Deceased
Michelle Williams
Demographics
46y, female
Coroner
Coroner Simon McGregor
Date of death
2016-08-29
Finding date
2020-12-18
Cause of death
Hypoxic brain injury in the setting of plastic bag asphyxia
AI-generated summary
Michelle Williams, a 46-year-old woman with longstanding schizophrenia and depression, died from hypoxic brain injury due to plastic bag asphyxia while an involuntary inpatient in a psychiatric unit. On 27 August 2016, during a period of heightened agitation about being unable to leave the ward, she had access to a plastic bag from a bin and was found unresponsive with the bag over her head. Critical failures included: insufficient staffing coverage led to a missed hourly observation at 12.00pm when two LDU nurses were assisting in the HDU; failure to perform a required formal risk assessment despite observable deterioration in her mental state that morning; and inadequate environmental controls allowing plastic bags in the inpatient unit. While PRN medication was not administered during her morning agitation, evidence suggests this would not have prevented her death. The coroner found these failures directly contributed to her death and commended subsequent systemic improvements in staffing protocols, risk assessment procedures, and environmental safety measures.
AI-generated summary — refer to original finding for legal purposes. Report an inaccuracy.
Specialties
Error types
Drugs involved
IN THE CORONERS COURT Court Reference: 2016 4083
OF VICTORIA AT MELBOURNE FINDING INTO DEATH WITH INQUEST Form 37 Rule 60(1) Section 67 of the Coroners Act 2008 Inquest into the death of: Michelle Williams Delivered on: 18 December 2020 Delivered at: Coroners Court of Victoria 65 Kavanagh Street, Southbank, Victoria Hearing date: 11 December 2020 Findings of: Simon McGregor, Coroner Representation: Roslyn Kaye of Counsel, instructed by K&L Gates For Bendigo Health Counsel Assisting the Coroner: Daniel Wright, Coroner’s Solicitor
Michelle’s death was reported to the Coroner as it fell within the definition of a reportable death in the Coroners Act 2008 (the Act). Reportable deaths include deaths that are unexpected, unnatural or violent or result from accident or injury, as well as deaths of persons who were, immediately before death, ‘persons placed in custody or care’ for the purposes of the Act.
As Michelle was detained in a designated mental health service within the meaning of the Mental Health Act 2014, she was a ‘person placed in custody or care’. In addition to making her death reportable, this required me to hold an inquest into her death pursuant to section 52(2)(b) of the Act.
The role of a coroner is to independently investigate reportable deaths to establish, if possible, identity, medical cause of death, and surrounding circumstances. Surrounding circumstances are limited to events which are sufficiently proximate and causally related to the death. The purpose of a coronial investigation is to establish the facts, not to cast blame or determine criminal or civil liability.
Under the Act, coroners also have the important functions of helping to prevent deaths and promoting public health and safety and the administration of justice through the making of comments or recommendations in appropriate cases about any matter connected to the death under investigation.
Victoria Police assigned an officer to be the Coroner’s Investigator for the investigation of Michelle’s death. The Coroner’s Investigator conducted inquiries on my behalf, including taking statements from witnesses – including family, the forensic pathologist and treating clinicians – and submitted a coronial brief of evidence, which was received on 4 April 2017.
As it was clear that there were concerns with the care provided by Bendigo Health, the Coroner’s Investigator was requested to obtain additional statements and records from Bendigo
Health staff. After these were all received in January 2018, the file was reviewed by mental health case investigators in the Coroners Prevention Unit (CPU).1
The CPU identified several issues and conducted further investigation, obtaining additional materials from Bendigo Health in response to targeted questions. Once this investigation and the review of its results were completed, the CPU provided a report to me in December 2019.
After considering this report, I determined that several issues remained unresolved regarding changes made by Bendigo Health after Michelle’s death. After further correspondence with Bendigo Health relating to this issue, they provided new statements which satisfied my concerns.
As there did not remain any disputes about the factual circumstances surrounding Michelle’s death, I considered it appropriate to hold a summary Inquest, without calling any witnesses, which occurred on 11 December 2020. In advance of this inquest, Counsel Assisting prepared submissions setting out a number of criticisms of Bendigo Health, and these were provided in advance to representatives of Bendigo Health so that their Counsel had opportunity to respond to these criticisms.
After hearing submissions from Counsel Assisting and from Counsel for Bendigo Health at inquest, as well as comments from Michelle’s family spokesperson Matthew Mulkearns, I am satisfied that no further investigation is necessary and that I am able to make the findings required under the Act.
This finding draws on the totality of the coronial investigation into the death of Michelle Williams, including evidence contained in the coronial brief, medical records, evidence obtained in further investigations, and submissions made at inquest. Whilst I have reviewed all the material, I will only refer to that which is directly relevant to my findings or necessary for narrative clarity. In the coronial jurisdiction, facts must be established on the balance of probabilities.2 1 The Coroners Prevention Unit (CPU) was established in 2008 to strengthen the prevention role of the coroner. The unit assists the Coroner with research in matters related to public health and safety and in relation to the formulation of prevention recommendations. The CPU also reviews medical care and treatment in cases referred by the coroner. The CPU is comprised of health professionals with training in a range of areas including medicine, nursing, public health and mental health.
2 Subject to the principles enunciated in Briginshaw v Briginshaw (1938) 60 CLR 336. The effect of this and similar authorities is that coroners should not make adverse findings against, or comments about, individuals unless the evidence provides a comfortable level of satisfaction as to those matters taking into account the consequences of such findings or comments.
