Deceased
Roy Roberts
Demographics
33y, male
Coroner
Decision ofDeputy State Coroner Lee
Date of death
2019-08-28
Finding date
2023-02-03
Cause of death
Prolonged epileptic seizure resulting in gastric content aspiration and cardiac arrest
AI-generated summary
Roy Roberts, a 33-year-old Aboriginal man, died at Long Bay Correctional Complex from a prolonged epileptic seizure resulting in aspiration and cardiac arrest. He had experienced three prior seizure-like episodes in April-May 2019, all undiagnosed. Each time he was taken to hospital, refused further assessment and investigations, and was returned to custody. The coroner found Roberts had underlying primary epilepsy. Clinical opportunities existed to explore reasons for his refusal of assessment and to counsel him more extensively on risks, particularly after the third episode. While clinicians mistakenly attributed episodes to drug use based on found paraphernalia and apparent naloxone response, expert evidence conclusively demonstrated these episodes were epileptic. Key failings included: inadequate consideration of differential diagnoses (particularly epilepsy), insufficient exploration of modifiable factors affecting his refusal of treatment, limited Aboriginal Health Worker access, and patient confidentiality concerns that may have discouraged him from engaging with care. Even with hindsight, it remains uncertain whether different clinical approaches would have changed the outcome.
AI-generated summary — refer to original finding for legal purposes. Report an inaccuracy.
Specialties
Error types
Drugs involved
CORONER’S COURT OF NEW SOUTH WALES Inquest: Inquest into the death of Roy Roberts Hearing dates: 28 & 29 March 2022; 6 to 9 September 2022 Date of Findings: 3 February 2023 Place of Findings: Coroner’s Court of New South Wales, Lidcombe Findings of: Magistrate Derek Lee, Deputy State Coroner Catchwords: CORONIAL LAW – cause and manner of death, diagnosis of primary epilepsy, seizure-like episode, methadone toxicity, opioid substitution therapy, unprescribed drug use, care and treatment, refusal of medical assessment and investigations, Aboriginal Health Worker, inmate patient confidentiality File number: 2019/269131 Representation: Ms J Curtin, Counsel Assisting, instructed by Mr P Armstrong (Crown Solicitor’s Office) Ms G Campagna (Aboriginal Legal Service) for the family of Mr Roberts Mr S Grey for Justice Health & Forensic Mental Health Network and Western New South Wales Local Health District, instructed by Hicksons Lawyers Mr T Hackett for Dr T Lipski, instructed by Avant Law Mr R Reitano for Mr G Lethbridge, instructed by WG McNally & Co Ms V Temelkovska (Department of Communities & Justice) for the Commissioner of Corrective Services New South Wales Mr B Wilson for Dr M Fleri, instructed by Meridian Lawyers
Findings: Roy Roberts died on 28 August 2019 at Long Bay Correctional Complex, Matraville NSW 2036.
The cause of Mr Roberts’ death was a prolonged epileptic seizure resulting in gastric content aspiration and cardiac arrest.
Mr Roberts died of natural causes whilst in lawful custody serving a sentence of imprisonment. The prolonged epileptic seizure on 28 August 2019 represented the fourth occasion since April 2019 that Mr Roberts had suffered a seizure event due to undiagnosed and underlying primary epilepsy.
Following the first three occasions, Mr Roberts declined any further assessment or treatment at hospital. Although the reasons for Mr Roberts’ decision are not known, opportunities existed during subsequent clinical reviews for these reasons to be explored with Mr Roberts with a view to performing further medial assessment.
However, even if these opportunities had been taken it is not known whether Mr Roberts would have been receptive to them, or whether they may have materially affected the eventual tragic outcome.
Recommendations made To the Chief Executive, Justice Health & Forensic Mental Health pursuant to section 82, Network: Coroners Act 2009
(a) the confidentiality requirements that attach to provision of any health information by an inmate to a clinician; and (b) the ability to request a consultation with a clinician in the absence of a Corrective Services New South Wales officer, subject to any safety and security considerations that may attach to such a request.
I recommend that consideration be given to providing ongoing education and training to all nursing and medical clinicians regarding the clinical features and identification of primary epilepsy.
I recommend that a review be conducted to ensure that Aboriginal Health Worker positions in New South Wales correctional centres are sufficiently staffed to provide inmate patients with access to culturally appropriate healthcare.
Non-publication orders: See Annexure A
Table of Contents
Were appropriate steps taken to ensure Mr Roberts’ well-being following the three previous
What was the source of the methadone that led to the concentration found in Mr Roberts’
Introduction 1.1 Roy Roberts was in lawful custody serving a sentence of imprisonment at the time of his death. In April and May 2019, Mr Roberts experienced three episodes where he was seen to be suffering from an apparent seizure. On each occasion Mr Roberts was taken to hospital but he declined any medical treatment once there and was returned to custody.
1.2 On 28 August 2019, Mr Roberts failed to attend an afternoon muster. He was found in his cell a short time later experiencing a fourth seizure-like episode. Emergency services were contacted and resuscitation efforts were initiated. However Mr Roberts could not be revived and was, tragically, pronounced deceased. He was just 33 years old at the time.
2.1 Under the Coroners Act 2009 (the Act) a Coroner has the responsibility to investigate all reportable deaths. This investigation is conducted primarily so that a Coroner can answer questions that are required to be answered pursuant to the Act, namely: the identity of the person who died, when and where they died, and what was the cause and the manner of that person’s death.
2.2 When a person is charged with an alleged criminal offence, or sentenced after being convicted of a criminal offence, they can be detained in lawful custody. By depriving that person of their liberty, the State assumes responsibility for the care of that person. Section 23 of the Act makes an inquest mandatory in cases where a person dies whilst in lawful custody. In such cases the community has an expectation that the death will be properly and independently investigated.
2.3 A coronial investigation and inquest seeks to examine the circumstances surrounding that person’s death in order to ensure, via an independent and transparent inquiry, that the State discharges its responsibility appropriately and adequately. This type of examination typically involves consideration of, where relevant, the conduct of staff from Corrective Services New South Wales (CSNSW) and Justice Health & Forensic Mental Health Network (Justice Health).
2.4 In Mr Roberts’ case, the coronial investigation identified a number of issues which warranted further exploration and consideration at inquest. These issues focused on the care and treatment provided to Mr Roberts, whilst in custody and at hospital, with respect to the three seizure-like episodes that he experienced in April and May 2019.
2.5 In this context it should be recognised at the outset that the operation of the Act, and the coronial process in general, represents an intrusion by the State into what is usually one of the most traumatic events in the lives of family members who have lost a loved one. At such times, it is reasonably expected that families will want to grieve and attempt to cope with their enormous loss in private. That grieving and loss does not diminish significantly over time. Therefore, it should be acknowledged that the coronial process and an inquest by their very nature unfortunately compel a family to re-live distressing memories several years after the trauma experienced as a result of a death, and to do so in a public forum. This is an entirely uncommon, and usually foreign, experience for families who have lost a loved one.
2.6 It should also be recognised that for deaths which result in an inquest being held, the coronial process is often a lengthy one. The impact that such a process has on family members who have many unanswered questions regarding the circumstances in which a loved one has died cannot be overstated.
3.2 Mr Roberts was a proud Bundjalung man who was raised and lived on Gadigal land. Mr Roberts was known as a “mulgul kid”, with “mulgul” meaning mischievous in Bunjum and Wiradjuri. He and his older brother, Lloyd, were raised by their maternal grandparents, Lester and Helen Roberts.
3.3 Mr Roberts had four brothers and six sisters and grew up in the heart of Redfern on The Block. His nickname around the neighbourhood was “the best kid on the block”, because from a young age he was known to look out for everyone, regardless of their age. According to his sisters, everyone from the neighbourhood had a soft spot for Mr Roberts.
3.4 Mr Roberts reportedly suffered from attention deficit hyperactivity disorder and experienced difficulties at school. He later attended a special need to school due to challenges with his literacy.
3.5 According to his family, Mr Roberts had a loud voice and a big personality - he would always be heard before he was seen. Mr Roberts was also described as a happy-go-lucky person with a heart of gold.
Even when he was dealing with his own personal challenges, he would do so with a positive attitude and a smile on his face, and he would never allow such challenges to prevent him from being there for his family.
3.6 Mr Roberts had 17 nieces and nephews and gave an enormous amount of time to them. He could never say no to any of them. No request was too big or small. There is no doubt that he was the best uncle that he could be to all of them, and that they still wish that their “Uncle Roy” was still here.
3.7 There is no doubt that Mr Roberts was a much-loved son, grandson, brother, uncle, nephew and cousin. There is equally no doubt that he is still greatly missed by his family and many friends, and that they are still healing and always will be.
4.2 During his most recent period in custody, Mr Roberts was housed at the Metropolitan Remand and Reception Centre (MRRC) and Bathurst Correctional Centre (Bathurst CC). On 26 August 2019, Mr Roberts was transferred to Long Bay Correctional Complex and housed in the Metropolitan Special Programs Centre (MSPC).
From 2006, Mr Roberts had participated in a methadone opioid substitution therapy (OST) program to address his use of heroin. In 2011, Mr Roberts began attending an OST program at St Vincent’s Hospital. He was prescribed varying amounts of methadone, including a daily dose of 120mg in the months immediately preceding his death.
5.2 Although Mr Roberts himself reported no health concerns, aside from a history of eczema, his medical records indicate that he had a history of osteomyelitis, pulmonary abscess, anaemia, central hypogonadism, septic arthritis and pancreatitis.
5.3 On or around 13 December 2018, Mr Roberts asked to see a Justice Health Drug & Alcohol (D&A) clinician regarding his methadone dosage. He was placed on a waitlist and later seen on 4 February 2019 by a D&A Registered Nurse (RN) Jo-Anne Asimus who noted that Mr Roberts was on a daily methadone dose of 65mg. Mr Roberts reported that he felt “like using” and asked to be prescribed his original dose of methadone of 135mg. An ECG was also performed at or around the time of this consultation.
5.4 On 7 February 2019, RN Asimus discussed Mr Roberts’ methadone dose with Dr John Kehoe, a Justice Health Visiting Medical Officer (VMO). Dr Kehoe did not examine Mr Roberts but had regard to his medical history, and determined that Mr Roberts’ dose of methadone could be increased to 70mg, with the option of increasing his dose by 5mg each fourth or fifth dosing, up to a maximum of 100mg.
Dr Kehoe later reviewed the earlier ECG results and found them to be within normal limits.
6.1 On 19 April 2019, Mr Roberts was captured on CCTV footage to walk towards a table whilst in 8 yard and sit down across from another inmate (the CCTV footage). After providing an item, possibly a cigarette, to another inmate, Mr Roberts’ facial expression changed, with his eyes appearing to roll back in his head, before he collapsed to the ground and suffered what appeared to be a tonic-clonic seizure (the 19 April 2019 episode). A number of other inmates came to Mr Roberts’ aid, turning him on his left side and continuing to assist him.
6.2 CSNSW officers who were made aware of the incident sought assistance from Justice Health staff who attended within five minutes. Mr Roberts was found lying on his left side, with vomit in and around his mouth and on the ground. On assessment, Mr Roberts was found to be responsive to speech and to have a Glasgow Coma Scale (GCS) score of 15, with his pupils equal and reactive. Mr Roberts was given oxygen and an ambulance was called. One of the attending nurses (RN Kirsty Jennings) notice that Mr Roberts was holding a small ball made of foil.
6.3 At around 12:40pm, RN Jennings administered 400micrograms of naloxone to Mr Roberts who appeared to respond by becoming more alert and agitated. His GCS was assessed as 15 before being transferred by ambulance to Bathurst Base Hospital (BBH) for assessment.
