STATE OF NEW SOUTH WALES v BURTON

(1) Allow the appeal and set aside the orders made in the District Court on 7 June and 15 June 2007.

(2) In lieu thereof give judgment for the plaintiff in the sum of $80,000, such judgment to have effect on and from 7 June 2007.

DECISION: (3) Give leave to the parties to file written submissions in respect of appropriate costs orders, if such orders cannot be agreed to, within the following timetable:

(a) Appellant to file and serve submissions on or before 12 December 2008; (b) Respondent to file and serve submissions on or before 19 December 2008,

such submissions to address the appropriate costs orders to be made in respect of the first trial, the remittal hearing and this appeal.

CATCHWORDS: APPEALS - assessment of damages - whether Court of Appeal should assess damages for itself or remit to lower court - where case already remitted twice before - just, quick and cheap resolution of real issues in proceedings - Civil Procedure Act 2005 (NSW), s 56(1) and (2) - DAMAGES - loss of a chance - burden of proof - intervention and treatment for post-traumatic stress disorder - assessment of loss of a chance of a "better outcome" - assessment of degree of likelihood of better outcome - where harm resulted from initial exposure to trauma and, through defendant's negligence, not ameliorated by appropriate care and treatment - likelihood of plaintiff revealing intensity of feelings and responses - what treatment would have been offered - chance of accepting treatment - likely effect of treatment - where aetiology of plaintiff's condition uncertain - effect better outcome would have had on plaintiff's financial loss - EVIDENCE - experts - intervention and treatment for post-traumatic stress disorder - evaluation of medical experts' evidence - where experts' evidence imprecise - whether earlier treatment leads to better outcome - whether "compulsory debriefing" would have assisted or hindered plaintiff - utility of relying on studies - PRACTICE & PROCEDURE - remittal - nature of court's jurisdiction on remittal - scope of order remitting for damages to be assessed on loss of a chance basis - whether trial judge bound to accept assumptions underlying order of remitter as factually correct or whether entitled to depart from them - whether statement of claim should have been amended to allege with greater precision better outcome - whether terms of remitter required fresh finding as to liability - WORDS & PHRASES - "remitter"

LEGISLATION CITED: Civil Procedure Act 2005 (NSW), s 56

Barnett v Chelsea and Kensington Hospital Management Committee [1969] 1 QB 428 Chapman v Hearse [1961] HCA 46; 106 CLR 112 The Commonwealth v Amann Aviation Pty Ltd [1991] HCA 54; 174 CLR 64 Gregg v Scott [2005] 2 AC 176 GW Sinclair & Co Pty Ltd v Cocks [2001] VSCA 47; [2001] ANZ Conv R 522 Houghton v Immer (No. 155) Pty Ltd (1997) 44 NSWLR 46 Kitchen v Royal Air Force Association [1958] 1 WLR 563 LJP Investments Pty Ltd v Howard Chia Investments Pty Ltd (1989) 24 NSWLR 499 CASES CITED: Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638 McGhee v National Coal Board [1973] 1 WLR 1 Murphy v Overton Investments Pty Ltd [2004] HCA 3; 216 CLR 388 Olympic Holdings Pty Ltd v Lochel [2004] WASC 61 Rufo v Hosking [2004] NSWCA 391; 61 NSWLR 678 Sellars v Adelaide Petroleum NL [1994] HCA 4; 179 CLR 332 State of New South Wales v Burton [2006] NSWCA 12; (2006) Aust Torts Repo 81-826 Waterways Authority v Fitzgibbon [2005] HCA 57; 79 ALJR 1816 Watts v Rake [1960] HCA 58; 108 CLR 158

Davidson JRT et al, "Posttraumatic Stress Disorder: Acquisition, Recognition, Course, and Treatment" J Neuropsychiatry Clin Neurosci 16:2; 135-147 (2004) TEXTS CITED: Harvey AG and Bryant RA, "The Relationship Between Acute Stress Disorder and Posttraumatic Stress Disorder: A 2-Year Prospective Evaluation", Journal of Consulting and Clinical Psychology (1999) Vol 67 (6), p 985-988 McFarlane AC, "Debriefing: Care and Sympathy Are Not Enough", Medical Journal of Australia, 2003; 178 (11): 533-534Tony Weir, Tort Law, (2002) p 76

PARTIES: State of New South Wales (Appellant) Christopher Burton (Respondent)

COUNSEL: P Menzies QC/DF Villa (Appellant) M Joseph SC/A Naylor (Respondent)

SOLICITORS: I V Knight, Crown Solicitor (Appellant) Oates and Smith (Respondent)

LOWER COURT DATE OF DECISION: 7 June 2007; 15 June 2007

IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL CA 40422/07 DC 9889/01

ALLSOP P BASTEN JA HANDLEY AJA

27 November 2008 STATE OF NEW SOUTH WALES v Christopher BURTON Headnote

Mr Burton ("the plaintiff") was a marksman in the Tactical Response Group of the New South Wales Police Service. On 2 December 1998, whilst deployed at a farm near Coolabah, in Northern New South Wales, he was fired upon by a farmer, who was holding his young son as a hostage.

In May 1999, the plaintiff saw a psychologist and was diagnosed as suffering from post-traumatic stress disorder ("PTSD"). The plaintiff sued the State of New South Wales as the body responsible for the Police Service's negligence. That negligence was identified as failure to take steps to ensure that the plaintiff obtained early counselling and appropriate psychological or psychiatric care.

On 22 October 2004, the plaintiff obtained judgment in the District Court for $392,182.20. The State appealed from that judgment, with respect to the findings of liability and the quantum of damages. On 10 February 2006, the Court of Appeal dismissed the appeal on liability, but set aside the judgment and remitted the matter to the District Court for damages to be assessed on a loss of a chance basis. On 7 June 2007, Puckeridge DCJ delivered judgment in favour of the plaintiff in an amount of $237,579, finding the State liable for 60% of the plaintiff's harm. In the present appeal, the State challenged the trial judge's assessment of damages.

(i) whether the trial judge properly assessed the damages of a loss of a chance;

(ii) whether the Court should assess the damages of a loss of a chance for itself; and

(iii) if so, what was the likelihood of a better outcome and the extent to which that outcome reduced the severity and duration of the plaintiff's symptoms.

