Shehabi v Attorney General

Background

1. Prof Shehabi first lodged his application seeking approval for the SPICE III study (Protocol version 2 dated 24 January 2013) on 1 August 2013 in the Tribunal's predecessor, the Guardianship Tribunal of New South Wales (proceedings CT 10/2013). As we understood it, such an application for approval was thought to be the most suitable vehicle for obtaining a decision as to whether or not the SPICE III study was a clinical trial for the purposes of Pt 5 of the Guardianship Act. Thus, although Prof Shehabi lodged this application for approval, his primary position was that the application should be dismissed on the basis that the SPICE III study was not a clinical trial that could be approved under s 45AA of the Guardianship Act. In the circumstances of this case, it was not an inappropriate manner in which to proceed.
2. A further version of the protocol for the SPICE III study (Protocol version 3 dated 14 June 2013) was adopted prior to proceedings CT 10/2013 being determined.
3. On 1 January 2014, the Guardianship Tribunal was abolished when the Civil and Administrative Tribunal of New South Wales (NCAT) was established and the functions of the abolished Tribunal came to be performed by the Guardianship Division of NCAT. NCAT was empowered to deal with the proceedings under the transitional provisions in Sch 1, cl 7 of the Civil and Administrative Tribunal Act 2013 (NSW) (the NCAT Act).
4. On 23 December 2014, the Guardianship Division of the Tribunal made the following orders in those proceedings: "1. The Tribunal has considered the clinical trial known as SPICE III: A Prospective Multicentre Randomised Controlled Trial of Early Goal Directed Sedation Compared with Standard Care in Mechanically Ventilated Patients in Intensive Care. 2. The Tribunal is satisfied that in relation to the above-named clinical trial, all the requirements of section 45AA(2) of the Guardianship Act have been met. 3. The Tribunal approves the clinical trial as one in which adults unable to consent to their own treatment may participate. Unless the Tribunal subsequently orders otherwise, this approval remains current until that trial has been concluded in New South Wales. 4. The Tribunal approves the clinical trial at the following sites: Prince of Wales Hospital, Nepean Hospital, Hornsby Ku-ring-gai Hospital, St Vincent's Hospital, Westmead Hospital and Lismore Hospital. This approval is subject to the following condition(s): 5. The Tribunal orders that 'persons responsible' for those unable to consent to their own medical treatment may not exercise the function of giving or withholding consent to the carrying out of medical treatment on the person they are 'person responsible' for in the course of the clinical trial. 6. The Tribunal is not satisfied that the forms for granting consent and the information sheet provided to the Tribunal by: Prince of Wales Hospital, Nepean Hospital, Hornsby Ku-ring-gai Hospital, St Vincent's Hospital, Westmead Hospital and Lismore Hospital. are sufficient to enable 'persons responsible' to decide whether or not it is appropriate that those persons they are 'person responsible' for should take part in the clinical trial. 7. The Tribunal orders that the Tribunal is required to exercise the function of giving or withholding consent for the carrying out of medical treatment on patients in the course of the clinical trial (s 45AB(1)(b)). "

• Order Approving Participation in Clinical Trial dated 23 December 2014 and the reasons for decision: Application for approval for adults unable to consent to their own treatment to participate in a clinical trial (SPICE III Trial) [2014] NSWCATGD 44.

1. It can be noted from orders 6 and 7 of those orders that the function of giving or withholding consent under s 45AB was to be exercised by the Tribunal. This came about because the Tribunal was not satisfied that the form for granting consent and the information available were sufficient to meet the requirements of s 45AB(2) so that the consent function could be exercised by the "persons responsible for the patients".
2. On 24 April 2015 the Applicant lodged a further application with the Tribunal in relation to the SPICE III study (proceedings CT 7/2015). In this second application Prof Shehabi sought: 1. to extend the study to certain sites in addition to those included in the Tribunal's orders of 23 December 2014; and 2. to have the function of giving or withholding consent to be exercised by persons responsible instead of the Tribunal.
3. With this application Prof Shehabi submitted a further protocol for the SPICE III study (version 3.6 dated 26 April 2015) together with the proposed forms and information to be provided to persons responsible for patients who were proposed to be participants in the study. This application did not seek to challenge the Tribunal's previous conclusion that the SPICE III study was a clinical trial for the purposes of Pt 5 of the Guardianship Act.
4. On 16 June 2015 Prof Shehabi submitted a further amended protocol for the SPICE III study (version 3.10 dated 8 June 2015) and amended proposed information and consent forms for persons responsible.
5. On 19 June 2015, the Guardianship Division made orders in proceedings CT 7/2015 in relation to the SPICE III study as follows: "1. The Tribunal has considered the clinical trial known as SPICE III: A Prospective Multicentre Randomised Controlled Trial of Early Goal Directed Sedation Compared with Standard Care in Mechanically Ventilated Patients in Intensive Care as described in Protocol v3.10 dated 8 June 2015. 2. The Tribunal is satisfied that in relation to the above-named clinical trial, all the requirements of section 45AA(2) of the Guardianship Act 1987 (NSW) have been met. 3. The Tribunal approves the clinical trial as one in which adults unable to consent to their own treatment may participate. Unless the Tribunal subsequently orders otherwise, this approval remains current until that trial has been concluded in New South Wales. 4. The Tribunal approves the clinical trial at the following sites: Prince of Wales Hospital, Prince of Wales Private Hospital, Nepean Hospital, Hornsby Ku-Ring-Gai Hospital, St Vincent's Hospital, Lismore Base Hospital, Westmead Hospital, Royal North Shore Hospital, Gosford Hospital 5. The Tribunal orders that 'persons responsible' for those unable to consent to their own medical treatment may exercise the function of giving or withholding consent to the carrying out of medical treatment on the person they are 'person responsible' for in the course of the above-named clinical trial. 6. The Tribunal is satisfied that the form for granting consent and the information sheet provided to the Tribunal by the applicant described as NSW Master PIS&CF (person responsible) version 3 dated 27 April 2015 is sufficient to enable 'persons responsible' to decide whether or not it is appropriate that those persons they are 'person responsible' for should take part in the above-named clinical trial."

• Application for additional orders relating to a clinical trial previously approved by the Tribunal and known as the SPICE III Trial [2015] NSWCATGD 24 at [1] and Appendix A.

The Appeal

1. On 28 July 2015, Prof Shehabi lodged with the Tribunal a Notice of Appeal signed by him on 19 July 2015, together with a letter dated 19 July 2015 headed "… SPICE III Study (10/2013 and 7/2015) Grounds for Appeal". The Notice of Appeal identified the date of the decision the subject of the appeal as 19 June 2015. This suggested that the only decision appealed from was the later decision in proceedings CT 7/2015. From other material in the notice of appeal and the letter, however, it is apparent that the decision dated 23 December 2014 in proceedings CT 10/2013 was also appealed against. For example, the heading to the letter of 19 July 2015, refers to both proceedings CT10/2013 and CT 7/2015 and in section 11C of the notice of appeal, under the heading "Orders the NCAT Appeal Panel should make", Prof Shehabi stated: "Due to the little differences and the many similarities that exists between SPICE III and ADRENAL, we believe the Tribunal should determine that SPICE III is not a "Clinical Trial" from the Guardianship perspective to provide consistency and clarity to researchers, governance officers and ethics committees."
2. In our view, there is no doubt that Prof Shehabi intended to appeal against the orders in both proceedings CT10/2013 and CT 7/2015 since the underlying basis for his challenge to all of the orders made has always been that the SPICE III study was not a clinical trial which could be approved under s 45AA of the Guardianship Act.
3. Rule 25(4)(c) of the Civil and Administrative Tribunal Rules 2014 requires an appeal of this nature to be lodged within 28 days from the day on which the appellant was notified of the decision to be appealed or given reasons for the decision (whichever is the later). In the circumstances of this appeal, the notice of appeal was lodged outside this period and would have been out of time unless an extension of time was granted under s 41 of the NCAT Act. On 3 November 2015, the Appeal Panel extended the time for filing the appeal to 31 July 2015, by consent, and thus the appeal is taken to have been lodged within time.
4. In both proceedings at first instance, Prof Shehabi was the only party and there was no contradictor. At a directions hearing on 31 August 2015, the Appeal Panel directed the Principal Registrar to refer the papers in the appeal to the Attorney General of New South Wales, in her capacity as the Minister administering the Guardianship Act, so that she could consider whether to intervene under s 44(4)(a) and (b) of the NCAT Act in order to act as the contradictor in this appeal. The Attorney General exercised her right to intervene and has acted as the contradictor in this matter.

