Secretary, NSW Treasury v C (GD)

[2004] NSWADTAP 6

Appeal Panel - Internal

CITATION:	Secretary, NSW Treasury v C (GD) [2004] NSWADTAP 6 revised - 20/02/2004
PARTIES:	APPELLANT Secretary, NSW Teasury RESPONDENT C
FILE NUMBER:	039066
HEARING DATES:	8/12/2003
SUBMISSIONS CLOSED:	12/15/2003
DATE OF DECISION:	02/20/2004
DECISION UNDER APPEAL:	C v Secretary, NSW Treasury (No 3) [2003] NSWADT 200
BEFORE:	Chesterman M - ADCJ (Deputy President); Robinson MA - Judicial Member; Blake C - Member
CATCHWORDS:	application of common law test - no evidence
MATTER FOR DECISION:	Principal matter
FILE NUMBER UNDER APPEAL:	013081
DATE OF DECISION UNDER APPEAL:	08/25/2003
LEGISLATION CITED:	Administrative Decisions Tribunal Act 1997 Freedom of Information Act 1989 Privacy and Personal Information Protection Act 1998
CASES CITED:	Re Applicant and Department of Education and Training, Children’s Youth and Family Services (1998) 53 ALD 509 Botany Council v The Ombudsman (1995) 37 NSWLR 357 Brown v Raphael [1958] Ch 636 Browne v Dunn (1893) 6 R 67 Bushell v Repatriation Commission (1992) 175 CLR 408 C v Secretary, NSW Treasury (No 3) [2003] NSWADT 200 Central Sydney Area Health Service v Crewdson [2001] NSWADTAP 44 Re Close and Australian National University (1993) 31 ALD 597 Cox and Dept of Defence (1990) 20 ALD 499 Crewdson v Central Sydney Area Health Service [2002] NSWCA 345 Director General Department of Community Services v S [2000] NSWADTAP 27 Re Gordon and Department of Social Security (1991) 25 ALD 33 Hewitt v Grabicki 794 F 2d 1373 Re Leverett and Australian Telecommunications Commission (1985) 8 ALN N135 Re Resch and Department of Veterans’ Affairs (1986) 9 ALD 380 RR v Department of the Army 482 F Supp 770 (1980) Smith v Land & House Property Corporation (1884) 28 Ch D 7 Re Warren and Department of Defence, unreported, Administrative Appeals Tribunal, 22 December 1992
REPRESENTATION:	APPELLANT M Allars, barrister RESPONDENT In person
ORDERS:	1. Appeal allowed in part.; 2. The Tribunal’s order, setting aside the Appellant’s decision to refuse to amend the report of Dr Lucire relating to the Respondent, is upheld.; 3. In substitution for that decision of the Appellant, a decision is made to attach a notation to the report, and to every copy of the report held by the Appellant, a notation in the following terms:; ‘The attached document is incorrect within the meaning of s 39(c) of the Freedom of Information Act 1989. Specifically, Dr Lucire’s diagnosis that Mr C (insert full name) seemed on clinical examination to have a Paranoid Delusional Disorder and on the documents to have a Narcissistic Personality Disorder has been withdrawn by her. For further details, see the decision of the Appeal Panel of the Administrative Decisions Tribunal in Secretary, NSW Treasury v C [2004] NSWADTAP 6.’; 4. If the Appellant discloses to any person (including any other agency or any Minister) any information contained in Dr Lucire’s report, the Appellant should ensure that the above notation is included when that disclosure is made.; 5. If either party wishes to make an application for costs, it should do so within 28 days of the date of this decision.; 6. If such an application is made, the other party is directed to file and serve any submissions in reply within 28 days of being served with the application. ; 7. Any application for costs will be determined by the Tribunal ‘on the papers’ pursuant to s 76 of the Administrative Decisions Tribunal Act 1997

1 In this case, the Appellant, the Secretary, NSW Treasury (hereafter ‘the agency’) appealed against an order made in review proceedings before the Tribunal, constituted by Magistrate N Hennessy, Deputy President (C v Secretary, NSW Treasury (No 3) [2003] NSWADT 200). The order set aside a decision of the agency refusing to amend a medical report pursuant to s 44 of the Freedom of Information Act 1989 (‘the FOI Act) and specified amendments to be made to the report. The report related to the Respondent. We will refer to him, as the Tribunal did, as ‘Mr C’, in order to protect his privacy.

2 The medical report, which is undated, was prepared by Dr Yolande Lucire, a consultant psychiatrist, following a consultation with Mr C on 6 January 2000. Dr Lucire had been engaged to prepare the report in connection with a workers compensation claim instigated by Mr C on 22 December 1998. The claim was against his then employer, the Department of Corrective Services, and alleged ‘systemic harassment’.

3 In support of his claim, Mr C lodged a medical report, dated 15 December 1998, by his treating psychiatrist, Dr Howard Napper. In it, Dr Napper expressed the opinion that Mr C suffered from adjustment, depression and anxiety. Dr Napper was called by Mr C as an expert witness in the present proceedings.

4 Dr Lucire was engaged to prepare her report by GIO General Ltd, which dealt with Mr C’s claim on behalf of the Treasurer. The letter of instructions to her, dated 15 December 1999, requested that she determine the ‘reasonableness’ of Mr C’s conduct. It contained the following passage:-

Would you please take a complete history and indicate whether the claimant is suffering from a diagnosable psychiatric condition due to work. If you feel the condition has substantially arisen out of work, would you indicate whether it has arisen as a result of any unreasonable action by the employer. Would you assess fitness for either normal or restricted duties and offer a prognosis for return to normal duties. Would you also assess current treatment and its appropriateness.

5 After receiving Dr Lucire’s report, the agency gave Mr C access to it. He considered that a significant quantity of the information contained in it was incorrect. On 26 February 2001, he applied to the agency under the FOI Act and the Privacy and Personal Information Act 1998, seeking amendment of the report on the ground that this information was ‘incomplete, incorrect, out of date or misleading’.

6 In so far as this application was made under the FOI Act, the relevant provision was s 39, which states as follows:-

39. A person to whom access to an agency’s document has been given may apply for the amendment of the agency’s records:

(a) if the document contains information concerning the person’s personal affairs, and

(b) if the information is available for use by the agency in connection with its administrative functions, and

(c) if the information is, in the person’s opinion, incomplete, incorrect, out of date or misleading.

7 Section 43(1) of the FOI Act sets out the manner in which an agency must determine an amendment application:

43. (1) An agency shall determine an application:

(a) by amending its records in accordance with the application, or

(b) by refusing to amend its records.

8 Section 44 provides that:

44. An agency may refuse to amend its records in accordance with an application:

(a) if it is satisfied that its records are not incomplete, incorrect, out of date or misleading in a material respect, or

(b) if it is satisfied that the application contains matter that is incorrect or misleading in a material respect, or

(c) if the procedures for amending its records are prescribed by or under the provisions of a legislative instrument other than this Act, whether or not amendment of those records is subject to a fee or charge.

9 The agency refused, both initially and in an internal review, to amend the report. The ground of refusal was that it was satisfied that the report was not incomplete, incorrect, out of date or misleading in a material respect.

10 By a letter dated 9 April 2001, Mr C then required the agency to add to its records a notation under s 46 of the FOI Act, to the effect that he disputed the correctness of the report. Section 46 is as follows:

46. (1) If an agency has refused to amend its records, the applicant may, by notice in writing lodged at an office of the agency, require the agency to add to those records a notation:

(a) specifying the respects in which the applicant claims the records to be incomplete, incorrect, out of date or misleading, and

(b) if the applicant claims the records to be incomplete or out of date setting out such information as the applicant claims is necessary to complete the records or to bring them up to date.

(2) An agency shall comply with the requirements of a notice lodged under this section and shall cause written notice of the nature of the notation to be given to the applicant.

11 The notation was added as requested. Mr C applied, however, to this Tribunal for review of the agency’s decision under s 44 to refuse to amend its records. Although in his letter of 26 February 2001 to the agency he had challenged the accuracy of a large proportion of the report’s 75 paragraphs, he ultimately indicated to the Tribunal that, in essence, what he sought to amend was the diagnosis of his condition that Dr Lucire had provided in the report.

