THE Tasmanian Mental Health Act defines chemical restraint as a “medication given primarily to control a person’s behaviour, not to treat a mental illness or physical condition”.

Acute agitation or behavioural disturbance is a common presenting problem to Australian emergency departments and hospitals. Sedative medication is used to manage the agitation and, in some cases, treat a mental illness or physical condition. However, in many cases, it is, by definition, a form of chemical restraint used to control a patient’s behaviour.

The implicit goal of chemical restraint at the hands of health care providers is to ensure the safety of the patient and those trying to care for them during a period of acute agitation that makes informed consent and adequate assessment difficult, if not impossible.

To ensure best practice and prioritise safety, one would think chemical restraint education and guidelines would be consistent and evidence-based across Australia.

They are not. There is no national guideline or standard.

Below are some examples of the variations across Australian states and territories in sedation or chemical restraint guidelines:

While there has been a concerted effort to move towards a universal consensus guideline, notably between Queensland and NSW, it is obvious that significant variation in chemical restraint practice remains.

This variation can lead to increased risk and the following four cases of deaths related to acute chemical restraint should be a wake-up call for all health care providers charged with the provision of this emergency duty.

The 2007 death of David Lee (page 70) was due to excessive sedation from midazolam infusion in Port Hedland hospital. He had schizophrenia and was under the involuntary status of the Mental Health Act. A follow-up report (page 13) concluded that a lack of adequate staffing should not be justification for using excessive chemical restraint measures.

A very similar case of a chemical restraint-related death was noted here in 2009 in WA (page 22), when an Indigenous man with schizophrenia was excessively sedated with midazolam infusion. The cause of death was pulmonary embolus, but it is likely that this was a result of prolonged immobilisation as a result of the excessive sedation, its resultant pulmonary aspiration syndrome and need for intensive care unit level admission.

In 2010, in Townsville, Lyji Vaggs died after chemical restraint with olanzapine and then midazolam for acute agitation in the mental health unit of the main hospital. Unlike Lee in WA, there was adequate staffing during the admission of Mr Vaggs, including police.

In 2015, David Dungay died in the psychiatric unit of the Long Bay prison after being physically restrained and then administered midazolam via injection. He had been diagnosed with schizophrenia while in prison and was improving on appropriate medication. The coronial inquest into his unnatural death is still pending.

These four men were all Indigenous and either under care of the prison system or the local mental health act.

The use of midazolam infusion has never been supported by any official guidelines in Australia and the origins of its use for acute behavioural disturbance remain a mystery. While the use of midazolam injection is supported by several guidelines, it is my opinion that it carries significant risks in inherently risky situations, such as the acutely agitated patient.

The push to remove midazolam from clinical guidelines for this high risk patient group should be supported, as Queensland Health and NSW Health have done in past 2 years.

We can do better. We must do better if patients are to trust that safe care is our priority, rather than variable care.

It is grimly ironic that midazolam has been used as a lethal injection substitute for capital punishment in the United States, since thiopentone sourcing has become problematic.

Dr Minh Le Cong works for the Royal Flying Doctor Service in Queensland and is the chief editor of the PHARM (Prehospital and Retrieval Medicine) blog site and podcast, the FOAM4GP blog site and the Ketamine Leadership Academy blog site.


To find a doctor, or a job, to use GP Desktop and Doctors Health, book and track your CPD, and buy textbooks and guidelines, visit doctorportal.



We need to do a better job addressing depression in our elderly patients
  • Strongly agree (81%, 35 Votes)
  • Agree (14%, 6 Votes)
  • Disagree (2%, 1 Votes)
  • Strongly disagree (2%, 1 Votes)
  • Neutral (0%, 0 Votes)

Total Voters: 43

Loading ... Loading ...

8 thoughts on “Chemical restraint or lethal injection?

  1. Philip Benjamin says:

    Having worked in psych assessment in a number of states over the past 20 years, my view on the abuse of parenteral sedation is that, despite mental health problems being the cause of 25% of presentations, most staff (especially junior staff) have no training in managing, or even understanding, the needs of these patients, or the way in which their own behaviours lead to escalation in the level of distress and agitation can cause. Surely remediation of this side of the issue is as important as the relatively simple matter of establishing best practice.

  2. Robert Reznik says:

    I could not agree more. While all mental health staff are told of the need to de-escalate situations there is often only lip service given to this process. Trying to understand the patients agitation especially if psychotically driven can help avoid such deaths. Agitated psychotic patients require quiet reassurance that they are safe, yelling at them and telling them about your rules only increases their level of fear and continues to drive their agitation.