Grounds of Appeal

1. Prof Shehabi was permitted to file a document headed "Amended Grounds of Appeal" on 2 December 2015. The grounds raised in that document were as follows: "1. The Tribunal's finding that the proposed Study known as SPICE III, ("the Study") is a clinical trial as defined in section 33 (1) of the Guardianship Act NSW ("the Act") is contrary to the evidence and the weight of the evidence, before the Tribunal; 2. In reaching its finding that the Study is a clinical trial as defined in the Act, the Tribunal failed to give sufficient weight to: (a) Evidence that the Study consists of established therapies and procedures considered current standard of care in most hospital Intensive Care Units ("ICUs") in New South Wales and elsewhere in Australia; (b) Evidence that the Study does not involve new or experimental treatment; but rather, a comparison of standard care sedation strategies to determine their long-term effects; (c) Evidence of the Chair and Deputy Chair of the South East Sydney Local Health District, Human Research Ethics Committee ("HREC") concerning the HREC view that that the Study is a comparison of two accepted forms of treatment and noting the equipoise of the Study; (d) Evidence that when a patient is admitted to an ICU and requires mechanical ventilation, sedation usually takes place as a matter of urgency and often without the consent (or the opportunity to obtain the consent) of the patient or the patient's responsible person; (e) Evidence to the effect, that the predominant use of dexmedetomidine in Australia is for sedation of ventilated patients, medical or surgical, in Intensive Care and for longer than 24 hours; (f) Evidence that the Study is not "blinded" and hence, clinicians and other carers are fully aware of the treatments administered within the Study, consistent with relevant clinical practice; and the relevant physician can therefore adjust the administered treatment at any time according to a patient's clinical needs; (g) Evidence that the study is randomised as essentially, a scientific and academic way of reducing bias, with the relevant clinical using similar agents and procedures to achieve a similar sedation level to that which would have been achieved without the patient being part of the Study; (h) Evidence that to cease use of a sedative agent such as dexmedetomidine providing needed ongoing comfort and sedation to any patient because the period of its infusion has reached a 24 hour duration is both unethical and dangerous; (i) Evidence that either treatment arm of the Study is equivalent for the particular patient; and as an emergency intervention for patients who need sedation, the care in both Study arms will achieve the same objective; (j) Evidence that the intention of the treatments within the Study is to achieve the alleviation of suffering and to allow life-sustaining therapies to be provided to patients; (k) Evidence that the regulatory framework created by the Therapeutic Goods Administration ("TGA"), including the approved Product Information ("P.I."), is commonly not congruent with actual clinicians' use and current standard practice; and that such actual use and practice can be standard practice despite a lack of TGA approval for that specific indication; but that use for such indication cannot be marketed commercially; 3. In reaching its conclusion that the Study is a clinical trial as defined in the Act and having regard for the objects of the Act, the Tribunal took into account an irrelevant consideration by relying on evidence that the Clinical Trial Notification (CTN) Scheme (Australian Government, Department of Health, TGA) is engaged in relation to the Study; 4. The Tribunal took into account irrelevant considerations in concluding that whilst dexmedetomidine retains its current TGA registration and is therefore captured by the CTN Scheme, then the Study proposing to use the drug "…beyond its licensed parameters in the context of a randomised trial…amounts to a 'clinical trial' for the purposes of the Act." 5. In concluding on the evidence before it, that the Study is a clinical trial for the purposes of the Act, the Tribunal failed to apply the principles and reasoning, applied by the then, Guardianship Tribunal in its decision delivered on 23 November 2004, in: 'Application for Approval for Adults unable to consent to their own treatment in a clinical trial' concerning a 'Multi-Centre, Open Label, Randomised, controlled trial of two target ranges for glycaemic control in Intensive Care Unit Patients' (the "NICE Study"), Trial No. T/2004/12."
2. This document also indicated that Prof Shehabi was seeking that the Appeal Panel make the following orders to dispose of the appeal: "1. That the SPICE III study does not constitute a clinical trial pursuant to the Guardianship Act 1987 (NSW) 2. That the application for approval of adults, unable to give consent to their own treatment, to participate in the SPICE III study is dismissed. 3. Any other orders the Appeal Panel sees fit."
3. The Attorney General filed her reply to appeal on 21 December 2015 and responded to the grounds of appeal raised by Prof Shehabi as follows: "Ground 1: The Tribunal's finding that the study known as SPICE III: A Prospective Multicentre Randomised Controlled Trial of Early goal Directed Sedation Compared with Standard Care in Mechanically Ventilated Patients in Intensive Care is a "clinical trial" is supported by the evidence. Ground 2: In reaching its finding that the study was a "clinical trial", the Tribunal gave sufficient weight to the evidence. Grounds 3 and 4: The Tribunal did not take into account irrelevant considerations in considering that study was a "clinical trial". Ground 5: The Tribunal was not in error in its application of principles and reasoning."
4. As formulated in the Amended Grounds of Appeal, Prof Shehabi's grounds raise the following questions: 1. whether the Tribunal's findings were against the weight of the evidence or gave sufficient weight to certain evidence - grounds 1 and 2; 2. whether the Tribunal took into account irrelevant considerations - grounds 3 and 4; and 3. whether the Tribunal should have adopted the same construction of s 45AA as it applied to clinical trials as was adopted in the NICE study decision which proceeded on the basis that clinical trials that do not involve new or non-standard treatment do not fall within s 45AA of the Guardianship Act - ground 5.
5. The Appeal Panel was not convinced that grounds 1 and 2 and grounds 3 and 4 were suitably formulated to capture the real bases of Prof Shehabi's appeal.
6. As has already been noted, Prof Shehabi's underlying basis for challenging all of the orders made in both proceedings was that the SPICE III study was not a clinical trial that could be approved under s 45AA of the Guardianship Act. This proposition can be seen as resting upon two contentions, the substance of which was put forward at various times on behalf of Prof Shehabi: 1. On the proper construction of s 45AA, clinical trials which can be approved under that section do not include all trials of drugs or techniques that necessarily involve the carrying out of medical (or dental) treatment on the participants in the trial but only trials of drugs or techniques that are new or experimental or treatments that are new or non-standard; 2. On the facts as found by the Tribunal, the SPICE III study is not a clinical trial to which s 45AA applies in that the treatments (both drugs and their mode of use) used in both arms of the SPICE III study are existing, accepted treatments for patients in intensive care and the fact that dexmedetomidine is used outside its product information approved under the TG Act (often known as off label use) in a randomised trial, and consequently notification under the Therapeutic Goods Act 1989 (Cth) (the TG Act) is required, does not mean that the SPICE III study is a trial of drugs or techniques that are new or experimental or treatments that are new or non-standard.
7. During the hearing the Appeal Panel raised the formulation of the grounds of appeal with the parties and both sides were content to proceed on the basis that the substance of Prof Shehabi's grounds of appeal could be captured in two reformulated grounds to the following effect: 1. That the Tribunal below erred in so far as it did not construe s 45AA as applying only to trials of drugs or techniques that were new or experimental or treatments that were new or non-standard (based on ground 5); 2. That the Tribunal below erred in finding that, because the SPICE III study was required to be notified under the TG Act as a result of dexmedetomidine being used in a randomised trial outside its approved product information, the SPICE III study was a trial of new or experimental drugs or techniques and thus was a clinical trial falling within s 45AA (based on grounds 1 - 4).
8. Formulated in this way, the first reformulated ground raises the proper construction of the expression "clinical trial" in the Guardianship Act. Where, as in the present case, the words "clinical trial" used in s 45AA and elsewhere in Pt 5 of the Guardianship Act derive shades of meaning from their context and the syntax of the sentences, resolving their meaning involves a question of law: Duffy v Da Rin [2014] NSWCA 270; 87 NSWLR 495 at [30].
9. The second reformulated ground raises the question of whether the facts as fully found by the Tribunal in relation to the SPICE III study meant that, on the proper construction of s 45AA and Pt 5, the study was not a clinical trial that could be approved under that section. This also involves a question of law: Hope v Bathurst City Council [1980] HCA 16; 144 CLR 1 at 7.

Internal Appeal - Leave to Appeal Not Required

1. The decisions made in proceedings CT 10/2013 and CT 7/2015 were general decisions of the Tribunal within the meaning of s 32(1)(a) of the NCAT Act and thus were internally appealable decisions. Consequently, Prof Shehabi was entitled to appeal against those decisions to the Appeal Panel under s 80(1) of the NCAT Act. Furthermore, as the decisions were not interlocutory decisions (as defined in s 4(1) of the NCAT Act), Prof Shehabi was entitled to appeal as of right on any question of law and by leave of the Appeal Panel on any other ground, under s 80(2)(b).
2. Both of the reformulated grounds involve a question of law and, accordingly, Prof Shehabi was entitled to appeal as of right in respect of those grounds and leave to appeal, although sought in the notice of appeal, was not required.

Material and Hearing before the Appeal Panel

1. In addition to the notice of appeal, reply to appeal, amended grounds of appeal and written submissions by both parties, the Appeal Panel had before it: 1. an "Agreed Bundle of Documents" in two volumes, prepared by the parties and filed on 26 February 2016; and 2. a bundle of Supplementary Documents prepared by the appellant's solicitors and filed on 22 March 2016. The intervener did not object to this bundle of Supplementary Documents being relied upon in this appeal.
2. At the hearing on 27 May 2016, Mr Dwyer of counsel appeared for Prof Shehabi and Ms Kumar of counsel appeared for the Attorney General.

First Reformulated Ground - Proper Construction of "Clinical Trial" in Pt 5

1. The first reformulated ground of appeal requires us to consider the proper construction of the expression "clinical trial" as it is used in Pt 5 and, more specifically, ss 35 and 45AA of the Guardianship Act.
2. At first instance, the Tribunal held at [92]: "It is not, however, necessary to undertake a detailed examination of the context and legislative purpose of the clinical trial provisions in Part 5 of the Guardianship Act to determine whether the meaning of 'trial' should be limited in the manner suggested by the applicant and as referred to in earlier Tribunal decisions. This is because the Tribunal was satisfied that even if, for the sake of argument, the meaning of 'clinical trial' was to be limited in the manner suggested by the applicant, SPICE III satisfies this narrower view of the term on the basis that the trial proposes the use of dexmedetomidine beyond its TGA registration."
3. The "narrower view of the term" referred to in that passage is a construction of Pt 5 that would limit the clinical trials that could be approved under s 45AA to trials of "new or experimental drugs or treatments" (see the reasons for decision at first instance (Reasons) at [81]). The reference to earlier Tribunal decisions may be a reference to decisions such as the NICE decision of the Guardianship Tribunal (Orders made on 23 November 2004, Reasons for decision dated 8 December 2004, unreported) which held: "The [NICE] study seeks to compare the administration of two standard forms of treatment to ensure the maintenance of a designated blood glucose level concentration in patients in intensive care, which does not involve major treatment, it does not involve non-standard or new treatment, nor does it involve any significant medical risks. Accordingly, the Tribunal was not satisfied that the NICE study (Normoglycaemia in Intensive Care Evaluation Study) involves a trial of new or non-standard treatment. The Tribunal was not satisfied that the NICE study is a clinical trial within the meaning of the Guardianship Act and the Tribunal therefore determined to dismiss the application for approval of that trial."
4. Before the Appeal Panel, Prof Shehabi submitted that the Tribunal below should have held that s 45AA only permitted the Tribunal to approve a clinical trial if it was a trial of a new or experimental treatment. In this regard, he expressly adopted the reasoning of the Guardianship Division in the later decision of Application for approval for adults unable to consent to their own treatment to participate in a clinical trial (ADRENAL Trial) [2015] NSWCATGD 23 (ADRENAL), especially at [107] to [125]. In ADRENAL, the Tribunal concluded at [125]: "A textual reading of the provisions relating to clinical trials in Part 5 of the Guardianship Act has led the Tribunal to the conclusion, on balance, that the definition of 'clinical trial' in section 33(1) of the Act, despite the breadth of the definition in that section and the language used in the definition of 'special medical treatment' in the same definitional section, should be interpreted as excluding trials of a currently accepted treatment for the condition in question."
5. The Attorney General submitted that there was no error in the approach of the Tribunal at first instance in its reasons for decision in relation to the SPICE III study concerning the construction of "clinical trial" and its application in this case. She referred the Appeal Panel to [87] to [99] of the Tribunal's Reasons which, it was submitted, disclosed no error and were based on an appropriate approach in the circumstances of this matter.

Applicable Principles

1. The relevant principles of statutory construction are well established. They include: 1. The process of statutory construction begins with the statutory text and, in a sense, also ends with the statutory text, considered in its context: Commissioner of Taxation v Consolidated Media Holdings Ltd [2012] HCA 55; 250 CLR 503 at [39]; 2. The context of the statutory text includes: 1. extrinsic material to which reference may be had under s 34 of the Interpretation Act 1987 (NSW) but only so far as it assists in ascertaining the meaning of the statutory text. Such material cannot displace the meaning of the text: Alcan (NT) Alumina Pty Ltd v Commissioner of Territory Revenue [2009] HCA 41; 239 CLR 27 at [47]; Bignill v Director of Public Prosecutions [2016] NSWCA 13 at [25]; 2. the legislative history: Thiess v Collector of Customs [2014] HCA 12; 250 CLR 664 at [22]; Bignill v Director of Public Prosecutions [2016] NSWCA 13 at [25]; and 3. the objectively determined purpose and policy of the provision and the mischief it is seeking to remedy: Alcan (NT) Alumina Pty Ltd v Commissioner of Territory Revenue [2009] HCA 41; 239 CLR 27 at [47]; Thiess v Collector of Customs [2014] HCA 12; 250 CLR 664 at [23]; Bignill v Director of Public Prosecutions [2016] NSWCA 13 at [25]; 1. Ordinarily the legal meaning will correspond with the grammatical meaning of a provision but not always. The context of the words, the consequences of a literal or grammatical construction, the purpose of the provision or the principles of construction may require the words of a legislative provision to be read in a way that does not correspond with the literal or grammatical meaning: Project Blue Sky v Australian Broadcasting Authority [1998] HCA 28; 194 CLR 355 at [78]; 2. The primary object of statutory construction is to construe the relevant provision so that it is consistent with the language and purpose of all the provisions of the statute and on the prima facie basis that its provisions are intended to give effect to harmonious goals: Project Blue Sky v Australian Broadcasting Authority [1998] HCA 28; 194 CLR 355 at [69] and [70]; 3. The context, the general purpose and policy of a provision and its consistency and fairness may be surer guides to its meaning than the logic with which it is constructed: Commissioner for Railways (NSW) v Agalianos [1955] HCA 27; 92 CLR 390 at 397.
2. In relation to the use of statutory definitions in construing a provision, it is important to bear in mind the words of McHugh J in Kelly v The Queen [2004] HCA 12; 218 CLR 216 at [103]: "… the function of a definition is not to enact substantive law. It is to provide aid in construing the statute. Nothing is more likely to defeat the intention of the legislature than to give a definition a narrow, literal meaning and then use that meaning to negate the evident policy or purpose of a substantive enactment. There is, of course, always a question whether the definition is expressly or impliedly excluded. But once it is clear that the definition applies, the better - I think the only proper - course is to read the words of the definition into the substantive enactment and then construe the substantive enactment - in its extended or confined sense - in its context and bearing in mind its purpose and the mischief that it was designed to overcome. To construe the definition before its text has been inserted into the fabric of the substantive enactment invites error as to the meaning of the substantive enactment."
3. Applying these principles to s 45AA of the Guardianship Act requires: 1. an appreciation of the purpose of the Guardianship Act as a whole; 2. a detailed understanding of Pt 5 of that Act, which is the only Part in which "clinical trials" are mentioned, and the definition of "clinical trial" contained in that Part; and 3. consideration of the legislative history of the clinical trial provisions and the mischief that the introduction of those provisions was designed to overcome.