6.4 RN Jennings completed some paperwork in relation to the incident in which she noted that her provisional diagnosis was “drug overdose” of an “unknown substance”. In other paperwork relating to the reason for Mr Roberts’ transfer to an external hospital, RN Jennings recorded a provisional diagnosis of “seizure head injury”, with the reason for transfer being “? Seizure? Drug overdose head injury”.
Transfer to hospital 6.5 Mr Roberts arrived at BBH at around 1:16pm. The New South Wales Ambulance (NSWA) records relevantly record that after Mr Roberts fell backwards and hit his head he reportedly “started to fit and started having convulsions”. It was also noted that Mr Roberts denied having a fit or hitting his head and indicated that he did not want to attend hospital. The NSWA records note the assessment by the attending paramedics was of a “generalised tonic-clonic seizure”.
6.6 At triage, queries were raised as to whether Mr Roberts had experienced a tonic-clonic seizure or a drug overdose.
6.7 Dr Thomas Lipski reviewed Mr Roberts in the company of two CSNSW officers. Mr Roberts was found to be fully alert, with clear speech and did not appear to be intoxicated. Mr Roberts’ vital signs were noted to be within normal limits. During the course of the assessment, Dr Lipski found a tablet in Mr Roberts’ left sock and half a tablet in his trouser pocket. Dr Lipski showed the tablets to one of the CSNSW officers, who showed no interest in them, and then disposed of them. An electrocardiogram (ECG) was performed which showed nil sinus rhythm and no acute changes. Follow in his assessment, Dr Lipski recorded an entry in Mr Roberts’ progress notes which read, “query oiverdose [sic] ?”.
1902039179 6.8 At around 1:50pm, Mr Roberts indicated that he did not wish to remain at hospital and discharged himself against medical advice. He was returned to Bathurst CC where he was reviewed by RN Amanda Franklin at around 2:45pm. Ms Roberts was placed in a camera cell for continuous monitoring overnight and arrangements were made for him to be reviewed the following day.
Subsequent reviews during April and May 2019 6.9 Following review by a Justice Health RN on the morning of 20 April 2018, Mr Roberts was cleared to return to his usual cell. Mr Roberts was reviewed again on 23 April 2019, when it was noted that he had experienced a “recent possible overdose/intoxication [event]” and that he denied “any accidents or use of illicit injectable substances”. Arrangements were made for Mr Roberts to be seen by a D&A VMO on 3 May 2019.
6.10 In the interim, Mr Roberts was reviewed again by a Justice Health RN on 28 April 2018. The notes from that review record that Mr Roberts denied overdosing on 19 April 2018. Mr Roberts’
observations were found to be normal and it was noted that no further review was required unless Mr Roberts experienced “an additional incident”.
6.11 Dr Kehoe reviewed Mr Roberts on 3 May 2019 via AVL, with the assistance of RN Asimus. Dr Kehoe reviewed Mr Roberts’ ECG results from February 2019 and considered them to be normal. Dr Kehoe recorded the following in Mr Roberts’ progress notes: “? Seziure ? OD on 19/4/19”.
6.12 Mr Roberts asked for his methadone dose to be increased to its previous level. Dr Kehoe prescribed a further incremental increase of the dosage by 5mg every 4 to 5 days up to a maximum of 120mg.
Dr Kehoe suggested that a repeat ECG be performed on 15 May 2019 given that by that date, Mr Roberts’ dosage would be close to 120mg.
7.1 At around 2:28pm on 5 May 2019, CSNSW officers found Mr Roberts experiencing a seizure-like episode on the ground of the prison yard (the 5 May 2019 episode). A call for assistance was made to Justice Health and RN Franklin responded. Mr Roberts was found to have rigidity in his muscles, to be frothing at the mouth and to be verbally unresponsive, although no actual seizing or fitting was observed by RN Franklin.
7.2 Initially at 2:43pm, Mr Roberts’ GCS score was noted to be 3 but 10 minutes later it was found to be
Seizure” and “? Drug use” in paperwork relating to Mr Roberts’ subsequent transfer to hospital, and that he had been found with “white tablet ½ one found on him”.
Transfer to hospital 7.3 Mr Roberts was taken by ambulance to BBH at around 3:15pm. The NSWA records indicate that Mr Roberts was found lying prone in the yard and that Justice Health had stated that Mr Roberts had a similar episode three weeks earlier in relation to suspected drug use. At triage, it was noted that Mr Roberts was “post ictil post seizure” and reference was made to a syringe and half a tablet being found next to him.
7.4 Mr Roberts was again seen by Dr Lipski on this occasion. However, Mr Roberts refused to be offloaded from the ambulance trolley or be brought into the emergency department for assessment.
Dr Lipski informed Mr Roberts of the risks of not undertaking medical investigations and treatment, namely recurrent seizures, possible death from a medical complication, and having an unrecognised, undiagnosed and untreated serious medical. Despite this, Mr Roberts remained adamant that he did not wish to be assessed and signed a form at around 3:40pm confirming that he wished to be discharged against medical advice.
Reviews upon return to custody 7.5 Mr Roberts was returned to Bathurst CC where he was reviewed by RN Franklin. Arrangements were made for Mr Roberts to be placed in a camera cell for 24 hours for observation. It was noted that Mr Roberts had experienced seizures which were possibly drug-related and that he required medical review.
7.6 RN Asimus reviewed Mr Roberts the following morning on 6 May 2019. It was noted that the two seizure-like episodes experienced by Mr Roberts “may well be attributed to illicit benzodiazepine use – possibly rivitrol”. Mr Roberts was informed of the risks and consequences for his health in relation to the concurrent use of opiates and benzodiazepines.
8.1 At around 1:30pm on 30 May 2019, Mr Roberts collapsed in the yard and suffered another seizurelike episode (the 30 May 2019 episode). CSNSW and Justice Health staff again responded to the incident. Mr Roberts was found lying on the ground, with a reduced level of consciousness, unresponsive to verbal stimuli, appearing to be frothing at the mouth and with his eyes rolling back in his head. His GCS score was noted to be 3, his airway was suctioned and oxygen was administered.
8.2 Mr Roberts’ condition improved marginally so that by 1:45pm his GCS score was noted to be 6.
Nursing staff administered 400 mcg of naloxone to Mr Roberts. After approximately five minutes, Mr Roberts’ level of consciousness improved, he became more alert and sat up, appearing to move normally. NSWA paramedics arrived on scene at 1:55pm and by this time Mr Roberts GCS score was noted to be 15.
Transfer to hospital 8.3 Mr Roberts was taken by ambulance to BBH, arriving at 2:24pm. Entries in Mr Roberts’ progress notes make reference to Mr Roberts having experienced “syncope/seizure” as a result of a “suspected narcotic overdose”. It was also noted that Mr Roberts had two previous presentations to the emergency department “for sam[e]”.
8.4 Mr Roberts was reviewed by Dr Arnav Gupta, a junior medical officer, and Dr Oksana Williams, a VMO.
However, Mr Roberts declined any further investigations or treatment. Dr Gupta and Dr Williams explained the associated risks in doing so, but Mr Roberts maintained his stance. At around 3:25pm, Mr Roberts signed paperwork confirming his wish to be discharged against medical advice.
Subsequent review and follow-up upon return to custody 8.5 On his return to Bathurst CC, Mr Roberts was brought to a Justice Health clinic by CSNSW officers and spoke to RN Sandra Abbe. It appears that Mr Roberts maintained his reluctance to engage with medical professionals, although it is not clear whether Mr Roberts was waitlisted for further review by a general practitioner (GP).
8.6 On 14 June 2019, Ms Roberts was scheduled to attend the GP clinic at Bathurst CC so that he could be assessed by Dr Mari Fleri regarding the need for any treatment “for recent ? seizure”. Mr Roberts refused to attend and Dr Fleri noted that he should be classified for group cell placement, meaning that he was not to be housed in a cell on his own.
8.7 On 23 July 2019, Mr Roberts was transferred from Bathurst CC to the MRRC. On 23 August 2019, Dr Fleri saw Mr Roberts in the GP clinic to follow up on his hospital attendances. Dr Fleri noted that Mr Roberts had suffered a loss of consciousness on 19 April 2019 and 30 May 2019, but was unaware of the 5 May 2019 episode. Dr Fleri considered that both events were “drug-related”, and that no further action was required.
9.1 Between around 8:30am and 9:00am on 28 August 2019, Mr Roberts spoke to his cellmate and nothing amiss was noted. Sometime at around 9:00am, Mr Roberts received his daily dose of methadone, administered by RN Jocelyn Cadornigara.
9.2 By around 11:00am, Mr Roberts was at the basketball court connected to Wing 17 of the Metropolitan Programs Centre Area 3. A muster was conducted by CSNSW First Class Correctional Officer Grant Lethbridge, who marked Mr Roberts off as being present. Nothing unusual was noted at the time.
9.3 At around 1:30pm, Mr Lethbridge was conducting a check of inmate account balances and marked Mr Roberts’ name off the list. Mr Lethbridge conducted a second muster on the basketball court at around 2:20pm. Mr Roberts was unaccounted for and at around 2:25pm, Mr Lethbridge went to Mr Roberts’ cell to check on him.
9.4 The cell door was found to be slightly ajar, and Mr Roberts was found lying on the ground, partially or completely on his left hand side, apparently having a seizure and with his eyes wide open (the 28 August 2019 episode). Mr Roberts was also noted to be unresponsive and frothing at the mouth. Mr Lethbridge immediately called for an emergency medical response. Justice Health nurses arrived at the cell within several minutes and commenced treating Mr Roberts. An ambulance was called at 2:30pm.
9.5 Following the arrival of further Justice Health nurses, Mr Roberts was placed on his side in the recovery position in and was intubated in order to drain the fluid that was pooling in his mouth. His airway was cleared and oxygen therapy was administered. Mr Roberts was moved from his cell to the corridor, to allow for more room for treatment, and two doses of naloxone were administered with no effect. Cardiopulmonary resuscitation was commenced at around 2:40pm and suction equipment was used to keep Mr Roberts’ airway clear.
9.6 Paramedics arrived at the scene at around 2:50pm and took over the resuscitation efforts. A defibrillator was used but it was found that Mr Roberts did not have a shockable rhythm. Further oxygen therapy was provided and Mr Roberts was also administered adrenaline. Resuscitation efforts continued until 3:09pm when Mr Roberts was pronounced deceased.
9.7 After being notified of Mr Roberts’ death, New South Wales Police Force (NSWPF) officers arrived at the scene at around 4:16pm. Mr Roberts’ cell was searched at around 7:00pm, and no evidence of drug use was found.
(a) evidence of widespread gastric content aspiration into the lungs, together with changes in the lungs that are typically found in intravenous drug users;
(b) multiple calcified nodules contained in the pancreas, which was extremely fibrous, in keeping with chronic pancreatitis;
(c) small foci of chronic inflammation, many associated with blood vessels, in some sections of the brain; and
(d) routine toxicology identified a concentration of methadone (1.8 mg/L) within the reported lethal range (the postmortem methadone concentration), together with a negligible concentration of quetiapine (<0.05 mg/L) (the postmortem quetiapine concentration).
10.2 In the autopsy report, Dr Burger relevantly noted the following matters:
(a) Whilst it is typically quite difficult to interpret methadone levels obtained from postmortem blood, due to the wide and overlapping concentration ranges, the concentration found in Mr Roberts’ case “is still much higher than the toxic range quoted in the literature”.
(b) The foci of inflammation noted in the brain could “possibly be a consequence of previous intravenous drug use” and that this change was not present in the medial temporal structures, which are typically associated with epilepsy.