MATTERS IN RELATION TO WHICH A FINDING MUST, IF POSSIBLE, BE MADE Circumstances in which the death occurred
She was born in Moonee Ponds and attended Aberfeldie Primary School then Buckley Park Secondary School.3
She began to have mental health issues as an adolescent, and received inpatient treatment, eventually being diagnosed with paranoid schizophrenia and depression. She also developed substance abuse issues at around this time which continued throughout her life.4
Michelle had three children with Richard McInnes: Justin, Ricky and Corey, and she later had another child, Tia.5
Through the 1990s Michelle had a chaotic life with multiple psychiatric admissions and at least one suicide attempt. However, from around 2000 to 2011 she reportedly seemed stable and was on regular medication.6
From 2011 onward her condition began to decline again.7 She was referred to the community mental health team at Bendigo Health in October 2014 following a relapse in psychotic symptoms and alcohol and other substance misuse. She was referred to the mental health team again by her GP on 2 August 2016.8 Admission to Alexander Bayne Centre
Around 9 August 2016, Michelle’s friend Fiona Harris became concerned about her mental state. She had recently ended a relationship, she was drinking alcohol to excess and was also using methamphetamine, cannabis and excess diazepam. She was no longer taking her 3 Statement of Ian Williams dated 5 December 2016.
4 Statement of Ian Williams dated 5 December 2016.
5 Statement of Ian Williams dated 5 December 2016.
6 Statement of Ian Williams dated 5 December 2016.
7 Statement of Ian Williams dated 5 December 2016.
8 Statement of Dr Allison Newman dated 16 January 2018.
prescribed antidepressant and antipsychotic medications and admitted to hearing voices, hearing birds singing and seeing vivid images.9
Michelle was also considered a risk to herself as she was walking in front of traffic and talking about the world hating her. She reportedly felt that everything that happened was her fault and that the world would be better off without her. She was highly agitated and thought disordered, was banging her head and mumbling, and had spoken about driving her car into a tree.10
On 9 August 2016 Michelle was admitted to the ABC, the acute inpatient mental health unit at Bendigo Health. She was admitted as a voluntary patient, but when she was reviewed on the next day by Consultant Psychiatrist Dr Allison Newman, Dr Newman considered that she was distressed and psychotic without the capacity to make treatment decisions. Dr Newman therefore changed her status to that of an involuntary patient.11
Dr Newman placed Michelle on the open ward in the ABC and instituted hourly observations.
Michelle expressed vague plans to self-harm but no intent, and her suicidality and self-harm risks were assessed as moderate.12
Dr Newman reinstated Michelle’s regular medications (which she had not been taking), 5mg twice daily of the antipsychotic olanzapine and 50mg once daily of the antidepressant desvenlafaxine. She also instituted an alcohol and benzodiazepine detoxification schedule, alcohol withdrawal scale monitoring and nicotine replacement therapy.13
Over the next days Michelle’s mental state appeared to gradually improve.14 Incident on 12 August 2016
On 12 August 2016, a nursing risk assessment found Michelle’s risk of suicide to be low.15
One of the nurses on the evening shift that day was Rochelle Battersby. At around 9.00pm Nurse Battersby was checking on her patients when she went to see Michelle. Nurse Battersby 9 Statement of Dr Allison Newman dated 16 January 2018.
10 Statement of Dr Allison Newman dated 16 January 2018.
11 Statement of Dr Allison Newman dated 16 January 2018.
12 Statement of Dr Allison Newman dated 16 January 2018.
13 Statement of Dr Allison Newman dated 16 January 2018.
14 Statement of Dr Allison Newman dated 16 January 2018.
15 Statement of Dr Allison Newman dated 16 January 2018.
found Michelle with what she describes as an ‘orange bag’ which was ‘very frail’ around her neck – the bag was not tied, but was draped over Michelle’s shoulders and back.16
Michelle denied being suicidal, but was very distressed and apparently delusional. Nurse Battersby comforted her, then asked her for the bag and Michelle gave her the bag without objection.17
After this, Nurse Battersby spoke with the shift manager about these events, then the medical officer. She showed the bag to each of them and asked if they wanted to keep it or find out what kind of bag it was, but neither of them pursued this. As Michelle had had gastroenteritis, they asked Nurse Battersby to dispose of the bag in a medical waste bin. They then asked her to check on Michelle more regularly.18
Nurse Battersby went back to see Michelle again and found that she had settled. She then handed Michelle’s care over to the night shift.19 Treatment between 12 August 2016 and 25 August 2016
Despite this event, Michelle’s suicide risk continued to be assessed by staff as low. As she seemed to struggle with being on the ward, Dr Newman granted her 15 minutes unescorted leave on the grounds – however, Michelle reportedly ‘became distressed and disoriented on the grounds’, so no further unescorted leave was given.20
Due to Michelle’s agitation and distress, she was given doses of diazepam and olanzapine on a ‘prn’ basis.21
On 19 August 2016 she informed a psychiatrist that she felt ‘90% normal’ and was ‘ready to go home’. She denied psychotic symptoms and showed no evidence of a thought disorder.
Although angry and disappointed that she couldn’t be released that day, she presented as positive and future-planning and her suicide risk was assessed as low.22
17 Statement of Rochelle Battersby dated 5 March 2017.
18 Statement of Rochelle Battersby dated 5 March 2017.
19 Statement of Rochelle Battersby dated 5 March 2017.
20 Statement of Dr Allison Newman dated 16 January 2018.
21 pro re nata or ‘as needed’. Statement of Dr Allison Newman dated 16 January 2018.
22 Statement of Dr Allison Newman dated 16 January 2018.
hours expressing an intention to take her own life by driving her car into a tree. Considering this, Dr Newman determined not to discharge Michelle from the ABC.23
Michelle continued to be distressed over the following days, and continued to contact friends expressing an intent to end her own life. On 24 August Michelle’s treatment was discussed with her father, Ian Williams, by telephone.24 Incident on 25 August 2016
At 3.00am on 25 August 2016, Nurse J Daly was performing a regular round and found Michelle lying on the ground with a sash from her dressing gown around her neck. Nurse Daly activated a duress alarm and staff attended Michelle’s room. Michelle allowed staff to remove the sash, and after staff checked on her they found her physical observations within normal limits.25
When staff spoke with Michelle, she made a number of comments such as ‘I have to die’. Staff performed a risk assessment and concluded that Michelle had a high risk for suicidality and self-harm.26
Associate Nurse Unit Manager (ANUM) Gavin Pavey decided to transfer Michelle to the High Dependency Unit (HDU) for increased supervision and to restrict access to means of self-harm and suicide.27
As there were no available beds, Nurse Pavey directed for her to be nursed in a seclusion room with the door open and to have visual sightings every fifteen minutes. Michelle was to receive medication at 3.25am to decrease her level of agitation.28
Nurse Pavey notified the Duty Doctor, completed nursing notes and completed an incident report. His notes for the rest of the night indicate that Michelle slept intermittently without any notable incidents.29 23 Statement of Dr Allison Newman dated 16 January 2018.
