 0 3 14 3 11 28
2010–11 1 13 5 28 6 23 26
Total 8 6 97 17 105 15 702

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

Sex

As can be seen in Table 87, females are rarely shot by police, with just three percent (n=3) of all persons shot by police being female. Further, the last female to be shot by police was in 1995–96. In all other years, only males have died as a result of a police shooting.

Table 87 Persons shot by police by sex, 1989–90 to 2010–11 (n)
Male Female Total
1989–90 2 0 2
1990–91 5 0 5
1991–92 4 0 4
1992–93 4 0 4
1993–94 8 1 9
1994–95 5 1 6
1995–96 3 1 4
1996–97 7 0 7
1997–98 5 0 5
1998–99 2 0 2
1999–2000 11 0 11
2000–01 3 0 3
2001–02 2 0 2
2002–03 5 0 5
2003–04 7 0 7
2004–05 6 0 6
2005–06 3 0 3
2006–07 3 0 3
2007–08 3 0 3
2008–09 5 0 5
2009–10 3 0 3
2010–11 6 0 6
Total 102 3 105

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

Age

Analysis of the data showed that of all persons shot by police since 1989–90 (n=105), almost two out of every three (64%; n=67) were persons aged less than 35 years at the time of death (see Table 88). Persons over 35 years comprise just 36 percent (n=38) of persons shot by police. Despite one recorded case in each of the last three financial years, persons aged 19 years or less are rarely shot by police and such cases remain infrequent.

Table 88 Persons shot by police by age category, 1989–90 to 2010–11 (n)
19 yrs or less 20–24 yrs 25–29 yrs 30–34 yrs 35–39 yrs 40–44 yrs 45+ yrs Total
1989–90 0 1 0 0 0 1 0 2
1990–91 0 2 2 1 0 0 0 5
1991–92 0 2 1 0 0 1 0 4
1992–93 0 0 0 0 0 1 3 4
1993–94 2 2 1 1 2 1 0 9
1994–95 0 0 0 2 0 3 1 6
1995–96 0 1 1 0 2 0 0 4
1996–97 0 1 3 1 2 0 0 7
1997–98 0 1 1 3 0 0 0 5
1998–99 1 0 0 0 1 0 0 2
1999–2000 0 0 8 0 1 2 0 11
2000–01 0 1 0 0 1 0 1 3
2001–02 0 1 0 0 0 1 0 2
2002–03 1 1 0 2 0 0 1 5
2003–04 0 0 2 3 0 1 1 7
2004–05 0 0 2 2 1 1 0 6
2005–06 0 1 1 0 1 0 0 3
2006–07 0 0 0 3 0 0 0 3
2007–08 0 0 1 0 0 0 2 3
2008–09 1 1 0 0 0 2 1 5
2009–10 1 0 0 0 1 0 1 3
2010–11 1 1 1 2 0 1 0 6
Total 7 16 24 20 12 15 11 105

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

Circumstances surrounding shooting deaths

Time of incident

Since 1989–90, most police shooting deaths occurred during the late afternoon and early morning (see Table 89). Where information was available, 29 percent (n=30) occurred between 4 pm and 8 pm, 18 percent (n=19) between 8 pm and midnight and 14 percent (n=15) between midnight and 4 am. The most unlikely time for a person to be shot by police was between the hours of 4 am and 8 am, with only eight percent (n=7) of shootings occurring during these hours.

Table 89 Police shooting deaths by time of incident, 1989–90 to 2010–11 (n)
12.00–4.00 am 4.01–8.00 am 8.01 am–12.00 pm 12.01–4.00 pm 4.01–8.00 pm 8.01–11.59 pm Total
1989–90 0 0 1 0 1 0 2
1990–91 3 0 1 0 1 0 5
1991–92 2 1 1 0 0 0 4
1992–93 0 0 0 2 0 2 4
1993–94 0 0 1 3 3 2 9
1994–95 1 0 2 1 1 1 6
1995–96 1 0 1 0 2 0 4
1996–97 3 1 2 0 0 1 7
1997–98 1 1 0 0 2 1 5
1998–99 0 1 0 0 1 0 2
1999–2000 1 1 3 1 4 1 11
2000–01 0 0 0 0 1 2 3
2001–02 0 0 0 0 2 0 2
2002–03 0 0 0 1 2 2 5
2003–04 1 1 1 3 1 0 7
2004–05 0 1 0 1 4 0 6
2005–06 0 0 2 0 0 1 3
2006–07 0 0 2 0 1 0 3
2007–08 0 0 0 0 1 2 3
2008–09 0 0 1 1 2 1 5
2009–10 1 0 1 0 1 0 3
2010–11 1 1 0 0 0 3 5
Totala 15 8 19 13 30 19 104