Purpose of the Guardianship Act and the Section 4 Principles

1. In the Appeal Panel's view some, at least, of the purposes of the Guardianship Act are encapsulated in the general principles stated in s 4 of the Act as follows: "4. General Principles It is the duty of everyone exercising functions under this Act with respect to persons who have disabilities to observe the following principles: (a) the welfare and interests of such persons should be given paramount consideration, (b) the freedom of decision and freedom of action of such persons should be restricted as little as possible, (c) such persons should be encouraged, as far as possible, to live a normal life in the community, (d) the views of such persons in relation to the exercise of those functions should be taken into consideration, (e) the importance of preserving the family relationships and the cultural and linguistic environments of such persons should be recognised, (f) such persons should be encouraged, as far as possible, to be self-reliant in matters relating to their personal, domestic and financial affairs, (g) such persons should be protected from neglect, abuse and exploitation, (h) the community should be encouraged to apply and promote these principles."
2. Of particular significance in the context of clinical trials involving persons who are unable to give consent to participation are those principles found in pars (a) and (g). As a consequence, the Appeal Panel, in construing Pt 5 as it applies to clinical trials, should consider which construction would promote the welfare and interests of persons with disabilities and the extent to which the construction serves to protect such persons from neglect, abuse and, especially relevant in the context of trials of medical treatments, exploitation.

Part 5 of the Guardianship Act

1. At the most general level, Pt 5 (ss 32 to 48) of the Guardianship Act deals with when consent to medical or dental treatment is required, and how it may be obtained in respect of adults with impaired decision making capacity.

Objects of Pt 5

1. The objects of Pt 5 are stated in s 32 of the Guardianship Act and give an indication of the purposes of that Part. Furthermore, as a result of the operation of s 33 of the Interpretation Act 1987 (NSW), a construction of Pt 5 that would promote the purpose or object underlying its provisions should be preferred to a construction that would not promote that purpose or object. Section 32 of the Guardianship Act is in the following terms: "32 Objects The objects of this Part are: (a) to ensure that people are not deprived of necessary medical or dental treatment merely because they lack the capacity to consent to the carrying out of such treatment, and (b) to ensure that any medical or dental treatment that is carried out on such people is carried out for the purpose of promoting and maintaining their health and well-being."
2. These objects indicate, in a case such as the present, that the preferred construction of the provisions in Pt 5 should be one that ensures that persons with impaired decision making capacity have access to drugs and techniques that are necessary for their health and well-being notwithstanding that those treatments are only available through clinical trials. This restricted availability might be anticipated to be primarily because those treatments are still in the developmental phase or they have not yet become standard or widely accepted.

Scope of Operation of Pt 5

1. The scope of operation of Pt 5 is set out in s 34(1) which provides that Pt 5 applies to patients who are 16 years of age or older and who are incapable of giving consent to the carrying out of medical or dental treatment.
2. "Medical … treatment" or "treatment" is relevantly defined in s 33(1) as meaning: "medical treatment (including any medical or surgical procedure, operation or examination and any prophylactic, palliative or rehabilitative care) normally carried out by or under the supervision of a medical practitioner … (and, in the case of treatment in the course of a clinical trial, is taken to include the giving of placebos to some of the participants in the trial) …[subject to a number of exceptions which are not relevant for present purposes]".
3. It was not in dispute in this matter and it was accepted by the Tribunal that: 1. The administration of drugs for the purpose of sedation as part of the SPICE III study was "medical treatment" for the purposes of Pt 5 and dental treatment was not involved; 2. The SPICE III study did not involve the giving of placebos to any participants; and 3. Patients 16 years and older in intensive care who require mechanical ventilation and who fall within the study's entry criteria will not be capable of giving consent to the carrying out of medical treatment and, consequently, persons who might be candidates for participation in the SPICE III study will be patients to whom Pt 5 applies.

Prohibition on Treatment without Consent or Authorisation

1. The central provision of Pt 5 is s 35. This section prohibits medical treatment for persons to whom the Part applies except with a relevant form of consent or other authorisation. Relevantly for the present appeal, s 35 provides as follows: "35 Offences (1) A person must not carry out medical … treatment on a patient to whom this Part applies unless: (a) consent for the treatment has been given in accordance with this Part, or (b) the carrying out of the treatment is authorised by this Part without any such consent, or (c) the treatment is carried out in accordance with an order made by the Supreme Court in the exercise of its jurisdiction with respect to the guardianship of persons. Maximum penalty: • in the case of special treatment or treatment in the course of a clinical trial (on conviction on indictment) - imprisonment for 7 years, or • in the case of minor or major treatment (on summary conviction) - imprisonment for 1 year or 10 penalty units, or both. …"
2. As a consequence of s 35 and in the absence of a Supreme Court order under s 35(1)(c), the medical practitioners conducting the SPICE III study will commit an offence by carrying out the medical treatment involved in the study on patients to whom Pt 5 applies unless: 1. the treatment is authorised by Pt 5 without consent; or 2. consent has been obtained in accordance with Pt 5 for the treatment.

Types of Treatment referred to in Pt 5

1. The provisions relating to medical (and dental) treatment for which consent is not required under Pt 5, or for which consent may be obtained under that Part, distinguish between 4 types of treatment: 1. Special treatment: 2. Treatment in the course of a clinical trial; 3. Major treatment; and 4. Minor treatment.

Special Treatment

1. Under s 33(1) of the Guardianship Act, "special treatment" in Pt 5 relevantly means: "… (b) any new treatment that has not yet gained the support of a substantial number of medical practitioners … specialising in the area of practice concerned, … but does not include treatment in the course of a clinical trial."
2. The other types of treatment that fall within the definition of "special treatment", namely treatments involving permanent infertility, termination of pregnancy, vasectomy or tubal occlusion or aversive stimuli, are not relevant in the present appeal (see pars (a) and (c) of the definition in s 33(1) and cl 9 of the Guardianship Regulation 2010).

Clinical Trial

1. Section 33(1) provides that "clinical trial" in Pt 5 relevantly means "a trial of drugs or techniques that necessarily involves the carrying out of medical … treatment on the participants in the trial". Further, as has already been noted, "treatment in the course of a clinical trial" is expressly stated to fall outside the definition of "special treatment" in s 33(1).

Major Treatment

1. The definition of "major treatment" in s 33(1) excludes any treatment that is special treatment or treatment in the course of a clinical trial but otherwise leaves the meaning of the term to be declared by the Guardianship Regulation. Clause 11 of the Regulation contains a list a treatments that are declared to be "major treatment" for the purposes of the definition in s 33(1). For this appeal and the treatment administered in the SPICE III study, the most significant type of treatment declared to be major treatment is described in cl 11(c) as follows: "(c) any treatment that involves the administration of a general anaesthetic or other sedation, but not treatment involving: (i) sedation used to facilitate the management of fractured or dislocated limbs, or (ii) sedation used to facilitate the insertion of an endoscope into a patient's body for diagnostic purposes unless the endoscope is inserted through a breach or incision in the skin or a mucous membrane,"
2. Notwithstanding that declaration in cl 11(c), if the SPICE III study treatment is treatment in the course of a clinical trial or special treatment, it will not be major treatment, because of the express exclusions from the definition of "major treatment" in s 33(1), referred to above. If, however, the SPICE III study treatment is not treatment in the course of a clinical trial or special treatment, it will be major treatment, assuming that it falls within cl 11(c) of the Guardianship Regulation.

Minor Treatment

1. Finally, under s 33(1), "minor treatment" means treatment that is not treatment in the course of a clinical trial, special treatment or major treatment.

Treatment Authorised by Pt 5 without Consent

1. Section 37 is the only section in Pt 5 that permits medical treatment to be carried out on a patient to whom the Part applies without consent given in accordance with that Part or a Supreme Court order. Under s 37, there are two situations in which consent is not required.
2. First, treatment without consent is permitted in cases of urgent necessity. In this regard, s 37(1) provides: (1) Medical … treatment may be carried out on a patient to whom this Part applies without consent given in accordance with this Part if the medical practitioner … carrying out or supervising the treatment considers the treatment is necessary, as a matter of urgency: (a) to save the patient's life, or (b) to prevent serious damage to the patient's health, or (c) except in the case of special treatment - to prevent the patient from suffering or continuing to suffer significant pain or distress.
3. Paragraph (c) is different from pars (a) and (b) in that par (c) contains an express exception for cases involving "special treatment". Thus, consent is required in accordance with Pt 5, even if the treatment is urgent and necessary, when it is special treatment and its only purpose is to prevent the patient suffering significant pain or distress. The same does not apply if the treatment is in the course of a clinical trial rather than special treatment.
4. Before any treatment can be administered without consent under s 37(1), the medical practitioner carrying out or supervising the treatment must have considered the treatment to be necessary, as a matter of urgency for one of the purposes in s 37(1)(a), (b) or (c) in the circumstances of the particular case. Whether or not the treatment was part of a "clinical trial" is not determinative of whether or not the treatment would fall within s 37(1). Indeed, it might well be more difficult to be legitimately satisfied that the treatment was necessary in the case of treatment in the course of a clinical trial or special treatment, where such treatment was new treatment that had not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned.
5. Secondly, minor treatment may be carried out on a patient without consent in the circumstances set out in s 37(2) and (3). As has been noted above, the SPICE III study treatment would not fall within the definition of "minor treatment". Consequently, s 37(2) and (3) are not relevant on this appeal.
6. If the medical treatment for patients to whom Pt 5 applies is not authorised to be performed without consent under either of the two limbs of s 37 and there is no relevant Supreme Court order, consent has to be obtained in accordance with the other provisions of Pt 5.