The Court held, allowing the appeal (Basten JA, Allsop P and Handley AJA agreeing):

The trial judge assessed the "better outcome" as being no injury at all and placed a percentage on the likelihood of that better outcome. It was not open to the trial judge to do this. The trial judge did not make an express finding as to the nature of the better outcome, nor did he consider whether his approach was consistent with the scope of the remitter or whether, given the evidence before the trial judge, some departure from the terms of the remitter was justified or required: [16], [18], [37]. Murphy v Overton Investments Pty Ltd [2004] HCA 3; 216 CLR 388, applied.

To the extent that the trial judge accepted Professor McFarlane's evidence, to the exclusion of contrary views, it seems likely that he failed to adopt a correct approach. He did not explain why he placed little weight on the evidence of the State's experts: [53]-[54].

Neither party having submitted that there was any significant obstacle to the Court making its own assessment of the value of the lost chance of a better outcome, it is appropriate that the Court reassess the evidence for itself, taking account of relevant findings made by the trial judge and making further findings as required: [39].

The question to be addressed in the medical evidence was the likely consequence of receiving delayed treatment, measured in terms of possible improved outcomes, and the chance of that occurring: [62].

There was a lack of clear evidence as to the effect of delay in providing treatment for PTSD: [62]. Generally, the experts were unable to assess and/or quantify the chance of a better outcome for the plaintiff: [82]. This imprecision did not mean that the plaintiff failed to prove any chance of a better outcome resulting from psychological debriefing and monitoring, followed by earlier treatment: [83]. Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638, applied.

The plaintiff would not, in December 1998, have revealed sufficient of his symptoms to allow a police psychologist to have decided that he should be placed on restricted duties or stood down: [100].

Given appropriate debriefing and monitoring, cognitive behaviour therapy would have been more likely to commence in late January 1999. If that had occurred, there would have been an improvement of four months in the commencement date of intervention and treatment: [101].

Loss of earning capacity was agreed between the parties and the trial judge assessed general damages at $90,000, without attributing the figures to any particular period. The best that the Court can do is to work with the global figures and assume that any improvement in the severity or duration of the PTSD would be directly reflected in the damages as assessed or agreed: [106].

Because of the uncertain aetiology of the plaintiff's condition, combined with the absence of any clear medical evidence as to the effects of the four month delay in providing psychological treatment, the chance of some better outcome is highly speculative. Because the Police Service's psychologist's intervention appears to have been of some assistance, it may be assumed that earlier intervention would have alleviated the symptoms for a longer period. However, there is little chance that a chronic condition would have been avoided: [109]. Murphy v Overton Investments Pty Ltd [2004] HCA 3; 216 CLR 388; Houghton v Immer (No. 155) Pty Ltd (1997) 44 NSWLR 46; LJP Investments Pty Ltd v Howard Chia Investments Pty Ltd (1989) 24 NSWLR 499, applied.

Taking into account these factors, the value of the plaintiff's lost chance is 20%: [110]. IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL CA 40422/07 DC 9889/01