a: 1 case has been excluded due to missing information about the time of shooting

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

Manner of death

This section presents information about the circumstances in which people were shot by police. The majority of police shooting deaths (93%; n=98) since 1989–90 were determined to be justifiable homicides (see Table 90). A justifiable homicide is that which occurs under circumstances authorised by law and in such cases, generally no charges against the officers involved are recommended by the Director of Public Prosecutions. Currently, the manner of death in one police shooting is being investigated by the Director of Public Prosecutions in New South Wales. Consequently, this case has been excluded pending the outcome of that investigation.

Table 90 Persons shot by police by manner of death, 1989–90 to 2010–11 (n)
Unlawful homicide Justifiable homicide Other Total
1989–90 0 1 1 2
1990–91 0 3 2 5
1991–92 0 4 0 4
1992–93 0 4 0 4
1993–94 0 9 0 9
1994–95 0 6 0 6
1995–96 0 4 0 4
1996–97 0 7 0 7
1997–98 0 5 0 5
1998–99 0 2 0 2
1999–2000 0 11 0 11
2000–01 0 2 0 3
2001–02 1 2 0 2
2002–03 0 4 1 5
2003–04 0 6 1 7
2004–05 0 6 0 6
2005–06 0 3 0 3
2006–07 0 3 0 3
2007–08 0 3 0 3
2008–09 0 4 0 4
2009–10 0 3 0 3
2010–11 0 6 0 6
Totala 1 98 5 104

a: 1 case has been excluded due to the fact that the manner of death is still under investigation

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

Mental illness

Of the 105 persons shot by police, information is available in 57 cases about whether the deceased had a mental illness at the time of the shooting. For those persons shot by police where the available information did not indicate the presence of a mental health issue, the lack of information should not be taken as confirmation that the person had no mental health issues. It may be the case that information about the deceased’s mental health was not included in the coronial finding because it had little or no impact on the circumstances of the death. It is also possible that during this period, other people were shot by police that had mental health issues but that these issues were not diagnosed and/or were not recorded in the available information. For this reason, findings presented here should be treated as a conservative estimate of the prevalence of mental illness among persons shot by police. The assignment of mental illness, disorders or conditions was made through judgement by NDICP research staff based on often limited available information.

Over the last 22 years, 44 persons suffering a mental illness have been shot by police, representing just over two in every five (42%) persons shot by police over this period. Psychotic disorders such as schizophrenia were by far the most prevalent (59%; n=26). According to the US National Institute of Mental Health, persons suffering from schizophrenia

may believe other people are reading their minds, controlling their thoughts, or plotting to harm them…people with the illness [can become] withdrawn or extremely agitated (NIMH 2012: np).

Responding to situations where a person is suffering the acute symptoms of schizophrenia, such as behaving erratically or debilitated by paranoid delusions, can be a very difficult and dangerous task for police officers.

Available research indicates that persons with schizophrenia may be overrepresented among those persons who have a mental illness and come in contact with police. Although limited Australian research exists on this topic, other international studies point to the over-representation of people suffering schizophrenia in the criminal justice system. In the United States, for example, a study of 172 individuals with schizophrenia or a schizoaffective disorder found that over three years, 48 percent (n=83) had contact with police; 22 percent of whom (n=37) had had charges filed against them (Brekke et al. 2001).

One possible explanation for the over-representation of people suffering from schizophrenia is their heightened level of involvement in violent or potentially violent activities warranting police intervention. In a Finnish study conducted by Eronen, Tiihonen and Hakola (1996: 1), it was found that

the risk of committing a homicide was about 10 times greater for schizophrenia patients and that the risk was further increased for males where schizophrenia coexisted with alcoholism.

