Finding into death of Frank Edward Frood

IN THE CORONERS COURT OF VICTORIA AT MELBOURNE

Court Reference: COR 2012 4080

FINDING INTO DEATH WITHOUT INQUEST

Form 38 Rule 60(2)

Section 67 of the Coroners Act 2008

J, AUDREY JAMIESON, Coroner having investigated the death of FRANK EDWARD FROOD

without holding an inquest:

find that the identity of the deceased was FRANK EDWARD FROOD born 2 September 1965

and the death occurred on 28 September 2012

at Hampton Park VIC 3976

from:

1(@) BRONCHOPNEUMONIA ON A BACKGROUND OF METHADONE AND BENZODIAZEPINE USE

Pursuant to section 67(1) of the Coroners Act 2008, I make findings with respect to the following circumstances:

1. Mr Frank Edward Frood was 47 years of age at the time of his death. He lived in Hampton Park with his parents and was on a disability pension. Mr Frood was an asthmatic and had been

diagnosed with bipolar disorder. He had a history of illicit drug abuse, including heroin.

2: Prior to 10.00am on 28 September 2012, Mr Frood was located by his mother on the lounge room floor, curled up on his side. Mr Frood’s teenage son rolled him over and liquid came out of his mouth. Emergency services were called at approximately 9.55am. Attending paramedics were unable to resuscitate Mr Frood and declared him deceased. Paramedics moved Mr Frood to a bedroom, due to a number of family members being present and walking around the

premises. Police arrived shortly afterwards.

1 of 14

INVESTIGATIONS

Forensic pathology investigation

Dr Yeliena Baber, Forensic Pathologist at the Victorian Institute of Forensic Medicine, performed a full post mortem examination on the body of Mr Frood and referred to the Victoria Police Report of Death, Form 83. At autopsy, Dr Baber identified thick purulent material within Mr Frood’s main bronchi and within smaller airways. Histological examination confirmed

widespread bronchopneumonia within both of Mr Frood’s lungs.

Toxicological analysis of post mortem blood detected multiple drugs at low levels, including methadone,’ diazepam,’ oxazepam,’ tramadol,* olanzapine’ and methylamphetamine.° The C-

reactive protein’ was high at 148mg/L, confirming that infection was present.

Pp g Pp

Dr Baber reported to the Coroner that Mr Frood’s death was due to widespread bronchopneumonia, however the contribution of a respiratory depressant effect of methadone,

tramadol and two benzodiazepines in combination could not be excluded.

Police investigation

The circumstances of Mr Frood’s death have been the subject of investigation by Victoria Police on my behalf. Police obtained statements from Mr Frood’s mother Judith Frood and General

Practitioner at Frankston Healthcare Dr Andrew Taylor.

In the course of their investigation, police learned that Mr Frood had been a heroin user for at least 20 years. Mrs Frood reported from time to time, he would stop taking heroin and start taking methadone.

In a letter to Frankston Magistrates’ Court dated 9 May 2011, Dr Taylor had written that alongside a long standing opiate addiction, Mr Frood had coronary artery disease, mixed

valvular disease and mental illness including bipolar disease.

' Methadone is a synthetic narcotic analgesic and is used for the treatment of opioid dependency or for the treatment of severe pain,

? Diazepam is a sedative/hypnotic drug of the benzodiazepines class.

3 Oxazepam is a sedative/hypnotic drug of the benzodiazepine class.

• Tramadol is a narcotic analgesic used for the treatment of moderate to severe pain.

5 Olanzapine is indicated for the treatment of schizophrenia and related psychoses. It can also be used for mood stabilisation and as an anti-manic drug.

5 Methylamphetamine is a central nervous system stimulant structurally related to dexamphetamine.

7 The C-reactive protein is a marker of infection or inflammation.

2 of 14

9. Mr Frood’s mother would accompany him to collect methadone from the pharmacy and she

reported that prior to his death, he was trying to cease taking the drug.

10. Mrs Frood also stated that her son had endured a chest infection for months before he died, and that he had been on antibiotics for some time. Mr Frood was coughing up sputum and having difficulty breathing. He had been a severe asthmatic since he was a young child, and this

condition worsened during his chest infection.

11. Mrs Frood had observed that for two nights prior to his death, Mr Frood had been reluctant to

go to sleep. He had been scared to lie down because of his difficulty with breathing.

Coroners Prevention Unit investigation

12. The Coroners Prevention Unit (CPU),® also investigated the circumstances of Mr Frood’s death on my behalf, in particular in relation to concerns that he was prescribed methadone for opioid replacement therapy (ORT) to treat his longstanding opioid dependence, while suffering ongoing respiratory issues including asthma and chronic chest infection. I asked the CPU to investigate whether the management of the ORT was appropriate, given that Mr Frood died

from bronchopneumonia in a setting of methadone and benzodiazepine use.

Recent context of prescriptions

13. The evidence indicates that proximal to Mr Frood’s death, he was attending the Frankston Pharmacy two days per week for supervised methadone dosing, and was dispensed five methadone doses per week for unsupervised consumption. Mr Frood had a long history of opioid dependence, and was prescribed methadone in ORT to treat this dependence across a period of approximately 22 years. His most recent ORT prescribing clinicians were Dr Robert Weiss (November 2007 to October 2011) and Dr Andrew Taylor (October 2011 to September

2012. at Frankston Healthcare. Dr Taylor held a valid Schedule 8 permit to prescribe ORT methadone to Mr Frood at the time of his death.

14, In addition to methadone, Frankston Healthcare clinicians Dr Weiss, Dr Taylor and their colleague Dr Emad Tadros, regularly prescribed the benzodiazepine diazepam to Mr Frood. The

clinical indication repeatedly recorded in the Frankston Healthcare medical records for

8 The Coroners Prevention Unit (CPU) was established in 2008 to strengthen the prevention role of the coroner. The unit assists the coroner with research in matters related to public health and safety and in relation to the formulation of prevention recommendations, as well as assisting in monitoring and evaluating the effectiveness of the recommendations. The CPU comprises a team with training in medicine, nursing, law, public health and the social sciences.

3 of 14

prescribing the diazepam was “opiate addiction” (with the exception of 30 December 2011,

when Dr Weiss recorded the clinical indication as “anxiety — generalised”).

15. In the six months leading up to Mr Frood’s death, his Pharmaceutical Benefits Scheme (PBS) Patient Summary shows he was also dispensed diazepam from doctors at other clinics, including Dr Alan Lim, Dr Sarvanan Shanmugam, Dr Omid Khavari, Dr Shahroze Khan and Dr Thoung Sein at First Health Medical Centre in Hampton Park; Dr Martin Hill and Dr Jia Li at Stud Road Medical Centre in Dandenong; and Dr Peter Williams at Duff Street Medical Clinic in Cranbourne. The clinical records from these doctors were not obtained, therefore I am unaware of whether the clinical indication for their prescribing was “opiate addiction”, anxiety, or

another condition.

16. Together with his opioid dependence, Mr Frood had a long history of asthma and respiratory disease. In the five years leading up to his death, Dr Weiss and Dr Taylor diagnosed him at various times with acute bronchitis and other chest infection; they prescribed regular medication to treat his asthma as well as a range of antibiotics (including cephalexin, amoxicillin and roxithromycin). In addition, in the six months leading up to Mr Frood’s death, his PBS Patient Summary shows he was also dispensed antibiotics including amoxicillin and roxithromycin from doctors at other clinics, including Dr Wei GU at First Health Medical Centre and Dr Melissa Soars and Dr Martin Hill at Stud Road Medical Centre.