ALLSOP P BASTEN JA HANDLEY AJA

27 November 2008 STATE OF NEW SOUTH WALES v Christopher BURTON Judgment 1 ALLSOP P: I have had the privilege of reading the reasons in draft of Basten JA. I agree with them and with the orders that his Honour proposes. His Honour's reasons reveal, with clarity, the complexity that can arise in loss of a chance analysis in this field. In particular, they reveal the need to be precise, as far as one can be, about the nature of the so-called "better outcome", because it is the identification of this element which affects so directly the calculation of the percentage chance. Without knowing what the "better outcome" is said to be, a calculation of a percentage loss of chance risks being a meaningless exercise. 2 BASTEN JA: In 1998 Mr Christopher Burton ("the plaintiff") was a marksman in the Tactical Response Group of the New South Wales Police Service. Whilst deployed at a farm near Coolabah in northern New South Wales, he was fired upon by the farmer, who was holding his young son as a hostage. The plaintiff was visibly shaken at the time of the incident and, when he eventually saw a psychologist some six months later, was diagnosed as suffering from post-traumatic stress disorder ("PTSD"). 3 There was no doubt that a cause of the plaintiff's psychiatric disorder was the criminal act of the farmer in firing upon the plaintiff. He did not, however, sue the farmer for his loss. Rather, he sued the State of New South Wales as the body responsible for negligence on the part of the Police Service. That negligence was identified as failure to take steps to ensure that he obtained early counselling and appropriate psychological or psychiatric care. The State did not seek contribution from the farmer. 4 The issue in the present appeal was whether the trial judge had properly assessed Mr Burton's damages, as recoverable from the State. Procedural background 5 The incident at Coolabah occurred on 2 December 1998. Almost two weeks later, the plaintiff was deployed in similar circumstances at a farm near Nimmitabel, where nothing untoward happened, but he demonstrated a degree of nervousness at an unexpected event. 6 The plaintiff saw a psychologist working with the Police Service in May 1999. He was diagnosed as then suffering from PTSD and on 4 June 1999 filed a claim for benefits as a result of being hurt on duty. 7 The claim for common law damages, lodged in February 2001, came to trial in September 2004. On 22 October 2004 he obtained a judgment in the District Court for $392,182.20. 8 The State appealed from that judgment, both with respect to the findings of liability and the quantum of damages. On 10 February 2006 this Court dismissed the appeal on liability, but set aside the judgment and remitted the matter to the District Court for a further assessment of damages: State of New South Wales v Burton [2006] NSWCA 12; (2006) Aust Torts Rep ¶81-826 ("Burton No. 1"). On 7 June 2007, following a hearing which ran for seven days, Puckeridge DCJ delivered judgment in favour of the plaintiff in an amount of $237,579, being 60% of the award at the first trial, including interest to the date of the second judgment. 9 The issue on the present appeal is whether that assessment was properly undertaken. The answer to that question requires consideration of the terms of the remitter from this Court. Nature of jurisdiction on remittal 10 In this Court, on the first appeal, Spigelman CJ (with whom Hunt AJA relevantly agreed) held that damages had been assessed on an inappropriate basis and that the matter should be remitted "for damages to be assessed on a loss of a chance basis": at [24]. His Honour continued: "25 The PTSD was not caused by the failure to have counselling, or take other ameliorative steps. It was caused by the shooting. The relevant loss in this case is the loss of a chance of a better outcome. The focus of attention must be on the extent to which the Respondent's condition would or could have been improved had earlier counselling been provided. 26 Any such reassessment should not extend to the loss of a chance that the [plaintiff] would not have suffered PTSD at all. I do not understand Dr Wright to advance the proposition that that was a significant factor. 27 The onus was throughout on the Respondent, as plaintiff, to establish the existence and extent of the loss of the chance. On the basis of his Honour's findings it appears that there was a loss of such a chance and that, as a matter of causation, the Respondent has suffered some loss or damage. … 28 His Honour made no assessment of the extent of the loss of the chance. …" 11 At the first trial, evidence had been obtained from a number of psychiatrists but the assessment had been undertaken largely by reference to the opinions of Dr Wright. It was Dr Wright's evidence which formed the basis of the assessment held in this Court to be erroneous and which informed the conclusion referred to by the Chief Justice that the PTSD was caused by the shooting and not by the Appellant's negligence. As explained by the Chief Justice at [21]: "The thrust of Dr Wright's evidence is clearly directed to establishing that the omissions on the part of the Appellant were such as to increase the severity and duration of the PTSD which was caused by the shooting incident itself. The references to the possibility that early intervention would have reduced the risk of the Respondent suffering PTSD at all are at best tangential and provide no evidentiary basis for an inference that such contribution to incurring the condition as was made by the Appellant's breach of duty was a 'material contribution'." 12 The terms of the remitter required that the District Court assess the loss of a chance of a "better outcome", being an outcome which assumed that the plaintiff would have suffered from PTSD regardless of the negligence of the State. These remarks gave rise to a difficulty on remittal; the trial judge, being presented with a range of additional medical evidence, was obliged to consider the extent to which he was bound to accept the assumptions underlying the order of remitter as factually correct, or whether he was entitled to depart from them. 13 The circumstances of the present case are not dissimilar to those which arose in Murphy v Overton Investments Pty Ltd [2004] HCA 3; 216 CLR 388. In that case the High Court concluded that the decision below, that the appellants had proved no loss or damage, was erroneous: at [74]. The joint judgment continued: "Though at the end of the day the appellants may fail to prove any loss or damage, it is possible that they will demonstrate that they have suffered loss or damage. The matter should be remitted to the trial judge to assess the damages, if any, to be allowed to the appellants. The trial judge, while hearing the remitted issues, may think it proper to draw inferences in favour of the appellants, if it is concluded that the respondent's wrong itself made quantification difficult." 14 The appropriateness of drawing inferences in such circumstances will be adverted to below. In relation to the scope of the remitter, the joint judgment stated at [75]: "The order for remitter is not an order for retrial. Nothing we have said, however, should be understood as indicating any view about whether, on remitter, an application by either side for leave to reopen its case to lead further evidence should be granted." 15 Assuming the further hearing proceeded on the same basis as the first trial, the terms of the remitter required the identification of the "better outcome" and an assessment of the degree of likelihood of that better outcome, absent the negligent omission of the State. 16 The approach adopted at the first trial imposed liability on the State for the whole of the loss suffered. It did so on the conventional basis that the trial judge was to be satisfied on the balance of probabilities that the negligence of the State materially contributed to the harm suffered. Accordingly, the State was found liable for the whole of that harm. The finding on the second trial that the State was liable for 60% of that harm appears to have assessed the "better outcome" as being no injury at all and to have placed a percentage on the likelihood of that better outcome. In fact, there was no express finding as to the nature of the better outcome, nor was there any discussion as to whether the approach adopted was in fact consistent with the scope of the remittal or whether, given the evidence before the trial judge, some departure from the terms of the remitter was justified or required. 