Coroner's Finding: SMITH Gordon Charles

CORONERS ACT, 2003 SOUTH AUSTRALIA FINDING OF INQUEST An Inquest taken on behalf of our Sovereign Lady the Queen at Adelaide in the State of South Australia, on the 1st and 13th days of July 2016 and the 2nd day of August 2016, by the Coroner’s Court of the said State, constituted of Anthony Ernest Schapel, Deputy State Coroner, into the death of Gordon Charles Smith.

The said Court finds that Gordon Charles Smith aged 91 years, late of Elder Care Kirkholme Nursing Home, 7 Victoria Avenue, Goodwood, South Australia died at the Repatriation General Hospital, 216 Daws Road, Daw Park, South Australia on the 30th day of July 2014 as a result of multi-organ failure due to inanition complicating prolonged delirium of uncertain cause with contributing ischaemic and hypertensive heart disease, congestive cardiac failure, cerebrovascular disease, chronic renal impairment and type 2 diabetes melllitus. The said Court finds that the circumstances of his death were as follows:

1. Introduction 1.1. Gordon Charles Smith was 91 years of age when he died on 30 July 2014. He died at the Repatriation General Hospital (the RGH) where he had been admitted since 3 July 2014.

1.2. Prior to his admission to the RGH Mr Smith had been residing at the Elder Care Kirkholme Nursing Home in Goodwood (the nursing facility). He had been residing at that facility since April 2014. Mr Smith had experienced a number of illnesses throughout his life including type 2 diabetes, ischaemic heart disease and an anxiety disorder. In 2010 Mr Smith had been diagnosed as suffering from Post-Traumatic Stress Disorder for which he was being treated with mirtazapine. In 2013 the deceased had a number of voluntary admissions to Ward 17 of the RGH for cognitive impairment.

In 2014 staff at the RGH noted that the deceased was showing signs of dementia. As well, an episode of pneumonia was complicated by delirium. It was these health concerns that prompted Mr Smith’s movement from private accommodation into the nursing facility.

1.3. By the commencement of July 2014 nursing facility staff and RGH nursing staff entertained concerns that Mr Smith was experiencing significant confusion. His baseline cognitive impairment and multiple co-morbidities were taken into account, as was his delirium.

2. Reason for Inquest 2.1. On 10 July 2014 Mr Smith was placed on a Level 1 inpatient treatment order pursuant to section 21 of the Mental Health Act 2009. This would be the first of a series of inpatient treatment orders that were imposed upon Mr Smith at the RGH between 10 July 2014 and the day of his death. The most recent inpatient treatment order proximate to his death was a Level 2 inpatient treatment order imposed on 28 July 2014.

As at the day of his death Mr Smith was still subject to that Level 2 order. The detention pursuant to that order meant that Mr Smith’s death occurred while he was in custody as contemplated by the Coroners Act 2003. This had two legal consequences. Firstly, the death was required to be reported to the State Coroner and, secondly, an Inquest into the cause and circumstances of his death was mandatory. These are the findings of that mandatory Inquest.

3. The delay in reporting Mr Smith’s death 3.1. On 30 July 2014, the day of Mr Smith’s death, Dr Hamish Eske who at that time was a resident medical office at the RGH and who had been involved in Mr Smith’s care, issued a doctor’s certificate of cause of death pursuant to section 36 of the Births, Deaths and Marriage Registration Act 1996 (the BDM Act). On the same day Dr Eske also issued a certificate pursuant to section 12(2) of the Burial and Cremation Act 2013.

A second medical practitioner also signed a certificate necessary for cremation.

Dr Eske’s death certificate cited the cause of death as follows: 'A bi-lateral (word unintelligible) lobe pneumonia - organism unknown - 2 days; B worsening dementia - 6 weeks; C congestive cardiac failure - long term; D type 2 diabetes - long term.'

3.2. Apart from the above no other conditions are cited on the certificate. As this was a reportable death in custody, the death certificate and written notice of Mr Smith’s death should not have been given to Registrar of Births, Deaths and Marriages1. Instead, Mr Smith’s death should have been notified to the State Coroner pursuant to section 28(1) of the Coroners Act 2003. The obligation to notify the State Coroner of a death must be complied with by any person who becomes aware of a death that is or may be a reportable death. The obligation is an immediate obligation. Thus, the obligation in Mr Smith’s case was imposed upon any medical practitioner at the RGH who became aware of Mr Smith’s death and of his circumstances of detention as at the time of his death.