Methadone prescribing

17. The initial clinical issue I explored in my investigation was how Dr Weiss and Dr Taylor managed the risks entailed in prescribing methadone to a patient who also suffered significant respiratory disease. Methadone has strong depressive effects on the central nervous system, including respiratory depression. In a person with pre-existing respiratory disease, there is a risk that methadone might further — and fatally - compromise respiration. Benzodiazepines enhance

the respiratory depressant effects of methadone, thus increasing this risk.

18. I determined this risk was directly relevant to Mr Frood’s cause of death, as he was suffering a particularly severe chest infection proximal to death, to the point where his mother stated he was “scared to lie down because of how bad his breathing was”; and his medical cause of death was

bronchopneumonia on a background of methadone and benzodiazepine (diazepam) use.

19. My concern regarding methadone prescribing to a person who suffers significant respiratory

disease, is reflected both directly and indirectly in a range of methadone prescribing guidelines

4 of 14

and information. For example, the Consumer Medicine Information for Biodone Forte (the main

brand of methadone liquid used in ORT) warns that:

a. “Biodone Forte is not suitable for everyone. Your doctor will take a full assessment of your condition before prescribing Biodone Forte. You should not use Biodone Forte if you: [...] Have any type of breathing problems especially if you suffer from blue discoloration of the skin, or plenty of mucus in your airways [...or] are suffering from an

asthma attack.”?

20. The contraindications in the Product Information for Biodone Forte include:

a. “Like other opioids, methadone is contraindicated in patients with respiratory depression, especially in the presence of cyanosis and excessive bronchial secretions.

Methadone should not be given during an attack of bronchial asthma.”!°

21. One precaution for methadone administration listed in the Biodone Forte Product Information is that “the major side effect of methadone is respiratory depression”. The product information

also warns that:

a. “The general depressant effects of methadone may be enhanced by other centrally-acting agents such as alcohol, barbiturates, neuromuscular blocking agents, phenothiazines and

tranquillisers.”!!

22, During the period when Mr Frood was being prescribed ORT methadone at Frankston Healthcare, the Commonwealth Department of Health Clinical Guidelines and Procedures for the Use of Methadone in the Maintenance Treatment of Opioid Dependence (2003) were the primary reference for administering ORT in Australia, establishing the overarching clinical

framework within which individual states’ ORT policies were developed.

23. The Clinical Guidelines noted that some categories of patients “are not suitable for treatment with methadone”, including:

a. “Other contraindications identified by the manufacturers of methadone include severe respiratory depression, acute asthma, acute alcoholism, head injury and raised

intracranial pressure, ulcerative colitis, biliary and renal tract spasm, and patients

° McGaw Biomed Pty Ltd, “Biodone Forte methadone hydrochloride: consumer medicine information”, 24 November 1999 (updated August 2007).

© Biomed Australia Pty Ltd, “Biodone Forte product information”, amended 7 April 2015.

'! Biomed Australia Pty Ld, “Biodone Forte product information”, amended 7 April 2015.

5 of 14

receiving monoamine oxidase inhibitors or within 14 days of stopping such treatment. It

is recommended that specialist advice be sought in these cases.”!?

24. The Clinical Guidelines further emphasised that “particular caution should be exercised by

prescribers” when patients had certain clinical conditions including the following:

a, “Asthma and other respiratory conditions: in such patients even usual therapeutic doses of opioids may decrease the respiratory drive associated with increased airways

resistance.

b. Poor compliance: patients who exhibit poor compliance with treatment for major intercurrent illness such as asthma or diabetes pose a particular challenge in MMT

[methadone maintenance therapy].”!

25. In April 2014, approximately 18 months after Mr Frood’s death, updated National Guidelines for Medication-Assisted Treatment of Opioid Dependence were released. The guidance

regarding respiratory disease was:

a. “Mild asthma and emphysema are not contraindications to substitution treatment and changing the dose of substitute medication is generally not necessary, but it is appropriate to review other factors that might contribute to respiratory distress. If

concerned, seek specialist advice or referral,”'*

26. The 2014 National Guidelines do not provide any further information about appropriate management of ORT clients with respiratory disease, however the following advice regarding

methadone, buprenorphine and respiratory depression more generally:

a. “As a partial agonist, buprenorphine is a safer opioid than methadone with regard to the potential for over-sedation, respiratory depression and overdose. Hence, dose increases can be more rapid and, in general, most patients can achieve their target dose within two

to three days.”

2 Wenry-Edwards S, Gowing L, White J, et al, Clinical Guidelines and Procedures for the Use of Methadone in the Maintenance Treatment of Opioid Dependence, Commonwealth Department of Health, August 2003, p.6.

3 Henry-Edwards S, Gowing L, White J, et al, Clinical Guidelines and Procedures for the Use of Methadone in the Maintenance Treatment of Opioid Dependence, Commonwealth Department of Health, August 2003, p. 7.

Gowing L, Ali R, Dunlop A, et al, National Guidelines for Medication-Assisted Treatment of Opioid Dependence, Commonwealth Department of Health, April 2014, p.38.

'S Gowing L, Ali R, Dunlop A, et al, National Guidelines for Medication-Assisted Treatment of Opioid Dependence, Commonwealth Department of Health, April 2014, p.25.

6 of 14

27,

CO

At a state level, the Victorian Department of Human Services Policy for Maintenance Pharmacotherapy for Opioid Dependence (2006) was the relevant policy for ORT provision during the period when Mr Frood was being prescribed ORT methadone at Frankston Healthcare. The Policy did not contain any general guidance for methadone administration to ORT clients with respiratory disease. However, the Policy listed “unstable medical conditions (for example, decompensated cirrhosis, pneumonia)” as a contraindication to takeaway dosing." Additionally, the Policy noted that “concerns re: medical condition (severe liver / respiratory disease)” were a reason for restricting client access to only one to two non-consecutive

takeaway methadone doses per week.”

. A revised Department of Health (now Department of Health and Human Services) Policy for

Maintenance Pharmacotherapy for Opioid Dependence was released in 2013 and includes identical warnings to the 2006 Policy regarding respiratory disease and takeaway dosing,'® but

again does not contain any general guidance on ORT clients with respiratory disease.

Unsupervised methadone dispensing and role of benzodiazepines

In my investigation, the initial focus was on the low level of supervised methadone dispensing to Mr Frood (he accessed five unsupervised doses per week regularly over the 12 months leading up to his death) because I was concerned that it led to a missed opportunity for prevention; specifically, if Mr Frood had attended more regularly for methadone dispensing, his pharmacist may have noted the significant worsening in his respiratory disease and may have

contacted his treating clinician or directed him to appropriate treatment.

However, in the course of this investigation, I noted that there were other reasons to question Dr Taylor’s decision to allow Mr Frood regular access to five unsupervised methadone doses per week. In particular, Dr Taylor and his colleagues at Frankston Healthcare made multiple clinical notes proximal to Mr Frood’s death indicating that he attended inappropriately to seek the benzodiazepine diazepam. Dr Taylor additionally made a clinical note about Mr Frood

fraudulently modifying a referral to support his drug seeking. The 2013 Victorian Department of

‘6 Drugs and Poisons Regulation Group, Victorian Department of Human Services, Policy for Maintenance Pharmacotherapy for Opioid Dependence, 2006, p.22.