(c) Accordingly, there was “no indication that the inflammatory changes were responsible for the convulsions suffered [by Mr Roberts] in the last few months” of his life.
(d) Mr Roberts’ terminal convulsions, which were observed when he was found in his cell, “could be explained by terminal anoxia in the brain secondary to gastric content aspiration, and do not necessarily imply a separate neurological disease process at the time of death”.
10.3 Ultimately, Dr Burger opined that the cause of Mr Roberts’ death was gastric content aspiration due to methadone toxicity.
11.1 Prior to the commencement of the inquest a list of issues was circulated amongst the sufficiently interested parties, identifying the scope of the inquest and the issues to be considered. That list identified the following issues for consideration: (1) What was the likely cause (or causes) of the seizure-like event that led to Mr Roberts’ death?
(2) What was the likely cause (or causes) of the three previous seizure-like events Mr Roberts experienced whilst in Bathurst Correctional centre?
(3) Whether appropriate steps were taken to ensure Mr Roberts’ well-being following the seizurelike events suffered by him in Bathurst Correctional Centre?
(4) Whether Mr Roberts’ participation in the methadone Opioid Substitution Treatment program was properly managed, including having regard to his onset of recurrent unprovoked seizurelike events whilst in Bathurst Correctional Centre?
(5) What was the likely source of the methadone that led to Mr Roberts’ concentration of methadone in his blood at the time of death?
(6) Was the response to Mr Roberts being found in his cell suffering from an apparent seizure adequate, and was CPR commenced in a timely manner and consistent with relevant policies and procedures?
11.2 Each of the above issues is discussed in detail below, and it will be convenient to consider some of the issues together and in chronological order.
11.3 In order to assist with consideration of some of the above issues, independent opinions were sought from the following experts:
(a) Associate Professor Mark Adams, consultant cardiologist;
(b) Professor Mark Cook; consultant neurologist and epileptologist; and
(c) Associate Professor Darren Roberts; consultant clinical toxicologist and pharmacologist.
12.2 Before going on to consider each of the discrete issues which the inquest considered, it would be useful to set out the evidence upon which the eventual conclusions in relation to each issue has been reached. The relevant evidence can be divided into the following categories:
(a) Physical evidence of unprescribed drug use;
(b) Mr Roberts’ apparent response to the administration of naloxone;
(c) Observations of Mr Roberts from the CCTV footage;
(d) The significance of the postmortem methadone concentration;
(e) Clinical features observed in relation to the three previous episodes; and
(f) The likelihood of a cardiac or neurological cause of the three previous episodes and the 28 August 2019 episode.
The 19 April 2019 episode 13.1 Whilst attending to Mr Roberts on 19 April 2019, RN Jennings noticed that he was holding a small ball made of foil. RN Jennings gave evidence that she did not take a closer look at the item, did not know what happened to it, and agreed that it could have been anything.
13.2 The NSWA electronic records relevantly note the following: Pt found with a home-made bong that is used to smoke nicorret [sic], staff state that a used needle was also found in his cell.
13.3 The BBH triage notes similarly record that a “[n]icorette lozenger made into [a] bong” was found in Mr Roberts’ hand, and that a “syringe” was found in his cell. Notwithstanding, RN Jennings gave evidence that she did not find anything similar to either item around Mr Roberts at the time.
13.4 When examining Mr Roberts, Dr Lipski found a white tablet in Mr Roberts’ left sock, and half tablet in his trouser pocket. Dr Lipski could not identity either item and later discarded both items (after showing the tablet to one of the CSNSW officers who accompanied Mr Roberts to hospital).
The 5 May 2019 episode 13.5 RN Jennings gave evidence that on 5 May 2019, CSNSW officers reported finding a white tablet on Mr Roberts’ person. RN Jennings accordingly made the following entry in Mr Roberts’ progress notes: …white tablet ½ one found on him. CSNSW state same tablet found on him last time he had seizure 19/4/19.
Expert evidence 13.6 Associate Professor Roberts explained that a significant amount of sedative medication would have been required to account for the three previous episodes and the clinical features observed in Mr Roberts at the time. Associate Professor Roberts explained:
We’re not talking about one or two tablets of a prescribed medicine or medicines available on prescription, it would have to be an overdose.
13.7 Further, Associate Professor Roberts considered that there are two matters relevant to the possibility that Mr Roberts may have used a syringe on 19 April 2019 to inject a substance such as methadone, or another opiate:
(a) Mr Roberts could not have used a plastic syringe on its own to inject a substance, as both a syringe and needle would be required.
(b) If Mr Roberts had used a syringe to inject an opiate such as methadone or heroin, the effects of such drugs would have been evident within seconds. Based on the rate that Mr Roberts was observed to have been walking on the CCTV footage, this means that he would have had to inject such a drug within 10 metres of where he eventually sat down. Further, if Mr Roberts had so injected the drug, signs of intoxication would have been observed on the CCTV footage prior to the seizure-like episode 13.8 The small ball made of tin foil, the home-made bong, the syringe and the whole and partial tablets were never examined and there is no reliable evidence as to their origin or use. Whilst it may be accepted that items on their own raise suspicions regarding the self-administration of unprescribed drugs by Mr Roberts, there is no evidence to establish that any of these items had any bearing on the nature of the 28 August 2019 episode or the three previous episodes.
13.9 Further, what is captured on the CCTV footage is inconsistent with the self-administration of any unprescribed opiate by Mr Roberts. The CCTV footage does not show any signs that Mr Roberts was experiencing the intoxicating or sedating effects of any opiate.
However, additional consideration should be given to the following matters regarding this apparent response:
(a) Naloxone was administered to Mr Roberts on 19 April 2019 and 30 May 2019, but not on 5 May
(b) Associate Professor Roberts explained that naloxone has a short half-life compared to methadone, meaning that the sedating effects of methadone would be apparent within up to two hours following the administration of naloxone. However, on 19 April 2019 and 30 May 2019, Mr Roberts showed no signs of sedation either at BBH or upon his return to Bathurst CC after the short-acting effects of naloxone wore off.
14.2 In evidence, Professor Cook was asked whether he agreed that the rate at which Mr Roberts apparently recovered following naloxone administration supported the possibility that methadone toxicity caused the three previous episodes. Professor Cook explained: I don't think it helps you distinguish because typically people will recover a few minutes after a seizure though most usually they are in a confused and sometimes agitated or combative state and I think that fits the description of what we've seen here very well so having been administered and then getting better a few minutes later doesn't help you make the distinction at all.
14.3 The opinions expressed by Associate Professor Roberts and Professor Cook establish that the administration of naloxone to Mr Roberts on 19 April 2019 and 30 May 2019 had no meaningful effect.
Given the similar features of the three previous episodes and the expert evidence (discussed further below) regarding the likelihood of an underlying health condition which caused each episode as well as the 28 August 2019 episode, it is most likely that these episodes were not the result of opioid overdose.
(a) There was no evidence of any incapacitation or intoxication on the part of Mr Roberts at the time given that he was able to approach a table with a normal gait, manoeuvre around other people in a coordinated manner, and sit down in a “fairly small chair”;
(b) Mr Roberts appeared to display fine motor activity with appropriate trajectories and showed normal coordination with his hands; and
(c) Mr Roberts appeared to have normal interactions with other people.
15.2 From these observations Associate Professor Roberts could see no indicators that Mr Roberts demonstrated features of opiate intoxication such as displaying a slow and steady gait, inability to maintain tone, or adverse impact on fine motor abilities.
15.3 The opinions expressed by Associate Professor Roberts therefore establish that the 19 April 2019 episode was not caused by an opioid overdose.
(a) Mr Roberts had participated in methadone OST from around 2006 and had been prescribed daily doses ranging from 40mg to 135mg;
(b) Following a period of slow dose uptitration, instituted by Dr Kehoe from February 2019, to ensure tolerance to the opioid adverse effects of methadone, Mr Roberts had stabilised on a daily dose of 120mg;
(c) Due to overlapping ranges reported in the literature as to what may represent a therapeutic, toxic or lethal concentration, there is not always a direct correlation between a concentration and its likely clinical effects;
(d) Concentrations from antemortem blood samples may not be readily comparable to postmortem blood samples;
(e) There is extremely wide variability in the reported relationship between methadone concentrations and toxicity;
(f) Whilst a methadone concentration of 1.8mg/L was a “very high concentration” and would normally be considered supratherapeutic, patients develop tolerance to the opioid effects of methadone with long term use; in other words, a concentration of 1.8mg/L may not necessarily be a lethal concentration in a tolerant individual;
(g) It is possible that the concentration of 1.8mg/L observed in Mr Roberts could have resulted from his prescribed daily 120mg dose about six hours before his death.
16.2 Having regard to the above, it can be accepted that Mr Roberts most likely developed tolerance to the effects of methadone by virtue of his participation in a methadone OST program during the 12 years preceding his death. Accordingly, whilst the postmortem methadone concentration may be regarded as very high as a bare measurement, it is not possible to draw a direct correlation between such a measurement and its likely toxic effects. Importantly, the opinion expressed by Associate Professor Roberts establishes that Mr Roberts’ prescribed daily dose of methadone could have entirely accounted for the post-mortem methadone concentration.
16.3 In addition, Professor Cook explained that seizures provoked by methadone use are rare and that the postmortem methadone concentration did not alter his opinion that a prolonged epileptic seizure preceded Mr Roberts’ death. Associate Professor Adams similarly noted that methadone is not known to cause epileptic seizures.
(a) There were no clinical markers of opioid toxicity, such as miosis (small-sized pupils) and decreased respiratory rate, present;
(b) Following each episode, Mr Roberts’ observations were largely normal, including a sustained normal level of consciousness; and
(c) There was no evidence of a pattern of respiratory rate (“low, lower, higher, lower”) consistent with the administration of naloxone in the context of methadone toxicity.
17.2 It can be seen then that during each of the three previous episodes, Mr Roberts did not display clinical features consistent with opioid toxicity.
18.2 Having formed the view that none of the episodes were drug-related, Associate Professor Roberts considered that Mr Roberts was likely suffering from a primary seizure disorder or a cardiac dysrhythmia. The question that follows is whether there is sufficient evidence to reach a conclusion as to the likelihood of a cardiac or neurological cause for each of the episodes.
18.3 Associate Professor Adams also considered that the cause of Mr Roberts’ death was most likely related to either a seizure disorder or a cardiac arrhythmia, and ultimately indicated that he favoured a cardiac cause of death. Associate Professor Adams considered that the episodes experienced by Mr Roberts were the result of recurrent cardiac arrythmias, likely due to the development of a condition known as torsade de pointes. This is a type of ventricular tachycardia known to be triggered by methadone causing a dose-dependent increase in the QT interval.
Associate Professor Adams acknowledged that the ECG conducted at Bathurst Base Hospital on 19 April 2019 did not show significant QT prolongation. However, he explained that torsade de pointes is relatively common amongst methadone users and can cause syncope if it is self-limiting or can be fatal where it persists.
18.4 Associate Professor Adams also noted that:
(a) cardiac arrhythmias can trigger epileptic-like seizures due to cerebral hypoxia and can mimic a generalised epileptic seizure; and
(b) it is not uncommon for individuals who die suddenly from a known cardiac cause to experience multiple syncopal episodes prior to a fatal arrhythmia developing.
18.5 Associate Professor Adams noted that there were no clinical features that would positively support a conclusion that a cardiac arrhythmia was the underlying cause of the episodes. Notwithstanding, Associate Professor Adams remained of the view that Mr Roberts’ death was more likely to be cardiac in nature but did not discount the possibility that the death may be related to an underlying seizure disorder. In this regard, Associate Professor Adams indicated that he was content to defer to Professor Cook regarding whether the clinical features of each of episodes were sufficient to allow for a diagnosis of epilepsy.