3.3. In the event Mr Smith was cremated on 12 August 2014. A notification to the State Coroner was not made until 2 September 2014. The belated notification of Mr Smith’s death in custody to the State Coroner meant, among other things, that the State Coroner was denied the opportunity for Mr Smith’s remains to be examined by way of an autopsy or for an independent opinion as to cause of death to be obtained and considered while Mr Smith’s remains were still available for autopsy if it became necessary in light of that opinion.

3.4. I accept that the failure to effect immediate notification of Mr Smith’s death to the State Coroner was the result of a lack of awareness on the part of RGH medical practitioners of the circumstances that rendered the death of Mr Smith a death in custody. In another section of these findings I will discuss the circumstances in which this failure occurred.

4. Mr Smith’s decline in health and detention 4.1. On 3 July 2014 Mr Smith was admitted as a voluntary patient to the RGH. In the lead-up to this admission staff from his nursing facility expressed concern that he had become more confused. The plan for this admission was to investigate whether Mr Smith was developing worsening dementia, or whether there were other issues such as delirium.

4.2. On 3 July 2014 Dr Christopher Veale examined Mr Smith and suggested that it was clear that Mr Smith was suffering from delirium.

1 Section 36(2)(b) of the BDM Act

4.3. Dr Veale indicates in his witness statement2 that he noticed a marked difference in Mr Smith compared with a previous presentation in 2013. Dr Veale indicates that Mr Smith did not recognise him, notwithstanding that he had treated him for a number of years. Mr Smith appeared to have forgotten that his wife had died. He was wandering around the ward in a very confused state and appeared to be looking for her.

Whereas prior to July 2014 Mr Smith had been very calm and amicable, during this admission he became quite agitated. His behaviour was markedly different. He was very disoriented and his memory was adversely diminished.

4.4. On 10 July 2014 Mr Smith was found to have left the ward and have wandered into the carpark. Out of concern that he would do this again, a Level 1 inpatient treatment order was imposed. On 11 July 2014 Dr Veale confirmed this order and requested Mr Smith’s transfer to a secure ward within the RGH. Dr Veale had this to say about Mr Smith’s presentation as of 11 July 2014: 'He was grossly confused, didn’t recognise me and didn’t understand the process. He did not appear depressed, anxious or suicidal. I recommended he be kept under close observation and remain with a nurse special, one on one.'

4.5. On 12 July 2014 a MET call was made in response to Mr Smith’s very low blood pressure and marked drowsiness. He was transferred to Ward 2 for medical treatment.

On 14 July 2014 it was noted that Mr Smith seemed to be suffering from a lower respiratory tract infection. On 17 July 2014 Dr Veale believed that there was some improvement in Mr Smith, although he was nowhere near baseline. Due to the fact that at that time there was no longer concern that Mr Smith might harm himself, and due to his improved level of cooperation, the inpatient treatment order was allowed to lapse.

He remained under close observation with a nurse special. The plan was to use the least restrictive means to care for Mr Smith.

4.6. On 19 July 2014 Mr Smith’s behaviour was such that there was concern that he was suffering from some other form of infection. On 22 July 2014 he was reviewed by Dr Devon Marshman. This doctor imposed a fresh Level 1 inpatient treatment order on Mr Smith. Later that day Dr Veale reviewed Mr Smith and confirmed the order. At that time Mr Smith was delirious and showed aggression towards staff. Dr Veale’s last consultation with Mr Smith was on 25 July 2014 when he suggested that Mr Smith be commenced on clonazepam to control his disturbed behaviour. Dr Veale recommended 2 Exhibit C6

one to two drops four times a day and titrated the dosage based on Mr Smith’s symptoms, reducing it to only one drop if he appeared adequately sedated. On 23 July 2014 Dr Richard Weeks, a senior staff specialist psychiatrist and psychogeriatrician at the RGH, assessed Mr Smith. In his witness statement he describes Mr Smith at that time as follows: 'His presentation was consistent with an active delirium where he was acutely confused, fluctuating between being agitated and aggressive, and being sedated at other times. There have been a number of code black incidents in response to significant episodes of aggression requiring team management and sometimes medication. He was frequently experiencing visual hallucinations, and reaching out for unseen objects which is often a feature of delirium.'

4.7. Dr Weeks recommended that Mr Smith’s dosage of risperidone be increased from 0.25mg daily to 0.25mg twice a day. On 28 July 2014 Dr Weeks reviewed Mr Smith for the purpose of considering whether a Level 2 inpatient treatment order should be imposed. At that time Mr Smith was still very unsettled, restless and slightly agitated.