17 Drugs and Poisons Regulation Group, Victorian Department of Human Services, Policy for Maintenance Pharmacotherapy for Opioid Dependence, 2006, p.26,

18 See Drugs and Poisons Regulation, Victorian Department of Health, Policy for Maintenance Pharmacotherapy for Opioid Dependence, 2013, pp. 22, 26.

7 of 14

Health and Human Services Policy explicitly warns that unstable benzodiazepine use and

dependence are contraindications to unsupervised methadone dosing.

Dr Taylor’s explanation of his prescribing practice

31.1 directed the CPU to write to Dr Taylor and Dr Weiss, requesting statements regarding how

they managed the risks of prescribing methadone to a person suffering significant respiratory

disease. The Court received a response from Dr Taylor, dated 31 December 2015, but not from

Dr Weiss; I determined that as Dr Taylor was the responsible ORT clinician proximal to Mr

Frood’s death, his statement was sufficient to inform my investigation.

32.1 have annexed the statement questions as Attachment A to this finding, and Dr Taylor’s

statement as Attachment B. The following is a summary of his main points:

33. Dr Taylor provided a statement to the Court dated 31 December 2015 in which:

a.

Dr Taylor indicated his methadone prescribing to Mr Frood was clinically appropriate,

as was his management of Mr Frood’s lung disease.

Dr Taylor maintained that Mr Frood died because “he did not access care available for his chest infection”. The reasons for his not accessing care were poverty and mental illness. Dr Taylor proposed that the underlying cause of Mr Frood’s death could be formulated as “severe mental illness, drug addiction (likely related) and consequent self

neglect”.

Dr Taylor stated that opioid dependence and severe lung disease are “very frequent

comorbidities” and he has extensive experience in managing them together.

Dr Taylor stated he was unaware of any specific guidelines for treating opioid dependence among people who suffer chronic respiratory disease, and further indicated

that no such guidelines are needed.

Dr Taylor explained that he supported unsupervised methadone dosing for Mr Frood because unsupervised dosing is “vital to engagement” in treatment for drug dependence.

Dr Taylor did not believe the 2006 DHS Policy regarding severe respiratory disease and access to unsupervised dosing was applicable to Mr Frood, because “he did not have

opioid induced depression of respiratory drive.”

8 of 14

Prescription shopping

34. While the central focus of the CPU investigation was on how the Frankston Healthcare clinicians managed Mr Frood’s opioid dependence in combination with his serious lung disease, other issues were noted and documented in the course of the case review. Mr Frood’s propensity for prescription shopping for benzodiazepines was noted in particular, and the extra risk that they posed to his health, given their enhancement of the respiratory depressant effects of

methadone.

35. The PBS Patient Summary of PBS benefits paid for medications dispensed to Mr Frood in the two years leading up to his death, runs to 26 pages. A review of the Patient Summary for the 12 months leading up to Mr Frood’s death showed he was dispensed the following drugs of

dependence in addition to what was prescribed by Frankston Healthcare clinicians: a. At Stud Road Medical Centre in Dandenong: i. Codeine on one occasion from Dr Martin Hill ii. Diazepam on two occasions from Dr Jia Li iti. Diazepam on eight occasions from Dr Martin Hill iv, Diazepam on one occasion from Dr Suzette Meshreky v. Oxycodone on one occasion from Dr Martin Hill vi. Tramadol on three occasions from Dr Martin Hill vii. Tramadol on one occasion from Dr Suzette Meshreky b. At First Health Medical Centre in Hampton Park: i. Diazepam on two occasions from Dr Omid Khavari ii. Diazepam on five occasions from Dr Alan Lim iii. Diazepam on two occasions from Dr Saravanan Shanmugam iv. Oxycodone on one occasion from Dr Shahroze Khan v. Oxycodone on one occasion from Dr Thoung Sein vi. Oxycodone on six occasions from Dr Alan Lim c. At Duff Street Medical Clinic in Cranbounre:

i. Diazepam on two occasions from Dr Peter Williams

9 of 14

At Young Street Medical and Dental Centre in Frankston: i. Diazepam on one occasion from Dr Zoran Zec At Casey Superclinic in Berwick: i. Diazepam on one occasion from Dr Milan Katic ii. Diazepam on one occasion from Dr Mahbub Mazumder iii, Tramadol on one occasion from Dr Myint Maung At Marina Medical Centre in Patterson Lakes: i. Diazepam on one occasion from Dr Michael Croce At Dandenong Superclinic in Dandenong: i. Diazepam on one occasion from Dr Roger Berard ii. Temazepam on one occasion from Dr Mojtaba Sebti At Southern Cross Medical Centre in Hampton Park: i. Oxyocodone on one occasion from Dr Sanskruti Joshi At Hallam Family Practice in Hallam:

i. Tramadol on one occasion from Dr Ngoc Le

COMMENTS

Pursuant to section 67(3) of the Coroners Act 2008, I make the following comments connected

with the death:

1,

As Dr Taylor indicated in his statement, he has “much experience” in prescribing methadone to patients with comorbid opioid addiction and severe lung disease. His opinions that there are no inherent issues in prescribing methadone to patients with these comorbidities; that guidelines for methadone prescribing to patients with severe lung disease are not needed; and that severe lung disease should not mediate access to unsupervised dosing; are presumably based on this

extensive experience and associated clinical knowledge.

However, while acknowledging Dr Taylor’s experience, J must also acknowledge the various prescribing advice — for example in consumer medicine information and product information for methadone syrup; in the Commonwealth Department of Health Clinical Guidelines and

National Guidelines; and in the Victorian Department of Health and Human Services Policy —

10 of 14

which would not appear to be consistent with Dr Taylor’s opinions. I also note data from the Coroners Prevention Unit, which indicates that between 2000 and 2015 there were 52 deaths investigated by Victorian coroners where the cause of death included acute methadone toxicity in combination with pre-existing respiratory disease; in seven of these deaths asthma was

explicitly nominated as being a contributory factor.

3. Focusing on the Frankston Healthcare clinicians’ benzodiazepine prescribing independently of

the methadone prescribing, I noted the following concerns:

a. The Frankston Healthcare notes show that between June 2009 and Mr Frood’s September 2012 death, clinicians there prescribed diazepam to him continuously.

Continuous prescribing for extended periods (longer than six to eight weeks) is contraindicated in all but exceptional cases because it is associated with patients

developing benzodiazepine dependence.

b. According to the Frankston Healthcare notes, the diazepam was almost always prescribed to treat Mr Frood’s “opiate addiction”. “Opiate addiction” is not a Therapeutic Goods Administration approved clinical indication for diazepam, nor is it specified in the 2013 edition of the Australian Medicines Handbook. Therefore, presumably the diazepam was being prescribed off-label. However, off-label prescribing is not eligible for PBS subsidies, whereas the PBS Patient Summary shows that a PBS

benefit was claimed for dispensing the prescribed diazepam on each occasion.

c. According to the Frankston Healthcare clinical notes, in the six months leading up to Mr Frood’s death he was prescribed a Smg tablet of diazepam twice daily. Over the course of six months this would amount to a requirement for 360 tablets (two tablets per day for 180 days). However, during this period the Frankston Healthcare clinicians provided him scripts for 600 tablets, which was nearly twice this amount, while simultaneously

making clinical notes that he was clearly benzodiazepine dependent and drug seeking.

d. There is no evidence that any clinician at Frankston Healthcare contacted Drugs and Poisons Regulation at the Victorian Department of Health and Human Services to notify intention to supply a drug of dependence, diazepam, to a drug dependent person, Mr Frood, as required under Section 33 of the Drugs Poisons and Controlled Substances

Act 1981 (Vic).