From data collected on police shootings in Australia over the last 22 years, it can be seen that more than half of all persons with a mental illness who were shot by police had schizophrenia (59%; n=26; see Table 91) and of these, more than four in five (81%; n=21) were committing violent offences at the time when they were shot. In responding to such incidents, police officers may not know the individual has schizophrenia; all they are presented with is a person committing violent offences and behaving irrationally. Over the last decade, police officers in every jurisdiction have received special training to identify the symptoms of mental illness, as well as practical training in responding effectively to such incidents. It should also be noted that the overwhelming majority of interactions between police and persons suffering a mental illness end constructively.

Table 91 Persons shot by police by type of mental illness and year as a proportion of all persons shot by police, 1989–90 to 2010–11
Adjustment disorder Anxiety disorder Mood disorder Personality disorder Psychotic disorder All other disorders Total % of all persons shot by police
1989–90 0 0 0 0 0 0 0 0.0
1990–91 0 0 0 0 1 0 1 20.0
1991–92 0 1 0 1 0 0 2 50.0
1992–93 0 0 0 0 0 0 0 0.0
1993–94 0 0 0 0 3 2 5 55.6
1994–95 1 0 2 1 0 0 4 66.7
1995–96 0 0 0 0 2 0 2 50.0
1996–97 0 0 0 0 3 0 3 42.9
1997–98 0 0 0 0 1 0 1 20.0
1998–99 0 0 0 0 0 0 0 0.0
1999–2000 0 0 1 0 1 1 3 27.3
2000–01 0 0 0 0 0 1 1 33.3
2001–02 0 0 1 0 1 0 2 100.0
2002–03 0 0 1 0 0 1 2 40.0
2003–04 0 0 0 1 4 0 5 71.4
2004–05 0 0 0 0 3 0 3 50.0
2005–06 0 0 0 0 1 0 1 33.3
2006–07 0 0 0 0 2 0 2 66.7
2007–08 0 0 1 0 0 1 2 66.7
2008–09 0 0 0 0 2 0 2 40.0
2009–10 0 0 0 0 1 1 2 66.7
2010–11 0 0 0 0 1 0 1 16.7
Totala 1 1 6 3 26 7 44 41.9

a: 48 cases have been omitted due to missing information about whether the deceased had a mental illness and 13 cases have been excluded because the deceased was not recorded in the available information as having a mental illness

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

Most serious offence

Since 1989–90, more people have been shot by police following or during the commission of a violence offence than any other type of offending (69%; n=72; see Table 92). The majority of these violent offences involved homicides, serious assaults or persons with deadly weapons. Of the 14 persons shot by police in the last three financial years, nine had committed or were committing violence offences at the time when they were shot, with the remaining five cases involving a theft-related, good order or other offences, with one case missing information about the offence committed.

Table 92 Persons shot by police by most serious offence, 1989–90 to 2010–11 (n)
Violent Theft-related Drug-related Traffic Good order Other Total
1989–90 1 1 0 0 0 0 2
1990–91 4 1 0 0 0 0 5
1991–92 1 1 0 0 1 1 4
1992–93 2 2 0 0 0 0 4
1993–94 6 3 0 0 0 0 9
1994–95 5 0 0 1 0 0 6
1995–96 2 1 0 0 0 1 4
1996–97 4 2 1 0 0 0 7
1997–98 4 0 0 0 0 1 5
1998–99 1 1 0 0 0 0 2
1999–2000 9 2 0 0 0 0 11
2000–01 1 1 0 0 1 0 3
2001–02 2 0 0 0 0 0 2
2002–03 1 1 0 0 1 2 5
2003–04 7 0 0 0 0 0 7
2004–05 6 0 0 0 0 0 6
2005–06 3 0 0 0 0 0 3
2006–07 2 1 0 0 0 0 3
2007–08 2 0 0 0 1 0 3
2008–09 3 1 0 0 0 1 5
2009–10 2 0 0 0 0 1 3
2010–11 4 0 0 0 1 0 5
Totala 72 18 1 1 5 7 104

a: 1 case has been omitted due to missing information about the most serious offence

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

Location of death

Most police shootings occur in public places (45%; n=47) or on private property (35%; n=37; see Table 93) than any other location. Occasionally, persons are shot and then transferred to hospital where they later succumb to the gunshot wound(s) (16%; n=17). The four deaths in the ‘other’ category were all deaths occurring in an ambulance en route to the hospital.