12 The precise content of this diagnosis will be examined later in this judgment. In its judgment, at [1], the Tribunal treated it as sufficiently set out in the first paragraph of the report (with ‘Mr C’ substituted for the Respondent’s real name). This is as follows:-

Mr C, on clinical examination, seemed to have a Paranoid Delusional Disorder. On the documents it is more like Narcissistic Personality Disorder. He also has compulsive traits.

13 In summary, the preferred means whereby Mr C wished to amend the report was through (a) the deletion of all the words or phrases in it which attributed paranoia, narcissism, delusions or other psychiatric problems to him and (b) the addition of a statement that these words had been deleted in consequence of the proceedings in the Tribunal and that they formed part of a ‘biased and unsubstantiated opinion’. The precise terms of the amendments that he sought are set out in the Tribunal’s judgment, C v Secretary, NSW Treasury (No 3) [2003] NSWADT 200, at [100 – 101].

The Tribunal’s consideration of preliminary matters and basic principles

14 The Tribunal noted, at [11], that Dr Lucire’s report contained information concerning Mr C’s personal affairs and that this information was available for its use in connection with its administrative functions. It implicitly concluded that the requirements of subparagraphs (a) and (b) of s 39 of the FOI Act were accordingly satisfied.

15 It rejected, at [16 – 18], a contention by the agency that Dr Lucire’s report did not form part of the agency’s ‘records’, for the purposes of s 39.

16 It then went on to consider, at [20 – 44], the legal principles governing amendment of records under the FOI Act and their application to the circumstances of this case. Having noted that under s 61 of this Act, the onus lay on the agency to justify its decision not to amend the report, it set out, at [20] and [34], several important paragraphs from the judgment of Handley JA (with whom Ipp JA and Davies AJA agreed) in Crewdson v Central Sydney Area Health Service [2002] NSWCA 345. This is, as far as we are aware, the only Court of Appeal decision, and therefore the leading authority, on the meaning of the key phrase ‘incomplete, incorrect, out of date or misleading’ in ss 39(c) and 44(a).

17 The paragraphs that the Tribunal quoted, [19], [31] and [34 – 37], are as follows:-

[19] The alleged unfairness in the procedures which led to the appellant’s psychiatric examination and in the course of that examination, including the alleged bias of Dr Roberts, do not raise issues directed to the accuracy of the statement. One cannot establish that a medical opinion is “incorrect ... or misleading” by establishing that the procedures followed were legally flawed or the doctor was biased. The opinion may still be correct.

[31] A statement of opinion by a skilled person is not only a representation that he holds that opinion, it will also generally be an implied representation that he “knows facts which justify his opinion”. See Smith v Land & House Property Corporation (1884) 28 Ch D 7 CA, 15; and Brown v Raphael [1958] Ch 636 CA, 642, 644, 645.

[34] Even if the Tribunal accepted other experts who had a different opinion that would not make “incorrect”, for the purposes of s 39(c), an accurate statement of the opinion held by Drs Roberts and Jagger. See Re Resch (1986) 9 ALD 380, 389 (Hall DP); Re Gordon (1991) 25 ALD 33; and Re Close and Australian National University (1993) 31 ALD 597, 609-10. The Act is not a vehicle for the determination of disputed questions of expert or other opinion when the recorded opinion was actually held and accurately entered in the official records.

[35] The position might be different if an expert whose opinion had been accurately recorded recognised later that it was incorrect at the time and withdrew it. However the proper course would be to add a notation that the opinion had been withdrawn rather than to remove the original opinion. See Cox and Dept of Defence (1990) 20 ALD 499, 500, 501 (Todd DP). An amendment in the latter form would falsify the records and attempt to rewrite history. See Botany Council v The Ombudsman (1995) 37 NSWLR 357, 369 and Re Close (1993) 31 ALD 597, 609. Without the original opinion the records would not tell the whole story, and would be misleading.

[36] Some decisions support the view that an accurate statement of an opinion, expert or otherwise, that was genuinely held could be incorrect or misleading for present purposes if it was based solely or perhaps substantially on information which was shown to be incorrect. See Director General Department of Community Services v S [2000] NSWADTAP 27 which was referred to in the decision of the Appeal Panel in the present case. Compare RR v Department of the Army 482 F Supp 770 (1980), 774, 775.

[37] It is not necessary to consider these questions because it was not established that the opinion was based solely or substantially on information shown to be incorrect.

18 It is convenient for us to quote at this point a further paragraph in Handley JA’s judgment, [24], which in our view also provides important guidance:-

[24] The appellant’s attempt to use the Act as a vehicle for the collateral review of the merits or validity of official action should be rejected in any event. The Act is concerned with the accuracy of official records, not with the merits or legality of the official action recorded in them. Compare Re Resch and Dept of Veterans’ Affairs (1986) 9 ALD 380, 386, 387 (Hall DP) and Hewitt v Grabicki 794 F 2d 1373 (9th Circuit 1986), 1378.

19 Applying principles set out in Crewdson, the Tribunal rejected, at [21 – 23], claims by Mr C that Dr Lucire’s report was ‘incorrect’ or ‘misleading’ on the separate bases that (a) she had exhibited bias and (b) Dr Napper, in evidence to the Tribunal, had expressed strong disagreement with her diagnosis.

The Tribunal’s acceptance of the argument that Dr Lucire had withdrawn her diagnosis

20 The Tribunal then discussed, at [24 – 27], a claim by Mr C that Dr Lucire, in her evidence to the Tribunal, had withdrawn the diagnosis of his condition contained in her report. The Tribunal’s observations in these paragraphs should be quoted, as they are central to this appeal:-

[24] Withdrawal or modification of opinion . The applicant submitted that Dr Lucire modified her expert opinion in oral evidence. She said in the report that Mr C “seemed to have a Paranoid Delusional Disorder” and “on the documents it is more like Narcissistic Personality Disorder. He also has compulsive traits.” Later, in the report, Dr Lucire says that “Mr C genuinely believes that he is not getting what he is entitled to. This attitude is a manifestation of his personality disorder which is primarily narcissistic but occasionally he is paranoid as well.”

[25] These and other statements demonstrate that Dr Lucire made a diagnosis of Narcissistic Personality Disorder and Paranoid Delusional Disorder. Dr Lucire modified her diagnosis in her oral evidence. She said in her evidence that:

I’m prepared to say that Mr C has narcissistic personality traits of significant intensity and possibly a personality disorder and I would be quite happy to say that.
[26] The following exchange occurred between the agency’s counsel and Dr Lucire:
Q: Your diagnosis of Mr C at the time you examined him is as set out in your report, that he had a paranoid delusional personality or narcissistic personality?

A: No, he has a personality disorder which – well, narcissistic personality traits which are very prominent. I also thought possibly at the time that I saw him or possibly at various times close to the time I saw him he had been frankly paranoid, that is, he was seeing conspiracies and so on, and that is called a paranoid disorder. . .

[27] By answering “No” to this question and by her subsequent explanation and other evidence, I find that Dr Lucire withdrew her diagnosis of a Narcissistic Personality Disorder and Paranoid Delusional Personality Disorder. Her modified position is that Mr C has traits of a Narcissistic Personality Disorder and that it is possible that he has a Paranoid Delusional Personality Disorder. This is so despite the more general statement later in her evidence that she genuinely believed that her diagnosis of the applicant “was a good one”. The effect of Dr Lucire’s withdrawal, or at least modification, of her opinion is dealt with further below.

21 At paragraph [88] in its judgment, the Tribunal returned to this issue. Before quoting again paragraph [35] of Handley JA’s judgment in Crewdson (set out in the present judgment at [17] above), it observed as follows:-

[88] There are three bases for concluding that Dr Lucire’s opinion that Dr C has a Paranoid Delusional Disorder and a Narcissistic Personality Disorder is incorrect. The first, and perhaps the most fundamental, is that Dr Lucire withdrew this opinion in oral evidence. Even if it cannot be said that she withdrew it, she certainly modified it so that she was no longer giving a definite diagnosis of Paranoid Delusional Disorder or a Narcissistic Personality Disorder. In Crewdson , the Court of Appeal anticipated such a situation and said at [35]….