Treatment with Consent in accordance with Pt 5

1. Sections 36, 40 and 42 to 45AB establish by whom consent to the carrying out of medical treatment on a patient to whom Pt 5 applies may be given and how the consent may be obtained. In summary: 1. The "person responsible" in respect of the patient may consent to minor or major treatment: s 36(1)(a). Section 33A specifies who the "person responsible" is for a number of categories of people and provides a hierarchy of "persons responsible" for those other than children or persons in the care of the Secretary. The method of obtaining that consent, the information to be provided and other steps required to be taken are set out in s 40. 2. The Tribunal may consent in any case: s 36(1)(b). How applications for consent are to be made to the Tribunal and dealt with is set out generally in ss 42 to 45. A special regime for obtaining consent to participation in clinical trials is established by ss 45AA and 45AB. Use of this special regime can obviate the need for an application to the Tribunal for consent for each individual patient to whom Pt 5 applies who is proposed to be a participant in the clinical trial. 3. The guardian of a patient may consent in certain cases involving continuing or further special treatment, under s 36(2), if the Tribunal has already consented to the special treatment and authorised the guardian to consent to the continuing or further treatment. How that power is to be exercised is governed by s 45A.
2. It follows from these provisions that, unless the treatment is authorised under s 37 or a Supreme Court order: 1. If medical (or dental) treatment is neither treatment in the course of a clinical trial nor special treatment, it will be major or minor treatment to which the "responsible person" may consent under s 36(1)(a), provided that the requirements of s 40 are complied with. Alternatively, the Tribunal may consent if an application for consent is made to it under s 42 and, in such a case, the process to be followed is set out in ss 43 to 45. 2. If the treatment is special treatment, only the Tribunal can consent, at least initially and in the absence of a guardian appointed to provide consent for further or continuing treatment, s 36(1)(b) and (2). Obtaining consent from the Tribunal on a case by case basis is governed by ss 42 to 45. Section 45A deals specifically with consent by a patient's guardian authorised by the Tribunal to consent to further or continuing special treatment. 3. If the treatment is treatment in the course of a clinical trial, the Tribunal can approve the trial under s 45AA and, if it does so, consent can be obtained in the manner determined by the Tribunal under s 45AB. In appropriate cases, the Tribunal can permit the "person responsible" to consent, thus avoiding the need for an application to the Tribunal on a case by case basis.
3. Consequently, whether the treatment to be carried out as part of the SPICE III study is part of a clinical trial, special treatment or major treatment has significant implications for obtaining consent under Pt 5.
4. In this context, the appellant, Prof Shehabi, applied for approval of the SPICE III study under s 45AA but his primary position was that the study was not a clinical trial falling within that section and his application should be dismissed. It appears to have been assumed that if it was not a clinical trial, the treatment involved in the SPICE III study would also not be special treatment. Thus, it would be major treatment to which persons responsible could consent under s 40 and there would be no need for individual applications to the Tribunal for every patient. Alternatively, the treatment might be authorised if the requirements of s 37(1) were satisfied.
5. The Appeal Panel now turns to consider ss 45AA and 45AB and the "clinical trials" that fall within those sections.

The Approval and Consent Regime under ss 45AA and 45AB

1. Sections 45AA and 45AB govern the approval of clinical trials and how consent to participation is to be obtained. Those sections provide: 45AA Tribunal may approve clinical trials (1) The Tribunal may approve, in accordance with this section, a clinical trial as a trial in which patients to whom this Part applies may participate. (2) The Tribunal may give an approval under this section only if it is satisfied that: (a) the drugs or techniques being tested in the clinical trial are intended to cure or alleviate a particular condition from which the patients suffer, and (b) the trial will not involve any known substantial risk to the patients (or, if there are existing treatments for the condition concerned, will not involve material risks greater than the risks associated with those treatments), and (c) the development of the drugs or techniques has reached a stage at which safety and ethical considerations make it appropriate that the drugs or techniques be available to patients who suffer from that condition even if those patients are not able to consent to taking part in the trial, and (d) having regard to the potential benefits (as well as the potential risks) of participation in the trial, it is in the best interests of patients who suffer from that condition that they take part in the trial, and (e) the trial has been approved by a relevant ethics committee and complies with any relevant guidelines issued by the National Health and Medical Research Council. (3) The fact that a clinical trial will or may involve the giving of placebos to some of the participants in the trial does not prevent the Tribunal from being satisfied that it is in the best interests of patients that they take part in the trial. (4) The Tribunal's approval of a clinical trial under this section does not operate as a consent to the participation in the trial of any particular patient to whom this Part applies. The appropriate consent must be obtained under Division 3 or 4 before any medical or dental treatment in the course of the trial is carried out on the patient. (5) In this section: ethics committee means: (a) for so long as there is any relevant Institutional Ethics Committee registered by the Australian Health Ethics Committee established under the National Health and Medical Research Council Act 1992 of the Commonwealth - an Institutional Ethics Committee so registered, or (b) in the absence of such a committee, an ethics committee established by: (i) a local health district or a public hospital, or (ii) a university, being an ethics committee concerned, wholly or partly, with medical research, or (iii) the National Health and Medical Research Council. 45AB Consent for participation in clinical trials in individual cases (1) If the Tribunal is satisfied as to the matters specified in section 45AA (2) in relation to a clinical trial, it may, by order, determine: (a) that the function of giving or withholding consent for the carrying out of medical or dental treatment on patients in the course of the trial is to be exercised by the persons responsible for the patients (in which case Division 3 applies), or (b) that the Tribunal is to exercise that function itself (in which case Division 4 applies). (2) Before making a determination referred to in subsection (1) (a), the Tribunal must be satisfied that the form for granting consent and the information available about the trial provide sufficient information to enable the persons responsible to decide whether or not it is appropriate that the patients should take part in the trial.
2. These sections establish a three step process for obtaining consent to medical treatment as part of a clinical trial for patients to whom Pt 5 applies. 1. First, the Tribunal must determine whether to approve the trial, under s 45AA but that approval does not operate, by itself, as consent to the participation in the trial by any particular patient. 2. Secondly, if the Tribunal is satisfied in accordance with s 45AA(2) that the clinical trial should be approved, the Tribunal then has to determine whether consent to treatment as part of that trial should be given by the "person responsible" or by the Tribunal. 3. Thirdly, depending on whether the consent is to be obtained from the person responsible or the Tribunal, consent must then be obtained under s 40 or under ss 42 to 45, respectively, for the medical treatment in the course of the trial to be carried out on the patient.

Meaning of "Clinical Trial" in the Context of Pt 5

1. As has already been noted, the term "clinical trial" is relevantly defined in s 33(1) as "a trial of drugs or techniques that necessarily involves the carrying out of medical … treatment on the participants in the trial". Read in isolation, that definition is extremely broad and could include every test or study relating to the use of drugs or techniques so long as it involves medical treatment, no matter how old, well accepted or widely available the drug or technique might be. For example, read broadly the definition could include a trial or study as to the differences in the cost or ease of administering a certain well established drug to treat dementia in a hospital environment compared to treatment in a home or nursing home environment.
2. To attempt to construe the words "clinical trial" by looking at the definition in isolation is not, however, appropriate. As explained in Kelly v The Queen [2004] HCA 12; 218 CLR 216 at 253 [103], the proper course is to "read the words of the definition into the substantive enactment and then construe the substantive enactment - in its extended or confined sense - in its context and bearing in mind its purpose and the mischief that it was designed to overcome."
3. In Pt 5, the definition of "clinical trial" plays a substantive role in s 45AA, in the definitions of "special treatment" and "medical … treatment" in s 33(1) and in the offence provision, s 35(1).
4. Reading the relevant words of the definition of "clinical trial" into s 45AA yields a provision which in effect provides, in part: 1. The Tribunal may approve, in accordance with this section, a trial of drugs or techniques that necessarily involves the carrying out of medical treatment on the participants in the trial as a trial in which patients to whom this Part applies may participate; 2. The Tribunal may give an approval under this section only if it is satisfied that: 1. the drugs or techniques being tested in the trial of drugs or techniques that necessarily involves the carrying out of medical treatment on the participants in the trial are intended to cure or alleviate a particular condition from which the patients suffer, and 2. the trial will not involve any known substantial risk to the patients (or, if there are existing treatments for the condition concerned, will not involve material risks greater than the risks associated with those treatments), and 3. the development of the drugs or techniques has reached a stage at which safety and ethical considerations make it appropriate that the drugs or techniques be available to patients who suffer from that condition even if those patients are not able to consent to taking part in the trial, and 4. having regard to the potential benefits (as well as the potential risks) of participation in the trial, it is in the best interests of patients who suffer from that condition that they take part in the trial, and 5. the trial has been approved by a relevant ethics committee and complies with any relevant guidelines issued by the National Health and Medical Research Council. 1. The fact that a trial of drugs or techniques that necessarily involves the carrying out of medical treatment on the participants in the trial will or may involve the giving of placebos to some of the participants in the trial does not prevent the Tribunal from being satisfied that it is in the best interests of patients that they take part in the trial.
5. Section 45AA(2)(b) contains a number of indications that it is not just any testing that is intended to be covered. First, the express reference to "known" substantial risk implies that there may also be unknown risks for patients involved in the trial. This suggests that the trials are not of drugs or techniques that are well accepted or have been used for some time so that the risks associated with them are all well understood.
6. Secondly, the words "if there are existing treatments …" and the assumption that the risks associated with those treatments are sufficiently well understood, so as to be able to be taken into account as envisaged, indicate that the trials to which s 45AA is intended to apply are not trials of "existing treatments" but of treatments (whether drugs or techniques) which are not presently accepted for use in the manner proposed in the trial.
7. The reference in s 45AA(2)(c) to the stage of development which the drugs or techniques being tested has reached is a further indication that the trials to which the section applies are trials of treatments that are still in the developmental stage and do not include trials of fully developed and accepted drugs and techniques. The stage of development which the drugs or techniques should have reached is one where "safety and ethical considerations make it appropriate that the drugs or techniques be available to patients who suffer from that condition even if those patients are not able to consent to taking part in the trial". This is consistent with the trials the subject of s 45AA being limited to those that are of new or experimental treatments.
8. The need for approval by a "relevant ethics committee" and compliance with "any relevant guidelines" of the National Health and Medical Research Council referred to in s 45AA(2)(e) is also consistent with the trials which are the subject of that section being limited to those of drugs or techniques which are not existing or widely accepted treatments or fully developed drugs or techniques.
9. In s 45AA(3) there is the specific provision that the fact that the trial may involve the giving of placebos to some of the participants in the trial does not prevent the Tribunal from being satisfied that it is in the best interests of patients that they take part in the trial. Similarly, in the definition of "medical … treatment" in s 33(1) it is stated that medical treatment "in the course of a clinical trial, is taken to include the giving of placebos to some of the participants in the trial". The use of placebos is common as a control when testing a new drug but would appear to be unlikely to be used in trials involving already widely accepted drugs or techniques. It would be unlikely because using placebos in that context would entail not treating a patient's medical condition in circumstances where there is an established or accepted treatment for that condition and no useful information would be likely to be gained from that form of non-treatment. Thus, this provision gives a further indication that the trials the subject of s 45AA are primarily intended to be trials of new or untested drugs or techniques.
10. The definition of "special treatment" (s 33(1)) is also relevant. In addition to the categories of special treatment relating to permanent infertility, termination of pregnancy, vasectomy or tubal occlusion or aversive stimulus treatment, special treatment covers (under par (b) of the definition) any new treatment that has not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned, except when such treatment is part of a clinical trial.
11. The exclusion of treatment in the course of a clinical trial from "special treatment" makes considerable sense if clinical trials referred to in Pt 5 are those that involve testing new treatments that have not yet gained substantial support from relevant practitioners. Read in this way, treatment in the course of a clinical trial could be seen as a subset of treatment that is new and has not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned.
12. The offences created by s 35, in Pt 5, also provide useful insight in this regard. The offences fall into two categories: 1. Those punishable on conviction on indictment with imprisonment for up to 7 years; 2. Those punishable on summary conviction with imprisonment for up to 1 year or 10 penalty units or both.
13. The offence punishable by imprisonment for up to 7 years is committed where special treatment or treatment in the course of a clinical trial is carried out without consent under Pt 5 or other authorisation under Pt 5 or not in accordance with a Supreme Court order. By way of contrast, unapproved major or minor treatment is only punishable with the lesser penalties. This indicates that the legislature considered that unapproved special treatment and unapproved treatment in the course of a clinical trial were of the same order of seriousness. They both appear to be considerably more serious than unapproved major or minor treatment. The fact that special treatment and treatment in the course of a clinical trial are similarly serious is consistent with their both being different forms of treatment that is new and has not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned. Otherwise, there could be treatment in the course of a clinical trial that did not involve new treatment and which would, as a consequence, be similar to major or minor treatment but which would, however, be punishable on indictment and subject to a much more serious penalty than applies to unauthorised major or minor treatment.
14. One reason why unapproved special treatment falling within par (b) of the definition should be seen as so serious is because subjecting persons who lack capacity to consent to new and not yet accepted treatments, without consent or other authorisation under the Guardianship Act, involves a high risk of exploitation of those persons. Because of their disability, they are open to being exploited in order to test, or experiment with, new treatments.
15. Unapproved treatment in the course of a clinical trial carries the same risk of exploitation where the trials involve drugs or techniques that are new treatments that have not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned. On the other hand, the unapproved treatment of persons to whom Pt 5 applies in the course of a trial involving well known and accepted drugs or techniques is unlikely to involve a similar potential for exploitation and would not generally be seen as being of equal seriousness as unapproved experimentation on persons who lacked the capacity to consent. The degree of seriousness involved in a trial of well understood and generally accepted drugs or techniques is comparable to that of unapproved major or minor treatment.
16. These considerations based on the use of "clinical trial" in the context of s 35(1) are a further indication that clinical trials the subject of s 45AA, in particular, and Pt 5, in general, should be seen as being limited to those of drugs or techniques where the treatment is new and has not yet gained the support of a substantial number of practitioners specialising in the area of practice concerned.
17. In summary, if the guiding principles in s 4 of the Guardianship Act and the objects and provisions of Pt 5 are read as a whole, it can be seen that a construction of the words "clinical trial", in ss 45AA, 33(1), 35(1) and elsewhere in Pt 5, which is limited to trials of drugs or techniques that necessarily involve new medical (or dental) treatment that has not yet gained the support of a substantial number of medical practitioners (or dentists) specialising in the area of practice concerned would be consistent with the text of s 45AA, gives the offence provisions in s 35 and other provisions of Pt 5 an harmonious operation and does not lead to incongruous or absurd results.
18. Limiting "clinical trials" in this way is also consistent with the apparent legislative purposes of the Guardianship Act as revealed in the s 4 principles and in s 32. In particular, such a construction would assist to ensure that persons with impaired decision making capacity: 1. are not exploited in order to test new and experimental drugs and techniques; 2. are not disadvantaged by not having access to new and experimental drugs and techniques that are being trialled, in appropriate cases with appropriate protections; and 3. are not disadvantaged by not having ready access to well understood and standard treatments even in the context of a trial of drugs or techniques necessarily involving medical (or dental) treatment.
19. Whether this construction should be adopted should also be considered in the light of the legislative history of the clinical trial provisions of Pt 5, the relevant extrinsic material, the mischief the introduction of the clinical trial provisions were designed to cure and any previous decisions relating to "clinical trials".