He was achieving minimal sleep, was still acutely confused and was unable to speak coherently. Dr Weeks states that he was unable to interpret what Mr Smith was saying and Mr Smith was unable to answer any questions asked of him and spoke no intelligible words. Other matters taken into account in considering the imposition of a Level 2 order were that Mr Smith was ripping his bed clothes off, he was constantly and actively hallucinating as reported by his nurse special and that there were reports that he was agitated with poor oral intake. There was difficulty in achieving adequate fluid intake and sustenance.

4.8. Mr Smith had a moist cough for which a chest X-ray was undertaken. He had ongoing fluctuation in his mental state with intermittent agitation. Due to a tendency on Mr Smith’s part to remove the intravenous fluid line, he was occasionally soft shackled.

Dr Weeks said this: 'In my assessment Mr Smith was suffering from a persisting delirium of unknown origin, on-going poor oral intake, probable dehydration and possible respiratory tract infection.

On the basis of his symptoms that were presented, I decided to confirm the ITO and initiate a Level 2 ITO.’

4.9. Ms Barbara Fenech, who is Mr Smith's niece, and a Mr Robert Smith, who is a nephew of Mr Smith, had both been granted enduring power of attorney for Mr Smith. On 29 July 2014 both Ms Fenech and Mr Robert Smith had a discussion with Dr Eske concerning Mr Smith's ongoing treatment. As a result of this discussion agreement was

reached that active treatment should cease and that comfort care only should be instituted.

4.10. That of course was acted on and Mr Smith was given only comfort medications thereafter. Perhaps not unexpectedly he then deteriorated quite rapidly but remained under the care of a nurse special at all times. He was last seen alive by the allocated nurse at 5am on 30 July 2014 and was discovered by that same nurse to be deceased at approximately 6am.

5. The cause of Mr Smith’s death 5.1. I have already referred to the certificate of Dr Eske and the cause of death that is described within that document.

5.2. After Mr Smith’s death was notified to the State Coroner a review of his cause of death was sought from Forensic Science South Australia. The review was sought based upon Mr Smith’s clinical history. As I have already indicated, a post-mortem examination was not available due to the fact that by the time the State Coroner was notified of the death Mr Smith’s remains had already been cremated.

5.3. In the pathology review of Dr Iain McIntyre, as discussed with Dr John Gilbert who is a forensic pathologist3, Mr Smith’s cause of death is described as follows: 'Multi-organ failure due to general inanition in a man with ischaemic heart disease, chronic cardiac and renal failure and delirium.' The review added a rider that the contribution to Mr Smith’s death of sedation and shackling was uncertain. The review thus contained a recommendation that an autopsy be performed with toxicology studies. Of course, this recommendation could no longer be carried out due to Mr Smith’s cremation. In making this recommendation Forensic Science South Australia were unaware of the fact that Mr Smith’s remains were no longer available for the purposes of autopsy. The review also recommended the retention of ante-mortem blood samples. In the event no such samples were available for analysis.

5.4. In the light of the fact that Mr Smith remains were no longer available for autopsy, and due to the unavailability of ante-mortem blood samples, a supplementary report was 3 Exhibit C1a

sought from Dr Gilbert4. In that supplementary report Dr Gilbert acknowledges the unavailability of the remains for the purposes of autopsy and within the document sets out a revised cause of death as follows: 'Ia. multi organ failure due to inanition complicating Ib. physical and pharmacological restraint for delirium of uncertain cause II. ischaemic and hypertensive heart disease, cerebrovascular disease, chronic renal impairment, type 2 diabetes mellitus.'

5.5. This cause of death is expressed in death certificate format and when extrapolated would be expressed as follows; multi-organ failure due to inanition complicating physical and pharmacological restraint for delirium of uncertain cause with contributing ischaemic and hypertensive heart disease, cerebrovascular disease, chronic renal impairment, and type 2 diabetes mellitus.

5.6. As a result of Dr Gilbert being supplied with further records in relation to Mr Smith’s clinical history at the RGH, he provided a further pathology review5 in which he indicates that in his view the frequency of physical restraint was not great. In addition, having accessed Mr Smith’s OACIS medical records covering radiology and blood tests, it was now apparent to Dr Gilbert that physical and pharmacological restraint were not major contributing factors to Mr Smith’s inanition. Within the revision document Dr Gilbert expresses the cause of death as: 'Ia. multi-organ failure due to inanition complicating prolonged delirium of uncertain cause.

