Skip to main content

Deaths in police custody: A review of the international evidence

Research Report 95

This report provides a summary of the published research literature and administrative data on deaths in or following police custody, and apparent suicides following police custody, to support the independent review chaired by the Rt Hon. Dame Elish Angiolini DBE QC, which was announced in July 2015. The review’s terms of reference include:

  • reviewing the processes and procedures surrounding deaths in police custody in England and Wales; and
  • identifying why investigations following such deaths have, according to the then Home Secretary, “fallen short of many families’ needs”.

This report provides a review of statistics and research to answer five key questions.

  • What are the extent and trends in deaths in or following police custody, and apparent suicides following police custody, in England and Wales?
  • What are the extent and trends of such deaths in comparable Western countries?
  • What are the main causes of deaths in police custody, and suicides following police custody, in England and Wales?
  • What research evidence is there for procedures or ‘good practice’ to prevent or reduce deaths in or following police custody?
  • What evidence is there of ‘good practice’ for the running and management of investigations into deaths in or following police custody?

There were 14 deaths in or following police custody in England and Wales in 2015/16, in line with the average for the last 7 years, but lower than the levels seen in the late 1990s and early 2000s, according to statistics collected by the Independent Police Complaints Commission (IPCC). There were 60 apparent suicides following police custody in 2015/16. Apparent suicides following police custody increased markedly between 2011/12 and 2012/13, from 38 to 64, although this is thought to be in part the result of improved identification of these cases.

Various factors are thought to have contributed to the previous reductions in deaths in or following police custody. The sharp reduction in deaths between 1998/99 and 1999/2000 (49 to 31) was largely due to a reduction in suicides within police custody. This has been attributed to several developments. These include the removal of ligature points in cells and increased use of CCTV in cells, which in turn reduced hangings. During the period from 2003/04 to 2008/09, the introduction of the revised Police and Criminal Evidence Act 1984 (PACE) Code of Practice C regarding the handling of detainees, and the reduction in arrestees going through police custody suites, may both have contributed to reductions in deaths in custody. However, there is no definitive evidence to link these developments to the fall in these deaths.

International comparisons of rates of deaths in police custody are complicated by the use of differing definitions and the lack of accessible data. However, there are a small number of countries where broadly comparable data are available. The rate of deaths in or following police custody in England and Wales was similar to rates of deaths in police custody in New Zealand during the 2000s (rate of 0.6 deaths per million population per year in New Zealand, compared with an equivalent rate of 0.5 in England and Wales). They were also similar to Australia between 2003/04 and 2012/13 (0.3 in Australia compared with 0.4 in England and Wales). Data for a more limited period (2013/14 and 2014/15) suggest rates of deaths in police custody were higher in Scotland compared with England and Wales (0.8 in Scotland compared with 0.2 in England and Wales).

Those who die in police custody in England and Wales are typically male, aged between 31 and 50, and from a White ethnic background. Other comparable Western countries for which data was found show a similar demographic profile
Those who die from suicide following police custody in England and Wales are also typically male, aged between 31 and 50, and from a White ethnic background. Individuals arrested for sexual offences are much more likely to die from an apparent suicide following police custody compared with those arrested for other offence types (12 times higher than the average). International evidence also suggests that people arrested for sex offences, particularly child sex offences, appear to have a higher risk of committing suicide following police custody.

Natural causes have been the most common known cause of deaths in police custody in England and Wales between 2004/05 and 2014/15, accounting for 51 per cent of deaths in this period. Drugs and/or alcohol also featured as causes in around half of deaths (49%). An even higher proportion of those who died had an association with drugs or alcohol (82%)

The predominance of these types of causes of death suggests that screening processes for arrestees on their reception into custody are important in identifying risks for police detainees. Studies in England and Wales have found that both risk assessment tools and their use have sometimes been deficient. Evidence from international reviews of deaths in police custody also suggests that screening needs to be followed up by regular monitoring of those in custody and the communication of risk to other staff involved in detainee care.

Technology, such as CCTV, or life signs monitoring equipment, can be installed in cells to enhance monitoring of detainees. There is anecdotal evidence that this has prevented deaths in custody but little robust evidence for effectiveness. Research on the use of this technology emphasises that it can only enhance, not replace, monitoring of detainees by custody officers and staff.

Providing alternatives to police custody for vulnerable individuals is another key theme in the international literature.

  • In some parts of Australia, ‘sobering-up’ centres are established alternatives to police custody for those who are intoxicated. There is some indication that these centres can reduce the number of intoxicated people taken into police custody, but there is no robust evidence around improved health outcomes or reduced deaths of detainees.
  • There have been promising results from pilots of ‘street triage’ in England, where mental health nurses accompany officers to relevant incidents, in reducing police custody as a place of safety for those detained under the Mental Health Act 1983. But here too, the strength of evidence is weak.
  • A review of the evidence on Crisis Intervention Teams, an intervention developed in the US involving training officers in dealing with people with mental health needs in combination with a designated referral point, found no overall effect on arrests of mentally ill persons.

Police use of restraint against detainees was identified as a cause of death by post-mortem reports in 10 per cent of deaths in police custody between 2004/05 and 2014/15. However, a higher proportion of deaths would have restraint used at some point during detention. Use of restraint has been found to be more prevalent in cases of Black and Minority Ethnic (BME) individuals who have died in police custody than in deaths of White people. Police use of force has also been found to be greater amongst those with mental health problems. To reduce police use of force, de-escalation techniques are seen to be the most appropriate intervention. Training police officers in such techniques shows promise in reducing use of force, but further evaluations are needed.

Several international institutions have established principles for the effective investigation of deaths in police custody. These principles include the independence of investigations and the involvement of next of kin in the investigative process. Research has identified some barriers to the perceived independence of the investigation through the employment of ex-police officers by police complaint bodies, while recognising the valuable knowledge and skills they bring to investigations. The limited available research into the views of the families of those who have died in police custody has found that they have felt that communications from investigators were inadequate. Such criticisms have previously been directed at the IPCC. The IPCC has recognised these issues in its review of its handling of deaths following police contact, and has worked to address them.