4. The review of Mr Frood’s PBS Patient Summary of PBS benefits paid for medication dispensed to him in the 12 months leading up to his death provides further evidence that he had developed

il of 14

a serious benzodiazepine dependence and was going to great lengths to “prescription shop” to obtain diazepam. While neither oxycodone nor tramadol contributed to Mr Frood’s death, both drugs would have fed his opioid dependence at a time when it was being treated with

methadone.

I note that neither medical records not statements were sought from any of the listed prescribers in the course of my investigation. However, I concluded that this material was not necessary because Mr Food was clearly engaged in prescription shopping and at the time of his death, as is the case even today, there were no effective tools available for prescribing doctors to establish who else a patient had attended for scripts, other than relying on all doctors to make reports to

Drugs and Poisons Regulation in line with relevant legislation.

I note Coroners Prevention Unit data that indicates between 2009 and 2015 there have been an average of 376 overdose deaths per year in Victoria. The annual frequency of overdose deaths has increased every year since 2010, reaching 420 deaths in 2015. Of these deaths, pharmaceutical drugs consistently contributed in approximately 80 percent of the Victorian overdose deaths annually. Among pharmaceutical drugs, benzodiazepines were the most frequently contributing drug group. The 2009-2015 data shows that the dominant role of pharmaceutical drugs in Victorian overdose deaths has remained undiminished over time, despite recent safety-focused initiatives such as improved prescribing guidelines, drug rescheduling and reformulation of some drugs into purportedly safer preparations. I have annexed the Coroners Prevention Unit’s data summary relating to Victorian overdose deaths

from 2009 to 2015 as Attachment C.

There is thus an ongoing, urgent need for Victoria to implement a real time prescription monitoring (RTPM) system to achieve reductions in pharmaceutical drug related harms and deaths. An RTPM system will, for the first time, enable prescribers and dispensers to find out what drugs a presenting patient has been dispensed in what quantities, when and by whom. This will have enormous benefits in assisting clinicians to make informed prescribing and dispensing decisions, and to coordinate the care they provide to patients. In addition, an RTPM system will enable a range of other prevention-focused interventions alongside identifying prescription shoppers: for example, identifying doctors whose prescribing practices might be clinically suboptimal so they can be provided targeted education and providing automated warnings to

prescribers and dispensers regarding potential issues with drug interactions and drug quantities.

12 of 14

8, Inote that the implementation of an RTPM system is an undoubtedly complex endeavour, and

that progress appears to have been made, at least in planning. However, I also emphasise that every year since 2010, more Victorians have died from fatal overdoses involving pharmaceutical drugs, and that if Victoria continues to link its RTPM efforts with nationally coordinated initiatives then there could be ongoing significant delays causing the deaths to

continue unabated.

RECOMMENDATIONS

T have concluded that best clinical practice in methadone prescribing to drug dependent patients with severe lung disease may be an area where there are differing clinical opinions, and have determined that it might be most appropriately considered by the relevant authorities. Therefore, I recommend that the Victorian Department of Health and Human Services review the Policy for Maintenance Pharmacotherapy for Opioid Dependence (2013) to ensure it provides adequate and explicit guidance to clinicians on how to manage maintenance pharmacotherapy in

patients with asthma or other respiratory conditions.

. And I further recommend that the Commonwealth Department of Health review the National

Guidelines for Medication-Assisted Treatment of Opioid Dependence (2014) to ensure they provide adequate and explicit guidance to clinicians on how to manage maintenance

pharmacotherapy in patients with asthma or other respiratory conditions.

. Again, I determined that the question as to whether or not the Frankston Healthcare clinicians’

methadone and benzodiazepine prescribing to Mr Frood was consistent. with accepted clinical practice and extant legislation, would most appropriately be examined by relevant authorities.

Therefore I recommend that the Victorian Department of Health and Human Services direct Drugs and Poisons Regulation to review the Frankston Healthcare clinicians’ diazepam and methadone prescribing to Mr Frood, and determine whether the clinicians require any further traiming in maintenance pharmacotherapy, prescribing to drug-dependent patients, or the

obligations of prescribers under the Drugs Poisons and Controlled Substances Act 1981 (Vic).

. And I further recommend that the Australian Health Practitioner Regulation Agency review the

treatment provided (and particularly drugs prescribed) to Mr Frood at Frankston Healthcare and consider whether this treatment raises any issues relating to the conduct of the practitioners

involved.

Mr Frood’s death further reinforces the immediate need for a real-time prescription monitoring

system to assist doctors in their clinical decision-making around drug prescribing, which should

13 of 14

not await the involvement of all other states and territories. With this in mind, I recommend that the Victorian Department of Health and Human Services immediately proceed with implementing a real time prescription monitoring system in Victoria to tackle the ever-

increasing toll of pharmaceutical drug related deaths in the state.

FINDINGS

I find that the prescription of methadone and diazepam, in light of Mr Frood’s ongoing respiratory problems, may have exacerbated his decline in health. In the circumstances, I find that a real time prescription monitoring system may have assisted clinicians to understand the extent of Mr Frood’s prescription shopping for diazepam in the lead up to his death. However, on the evidence available

to me, I am unable to find that Mr Frood’s death was preventable.

I accept and adopt the medical cause of death as identified by Dr Yeliena Baber and find that Frank

Frood died from bronchopneumonia, on a background of methadone and benzodiazepine use.

Pursuant to section 73(1A) of the Coroners Act 2008, | order that this Finding be published on the internet.

I direct that a copy of this finding be provided to the following:

Mrs Judith Frood

Dr Andrew Taylor, General Practitioner

Dr Robert Weiss, General Practitioner

Victorian Department of Health and Human Services

Australian Health Practitioner Regulation Agency

Commonwealth Department of Health

Senior Constable Viktoria Maley

Signature:

AUDREY JAMIESON

Date: 4 April 2016

14 of 14

Attachment A

Questions put to Dr Andrew Taylor

Attachment A

Background

You and your colleague Dr Ronald Weiss treated Frank Frood’s opioid dependence with opioid replacement therapy using methadone. Additionally you treated Frank Frood for chronic respiratory disease including asthma and bronchitis.

Prescribing methadone to a person who suffers serious respiratory disease is recognised to entail some risk. For example, the Product Information for methadone syrup includes the following contraindication:

Like other opioids, methadone is contraindicated in patients with respiratory depression, especially in. the presence of cyanosis and excessive bronchial secretions. Methadone should not be given during an attack of bronchial asthma.