22 At [28 – 44], the Tribunal then stated that the ‘remaining issue’ was whether an agency’s records may be held to be ‘incomplete, incorrect, out of date or misleading in a material respect’ on the ground that, quoting Handley JA’s judgment in Crewdson at [36], they are ‘based solely or perhaps substantially on information which was shown to be incorrect’. It made the preliminary observation that, according to his Honour’s judgment at [37], the Court of Appeal did not need to resolve this issue.

23 The Tribunal then referred, at [29], to a discussion of this issue in the judgment of the Tribunal’s Appeal Panel in the Crewdson case (Central Sydney Area Health Service v Crewdson [2001] NSWADTAP 44). It said that the Appeal Panel rejected what the Tribunal called a ‘restrictive approach’, namely, that ‘an expert report based solely on discredited facts cannot be allowed to stand’. As summarised by the Tribunal, the ‘more generous approach’ preferred by the Appeal Panel was as follows:-

… it is sufficient if the factual substratum for the expert opinion is substantially disproved and the Tribunal is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert.

24 The Tribunal noted that the Appeal Panel quoted an ACT Administrative Appeals Tribunal case, Re Applicant and Department of Education and Training, Children’s Youth and Family Services (1998) 53 ALD 509, in support of this formulation. It cited para [44] of the Appeal Panel’s judgment as the source of the formulation. This is clearly an error: the reference should apparently be to [96] (the relevant part of which is quoted below at [28]).

25 In the ensuing paragraphs, [30 – 36], the Tribunal then outlined the arguments put to it by each of the parties regarding the test that it should apply. The principal features of these arguments, as outlined, were as follows:-

· Mr C contended that it was sufficient for him to show that the factual substratum for Dr Lucire’s opinion had been substantially disproved, rather than thoroughly discredited. As a qualification to this, he claimed (see [30]) that it was enough for him to show that ‘his perceptions of his treatment by his employer were not abnormal and had some basis in fact’.

· The agency’s primary contention was that a professional medical report could only be held incorrect or misleading within s 44(a) of the FOI Act if the opinion expressed was (a) not genuinely and actually held by the medical practitioner or (b) was not accurately entered in the official records. In the alternative, it contended that the Tribunal should adopt the ‘restrictive approach’, claiming that the ‘generous approach’ outlined by the Appeal Panel was not supported by Australian authorities and had been implicitly rejected by the Court of Appeal in Crewdson. In addition, what would appear to have been a ‘fall-back’ position maintained by the agency (though the Tribunal’s judgment is not entirely clear on this point) was that either the ‘sole fact’ on which a report is based or ‘a substantial part of the factual basis for the report’ would have to be ‘destroyed’ before the relevant opinion could he held incorrect. The agency argued that on a proper reading of Handley JA’s judgment in Crewdson, he must be taken to have rejected any less rigorous test (see [36]). It referred also (see [34]) to statement by Handley JA (at [31]) to the effect that a statement of opinion by a skilled practitioner is not only a representation that he or she holds the opinion, but also an implied representation that he or she knows facts justifying the opinion.

26 At [37 – 44], the Tribunal set out its response to these competing contentions. First, it rejected the agency’s primary submission, saying that there was no reason to treat medical opinions as falling outside the general principle that ‘facts which are non-existent or infected with inaccuracy may render incorrect an opinion based on those facts or assumed facts’ (see [37]). In support of this proposition, it cited three authorities: Re Leverett (1985) 8 ALN N135 at N136; Re Warren and Department of Defence, unreported, Administrative Appeals Tribunal, 22 December 1992; and RR v Department of the Army 482 F Supp 770 (1980). It also indicated its dissent from the agency’s submission that Handley JA’s judgment in Crewdson implicitly indicated a preference for a particular approach, saying that the Court of Appeal ‘did not decide the issue’ ([38]).

27 At [39], the Tribunal noted that the American decision in RR v Department of the Army had provided guidance to Australian tribunals. It quoted the following passages from the judgment of Gesell J:-

It would defy common sense to suggest that only factually erroneous assertions should be deleted or revised, while opinions based solely on these assertions must remain unaltered in the individual’s official file. An agency may not refuse a request to revise or expunge prior professional judgments once all the facts underlying such judgments have been thoroughly discredited. [at 774]

Where matters of professional judgment such as this are concerned and the factual predicates of such opinions are diverse, it is next to impossible to reconstruct the process by which the opinion was formulated and determine what the opinion would have been. This difficulty is particularly evident where, as here, the opinion was formulated in good faith and was based on observations made nearly 30 years ago. Under such circumstances the challenged medical judgments are not so thoroughly discredited as to justify their deletion or contradiction in the record. [at 775].

28 The Tribunal then stated (at [40]) that ‘the so-called “generous” approach seeks to overcome the difficulty identified in RR v Department of the Army’. It identified this as being the difficulty of determining the facts on which an opinion was based. It suggested that in some instances, with the help of an appropriately qualified expert (see [41]), this difficulty could be overcome and ‘a determination made as to what the opinion would have been’. While noting (at [40]) that this process would not be possible in every case, it quoted the following passage from the Appeal Panel’s judgment in Crewdson (at [96]) as indicating how it could operate:

... where the facts upon which an expert opinion is based have not been substantially disproved, then we agree that the tribunal should not take on the mantle of expert. It should only proceed to interfere with the original expert’s opinion where it is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert.

29 The Tribunal then set out, at [42], its general conclusions as follows:-

In my view, on closer examination, the so-called “restrictive” and “generous” tests are not mutually exclusive and both provide a valid starting point for identifying at least two of the circumstances in which an opinion may be incorrect or misleading. My conclusion on the basis of the case law and the above analysis is that an expert opinion may be “incorrect” where either:
• all the facts underlying the opinion have been disproved; or
• some of the facts underlying the opinion have been disproved and either the opinion cannot stand on the basis of the facts that remain or the Tribunal is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert.

30 Next, the Tribunal stated at [43] that although a statement of opinion by a skilled practitioner would generally incorporate an implied representation that he or she knew facts justifying the opinion, there would still be many cases where it could be shown that the facts on which the opinion was based were incorrect.

31 Concluding, at [44], its discussion of the legal principles to be applied, the Tribunal rejected what it took to be an implicit argument by the agency that the onus lay on Mr C to ‘thoroughly discredit any possible factual basis’ for Dr Lucire’s opinion. It reiterated that under s 61 of the FOI Act, the onus lay on the agency to demonstrate that the report was not ‘incomplete, incorrect, out of date or misleading in a material respect’ within the meaning of s 44.

32 In applying these principles, the Tribunal considered a number of factual errors allegedly underlying Dr Lucire’s report. It concluded that there were in fact two material errors, but dismissed Mr C’s claims in relation to the remainder. In this judgment, it is sufficient to deal with the two errors which the Tribunal found to have been present.

33 The first error was that, as the Tribunal expressed it at [52], Dr Lucire ‘must have found (or assumed)’ that Mr C had suffered since early adulthood from Narcissistic Personality Disorder and Paranoid Personality Disorder. (Dr Lucire’s diagnosis was, it will be recalled, that Mr C ‘seemed to have a Paranoid Delusional Disorder’ and that ‘on the documents’ it was ‘more like Narcissistic Personality Disorder’.) The Tribunal’s view was that there was ‘no evidence whatsoever’ to support her finding or assumption.

34 The grounds for this conclusion by the Tribunal were outlined as follows at [51 – 52]: (a) its finding that Dr Lucire did not take a medical history from Mr C; (b) her evidence that, as far as she could ascertain, he had had no symptoms or episodes of either disorder until 1992 (during which year he reached the age of 43); and (c) the fact that none of the other medical reports in evidence referred to any family or personal history of either disorder. The Tribunal noted a statement by Dr Lucire in her evidence that paranoiac or narcissistic personality traits may be present without giving problems, or may give problems without being observed or presenting clinically. It did not treat this assertion as detracting from its conclusion.