24 Statement of Dr Allison Newman dated 16 January 2018.
25 Statement of Gavin Pavey dated 3 March 2017.
26 Statement of Gavin Pavey dated 3 March 2017.
27 Statement of Gavin Pavey dated 3 March 2017.
28 Statement of Gavin Pavey dated 3 March 2017.
29 Statement of Gavin Pavey dated 3 March 2017.
She continued to be nursed in the HDU throughout the next night. She again slept intermittently, occasionally waking with agitation and disorganised thoughts.30
On 26 August 2016 she was reviewed in the HDU by Dr Newman. Michelle told Dr Newman ‘how she was finding it very noisy and unsettling in HDU’, and they discussed medication options and the possibility of using the ‘sensory room’ in the ABC to manage her distress.
Michelle denied suicidal ideation but claimed that there was a conspiracy against her.31
Dr Newman transferred Michelle to the Low Dependency Unit (LDU), where she would be less constrained. As Michelle was agitated about not being able to smoke while on the ward and not having any leave, Dr Newman provided nicotine replacement therapy and gave ‘prn’ antipsychotic medication to help manage her distress.32 Events of 27 August 2016
On the morning of 27 August 2016, the ANUM in charge of the Alexander Bayne Centre was Heather Somerville. Two nurses, Vicki Abbott and Sarah Toma, were assigned to work with patients on the LDU. Three other nurses were allocated to work in the HDU. Nurse Abbott was allocated as Michelle’s contact nurse in the LDU, and Michelle was to receive visual sightings each 60 minutes.33
At around 7.30am Michelle presented to the nurses’ station asking if she could go outside to smoke a cigarette. Nurse Somerville checked if Michelle had a leave form permitting her to leave the building for either a short 10-15 minute break or for extended leave, and she did not locate any. As Michelle was an involuntary inpatient, Nurse Somerville informed her that she could not go outside to smoke.34
During that morning Michelle called multiple people, including Ms Harris and her father, informing them that she was going to get out that day, and she asked her father to come pick her up. When Mr Williams contacted the ABC that morning to enquire about this, they informed him that Michelle was not being discharged.35
Nurse Abbott states that Michelle seemed ‘very irritable and wanted to leave’ that morning, insisting that she was a voluntary patient and that she could go. Nurse Abbott repeatedly 30 Statement of Gavin Pavey dated 3 March 2017.
31 Statement of Dr Allison Newman dated 16 January 2018.
32 Statement of Dr Allison Newman dated 16 January 2018.
33 Statement of Heather Somerville dated 17 March 2017; Statement of Sarah Toma dated 6 October 2016.
34 Statement of Heather Somerville dated 17 March 2017.
35 Statement of Heather Somerville dated 17 March 2017; Statement of Ian Williams dated 5 December 2016.
explained to Michelle that she was an involuntary patient, and so could not leave. Michelle also repeatedly asked to go outside to smoke, and was declined. Nurse Abbott offered her nicotine spray, but Michelle refused it.36
Nurse Abbott recalls that Michelle seemed frustrated, but did not appear suicidal. Nurse Abbott states that she checked if Michelle had any authorised ‘prn’ medications for use when she was unusually agitated and frustrated. Nurse Abbott states that she found that none were listed, so she did not provide Michelle with any additional medication.37
An entry on the observation chart notes that Michelle was sighted at the nurses’ station at 11.00am.38
At around 11.50am, Nurse Abbott and Nurse Toma went into the HDU to assist staff with restraining a patient who was ‘screaming and abusive’. This process took some time because the patient continued to struggle and required chemical sedation.39
Shortly after noon, kitchen staff informed Nurse Somerville that lunch would be late. Nursing staff, including Nurse Abbott and nurse Sarah Toma, began checking the female end of the low dependency unit to inform patients that lunch would be late.40
As Nurse Abbott was checking names, she noticed that Michelle was missing. At around this time, a patient informed Nurse Toma that one of the showers had been on for a long time. Nurse Abbott and Nurse Toma then went to check on the communal toilet and shower for rooms ten and eleven.41
Nurse Toma opened the door and saw Michelle lying on the floor, fully clothed, with a pink plastic bag secured over her head. The shower was running with warm water and Michelle’s head was under the stream of water from the shower.42 36 Statement of Heather Somerville dated 17 March 2017; Statement of Vicki Abbott dated 3 March 2017.
37 Statement of Vicki Abbott dated 3 March 2017.
38 Bendigo Health Medical Records.
39 Statement of Heather Somerville dated 17 March 2017; Statement of Vicki Abbott dated 3 March 2017.
40 Statement of Heather Somerville dated 17 March 2017.
41 Statement of Heather Somerville dated 17 March 2017.
42 Statement of Vicki Abbott dated 3 March 2017; Statement of Sarah Toma dated 6 October 2016.
Nurse Abbott pulled her duress alarm and additional nursing staff quickly attended. Nurses saw water pooled in the bag around Michelle’s neck and Nurse Abbott pulled the bag off of her head. Nurses moved Michelle from the bathroom and began CPR, while a ‘Code Blue’ cardiac arrest emergency call was sounded.43
The Bendigo Health Code Team attended and took over CPR. Michelle was transferred to the Intensive Care Unit at 1.10pm.44
Although she was treated in ICU for the next several days, it became clear that she would not recover from her brain injuries. In accordance with Michelle’s wishes, she was approved as an organ donor after cardiac death and life supporting measures were withdrawn at 2.20pm on 29 August 2016.45 Identity of the deceased
On 28 August 2016, while Michelle was still in ICU, she was visually identified by Justin McInnes. Identity is not in dispute and requires no further investigation.