18.6 As to this matter, Professor Cook considered the following clinical features to be relevant to his opinion that the most likely cause of the episodes was primary epilepsy:
(a) The 19 April 2019 episode occurred from wakefulness, and Mr Roberts was found to be frothing at the mouth with no change in pupillary size or respiration noted;
(b) As to the 5 May 2019 episode, Mr Roberts was found to be frothing at the mouth and to exhibit prolonged convulsive activity and generalised rigidity, with confusion and agitation following the episode - all are considered to be typical features of an epileptic seizure;
(c) The same features were present in relation to the 30 May 2019 episode as previously, with Mr Roberts complaining of nausea prior to the event. Professor Cook explained that this is “a very common description of the lead into a convulsive episode and it’s a valuable clue when interrogating someone who has had a convulsive episode because it suggests that the focal seizure is the cause of it”;
(d) Mr Roberts’ eyes were observed to be open during the 28 August 2019 episode and he was producing copious amounts of fluid from his mouth. Professor Cook considered these clinical features to distinguish the seizure-like episode as epileptic rather than cardiac in origin.
18.7 Further, Professor Cook considered that each episode represented a repeating pattern, with similar clinical features present, and with the likelihood of a diagnosis of primary epilepsy increasing with each subsequent event. Professor Cook explained: So having had only a single event you couldn’t be sure but the recurrence of the events and the additional information makes the diagnosis very strong.
18.8 In addition, Professor Cook emphasized the importance of the CCTV footage which he described as being “absolutely central” to his opinion. For context, Professor Cook has reviewed and reported on over 5,000 video recordings of patients, as part of his routine diagnostic work, to determine whether a patient has experienced a seizure, and the nature of any such seizure.
18.9 Professor Cook considered that Mr Roberts experienced a “minor episode” 28 seconds into the CCTV footage. He explained: The event which starts 28 seconds into the video provided where Mr Roberts can be seen to turn his head and raise his right arm slightly is unquestionably a minor seizure, and probably a focal event originating in the left hemisphere of the brain, where the electrical activity there will cause the head to turn to the opposite side. This brief burst of activity then subsides, and Mr Roberts appears to return to a normal position, but it isn’t clear that he is completely responsive. Then another event begins which is identical in onset, with the head turned and the right arm raised, but this episode progresses to a generalised convulsion, confirming that the minor event moments prior earlier was [a] very brief episode of seizure activity (given the identical movements that are observed, allowing confirmation of the association between the two events). Seizures often occur in clusters, and it is frequently the case that flurry of minor seizures proceeds a major convulsion.
18.10 Ultimately, Professor Cook considered that all four episodes were manifestations of a previously undiagnosed primary seizure disorder, were epileptic in origin and could properly be described as seizures. Professor Cook opined that the cause of Mr Roberts’ death was a prolonged epileptic seizure, resulting in aspiration of vomitous and cardiac arrest. Professor Cook considered that a cardiac cause of death relating to an arrhythmia induced by methadone was possible, but less likely.
18.11 When asked whether he might be more inclined, as a neurologist and epileptologist, to consider that the cause of each episode was neurological, Professor Cook explained: …Over the last five years or so that we have monitored, you know, many thousands of patients in this sort of situation and we actually do cardiac monitoring on all of them and I would see, maybe we would monitor maybe two or 300 patients a month and I would see a cardiac event responsible for seizures perhaps once every two months so I would have a better insight in that we monitor people with blackouts of all causes and capture cardiac and EEG information so I’m probably in a slightly better position than most to make the correlation.
19.1 The following matters are relevant to consideration of this issue:
(a) Whilst the drug paraphernalia items purportedly found on and around Mr Roberts raise the possibility of unprescribed drug use, there is no evidence as to the nature or use of any such drug. Accordingly, there is no reliable evidence that unprescribed drug use was the sole or dominant cause of any of the three previous episodes.
(b) The CCTV footage of Mr Robert’s behaviour and interaction with other persons on 19 April 2019 is inconsistent with the sedative effects of an opioid drug consumed prior to the events depicted in the footage.
(c) Although clinicians considered at the time that Mr Roberts showed a positive response to the administration of naloxone during two of the episodes, the consistent rate of recovery across all three episodes and the absence of any sign of sedation after the effects of the naloxone ceased demonstrate that Mr Roberts had no meaningful response to naloxone administration.
(d) Mr Roberts did not display clinical features across each of the three episodes that were consistent with opioid toxicity.
(e) Rather, the clinical features displayed by Mr Roberts were consistent with a primary diagnosis of epilepsy.
(f) The three previous episodes were temporally connected, with each caused by the same underlying health condition and with each successive episode confirming the likelihood of this diagnosis.
19.2 Conclusion: Having regard to each of the matters set out above the most likely cause of the three previous episodes was underlying and undiagnosed epilepsy.
20.1 Consideration of this issue also necessarily involves reaching conclusions as to the manner and cause of Mr Roberts’ death. In addition to the matters already discussed above, there are several matters particular to the 28 August 2019 episode that are relevant to consideration of this issue.
20.2 First, Mr Roberts likely had his daily dose of methadone prior to 9:00am on 28 August 2019. He was later seen by two CSNSW officers at around 8:30am to 9:00am and 11:00am to be behaving normally, with no signs of sedation. It is most likely that Mr Roberts was seen on a third occasion at around 1:30pm when Mr Lethbridge marked his name off a list. Although Mr Lethbridge had no specific recollection of seeing Mr Roberts on this third occasion, if Mr Roberts had displayed any signs of sedation, it is most likely that Mr Lethbridge would have noted it.
20.3 Associate Professor Roberts explained that the time course of methadone absorption is relatively low, peaking between two and five hours following administration. Therefore, if Mr Roberts had taken an additional dose of methadone on 28 August 2019, he would have displayed progressive signs of toxicity prior to his seizure-like episode. In the absence of any such signs observed by either of the two CSNSW officers, the evidence does not support a conclusion that the 28 August 2019 episode was due to methadone toxicity.
20.4 Second, Associate Professor Roberts acknowledged the possibility that Mr Roberts’ death was due to methadone overdose administered intravenously. Such administration would be associated with a rapid decrease in a person’s level of consciousness, coma, respiratory failure, and death. However, due to Mr Roberts’ existing methadone dose and his tolerance to the drug, he would have required a reasonably high additional dose to be at risk of life-threatening methadone toxicity. However, the postmortem methadone concentration did not likely represent a lethal concentration for a tolerant individual like Mr Roberts and could be accounted for by Mr Roberts’ prescribed daily dose.
20.5 Third, given the expert evidence as to the three previous episodes being most likely caused by an underlying and undiagnosed health condition, it is incongruous and most unlikely that Mr Roberts experienced an isolated incident of methadone toxicity on 28 August 2019 that was separate and distinct.
20.6 Fourth, Mr Roberts’ cell was inspected by Mr Lethbridge and searched by NSWPF officers. No evidence of suggested drug use was found during the inspection and search. In addition, attending paramedics did not observe any drug paraphernalia around Mr Roberts or any obvious signs of drug use on his body.
20.7 Fifth, whilst Mr Roberts had puncture marks on his inner right arm and on both hands, it appears that these were caused during the resuscitation attempts made by Justice Health and NSWA personnel. Two of the NSWA paramedics who attended on 28 August 2019 noted that multiple attempts were made by paramedics and Justice Health staff to cannulate Mr Roberts (due to an initial inability to find an appropriate vein). One of the paramedics observed some marks on Mr Roberts’ arms but did not consider the marks to be a result of intravenous drug use. There is therefore no evidence of any puncture mark or injury on Mr Roberts as a result of intravenous drug use proximate to the time of his death.
20.8 Sixth, as to the suggestion that the 28 August 2019 episode was caused by the toxic effects of a drug other than an opioid, it should be noted that the postmortem concentration of quetiapine was less than 0.05 mg/L. This concentration was too low to be quantified. Accordingly, Associate Professor Roberts considered that the concentration was so low as to be irrelevant to the 28 August 2019 episode. Professor Cook similarly considered that this trace amount of quetiapine was unlikely to be relevant to the cause of Mr Roberts’ death.
Submissions on behalf of the interested parties 20.9 It was variously submitted on behalf of Justice Health and Western NSW Local Health District, Dr Fleri and Dr Lipski that, “the complexities in determining [the] cause of [Mr Roberts’] death are multifactorial”, the manner and cause of Mr Roberts’ death is “controversial and complicated”, and that “the cause of [Mr Roberts’] death remains uncertain”. In essence, the submissions urge careful consideration of the opinions expressed in particular by Professor Cook and Associate Professor Adams, and that the evidence may not allow for any conclusion to be reached as to the cause and manner of Mr Roberts’ death.
20.10 It is acknowledged that even with the benefit of hindsight, and the assistance of three eminent experts who are specialists in their respective fields of medicine, determination of the cause and manner of Mr Roberts’ death is not without its complexities. So much is clear from the discordance regarding some aspects of the opinions expressed by the experts, already described above.
20.11 To the extent that the submissions advanced on behalf of the above parties suggest that the cause of Mr Roberts’ death was related to opioid toxicity, these submissions ignore:
(a) the concordance amongst the three experts as to the likelihood of an underlying and undiagnosed health condition which caused all four episodes, and the unlikelihood of an isolated and unrelated cause of the 28 August 2019 episode; and
(b) the proportionate correlation between occurrence and likelihood, meaning that the recurrence of each episode increases the likelihood of an underlying and undiagnosed health condition.
20.12 It has already been acknowledged that Associate Professor Adams leaves open possibility that the cause of Mr Roberts’ death was an arrhythmic event of cardiac origin. It is unsurprising that if Mr Roberts had experienced such an event that it could not be demonstrated at autopsy. In other words, a cardiac arrhythmia could only have been diagnosed in life. Even so, Associate Professor Adams did not identify any clinical features regarding the three previous episodes consistent with a diagnosis of cardiac arrhythmia, noting that the ECG performed on 19 April 2019 showed no convincing evidence of QT prolongation. In this regard, Associate Professor Adams was appropriately content to defer to Professor Cook regarding any identifiable clinical features consistent with a diagnosis of epilepsy. These features have already been discussed above.
20.13 It was submitted on behalf of Dr Fleri that:
(a) “…serious allegations…have recently been levelled against Dr Fleri” and that “Dr Fleri’s alleged wrongdoing” is relevant to determination of the cause and manner of Mr Roberts’ death;
(b) accordingly, “the Briginshaw standard is now of some relevance”; and
(c) “were the court required to reach such findings on the Briginshaw standard…the evidence would not satisfy this more onerous standard”.
20.14 First, to the extent that these submissions suggest that the function and purpose of an inquest is to “level” serious allegations against an individual, or allege wrongdoing on the part of an individual, is to entirely misunderstand and misrepresent the nature of coronial proceedings and the inquest process. Allegations of wrongdoing, negligence or criminal conduct, and apportionment of blame are matters which are wholly incongruous with the nature of the coronial jurisdiction. It clearly follows then that determination of the cause and manner of Mr Roberts’ death is to be made on the balance of probabilities and not on the Briginshaw standard.
20.15 Second, for abundant caution, it should be made clear that Dr Fleri’s interaction with Mr Roberts on 23 August 2019 (discussed further below) had no bearing on the cause and manner of Mr Roberts’ death.
20.16 Third, it is further submitted on behalf of Dr Fleri that if the Briginshaw standard does not apply to determination of the cause and manner of Mr Roberts’ death then “it is open to the Court to find that the available evidence goes close to and perhaps just satisfies such a standard”. To the extent that this submission acknowledges that is open on the available evidence to make a finding as to the cause and manner of Mr Roberts’ death on the balance of probabilities, it is indeed correct.