The 2003 Commonwealth Department of Health Clinical Guidelines and Procedures for the Use of Methadone in the Maintenance Treatment of Opioid Dependence, which were in place when you were prescribing methadone, warned that some categories of patients “are not suitable for treatment with methadone’, including:

Other contraindications identified by the manufacturers of methadone include severe respiratory depression, acute asthma, acute alcoholism, head injury and raised intracranial pressure, ulcerative colitis, biliary and renal tract spasm, and patients receiving monoamine oxidase inhibitors or within 14 days of stopping such treatment. it is recommended that specialist advice be sought in these cases

The more recent 2014 National Guidelines for Medicatior-Assisted Treatment of Optoid Dependence state that:

Mild asthma and emphysema are not contraindications to substitution treatment and changing the dose of substitute medication is generally not necessary, but it is appropriate to review other factors that might contribute to respiratory distress. if concerned, seek specialist advice or referral

Coroner Jamieson is seeking answers to the following questions, to gain an understanding of how you negotiated the potential tension between prescribing methadone for opioid dependence and treating Frank Frood’s respiratory disease.

Questions

1, Were you concerned about prescribing methadone to Frank Frood in circumstances where he aiso suffered respiratory disease?

2. Were you ever concerned that Frank Frood’s respiratory disease was sufficiently serious to contraindicate methadone prescribing?

3. Did you put in place any measures to manage the risk of taking methadone in a setting of respiratory disease? In particular, did you have any strategies in place to manage or monitor the effects of methadone if Frank Frood experienced a clinical worsening of his respiratory disease?

4, Are you aware of any practice guidelines or other resources for treating opioid dependence among people who suffer chronic respiratory disease? Do you think there is aneed for detailed guidelines? :

5. Unsupervised or ‘takeaway’ dosing is believed to encourage client engagement in opioid replacement therapy, but supervised dosing creates opportunities for the pharmacist to

Attachment A Lof2

monitor the client’s day-to-day presentation. What regard did you have to these considerations when determining Frank Frood should have access to five unsupervised methadone doses per week?

6. The DHHS Policy for Maintenance Pharmacotherapy for Opiotd Dependence (2006, revised 2013) lists severe respiratory disease as a reason for restricting client access to unsupervised dosing. What regard did you have to this when determining Frank Frood should have access to five unsupervised methadone doses per week?

7, In the days immediately preceding Frank Frood’s death, evidence suggests he was too scared even to lie down because he was in such severe respiratory distress. Were you aware of this? If so, did you take any steps to manage this? If not, what steps might you have taken had somebody (for example a family member of pharmacist} alerted you to his clinical state?

8. If you have any additional comments for Coroner Jamieson on methadone prescribing to a patient who suffers respiratory disease, these would be welcome.

Attachment A 2 of 2

Attachment B

Dr Andrew Taylor’s Statement

Mai /02/2011 3:61 4397768014 4 PAGE 91/04

0397700014

“JAN 2018 , SIN” @ Frankston Healthcare <3 CORONER'S” 4/42 Young Btreet

PRE ORS Pad PE Pd Pad OE Pad ed PSA ME OMENS OE EL Pd ab rab at Pd fe Pay nd Pa Pr ad at Pad as Pt Ah Ma) tt Pt Pat Prat Pot Pat Pad Oa Pf Pd tt Pad et

Ph 97700023 Fax 97700014 31/12/2015

Coroners Court Att Jeremy Dwyer

Re: Mr Frank rood’ Reference: COR 20412 004080 :

itis over 3 years since Frank Frood, DOB 2/9/1966, died. |! sent a letter, through the investigating policeman, at about that time. It is unfortunate that my merory of Frank, and the events around his demise , are not as clear now as they once were. . a

Lam informed that Frank died as resuit of a chest infection. jFrom your letter it would seem that you believe that asthma was actually the underlying cause of the infectipn, and death. Suffering a chest infection may be a consequence of him having asthma, but in fact is significantly different in that death from infection, as opposed to asthma, is not usually immediate/ rapid. i .

| suggest that the underlying cause of death was, in my opinion, severe mental illness, drug addiction ( likely related} and consequent self neglect. He did not seek treatinent for a chest infectrion which ‘was so severe that he could not fie flat for a number of days". :

Mr Frood was extremely grandiose when | last saw him, anf showing profound cognitive dissonance. tn hindsight it is possible he was methamphetamine affected}! would be interested in toxicology results.

_ Inresponse to your questions . ; a

4. No, Not at all. Opiate addiction and severe lung disease pre very frequent comdrbidities. | have much experience in prescribing for such dual diagnosis patients. poth diseases are readily and easily treated with subsidised medications. However it is my experience that patients on methadone (and unemployed) can rarely afford all the chest medications that they require. Mefhadone at $150+ pet. month, plus rent, leaves little money for other essentials, Not prescribing methadong leaves the patient to die.

5. No. This question suggests alternative methods of treatitg opiate addiction exist, They really don't in this State at this time. oe :

6. Both respiratory and addiction conditions were treated. What was not well treated and cannot be within current frameworks is co-morbid poverty, and evolving/ wofsening psychiatric illness ‘

7. No, | am not aware of any guidelines and see no need fdr them. Suboxone may be a safer alternative for patients with oplate addiction and respiratory dissease. Un rtunately most patients do not find that Suboxone and methadone are equally efficacious and most have a stfong preference for one or the other. .

Mr Frood died because he did not access care available fot his chest infection.

5, Take aways, or independent self medication and self treatment of drug addiction is vital to engagement as is recognised. Take away dosing serves to free people up ip be with family and werk, and to minimise contact with "old friends” ( people on same drug treatment }. { hypothetically ask you if it would it be sensible to suggest that people with severe asthma attend the phartnacist daily and be-required to use their preventative medications in front of same ? That is what is being suggested, | ' 6, | presumed severe respiratory disease to allude to those patients whereby gpioid induced depression of respiratory drive or cough reflex to be both likely and potentially lethal, If Mr Frood: could not fie flat fora number of nights due his breathlessness he did not have opioid Induced depressidn of respiratory drive!

7, Mr Erood, nor hig family and friends, did not contact me fegarding his terminal illness. Itis clear that Mr Frood required hospital treatment. 4

8. Making medications like methadone free to patients Is aldesperately required intervention that will save lives and reduce drug dealing. .

Yours sincerely, Dr Andrew J ay

037763K;

Attachment C Coroners Prevention Unit Data Summary

Re: Victorian Overdose Deaths, 2009-2015

Coroners Court of Victoria

Coroners Prevention Unit Data Summary

Author: Jeremy Dwyer, Coroners Prevention Unit Date: 16 March 2016 Re: Victorian overdose deaths, 2009-2015

1, Background

This data summary provides an introduction to the drug types, drug groups and individual drugs that contributed to overdose deaths in Victoria 2009-2015. Data was drawn from the Victorian Overdose Deaths Register (the Register} created by the Coroners Prevention Unit (CPU); the composition of the Register is described in Attachment A to the data summary.

2. Annual frequency of overdose deaths, Victoria 2009-2015

Table 1 shows that in the period 2009-2015 the annual frequency of Victorian overdose deaths ranged between 342 deaths (in 2010} and 420 deaths {in 2015) with an average of 376 deaths per year. The annual frequency of deaths increased each year between 2010 and 2015. Approximately 30% of overdose death each year involved a single drug, and 70% involved the combined toxic effects of multiple (two or more} drugs.