Medical cause of death
Forensic Pathologist Dr Yeliena Baber from the Victorian Institute of Forensic Medicine (VIFM), conducted an examination of Michelle and provided a written report of her findings.
Dr Baber provided an opinion that the medical cause of death was ‘1(a) Hypoxic brain injury in the setting of plastic bag asphyxia’.
Toxicological analysis of post-mortem samples identified the presence of olanzapine, quetiapine and desmethylvenlafaxine, in accordance with Michelle’s treatment
43 Statement of Heather Somerville dated 17 March 2017.
44 Statement of Heather Somerville dated 17 March 2017.
45 Statement of Dr Emma-Jane Broadfield dated 30 November 2016.
This review identified a number of issues, including ‘prn’ medication, the performance and accuracy of risk assessments, inappropriate staffing levels between units, next-of-kin contact issues and Michelle’s access to means of suicide.
Submissions were made at inquest in regard to these issues by Counsel Assisting and by Counsel for Bendigo Health. In addition, prior to the inquest representatives of Bendigo Health provided a supplementary statement from Associate Professor Philip Tune, Clinical Director of Mental Health Services at Bendigo Health, which addressed some of these issues.
‘PRN’ medication
When, on 27 August 2016, Nurse Abbott checked to see if ‘prn’ medication was available for Michelle to take in order to limit distress, she did not see that there was any listed.
However, later review of records showed that her medication chart did have ‘prn’ medication prescribed and available.
It is unclear whether there was an error on the part of Nurse Abbott, an error on the part of some other staff member or whether Bendigo Health policies or protocols were responsible.
Submissions at inquest
At inquest, Counsel Assisting submitted that Michelle’s agitation and distress would probably have been reduced had she received her ‘prn’ medication, which in turn would have reduced the likelihood of her attempting to cause her own death.
Counsel for Bendigo Health submitted in response that this was ‘highly speculative and without evidentiary support’. She noted that Michelle had received her ‘baseline’ level of medication, and so was not entirely unmedicated.
In response to this point, Counsel Assisting submitted that it was clear that Michelle had a sufficiently high degree of agitation that a greater dose of medication than ‘baseline’ was justified, which was supported by the fact that Nurse Abbott had found it reasonable to check whether ‘prn’ medication was available.
Counsel for Bendigo Health also raised several key points from A/Prof Tune’s supplementary statement:
(a) that it is not only agitation that drives patients to attempt to take their own life;
(b) that the ‘prn’ medications prescribed to Michelle, olanzapine and quetiapine, do not have definite or direct anti-suicidal properties; and
(c) that at times medication can have little or no effect on a patient.
Conclusions
The evidence supports the conclusion that Michelle’s agitation and distress on the morning of 27 August 2016 were such that it would have been reasonable to provide her ‘prn’ medication in addition to her ‘baseline’ prescribed dosage.
However, Counsel for Bendigo Health is correct that it is ‘highly speculative’ to argue that ‘prn’ medication would have had a substantive effect on the likelihood of Michelle taking her own life. Especially considering the points raised by A/Prof Tune in his supplementary statement, there is insufficient basis to find that the failure of Bendigo Health staff to provide ‘prn’ medication to Michelle on the morning of 27 August 2016 played a causative role in her death.
Performance of risk assessments
The Bendigo Health Psychiatric Clinical Risk Assessment Policy states that a formal risk assessment should be recorded on day shift, evening shift, night shift (unless the patient sleeps through) and following any significant deterioration in mental state.
The CPU advised that Michelle experienced a deterioration in mental state on the morning of 27 August 2016 that would require a risk assessment to be completed as per Bendigo Health policy, and that this was not completed.
Submissions at inquest
Counsel Assisting submitted that I should accept the CPU’s advice on this issue.
Counsel for Bendigo Health contended that Michelle’s condition on the morning of 27 August 2016 did not constitute a significant deterioration such as would justify a formal risk assessment beyond those regularly scheduled.
She referred to clinical records’ description of Michelle that morning as ‘angry, frustrated and irritable that she was not permitted to go on leave’, and compared this to Michelle’s mental
state on the preceding days. In particular, Counsel compared this state to Michelle’s condition on 26 August, when she had ‘appeared anxious with low mood’, and to her condition on 25 August, when she had ‘deteriorated to the extent that she had placed a dressing gown cord around her neck (and was dealt with appropriately at the time, having been transferred into HDU and onto 15 minute observations)’.
She submitted that the question of Michelle’s deterioration should be considered in comparison to these dates, and that therefore there was ‘no omission’ on Bendigo Health’s part in not making a formal risk assessment.
Counsel Assisting contended that it would be inappropriate to compare Michelle’s condition to her condition on 25 August, for the purposes of determining whether she had suffered a significant deterioration that would justify a formal risk assessment. Michelle’s condition had improved since that time, to the point that she was transferred back to the LDU.
Further, Counsel Assisting submitted that the relevant benchmark for Michelle’s condition on 26 August should not be her condition early in the day, as in fact the noisy and unsettling conditions in HDU were recorded as contributing to her distress, but rather that the question of deterioration on the morning of 27 August should be measured against her condition late on the day of 26 August after her transfer back to the LDU.
He raised two points of evidence that supported the conclusion that Michelle’s condition had observably deteriorated on the morning of 27 August:
(a) that Nurse Abbott had considered the possibility of providing ‘prn’ medication in addition to Michelle’s baseline medication; and
(b) that Nurse Toma had specifically stated, of the morning of 27 August, that ‘this was not [Michelle’s] usual behaviour, she had been getting quite well’.46 Conclusions
46 Statement of Nurse Sarah Toma dated 6 October 2019.
If her condition at any point that morning had been observed as identical to her condition on 25 August when she required transfer to the HDU and observations each 15 minutes, then it would have been clearly inappropriate to continue treating her as according to her risk of suicide as assessed the previous night.