35 The reason why the Tribunal regarded this error as significant was that Dr Lucire’s diagnosis of Narcissistic Personality Disorder and ‘Paranoid Personality Disorder’ relied (explicitly in the former case, implicitly in the latter) on diagnostic criteria which include the phrase ‘beginning by early adulthood’. These criteria are set out in the Diagnostic and Statistical Manual of Mental Disorders (DSM – IV), 4th edition, published by the American Psychiatric Association, Washington, DC. The text of the criteria for Narcissistic Personality Disorder was in fact annexed to Dr Lucire’s report.

36 At [90], [92] and [95], the Tribunal summarised as follows its reasons for holding that this erroneous finding or assumption by Dr Lucire constituted a ground for concluding that her diagnosis was incorrect in relation to both disorders:-

[90] The second basis for concluding that Dr Lucire’s opinion is incorrect is that neither of the disorders which Dr Lucire diagnosed began in early adulthood. For the reasons set out in paragraph 52 above, Dr Lucire’s assumption that the disorders did begin at that time (if indeed she ever directed her mind to that issue) is incorrect. Because early adult onset is a pre-requisite to the diagnosis of both disorders under DSM-1V, Dr Lucire’s opinion cannot stand and it is not necessary to obtain the opinion of another expert to find that it is incorrect. If it is necessary for the Tribunal to be persuaded by an alternative expert opinion, I am persuaded by the opinion of Dr Napper on this point. (See paragraph 92 below.)

[92] Dr Napper is a consultant psychiatrist who treated Mr C on 22 occasions between October 1998 and June 2000. When asked about Dr Lucire’s medical diagnosis of the applicant, he stated that he did not agree with Dr Lucire’s report on any ground. He confirms the Tribunal’s conclusion above, that in order to make a diagnosis of Narcissistic Personality Disorder or Paranoid Personality Disorder “it must have been present for a long time right from early developmental periods. It’s pervasive in that it affects all aspects of life in terms of psycho-social occupational functioning. It’s not just something that occurs that comes on later on in life.”

[95] Dr Napper’s evidence was credible and convincing. It firstly confirms the Tribunal’s finding that early adult onset is a mandatory element of both disorders, if the practitioner is relying on DSM-1V….

37 The second error by Dr Lucire that the Tribunal discerned (at [53]) was a finding expressed most clearly in paragraph 20 of her report (it was mentioned also in paragraphs 22, 38 and 29). This was, in the terms of paragraph 20:-

I believe Mr C exaggerates his achievements when he speaks of ‘making steady progress in the Australian corporate sector.’ This is a narcissistic characteristic of narcissistic personality disorder.

38 The Tribunal referred, at [54 – 55], to evidence indicating that Mr C’s capacities and achievements had in fact been very highly regarded by his superiors in the Department of Corrective Services. On the basis of this evidence, it made the following finding, at [56]:-

There was no evidence whatsoever which supports Dr Lucire’s finding or assumption that Mr C was exaggerating his achievements or that he expects to be recognised as superior without commensurate achievements. Consequently, one of the factual bases on which Dr Lucire based her opinion that Mr C has a Narcissistic Personality Disorder has been disproved.

39 At [91], [93] and [95], the Tribunal summarised as follows its reasons for holding that this erroneous finding or assumption by Dr Lucire constituted a further ground for concluding that her diagnosis of a Narcissistic Personality Disorder was incorrect:-

[91] The third basis for the Tribunal’s conclusion that the diagnosis of Narcissistic Personality Disorder is incorrect is that one of the elements of that disorder contained in DSM-1V, on which Dr Lucire relied, is not supported by the facts. The assumptions on which Dr Lucire based her opinion that Mr C has a grandiose sense of self-importance were incorrect. Dr Lucire did not expressly identify the five traits listed for Narcissistic Personality Disorder in DSM-1V on which she relied. However, reading her report, it is clear that as well as a “grandiose sense of self importance,” she regarded him as having a “sense of entitlement: unreasonable expectation of especially favourable treatment” and as reacting to criticism with “feelings of rage, shame, or humiliation”. The presence of two of the traits listed in DSM-1V is insufficient to justify a diagnosis of Narcissistic Personality Disorder, but this is a case where the Tribunal must seek the opinion of an expert as to whether, based on the facts that remain, such an opinion could be held by any competent expert.

[93] Dr Napper’s second point was that in order to conclude that a person has a particular personality disorder then a certain number of the personality traits listed in DSM-1V has to be present. He said that if a medical practitioner uses DSM-1V he or she has to have the evidence to support the diagnostic categories. He went through the features of Narcissistic Personality Disorder in DSM-1V and concluded that “. . . to make that diagnosis you need to have five or more of those nine features. Now in my opinion Dr Lucire’s report falls far short in terms of not being able to come up with these features.” Dr Napper expressed the opinion that a competent psychiatrist could not come to the conclusion to which Dr Lucire came, based on the available evidence.

[95] Dr Napper’s evidence was credible and convincing…. Secondly, the Tribunal is persuaded that based on the facts that remain after eliminating a “grandiose sense of self-importance” as a trait exhibited by Mr C, no competent expert could find that Mr C exhibits the requisite number of traits to justify a diagnosis of Narcissistic Personality Disorder.

40 In the appeal, Ms Allars, counsel for the agency, did not dispute the proposition that, to quote from Handley JA’s judgment in Crewdson at [35], a finding of incorrectness under s 44 of the FOI Act might be made ‘if an expert whose opinion had been accurately recorded recognised later that it was incorrect at the time and withdrew it’.

41 Ms Allars argued, however, that the Tribunal erred in law in making its finding that during the hearing at first instance Dr Lucire retracted her diagnosis. Her primary submission was that this was an incorrect finding on the evidence. An alternative formulation of this submission was that there was in fact no evidence on which the finding could be made. Ms Allars referred to various passages in Dr Lucire’s report and to the transcript of Dr Lucire’s evidence before the Tribunal.

42 Ms Allars submitted two further propositions to us: (a) that because it was not put to Dr Lucire in cross-examination that she had withdrawn the diagnosis, it was not open to the Tribunal to make this finding, on account of the rule in Browne v Dunn (1893) 6 R 67; and (b) that Handley JA’s dictum should be interpreted as requiring a formal withdrawal of the diagnosis, as opposed to one or more ‘loose’ statements suggesting that the maker of the diagnosis had had second thoughts about it.

43 Mr C’s arguments in response chiefly took the form of directing our attention to passages in the transcript of Dr Lucire’s evidence where her opinions seemed to differ from those expressed in her report. He also pointed out that in another passage Ms Allars herself appeared to concede that Dr Lucire’s opinion appeared to have changed.

44 The starting-point for our consideration of this issue must be the precise content of the diagnosis given by Dr Lucire in her report. As indicated above at [12], the formulation on which the Tribunal based its judgment was that set out in paragraph [1] of the report. It is as follows:-

Mr C, on clinical examination, seemed to have a Paranoid Delusional Disorder. On the documents it is more like Narcissistic Personality Disorder. He also has compulsive traits.

45 This is not, however, the only formulation of a diagnosis to be found in what is on any view a distinctly incoherent and disjointed report. Numerous references to alleged disorders or personality traits of Mr C, involving paranoia, delusions and/or narcissism, are scattered throughout the report. Since (a) the records that are the subject of this litigation comprise the whole of Dr Lucire’s report, not just the first paragraph and (b) the terminology of the first paragraph is far from precise, it appears to us that in determining this issue relating to a claimed withdrawal or retraction of her diagnosis, we must take due account of all the statements in her report that in some way convey her opinion regarding Mr C’s condition.

46 So far as paranoia is concerned, the following additional paragraphs of the report contain statements elaborating on or modifying what Dr Lucire said in the first paragraph: 7, 14, 17, 18, 20, 24, 26, 27, 61 and 69.

