The use of force by security personnel to restrain people is a hot topic in many circles. Occasionally tragic circumstances arise whereby the restrained subject dies during the incident and the finger is usually pointed at the security personnel involved. However, deaths proximal to episodes of physical restraint are actually rare occurrences, especially given the number of people who are restrained by security personnel on a weekly basis. If you consider the number of restraint episodes involving the private sector (that is, security personnel or private corrections organisations) plus those performed by police, corrections, community health, mental health, and other government bodies, then the number of physical restraints annually is staggering and most of these end with no significant injury. However, we must remember that the possibility of injury or death exists any time a security officer is required to physically intervene and physically restrain a person.

Tragically some of these restraint episodes do result in the death of the person being restrained. This article will not focus on deaths involving security personnel where the cause of death was attributed to a strike and/or the deceased falling down after being struck and suffering catastrophic head injuries: cases such as the death of David Hookes or more recently, the death of Wilson Castillo in Kings Cross.

Specifically this article will address deaths during direct physical restraint by security personnel, and the most common mechanisms that contribute to those deaths. Some of the high profile cases of recent years include the following tragic deaths resulting in criminal charges against the security providers: Domenic Esposito at the Ramsgate Hotel in South Australia; Dean Eustice at an Adelaide Shopping Centre; Terii Tararo at the Fishermans Wharf Tavern on the Gold Coast; and William Amaya at Brisbane’s Royal Exchange Hotel.

Any untimely death is a tragedy and a death during a restraint while in custody or care seems even more so. Normally, any death that occurs during custody or care will come under the scrutiny of the coroner (depending on the legal jurisdiction) and will be subject to intense examination. They also tend to receive lengthy public and media scrutiny and a search for blame. In purely financial terms, where there is blame today’s litigious society is also looking for compensation.

‘Restraint related death’ is a broad term. It covers the identified contributing factors that combine to result in the death of a person during, or shortly after restraint. The focus of this article is on the phenomenon of restraint related death and a review of the identified causes, rather than the legal and ethical issues surrounding restraint. Suffice to say, a reduction in the number of restraint episodes would certainly lead to a decrease in exposure to the risk of subject death and staff injury during restraint.

The hypothesis that positional asphyxia was the cause of death during restraints was initially proposed in a 1988 journal article (Reay et al. 1988); it was a theme that Reay followed for over a decade (Reay 1998) and a number of newspaper articles and pseudo-scientific reports joined in during that period. This led to the development of the hypothesis that the prone position of restraint (face down) impaired the subject’s ability to breathe and subsequently led to asphyxiation; proponents of the hypothesis argued that prone restraint should be outlawed. It is evident now that these authors drew premature and incorrect conclusions from their highly subjective reports with the conclusion that there was, or is, a single cause of restraint related death.

One clear problem with the previously examined data is that at the time (and indeed, today), almost all restraints were deliberately taken to the floor due to the tactical advantage this gives the restraint team via the effective use of biomechanics and leverage. This meant that almost all physical restraints of violently struggling people were prone. It is therefore not surprising that prone restraints feature prominently in the data. Furthermore, the term ‘prone restraint’ does not describe just one procedure but rather a wide range of situations in which the person is held on the floor ‘face down’. It must be noted too that many of the studies and subsequent discussion focused on the US practice of the hobble or hogtie restraint, which varies greatly from the techniques used in Australia.

Other researchers (Chan et al. 1998) established through their research that the prone position alone had little to no measurable effect on breathing, while later authors used objective criteria in their studies and literature reviews and subsequently identified far more factors which, alone or combined, were more likely to cause death during or immediately after restraint. Stratton et al. (2001) conducted research related to the sudden and unexpected death of people requiring restraint for excited delirium and found that most people restrained in the prone position did not die (91%); and of those who did, all of the deaths were preceded by violent struggle requiring forced restraint; most had stimulant drugs in their system (78%); more than half had chronic disease or were classified as obese (56% each).

Other researchers (Mohr & Mohr, 2000; Mohr et al. 2003) have identified even more causes of restraint related deaths. In most cases, pre-existing conditions, medications and intoxication were found to be causative factors.

Confusion often surrounds the causes of restraint related death because in many cases the causes do not provide typical pathologic evidence that can easily be investigated at autopsy. Often the only sources for a determination of death are the events surrounding the physical struggle. Obviously, this is an inexact science and often leads to incorrect conclusions or the identification of a single cause when it has become increasingly evident that there are concomitant factors involved in causing death.

Most reported cases of sudden and unexpected death during restraint involve young men in an ‘excited’ state or one of ‘agitated delirium’ as a result of psychiatric illness or intoxication from illicit drug use. These individuals were combative, violent, and often struggled or suffered traumatic injuries as a result of a confrontation with law enforcement before their placement in the restraint position. Similar descriptions usually precede restraint related deaths involving security in healthcare environments.

Excited delirium syndrome is a term that appears in much of the restraint related death literature. This is a highly contentious description of a number of symptoms that are used to describe someone who is at risk of restraint related death. It is a term often used post mortem. Excited delirium syndrome is also referred to as agitated delirium syndrome. It must be noted that excited delirium syndrome is not a diagnosis or a disease; it is a visible collection of risk factors associated with restraint related death. The signs and symptoms of excited delirium can be found with even the most cursory of internet searches or reviews of medical literature.