It therefore cannot be correct to define a ‘significant deterioration’ as a deterioration compared to her condition when she had required greatly different treatment. That is akin to comparing apples with oranges. If she was to receive regular formal risk assessments at each shift for the purpose of ensuring her treatment was suited to her risk, then the benchmark for comparison when defining a ‘significant deterioration’ must be her condition at the time of her previous regular scheduled assessment within her current treatment paradigm, namely the evening shift on 26 August.
In this light, it is correct to view Nurse Abbott’s and Nurse Toma’s observations on the morning of 27 August as describing a ‘significant deterioration’. A formal risk assessment was therefore required under Bendigo Health policies, and the failure of Bendigo Health staff to perform one constituted an omission.
Accuracy and reliability of risk assessments
Bendigo Health risk assessments are rated across four categories: no current risk, low, medium and high. According to the Bendigo Health Psychiatric Clinical Risk Assessment policy, the risk of suicide, self-harm and aggression is greater in individuals with a past history of these behaviours.
Michelle was rated as having no current risk of suicide on 15 August 2016, twice on 18 August 2016 and 19 August 2016, with ratings completed by a number of different nurses. On each of these occasions, a past history of suicide attempts was also identified. The CPU have advised that these risk assessments were not in accordance with Bendigo Health policy.
Submissions at inquest
Counsel Assisting submitted that I should accept the CPU’s advice, and that considering the policy’s acknowledgement that a past history of these behaviours indicates greater risk, patients with a past history should be rated as low risk even when no current risks are identified.
Counsel for Bendigo Health criticised this approach. While she acknowledged that it might be arguable that ticking a box for ‘low risk’ was preferable to ticking a box for ‘no current risk’, this made no material difference for two reasons. One was that the difference between these two boxes was not great, and the other was that these risk assessments were more than a week prior to the events of 27 August and were therefore not materially causal.
Counsel Assisting conceded that the process of risk estimation is a holistic one, and not a simple ‘tick box’ exercise. He submitted, however, that the ticking of these boxes is a measurable contemporaneous record of these holistic risk assessments, and should be viewed as an indicium of the approach taken by various staff at various times in assessing Michelle’s risk of self-harm.
Conclusions
I acknowledge that, on each individual occasion, the distinction between recording Michelle’s risk as ‘low risk’ rather than ‘no current risk’ would not have significantly affected her care over the following several weeks. I also acknowledge that this is a limited record of a complex and dynamic process of continually assessing risk and adjusting treatment to it.
However, I do consider that it is evidence of a long-term tendency of underestimating, to some degree, the relevance of Michelle’s history of self-harm to her risk of self-harm at the moment.
Although the ticking of boxes is of course not the entirety of risk assessment, I consider it possible to infer from the data we have, the record of boxes ticked, that the holistic risk assessment as a whole suffered from this underestimation, and that this underestimation was inconsistent with Bendigo Health policy.
Systemic aspects of risk assessment policy compliance
In addition to noting that the risk assessments in which Michelle was noted as having no current risk of suicide despite having a past history of suicide were not in accordance with Bendigo Health policy, the CPU observed that these assessments were completed by a number of different nurses, with the same issue replicated.
The CPU advised, on the basis of their review of records, that staff compliance with Bendigo Health policy on risk assessment should be viewed as a systemic issue throughout the Alexander Bayne Centre.
Submissions at inquest
Counsel for Bendigo Health ‘strongly refute[d]’ this position. They noted the limited amount of evidence from which it is drawn, being four occasions in which a single box was ticked ‘no current risk’ rather than ‘low risk’.
Beyond this, she submitted that ‘a broad-ranging inquiry into “issues” with risk assessment at the Alexander Bayne Centre is neither warranted in view of the above, nor is it within the scope of this investigation’, and that ‘There is no evidence of any causal connection between any such “systemic issues” (even if they existed, which is denied by Bendigo Health) and the present matter’.
On the matter of the scope, Counsel Assisting submitted that such an inquiry was not too remote for my investigation, as in order to understand Michelle’s treatment at the Alexander Bayne Centre, it is necessary to consider the manner in which the staff as a whole assessed her risk of suicide, and it is therefore necessary to consider systemic aspects in addition to individual instances of assessment by a single staff member.
Conclusions
I accept Counsel Assisting’s submissions that such an inquiry is not too remote for my investigation. However, Counsel for Bendigo Health is correct that a causal connection between such issues and Michelle’s death would be difficult to establish to the relevant Briginshaw standard. 47
Although Bendigo Health policies state that it is acceptable for patients with low risk in the LDU to be given observations at three-hour intervals, Michelle’s clinicians consistently directed that she be observed at 60-minute intervals or 15-minute intervals. Insufficient frequency of observation is the most concrete way in which underassessment of her risks might 47 Briginshaw v Briginshaw (1938) 60 C.L.R. 336 esp at 362-363. “The seriousness of an allegation made, the inherent unlikelihood of an occurrence of a given description, or the gravity of the consequences flowing from a particular finding, are considerations which must affect the answer to the question whether the issues had been proved to the reasonable satisfaction of the tribunal. In such matters “reasonable satisfaction” should not be produced by inexact proofs, indefinite testimony, or indirect inferences…”
have been causative of her death, and although this did in the end occur, it was due to issues unrelated to her assessed risk of suicide.
Because of this, I do not find that these apparent systemic issues, even if they existed, played a direct causative role in Michelle’s death.
As such, it is not appropriate for me to determine whether the limited evidence of systemic issues is sufficient to ground findings. However, I do consider that there is sufficient evidence to justify a comment on the matter under section 67(3) of the Act, as these possible systemic issues are relevant to public health and safety.
Staffing levels and observations on 27 August
Submissions at inquest
Counsel Assisting submitted that Bendigo Health’s staffing practices failed to ensure that sufficient clinical staff were present at all times to observe residents in the LDU, and that this lack of observation contributed directly to Michelle being able to take her own life.