Table 1: Annual frequency of overdose deaths, Victoria 2009-2015

Year : 2009 | 2010 2011 2012 2013 2014 2015 All overdose deaths 379° 342 362 367 380 387 420 Single drug deaths 127 122 133 114 118 101 121 Multiple drug deaths 252 220 229 253 262 286 299

3. Overdose deaths by contributing drug types

Table 2 shows the annual frequency of Victorian overdose deaths involving pharmaceutical drugs, illegal drugs and alcohol. Data is presented across all overdose deaths, then disaggregated into single and multiple drug deaths.

Pharmaceutical drugs were overall the most frequent contributors to overdose deaths in Victoria during 2009-2015; they consistently played a role in around 80% of deaths each year. Their contribution was more prevalent in multiple drug deaths than single drug deaths; on average more than 95% of multiple drug deaths each year involved one or more contributing pharmaceutical drugs.

illegal drugs consistently played a role in approximately 42% of annual overdose deaths in the period 2009-2014; this increased to just over half of overdose deaths in 2015 (217 of 420, 51.7%}. The increase in illegal drug contribution from 164 deaths in 2014 to 217 deaths in 2015, appears to account for the overall increase in the frequency of overdose deaths between 2014 and 2015.

CPU COURTS IN CONFIDENCE 1/10

Alcohol consistently played a role in approximately 24% of annual overdose deaths across the period 2009-2015. Alcohol contribution was more prevalent in multiple drug deaths {an average 28% annually) than single drug deaths (an average 16% annuaily}.

Table 2: Annual frequency of overdose deaths by contributing drug types, Victoria 2009-2014

Drug types 2009 2010 2011 2012 2013 2014 2015 All overdose deaths 379 342 362 367 380 387 420 Pharmaceutical 295 266 275 306 313 316 330 legal 147 149 153 133 166 164 217 Alcohol! 94 85 88 80 94 94 97 Single drug deaths 127 122 133 114 118 101 t2!

Pharmaceutical 58 53 58 60 55 49 46 Illegal 45 48 56 35 St 34 56 Alcoho! 24 21 19 19 12 18 19 Multiple drug deaths 252 220 229 253 262 286 299 Pharmaceutical 237 213 217 246 258 267 284 legal 102 101 97 98 15 130 161 Alcohoi 70 64 69 6} 82 76 78

4, Interactions between contributing drug types

Table 3 shows the combinations of drug types that contributed in Victorian overdose deaths 2009-2015. Overall, 40.2% of overdose deaths involved pharmaceutical drugs only, compared to 13.8% of overdose deaths that involved only illegal drugs {and 5.0% of deaths that were alcohol only}. A further 22.0% involved pharmaceutical drugs in combination with illegal drugs, and 11.9% involved pharmaceutical drugs in combination with alcohol.

Table 3: Overall frequency and proportion of overdose deaths by combinations of

contributing drug types, Victoria 2009-2015

“Combinations of — Single dug Multipledtug All overdose contributing drug pes se. oF Total overdose deaths 836 100.0 1801 100.0 2637 100.0

Pharma only 379 45.3 682 37.9 1061 40.2 Pharma + illegal 0 0.0 581 32.3 581 22.0 illegal only 325 38.9 38 2.1 363 13.8 Pharma + alcohol 0 0.0 315 17.5 315 ing Pharma + illegal + alcohol 0 0.0 144 8.0 144 5.5 Alcohol] only 132 15.8 0 0.0 132 5.0 iHlegal + alcohol 0 0.0 41 2.3 41 1.6

5. Overdose deaths by contributing pharmaceutical drug groups

Pharmaceutical drugs were disaggregated into drug groups using a modified version of the Drug Abuse Warning Network (DAWN) level 2 drug categories classification system (the main modifications were that the ‘analgesics’ category was split into opioid and noropioid analgesics, and the ‘anxiolytics’ category was split into benzodiazepine and non-benzodiazepine anxiolytics).

CPU COURTS IN CONFIDENCE 2/10

Table 4 shows the annual frequency of Victorian overdose deaths 2009-2015 by contributing drug groups, with illegal drugs and alcohol included for context. Overall benzodiazepines were the most frequent contributing drug group, contributing in an annual average of 51.3% of all overdose deaths. The next most frequent pharmaceutical drug groups were opioid analgesics (an annual average of 48.5% of all overdose deaths}, antidepressants (annual average 34.0%) and antipsychotics {annual average 19.2%} :

Table 4: Most frequent contributing drug groups to overdose deaths, Victoria 2009-2015

Drug groups 2009 2010 2011 2012 2013 2014 2015 Ail overdose deaths 379 342 362 367 380 387 420 Benzodiazepines 160 169 180 199 212 215 220 Opioid anaigesics 177 145 183 212 192 186 183 legal! drugs 147 149 153 133 166 164 217 Antidepressants 122 106 101 142 134 144 151 Alcohol 94 85 88 80 94 94 97 Antipsychotics 63 64 65 78 75 81 82 Non-benzo anxiolytics 35 28 33 38 56 48 56 Non-opioid anaigesics 26 25 30 52 41 49 43 Anticonvulsants 18 14 13 10 37 45 44

6. Overdose deaths by individual! contributing drugs

Table 5 shows the most frequent individual contributing drugs to Victorian overdose deaths 2009-2015 within each of the most frequent contributing drug groups.

Table 5: Most frequent contributing individual drugs by drug groups in overdose deaths,

Victoria 2009-2015

Year oS ‘ 2009 2010 201 1 2012. 2013. 2014 2015 All Benzodiazepines 160 169 180 199 212 215 220 Diazepam 104 109 124 133 164 169 176 Alprazolam 62 56 43 57 45 28 21 Temazepam 28 22 48 35 22 20 25 Oxazepam 18 19 44 41 17 19 28 Nitrazepam 17 16 i 24 26 13 17 Clonazepam 7 9 14 18 19 25 31 All opioid analgesics 177 145 183 212 192 186 183 Codeine 76 57 66 93 71 54 60 Methadone 50 55 72 75 70 67 64 Oxycodone 41 39 46 46 él 46 53 Tramadol! 22 9 5 18 24 23 31 Morphine 22 in| 10 13 7 12 7 Fentanyl 1 2 5 17 V1 V1 20 Buprenorphine 3 4 14 4 3 7 4 Propoxyphene 10 10 7 3 1 2 0

(Table 5 continued over page}

CPU COURTS IN CONFIDENCE 3/10

(Tabie 5 continued from previous page)

Year : : : 2009. 2010 2011 Z0l2 2013 2014 2015 All illegal drugs 147 149 153 133 166 164 217 Heroin 127 139 129 Itt 132 ‘137 168 Methamphetamine 23 14 29 36 51 53 67 Amphetamine 4 4 19 i 10 8 Cocaine 7 1 2 4 5 7 15