17 Two procedural questions arose from the terms of the remitter: the first was whether the statement of claim should have been amended so as to allege with greater precision the outcome, the chance of achieving which the plaintiff had lost through the conduct of the State. Secondly, attention might have been directed to the evidential basis upon which the newly formulated claim was to be addressed. It is clear from the majority judgments on the first appeal that the terms of the remitter derived from the assessment made by this Court of the evidence then available, being the evidence called at the first trial. As the Chief Justice noted at [28]: "His Honour made no assessment of the extent of the loss of the chance. No submissions were made in this Court on the appropriate assessment. This Court is in no position to make such an assessment. There is a significant dispute on the evidence of the medical witnesses as to the efficacy of early intervention. Further consideration of this evidence should form part of any such assessment. The matter should be remitted to the District Court to assess the loss of a chance of a better outcome." 18 The rehearing proceeded, as it appears, without amendment to the pleadings and on the common assumption that both parties were entitled to call further medical evidence. Understandably, that course was not challenged on appeal, but it had the potential to give rise to a degree of inconsistency with respect to findings made at the first hearing, particularly in relation to what the plaintiff would have been offered and what his response would have been if particular forms of treatment had been offered. At the first trial, his "contributory" negligence was assessed at 10% because he had failed to report symptoms of stress and to take steps to seek counselling prior to May 1999: Burton No. 1 at [81]. This finding was not varied on appeal and was treated as beyond review at the remittal hearing. To similar effect, it was not open to the trial judge to conclude that the loss of a chance extended to an outcome that the plaintiff "would not have suffered PTSD at all": at [26]. That finding was reached on the basis of the medical evidence at the first trial. 19 A separate question was whether the terms of the remitter required a fresh finding as to liability, once the harm suffered was identified as something different to that which had been pleaded. It was concluded in this Court that "it appears that there was a loss of such a chance and that, as a matter of causation, the [plaintiff] has suffered some loss or damage": at [27]. On the further hearing, the State continued to assert (as it had in this Court during the first appeal) that no real chance of a better outcome had been lost through its conduct. 20 There was no suggestion in this case that the difficulties caused by a remitter of part only of a matter caused insuperable difficulties at the second hearing: cf Waterways Authority v Fitzgibbon [2005] HCA 57; 79 ALJR 1816 at [14]-[20] (Gleeson CJ, McHugh, Gummow and Hayne JJ agreeing). Nevertheless, it seems possible that error arose on the remittal, because some of the issues had not been sufficiently identified and the consequences of particular findings addressed. Nature of assessment required on remitter 21 Because it is doubtful that the steps taken by his Honour in assessing damages at the second hearing complied with the terms of the remitter, it is necessary to consider in more detail precisely what exercise was required to be undertaken and whether the approach adopted by his Honour caused the further hearing to miscarry. 22 An element of the tort of negligence is damage to the victim. Damage may be physical, psychiatric or financial. According to conventional principle, it must be identifiable and its existence must be established, as must the causal connection to a breach of duty. 23 In contract claims, damages may also be discovered in loss of an economic opportunity: that is, the loss of that promised under the bargain: The Commonwealth v Amann Aviation Pty Ltd [1991] HCA 54; 174 CLR 64. A similar approach has been adopted in relation to a claim under s 82 of the Trade Practices Act 1974 (Cth) for relief in respect of misleading or deceptive conduct: see Sellars v Adelaide Petroleum NL [1994] HCA 4; 179 CLR 332. In relation to solicitors' negligence, a client whose case is not filed in time, through the negligence of a solicitor, may recover for the loss of a chance to bring the proceedings to fruition: Kitchen v Royal Air Force Association [1958] 1 WLR 563. What is lost in the latter case, and is to be valued, is a chose in action. 24 The approach mandated by this Court on the first appeal required that the harm for which the State was responsible was the loss of a chance of a better outcome. This loss, it was held on the balance of probabilities, had been caused by the State's breach of duty. It was a sufficient harm to constitute the tort of negligence, other elements being established. Liability no longer being in issue, the purpose of the remitter was to allow that chance to be evaluated. 25 Nevertheless, there were elements of the harm which remained unidentified and hence the causal link with the breach of duty remained in issue. In cases involving a breach of contract causation must be established on the balance of probabilities. In GW Sinclair & Co Pty Ltd v Cocks [2001] VSCA 47; [2001] ANZ Conv R 522 at [29], Charles JA (Brooking and Buchanan JJA agreeing) held that where an element of the loss involved a choice to be made by the plaintiffs, that was to be established on the balance of probabilities. That approach was adopted by McLure J in Olympic Holdings Pty Ltd v Lochel [2004] WASC 61 at [123], in a passage expressly relied on by this Court in Burton No. 1 at [27], although referring generally to the plaintiff's onus: see also Sellars at 355; Rufo v Hosking [2004] NSWCA 391; 61 NSWLR 678 at [9] (Hodgson JA). Without identifying the care or treatment that should have been offered, and might have been effective, it is difficult to suppose that the plaintiff had established what he would have done had it been proffered. If this issue were to be treated as part of the assessment of loss, as a past hypothetical, it would not be a fact requiring proof on the probabilities. If, however, it were to be treated as a link in a causal chain, despite being a past hypothetical, it must be so proved. The answer will depend (in principle) on how the lost chance is identified: see Sellars at 360-364 (Brennan J). 26 The required approach appears to conform to that adopted by the dissenting members of the House of Lords in Gregg v Scott [2005] 2 AC 176. Thus, at [17] Lord Nicholls of Birkenhead stated: "In order to achieve a just result in such cases the law defines the claimant's actionable damage more narrowly by reference to the opportunity the claimant lost, rather than by reference to the loss of the desired outcome which was never within his control. In adopting this approach the law does not depart from the principle that the claimant must prove actionable damage on the balance of probability. The law adheres to this principle but defines actionable damage in different, more appropriate terms. The law treats the claimant's loss of his opportunity or chance as itself actionable damage. The claimant must prove this loss on balance of probability. The court will then measure this loss as best it may. The chance is to be ignored if it was merely speculative, but evaluated if it was substantial: see Davies v Taylor [1974] AC 207, 212, per Lord Reid." 27 The medical evidence, in cases of unknown aetiology, is likely to be imprecise and multifactorial. Further, the likelihood of a better outcome may be inversely proportional to the degree of success, perhaps because the treatment works in some cases but not others and it is not possible to place the plaintiff into one category rather than another. For example, early treatment might have a 20% chance of an immediate and complete cure, a 40% chance of halving the severity of the symptoms and a 60% chance of a significant but moderate reduction in either the length or severity of the symptoms. 28 In Gregg v Scott, Lord Nicholls noted at [37]: "Identifying the nature and extent of Mr Gregg's cancer at the time of the mistaken diagnosis (the first question), so far as this could be achieved with reasonable certainty, did not provide a simple answer to what would have been the outcome had he been treated promptly (the second question). There were several possible outcomes. Recourse to past experience in other cases, that is statistics, personalised so far as possible, was the best that could be done. These statistics express the various possible outcomes in percentage terms of likelihood." 29 His Lordship had earlier remarked that, "losing a chance of saving a leg is not the same thing as losing a leg": at [25], referring to Tony Weir, Tort Law (2002) at 76. However, he did not take the question of assessment further. His view as to the correct approach was shared by Lord Hope of Craighead, but they did not constitute the majority in the House of Lords. Lord Nicholls concluded his judgment in the following manner at [60]: "Had a majority of your Lordships shared this view the case would have been remitted to the Queen's Bench Division for damages to be assessed. In that event some general guidance would have been required on the assessment of damages in this type of case." 30 The trial judge was potentially required to answer a number of factual questions, which may be identified in the following broad terms:

(a) What was the likelihood that the plaintiff would reveal the intensity of his feelings and responses, so as to permit diagnosis and an offer of treatment, and when might that have occurred? (b) At the time so identified, what treatment would have been offered and with what chance of acceptance? (c) What would have been the likely effects of such treatment if undertaken? (d) If ineffective, would alternative treatment have been offered and, if so, when? (e) What were the likely effects of any alternative treatment, if undertaken? (f) If treatment would have led to a better outcome, what effect would that better outcome have had on the financial loss suffered by the plaintiff? (g) What was the chance of the better outcome eventuating? 31 It appears that the first question was treated as having been answered at the first hearing. In relation to the second question, his Honour appears to have accepted that early intervention would have involved both providing information as to available welfare assistance and monitoring the plaintiff's condition. The initial "treatment" was identified as psychological debriefing and psychological counselling. A further treatment which, on the medical evidence, was more likely to be effective in a serious case, namely cognitive behaviour therapy ("CBT") was, according to the State, unlikely in 1998 and 1999 to have been made available until "several weeks after the trauma and after other treatments had been found to be non-effective": at [103]. Whether or not his Honour accepted that position is not entirely clear but he did hold at [104]: "On the medical evidence before me CBT as presently performed would probably not have been available for the plaintiff in December 1998." 32 This imprecision was significant, to the extent that his Honour accepted the expert opinion of Professor A C McFarlane that the plaintiff suffered a 60% loss of chance of an improvement from treatment available at the time of the Coolabah incident, to which he added a further 20% if the plaintiff had received "the treatment early": at [110]. 33 In relation to the effect of early intervention (the third question), his Honour noted the State's submission that it was "unlikely to be beneficial": at [102]. That submission was not accepted. His Honour noted the agreement of the experts that, in the relevant period, effective treatments included "exposure treatment", "cognitive restructuring", medication and EMDR (eye movement desensitisation and reprocessing): at [104]. The primary focus appears to have been on CBT, a term not included in the list at [104], but which was apparently covered by the first two terms. The likely effect of treatment needed to be assessed against the characteristics usually conducive to success. Thus, it was recognised that there is a group of potential patients who are not responsive to CBT; it appears there may also be a group who are not responsive to any form of treatment: at [103]. His Honour noted evidence of Professor McFarlane that the plaintiff "did not have many of the characteristics which would be associated with potentially poor treatment outcomes": at [111]. It is not clear how that was taken into account, but as his Honour largely accepted Professor's McFarlane's conclusions, it should be assumed that, although the plaintiff did not respond to treatment provided in and after late May 1999, he was considered not to be amongst the group who would have been non-responsive to early intervention. 34 More importantly, the effects of the forms of intervention which would have been provided absent negligence were not identified with any degree of precision. Accepting Professor McFarlane, his Honour held that, had the plaintiff been treated from 1 January 1999 (after the Nimmitabel incident) "there was a 60% chance of improvement": at [124]. Elsewhere, his Honour found a "significant chance" or "real possibility" that early intervention and treatment would have "reduced the subsequent severity and duration of the condition: at [127], [128]. These critical findings were too imprecise to constitute a proper answer to the matter remitted for determination. In particular, they did not permit an answer to the final question (and none was given), namely the extent to which the possible reduction in the duration and severity of the condition would have affected the financial loss suffered by the plaintiff. In this context, it is to be recalled that the basis of remitter was that the post-traumatic stress disorder would have occurred in any event and that the loss of a chance did not extend to an absence of PTSD. 35 The finding that, absent negligence, the plaintiff had a 60% chance of a better outcome said nothing about the nature and extent of the better outcome: it merely indicated that the plaintiff had a 40% chance of no better outcome. 36 By notice of contention, the plaintiff sought to uphold the ultimate conclusion on the basis that the "consequences from the breaches of duty by the Appellant, would have been reduced by 40% if early counselling had been provided": contention 6. The misconception underlying that contention may indicate why his Honour sought to resolve the issues on remittal in the erroneous manner indicated above. 37 The failure to make critical findings, especially as to the extent of the anticipated better outcome, require that the appeal be upheld and the judgment below set aside. The consequential question is what steps should be taken to resolve the outstanding issue. Resolution of appeal 38 The trial judge, whilst assessing conflicts in the medical evidence, did not suggest that his assessment of the experts turned in any specific respect on their presentation in the witness box. That is not to deny the possibility that, even in respect of experts, the subtle influences of demeanour may not be significant. However, the present question is not whether the findings themselves demonstrated error, but whether the matter should be remitted for a third hearing or whether the better course is for this Court to make its own assessment of the evidence. The latter course is not ideal, but may accord better with the court's duty to exercise its powers so as "to facilitate the just, quick and cheap resolution of the real issues in the proceedings": Civil Procedure Act 2005 (NSW), s 56(1) and (2). That obligation may not provide certain answers in cases where factors pull in different directions, but it is a direction which must be given significant weight in its own right and will often favour disposal of a matter in this Court on a rehearing. 39 Neither party having submitted that there was any significant obstacle to the Court making its own assessment of the value of the lost chance of a better outcome, it is appropriate that the Court reassess the evidence for itself, taking account of relevant findings made by the trial judge and making further findings as required. 40 Before considering the evidence proffered at the second trial and the manner in which the trial judge assessed it, it is necessary in these circumstances to identify more precisely the legal approach which was mandated by the earlier decision of this Court. Legal principles: recovery for loss of a chance 41 The factual difficulties presented by the present case result to a significant extent from the uncertain aetiology of the mental illness. That difficulty is exacerbated by potentially cumulative contributions from two separate events or courses of conduct. In some circumstances, a later event is a foreseeable outcome of the earlier negligent event: see, eg, Chapman v Hearse [1961] HCA 46; 106 CLR 112. In other cases an initial disease or injury may be exacerbated by subsequent negligent medical intervention. The present case falls into a category in which the compensable harm results from an initial exposure to trauma or injury which, through negligence of the defendant, is not ameliorated by appropriate care or treatment. This category includes cases such as McGhee v National Coal Board [1973] 1 WLR 1 (in which a worker was exposed to brick dust which carried the risk of dermatitis, a risk which could have been mitigated by the provision of washing facilities which the employer negligently failed to provide) and Barnett v Chelsea and Kensington Hospital Management Committee [1969] 1 QB 428 (in which a night-watchman, suffering from arsenic poisoning, was misdiagnosed at a local hospital and died some hours later). 