Counsel for Bendigo Health submitted that: ‘It would involve a degree of speculation to draw a connection between the absence of two of the LDU nurses between 11.50am and 12.20pm, when they were assisting in the HDU, and [Michelle’s] actions. It is unknown when [Michelle] entered the showers or when she placed the plastic bag over her head, and whether that occurred at some point prior to 11.50am or afterwards. Even if the two nurses had remained on the LDU for the entire period, they would have had other duties to attend to there. As Associate Professor Tune points out, the only way to completely obviate a risk of attempted suicide is to place a patient on continuous observation, but that was not indicated at the time in the case of [Michelle].’
Counsel Assisting, in reply, firmly disagreed, submitting that ‘the reason we do not know exactly when [Michelle] undertook the acts leading up to, and resulting in, taking her own life is precisely because the observation at 12.00pm did not occur’ and that, if these arguments
were accepted, the failure to perform a clinically indicated observation would result in the impossibility of making findings that that very failure contributed to a death.
(a) a clinical decision had been made that, considering Michelle’s risk of suicide, the proportionate frequency of observation was once every 60 minutes; and
(b) this implies that it was a reasonable clinical assessment that her risk of taking her own life was such that, if she were to be reliably observed once an hour, staff would either be able to interrupt an attempt or to observe that Michelle’s condition was deteriorating in such a manner than an attempt was imminent, and intervene.
In conclusion, [I] submit that the proportionate frequency of observation to limit [Michelle’s] reasonably assessed risk of suicide was once an hour, an assessment made by Bendigo Health staff in accordance with, or even exceeding, Bendigo Health procedures. This proportionate frequency of observations did not occur, and the risk of a tragic outcome, to which the frequency was proportionate, did in fact eventuate. I maintain my submission that this constitutes a direct contribution.’ Conclusions
I do not accept the arguments made by Counsel for Bendigo Health on this issue.
As for the time at which Michelle undertook the actions resulting in her death, I note that Michelle was still able to be resuscitated at the time she was found at around 12.20pm. Even if she had placed the bag over her head prior to 12.00pm, if staff had discovered her twenty minutes earlier it is certain that her chances of survival would have been distinctly higher.
As for the likelihood of staff being able to identify Michelle’s heightened risk of suicide and being able to intervene if they had performed an observation at 12.00pm, I accept Counsel Assisting’s submission that the appropriate evidence for this probability is the proportionate frequency of observation determined by Michelle’s clinicians at Bendigo Health in response to her assessed risk of self-harm.
There is no indication that Michelle’s clinicians overestimated her risk of self-harm based on the information available to them.
A finding that the failure to perform this observation contributed directly to Michelle’s death would be strongly adverse to Bendigo Health. As I set out above, the standard of proof for coronial findings of fact is the civil standard of proof on the balance of probabilities, with the Briginshaw gloss or explications. Adverse findings or comments against individuals in their professional capacity, or against institutions, are not to be made with the benefit of hindsight but only on the basis of what was known or should reasonably have been known or done at the time, and only where the evidence supports a finding that they departed materially from the standards of their profession and, in so doing, caused or contributed to the death under investigation.
Based on what was known at the time, Bendigo Health clinicians made the reasonable judgment that observations once an hour would be able to prevent Michelle from taking her own life. All of the factual evidence indicates that this judgment was correct.
Failing to ensure that this observation occurred was a material departure from the standards to be expected of an acute inpatient psychiatric unit, and there is sufficient evidence for me to conclude that it directly contributed to Michelle’s death.
Access to means of suicide
At the time of Michelle’s death, plastic bags were known by Bendigo Health to be an object that could be used to assist in a suicide attempt, and patients were required to be supervised by staff if using a plastic bag.48
Nonetheless, it is clear that a bin within the inpatient setting was lined with a plastic bag. There is no indication that this was a one-time aberration rather than a consistent failure in Bendigo Health’s practices with regard to bins in the Alexander Bayne Centre.
However, while it was determined that the plastic bag which Michelle used to take her own life was taken from that bin, it is not clear whether the plastic bag which Nurse Battersby found on her person on 12 August 2016 came from the same source. It does not appear that, at that time, any actions were undertaken to determine the source of that bag.
Submissions at inquest
(a) in allowing a bin to be lined with a plastic bag, within an inpatient setting, Bendigo Health’s facilities management practices failed to protect their patients, and contributed directly to Michelle being able to take her own life; and
(b) the lack of investigation of the source of the plastic bag on 12 August was a failure to take actions to prevention patients in the Alexander Bayne Centre from having access to means of suicide.
Counsel Assisting acknowledged, however, that as it is unclear whether the source for the plastic bag of 12 August was the same as the source for the plastic bag on 27 August, there is insufficient evidence to ground the adverse finding that this failure to investigate contributed directly to Michelle being able to take her own life.
Counsel for Bendigo Health did not directly oppose the first submission, and concurred with the acknowledgement that the lack of investigation is insufficient to ground an adverse finding.
Additional submissions were made by Counsel Assisting and Counsel for Bendigo Health regarding the degree of likelihood that a proper investigation would have prevented the death, but neither submitted that this degree was sufficient to justify an adverse finding against Bendigo Health.
48 See statement of Sarah Toma dated 6 October 2016, Coronial Brief.
Conclusions
I accept Counsel Assisting’s two submissions set out above, including the acknowledgement regarding the difficulty of causation with respect to the investigation of the source of the plastic bag.
With regard to the matter of whether there is sufficient evidence to ground the adverse finding that the presence of the bag contributed directly to the death, the issues here are much clearer than with respect to the failure to make an observation.
The fact that the bag was present was incontrovertible, as well as the fact that it should not have been. As the inpatient psychiatric unit should have been otherwise free of available means of suicide, it is highly probable that Michelle would not have been able to take her own life if the bag were not present.