MDMA 5 1 I 1 3 4 GHB 3 i¢] 3 I (@] j All antidepressants 122 106 101 142 134 144 151 Mirtazapine 23 21 23 26 30 29 47 Amitriptyline 24 26 22 32 25 41 26 Citalopram 17 22 21 25 24 25 22 Venlafaxine 25 12 16 15 20 19 10 Fluoxetine 8 9 8 14 10 7 12 Duloxetine 3 5 7 1s it 12 12 Sertraline 6 6 4 12 13 9 il Desveniafaxine (e] 1 3 6 8 11 13 Doxepin 7 6 6 8 6 4 4 Aicohoi 94 85 88 80 94 94 97 All antipsychotics 63 64 65 78 75 8! 82 Quetiapine 28 37 . 34 41 41 48 45 Olanzapine 19 18 17 22 1s 21 28 Risperidone 6 3 i 8 10 7 9 Chlorpromazine 5 2 4 10 6 3 4 Zuclopenthixoil 5 4 4 6 3 3 4 Clozapine 5 5 0 4 6 2 2 Amisuipride i 3 6 3 2 4 1 All non-benzo anxiolytics 35 28 33 38 56 48 56 Doxylamine 13 16 it 21 23 13 14 Zopicione : 6 3 6 13 14 ti 16 Pentobarbitone 4 5 VW 1 8 1s 17 Zolpidem 11 3 5 5 6 10 Diphenhydramine 5 1 2 7 5 4 All non-opioid analgesics 26 25 30 52 41 49 43 Paracetamol 23 21 24 50 39 37 39 Ibuprofen 5 5 4 5 2 7 5 Naproxen I 2 2 2 1

(Table 5 continued over page)

CPU COURTS IN CONFIDENCE 4/10

{Table 5 continued from previous page}

Drugs Q 2009 2010 2011 2012 2013 2014 2015 All anticonvulsants 18 14 13 10 37 45 44 Pregabalin 0 ie) 0 0 17 27 31 Vailproic Acid 9 9 5 6 13 9 7 Carbamazepine 7 3 6 1 3 3 i Lamotrigine 1 2 i 2 2 2 1 Levetiracetam i} i} i i} 2 1 6 Topiramate i} i} 0 0 4 4 2

To summarise Table 5, the 10 overall most frequent contributing individual drugs to Victorian overdose deaths 2009-2015 were:

• Diazepam, a benzodiazepine {which contributed in 979 overdose deaths across the period).

• Heroin, an illegal drug (943 deaths}

~ Alcohol (632 deaths)

• Codeine, an opioid analgesic (477 deaths}

• Methadone, an opioid analgesic (453 deaths}

• Oxycodone, an opioid analgesic (332 deaths}

• Alprazolam, a benzodiazepine (312 deaths)

• Quetiapine, an antipsychotic {274 deaths}

• Methamphetamine, an illegal drug (273 deaths)

• Paracetamol, a non-opioid analgesic (233 deaths}

6. Overdose deaths by location of fatal incident

The Register includes detailed coded information on locations (street address, suburb and local government area} where fatal overdose incidents occur, and where deceased usually reside. For the purpose of this report, the CPU extracted basic overdose death frequencies (overall overdoses and subsets involving pharmaceutical drugs, illegal drugs and alcohol} by local government area {LGA} for the period 20092015.

The CPU established the population of each LGA as at 2011 according to the Australian Bureau of Statistics (ABS), and then calculated average annual overdose death rate per 100,000 population for each LGA using these steps:

(a) Overall frequency of overdose deaths in LGA for 2009-2015, -

(b) divided by 2011 population of LGA, - {c} multiplied by 100,000 (to yield seven-year overdose rate per 100,000), -

(d) divided by seven (to yield average annual overdose rate per 100,000 for the period 2009-2015}.

Table 6(a) shows the overall frequéncy of overdose deaths and average annual overdose rate per 100,000 population across the period 2009-2015, for all metropolitan Victorian LGAs. Across all metropolitan LGAs the average annual rate of overdose death per 100,000 population was 6.9 deaths.

CPU COURTS IN CONFIDENCE 5/10

Table 6a: Overal! frequency and average annual rate per 100,000 population of overdose

deaths in metropolitan local government areas, Victoria 2009-2015

| deaths drugs drugs Rate All metropolitan deaths 1985 1533 945 473 4,108,837 6.9 Banyule 48 39 20 9 122,983 5.6 Bayside 31 26 8 6 96,119 4.6 Boroondara 6t 50 23 14 167,062 5.2 Brimbank 103 74 62 17 191,496 77 Cardinia 27 19 9 7 75,831 5.4 Casey 69 55 22 12 261,282 3.8 Darebin 85 67 38 7 142,942 8.5 Frankston 101 87 42 26 130,350 TEI Glen Eira 44 28 19 13 137,152 4.6 Greater Dandenong 92 70 46 25 142,167 9.2 Hobsons Bay 35 27 19 6 87,395 5.7 Hume 56 44 29 8 174,290 4.6 Kingston 46 38 19 8 148,304 4.4 Knox 64 54 25 14 154,625 5.9 Manningham 27 21 11 4 116,750 3.3 Maribyrnong 73 52 47 18 75,154 13.9 Maroondah 60 46 15 19 107,323 8.0 Melbourne 129 96 89 31 100,240 18.4 Melton 29 24 10 3 112,643 3.7 Monash 59 44 25 12 177,345 4.8 Moonee Valley 40 25 27 12 112,180 5.1 Moreland 57 43 25 19 154,247 5.3 Mornington Peninsula 78 70 1S 20 149,271 7.5 Nittumbik a 8 4 t 62,716 25 Port Phillip 132 94 79 38 97,276 19.4 Stonnington 54 43 29 16 98,853 7.8 Whitehorse 82 69 33 24 157,538 7.4 Whittlesea 47 39 23 6 160,800 4.2 Wyndham 58 45 25 6 166,699 5.0 Yarra 131 88 96 4] 78,903 23.7 Yarra Ranges 56 48 11 2t 148,901 5.4

The highest frequency of overdose deaths among metropolitan LGAs occurred in Port Phillip (132 deaths between 2009-2012} followed by Yarra (131 deaths) and Melbourne (129 deaths}. These three LGAs also had the highest average annual rates per 100,000 population: Yarra (23.7 deaths per 100,000 population per year on average} then Port Phillip (19.4) then Melbourne (18.4).

in analysing regional Victorian LGAs, the CPU notes there are recognised issues with calculating rates where there are low frequencies of deaths among small populations. These issues were clearly present when average annual rates were calculated for some regional LGAs. Therefore, the CPU determined to tabulate only the results for regional LGAs where at least seven overdose deaths occurred across the seven-year period 2009-2015. This ted to the exclusion of 23 regional LGAs from Table 6b. However, the boided row “All regional deaths” in Table 6b shows the

CPU COURTS IN CONFIDENCE 6/10

frequencies and average annual rates aggregated across all regional LGAs, not just the LGAs included in the table.

Table 6(b) shows the overall frequency of overdose deaths and average annual overdose rate per 100,000 population across the period 2009-2015, for regional Victorian LGAs where at least seven overdose deaths occurred. Across all regional LGAs the average annual rate of overdose death per 100,000 population was 6.5 deaths, which was very close to the overall metropolitan rate (6.9).

Table 6b: Overall frequency and average annual rate per 100,000 population of overdose deaths in regional local government areas where at least seven overdose deaths occurred,

Victoria 2009-2015.