47 In some of these paragraphs, a diagnosis of Paranoid Delusional Disorder or Paranoid Disorder is given with little or no qualification. The following are some examples:-

[7] The explanation for this [i.e., a statement in paragraph [6] that ‘I could not call Mr C “a reasonable man”’] lies in his Paranoid Personality Disorder/Delusional Disorder (Persecutory Type). This is exacerbated by his obsessional features and the condition is well described in standard text books of psychiatry and frequently leads to litigious or querulous paranoia.

[14] Had I been a psychiatrist at HealthQuest, Mr C would have a medical retirement on the basis of his Personality Disorder. This is a mental condition at the best of times but, when under stress, he becomes obviously paranoid. Paranoia is a mental disorder.

[17] If Mr C contains himself, he simply looks like an obsessional and anxious person. If he opens up with his paranoid notions and his exaggeration, then the diagnosis is inescapable.

[24] When he failed to get what he wanted, Mr C interpreted this as ‘harassment in the form of disregard, isolation, intimidation and humiliation’. This is not normal but a feature of narcissism and paranoid personality, perhaps delusional.

48 In other parts of the report, the diagnosis of paranoia is offered in a less definite way and/or on a more restricted basis. The following passages (the last two of which appear under the heading ‘Diagnosis and Opinion’) exemplify this:-

[20] … The document [Mr C’s statement for the purposes of his workers compensation claim] contained a number of markers for paranoia, narcissism and obsessionality….

[26] On one hand, this [i.e., the expression by Mr C, described in paragraph [25], of ‘feelings of entitlement’] could be regarded as Paranoid Disorder, on the other hand, it is what one sees in Narcissistic Personality Disorder where people are paranoid, only in so far as they feel others are withholding what they are entitled to for malevolent reasons of their own.

[61] Mr C genuinely believes that he is not getting what he entitled to. This attitude is a manifestation of his personality disorder which is primarily narcissistic but occasionally he is paranoid as well.

[69] Mr C is paranoid only in so far as he sees others conspiring and others as being malevolent in order not to give him what he entitled to.

49 In addition to paragraph 1, the following paragraphs of Dr Lucire’s report contain references to narcissism: 16, 20, 23, 24, 26, 31, 57, 61 and 66. The last two of these appear under the heading ‘Diagnosis and Opinion’.

50 In virtually all of these paragraphs, a diagnosis of Narcissistic Personality Disorder is given, or at least referred to, with little or no qualification. In addition to paragraphs 24 and 61, quoted above at [47] and [48] respectively, the following are some examples:-

[20] The document [Mr C’s statement for the purposes of his workers compensation claim] contained a number of identifiers for narcissistic personality disorder. This document comprises part of the file. The document contained a number of markers for paranoia, narcissism and obsessionality. I believe that Mr C exaggerates his achievements when he speaks of ‘making steady progress in the Australian corporate sector’. This is a narcissistic characteristic of narcissistic personality disorder.

[31] At point 18 [in the statement by Mr C for his workers compensation claim], there is an allegation of staging… Mr C at this point is so grandiose that he can imagine that people go through false procedures in order to put on an act especially for him. One would need to be very self-centred, very narcissistic to believe this kind of thing. His response (sic) are those of a person with a Narcissistic Personality Disorder.

I felt humiliated, degraded and stressed.
[66] This [referring to a statement in paragraph 65 that ‘It seems that as soon as Mr C is offered what he wants, he wants more’] is all a manifestation of his Narcissistic Personality Disorder. This is constitutional and pre-existing and it will manifest with anxiety when a situation arises where Mr C feels that he should have more than society wants to, or can, give him.

51 In paragraph 26, however, the diagnosis of Narcissistic Personality Disorder is given as an alternative to Paranoid Disorder:-

[26] On one hand, this [i.e., the expression by Mr C, described in paragraph [25], of ‘feelings of entitlement’] could be regarded as Paranoid Disorder, on the other hand, it is what one sees in Narcissistic Personality Disorder where people are paranoid, only in so far as they feel others are withholding what they are entitled to for malevolent reasons of their own.

52 We turn now to consider the passages in the transcript of the hearing at first instance which bear upon the question whether, expressly or by implication, Dr Lucire withdrew her diagnosis. Most of these passages formed part of Dr Lucire’s own evidence.

53 As indicated above at [20], the Tribunal, in paragraphs [25 – 26] of its judgment, cited the following passages from the transcript:-

[A statement by Dr Lucire in cross-examination: Transcript, p 79, lines 37-40]: I’m prepared to say that Mr C has narcissistic personality traits of significant intensity and possibly a personality disorder and I would be quite happy to say that.

[An exchange between the agency’s counsel and Dr Lucire: Transcript, p 54, lines 47-56]:

Q: Your diagnosis of Mr C at the time you examined him is as set out in your report, that he had a paranoid delusional personality or narcissistic personality?

A: No, he has a personality disorder which – well, narcissistic personality traits which are very prominent. I also thought possibly at the time that I saw him or possibly at various times close to the time I saw him he had been frankly paranoid, that is, he was seeing conspiracies and so on, and that is called a paranoid disorder. . .

54 Mr C relied also on the following interchange between the Deputy President and counsel for the agency after Dr Lucire had completed her evidence (Transcript, p 93, lines 34-46):-

Deputy President: My understanding of what Dr Lucire was saying about narcissistic personality disorder, though she wasn’t questioned about this, was that he had some markings of the disorder. I think she modified her evidence in that regard rather than saying – I took her to be retreating from saying that he did have narcissistic personality disorder, that she was in fact saying he had some features of that disorder. Is that your recollection of the evidence?

Counsel for the agency: She was prepared to say he had narcissistic personality traits.

55 Deputy President Hennessy then asked counsel whether she was prepared to concede that. The witness being examined at the time, Dr Napper, intervened and this question was never answered.

56 A further passage of relevance which, as we see it, also supports Mr C’s submission is in the examination in chief of Dr Lucire (Transcript, p 50, lines 28-37):-

Q: Is it possible that what you term paranoia entitlement can be almost not discernible in a person for a period but then later re-emerge?

A: Certainly. If Mr C contains himself he presents simply as an anxious and depressed and wronged person. If one looks at more information which covers a longer period then characteristics emerge that would make a diagnosis of adverse personality traits perhaps even disorder inescapable.

57 In response, Ms Allars referred first to sections of the transcript (p 51, line 16; p 52, lines 45-50) in which Dr Lucire expressed a willingness to correct certain typographical errors in her report. She relied principally on the following answer (Transcript, p 53, lines 50-54) given by Dr Lucire to her question whether Dr Lucire genuinely believed in the contents of the report:-

Insofar as I have assessed it I believe that – I genuinely believe that this is a good assessment and I stand by it. If there are mistakes in it I’m happy to stand corrected but I genuinely believe that I’ve done it to the best of my ability within standard practice.

58 Our first conclusion on this ground of appeal is that the only ground on which we could overturn the Tribunal’s ruling on this matter is that there was no evidence to support its finding that Dr Lucire withdrew her diagnosis. As the Notice of Appeal made clear, all the grounds of appeal were based on alleged errors of law. No application was made under s 113(2) of the Administrative Decisions Tribunal Act 1997 for leave to extend the appeal to a review of the merits. In these circumstances, it is not sufficient for the Appellant merely to show that a preferable decision on the evidence would have been that Dr Lucire did not withdraw her diagnosis.

59 This conclusion on our part involves a rejection of the two ancillary submissions of Ms Allars outlined at [42] above. We do not believe that is a proper case in which to apply the rule in Browne v Dunn (1893) 6 R 67 so as to rule out any finding that Dr Lucire had withdrawn her diagnosis, since it was made abundantly clear to her during cross-examination that Mr C challenged this diagnosis. We also see no reason to interpret Handley JA’s statement of principle in Crewdson as requiring a formal withdrawal of the diagnosis.

60 Next, we should state our belief that, as the Tribunal implicitly held, a statement of expert opinion, such as a diagnosis, may, under the principle stated by Handley JA, be held to have been withdrawn, and on that ground ‘incorrect’, even though simultaneously another statement of opinion is put forward in substitution. The outcome may amount in substance to a modification of the original statement of opinion, as distinct from a total retraction. But for the purposes of the determination to be made in this case, the original statement of opinion must be treated as having been withdrawn.