Table 93 Police shooting deaths by location, 1989–90 to 2010–11 (n)
Public hospital Private property Public place Other Total
1989–90 0 1 1 0 2
1990–91 1 3 1 0 5
1991–92 1 2 1 0 4
1992–93 1 0 3 0 4
1993–94 1 3 5 0 9
1994–95 2 2 2 0 6
1995–96 1 2 1 0 4
1996–97 3 0 2 2 7
1997–98 0 2 2 1 5
1998–99 1 1 0 0 2
1999–2000 1 4 6 0 11
2000–01 0 2 1 0 3
2001–02 0 1 1 0 2
2002–03 1 1 2 1 5
2003–04 0 2 5 0 7
2004–05 2 1 3 0 6
2005–06 0 2 1 0 3
2006–07 0 2 1 0 3
2007–08 0 1 2 0 3
2008–09 0 3 2 0 5
2009–10 2 0 1 0 3
2010–11 0 2 4 0 6
Total 17 37 47 4 105

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

Summary of motor vehicle pursuit and shooting deaths

Motor vehicle pursuits

Since the NDICP began collecting data on motor vehicle pursuit deaths in 1989–90, there have been a total of 206 deaths, of which 44 (21%) involved Indigenous persons and 162 (79%) were of non-Indigenous persons. Of these 206 deaths, 193 deaths (94%) were of male, while the remaining 13 (6%) were of females. Closer examination of the data also revealed that a third of deaths involved young persons aged between 15 and 19 years (33%; n=68), and one in five involved a person aged 20–24 years (21%; n=42).

Another interesting finding was that of the 178 cases where the speeds involved in the pursuit were known, average motor vehicle pursuit speeds have remained fairly stable, with the average over the whole 22 years being 135.4 km/h. Moreover, where the duration of pursuits was known (n=163), the average duration of fatal pursuits was five minutes and 33 seconds. As such, a large proportion of pursuit accidents happen within a few minutes after the pursuit commences. Analysis of the time of day when pursuit fatalities occur showed that they most frequently happen between midnight and 4:00 am (36%; n=71).

Most people die from external/multiple trauma (74%; n=152) or head injuries (20%; n=41) sustained during a vehicle accident. In this way, the manner of death is almost always accidental (87%; n=179). Overall, the most common types of offences that led to a motor vehicle pursuit death were traffic-related (44%; n=88) or theft-related offences (44%; n=87). Of the theft-related offences leading to pursuits, 86 percent (n=75) were thefts of motor vehicles, usually the vehicle being pursued. For this reason, if motor vehicle thefts continue to decline in Australia, this may have a positive impact on the number of theft-related pursuit fatalities.

Shooting deaths

Since 1989–90, there have been 199 deaths resulting from a shooting in police custody or a custody-related operation. In total, more than half (53%; n=105) were persons shot by police, with just over two in five deaths (46%; n=92) being persons shooting themselves in the presence of police. In the remaining two shooting deaths, one was a non-Indigenous person shot by a government official during an armed robbery and in the other case, the coroner was unable to determine who fired the shot that killed the deceased. Over the last 22 years, there have been 13 shooting deaths involving an Indigenous person, of which eight (62%) were persons shot by police and five (39%) were self-inflicted shootings.

Of the 105 persons shot by police, 102 (97%) were male, while only three (3%) were female. In terms of age, persons shot by police were more likely to be 25 to 34 years (42%; n=44), while the least likely to be shot were those less than 20 years old and those over 49 years (7%; n=7 and 4%; n=4 respectively). Unlike motor vehicle pursuit deaths that are generally precipitated by theft or traffic offences, most police shootings occurred following a violent offence (69%; n=72). Finally, most police shooting take place in either a public place (45%; n=47) or private property (35%; n=37).