Legislative History, Extrinsic Material and the Mischief

1. When enacted, the Disability Services and Guardianship Act 1987 (NSW), later renamed the Guardianship Act 1987, did not contain the term "clinical trial" or any provisions relating to the approval of clinical trials. Nor were new or non-accepted treatments included in any of the other defined terms, such as "major medical treatment" or "special medical treatment". Originally, the Act as made in 1987 contained a definition as follows: "special medical treatment" means - (a) any medical treatment that is intended, or is reasonably likely, to have the effect of rendering permanently infertile the person on whom it is carried out; or (b) any other medical treatment that is declared by the regulations to be special medical treatment for the purposes of this Part.
2. The Guardianship (Amendment) Act 1993 (NSW) removed the definition of "special medical treatment" in s 33(1) and replaced it with "special treatment" defined as follows; "special treatment" means: (a) any medical treatment that is intended, or is reasonably likely, to have the effect of rendering permanently infertile the person on whom it is carried out; or (b) any new treatment that has not yet gained the support of a substantial number of medical practitioners or dentists specialising in the area of practice concerned; or (c) any other kind of treatment declared by the regulations to be special treatment for the purposes of this Part."
3. That 1993 Act also significantly modified s 45 - Restrictions on the Board's power to give consent. The new s 45(3) provided: (3) In the case of: (a) special treatment of a kind specified in paragraph (b) of the definition of that expression in section 33 (1); or (b) prescribed special treatment (other than special treatment of a kind specified in paragraph (a) of that definition), the Board may give consent to the carrying out of the treatment if it is satisfied that: (c) the treatment is the only or most appropriate way of treating the patient and is manifestly in the best interests of the patient; and (d) in so far as the National Health and Medical Research Council has prescribed guidelines that are relevant to the carrying out of that treatment-those guidelines have been or will be complied with as regards the patient.
4. In the second reading speech for the Guardianship (Amendment) Bill 1993 (New South Wales Legislative Assembly, Parliamentary Debates (Hansard), 21 April 1993, 1379 (James Longley)), the Minister noted: "The definition of special medical treatment is extended to cover any kind of medical or dental treatment, the use of which has not yet gained the support of a substantial number of doctors or dentists specialising in the area of practice concerned. This amendment has the effect of extending the board's scrutiny to all treatments of this kind and, not as is presently the case, only to those which are outside the guidelines of the National Health and Medical Research Council. The Guardianship Board will be able to give consent to such treatments only where the treatment is manifestly in the best interests of the patient and represents the most appropriate alternative form of treatment."
5. It can be seen that, at this time, new treatment of the type referred to in par (b) of the definition of "special treatment" would have included such new treatment carried out as part of a clinical trial.
6. In 1997, a bill, the Guardianship Amendment Bill 1997, was introduced which contained provisions to allow the Tribunal to authorise (in certain circumstances) the participation in clinical trials of persons who lack the capacity to consent to their own medical and dental treatment. These provisions, however, were removed by the Legislative Council and subsequently referred to the Legislative Council Standing Committee on Social Issues (New South Wales Legislative Assembly, Parliamentary Debates (Hansard), 8 April 1998, 3847 (Faye Lo Po')). As it happens, they were later reintroduced in the Guardianship Amendment Bill 1998, which was passed by both Houses on 5 May 1998, becoming the Guardianship Amendment Act 1998 (NSW).
7. The Report of the Standing Committee on Social Issues entitled "Clinical Trials and Guardianship: Maximising the Safeguards" contained the following: "The potential involvement of people with decision-making disabilities in a trial involving a placebo was the impetus for developing the amendments relating to clinical trials, as explained by the President of the Guardianship Board: 'The reason the amendments were sought was that the Board was being asked to give consent to new treatments that were available only through clinical trial. We had anxieties about them. … Our anxieties are based on the fact that some people would be getting a placebo and not the actual treatment. It is very clear that where treatment is new and we know a person is being given that new treatment, we can give consent, provided certain substantial criteria are met. However, we were concerned that we needed statutory clarification for the placebo issues (O'Neill Evidence, 8 July 1997).'"
8. Among other changes, the 1998 Act inserted into the Guardianship Act the definition of "clinical trial" and ss 45AA and 45AB, which govern the Tribunal's approval of clinical trials, as well as making amendments to refer to clinical trials in the definitions of "special treatment", "major treatment" and "minor treatment" and in the offence provisions of s 35.
9. The Minister's second reading speech in relation to the 1998 Act made extensive reference to new treatments and "treatments available only through clinical trials" including the following (New South Wales Legislative Assembly, Parliamentary Debates, 8 April 1998, 3847-3848 (Faye Lo Po')): "The Guardianship Amendment Bill 1998 reintroduces those provisions in the Guardianship Amendment Bill 1997 which ensure that people who cannot consent to their own treatment are not denied access to new treatments available only through clinical trials. … The Government has been of the view, and now the all-party Standing Committee on Social Issues of the Legislative Council is unanimously of the view, that people who cannot consent to their own treatment should have access to new treatments, provided that certain stringent safeguards are met. … These provisions are primarily about giving previously healthy people access to new treatments for conditions that take away their capacity to live normal lives in the community. … The provisions in relation to access to treatments available only through clinical trials will be inserted into part 5 of the Guardianship Act… Without this legislation those who cannot consent to their own treatment will be denied access to new treatments that will manifestly be of benefit to them, only because they have lost the capacity to consent." (emphasis added)
10. After noting that "[s]ince the Guardianship Act came into force in 1989 it has always been possible to obtain a valid substitute consent to give new treatment", the Minister expressly identified the mischief that the introduction of those provisions was designed to overcome (New South Wales Legislative Assembly, Parliamentary Debates, 8 April 1998, 3847 (Faye Lo Po')): "The difficulty to be cured by this bill is as follows: the Therapeutic Goods Administration, the Commonwealth Authority responsible for overviewing the importation of medications into Australia, has been insisting that many new treatments should be available in Australia only through clinical trials in which most of the participants will get the new treatment but some, the control group, will not. In order for such clinical trials to be scientifically valid, they have to be double-blinded. This means that neither the treating doctors and nurses nor the trial administrators know which participants are receiving the treatment and which are in the control group. This means it is impossible to predict those participants in the trial who will receive the treatment and those who will not. In cases where there is no existing treatment, some participants will receive a placebo only. In cases where there is an existing treatment, some will get that treatment and not the new treatment. As I said, the present substitute consent provisions in relation to new treatment in the Guardianship Act are based on the premise that a person will receive the new treatment. Because of the therapeutic goods administration policy - and there is no argument with that policy - people can gain access to many new treatments only if they join a clinical trial, but it cannot be guaranteed that they will receive the treatment. If they do not join the clinical trial they have no chance whatsoever of getting the treatment."
11. Both the extensive references to "new treatments", and the mischief that the bill was designed to overcome, strongly support the conclusion that the clinical trial provisions of the Guardianship Act, including s 45AA and s 35, as well as the definitions of various types of treatment in Pt 5, should be construed so that "clinical trials" are limited to those which necessarily involve new medical or dental treatment that has not yet gained the support of a substantial number of medical practitioners or dentists specialising in the area of practice concerned.
12. The effect of the 1998 amending Act was to carve clinical trials out of new treatments that would fall within par (b) of the definition of "special treatment" in s 33(1), so that a different approval regime could be created in ss 45AA and 45AB which would allow persons with impaired decision making capacity to have the benefit of participating in trials of new drugs or techniques, including those involving placebos, subject to appropriate safeguards and protections against exploitation or risk of harm. It does not appear that it was Parliament's intention to broaden the scope of "clinical trials" beyond new treatment (falling within par (b) of the definition of "special treatment") to which, in some but not all cases, the Guardianship Board had previously been able to consent under s 45(3)(a) (in its pre-1998 form).
13. Accordingly, in our view, the legislative history, the extrinsic material and the mischief to be cured confirm that the proper construction of "clinical trial" in ss 35(1) and 45AA and elsewhere in Pt 5 is that this expression refers to a trial of drugs or techniques that necessarily involves the carrying out of new medical (or dental) treatment that has not yet gained the support of a substantial number of medical practitioners (or dentists) specialising in the area of practice concerned.

Previous Decisions of the Tribunal

1. Finally we turn to consider the previous decisions of the Tribunal (or its predecessor) dealing with the construction of "clinical trial" in Pt 5.