  3. Anonymous says:

    The variability of presentation in acutely agitated and possibly psychotic individuals diminishes the need for a rigid guideline that does not offer such variability of approach, per individual patient. The capacity to monitor a patient should they be given sedation, should also inform the type of sedation given. The need to de-escalate the patient is inherent in almost all presentations. However, there are circumstances where a patient is presenting with severe aggressive responses (e.g. amphetamine induced psychosis) that increases the risk of significant harm to self, staff and other patients. De-escalation may not impact at all on such patients, and I would be interested to hear what my colleagues would suggest as an alternative to medication in such presentations? Above and beyond this, appropriate use of medication to stabilise the patient is actually good practice. Guidelines therefore remain guidelines, to inform and guide us.

  4. Minh Le Cong says:

    Hi there and thanks for comments so far
    Guidelines that are evidence based from decent Australian research are needed because there is variability in approaches to this risky situation. The variable approach is due to inadequate education and training on how to best manage the undifferentiated acutely agitated patient as Philip points out. The variability in guidelines and lack of decent research in the past has perpetuated this conundrum.

    However as I wrote about and cited, the latest Queensland and NSW emergency department guidelines for acute severe behavioural disturbance, are utilising decent Australian emergency research and provide evidence based foundation towards best practice.

    Non pharmacologic interventions are not well taught, this is true. Pharmacologic interventions should be focussed towards safety and efficacy and the latest Australian research has informed us well. I suggest we at least follow our state based guidelines and try to adopt a more evidence based approach rather than saying its too hard to predict.

    Ideally I would like to see a national guideline adopted for behavioural disturbance and a consensus approach across the fields of psychiatry, emergency medicine, general practice and retrieval medicine.

  5. Michael King says:

    As an emergency practitioner working in Tasmania and having been involved in educating local staff on the Mental Health Act 2013, I am concerned that you used “our” definition of chemical restraint but then did not make any further comment about the Tasmanian experience in approaching patients with acute behavioural disturbance. I don’t agree with your comment that “in many cases, it is, by definition, a form of chemical restraint used to control a patient’s behaviour” when referring to sedative medication. By definition it is sedative medication, used to induce a calmer state. The specific goal is to relieve the distress associated with the agitated state, whatever the underlying cause for the agitation. The need or otherwise to modify the person’s behaviour is a secondary consideration and always related to their safety, not our convenience or to make up for a lack of staff.

  6. Minh Le Cong says:

    thanks Michael
    I wrote this in 3rd paragraph :’The implicit goal of chemical restraint at the hands of health care providers is to ensure the safety of the patient and those trying to care for them during a period of acute agitation that makes informed consent and adequate assessment difficult, if not impossible.”

    As for Tasmanian emergency department guidelines, please provide a reference because I could not find a publicly available one.

    sadly, tragically, the Indigenous deaths I cite from publicly available resources such as coroners websites, clearly show that despite the best intentions, chemical restraint is medically administered/authorised pharmacological sedation used at times to control agitated behaviour when insufficient staff and/or appropriate training/resources are lacking . For example, the rural hospital that lacks authorised secure mental health beds , necessitating aeromedical transfer of an involuntary patient. The case of David Lee is so relevant here. He was sedated excessively with midazolam infusion because the hospital staff were awaiting aeromedical transfer due to lack of secure mental health beds and there were insufficient staffing to allow safe endotracheal intubation and monitoring of a ventilated patient.

    My point is we have to admit there are risky situations arising in hospitals across Australia. Not every rural hospital has the resources of a major tertiary hospital yet have to deal with the same risks when an acutely agitated patient presents. The problem is guidelines have been variable across Australia and its when guidelines written for well resourced emergency departments are not just able to be easily translated to rural and less resourced settings, this is when the patient safety risk increases. Midazolam might be fine in a well staffed capital city emergency department with acute psychiatric services on tap and well monitored secure beds. It certainly didnt help David Lee or the others I cited.

    If we are saying that these tragic deaths are as a result of the best medical care we can offer in Australia to patients who cannot consent to their treatment nor the medications administered to them, then I say we should do better.

  7. Dr Jennifer Smith says:

    I am a rural GP working in acute ED as well as general office practice, and have been actively seeking out training in de-escalation and safe physical response training for my own and patient’s safety, but have not found it. Where is the training we can do? Why is it not easily available and advertised? Happy to follow up anyone’s recommendations in how to pursue training in this area if you have found courses.

  8. Minh Le Cong says:

    Hi Jennifer
    My rural colleague Casey Parker suggests this course

Leave a Reply

Your email address will not be published.