II. ischaemic and hypertensive heart disease with congestive cardiac failure, cerebrovascular disease, chronic renal impairment and type 2 diabetes mellitus.' which extrapolated is expressed as; multi-organ failure due to inanition complicating prolonged delirium of uncertain cause with contributing ischaemic and hypertensive heart disease with congestive cardiac failure, cerebrovascular disease, chronic renal impairment, and type 2 diabetes mellitus.

5.7. In his witness statement Dr Eske, who signed the original death certificate in error, deals with the apparent difference between the cause of death stated on his certificate and that ultimately stated within the pathology reviews conducted by Forensic Science South Australia. He explains that there is no material difference insofar as the 4 Exhibit C2a 5 Exhibit C2b

pneumonia that he cited in his death certificate had progressed to the multi-organ failure identified in the Forensic Science South Australia reviews. I understand and accept that explanation and have nothing further to add.

5.8. The confusion over the precision of Mr Smith’s cause of death could naturally have been avoided if, as Drs McIntyre and Gilbert originally recommended, a post-mortem examination comprising an autopsy could have been carried out in the first instance.

5.9. In any event, I find that the cause of death is as stated in Dr Gilbert’s revised supplementary pathology review6, namely multi-organ failure due to inanition complicating prolonged delirium of uncertain cause with contributing ischaemic and hypertensive heart disease with congestive cardiac failure, cerebrovascular disease, chronic renal impairment, and type 2 diabetes mellitus. I also find that any physical restraint and or pharmacological restraint were not major contributing factors in Mr Smith’s cause of death and had in any case been necessary in all of the circumstances.

6. Conclusions 6.1. Mr Smith’s death was investigated by Detective Brevet Sergeant Sherrie Modra of SAPOL Sturt CIB. I received into evidence her comprehensive report. The investigation does not reveal any circumstance surrounding Mr Smith’s death that is suspicious or otherwise questionable. Ms Modra’s report quotes with evident approval Ms Fenech’s conclusion as expressed in the latter’s own witness statement that the RGH tried to do their best to help Mr Smith and that all of his needs were met. In addition, Ms Fenech’s perception was that RGH clinical staff had tried very hard to find out what was wrong with Mr Smith and that they were trying to help him. She adds that over the years the RGH had always been very good at looking after Mr Smith and that she could trust the decisions that doctors made in relation to his care. I accept that evidence, as I do the evidence of Ms Modra.

6.2. I also find that at all times Mr Smith’s inpatient treatment orders were lawful and appropriate.

6 Exhibit C2b

7. The failure to notify Mr Smith’s death to the State Coroner 7.1. This issue was originally addressed in the statement of Dr Dirk Hoffman dated 13 March 20157. Dr Hoffman is a physician and senior staff specialist in internal medicine at the RGH. In this statement Dr Hoffman explains that his team assumed the care of Mr Smith on 28 July 2014 which was two days before he died. On assuming Mr Smith’s care Dr Hoffman’s team reviewed all his laboratory investigations and imaging studies. Naturally Mr Smith himself was examined. It was concluded that Mr Smith had a multitude of irreversible conditions including chronic kidney failure, heart failure and pulmonary infection. As well, he was refusing to eat or drink. A mutual conclusion was reached with his next of kin that it would be unreasonable to continue life sustaining measures including antibiotic treatment, intravenous hydration and nutrition. Therapy was therefore withdrawn and as seen earlier Mr Smith died in the early hours of 30 July 2014.

7.2. Dr Hoffman explains that at the time of Mr Smith’s transfer to his team’s care and at the time of his death, they were unaware that he was subject of the inpatient treatment order that had been imposed by a psychiatrist on the 28 July 2014 just prior to his transfer to Ward 2. Dr Hoffman explains that in the normal course of events a psychiatrist would be expected to withdraw such an order in Mr Smith’s circumstances.

When Mr Smith died on the morning of 30 July 2014 the intern who attended that morning was notified and certified life extinct. Dr Hoffman explains that death certificate was issued and that no-one thought to notify the Coroner because no-one considered him to be under an inpatient treatment order. Furthermore, this was a natural and expected death and there was no other reason for his death to be reported to the State Coroner in anyone’s view at that time.

7.3. In his statement Dr Hoffman states that the Electronic Patient Administration System (EPAS) had been implemented at the RGH three months prior to Mr Smith’s death.

Approximately 200 clinical entries were generated for Mr Smith during his three week admission. The entries are not distinctive in any way and all entries were of equal appearance and of equal apparent importance. Dr Hoffman states that this makes it difficult for staff to distinguish important from non-important information. When it ultimately became necessary to compile the notification to the State Coroner, 7 Exhibit C4

Dr Hoffman examined the approximate 200 entries on EPAS. Ultimately he located the impatient treatment order on the system. However, Dr Hoffman states there was nothing on the system to flag that the patient was being treated pursuant to an order.