. : : deaths drugs drugs coe Rate All regional deaths 643 562 177 156 1,422,355 6.5 Baliarat 34 29 14 10 95,185 5.1 Bass Coast 17 12 5 4 30,233 8.0 Baw Baw 21 19 2 3 43,389 6.9 Campaspe 12 12 1 1 36,855 47 Colac Otway 9 8 3 2 20,799 6.2 East Gippsland 22 20 8 8 * 42,826 7.3 Glenelg 18 14 4 5 19,848 13.0 Greater Bendigo 58 49 12 10. 101,995 8.1 Greater Geelong Tit 94 40 32 215,837 7.3 Greater Shepparton 38 34 tt 8 61,744 8.8 Hepburn 8 6 3 1 14,629 7.8 Horsham 8 8 I 1 19,523 5.9 Latrobe 50 49 13 8 73,788 9.7 Macedon Ranges 12 9 3 2 42,883 4.0 Mildura 25 22 4 8 51,822 69 Mitchell 17 5 7 2 35,105 6.9 Moira 8 7 2 2 28,406 4.0 Moorabool 8 6 2 1 28,670 4.0 Northern Grampians 8 8 i 2 12,054 9.5 South Gippsland 9 9 1 27,512 47 Wangaratta 12 12 3 5 27,212 63 Warrnambool 12 9 2 1 32,667 5.2 Wellington 17 15 7 6 42,068 5.8 Wodonga 2t 20 3 3 36,025 8.3

The highest frequency of regional overdose deaths occurred in Greater Geelong {111 deaths), but the highest average annual rate was in Glenelg (13.0 deaths per 100,000 population per year).

CPU COURTS IN CONFIDENCE 7/10

Attachment A

The Coroners Prevention Unit (CPU) created the Victorian Overdose Deaths Register (the Register) to support coronial investigations; this attachment describes the case identification and coding process used to populate the register.

A.| Definitions

The CPU definition of the term ‘drug’ is largely consistent with the Australian Bureau of Statistics (ABS) definition, encompassing substances that “may be used for medicinal or therapeutic purposes, or to produce a psychoactive effect’.! Like the ABS, the CPU excludes tobacco and volatile solvents such as petro! and toluene from its definition of a drug. However, the CPU considers alcohol to be a drug, whereas it is excluded under the ABS definition.

An overdose death is any death in which the acute toxic effects of one or more drugs played a causal or contributory role.

A.2Z Inclusion and exclusion criteria

The CPU includes as relevant any death where the expert death investigators (coroner, forensic pathologist and forensic toxicologist) advise the acute toxic effects of one or more drugs played a causal or contributory role.

The following types of deaths are included:

• Deaths caused by drug overdose in combination with an underlying natural disease process; for example “methamphetamine toxicity in a setting of cardiomegaly”,or “acute alcohol toxicity in an obese person”.

• Deaths caused by drug overdose in combination with another (non-overdose} mechanism; for example “effects of hypothermia and combined drug toxicity”, or “inhalation of motor vehicle exhaust in a person with fatally toxic levels of oxycodone and diazepam”.

The following types of deaths are excluded:

• Deaths that resulted from allergic reactions to drugs {allergic reaction and overdose are very different mechanisms of death, requiring different preventative countermeasures).

• Deaths associated with the behavioural effects of drugs, for example a motor vehicle collision or falling off a pier and drowning while intoxicated.

• Deaths linked to chronic drug abuse in the absence of an acute toxic effect, for example a death from liver disease brought about by chronic alcohol use.

• Deaths linked to a means of drug-taking rather than the toxic effects of the drug, for example foreign body granulomatosis caused by crushing and injecting tablets that contain insoluble binding agents. Note however that a death from foreign body granulomatosis in combination with acute drug toxicity would be included as relevant.

• Suspected overdose deaths where specific contributing could not be identified (for example because appropriate specimens could not be obtained for forensic toxicological examination).

1 Australian Bureau of Statistics, “Drug-induced deaths: a quide to ABS causes of death data”, 8 August 2002, p.2.

CPU COURTS IN CONFIDENCE 8/10

• Suspected overdose deaths where the cause of death was unascertained following coronial investigation.

A.3 Case identification

The CPU identifies potentially relevant deaths for inclusion in the Register through searches of the CCOV's case management system and death surveillance database, as well as the National Coronial Information System. The autopsy report, toxicology report and {for closed cases} finding in each potentially relevant death are reviewed to determine whether the death meets the inclusion criteria.

A.4 Coding

For each death that meets the inclusion criteria, the CPU uploads the following information into the Register (a custom Access database]: the local case number, deceased age and sex, cause of death, intent, and the date the death was reported.

For each death the CPU then uses the register interface to record each individual drug that the expert death investigators determined had played a causal or contributory role in the overdose. The coding rules are:

• Where the expert death investigators explicitly nominate the individual contributing drugs (for example, “an overdose of morphine in combination with diazepam’}, these are coded as contributory in the death.

• Where the expert death investigators nominate drug types or groups rather than individual drugs (for example, “an overdose of opioids and benzodiazepines”}, the toxicology report is reviewed and ail specific drugs detected that belong to that drug type or group are coded as contributory.

• Where the expert death investigators do not nominate contributing drugs at all, the toxicology report is reviewed and every specific drug detected is coded as contributory.

In most cases, the expert death investigators concur as to contributing drugs. The main exception is with respect to metabolites, where on occasion the forensic pathologist and coroner nominate a contributing drug that the toxicologist advises was actually a metabolite of another drug (recurring examples are risperidone and hydroxyrisperidone, which are both are drugs in their own right but the latter is also a metabolite of the former; and diazepam, temazepam and oxazepam, where the latter two drugs can also be present as metabolites of the diazepam). In such cases, the CPU follows the toxicologist's advice and codes only the contributing drugs, not the metabolites.

The CPU also uses special coding rules for determining the drug source where morphine is a contributing drug, because morphine can be present as a metabolite of heroin, as a metabolite of codeine, or as a drug in its own right. The following hierarchy of coding rules is applied:

• If 6monoacety! morphine {a distinctive metabolite of heroin} is detected in postmortem blood or urine, morphine is assumed to be present as a metabolite of heroin.

• If there is evidence of a high level of codeine in comparison to morphine in blood {as a rule of thumb, if the codeine level is more than six times higher than the morphine level), or if there is codeine in blood but morphine is only detected in urine, the morphine is assumed to be present as a metabolite of codeine.

• if there is evidence the deceased was prescribed morphine, or if morphinecontaining medications were found at the scene of death, the morphine is assumed to be present as a drug in its own right.

CPU COURTS IN CONFIDENCE 9/10

• If none of the above three conditions are met, the morphine is coded as being of unknown source. Morphine of unknown source is aggr

In the overwhelming majority (more than 90%) of deaths involving morphine of unknown origin, the circumstantial evidence including statements of witnesses and deceased drug use history indicates that the likely morphine source was heroin. For this reason, deaths involving morphine of unknown source are classified as heroin deaths for analysis.

A.5 Limitations

Coding in the Register is continually reviewed as coroners’ investigations progress and findings are made. Therefore, any data reported from the Register is subject to review and may subsequently change.

Combining heroin with morphine of unknown source for analysis may lead to an over-estimate of heroin involvement in Victorian overdose deaths. However, the magnitude of over-estimate is likely to be very small and is preferable to the large distortion in estimation of heroin overdose that would occur if morphine of unknown source was treated separately to heroin.

CPU COURTS IN CONFIDENCE 10/10