61 This brings us to the primary question to be resolved. In summary, the basis on which the Tribunal found the diagnosis in the report to have been withdrawn was as follows: (a) Dr Lucire said in the report that Mr C ‘seemed to have a Paranoid Delusional Disorder’, but said in oral evidence that it was possible that he had such a disorder; and (b) she said in the report that his disorder was ‘more like Narcissistic Personality Disorder’, but said in oral evidence that he had traits of such a disorder.

62 Our conclusions following our more detailed review (at [44 – 57] above) of both the report and the oral evidence are these:-

As to (a), it cannot be said that there is no evidence to support the Tribunal’s finding. If anything, our review of the relevant material confirms it. In the body of the report, there appears more certainty about the diagnosis of a Paranoid Disorder than in the paragraph quoted by the Tribunal. The additional sections of the oral evidence that we have identified further manifest an inclination on Dr Lucire’s part to substitute the opinion merely that Mr C might ‘possibly’ have had such a Disorder.

As to (b), the additional material that we have considered reinforces the Tribunal’s finding. In the body of the report, there is distinctly more certainty about the diagnosis of a Narcissistic Personality Disorder than in the paragraph quoted by the Tribunal. The additional sections of the oral evidence that we have identified add some weight – admittedly not a great deal – to the conclusion that she was inclined to substitute the opinion merely that he had narcissistic personality traits.

63 For these reasons, we reject the agency’s claim that the Tribunal erred in law in finding that in oral evidence Dr Lucire withdrew the diagnosis contained in her report on Dr C. There was, in our judgment, evidence available to the Tribunal that could support this conclusion. On this basis, if no other, the Tribunal’s order setting aside the agency’s decision to refuse to amend the report under s 44(a) of the FOI Act should be upheld.

The challenge to the finding that Dr Lucire’s diagnosis was based on incorrect information

64 The starting-point of the challenge made by Ms Allars to the Tribunal’s finding that Dr Lucire’s diagnosis was based on incorrect information was a contention that the Tribunal’s outline of the relevant legal principles was erroneous in law. In particular, Ms Allars disputed the Tribunal’s statement (at [38]) that the Court of Appeal’s judgment in Crewdson gave no guidance on the approach to be adopted when the factual basis of an expert opinion was alleged to be incorrect.

65 In her submission, the Court of Appeal did indeed convey a view on this matter. Although Handley JA’s judgment did not formulate a principle for determining the circumstances in which a medical opinion could be held incorrect on account of inaccuracies in the factual ‘substratum’ on which it was based, it clearly indicated that unless this substratum was shown to be at least ‘substantially’ incorrect, the opinion could not be held incorrect. If, to quote from Ms Allars’ written submission, ‘the proper test posed a lower threshold of disproof of underlying facts, then the Court would have had to consider whether the evidence met that threshold in order to satisfy itself that it was unnecessary to determine what was the test and apply the test…’ In fact, it did not do this.

66 Save on two matters, which we will mention shortly, issues of general principle were not specifically addressed in the submissions of Mr C.

67 The second way in which Ms Allars placed reliance on Crewdson was to maintain that certain well-established principles outlined by Handley JA ‘set the parameters within which a test may be developed’. These principles were that ‘amendment of records is a narrower exercise than merits review’ and that ‘such amendment does not extend to “correcting” expert opinions by replacing them’. In making this argument, she relied in particular on the following two observations of his Honour:-

[24] The appellant’s attempt to use the Act as a vehicle for the collateral review of the merits or validity of official action should be rejected in any event. The Act is concerned with the accuracy of official records, not with the merits or legality of the official action recorded in them.

[34]… The Act is not a vehicle for the determination of disputed questions of expert or other opinion when the recorded opinion was actually held and accurately entered in the official records.

68 The next step in Ms Allars’ argument was to assert that, through not taking sufficient account of these two important aspects of the Court of Appeal’s judgment in Crewdson, the Tribunal arrived at an erroneous formulation of the test that it should apply. The key passage in the Tribunal’s formulation, set out in its judgment at [42], is as follows:-

My conclusion on the basis of the case law and the above analysis is that an expert opinion may be “incorrect” where either:
• all the facts underlying the opinion have been disproved; or
• some of the facts underlying the opinion have been disproved and either the opinion cannot stand on the basis of the facts that remain or the Tribunal is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert.

69 Ms Allars focused her attack on the proposition in the second of these two dot-points. As she indicated, an evident basis for claiming that this proposition is inconsistent with the Court of Appeal’s judgment in Crewdson is that it does not require that the facts underlying the opinion in question should be ‘substantially’ disproved. It merely requires that ‘some’ of these facts must be disproved.

70 Ms Allars argued also that the proposition conflicted with the ‘general principles’ outlined in Crewdson in so far as it permitted the Tribunal to determine ‘disputed questions of expert… opinion when the recorded opinion was actually held and accurately entered in the official records’.

71 According to the first of the two alternatives set out in the second dot-point (dealing with the situation where ‘the opinion cannot stand on the basis of the facts that remain’), the Tribunal itself could make such a determination without the benefit of expert advice. In Ms Allars’ submission, the Tribunal was in error in so holding.

72 In addition, she submitted, the Tribunal erroneously relied on this statement of principle in reaching the following conclusion at paragraph [90] of its judgment (quoted above at [36]):

Because early adult onset is a pre-requisite to the diagnosis of both disorders under DSM-1V, Dr Lucire’s opinion cannot stand and it is not necessary to obtain the opinion of another expert to find that it is incorrect.

73 According to the second of the two alternatives set out in the second dot-point (‘the Tribunal is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert’), the Tribunal could not engage in the determination of ‘disputed questions of expert opinion’ without itself receiving expert evidence. It purportedly fulfilled this requirement by admitting the evidence of Dr Napper, on which (at [95]) it relied (a) as a second ground for holding that Dr Lucire’s diagnosis of both disorders was incorrect because there was no evidence of early adult onset; and (b) in holding that ‘on the facts that remain after eliminating a “grandiose sense of self-importance” as a trait exhibited by Mr C, no competent expert could find that Mr C exhibits the requisite number of traits to justify a diagnosis of Narcissistic Personality Disorder’.

74 Ms Allars argued that nonetheless the Tribunal simply chose to prefer the opinion of one expert over the opinion of another, and that this was erroneous in the light of the general principles stated by the Court of Appeal in Crewdson. In making this argument, she referred in particular to the following passage in her cross-examination of Dr Napper (Transcript, p 115, lines 14-38):-

Q: If another psychiatrist reaches a different diagnosis than the one you reached in relation to a patient would you conclude that that psychiatrist was not competent simply on that basis?

A: I would conclude that the psychiatrist was not competent to assess or value that particular patient, I wouldn’t say that they are incompetent in other aspects of that practice because I don’t have any evidence. But if I thought they disagreed with me and I thought that I was right, my opinion of the psychiatrist regarding that particular case would be fairly low.

Q: Sorry, fairly low?

A: Would be low, I’d have a low opinion of their ability to assess that case.

Q: You take that view even though you acknowledge that often psychiatrists disagree in their diagnosis of patients?

A: That’s right because I think there are some competent psychiatrists and there are some incompetent psychiatrists.

75 Ms Allars submitted that these answers revealed the true basis of Dr Napper’s statement during examination in chief that ‘in my opinion Dr Lucire’s views could not be held by a competent psychiatrist’ (Transcript, p 99, lines 28-29). This was, she said, that he treated any psychiatrist whose opinion differed from his own as incompetent. For this reason, the Tribunal had in her submission no sufficient basis for reaching the conclusion, for which the only expert support came from Dr Napper, that Dr Lucire’s diagnosis was incorrect on account of either of the two alleged inaccuracies in the factual substratum (i.e., those based on the lack of any or any sufficient evidence of early adult onset and of Mr C having a ‘grandiose sense of self-importance’).