NICE Study Case (23 November 2004)

1. In respect of the study called "An Multi-Centre, Open Label, Randomised, controlled trial of two target ranges for glycaemic control in Intensive Care Unit Patients" (the NICE Study), the Guardianship Tribunal found that the "proposed study did not constitute a clinical trial under the provisions of Section 33(1) of the Guardianship Act." The Guardianship Tribunal reasoned: "Prior to the 1998 amendments to the Guardianship Act, clinical trials were considered "special treatment" under Section 33(1)(b). That provided special treatment included "any new treatment that has not yet gained the support of a substantial number of medical practitioners or dentists specialising in the area of practice concerned." Amendments to the Guardianship Act were introduced to clarify that people could be enrolled in clinical trials that may be placebo-controlled, in which there was no guarantee the patient would receive the treatment being studied. Further, the purpose of the amendments to the Guardianship Act was to ensure that people with disabilities did not miss out on access to new treatments that may be of benefit to them, that were only available in the context of a clinical trial. … On its face this definition does not appear to limit clinical trials to those circumstances where new or experimental treatments are being evaluated. However, the Oxford Dictionary defines "trial" to include "experimental treatment". The Tribunal considered that the word "trial" in Section 33(1) in the context of Part 5 of the Guardianship Act and in particular, Division 4A, implied a test of new or experimental drugs or procedures. … The Tribunal was not satisfied that the provisions of the Guardianship Act relating to the requirements for the Tribunal to approve clinical trials before they can include people who cannot provide consent to their own treatment were intended to require the Tribunal to review all clinical studies that involve patients who cannot provide their own consent to treatment regardless of the type of medical treatment or procedure sought to be analysed in the conduct of the particular study. The Tribunal preferred the view that the legislative provisions should be interpreted more narrowly to provide that a clinical trial includes the assessment of new medical treatments or procedures that seek to assess non-standard treatments or a non-standard application of an accepted treatment."
2. Although the reasoning of the Guardianship Tribunal is not entirely spelt out, the reasons and the conclusion are consistent with the view we have reached above.

ARISE Study Case (5 August 2008)

1. The Guardianship Tribunal also considered whether to approve as a clinical trial the study called "An multi-centre randomised controlled trial of early goal-directed therapy in patients presenting to the Emergency Department with severe sepsis in Australasia" (the ARISE Study). That tribunal held that the study was a clinical trial in which patients to whom Pt 5 applies could participate under s 45AA of the Guardianship Act. After noting the definition of "clinical trial" and "medical treatment", the Guardianship Tribunal stated: "The Guardianship Act does not define trial any further. According to the Oxford Dictionary, trial can be defined as a method or procedure used to find a result, test a prediction or evaluate a hypothesis or as experimental investigation or treatment. In deciding whether a trial is a clinical trial in terms of the Guardianship Act, the Tribunal takes into account the way that clinical trials were dealt with prior to the 1998 amendment of that Act. At that time, clinical trials were considered as special treatment under section 33(1)(b). Special treatment included any new treatment that has not yet gained the support of the substantial number of medical practitioners or dentists specialising in the area or practice concerned. Amendments to the Act were introduced to clarify that people could be enrolled in clinical trials that may be placebo controlled, in which there was no guarantee the patient would receive the treatment being studied. Further, the purpose of amending the Act was to ensure that people with disabilities did not miss out on access to new treatments which may be of benefit to them but that were only available in the context of a clinical trial."
2. The Tribunal did not apparently elaborate any further on the proper construction of "clinical trial" as it appears in the provisions of Pt 5. After having "regard to the general objectives of Part 5", the Tribunal concluded: "The Tribunal is satisfied that the current proposal before it does come under the definition of a clinical trial. Whilst it is referred to as a study, it includes a new method of delivering therapeutic treatment by medical practitioners. It is also a method or procedure used to find a result, test a prediction or evaluate a hypothesis and can be seen as an experimental investigation or treatment. Whilst it is a method of delivering the same drugs or treatment, it involves a new way of doing so."
3. In this case, the Tribunal's emphasis on the "new method of delivering" is consistent with "clinical trials" being limited to those where the treatment was new and had not yet gained the support of a substantial number of practitioners or specialising in the area of practice concerned. It is consistent with the view we have reached.
4. Somewhat curiously, the decision in the ARISE Study Case was overturned by the Guardianship Tribunal on 22 December 2008. The orders dated 5 August 2008 were replaced with the following order: "The Tribunal is not satisfied that the above-named clinical trial, is a clinical trial within the meaning of sections 33(1) and section 45AA of the Guardianship Act."
5. We have not found the reasons for decision which explain why that order was made. Nonetheless, it is unlikely that the Guardianship Tribunal would have made such a new order if it took a wider view of the meaning of "clinical trial" than the previous position adopted when the earlier orders were made.

AMOUNT Rehabilitation Trial Case (06 January 2015; [2015] NSWCATGD 1)

1. When the Guardianship Division of the Tribunal considered "a randomised trial of the effect of affordable technology on physical activity levels and mobility outcomes in rehabilitation: The AMOUNT Rehabilitation Trial" (AMOUNT Rehabilitation Trial) it "was satisfied that the proposed trial was a clinical trial within the meaning of the Guardianship Act". The Tribunal reasoned (at [34]-[35]) as follows: "As already noted, the definition of clinical trial is very broad. It covers any medical or dental treatment and is not confined to experimental or intrusive procedures. The definition of treatment specifically excludes identified procedures such as non-intrusive examination for diagnosis purposes and first-aid medical and dental treatment. Relevantly, s 33(1)(g) provides for the regulations to exclude certain treatment from the clinical trial regime. It is therefore clear that Division 4A is intended to operate broadly. If the legislature had intended to confine the clinical trial regime it could have easily done so by excluding research or non-intrusive procedures from the definition of treatment. Similarly, such procedures could be excluded by regulation. There is no such exclusion in the Guardianship Regulation 2010 (NSW). The language used in Part 5 is unambiguously wide. Accordingly, there is no need or justification for reading down the provisions or seeking guidance from extrinsic sources (Alcan (NT) Alumnia Pty Ltd v Commissioner of Territory Revenue [2009] HCA 41; (2009) 239 CLR 27, [47]). A rehabilitation specialist may decide to use affordable technologies as part of a person's tailored rehabilitation plan. This would be minor medical treatment which could be approved by a person responsible if the person who is the subject of the proposed treatment could not themselves consent. It would not be a clinical trial, even if the specialist decided to record the outcome as part of a research project, unless the treatment also involved a test or evaluation of the techniques being used. It is implicit that evaluation involves comparison with other techniques. Randomisation is the most common procedure used to evaluate a drug or technique. Once there is randomisation of treatment, the treatment must necessarily be part of a trial. This is because the health care professional is not recommending that 'a particular course of treatment should be carried out', as provided by ss 40(2)(f) and 42(2)(f), but rather that the person participate in the trial. While participation may also involve consent to particular treatment, the administration of a placebo or consent to alternative treatment as randomised, the recommendation is not focussed on a particular course of treatment but on the trial."
2. The reasoning of the Tribunal in this case does not appear to have taken into account the provisions of Pt 5 other than the definition in s 33(1). It appears to us that the Tribunal may have fallen into the error identified by McHugh J in Kelly v The Queen [2004] HCA 12; 218 CLR 216, referred to above. For that reason and the other reasons we have given, we do not think that the approach or the reasoning in the AMOUNT Rehabilitation Trial Case should be followed.

TRANSFUSE Trial Case (03 June 2015; [2015] NSWCATGD 18)

1. In respect of a study known as "multi-centre randomised double blinded phase III trial of the effect of standard issue red blood cell blood units on mortality compared to freshest available red blood cell units" (the TRANSFUSE Trial), the Guardianship Division of the Tribunal "determined that TRANSFUSE is neither a trial of drugs nor a trial of techniques" and hence "does not fall within the definition of 'clinical trial' in section 33(1) of the Guardianship Act." (at [39]-[46]). The decision, to the extent that it has any relevance in the present case, is not inconsistent with the view we have reached.

ADRENAL Trial Case (19 June 2015; [2015] NSWCATGD 23)

1. Prof Shehabi placed particular emphasis upon the Guardianship Division's decision in respect of a study known as "randomised blinded placebo controlled trial of hydrocortisone in critically ill patients with septic shock" (the ADRENAL Trial). In that case, the Guardianship Division of the Tribunal held that "the ADRENAL trial does not fall within the definition of 'clinical trial' in section 33(1) of the Guardianship Act 1987 (NSW)." Significantly, the Tribunal considered substantially the same issues as arise in the present appeal.
2. In ADRENAL, the Tribunal noted the reasoning and conclusion reached in the SPICE III Case (the decision the subject of this appeal), but distinguished that case on the basis that "ADRENAL is not a trial involving a drug to be administered beyond its TGA registration". The Tribunal therefore undertook "an examination of the context and legislative purpose of the clinical trial provisions" (see [106]).
3. The Tribunal noted a number of considerations in favour of a broad construction of "clinical trial", including the breadth of the definition of "clinical trial" in section 33(1) and the definition of "special treatment" in section 33(1): [111] "The legislation makes explicit reference to 'new treatment' in the relevant part of the definition of 'special treatment'. It is arguable that if the legislature had intended that the clinical trial provisions in Division 4A have application only to drugs or techniques that are new treatments, then it could have made this explicit and adopted the same or similar wording as the definition of 'special treatment' and made reference to 'new treatment'. [In the Tribunal's view, the import of the language in the definition of 'special treatment' is not to be lightly discarded given that the employment of different language in the same Act may show that the legislature had in view different objects (Scott v Commercial Hotel Merbein Pty Ltd [1930] VLR 25, [30] (Irvine CJ)).]"
4. These considerations, however, were outweighed in the Tribunal's view by a textual reading of the provisions relating to clinical trials in Part 5 (see [114]-[121]). The Tribunal concluded (in part, at [125]): "An examination of the provisions in section 45AA(2) supports this narrower view of the definition of 'clinical trial'. As previously outlined, 45AA(2)(c) would appear to only have meaning and effect if what is under consideration is a trial of drugs or techniques that are not accepted forms of treatment for the condition in question. Sub-sections (2)(b) and (2)(d) of section 45AA section 45AA(2) also lend weight to this construction of 'clinical trial'."
5. The Tribunal also considered the legislative purpose of the provisions relating to clinical trials by examining the relevant extrinsic material (at [123]-[124]): "The Second Reading Speeches associated with the Guardianship Amendment Bill indicate that section 45AA and the other clinical trial provisions were introduced primarily as a result of the difficulties posed by clinical trials involving placebos. Prior to the amendments, consent could be sought for treatment that fell within the definition of 'special treatment' in section 33(1)(b) as 'any new treatment that has not yet gained the support of a substantial number of medical practitioners or dentists specialising in the area of practice concerned'. However, an application for such treatment could only be approved by the then Guardianship Tribunal if, amongst other things, it was satisfied that the patient would in fact receive the treatment. The amendments clarified that, in relation to clinical trials in which placebos were included, a person with a decision making disability could still be included in the trial. Relevantly, the Second Reading Speech makes reference to 'new treatments' and 'that people who cannot consent to their own treatment should have access to new treatments provided certain stringent safeguards are met.'"
6. The analysis of s 45AA and the legislative purpose is consistent with the reasoning of the Appeal Panel in the present case, as set out above, although we differ slightly from the Tribunal in the ADRENAL Trial Case by preferring a construction formulated by reference to the wording in par (b) of the definition of "special treatment", in the light of s 35(1) and Pt 5, read as a whole, the legislative history relating to the introduction of the concepts of special treatment and clinical trials into the Guardianship Act and the apparent Parliamentary intention.
7. In summary, the previous decisions of the Tribunal dealing with "clinical trials" under Pt 5 provide no reason to depart from our view as to the proper construction which we have indicated above. On the contrary, a number of those decisions, especially the ADRENAL Trial Case, support the conclusion we have reached.