A review of the literature indicates that restraint related death is usually due to a small number of common factors playing a much bigger role than previously anticipated. The three most significant factors are:

#1 The duration of the event:

Simply put, the longer it takes to control the violent person the more at risk they are of dying. There are a number of mechanisms involved, primarily rhabdomyolysis, metabolic acidosis and the catecholamine effect.

Rhabdomyolysis is the rapid breakdown of skeletal muscle due to muscle tissue injury. The most common causes of rhabdomyolysis are alcohol abuse, muscle overexertion, muscle compression and the use of certain medications or illicit drugs. Amphetamines, ecstasy and cocaine increase the risk of toxicity. Methamphetamine and cocaine in particular are implicated as common causes of rhabdomyolysis. Security officers in healthcare facilities dealing with mental health patients must also note that drugs used in mental health care are all thought to predispose patients to the development of rhabdomyolysis.

Prolonged intense exercise of the type associated with a restraint episode (fleeing or fighting; or perhaps both) should be seen as a warning sign for rhabdomyolysis in any person. Rhabdomyolysis is also more likely if a person has underlying medical issues such as heart disease, hypertension, diabetes, obesity; is using illicit drugs; or is on psychotropic medication. Take note: the longer the episode, the more likelihood of rhabdomyolysis.

Delirium, induced by psychosis or drugs, may alter pain sensation and allow the subject to struggle harder and longer, demonstrating a level of physical exertion far beyond normal limits. This may result in severe metabolic acidosis and death.

During violent activity there is an abundant release of the neurotransmitter catecholamine into the blood stream, which can sensitise the heart and promote rhythm disturbances and enhance the toxicity of some stimulant drugs, leading to seizures, respiratory arrest, and cardiac arrest. The catecholemine effect was recently cited as a major contributing factor to the restraint related death of William Araya in Brisbane in 2006.

#2 Drug interactions:

It is widely accepted in the medical community that illicit stimulant drugs, as well as some psychotic prescription drugs, may cause terminal heart defects. In Australia, methamphetamine rather than cocaine is likely to be the stimulant causing the toxic interaction resulting in excited delirium-type symptoms and result in hospital admission. Australia has one of the highest levels of methamphetamine use in the world; in recent years usage has increased with more people dying of methamphetamine than cocaine overdose in this country.

Researchers have proved that methamphetamine is toxic to the heart and increases catecholamine activity in the branch of the peripheral nervous system responsible for modulating heart rate and blood pressure. This may lead to long term toxicity causing sympathetic overdrive, cardiovascular collapse, and death.

Notably, many restraints in entertainment precincts involve subjects who are experiencing a methamphetamine induced psychotic episode. Given the damage that the abuse of these stimulants causes, the cardiac stress of the struggle usually involved with a restraint episode obviously raises the risk of death.

#3 Untrained and/or chaotic response:

Untrained responses to restraint situations frequently feature in restraint deaths. A review of the available literature and of videos (the death of Peter Dalamangous at Star City Casino, 1998; the death of Faisal al Ani, 2009) and media reporting (Weiss in the Harcourt Courant, 1998) demonstrates that in many cases of death proximal to a restraint, the response was chaotic, untrained and had no fixed plan or methodology. This often leads to abnormal positioning, excessive weight on the torso, neck compression and prolonged restraint. For example, the inquest into the death of Sparka Isarva Huntington found that “officers who responded to calls for assistance were untrained for the task given to them. An uncoordinated, chaotic, prolonged struggle only ended when the patient died.”

The ‘stacks on’ or use of raw bodyweight to restrain has led to a number of deaths, both here and abroad. This type of restraint often leads to compression asphyxia, where the victim simply cannot breathe due to the weight on their torso. A trained and deliberate response goes a long way to reducing the risk of death.

Remember, any restraint holds inherent risks and any restraint can end in a death, therefore you should only restrain someone if it is absolutely necessary for the protection of that person, yourself, or other people, and when there is no viable less restrictive option. A ‘cordon and contain’ approach is safer for all concerned. If a restraint is necessary:

  1. Use a planned and calm response
  2. Maintain communication with the subject
  3. Have additional staff available who are not involved in the restraint to monitor the situation
  4. Complete the takedown procedure as quickly as possible to minimise the time in a restrictive position
  5. Avoid weight on the subject’s torso once on the ground
  6. Avoid putting the subject in a position which elevates their risk of positional asphyxia (up against a wall, jammed in a corner, or against a piece of furniture)
  7. If pain compliance is required to gain control, attempt mandibular angle and less invasive techniques first
  8. If you are in a healthcare setting or dealing with known mental health issues, be aware of the subjects underlying medical conditions where possible

Medical attention should be sought any time a person is restrained, particularly if they exhibit the signs and symptoms associated with excited delirium syndrome. It must be remembered that the factors that contribute to restraint deaths may continue to have an invisible effect hours after the actual restraint episode.

All staff involved in a restraint episode should be currently certified as competent in a recognised, evidence based restraint system that has been specifically developed for use as a restraint system, as opposed to a martial arts, combatives or self defence system. They should also have undergone an extensive risk management process and be able to provide the relevant support and documentation.

None of these measures can guarantee the safety of the subject, there are simply too many hidden factors involved, however they can minimise the risk of restraint related death.

A full reference list is available on request from