42 In McGhee, the initial exposure to brick dust was not a breach of duty on the part of the employer: the failure to provide washing facilities was. As explained by Lord Simon of Glaisdale, the law holds that a plaintiff "is entitled to damages for the injury if he proves on a balance of probabilities that the breach or breaches of duty contributed substantially to causing the injury": at p 8. His Lordship continued: "If such factors so operate cumulatively, it is, in my judgment, immaterial whether they do so concurrently or successively." 43 McGhee was not a case where reliance was placed upon loss of a chance of avoiding injury. In circumstances where the risk of injury had materialised, as Lord Reid explained, "I can see no substantial difference between saying that what the defender did materially increased the risk of injury to the pursuer and saying that what the defender did made a material contribution to his injury": at 5. 44 McGhee was not applied in the present case because, as the Chief Justice explained, the evidence at the first trial did not justify a finding that the breach of duty of the State substantially contributed to the psychiatric injury. The State is therefore not responsible for PTSD as the harm caused by its negligence. Nor is it responsible for a divisible part of the harm, the harm being indivisible. Yet the finding of the trial judge on the remitter was consistent only with the inference that the breach of duty had made a quite substantial contribution (60%) to the whole of the injury suffered by the plaintiff. 45 The finding of the trial judge was that, despite being shot at and suffering a traumatic response, if offered counselling and the opportunity for treatment, (a) the plaintiff would have availed himself of the offer and (b) he stood a 60% chance of avoiding any harm at all. Of course, a similar outcome would result from a finding that the plaintiff stood a 100% chance (absolute certainty) of avoiding all but 40% of the loss, which could be ascertained by reference to a number of weeks off work, a fresh assessment of pain and suffering and other items. The latter course was not the approach adopted by the trial judge. 46 The component of speculation involved in such an exercise is not diminished when it is taken to the next stage, which is to determine the extent, for example, that the victim's earning capacity has been diminished by the exacerbation of the condition. The overall exercise is a pragmatic attempt to require the tortfeasor to pay compensation proportional to the loss suffered. Even if this loss is identified as the loss of a chance, the assessment of its value inevitably requires, in the case of personal injury, assigning a monetary value to an actual or anticipated degree of impairment. 47 If the contributing causes were not an existing condition and medical negligence, but the concurrent or sequential effects of two tortious events, the victim would not be required to apportion responsibility. In the present case there were, it appears, two tortfeasors and the obligation to apportion responsibility fell on the plaintiff because he sued only one. 48 If this had been a case where the tortfeasor's conduct did not exacerbate an existing condition, but led to a result which would probably have occurred in any event, the tortfeasor would be required to provide an evidential basis for disentangling the consequences of the pre-existing condition: see Watts v Rake [1960] HCA 58; 108 CLR 158. In a case where the tortfeasor's conduct aggravates an existing condition, so long as the aggravation is foreseeable, the tortfeasor will be liable for the whole of the loss, subject to its ability to establish an appropriate allowance for the possibility that the disability would have eventuated by other means. However, to impose recovery for a loss of a chance places the burden of proof on the plaintiff in establishing liability to demonstrate, presumably on the balance of probabilities, the nature and extent of the better outcome which may have been foregone. Assessment of medical evidence 49 As noted above, his Honour carefully summarised the evidence of the psychiatric experts called by the parties. He also identified some relevant material in the literature which had been tendered before him. In assessing that evidence, at [101]-[122], his Honour accepted the evidence called for the plaintiff from Professor McFarlane and Dr Wright, and to the extent that they made concessions favourable to the plaintiff, also relied upon the experts called for the State. At [113] his Honour commented: "If the court were to adopt the submission of the defendant that an assessment of the loss of a chance of a better outcome from early intervention was minimal and so small as to be in effect speculative the court would have to place little weight on the evidence of both Professor [McFarlane] and Dr Wright and not take account of the concessions made by Professor Tennant and Dr Parmegiani if the plaintiff continued to have stressful symptoms." 50 His Honour also considered some of the medical literature, which emanated from highly reputable sources. At [116] his Honour noted: "The medical literature that has been tendered to the court shows that whilst there is a debate within the psychological and psychiatric profession as to the efficacy and benefits of intervention and/or early intervention, the views of Professor [McFarlane] and Dr Wright are not totally unacceptable views within the profession." 51 The existence of debate is incontrovertible: there was no suggestion that the views of any medical expert called at the trial stood so far from the mainstream of the debate as to be totally unacceptable, or even unrepresentative. Nor was it entirely appropriate to excise their views from those expressed in the literature, if that were intended. More importantly, however, his Honour continued at [120]: "From the medical literature before the court treatment for PTSD sufferers is being appraised within the profession and the benefits of early intervention as compared to later intervention is being questioned. It is not for the court to arbitrate any ongoing dispute within the profession and find as a fact the benefit of particular treatment and/or the benefit of early intervention." 52 True it is that the role of the court in a civil proceeding brought by a claimant against an employer does not in any respect call for the court to "arbitrate", or otherwise determine, which side of a professional dispute is right or wrong in principle. Nevertheless, the court is required to form a view where one side of the debate affirmatively supports the plaintiff's claim and the other side does not. If the court is unable to form a firm preference for the view which supports the plaintiff, then the plaintiff must fail because he or she bears the burden of proof. On the other hand, in assessing the nature of the chance which has been lost, the trial judge is well able to take differing sides of a debate into the balance. If one side supports a likelihood of success of a particular treatment at 60%, while the other assesses the likelihood at 20%, if the court is unable to express a preference for either view, it may well assess the likelihood at 40%. If it forms a preference for one view or the other, it does not follow that the court should move to one position or the other, but rather suggests that it may wish to move from the mid-point to some degree. It will only adopt the percentage proposed by one camp if it is satisfied that the other must be wholly excluded from the calculation. 53 To the extent that the trial judge accepted the evidence of Professor McFarlane, to the exclusion of contrary views, it seems likely that he failed to adopt a correct approach. To the extent that the medical literature reflected views not consonant with those of Professor McFarlane, it would appear that his Honour disregarded it. 54 To the extent that his Honour's adoption of the views of Professor McFarlane and Dr Wright involved placing little weight on the evidence of the State's experts, it is curious that his Honour made no reference to that inference, nor indicated why he might have been happy to adopt such an approach. History of treatment 55 The breaches of duty identified at the first trial were succinctly stated at [125] in the judgment on the remittal hearing as:

(a) failing to monitor the plaintiff, particularly through the agency of Sergeant Jennings, and (b) failing to instruct the plaintiff of the benefit of debriefing, psychological counselling and psychological treatment following the traumatic incident at Coolabah. 56 Absent those breaches of duty, his Honour found that there would have been "a significant chance" or "a real possibility" that he would have revealed the symptoms of stress which he experienced after the shooting incident. 57 It is not entirely clear, however, when that would have occurred. His Honour noted at [99] that, in his first judgment, he "did not consider that the plaintiff would have availed himself of the welfare facilities within the department at the time he was required to attend at Nimmitabel as a marksman as he had at that time decided to 'tough it out'": at [99]. He does not appear to have expressly departed from that finding in the second judgment and it may therefore be assumed that the plaintiff availing himself of welfare facilities would have occurred after 15 December 1998. On the other hand, at [126], in his final conclusions, his Honour stated: "I consider that there was a significant chance that had the plaintiff been instructed as to the benefit of debriefing and/or early psychological counselling he would have informed the counsellor of the stress symptoms he was experiencing and a consequent possibility that he would not have been required to attend the siege at Nimmitabel. As I have stated previously the plaintiff was deterred from speaking openly about his problems as a result of what Inspector Ruming said to him at Coolabah." 58 To the extent that there is variation from the earlier finding at [99], it appears to have been based on the view that, in relation to choices to be made by the plaintiff, his Honour was not considering the issue on the balance of probabilities, but by reference to possibilities. As I would understand the law, as set out in Burton No. 1, the latter approach was erroneous. 59 Further, at [124] his Honour had stated: "I consider that had the plaintiff been monitored properly there was a significant chance that the plaintiff would have been relieved of operational duties and would not have been required to attend the siege at Nimmitabel. It was the opinion of Professor [McFarlane] that had he been treated from the 1st January 1999 such treatment would have involved withdrawing the plaintiff from the tactical response unit and had that occurred there was a 60% chance of improvement." 60 For present purposes, it may be inferred that treatment was thought likely to have occurred from the beginning of 1999 on the basis that the plaintiff's acute stress symptoms would have been identified at a psychological debriefing in mid-December and subsequent monitoring over the next two weeks. It is of some importance that "stress debriefing" allows for revelation of symptoms and diagnosis of psychological difficulties. It is not an effective form of treatment in itself. Indeed, in a paper published by Professor McFarlane in 2003, he referred to a report of a further study which, he stated, "adds to the now substantial evidence that psychological debriefing has no value in prevention": McFarlane AC, "Debriefing: Care and Sympathy Are Not Enough", Medical Journal of Australia, 2003; 178 (11): 533-534. 61 As a result of an increasing irritability in early May 1999, the plaintiff "referred himself" to the police psychologist: see Report, Dr Murray Wright, 28 July 2000, p 3.5. The psychologist, Senior Sergeant Lette, made arrangements for him to receive psychological counselling from Mr Nick Cocco. Following five sessions of treatment, Mr Cocco reported on 5 July 1999 stating that the plaintiff required ongoing treatment for PTSD and requesting approval for additional sessions. In a report prepared in April 2003, Dr Lana Kossoff described his treatment with Mr Cocco in the following terms, on the basis that it had continued from 27 May 1999 until 8 February 2001 (p 5): "He was treated with a combination of cognitive behaviour therapy, hypnosis and EMDR. He also had psychoeducational input regarding the nature and treatment of PTSD and supportive counselling regarding his adjustment to being off work. Mr Burton was placed on Zoloft (? dose) by his general practitioner which helped him to sleep but he ceased it after about one month as he states he 'hates medication' and spoke to his doctor about weaning off it. He has not tried any other medication." 62 It would appear that the relevant delay in commencing potentially effective treatment, on the plaintiff's case, was a little under six months. The question to be addressed in the medical evidence was, therefore, the likely consequence of such a delay (measured in terms of possible improved outcomes) and the chance of that occurring. As his Honour recognised, there was a lack of clear evidence as to the effect of delay in providing treatment for PTSD. There was evidence available in 1998 confirming the utility of a diagnosis of acute stress disorder as a precursor to chronic PTSD: see Harvey AG and Bryant RA, "The Relationship Between Acute Stress Disorder and Posttraumatic Stress Disorder: A 2-Year Prospective Evaluation", Journal of Consulting and Clinical Psychology (1999) Vol 67 (6), p 985-988. Professor McFarlane noted that the plaintiff had symptoms of acute stress immediately following the incident at Coolabah. He stated in the report of the joint experts: "Acute stress disorders can be diagnosed as early as two days after the exposure to a traumatic event (Bryant and Harvey, Acute Stress Disorders: A Critical Review of the Diagnostic Issues, Clinical Psychological Review 1997, Vol 17 pp. 757 to 773). Bryant et al 1999 state that: 'A major use of this diagnosis (acute stress disorder) is that it can identify many individuals in the acute phase who will subsequently develop chronic post traumatic stress disorder (PTSD). For example, between 78% and 82% of motor accident survivors who satisfy the criteria for acute stress disorder suffer PTSD six months post trauma'." 63 There was a difference of view between Professor McFarlane and Professor Tennant as to whether very early symptoms were predictive of other disorders, but that does not appear to have affected the outcome. There was also a difference of view as to whether "compulsory debriefing" would have assisted or hindered: it was part of the acknowledged breach of duty that there was no psychological debriefing in the days following Coolabah. 64 There was general agreement as to the treatments which should have been offered between January and May 1999. In substance the treatments were not different from those noted above, which were provided by Mr Cocco from late May 1999. 65 Both Professor Tennant and Dr Brown indicated in answer to the question of the effect of delay, in the joint conference of experts, that the definitive study undertaken by the British Psychological Society and the Royal College of Psychiatrists in Britain, known as the National Institute for Clinical Excellence (NICE) Guidelines, published in 2005, found that only two studies addressed the issue of early intervention and both indicated that early intervention is no more effective than later intervention. The NICE Guidelines indicated there were no randomised control trials and the view was based upon "observational studies". It is clear, therefore, that Professor Tennant and Dr Brown believed that there was no quantifiable prospect that early intervention would have resulted in a better medical outcome. Dr Parmegiani was of the view that, in accordance with findings of the Cochrane Collaboration Study, first published in 1997, early intervention was likely to have a deleterious effect. 