Conclusion on First Reformulated Ground of Appeal

1. For these reasons, in our view, the proper construction of the expression "clinical trial", as used in Pt 5 of the Guardianship Act, refers to a trial of drugs or techniques that necessarily involves the carrying out of new medical (or dental) treatment that has not yet gained the support of a substantial number of medical practitioners (or dentists) specialising in the area of practice concerned.
2. In so far as the Tribunal below did not adopt that construction, it erred in law. On that basis, we would uphold the appeal on the first reformulated ground of appeal.

Second Reformulated Ground of Appeal - Was the SPICE III Study a "Clinical Trial"?

1. The second reformulated ground involves the contention that on the facts as found by the Tribunal and on the proper construction of the expression "clinical trial" in Pt 5, the SPICE III study is not a clinical trial that could be approved under s 45AA.
2. More specifically, it was contended that: 1. the treatments (both drugs and their mode of use) used in both arms of the SPICE III study are existing, accepted treatments for patients in intensive care and consequently the study was not a "clinical trial"; and 2. the fact that dexmedetomidine is used in the SPICE III study in a randomised trial outside its approved product information, and consequently notification under the TG Act is required, does not mean that the SPICE III study is a trial of drugs or techniques that are new or experimental or treatments that are new or non-standard and thus a "clinical trial".
3. We observe here that our construction of "clinical trial" as set out above does not employ exactly the same words as the construction for which Prof Shehabi contended or as used by the Tribunal in the ADRENAL Trial Case. Nonetheless, we do not think that there is any difference in substance between: 1. A trial of drugs or techniques involving new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned, as referred to in our construction of "clinical trial" as used in Pt 5; and 2. A trial of drugs or techniques that are new or experimental or treatments that are new or non-standard or expressions to a similar effect, as used in ADRENAL decision, the oral argument during the hearing and the written submissions of the parties.
4. We prefer the first formulation in the preceding paragraph as it uses expressions and concepts already found in the Guardianship Act and it implicitly acknowledges that treatment in the course of clinical trials for the purposes of Pt 5 was carved out of special treatment by the 1998 amendments and treatment in the course of these trials would still fall within par (b) of the definition of "special treatment" if it were not for its express exclusion from that definition in s 33(1) because of the words "but does not include treatment in the course of a clinical trial".
5. This second reformulated ground, therefore, requires us to consider whether: 1. On the facts as found by the Tribunal below the SPICE III study was a trial of drugs or techniques involving new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned; and 2. Because dexmedetomidine was being used in a randomised trial outside its product information approved by the TGA and thus the SPICE III study was required to be notified under the TG Act, the SPICE III study was a trial of drugs or techniques involving new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned.
6. In his written submissions, Prof Shehabi contended that the "two arms of the study use different combinations of sedatives that are commonly and routinely used by medical practitioners on ICU patients". He also drew the Appeal Panel's attention to the Tribunal's recognition at first instance that the proposed use of dexmedetomidine was "in accordance with existing treatments".
7. In relation to the use of dexmedetomidine beyond its approved product information and the engagement of the TG Act's CTN Scheme, Prof Shehabi's submitted that such approval is a "marketing approval", that there is "high level evidence" that supports the off-label use of dexmedetomidine and that there is "no provision to allow the TGA to recommend changes to the registration or indication of a registered therapeutic good, even if there is substantial medical evidence to support a different use of the good." As we understood them, these submissions concerned the fact that where therapeutic goods are registered under the TG Act, the Secretary must approve product information in relation to those goods, under s 25AA of the TG Act. In this context, "product information" means "in relation to therapeutic goods, … information relating to the safe and effective use of the goods, including information regarding the usefulness and limitations of the goods": TG Act s 3(1). Prof Shehabi was submitting that the fact that product information is so approved does not mean that the product in question cannot be used otherwise than in accordance with that product information.
8. The Attorney General, in her role as contradictor, submitted that it was "open to the Tribunal to take into consideration that the study proposed to use a drug outside the terms of its TGA registration, and the TGA marketing approval [approved product information] of the drug" and otherwise developed the submission that there was no error in the reasoning or findings of the Tribunal below.

Findings of the Tribunal Below concerning SPICE III Treatment

1. The Tribunal at first instance, in its Reasons, made a number of relevant findings concerning the SPICE III study.
2. The findings of the Tribunal included that the hypothesis sought to be tested in the SPICE III study is that early goal directed sedation, compared to standard care sedation, reduces 90-day all-cause mortality in critically ill patients who require mechanical ventilation for longer than 24 hours - Reasons [3].
3. The sedative agents, being midazolam, propofol, dexmedetomidine, used in both arms (that is, early goal directed sedation and standard care sedation) individually or in combination are accepted sedatives currently used for mechanically ventilated intensive care unit (ICU) patients - Reasons [4], [43] - [45] and [104].
4. The exclusion criteria applied as part of the SPICE III study ensure that patients who have a known sensitivity to any of the study medications or constituents of propofol will be excluded from the study. Patients with other specified conditions (including cardiovascular conditions, acute fulminant hepatic failure and proven or suspected acute primary brain lesion) will also be excluded - Reasons [105].
5. One of the agents, dexmedetomidine, has been used as a sedative agent in ICUs for a number of years. The relevant approved product information specifies that it can be used for sedation of initially intubated patients during treatment in an intensive care setting for up to 24 hours. It can also be used for sedation of non-intubated patients prior to and/or during surgical or other procedures - Reasons [5] and the approved Product Information in respect of Precedex, which is dexmedetomidine hydrochloride, (Version 5.0 p 8 under the heading "INDICATIONS"), a copy of which was included in the material before the Appeal Panel.
6. Patients randomised to the early goal directed sedation arm of the SPICE III study will receive a sedative infusion of dexmedetomidine commencing without a loading dose at a rate of 1.0 mcg/kg/hour and will be varied between 0-1.0 mcg/kg/hour to maintain a light sedation. Dexmedetomidine infusion will be continued until sedation is no longer clinically indicated up to a maximum of 28 days after enrolment - Reasons [102].
7. Thus, the SPICE III study proposes the use of dexmedetomidine for a period longer than 24 hours and is thereby proposed for use outside of its approved product information, referred to by the Tribunal below as its "TGA licensed parameters" - Reasons [6].
8. The use of dexmedetomidine has increased to such an extent that it is used regularly and widely as an alternative sedative agent to benzodiazepines in ICUs in NSW and Australia. The evidence, accepted by the Tribunal, was that physicians regularly administer dexmedetomidine for individual patients as long as is clinically indicated, which may be for longer than 24 hours, and at dosages greater than 1mcg/kg/hr - Reasons [103] and see [48].
9. Further, the Tribunal accepted (at Reasons [104]) the overview of evidence relating to the safety and efficacy of dexmedetomidine set out in the Study Protocol (pp 13 and 14) referring to the results of a number of Phase II clinical randomised control trials and, notably, to a 2012 randomised control trial (Propofol vs dexmedetomidine and Midazolam v dexmedetomidine, PRODEX and MIDEX) that: "confirmed the safety and efficacy of dexmedetomidine as an alternative sedative drug for mechanically ventilated patients for longer than 24 hours. It also confirmed the safety of dosing dexmedetomidine up to 1.5µg/kg/hour. These studies have led to the registration of dexmedetomidine in Europe for use in ICU sedation to a maximum dose of 1.5µg/kg/hour with no time restriction. Based on the body of evidence available to-date, the most recent 2012 International Sedation Clinical Practice Guidelines produced by a special taskforce of the Society of Critical Care Medicine suggested that midazolam, propofol, dexmedetomidine or combinations be used for ICU sedation in mechanically ventilated critically ill patients."
10. The evidence accepted by the Tribunal below also included a longitudinal prospective cohort study on sedation practice in Australia and New Zealand conducted in 2010 as part of the SPICE Project which showed that Australian and New Zealand intensive care clinicians used midazolam and/or propofol and/or dexmedetomidine in 66.7%, 80.1% and 27% of patients respectively, commonly in combination. The Study Protocol (p 8) noted the use of dexmedetomidine is unlikely to have decreased since that study was conducted in 2010 - Reasons [36].
11. There was also evidence (which was accepted and relied upon by the Tribunal at Reasons [93]) from the Chairman of the South Eastern Sydney Local Health District (SESLHD) Human Research and Ethics Committee (HREC), Professor James Isbister, that (at Reasons [55]): "In August 2013 the SESLHD HREC reviewed and approved protocol version 3 dated June 24 2013. The HREC also reviewed supporting documents including previous randomised controlled trials of dexmedetomidine, propofol and midazolam and current international guidelines. The HREC was satisfied that the study protocol is consistent with current and standard practice for sedation of mechanically ventilated patients in an intensive care unit. The HREC acknowledges that the study inclusion and exclusion criteria are in line with safety profile of all sedative agents used including dexmedetomidine."
12. The Tribunal below also "placed weight on the evidence", and, we understand, accepted, that since 2011, the use of dexmedetomidine has increased to such an extent that it is used regularly and widely as an alternative sedative agent to benzodiazepines in intensive care units in NSW and Australia by physicians treating the clinical needs of individual patients - Reasons [93]. That evidence was summarised at [56] in the reasons as follows: "• The applicant gave evidence that in Australia, dexmedetomidine is used primarily in ICUs for sedation and very little use is made of it outside ICUs. • The applicant attributed the increased utilisation of dexmedetomidine over this four year period to the results of two pivotal randomised control trials. The first was conducted in 2009 and compared the use of dexmedetomidine and midazolam. The second rise in useage is, according to the applicant, attributable to the published results of European studies that compared the use of dexmedetomidine and midazolam with dexmedetomidine and propofol. • According to the applicant, the increased use of dexmedetomidine in ICUs is also attributable to the guidelines for sedation practice published in 2013 by the US Society of Critical Care. These guidelines listed dexmedetomidine, midazolam and propofol as equal agents to be used for the sedation of critically ill patients. • In August 2014, the ANZICS Clinical Trials Group conducted a survey amongst the nine proposed sites at which the current study would be carried out. In response to a question from the Tribunal, Dr Shehabi confirmed that the study was specifically carried out for the purposes of the hearing on 15 October 2014. • According to Dr Shehabi, the results of the survey showed that: - The default target sedation in all patients in the nine ICUs was light sedation - Seven of the nine ICUs use the following three agents: dexmedetomidine, propofol and midazolam - Two of the nine ICUs use dexmedetomidine and propofol and another, with midazolam used as a rescue therapy if needed - All ICUs reported a significant increase in the use of dexmedetomidine in their ICU over the last two years. In addition: The dose range in all units is ≥ 1mcg/kg/hr All units use dexmedetomidine for longer than 24 hours and as long as clinically indicated - Once a patient is critically ill and ventilated, whether they are a ventilated surgical ICU patient or a ventilated medical ICU patient, the approach to sedation in the nine sites is the same. According to Dr Shehabi, this indicates that whether the patient comes to the ICU with a surgical background or with a medical background, the survey results indicate that there is no difference in terms of the choice of sedation medication or intensity - Current practice in ICUs mirrors the SPICE III protocol"
13. There were no findings that tended to indicate that the drugs or techniques used in either arm of the SPICE III study were new or experimental or had not yet gained the support of a substantial number of medical practitioners specialising in sedation of intensive care patients.