20.17 It was submitted on behalf of Justice Health that any finding as to the cause and manner of Mr Roberts’ death “turns heavily on the opinion of one expert, Prof Cook, and more specifically, on his interpretation of a particular aspect of the 19 April CCTV footage”, namely what is described as the “minor event” at 28 seconds into the footage, “and then making a further connection between that event and the others in May with the events immediately preceding Mr Roberts’ death”.
20.18 First, whilst Professor Cook described this aspect of the CCTV footage to be “absolutely central” to his ultimate opinion, it is clear that Professor Cook also had regard to a number of other matters set out at [18.6] and [18.7] above. It is equally clear that Professor Cook has considerable and particular experience in reviewing video recordings of persons of the nature of the CCTV footage and determining their utility in arriving at a potential diagnosis of epilepsy. Accordingly, considerable weight should be placed upon Professor Cook’s particular expertise.
20.19 Second, it is evident that prior to viewing the CCTV footage, Professor Cook had already formed the view that it was highly likely that Mr Roberts suffered from undiagnosed epilepsy, which was causative of each of the four episodes. Professor Cook’s viewing of the footage served to confirm this preliminary view with a degree of considerable certainty. As Professor Cook explained:
I think if it had only been in the single terminal event then one could have been less certain but combined with the earlier history and particularly with the information you have provided me more recently describing that after those how he was confused and vomited, had a warning of the events described on at least one occasion and the video of course that makes my certainty much greater.
20.20 Third, as noted at [12.2] above, the opinions expressed by Professor Cook formed one part of the totality of the evidentiary matters relevant to consideration of the cause of each of the four episodes, and ultimately, the cause and manner of Mr Roberts’ death.
20.21 Conclusion: The cause of the 28 August 2019 episode was Mr Roberts’ underlying and undiagnosed epilepsy. This manifested in Mr Roberts experiencing a prolonged epileptic seizure on the 28 August 2019 resulting in gastric content aspiration and cardiac arrest, leading to his death. Accordingly, Mr Roberts died of natural causes.
21.1 Consideration of this issue will require examination of specific features in relation to the three previous episodes. It will also be necessary to examine the steps taken by clinicians at BBH as well as at Bathurst CC.
The 19 April 2019 episode 21.2 Dr Lipski gave evidence that he read the NSWA records and BBH triage notes before reviewing Mr Roberts. Following his own observations of Mr Roberts, Dr Lipski formed the view that Mr Roberts may have experienced a drug-induced seizure having regard to Mr Roberts’ history of substance use, reference in the NSWA records to drug paraphernalia (a home-made bong and used needle) being located near Mr Roberts, and Mr Roberts’ apparent response to the administration of naloxone.
21.3 Although Dr Lipski acknowledged that his assessment of Mr Roberts was “also coloured by the fact that [he] found some tablets on [Mr Roberts ] which he may have concealed”, Dr Lipski clarified that his assessment was not to the exclusion of other possible diagnoses. Notwithstanding, Dr Lipski acknowledged that he did not give any meaningful consideration to a differential diagnosis because his assessment of Mr Roberts, and the investigations performed at BBH, were limited.
21.4 As part of his assessment, Dr Lipski explained that he informed Mr Roberts that he would like to conduct further investigations including a head CT scan (to exclude an intracranial bleed or other structural pathology), urine screen and the blood tests (to identify possible causes of the 19 April 2019 episode). However, Mr Roberts declined further investigations of any kind and informed Dr Lipski that he wished to leave hospital against medical advice.
21.5 Having formed the view that Mr Roberts had the capacity to decline further investigations, Dr Lipski informed him of the risks of self-discharge.
The 5 May 2019 episode 21.6 Dr Lipski performed an assessment of Mr Roberts, who refused to be offloaded from the NSWA trolley or enter the emergency department. On this occasion, Dr Lipski again explained to Mr Roberts the risks of recurrent seizures, and having an unrecognised, undiagnosed and untreated serious medical condition.
21.7 Mr Roberts again declined any further investigations and refused any further assessment. As he had several weeks earlier, Dr Lipski formed the view that Mr Roberts had the capacity to indicate such a refusal and to effectively discharge himself against medical advice.
The 30 May 2019 episode 21.8 Dr Williams considered that opioid use may have been a factor in Mr Roberts’ presentations to hospital given his apparent response to the administration of naloxone. However, Dr Williams also noted “conflicting information” in the NSWA records which indicated that Mr Roberts did not present with respiratory depression or pinpoint pupils suggestive of opioid toxicity. Accordingly, Dr Williams gave evidence that at the time of her review of Mr Roberts, she had not reached a clear diagnosis as to the cause of his presentation, noting that there was a “wide variety” of possible causes.
21.9 After Mr Roberts again refused any further investigation and treatment, both Dr Williams and Dr Gupta explained the associated risks in Mr Roberts effectively discharging himself. As with the two previous presentations to hospital, Dr Williams also formed the view that Mr Roberts had the capacity to refuse further treatment.
21.10 Associate Professor Roberts considered that following the 19 April 2019 episode, it would be reasonable for a blood glucose level, ECG and routine blood tests (to identify potential kidney or liver dysfunction) to be performed. Associate Professor Adams agreed that an ECG and blood tests would form part of any typical assessment process, as well as taking a careful history.
21.11 Whilst an ECG was performed on 19 April 2019, Mr Roberts declined an ECG during his two subsequent presentations to BBH. A blood glucose test was performed during all three presentations.
21.12 Conclusions: On each of Mr Roberts’ presentations to BBH, there was insufficient information before the treating clinicians to allow for a definitive diagnosis to be reached. The evidence establishes that the need for further investigations was recognised and that some minimal investigations were performed, to the extent that Mr Roberts gave his agreement for these to occur. However, Mr Roberts declined further investigations that may have assisted in arriving at a clear diagnosis, or at least a more informed clinical view regarding the cause of each episode and his repeated presentations.
21.13 The BBH clinicians explained to Mr Roberts the utility of such further investigations and the risks associated with investigations not being performed and Mr Roberts discharging himself. Having formed the view that Mr Roberts had the capacity to make decisions regarding his own care and treatment, there was little else that the clinicians could have reasonably done. Accordingly, to the extent that Mr Roberts permitted them to do so, the treating clinicians took appropriate steps to ensure Mr Roberts’ well-being.
Steps taken by Justice Health nursing staff 21.14 Upon his return to Bathurst CC following his first two presentations to hospital, Mr Roberts was placed in a camera cell for overnight observations. On each occasion, Justice Health nurses either placed Mr Roberts on GP and D&A waitlists or checked that Mr Roberts was already on such lists and would be seen by medical staff as soon as possible.
21.15 RN Asimus saw Mr Roberts on 23 April 2019, with a further review by Justice Health nurses on 28 April
21.16 The question that arises is whether appropriate consideration was given to any factors which may have contributed to Mr Roberts’ decision to refuse further medical treatment at hospital, and whether it may have been possible to overcome these factors, or at least counsel Mr Roberts regarding them.
21.17 In his first expert report, Professor Cook noted the following: Mr Roberts’ circumstances, past history of drug use, and current methadone use with the possibility of ingestion of additional non-prescribed drugs or further methadone seem to have distracted clinical staff from the identification of a primary seizure disorder. Whilst understandable in the circumstance, this likely prevented Mr Roberts being offered appropriate investigation and therapy.
It is possible from what is described regarding his response to advice in hospital that he may have declined further assessment, but after a seizure people are often still confused to a degree for some hours, and not able to make informed judgements in this period. It is unfortunate his situation was not pursued more vigourously by those caring for him after these are seizures, as appropriate therapy may have changed the course of events.
21.18 It was submitted on behalf of Justice Health and Western NSW Local Health District that “on each and every episode, Mr Roberts did not exhibit any signs of confusion or being in the post-ictal phase”.
This is because, it was submitted:
(a) following the 19 April 2019 and 5 May 2019 episodes, Dr Lipski noted that Mr Roberts’ observations were stable, he was alert, cooperative, communicating and not in an obvious postictal state; and
(b) following the 30 May 2019 episode, Dr Williams followed her usual practice in determining whether Mr Roberts was alert, responding appropriately, and oriented in time and place.
21.19 Professor Cook appropriately acknowledged in evidence that he was not in a position to contradict the observations of clinicians who had the opportunity to assess Mr Roberts as a patient before them. However, Professor Cook made this important observation: … This is the problem for everyone seeing patients in an emergency setting is that [they] are unfamiliar to the - to you and you often don’t know what they’re like normally so it can be hard to judge sometimes… 21.20 Professor Cook also elaborated further: Well typically after an epileptic seizure people are confused for some time afterwards, sometimes very confused. It might not be apparent and I’ve made this point later that it might not be apparent to those who aren’t familiar with [Mr Roberts] whether he’s returned to his normal self after a seizure when assessed and I wondered whether when assessed at the hospital he was in fact back to his normal self but that can be a difficult point to be certain of… 21.21 For clarity, counsel for Justice Health and Western NSW Local Health District asked Professor Cook whether his evidence regarding confusion during the post-ictal stage was given as a general medical proposition or specifically in relation to Mr Roberts. Professor Cook responded in this way: I, that’s absolutely right. As a general proposition people are confused after seizures. Certainly that was described with Mr Roberts that he was confused after them and I thought an additional feature in favour of them being epileptic seizures.
21.22 The above evidence therefore establishes that Mr Roberts most likely demonstrated some confusion after each of the three previous episodes. This raises the possibility that this confusion may have been a contributing factor to Mr Roberts’ decision to decline further assessment and treatment at hospital. This in turn suggests that that an opportunity may have existed following Mr Roberts’ return to a correctional centre, after his confusion had subsided, for the topic of further assessment and treatment to be revisited with him when reviewed by a Justice Health clinician.
Inmate patient confidentiality 21.23 One matter which may have been a factor in Mr Roberts’ stance regarding further treatment is the presence of a CSNSW officer during reviews conducted by Justice Health nursing staff.
21.24 RN Franklin reviewed Mr Roberts on 19 April 2019. She could not recall whether a CSNSW officer was present at this time but indicated that it was usual protocol for a CSNSW officer to be present during such reviews. RN Franklin gave evidence of her awareness that an inmate may request that a CSNSW not be present during such a review. However, RN Franklin indicated that she did not have any personal experience of an inmate making such a request and was unaware whether inmates are informed of their entitlement to do so.
21.25 RN Abbe reviewed Mr Roberts on 30 May 2019 when a CSNSW officer was present. RN Abbe gave evidence that this is consistent with usual practice where a CSNSW officer is present at all times during inmate consultations at the Bathurst CC clinic. RN Abbe also gave evidence that it is her usual practice to:
(a) Invite inmates to indicate whether they would prefer that the CSNSW officer wait outside the clinic during such consultations. It should be noted that in the event of such a request being made, a CSNSW officer is not obliged to accede to it.
(b) Inform inmates that CSNSW officers are not given automatic access to an inmate’s medical records, and that what an inmate discloses to her is not in turn disclosed to a CSNSW Officer. RN Abbe gave evidence that in her experience, inmates have expressed surprise when given this information.
21.26 Two Justice Health policy documents are relevant to consideration of this issue: Guidelines on the use and disclosure of inmate/patient medical records and other health information (January 2018) and Requesting and Disclosing Health Information (September 2017). Neither document expressly provides for Justice Health staff to inform inmates that:
(a) their health information will not be shared with CSNSW; or
(b) they may request a consultation with Justice Health staff without a CSNSW officer being immediately present.
21.27 Dr Gary Nicholls, Justice Health Clinical Director of Primary Care Medicine, gave evidence that “in general” it is the case that Justice Health clinicians inform inmate patients that their health information will not be shared with CSNSW (except in acute situations requiring disclosure of lifesaving information).