There is sufficient evidence to ground the adverse finding that Bendigo Health’s failure to prevent Michelle from having access to this plastic bag on 27 August directly contributed to her death.
Family and external support person contact
The CPU noted that Bendigo Health’s communication with Michelle’s ‘next of kin’ was problematic. In particular, it appears that Michelle’s family were not directly notified after the incidents on 12 August 2016 or 25 August 2016, nor were they notified of Michelle’s transfers into and out of the HDU.
According to Tim Lenten, Director of Nursing for Psychiatric Services, the Bendigo Health’s Patient & Carer Rights, Responsibilities & Information Provision Policy states that patients’ ‘carers, family members, and or nominated persons will be informed of changes relating to a patient’s care and treatment as they occur. Information about rights in relation to these changes are to be explained/provided at this time’.
Specifically, the Bendigo Health Psychiatry Inpatient High Dependency Unit (HDU) Admission and Management Protocol states that ‘support should be offered to family/carers of patients who have been placed in High Dependency Unit and an explanation given, in addition to reassurance’. This protocol does not provide a specific timeframe for this.
Fifty-seven hours passed between Michelle’s final transfer to the HDU and her death, and 24 hours passed between her being transferred back to the LDU and her death. The CPU consider that this is an unacceptable timeframe to wait before discussing with the next of kin.
With regard to family contact following attempted suicide or self-harm incidents, Mr Lenten has advised that Bendigo Health protocols have advised that the person in charge at the time of an incident is responsible for notifying family members, ‘as appropriate’. He has stated that he believes this would involve consultation with the treating team and with the patient.49
There is no record of any consultation occurring with Michelle or with her treating team following her incidents on 12 or 25 August 2016 regarding informing her next of kin.
The CPU advise that, while is it unlikely that notifying Michelle’s next of kin after these incidents would have prevented future suicide attempts, it is nonetheless important. Wellinformed family and support persons play a critical role in the delivery of mental health services.
Submissions at inquest and conclusions
Counsel Assisting submitted that I should accept the CPU’s advice on these issues. Counsel for Bendigo Health did not oppose this submission.
I accept this submission, and find that the CPU’s conclusions were correct.
In general, I commend Bendigo Health for its earnest and mature engagement with this investigation. In particular, statements from A/Prof Tune and Mr Lenten have been particularly helpful. Submissions made at inquest have also assisted me in this respect.
‘PRN’ medication
renewed focus on clinical escalation pathways will serve to prevent patient agitation from being undertreated, such as Michelle’s agitation was, in the future.
Staffing levels
A/Prof Tune has described how Bendigo Health have developed dedicated staffing profiles for the LDU and the HDU to formalise the management of the moving of staffing resources between them for resource-intense events such as seclusion. They have provided education to staff about their responsibilities in each of these areas and identified specific responders from both within the unit and adjacent units, including a security response, in the event of resourceintense events occurring. 50
I am satisfied that these changes will help to prevent future incidents where misallocation of staff causes patients to remain insufficiently supervised.
Risk assessments
Mr Lenten noted that the key document guiding clinical risk assessments is the ‘Psychiatric Clinical Risk Assessment Policy’, which is referenced throughout other relevant guiding documents including the ‘Searches of Patients and Visitors in Psychiatry Inpatient and Residential Protocol’, the ‘Psychiatry Inpatient High Dependency Unit (HDU) Admission and Management Protocol’, the ‘Visual Observation and Engagement in Psychiatry Inpatient and Residential Units Protocol’ and the ‘Restrictive Interventions Protocol’.
He advised that staff compliance with these protocols is monitored in two primary ways: routine annual auditing, and as part of the investigation processes for all clinical incidents. The routine annual auditing is reported to the Mental Health Quality and Risk Committee, which identifies issues with non-compliance and takes action to increase compliance and re-audit until compliance is raised.
He also noted that, as the service has transitioned from paper records to electronic records, they have used this process to set up the electronic risk assessment tool so that all areas of risk must be documented against, and a comment entered for all ‘Medium’ or ‘High’ risk indicators, before the system will allow a clinician to sign off on their documentation.
50 Statement of Associate Professor Philip Tune dated 13 September 2019.
More broadly, he noted that they are in discussion with the Office of the Chief Psychiatrist to further explore the reliability of the clinical risk assessment as an acute and predictive clinical tool.
I am satisfied that that these efforts set out by Mr Lenten show that Bendigo Health has put robust processes in place to prevent non-compliance with risk assessment guidelines in the future, and that they are additionally working to improve the accuracy and utility of the risk assessment process itself.
Access to plastic bags
According to Mr Lenten, in January 2017 Bendigo Health relocated Mental Health inpatient services to the Bendigo Hospital, and entered a partnership with the facilities service provider ‘Spotless’, which became primarily responsible for key functions that would participate in keeping the inpatient units free of plastic, including cleaning, portering and procurement and maintenance of appropriate furnishing.
Mr Lenten set out a list of environmental requirements put in place to ensure that plastic bags, as well as removable plastic covers for pillows, would not be brought into the inpatient unit.
He noted that Bendigo Health worked closely with key Spotless stakeholders to ensure that the inpatient environment would be free of possibly dangerous plastic items. He noted this included monitoring and auditing of changes to align with the requirement and the education of all Spotless staff who will be undertaking these roles.
At inquest, Counsel for Bendigo Health advised of further efforts that have taken place with regard to environmental requirements to make bins used in mental health units immediately and easily distinguishable from those with a liner, as well as storage arrangements to mitigate the risk of the wrong bin being accidentally used.
It is clear that, since the change in how Bendigo Health facilities are managed, the errors leading to Michelle gaining access to a plastic bag while in the inpatient unit will not be exactly repeated. Nonetheless, those errors do demonstrate the importance of vigilance in this area.
Auditing of compliance with protocols
He therefore submitted that I should make a recommendation to Bendigo Health that they initiate such a program, in order to prevent, as far as possible, mental health inpatients from accessing dangerous items while in the care of Bendigo Health.