SPICE III Treatment Not New, Unsupported Medical Treatment

1. In these circumstances, the Appeal Panel is satisfied that the drugs used in the SPICE III study and the manner of their usage (including the use of dexmedetomidine for more than 24 hours), to the extent that it constitutes a technique, are now well accepted or standard treatments for patients in intensive care that have the support of a substantial number of medical practitioners specialising in sedation of intensive care patients. This leads us to conclude that the SPICE III study is not a trial of drugs or techniques that necessarily involves new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned.

Was the SPICE III Study a "Clinical Trial" because of Randomisation and TGA Notification?

1. It is now necessary to consider whether, because dexmedetomidine was being used in a randomised trial outside its approved product information and thus the SPICE III study was required to be notified under the TG Act, the SPICE III study should therefore be found to be a "clinical trial" for the purposes of Pt 5 of the Guardianship Act, notwithstanding our conclusion in the preceding paragraph.
2. Off label use of dexmedetomidine in a randomised trial is significant in the present appeal because the Tribunal below concluded that "whilst dexmedetomidine retains its current TGA registration [approved product information] and is therefore captured by the clinical trial notification scheme of the TG Act, then the SPICE III trial, which proposes to use the drug beyond its licensed parameters [approved product information] in the context of a randomised trial, amounts to a 'clinical trial' for the purposes of the Guardianship Act" (Reasons [98]). Prof Shehabi contended that this reasoning was erroneous. The contradictor in this appeal, entirely appropriately, sought to uphold the Tribunal's decision in this regard.

Findings of the Tribunal Below concerning the SPICE III Study Randomisation and TGA Notification

1. The Tribunal found that in the SPICE III study the role of randomisation was to allocate patients to one of the two arms of the study as follows: "[58] … According to the Study Protocol (at page 19), patients who satisfy the inclusion criteria and have no exclusion criteria 'will be randomly assigned in a 1:1 ratio to either Early Goal Directed Sedation or to standard care sedation using a block randomisation with variable block size'. [59] Randomisation 'will be conducted through a password protected, secure website using a central, computer based randomisation program. Treatment allocation will be stratified by site and by whether there is presence or absence of suspected/proven sepsis'."
2. In relation to randomisation, the Appeal Panel understands that the Tribunal below accepted what was referred to as the "equipoise" of the SPICE III study - Reasons [97]. In this context it appears to us that the Tribunal below used the term "equipoise" to refer to the following circumstances or approach (see Reasons [62] ff): 1. "before a patient's entry into the study and randomisation, the physician must believe that neither one of the protocols would be better for the particular patient and that both treatments are equivalent in his or her mind for the particular patient" - Reasons [63]; 2. "either treatment arm is equivalent for the particular patient" - Reasons [64].
3. The Tribunal also noted (at Reasons [66]) that: "Professor James Isbister, the Chairman of the SESLHD HREC, advised that the HREC was very happy with the question of equipoise in this study and the survey conducted in August 2014 confirmed that the clinicians who administer these treatments are also very comfortable that there is no difference between the two treatment arms."
4. The Tribunal below understood that the SPICE III study involved an approach to randomisation and equipoise as follows (Reasons [64]): "a decision is in fact made about the treatment for a particular patient and … this occurs prior to the act of randomisation. The relevant decision is that either treatment arm is equivalent for the particular patient."
5. The Tribunal below also made findings, which were not in dispute before the Appeal Panel, concerning the national framework relevant to "clinical trials" as that term is defined in the National Health and Medical Research Council's National Statement on Ethical Conduct in Human Research 2007 (as updated from time to time) (the NHMRC National Statement). It may be noted here, as a matter of interest, that the NHMRC National Statement contains the guidelines referred to in s 45AA(2)(e) of the Guardianship Act.
6. The Tribunal found (Reasons [28] ff) that part of the national framework relevant to NHMRC clinical trials is the clinical trial notification scheme and the clinical trial exemption scheme, created pursuant to the TG Act. Under the TG Act, therapeutic goods for human use that are imported, manufactured in Australia, supplied by a corporation, supplied interstate or to the Commonwealth, or exported must be included in the Australian Register of Therapeutic Goods (ARTG). However, some therapeutic goods are exempted under the TG Act from the requirement for inclusion in the ARTG. This includes therapeutic goods for use in clinical trials. The exemption is achieved by way of the clinical trial notification scheme (CTN) or the clinical trial exemption scheme (CTX).
7. Pursuant to the TG Act and Regulations, either a notification under the CTN scheme or an application under the CTX scheme is required for all clinical investigational use of a product where that use involves, relevantly: "use of a product beyond the conditions of its marketing approval, including new indications extending the use of a medicine to a new population group and the extension of doses or duration of treatments outside the approved range."
8. Because dexmedetomidine is proposed to be used for more than 24 hours, outside its approved product information in the SPICE III study, notification of the study was provided to the TGA under the CTN. Under the CTN scheme, the use of such products is referred to as being used "for experimental purposes in humans" - Therapeutic Goods Regulations 1990 (Cth), Schedule 5A, item 3.
9. The Tribunal further found that under the CTN scheme, the TGA does not review any data relating to the trial the subject of the notification. Rather, the HREC is responsible for assessing the scientific validity of the trial design, the safety and efficacy of the medicine or device and the ethical acceptability of the trial process, and for approval of the trial protocol. In some institutions a scientific review or drug subcommittee may review the proposal before consideration by the HREC. The institution or organisation at which the trial will be conducted, referred to as the "Approving Authority", gives the final approval for the conduct of the trial at the site, having due regard to advice from the HREC.

Consideration of the Implications of Randomisation and TGA Notification

1. The Appeal Panel accepts that the SPICE III study involves: 1. the randomised participation in the study of patients to whom Pt 5 applies; and 2. those patients receiving "medical treatment" because sedative agents are being administered to them.
2. The randomised nature of participation indicates that a trial or test, in the broad sense, is being conducted. The trial or test is examining the consequences of the use of different drugs or combinations of drugs. From this it could be concluded that the practitioners carrying out the SPICE III study are conducting a test or trial of drugs or techniques that necessarily involves medical treatment. It does not follow from this, however, that the SPICE III study is a "clinical trial" for the purposes of Pt 5.
3. To conclude that because the SPICE III study is a trial of drugs or techniques that necessarily involves medical treatment it is a "clinical trial" involves the error of reading the definition of "clinical trial" in s 33(1) in isolation and without regard to the specific use of that expression in ss 35 and 45AA, the context provided by Pt 5 and the purposes of the Guardianship Act as a whole, the legislative history of Pt 5 and the mischief that the "clinical trial" provisions were designed to address. As we have shown above, the proper construction of the expression of "clinical trial" for the purposes of the Guardianship Act is more limited and only includes a trial of drugs or techniques that necessarily involves new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned. Thus, the randomisation involved in the SPICE III study alone does not provide a sufficient reason to conclude that the study is a "clinical trial" for the purposes of Pt 5 of the Guardianship Act.
4. Turning now to the consequences of the fact that the SPICE III study was required to be notified to the TGA, the Appeal Panel considers that there are a number of reasons why a requirement for notification under the TG Act does not determine whether or not a study is a "clinical trial" for the purposes of the Guardianship Act. First and most significantly, the purpose and structure of the notification and exemption schemes under the TG Act are entirely different from, and largely unrelated to, the purpose and provisions of Pt 5 of the Guardianship Act.
5. Secondly, the expression "clinical trial" is not defined in the TG Act, although it is used in ss 24 and 32DH in the definition of "individual patient data". Nonetheless, the TG Act does refer to the National Health and Medical Research Council (NHMRC) in relation to ethics committees. The NHMRC has given a definition of "clinical trial" in the NHMRC National Statement (Ch 3.3 p 30) but it is in very different terms from the definition of "clinical trial" in s 33(1). Further and in any event, the NHMRC definition is not a legislative definition designed to assist in determining the proper construction of the Guardianship Act but is, rather, an explanation of what the NHMRC is seeking to cover or address in its guidelines. The NHMRC definition is as follows: "A clinical trial is a form of human research designed to find out the effects of an intervention, including a treatment or diagnostic procedure. A clinical trial can involve testing a drug, surgical procedure, other therapeutic procedures and devices, a preventive procedure, or a diagnostic device or procedure."
6. In our view, whether or not a study is required to be notified under the TG Act and whether or not it falls within the NHMRC's explanation of "clinical trial" cannot be determinative of whether the study is a "clinical trial" for the purposes of the Guardianship Act.
7. As Prof Shehabi explained, and the Tribunal below accepted, use of dexmedetomidine for more than 24 hours, although beyond its approved product information and thus requiring notification of the SPICE III study under the CTN scheme, is not a use that is new. Nor is it a use that does not have the support of a substantial number of medical practitioners specialising in the area of practice concerned. We have referred to the findings of the Tribunal in this regard above.
8. It may be the case that a particular study or trial is required to notified or exempted under the TG Act because it involves new drugs or techniques which have not gained the supported of a substantial number of practitioners specialising in the relevant area. In such a case, the study may also be a "clinical trial" within the meaning of Pt 5 of the Guardianship Act and thus require approval under s 45AA if patients to whom Pt 5 applies are to participate. There is, however, no necessary connection between notification or exemption under the TG Act and whether the study is a "clinical trial" for the purposes of the Guardianship Act.

Conclusions on Second Reformulated Ground of Appeal

1. For the purposes of an application for approval of a study or trial under s 45AA of the Guardianship Act, the answer to the question whether it is or is not a "clinical trial" falling within that section depends on the proper construction of "clinical trial" in the Guardianship Act and, thus, on whether the study is a trial of drugs or techniques that necessarily involves new medical treatment that has not yet gained the support of a substantial number of medical practitioners specialising in the area of practice concerned. It does not depend on whether the study in question involves randomisation, the need to be notified under the TG Act or both.
2. For the reasons set out above, Appeal Panel is of the view that: 1. The treatments (both drugs and their mode of use) used in both arms of the SPICE III study are not new but existing, accepted treatments for patients in intensive care and those treatments have gained the support of a substantial number of medical practitioners specialising in the area of practice concerned, which the Appeal Panel understands can be described as sedation of patients in intensive care; 2. The fact that dexmedetomidine is used in the SPICE III study in a randomised trial outside its approved product information, and consequently notification under the TG Act is required, does not mean that the SPICE III study is, as a matter of law or of fact, a "clinical trial" for the purpose of Pt 5 of the Guardianship Act.
3. Accordingly, the SPICE III study is not, on the proper construction of the expression "clinical trial" in Pt 5 of the Guardianship Act set out above, a clinical trial that could be approved under s 45AA.
4. To the extent the Tribunal below reached contrary conclusions, we would allow the appeal on the second reformulated ground.

Conclusion

1. We have decided that the SPICE III study is not a clinical trial that can be approved under s 45AA and that we should uphold the appeal on both reformulated grounds. As a consequence, it is appropriate to set aside the orders made in both proceedings CT 10/2013 and proceedings CT 7/2015. In their place, we propose to make orders that the applications in both proceedings be dismissed.

Orders

1. The orders of the Appeal Panel are: 1. The Appeal is allowed. 2. The orders made in proceedings CT 10/2013 on 23 December 2014 and in proceedings CT 7/2015 on 19 June 2015 are set aside. 3. The application in proceedings CT 10/2013 is dismissed. 4. The application in proceedings CT 7/2015 is dismissed.

I hereby certify that this is a true and accurate record of the reasons for decision of the Civil and Administrative Tribunal of New South Wales. Registrar

Amendments 27 June 2016 - [134] - amendment to clarify reference to reasons of Tribunal at first instance. 27 June 2016 - [161] - removed unnecessary word. DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 27 June 2016