21.28 Dr Nicholls was unable to identify any specific policy which requires inmates to be informed of their right to request that a consultation occur in the absence of a CSNSW officer. Instead, Dr Nicholls indicated that as part of their training Justice Health clinicians are expected to provide appropriate care in a confidential manner, subject to overriding security concerns. Accordingly, Dr Nicholls expressed his confidence that clinicians discuss issues of confidentiality with inmate patients.
Further, Dr Nicholls indicated that it is usually only in highly volatile areas of correctional centres that a CSNSW officer remains in close physical contact with a patient when care is provided by Justice Health staff. In a primary care clinic environment, where no specific risk exists, Dr Nicholls explained that in his experience: [A CSNSW officer] could be outside of the door or down the corridor and there would be confidentiality provided to, to the patient, and I've certainly witnessed that amongst you know, many of my colleagues and notably amongst drug and alcohol nurses and doctors and psychiatrists who are seeing patients. You'll often have a clinic set up where the officer will be in the corridor watching doors, but clinicians are in the rooms alone with inmates giving them a level of privacy 21.29 It was submitted on behalf of the Commissioner of CSNSW that any consideration regarding the issue of inmate patient confidentiality relevant to the presence of a CSNSW officer during a consultation between an inmate and a Justice Health clinician was not a matter about which any opportunity was extended to CSNSW to provide evidence about prior to the inquest. Accordingly, it
was submitted, “there is an element of procedural unfairness for CSNSW in being able to properly address this issue”.
21.30 First, it is acknowledged that consideration of matters relating generally to inmate patient confidentiality did not form a discrete issue on the issues list distributed to the parties prior to the inquest. However, it was an issue which arose during the course of the inquest in consideration of the discrete issue as to whether appropriate steps were taken to ensure Mr Roberts’ well-being following the three previous episodes.
21.31 Second, the Commissioner of CSNSW was legally represented at the inquest. The Commissioner’s legal representatives did not seek to ask any questions of RN Franklin or RN Abbe, after Counsel Assisting explored with them issues relating to inmate patient confidentiality concerning Mr Roberts specifically, and more generally. The legal representative for the Commissioner did ask Dr Nicholls a number of questions but none were related to the topic regarding the presence of a CSNSW officer during a consultation between an inmate and a Justice Health Clinician.
21.32 Having regard to the above, and to the fact that no adverse finding is made affecting the Commissioner’s interests regarding this issue, no element of procedural unfairness exists.
21.33 Counsel Assisting submitted that it is necessary or desirable for a recommendation to be made to Justice Health to give consideration to the creation of a specific policy which guides clinical staff regarding the issues of patient confidentiality discussed above.
21.34 It was submitted on behalf Justice Health that such a recommendation ought not be made because:
(a) “the practicalities of this recommendation have not been the subject of sufficient consideration”;
(b) there is an existing practice by which inmates are informed that their health information is kept confidential;
(c) there are important considerations relating to the safety and security of Justice Health clinicians, which have not been “considered in a fulsome manner”.
21.35 Further, it was submitted on behalf of the Commissioner of CSNSW that such a proposed recommendation “will impact on CSNSW operations” and that the Commissioner has not been afforded procedural fairness on this issue.
21.36 First, it is acknowledged that the practicalities of this proposed recommendation were not explored in evidence during the inquest. It is always intended that any recommendation made to a sufficiently interested party be capable of practical and effective implementation, and that any recommendation does not adversely impact upon the other responsibilities and operations engaged in by that party. However, it is not always possible within the confines of an inquest to give consideration to the mechanics and processes involved with implementation of a recommendation.
Rather, a recommendation typically invites a party to give appropriate consideration to the subject of the recommendation, rather than mandating a particular course of action.
21.37 Second, the existing practice referred to by counsel for Justice Health in submissions was not the subject of evidence in the inquest. Instead, the evidence given by Dr Nicholls established that as at September 2022, Justice Health did not have a specific policy which addressed the issues of inmate patient confidentiality discussed above.
21.38 Third, it is acknowledged that Dr Nicholls’ anecdotal evidence suggests that Justice Health clinicians, consistent with their training, are conscious of providing appropriate and confidential care, and that in many cases where no security risk exists, consultations between an inmate and clinician can occur in private. However, the evidence of both RN Franklin and RN Abbe casts some doubt regarding whether inmates are appropriately informed about the degree of confidentiality which attaches to health information they might disclose or discuss with a clinician.
21.39 Fourth, it is also acknowledged that the relevant evidence as to this issue is largely anecdotal and pertains to clinical care provided at Bathurst CC during a defined period. The inquest did not receive any evidence as to the practices of Justice Health clinicians at other correctional centres.
21.40 It is therefore desirable to make the following recommendation, recognising the matters referred to above.
21.41 Recommendation 1: I recommend to the Chief Executive, Justice Health & Forensic Mental Health Network, that, in consultation with Corrective Services New South Wales, appropriate steps be taken to ensure that adequate education and guidance is provided to Justice Health clinicians so that they may inform inmate patients of: (a) the confidentiality requirements that attach to provision of any health information by an inmate to a clinician; and (b) the ability to request a consultation with a clinician in the absence of a Corrective Services New South Wales officer, subject to any safety and security considerations that may attach to such a request.
Involvement of Aboriginal Health Worker 21.42 One additional matter concerns the possibility of Mr Roberts being referred to an Aboriginal Health Worker for a consultation. Whilst such health workers operate within correctional centres, Dr Nicholls gave evidence that their role is to act as a liaison between Justice Health staff and inmate patients to ensure that inmates are provided with culturally safe and appropriate care.
21.43 It is not clear whether an Aboriginal Health Worker was employed at Bathurst CC in 2019. However, Dr Nicholls gave evidence that Mr Roberts did not see such a worker in April and May 2019. RN Abbe gave evidence that if such a worker had been available at Bathurst CC at the time, she would have made a request for Mr Roberts to be referred.
21.44 Dr Nicholls identified a number of issues associated with Aboriginal Health Workers, namely that such positions are difficult to fill and undergo a degree of turnover, resulting in many vacant positions currently (as at September 2022). However, Dr Nicholls indicated his understanding that (again as at September 2022) an active recruitment process was underway to fill any vacant positions.
21.45 Conclusions: Justice Health nursing staff took appropriate steps to ensure Mr Roberts’ well-being upon his return to custody. Arrangements were made to ensure that Mr Roberts was reviewed by a medical officer in a timely manner or placed on a waitlist for such a review to occur.
21.46 The evidence establishes that Mr Roberts was likely demonstrating some confusion after each of the three previous episodes, or at least that the BBH clinicians may not have been able to accurately determine whether Mr Roberts’ presentation had returned to its usual baseline, due to their lack of familiarity with Mr Roberts. However, it is not known whether any confusion that Mr Roberts might have been experiencing contributed to his decision to decline further assessment and treatment at hospital.
21.47 It is also not known whether Mr Roberts held any concerns regarding the confidentiality of his health information, and whether this contributed to his decision to decline further medical assessment and investigations. Similarly, it is also not known whether having access to an Aboriginal Health Worker might have had some bearing on Mr Roberts’ decision-making process.
21.48 Ultimately, what the evidence demonstrates is that Mr Roberts had repeated presentations to hospital over a two month period without a definitive diagnosis being formulated by any clinician regarding the cause of these presentations, and in circumstances where Mr Roberts had declined any further assessment or treatment. At least by the time of the third presentation, it would have been reasonable for some consideration to have been given as to whether any modifiable factors, such as those described above, existed which were contributing to the stance taken by Mr Roberts.
The evidence indicates that no such consideration was given. It is difficult to reach a more definitive conclusion regarding this issue. This is because there is no specific evidence as to when such consideration might or should have been given, and by whom.
21.49 However, even if such consideration had been given it is not known whether any modifiable factor relevant to Mr Roberts’ decision-making process would have been identified. Further, even if a modifiable factor had been identified, it is not known whether Mr Roberts would have been receptive to counselling or attempts to revisit his decision-making process. Finally, there is no evidence that even if Mr Roberts had reversed his decision and permitted further medical assessment and investigations to be performed that this would have altered the eventual outcome.
Recommendations 21.50 It is apparent from all of the above that diagnosis of the underlying health condition which Mr Roberts suffered from was critical to his subsequent management (to the extent that he would have allowed such management to occur). Given the complexities involved in formulating a diagnosis of primary epilepsy, it is necessary to make the following recommendation.
21.51 Recommendation 2: I recommend to the Chief Executive, Justice Health & Forensic Mental Health Network, that consideration be given to providing ongoing education and training to all nursing and medical clinicians regarding the clinical features and identification of primary epilepsy.
21.52 It should be noted that Justice Health has already taken steps to progress this issue by seeking advice from the Department of Neurology at Prince of Wales Hospital for assistance with relevant training. Further, arrangements are being put in place to provide a clinical staff with video-recorded educational material from an eminent neurologist and epileptologist at Prince of Wales Hospital.
21.53 Counsel Assisting submitted that it would be necessary or desirable to make a recommendation that Justice Health give consideration to providing cultural awareness education and cultural safety training to all Justice Health staff so that they may be more culturally aware in a clinical setting. This proposed recommendation was welcomed and supported by Mr Roberts’ family.
21.54 It was submitted on behalf of Justice Health that such a recommendation is neither necessary or desirable because, in essence, cultural awareness education and cultural safety training is already occurring and that this issue has been considered by a number of previous inquests. In the submissions made on behalf of Justice Heath, reference was made to a number of training programs and initiatives, together with a Statement of Commitment to Aboriginal Health, previously commissioned and issued by Justice Health. Much of this material did not form evidence in, and was not considered by, the inquest.
21.55 Cultural awareness education and cultural safety training for clinicians responsible for the care of First Nations people is obviously an important issue, particularly where such care is provided in a correctional setting. However, the inquest did not receive any evidence regarding the education and training which may have available to Justice Health clinicians as at August 2019, or whether such education and training had any bearing on the clinical management of Mr Roberts. There is therefore no evidentiary basis to allow for a recommendation regarding these matters to be made.
21.56 It was submitted on behalf of Mr Roberts’ family that it is necessary or desirable to make a recommendation that Justice Health give due consideration to the importance and value of access to culturally appropriate and culturally competent health care by prioritising employment of Aboriginal Health Workers within correctional centres and providing staff with regular and ongoing training and support to improve recruitment and retention rates, and implement frequent and mandatory cultural competency training for all health staff. The question of training and education has already been dealt with above and there is no need to revisit this issue.
21.57 As to the proposed recommendation regarding prioritisation of employment of Aboriginal Health Workers, it should first be noted that:
(a) it is not known whether the ability to have access to such a worker in 2019 might have had any bearing on Mr Roberts’ decision to refuse medical assessment and treatment;
(b) there is no evidence as to what factors may have contributed to the difficulties encountered by Justice Health regarding retention of Aboriginal Health Workers;
(c) the evidence given by Dr Nicholls indicates that, as at September 2022, Justice Health was embarking on an active recruitment process to fill any vacant positions; and
(d) none of these matters were explored with Dr Nicholls in evidence by the legal representative for Mr Roberts’ family.
21.58 However, given the submission by counsel for Justice Health that it “strongly supports the principle” behind the recommendation proposed by Counsel Assisting, the challenges associated with retention rates for Aboriginal Health Worker positions and the absence of information regarding the outcome of the ongoing recruitment process, it is desirable to make the following recommendation.
21.59 Recommendation 3: I recommend to the Chief Executive, Justice Health & Forensic Mental Health Network, that a review be conducted to ensure that Aboriginal Health Worker positions in New South Wales correctional centres are sufficiently staffed to provide inmate patients with access to culturally appropriate healthcare.