76 Ms Allars put forward some ancillary arguments. In brief, they were (a) that the Tribunal should have been loath to intervene because in none of the cases on which it relied, and indeed in no Australian case, had a reviewing court or tribunal interfered with a medical diagnosis under freedom of information legislation; (b) that any incorrect assumption made by Dr Lucire about whether the diagnosed disorders commenced in or before early adulthood was irrelevant because Dr Lucire herself made no finding or assumption on this issue (Ms Allars referred here to the statement by Dr Lucire in her evidence that paranoiac or narcissistic personality traits may be present without giving problems, or may give problems without being observed or presenting clinically); (c) that Dr Lucire’s conclusion as to Mr C’s exaggeration of his achievements was a matter of opinion, not of fact, and therefore did not form part of the ‘factual substratum’; (d) that there was no evidence to support the Tribunal’s finding that this conclusion was incorrect; and (e) that in stating (at [44]) that the agency bore the onus of justifying its determination, the Tribunal overlooked dicta of Handley JA in Crewdson ([2002] NSWCA 345 at [32]) imposing an initial or preliminary burden on an applicant such as Mr C to produce evidence in support of the amendment.

77 Mr C’s written submission to us included passages disputing the first and the last of these propositions. He also contested Ms Allars’ claim that there was no evidence to support the Tribunal’s finding that he did not exaggerate his achievements.

Our conclusions regarding the finding that Dr Lucire’s diagnosis was based on incorrect information

78 The first contention made by Ms Allars was that the Court of Appeal in Crewdson did provide guidance on the test to be applied where inaccuracy in the factual substratum for an expert opinion is alleged. In our judgment, her arguments on this issue were compelling.

79 We would sum up their outcome as follows. The issue expressly left open by Handley JA in Crewdson involved a narrow choice as to whether the circumstances in which an expert opinion could be held incorrect on account of inaccuracy in the factual substratum (a) were limited to cases where the substratum was wholly incorrect or (b) extended also to cases where the substratum was substantially incorrect. Implicitly, his Honour held that proof of any lesser degree of inaccuracy would be insufficient.

80 We agree also with Ms Allars’ summary of the ‘well-established general principles’ that were restated by Handley JA in Crewdson.

81 This brings us to the crucial question whether and, if so, in what respects the Tribunal’s own formulation of the test to be applied was erroneous. In dealing with this matter, we consider that we must traverse a wider range of issues than were canvassed by Ms Allars. One reason for this is that Mr C, who appeared in person, was understandably not in position to assist us greatly on issues of legal principle.

82 We note at the outset that in Crewdson the Court of Appeal made no criticism of the judgment of the Appeal Panel. Handley JA simply stated, at [39], that ‘the appellant has not established that the decision of the Appeal Panel was vitiated by legal error…’.

83 In its judgment (Central Sydney Area Health Service v Crewdson [2001] NSWADTAP 44), the Appeal Panel engaged in a lengthy and valuable discussion, at [56 – 96], of the circumstances in which an opinion might be amended on grounds of incorrectness in the underlying facts.

84 It referred first to an earlier decision of an Appeal Panel: Director-General, Department of Community services v S [2000] NSWADTAP 27. It noted that in this case (a) the opinions involved were not based on any ‘special expertise’ and (b) amendments to them were held to be warranted where there was no factual basis for them [emphasis in the original]; and (c) the Panel’s comment on the test of ‘thorough discreditation’ stated by Gesell J in RR v Department of the Army 482 F Supp 770 (1980) (see [27] above) was that it might be too restrictive in some instances.

85 The next case discussed at any length by the Appeal Panel in Crewdson was Re Resch and Department of Veterans’ Affairs (1986) 9 ALD 380. Here Hall DP in the Administrative Appeals Tribunal commented that the proposition for which RR stood as authority was ‘very narrow’, but also said (at [38]) that ‘the medical opinions of the Departmental medical officers are not shown to be “incorrect” merely by producing medical opinions to the contrary’.

86 The Appeal Panel in Crewdson turned next to an ACT Administrative Appeals Tribunal case, Re Applicant and Department of Education and Training, Children’s Youth and Family Services (1998) 53 ALD 509. It quoted the following dictum of Curtis P at [65]:-

I see no reason why an opinion should be allowed to stand as correct if the principal factual support for the opinion is shown to be erroneous unless it appears that same opinion would have been given if the author of the opinion had not relied on the discredited facts.

87 The Appeal Panel then made the following comment (at [66] in its own judgment):-

We prefer this approach, one somewhat broader than the proposition expressed in RR v Department of Army and in Re Resch , though still a cautious one. It is consistent with the view expressed by the Appeal Panel in S’s case .

88 It then drew a distinction, relevant also in the present case, between facts provided to a doctor, either by the person being examined or by a third party, and the doctor’s own observations of the person being examined. The point that it made was that the doctor’s own observations were generally not amenable to being proved incorrect.

89 The Appeal Panel then cited a dictum of Brennan J in Bushell v Repatriation Commission (1992) 175 CLR 408 at 430, warning that a decision-maker in a situation comparable to the present is ‘bound to have regard to its own want of scientific expertise in comparison with the expertise of a responsible medical practitioner…’ It observed that this principle was applied by the Administrative Appeals Tribunal in Re Gordon and Department of Social Security (1991) 25 ALD 335, a case dealing with ‘incorrectness’ under Commonwealth freedom of information legislation.

90 After some discussion of the application of these principles to the case before it, the Appeal Panel in Crewdson set out as follows (at [94 – 96]) its conclusions on the specific issue with which we are presently concerned:-

[94] While our general view is that a Tribunal should only move to amend an expert opinion on the basis of credible expert evidence justifying that course, we acknowledge that there may be cases where the facts upon which the professional assessment was based may be shown to be ‘thoroughly’ discredited (to use the words of Gesell J) through ordinary fact-finding processes.

[95] It is conceivable that an expert opinion could be shown to have no satisfactory basis in fact and should be withdrawn.

[96] However where the facts upon which an expert opinion is based have not been substantially disproved, then we agree that the tribunal should not take on the mantle of expert. It should only proceed to interfere with the original expert’s opinion where it is persuaded by alternative expert opinion that based on the facts that remain, such an opinion could not be held by any competent expert. We agree, with respect, with the sentiment of Brennan J as reflected in the dicta in Re Bushell that the mere fact that another expert might have taken a different view on the same facts is not enough. The opinion under challenge must be shown to be one not open to a competent expert.

91 In support of the test which it formulated, the Tribunal in this case quoted, at [40], the first two sentences of the paragraph just reproduced. It also relied on two decisions not yet discussed in this judgment: Re Leverett and Australian Telecommunications Commission (1985) 8 ALN N135 and Re Warren and Department of Defence, unreported, Administrative Appeals Tribunal, 22 December 1993. Both of these are decisions of the Administrative Appeals Tribunal dealing with Commonwealth freedom of information legislation.

92 The relevant passage in Re Leverett is as follows (paragraph [13]):-

It is unnecessary to attempt to categorize the bases on which a professional opinion could be found to be ‘incomplete, incorrect or misleading’. To that of a demonstration of total inadequacy of underlying factual information, there could no doubt be added those of the existence of bias or ill will, incompetence or lack of balance or necessary experience in the person forming the opinion, or the existence of such a trivial factual substratum as to render the opinion formed dangerous to rely upon and likely to result in error, or where facts have been misapprehended.

93 We have been unable to identify any statement of principle in Re Warren that is of significant use in resolving the specific matter that we are now considering.

94 On the basis of this review of the authorities, we have concluded that the Tribunal’s formulation, at [42], of the test to be applied is defective in two respects. Both of the defects relate to the second of the two dot-points. They are more or less as Ms Allars claimed (see above at [68 – 71]).

95 The first defect is an erroneous statement, in this second dot-point, that it is not necessary that the facts underlying the opinion in question should be ‘substantially’ disproved, but merely that ‘some’ of these facts must be disproved.

96 The second defect is an erroneous statement, in the first of the two alternatives set out in the second dot-point (dealing with the situation where ‘the opinion cannot stand on the basis of the facts that remain’), that the Tribunal itself could make a determination to this effect without the benefit of expert advice.