Searches for prohibited items
Mr Lenten noted guidelines for searching visitors and patients to prevent them from bringing dangerous objects into the inpatient environment. However, this does not address the issue of how steps were not taken to determine the source of the plastic bag which was found with Michelle on 12 August 2016.
At inquest, Counsel Assisting submitted that I should consider making a recommendation that Bendigo Health review and amend the Searches of Patients and Visitors in Psychiatry Inpatient and Residential Units protocol to include that, when a patient is found with a prohibited item under this protocol, all reasonable efforts are made to identify how the patient accessed the item, that steps be taken to prevent future access to such items in similar circumstances, and that the steps taken to prevent future access to the prohibited item be documented in the patient’s medical record.
Counsel for Bendigo Health advised that it is already the expectation that this be done by Bendigo Health staff, but that they were open to amending their policies to formalise that expectation.
Next of kin
He specifically notes that, with Consumer and Carer Consultants employed within their
education team, this training is reviewed, developed and delivered by people with lived experience of having these support roles.
He also advised that follow-up review of incidents includes a review of whether communication with support persons would have been appropriate, and if so whether it occurred. Mechanisms exist for follow-up of individual errors with staff as well as making recommendations to address systemic issues.
Despite the issues with arose in Michelle’s case, I am satisfied that Bendigo Health are taking appropriate measures to ensure that support persons are made aware of important changes in treatment, and are best enabled to perform their roles in support of patients.
(a) the identity of the deceased was Michelle Williams, born 18 June 1970; and
(b) the death occurred on 29 August 2016 at Bendigo Health, 100 Barnard Street, Victoria from 1(a) Hypoxic brain injury in the setting of plastic bag asphyxia; and
(c) the death occurred in the circumstances described above.
Having considered all of the circumstances, I am satisfied that Michelle intentionally took her own life while an inpatient at the Alexander Bayne Centre.
Having considered all of the evidence, I am satisfied that Michelle’s agitation and distress on the morning of 27 August 2016 were such that it would have been reasonable to provide her ‘prn’ medication in addition to her ‘baseline’ prescribed dosage, but that there is insufficient basis to find that the failure of Bendigo Health staff to do so played a causative role in her death.
Having considered all of the evidence, I am satisfied that a formal risk assessment was required following Michelle’s significant deterioration in condition on the morning of 27 August 2016, and that Bendigo Health staff failed to do this.
Having considered all of the evidence, I am satisfied that Bendigo Health failed to ensure that sufficient staff were present to carry out Michelle’s hourly observation at 12.00pm on 27 August 2016, and that this failure directly contributed to her death.
Having considered all of the evidence, I am satisfied that Bendigo Health failed to prevent Michelle from having access to a plastic bag on 27 August while an inpatient at the Alexander Bayne Centre, and that this failure directly contributed to her death.
COMMENTS Pursuant to section 67(3) of the Act, I make the following comments connected with the death.
The circumstances of Michelle’s death revealed evidence of issues where multiple staff at the Alexander Bayne Centre insufficiently considered the role of a history of self-harm and suicide attempts in increasing the current risk of self-harm and suicide, leading to underestimation of current suicide risk. However, these issues do not appear to have had a direct effect on the treatment provided to Michelle.
It is clear that Bendigo Health have made robust efforts to improve their practices with respect to this issue, as well as with respect to the other issues identified in my investigation. I commend the work they have undertaken to protect their patients and prevent deaths such as Michelle’s from occurring in the future, and I am confident that this work will continue.
RECOMMENDATIONS Pursuant to section 72(2) of the Act, I make the following recommendations:
(i) that Bendigo Health formalise the inclusion of plastic bags in their regular ligature audit; and (ii) that Bendigo Health amend their Searches of Patients and Visitors in Psychiatry Inpatient and Residential Units protocol to include that when a patient is found with a prohibited item, all reasonable efforts are made to identify how the patient accessed the item, that steps be taken to prevent future access to such items in similar circumstances, and that such steps be documented in the patient’s medical record.
I convey my sincere condolences to Michelle’s family for their loss.
Pursuant to section 73(1) of the Act, I order that this finding be published on the Coroners Court of Victoria website in accordance with the rules.
I direct that a copy of this finding be provided to the following: Mr Matthew Mulkearns, family spokesperson.
Bendigo Health, c/o K&L Gates.
Office of the Chief Psychiatrist.
Detective Senior Constable Stuart Poulton, Coroner’s Investigator.
Signature: ______________________________________ SIMON McGREGOR
CORONER Date: 18 December 2020 NOTE: Under section 83 of the Coroners Act 2008 ('the Act'), a person with sufficient interest in an investigation may appeal to the Trial Division of the Supreme Court against the findings of a coroner in respect of a death after an investigation. An appeal must be made within 6 months after the day on which the determination is made, unless the Supreme Court grants leave to appeal out of time under section 86 of the Act.
Natasha Stojkoski, a 38-year-old sales assistant at The Good Guys, sustained a workplace injury in October 2018 involving a Work Assisted Vehicle incident affecting her back, shoulders and neck. Subsequently, she reporte…
Harrison William Ladd, 25, died from combined drug toxicity involving clonazepam, dexamphetamine, codeine, diazepam and pholcodine. He had complex mental health and substance use disorder requiring specialist addiction c…
Simon Gaskill, aged 50, was found deceased in sand dunes at Ocean Grove Beach in April 2022 after sleeping rough for months. He had been discharged from hospital in December 2021 following treatment for severe alcohol wi…
Source and disclaimer
This page reproduces or summarises information from publicly available findings published by Australian coroners' courts. Coronial is an independent educational resource and is not affiliated with, endorsed by, or acting on behalf of any coronial court or government body.
Content may be incomplete, reformatted, or summarised. Some material may have been redacted or restricted by court order or privacy requirements. Always refer to the original court publication for the authoritative record.
Copyright in original materials remains with the relevant government jurisdiction. AI-generated summaries are for educational purposes only and must not be treated as legal documents. Report an inaccuracy.