Steps taken by Dr Kehoe 21.60 On 3 May 2019, Dr Kehoe saw Mr Roberts via AVL for a D&A consultation. The purpose of the consultation was to review Mr Roberts’ participation in the OST program. Dr Kehoe was aware of the 19 April 2019 episode and that Mr Roberts had declined further investigations at hospital. Dr Kehoe considered that there was no evidence available to him to form a conclusion as to the likely cause of the episode. In addition, Dr Kehoe gave evidence that he could not recall whether he provided Mr Roberts with any counselling as to appropriate next steps including possible further investigations or treatment.
21.61 Dr Kehoe also gave evidence that he would have considered any matters relating to the episode to be a primary health issue before considering to be a drug and alcohol issue. Dr Kehoe indicated that he was unaware whether Mr Roberts had been placed on a waitlist to see a GP regarding the episode but assumed that this had occurred.
21.62 Conclusions: The purpose of the 3 May 2019 consultation was to review Mr Roberts’ participation in an OST program. Dr Kehoe was aware of the 19 April 2019 episode and Mr Roberts’ decision to refuse treatment at hospital and self-discharge against medical advice However, Dr Kehoe did not have any available information upon which to form a view as to the likely cause of the episode other than it was likely a personal health, rather than a drug and alcohol, issue. Accordingly, appropriate steps were taken on3 May 2019 to ensure Mr Roberts’ well-being in so far as they related to his OST management.
Steps taken by Dr Fleri 21.63 Mr Roberts was scheduled to see Dr Fleri at a GP clinic on 14 June 2019 to follow up on his recent presentations to hospital. Dr Fleri was only aware of the episodes on 19 April 2019 and 30 May 2019.
Although Mr Roberts refused to attend the appointment, Dr Fleri still recommended Mr Roberts for group cell placement due to the nature of his presentations. Dr Fleri gave evidence that she did not place Mr Roberts on a seizure management plan because he had declined to attend the 14 June 2019 appointment, he had similarly declined further investigations at BBH, and there was evidence of drug use associated with the 19 April 2019 episode.
21.64 On 23 August 2019, Dr Fleri reviewed Mr Roberts following his transfer to the MRRC. By this stage, Dr Fleri was still only aware of the 19 April 2019 and 30 May 2019 episodes based on her review of the hospital discharge summaries and the progress notes on Mr Roberts’ file. As to her lack of awareness of the 5 May 2019 episode, Dr Fleri raised two possible explanations:
(a) the relevant discharge summary had not been filed correctly in Mr Roberts’ health care file; and
(b) at the time, the progress note entry for the 5 May 2019 episode would have been handwritten, and it was not her usual practice, due to time constraints, to read every handwritten entry in a patient’s progress notes.
21.65 As to the 19 April 2019 and 30 May 2019 episodes, Dr Fleri formed a provisional diagnosis that they were “drug-related” and that no further action was required. Dr Fleri’s provisional diagnosis was based on the following:
(a) her reading of the NSWA records, progress notes and discharge summary regarding the 19 April 2019 episode;
(b) reference to a bong being found in Mr Roberts’ hand on 19 April 2019, which Dr Fleri considered to be evidence that Mr Roberts “was abusing drugs”;
(c) Mr Roberts’ apparent response on 19 April 2019 and 30 May 2019 to the administration of naloxone; and
(d) Mr Roberts’ participation in the methadone OST program which demonstrated to Dr Fleri that Mr Roberts had “been involved in illicit drug use”.
21.66 Dr Fleri acknowledged that, with particular regard to points (b) and (c) above, she did not consider any differential diagnosis for Mr Roberts’ previous episodes, although it would have been her general practice to do so. When asked why a differential diagnosis was not considered regarding Mr Roberts’ presentation on 23 August 2019, Dr Fleri explained: Because of the drug paraphernalia that he was found with, certainly on one occasion and because of the reversal of his symptoms on naloxone it appears that I didn’t.
21.67 Dr Fleri indicated that had she known that naloxone was not administered to Mr Roberts on 5 May 2019 this would not have influenced her provisional diagnosis. Dr Fleri went on to explain that the NSWA records regarding this discovery of a syringe and half a tablet appeared “to confirm that his seizure like activity was drug-related and settled a reasonably quickly without the need for administration of naloxone”.
21.68 Dr Fleri gave evidence that if she had suspected a neurological cause for the previous episodes, she would have referred Mr Roberts for an EEG and a brain CT scan, with review by a neurologist to follow. Dr Fleri indicated that even if such referrals had been made, the likely wait time would have been in the region of 6 to 12 months, suggesting that such referrals would likely have made no difference to the outcome, given that Mr Roberts’ death occurred five days after her review.
21.69 Professor Cook gave evidence that an EEG, an MRI, a test for autoimmune encephalitis and an ECG would all be required in order to make a formal diagnosis of epilepsy. Further, Professor Cook went on to explain: I think after a convulsive episode like this everyone needs to be assessed by a neurologist ideally someone who is experienced in epilepsy they should see a neurologist. I think that's very important, certainly after two events. Now, you know, at the hospital I work at we also have a prison ward and we see lots of prisoners as out patients as well and so I'm very familiar with the scenario and the reluctance of prisoners to want to engage in further assessment under these circumstances for a variety of reasons and that's what makes the management of epilepsy and its diagnosis in the prison system so very challenging. I don't know of any easy solution to that but I think had it been – it might be that had it been put to Mr Roberts that he had a serious and recurrent cause for these events that might be fatal, that he might have altered his thinking about it.
21.70 Dr Fleri initially gave evidence that she had no independent recollection of whether she attempted to counsel Mr Roberts regarding the utility of allowing further investigations to be conducted regarding his episodes. However, Dr Fleri later indicated that, with the benefit of hindsight, she would have counselled Mr Roberts “much more extensively”. This suggests that at the relevant time, little to no counselling was provided by Dr Fleri to Mr Roberts. Dr Fleri acknowledged that the following matters were relevant to the absence of more extensive counselling being provided to Mr Roberts:
(a) Her conclusion that the 19 April 2019 and 30 May 2019 episodes were self-induced due to Mr Roberts’ drug use;
(b) Mr Roberts refusal of further investigations and treatment at hospital; and
(c) Mr Roberts’ apparent lack of concern during the 23 August 2019 consultation.
21.71 Dr Fleri acknowledged that it would have been of assistance if Mr Roberts had been referred to an Aboriginal Health Worker. Although Dr Fleri observed that no such worker was employed at the MRRC in September 2019, there is no evidence that Dr Fleri gave any consideration to such a referral being made at the relevant time.
21.72 Conclusions: Dr Fleri did not make herself aware of the 5 May 2019 episode despite such information being available in Mr Roberts’ progress notes. This is perhaps understandable given that Dr Fleri was operating under time constraints, there was a voluminous amount of handwritten material to review in the progress notes and it appears that the notes were not ordered in a logical and chronological manner. In any event, Dr Fleri’s evidence is that awareness of the 5 May 2019 episode would not have made any difference to her provisional diagnosis. Notwithstanding, Dr Fleri remained aware that Mr Roberts had experienced two seizure-like episodes within a period of six weeks and that, by virtue of Mr Roberts’ refusal of medical treatment, no investigations had been conducted to identify the cause of the episodes.
21.73 The evidence establishes that Dr Fleri departed from her usual practice in not considering a differential diagnosis as to the cause of the two episodes suffered by Mr Roberts that she was aware of. According to Dr Fleri, this is because Mr Roberts had been found with drug paraphernalia and had apparently responded to naloxone administration. It is not clear on the evidence whether Dr Fleri was distracted in arriving at a provisional diagnosis, or not considering a differential diagnosis, in the manner described by Professor Cook at [21.17] above. To the extent that such distraction may have existed, Professor Cook acknowledges that it was “understandable in the circumstance”. In any event, if Dr Fleri had considered as part of a differential diagnosis that there was a neurological cause for the two episodes that she was aware of, the evidence establishes that she would have instituted appropriate investigations (with Mr Roberts’ agreement).
21.74 There is no evidence that Dr Fleri took any steps to seek to understand any modifiable factors that may have contributed to Mr Roberts’ refusal of further medical assessment, or counsel Mr Roberts regarding the utility of allowing such assessment to be conducted. This is because Dr Fleri had formulated her provisional diagnosis that Mr Roberts’ episodes were self-induced due to drug use.
It is only with the benefit of hindsight that Dr Fleri considered that she should have counselled Mr Roberts much more extensively.
21.75 Notwithstanding, given that Dr Fleri reviewed Mr Roberts on 23 August 2019 in her capacity as a GP and for the purpose of following up Mr Roberts’ earlier presentations to hospital, an opportunity existed to revisit with Mr Roberts the possibility of further investigations and treatment. This is particularly so because 23 August 2019 represented the first opportunity for Mr Roberts to be reviewed by a medical practitioner in the GP clinic (after Mr Roberts declined to attend an appointment on 14 June 2019 ) following his most recent presentation to hospital.
21.76 Regrettably, this opportunity was not taken by Dr Fleri although it is acknowledged that there is no evidence as to whether such an opportunity, if taken, would have materially altered the eventual outcome. Professor Cook acknowledged the reluctance of inmates to want to engage in further assessment in such circumstances without there being “any easy solution” to these challenges.
22.1 On 4 February 2019, Mr Roberts was seen for a D&A clinical review. An ECG was performed, and Dr Kehoe considered that the results were within normal limits. Associate Professor Adams expressed the view that this result (and other ECG results between 2015 and 2019) did not indicate any concerns for arrhythmia problems or structural cardiac disease, and that Mr Roberts was in normal sinus rhythm. Further, Associate Professor Adams opined that the ECG result should not have prompted further investigations or affected Mr Roberts’ continuing participation in the methadone OST program. Associate Professor Adams suggested that, in general, ECG testing of patients on a methadone OST program, particularly those on high doses or at times when the dosage is significantly increased, would be useful.
22.2 On 7 February 2019, Dr Kehoe increased Mr Roberts’ methadone dosage from 65mg to 70mg, and provided instructions to allow for the dose to be increased by 5mg each fourth or fifth dosing, up to a maximum of 100mg.
22.3 On 3 May 2019, Dr Kehoe applied the same methodology, allowing for further incremental increases of Mr Roberts’ methadone up to a maximum of 120mg. Associate Professor Roberts opined that this increase in dosage was appropriately managed.
22.4 Dr Kehoe did not require any further ECG testing on this occasion given that Mr Roberts had been on higher methadone doses in the past without complication, and that the ECG results from 4 February 2019 showed that Mr Roberts’ QT interval was in the lower range of normal. However, Dr Kehoe suggested that an ECG be repeated on 15 May 2019 given that it was expected that Mr Roberts’ methadone dose by that date would be around 120mg.
22.5 On 6 May 2019, Mr Roberts was seen by RN Asimus and it was decided that his participation in the methadone OST program should continue according to current arrangements. Mr Roberts was not referred to Dr Kehoe, or any other medical practitioner, for further review.
22.6 Following the 30 May 2019 episode, no D&A review was conducted with Mr Roberts. However, he was waitlisted for an appointment at the GP clinic on 14 June 2019.
22.7 Conclusions: The management of Mr Roberts’ methadone dosage increases was appropriate and consistent with the approach suggested by Associate Professor Adams. However, although Dr Kehoe recommended that a further ECG be performed on 15 May 2019, there is no evidence to indicate that this occurred.