The existence of the order would not have been apparent to any person who opened his file.

7.4. Dr Hoffman explained in his original statement that there were certain difficulties with the EPAS system. His understanding was that when EPAS was initiated all paper copies of records were closed and were meant to be have been scanned and added to EPAS. He identifies this as one of clinicians’ major concerns with EPAS. Instead of having paper records, Dr Hoffman explained that to find anything one has to go through hundreds of entries most of which are not relevant to what the clinician is searching for.

In Mr Smith’s case the ‘inherent problems of EPAS’ did not assist in avoiding the oversight of the need for Mr Smith’s death to be notified to the State Coroner.

Dr Hoffman goes on to say that information from ‘SA Health Executive’ was that more than 1500 submissions had been made by doctors regarding the short comings of EPAS and that many clinicians feel ‘that it is unsafe’. This witness statement represented Dr Hoffman’s views as of 2014.

7.5. In a further statement taken from Dr Hoffman dated 8 July 2016, which was taken in order to update Dr Hoffman’s views on EPAS, Dr Hoffman states that to his knowledge, at some point after July 2014, which was the month of Mr Smith’s death, a mechanism of ‘flagging’ inpatient treatment orders was established but that had since been removed during an upgrade to the system. Dr Hoffman explained that he has subsequently refamiliarised himself with EPAS and confirms that at present inpatient treatment orders are not flagged or highlighted on the system. He states ‘therefore in my personal opinion compared to July 2014 we are back to the previous state’ and suggests that the same oversight that had occurred with Mr Smith could happen again. Time will tell whether that prediction comes to pass.

7.6. Dr Hamish Eske was the medical practitioner who signed and issued Mr Smith’s death certificate. At the time with which this Inquest is concerned he was a resident medical officer on Dr Hoffman’s team. The statement of Dr Eske dated 8 July 2016 asserts that in his experience doctors’ awareness of the existence of an inpatient treatment order in respect of a patient can be influenced by various factors including the sufficiency of

handover procedures by other medical staff. He says that at times one might be told about the order and at other times one is not told. Dr Eske states that at no time does he recall being advised of the existence of the inpatient treatment order for Mr Smith.

He states: ‘Unfortunately, the Enterprise Patient Administration System (EPAS) did not assist in identifying the ITO. There is no flagging or highlighting within EPAS and the information can get buried with the other documents. Had there been a flagging system, I would have known at the time of Gordon’s death that he was on an ITO. If I had of been aware of this, I would have discussed the ITO with the psychiatry team and determined whether it should have been removed or remained, dependant on his condition’.

7.7. Dr Eske states that at the time he completed the doctor’s certificate as to cause of death he was not aware of the inpatient treatment order. I accept that evidence.

7.8. I accept the evidence of Drs Hoffman and Eske as contained in their various witness statements.

7.9. The failure to report Mr Smith’s death became the subject of a letter to the State Coroner dated 8 September 2014 from Professor Belinda Moyes who is the Chief Executive Officer of the Southern Adelaide Local Health Network8. The letter is exhibited to the statement of Professor Moyes. In her letter Professor Moyes apologised for the failure to report Mr Smith’s death in a timely fashion, adding that the mandatory obligation to notify the State Coroner of reportable deaths is taken very seriously and that staff would be educated in relation to this obligation. The letter explains the circumstances in which the failure to report Mr Smith’s death occurred. The explanation is in accordance with the other evidence that I have mentioned in that regard. In relation to the contribution of the shortcomings of EPAS it says this: ‘The recent adoption of an electronic medical record, EPAS at the RGH has necessitated significant and ongoing changes in work practices which would have, in part, contributed to the failure to recognise that the patient was under an ITO. The ITO was part of more than 200 electronic documents through which to search electronically.’ Professor Moyes went on to assert that this issue would be a focus for the Director of Medical Services to address in Southern Adelaide Local Health Network generally and the RGH specifically. It has been extensively reported that other public hospitals have 8 See Exhibit C10a

already adopted, or will be adopting, EPAS and hence the issue needs to be addressed within the public health system at large.

7.10. I draw this issue to the attention of the Minister for Health and the Chief Executive Officer of the Department of Health and Ageing.

Key Words: Death in Custody; Natural Causes; EPAS In witness whereof the said Coroner has hereunto set and subscribed his hand and Seal the 2nd day of August, 2016.

Deputy State Coroner Inquest Number 36/2016 (1526/2014)