97 We would make three additional observations about these conclusions.

98 First, what we have called the first defect may well have resulted from an ambiguity in the first two sentences in paragraph [96] of the Appeal Panel’s judgment in Crewdson (these being sentences that, as we have said above at [89], the Tribunal quoted in its judgment). In the first sentence, the Appeal Panel appears to maintain a requirement that intervention by the Tribunal depends upon proof that the facts underlying the relevant opinion have been ‘substantially disproved’. In the second sentence, however, this requirement is not mentioned. This second sentence, standing alone, appears to provide the basis for much of the second dot-point in the Tribunal’s formulation.

99 Secondly, we note that when the Tribunal, at [29], first formulated a summary of the principles applied by the Appeal Panel in Crewdson, it in fact included a requirement that the ‘factual substratum for the expert opinion’ should be ‘substantially disproved’.

100 Our third observation sets out what we believe to be a necessary corollary to our ruling that the Tribunal’s formulation suffers from what we have called ‘the second defect’. What we have said in that connection is that, when the factual substratum for an expert opinion has been substantially disproved, it is not open to a tribunal or court to determine without expert evidence that the opinion ‘cannot stand’ on the basis of the facts that remain. The corollary, as we see it, is that the tribunal or court would generally, if not invariably, need to call on expert evidence in order to decide the antecedent question: namely, whether, having found specific factual material forming part of the ‘substratum’ to be incorrect, it could properly conclude that the factual substratum had been proved ‘substantially’ incorrect.

101 A decision of a tribunal or court as to whether there had been ‘substantial’ disproof could hardly depend simply on a comparison of the number of individual factual assertions or assumptions held to be incorrect with the number of facts that ‘remained’. Instead, the issue would have to depend on the relative significance of each of these two categories of factual material (the ‘incorrect’ material and the ‘remaining’ material) for the expert opinion that had been formed in reliance on both categories taken in conjunction. Except perhaps in very clear cases, we do not see how the tribunal or court could make such an assessment of ‘substantial’ disproof without expert assistance.

102 In this context, we note the significance of comments by the Appeal Panel in Crewdson, mentioned above at [88]. These are to the effect that a doctor’s report on the condition of a patient will generally be based in part on the doctor’s own observations made during an examination of the patient and that this factual material will generally not be amenable to disproof. In such circumstances, it is likely to be difficult for someone challenging the doctor’s opinion under freedom of information legislation to produce evidence suggesting that the underlying factual material is ‘substantially’ incorrect.

103 Our attempt at a reformulation of the test, leaving aside the issue of onus of proof, is as follows:-

An expert opinion may be held ‘incorrect’ where either:

(a) all of the factual material underlying the opinion is incorrect; or

(b) (i) some of this factual material is incorrect; and

(ii) the court or tribunal determines, having due regard to expert evidence, that in consequence this factual material is ‘substantially’ incorrect; and

(iii) the court or tribunal determines, having due regard to expert evidence, that the opinion, if based only on the factual material that has not been held incorrect, could not be maintained by any competent expert.

104 In consequence of these two defects in the Tribunal’s formulation of the test to be applied, we must set aside its decision that Dr Lucire’s diagnosis of Mr C’s condition was incorrect on account of inaccuracies in the underlying facts.

105 The primary reason for this is that the Tribunal did not consider it necessary to make, and therefore did not make, an essential finding, namely, that the underlying facts had been ‘substantially’ disproved.

106 A further reason is that the Tribunal reached its own conclusion, without relying on expert evidence, that since (a) early adult onset was a pre-requisite to the diagnosis of both of the disorders that Dr Lucire diagnosed and (b) there was no evidence of early adult onset, her diagnosis could not stand.

107 The alternative basis on which the Tribunal reached this conclusion was that according to Dr Napper, no competent expert could arrive at a diagnosis of either of these disorders in the absence of evidence of early adult onset. The Tribunal also relied on Dr Napper’s evidence for its conclusion that no competent expert could arrive at a diagnosis of Narcissistic Personality Disorder on the facts that remained after eliminating a ‘grandiose sense of self-importance’ as a trait exhibited by Mr C.

108 In this connection, however, we accept a further submission by Ms Allars. This was that because Dr Napper, the sole expert witness addressing the issue, appeared to treat any psychiatrist whose opinion differed from his own as incompetent, the Tribunal had before it no evidence to support a finding, based on ‘the facts that remained’, that ‘no competent expert’ could have formed the opinions that Dr Lucire formed. Put simply, Dr Napper’s evidence did not address the specific question on which the Tribunal needed to make a determination.

109 In our view, there is a further reason why Dr Napper’s evidence did not constitute a proper basis for the Tribunal’s conclusion. What he did, in essence, was to contest vigorously the diagnosis at which Dr Lucire had arrived, giving reasons why he himself reached quite different conclusions. It was understandable that he should seek to address all aspects of the correctness of Dr Lucire’s diagnosis, since at the relevant time he was Dr C’s treating psychiatrist. But according to the test that the Tribunal adopted and, indeed, the test that we have outlined, his task instead was more limited. It was to consider such parts of the factual material underlying Dr Lucire’s opinion as remained after inaccurate assertions or assumptions on her part were put to one side, and to advise the Tribunal whether, on that reduced factual basis, her opinion could be held by a competent psychiatrist. His evidence undoubtedly traversed these aspects of the matter, but his ultimate conclusion – that her diagnosis was not that of a competent psychiatrist – took many other considerations into account as well.

110 In the light of these rulings, it is strictly not necessary for us to deal with what we have called (at [76]) the ancillary arguments put forward by Ms Allars. But we will indicate that we did not find any of them persuasive.

The orders to be made

111 The outcome of the foregoing reasons is that this appeal must be allowed in part. The first of the Tribunal’s orders, setting aside the agency’s decision to refuse to amend the report under s 44(a) of the FOI Act, should be upheld. But its second order, specifying the nature of the amendment to be made, must be set aside and a different order substituted.

112 In its judgment, at [97 – 106], the Tribunal discussed the scope of its functions in reviewing the agency’s decision and the form that its own orders should take by virtue of its finding that this decision should be set aside. Its conclusion on the latter issue was that the agency should be ordered to attach a notation to Dr Lucire’s report, and to every copy of the report held by the agency, that would indicate the nature of the amendment to be made under s 43(1) of the FOI Act. In the appeal, no challenge was made to this part of the judgment.

113 The notation that the Tribunal ordered to be attached (see [107]) was as follows:-

The attached document is incorrect in a material respect pursuant to the Freedom of Information Act 1989. In particular Dr Lucire’s diagnosis that Mr C (insert full name) has a Narcissistic Personality Disorder and a Paranoid Delusional Disorder is incorrect. For further details see the decision of the Administrative Decisions Tribunal in C v-Secretary, NSW Treasury (No 3) [2003] NSWADT 200.

114 We consider that, for the reasons advanced by the Tribunal, the amendment should continue to be effected by way of notation. But in the light of our reasons and conclusions, the notation should be as follows:-

The attached document is incorrect within the meaning of s 39(c) of the Freedom of Information Act 1989. Specifically, Dr Lucire’s diagnosis that Mr C (insert full name) seemed on clinical examination to have a Paranoid Delusional Disorder and on the documents to have a Narcissistic Personality Disorder has been withdrawn by her. For further details, see the decision of the Appeal Panel of the Administrative Decisions Tribunal in Secretary, NSW Treasury v C [2004] NSWADTAP 6.

115 The Tribunal also stated, at [108], that if the agency discloses to any person (including any other agency or any Minister) any information contained in Dr Lucire’s report, the agency should ensure that the above notation is included when that disclosure is made. In our judgment an order should be made to this effect.

116 At first instance, Mr C indicated that if he were successful he would wish to have an order for costs. The Tribunal set out (at [109]) a procedure whereby he might do so. As this appeal has been upheld in part, we do not believe that either party is likely to be able to establish that there are ‘special circumstances’ warranting an order, as required by s 88(1) of the Administrative Decisions Tribunal Act 1997. But we are prepared to make available to both parties the procedure that the Tribunal outlined in its judgment.

Decision revised 20 February 2004 - to correct citation in Order 3 and paragraph 114