22.8 Dr Kehoe’s last involvement in Mr Roberts’ management was on 3 May 2019. Following this, Mr Roberts’ experienced two further episodes on 5 May 2019 and 30 May 2019. Despite this, Mr Roberts was not scheduled for further D&A review by a medical officer. The evidence suggests that such a review would have been useful to ensure that Mr Roberts’ continuing participation in the methadone OST program was appropriately managed. However, the available evidence does not allow for any conclusion to be reached as to whether any such review, and the performance of an ECG on 15 May 2019, might have materially affected subsequent events.
23.1 As at 28 August 2019, Mr Roberts’ daily methadone dose was 120mg. It should be remembered that Associate Professor Roberts expressed the view that:
(a) this prescribed daily dose could have accounted for the postmortem methadone concentration;
(b) it could be expected that Mr Roberts’ usual methadone concentration was in the vicinity of that observed postmortem; and
(c) the postmortem methadone concentration could have resulted from the administration of Mr Roberts’ prescribed dose administered approximately six hours before his death.
23.2 As noted above, there is no evidence that Mr Roberts self-administered any unprescribed drug on 28 August 2019. Further, Associate Professor Roberts opined that if Mr Roberts had self-administered an unprescribed dose of methadone, its sedating effects would have been observable.
23.3 Conclusions: The postmortem methadone concentration can most likely be attributed to the daily dose of 120mg administered to Mr Roberts approximately six hours prior to his death.
24.1 Section 13.2 of the CSNSW Custodial Operations Policy and Procedures (COPP) provides that if an inmate is found unconscious or seriously injured, the first responding officer must call or initiate a call for an ambulance immediately, and then call Justice Health to provide urgent medical assistance. Immediately following these steps, first aid must be provided.
24.2 Mr Lethbridge gave evidence that he made an urgent call to Justice Health for medical assistance and commenced first aid. However, he did not call for an ambulance. Mr Lethbridge gave evidence that he could not recall any obligation on his part to do so. He also acknowledged that when using his radio to call for medical assistance, a request for an ambulance could have been made at the same time. However, Mr Lethbridge explained that his responsibility was to seek immediate medical assistance and to maintain observations of Mr Roberts.
24.3 As part of his CSNSW training, Mr Lethbridge completed a unit of competency in first aid on 15 November 2017 which deals with the signs, symptoms and management of seizures. Where an inmate is found to be suffering a seizure, CSNSW officers are to follow the Australian Resuscitation Council (ARC) Guidelines.
24.4 Guideline 9.2.4 of the ARC Guidelines provide that if a person is unconscious and actively seizing, the rescuer should:
(a) follow the persons seizure management plan, if there is one in place;
(b) manage the person according to ARC Guideline 3; and
(c) call an ambulance.
24.5 Additionally, both ARC Guidelines 3 and 9.2.4 provides that a rescuer should remove a person from danger, assist the person to the ground and position the person on their side when practical, and ensure the person’s airway is open. Mr Lethbridge gave evidence that he acted in accordance with his first aid training to ensure that Mr Roberts was in the recovery position, and that by maintaining observations of Mr Roberts he ensured that Mr Roberts did not injure himself or anyone around him.
24.6 Mr Lethbridge also gave evidence that:
(a) He initially found Mr Roberts to be “already pretty much” in the recovery position but indicated later in evidence that Mr Roberts was “enough on his side to be in the recovery position, especially the way his arm was across his body”;
(b) he did not touch or move Mr Roberts because he had been trained not to do so when a person is experiencing a seizure; and
(c) he considered that Mr Roberts’ airway was open.
24.7 Officer Elias Habib gave evidence that when he arrived on the scene, he saw that Mr Roberts was “in a half turned position so not completely on his side”. Officer Habib therefore raised with Mr Lethbridge they should turn Mr Roberts “over a little bit more” to help aid his recovery.
24.8 RN Paterson gave evidence that on her arrival she found Mr Roberts lying on his back with his head partially tilted by bed sheets that were rolled up next to him against the wall. RN Paterson indicated that with the assistance of another person, she turned Mr Roberts and tilted his head to one side.
24.9 Other medical staff arrived at Mr Roberts’ cell within a couple of minutes of Mr Lethbridge’s call for medical assistance. An ambulance was called at 2:30pm. Mr Roberts was placed in the recovery position and his airway was cleared. Oxygen therapy was administered together with two doses of naloxone. CPR was commenced at 2:40pm, which was continued upon the arrival of paramedics at 2:50pm. Attempts were made to deploy a defibrillator but it was found that Mr Roberts had no shockable rhythm.
24.10 It is submitted on behalf of Mr Lethbridge that COPP Section 13.2 did not apply to the events of 28 August 2019 in that Mr Roberts was not found by Mr Lethbridge to be “unconscious or seriously injured”. Therefore, it is submitted, there was no requirement for Mr Lethbridge to immediately call for an ambulance in accordance with the relevant COPP provisions. It is acknowledged that interpretation of the terms of this provision was not explored in evidence.
24.11 However, the evidence establishes that Mr Roberts was suffering from a seizure and was unaware of, and unresponsive to, the environment that he was in and any external stimuli. Accordingly, he could have been considered to be unconscious for the purposes of COPP Section 13.2. Indeed, the ARC Guidelines make clear that a person may be unconscious and actively experiencing a seizure.
24.12 Conclusions: Immediate medical assistance was appropriately sought after Mr Roberts was found unresponsive in his cell. Resuscitation was commenced in a timely manner and continued until all available life-saving measures had been exhausted.
24.13 Mr Lethbridge did not strictly comply with the provisions of COPP 13.2 which required that an ambulance be called upon finding Mr Roberts to be unconscious. Although Mr Roberts was experiencing an active seizure, he was also not aware of, and not responding to, his external environment. Further, it appears that Mr Roberts was not entirely in the recovery position when discovered by Mr Lethbridge given that nursing staff who arrived on the scene afterwards placed Mr Roberts into this position.
24.14 However, it is evident that Mr Lethbridge otherwise acted in accordance with his first aid training to ensure that medical assistance was sought immediately, that consideration was given to whether Mr Roberts’ airway was open and no attempt was made to move or touch Mr Roberts. There is no evidence to suggest that any earlier call for an ambulance or the slight delay in placing Mr Roberts in the recovery position would have materially altered the eventual outcome.
25.2 I also acknowledge the assistance of Senior Constable Claudia Crean, the NSWPF officer-in-charge of the investigation, in compiling the initial brief of evidence.
25.3 The findings I make under section 81(1) of the Act are: Identity The person who died was Roy Roberts.
Date of death Mr Roberts died on 28 August 2019.
Place of death Mr Roberts died at Long Bay Correctional Complex, Matraville NSW 2036.
Cause of death The cause of Mr Roberts’ death was a prolonged epileptic seizure resulting in gastric content aspiration and cardiac arrest.
Manner of death Mr Roberts died of natural causes whilst in lawful custody serving a sentence of imprisonment. The prolonged epileptic seizure on 28 August 2019 represented the fourth occasion since April 2019 that Mr Roberts had suffered a seizure event due to undiagnosed and underlying primary epilepsy.
Mr Roberts died of natural causes whilst in lawful custody serving a sentence of imprisonment. The prolonged epileptic seizure on 28 August 2019 represented the fourth occasion since April 2019 that Mr Roberts had suffered a seizure event due to undiagnosed and underlying primary epilepsy.
Following the first three occasions, Mr Roberts declined any further assessment or treatment at hospital. Although the reasons for Mr Roberts’ decision are not known, opportunities existed during subsequent clinical reviews for these reasons to be explored with Mr Roberts with a view to performing further medial assessment. However, even if these opportunities had been taken it is not
known whether Mr Roberts would have been receptive to them, or whether they may have materially affected the eventual tragic outcome.
26.2 I close this inquest.
Magistrate Derek Lee Deputy State Coroner 3 February 2023 Coroners Court of New South Wales
Inquest into the death of Roy Roberts Annexure A Non-publication Orders
(a) The names, addresses, phone numbers, Visitor Index Numbers and other personal information that identifies or might identify any family member, friend or person who visited Mr Roberts while in custody (other than legal representatives or visitors acting in a professional capacity).
(b) The names, Master Index Numbers and other personal information of any persons in the custody of Corrective Services New South Wales (CSNSW), other than Mr Roberts.
(c) The names of any persons who were protected by an Apprehended Violence Order against Mr Roberts.
(d) The direct contact details, including telephone numbers and email addresses, of CSNSW officers, employees and offices that are not publicly available.
(e) The following CSNSW policy material:
(i) Local Operating Procedure SMSPL32 – Area 1 Methadone Buprenorphine Procedure.
(ii) COPP 1.1 Reception Procedures: i. At [3.2] on page 8 of 36, telephone number and email address of the State Sentence Administration Processing Unit.
ii. At [3.2] on page 9 of 36, telephone number of the Court Escort Security Unit.
iii. At [4.12] on page 16 of 36, complete section.
iv. At [5.1] on page 19 of 36, After Hours (emergency) phone number and email address of the Department of Foreign Affairs and Trade.
(iii) COPP 13.2 Medical Emergencies: i. At [2.5] on page 8 of 17, email address of CSNSW Senior Human Resources Advisor.
ii. At [2.6] on page 9 of 17, email address of CSNSW Senior Human Resources Advisor.
iii. At [2.7] on page 9 of 17, Item 3 in the Table.
(iv) COPP 13.3 Death in Custody: i. At [2.4] on page 6 of 17, third sentence.
ii. At [6.1] on page 12 of 17, telephone number of the ASPU.
(v) COPP 13.8 Crime Scene Preservation
i. At [4.2] on pages 10-11 of 14, whole subsection other than the sentence beginning, “For forensic evidence on victims…”.
(vi) COPP 16.8 Health Security Centre: i. At [1.2] on page 4 of 8, the entire subsection.
ii. At [1.3] on page 4 of 8, dot points 5 and 6.
iii. At [1.4] on page 5 of 8, the entire subsection other than the sentence beginning, “JH&FMHN has their own policy…”.
iv. At [2.1] on page 5 of 8, part of the second paragraph commencing after the word “centre”.
v. At [2.2] on page 6 of 8, item numbers 1, 2, 4, 5 and 6 in the box titled “procedure”.
(vii) COPP 19.1 General Escort Procedure: i. At [4.1] on page 7 of 23, whole subsection other than the sentences beginning, “The Governor/GM/MOS must…” and “This section must be read…”.
ii. At [9.3] on page 16 of 23, the first two paragraphs in the subsection.
iii. At [12.1] on page 18 of 23, the whole subsection other than the sentences beginning, “The note should give…” and “Refer to COPP section 19.6…”.
iv. At [13.2] on page 19 of 23, item numbers 3, 5, 10, 11 and 12 in the box titled “procedure” and some words in item 8.
v. At [13.3] on page 20 of 23, information in the third and fourth rows of the table.
(f) Crime Scene photographs (pages 13, 14, 87-89, 92-98, 100-107, 109, 111-120 of Tab 56).
(g) Hand-held video camera footage.
(h) The entirety of the CCTV footage captured within the Bathurst Correctional Centre on 19 April 2019.
Magistrate Derek Lee Deputy State Coroner 3 February 2023 Coroners Court of New South Wales
Phillip Walton, a 48-year-old with schizophrenia well-managed on clozapine for nearly 20 years, suffered an acute psychotic episode likely precipitated by cannabis use at a circus. After aggressive behaviour requiring po…
Eve Brown, a 42-year-old Aboriginal woman, died of hypovolaemic shock from a ruptured splenic haematoma in the context of urosepsis at a regional hospital. She presented with UTI symptoms and was admitted to Lightning Ri…
A 32-year-old Vietnamese-speaking woman died from sepsis due to pyelonephritis following premature discharge from hospital onto a midwifery support program. Multiple missed diagnostic opportunities occurred: difficulty p…
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.