66 Of the two experts briefed by the plaintiff, Dr Wright remained imprecise, his relevant answer being in the following terms: "Dr Wright believes that the intervention referred to in question 1 and the treatments referred to in question 2 would have improved the outcome if delivered early as opposed to late on the basis of the literature which outlines that intervention in anxiety and other co-morbid conditions delivered early in the course of an illness often will lead to a better outcome than the same treatments delivered late in the course of an illness." 67 The opinion of Dr Wright, as conveyed in the evidence at the second hearing, was not significantly different from that provided at the first trial, which had led to the conclusion set out by the Chief Justice and noted above. 68 Professor McFarlane was more precise in response to the relevant question, although his answer needed to be understood in context. "To date, there is not an adequate literature in the form of properly conducted clinical trials that allocate people to early and late treatment in the weeks or months immediately following a traumatic event. The literature which exists, such as the Gillespie et al paper of the [Omagh] bombing, is about a clinically-based intervention where the people who presented early may have been very different from those who presented late for treatment." 69 On the basis of that material, which appears to be specific to PTSD, it would have been impossible to provide any numerical indication as to the effect of delay. However, Professor McFarlane continued: "On the basis of the treatment trials generally showing that two-thirds of patients sustain a significant improvement would indicate a 66 per cent expectancy of an improved outcome. There is a further 14 per cent of improved outcome probable in Mr Burton's case if he had not been re-exposed to the incident at Nimmitabel. This is not addressed in the standard treatment literature, hence this question is answered both with reference to the published clinical literature and reference to knowledge of the etiology of post traumatic stress disorder and sound clinical practice." 70 There was some degree of elaboration available from Professor McFarlane's written report dated 22 May 2006. He expressed the view (at p 7) that "a conservative estimate is that [the plaintiff] was deprived of an 80% lost chance of a better medical outcome" which he said was based upon a "presumption" that he continued to suffer from PTSD despite having been given treatment at a later time, that he therefore fell into a group of individuals who were relatively treatment unresponsive, that the PTSD had been preceded by an acute stress disorder and: "The delay in receiving treatment for his PTSD is also a further significant loss of chance of improvement." 71 Because the question related to the loss of chance of improvement due to the negligent delay, the last sentence casts doubt upon the percentage given earlier in the same paragraph. 72 In his written report, Professor McFarlane referred to the NICE Guidelines, noted the comparison of studies, but gave no indication that the outcomes were not supportive of his conclusion. However, in considering the question of treatment generally, he stated (at p 6): "Against this background, it is reasonable to presume a conservative estimate is that approximately 80% of patients given an evidence-based treatment for posttraumatic stress disorder, there is a significant probability that the individual will benefit to a significant degree. There is little systematic research also looking at the timing of the treatment. In general, it is very difficult to conduct treatment trials in the immediate aftermath of events given the problems of recruitment and the delayed seeking of treatment which is frequently observed." 73 Although the first sentence in this passage is difficult to understand, the suggestion appears to be that 80% of patients provided with an appropriate treatment will benefit. That is qualified by the following statement that there is little or no evidence as to the effect of timing. His conclusion at the joint conference of experts, referred to earlier, should presumably be understood as indicating a similar view in different language. 74 Professor McFarlane's figures, which were not sourced to any particular study and acknowledged as being based on no scientific evidence, were qualified in the course of cross-examination. Thus, he was asked about the percentages of persons diagnosed with PTSD who would recover wholly or in part. His response referred to a US study known as the "National Co-morbidity Study" which obtained a random selection of 8,500 people: Tcpt, 31/10/06, p 25. He said that the study "came to the conclusion that if you had a post-traumatic stress disorder you had a 60 per cent chance of it resolving, that's with or without treatment, but 40 per cent of people, once you've reached 72 months, continued to have the condition". He later clarified that answer noting that of the 60% who will recover anyway "if they had received treatment they would have suffered from the disorder for a briefer period of time": at p 26. They would also have a lower risk of subsequent disorder following treatment. He continued: "I would not agree about the remaining 50 per cent because I would say that of that 50 per cent, in other words the one[s] who will still have the condition, with treatment two-thirds of them would have a very significant improvement." 75 It is apparent from this evidence that Professor McFarlane had already put the plaintiff into the particular category of those who did not recover fully, with or without treatment, because he remained symptomatic some six years after the event, a circumstance referred to in his report as a "presumption". Asked to support his conclusion he stated: "What I rely upon to support that proposition is if you look at treatment trials, and again this is obviously a generalisation, that in most treatment trials you will have - two-thirds of people will have a significant improvement and one-third have minimal or no improvement." 76 Subsequently, Professor McFarlane was examined as to what he considered "reasonable treatment" in 1998. He emphasised that debriefing was not a treatment but an intervention. He noted that the incident at Nimmitabel may have exacerbated the symptoms the plaintiff was suffering, but emphasised that "the optimal intervention would have involved CBT probably several weeks later, and that would have been planned at the time. You wouldn't have done it right there and then": Tcpt, 30/10/06, p 71(50). 77 Professor McFarlane was taken to the statement in the NICE Guidelines to the following effect: "Under current practice, given that the efficacy of CBT for PTSD was only established in the late 1990s, early interventions of this kind are a new development and have only recently been the subject or research." 78 He was then taken to passages in the NICE Guidelines which indicated, as noted above, that there was an absence of evidence in support of the greater effectiveness of early intervention. In response, he took the Court to page 87 of the Guidelines that said "when trauma-focused CBT is delivered between 1 and 6 months after the incident, there is evidence suggesting that it is effective for people at risk of developing chronic PTSD compared with the effect of being on a waiting list": Tcpt, 30/10/06, p 80. (The phrase "on a waiting list" referred to a control group.) He then proceeded to explain why the aggregation of data may have resulted in a more negative report of the studies taken as a whole. 79 However, taking this response at face value, it is clear that the comprehensive study prepared by NICE failed to identify any differential outcome resulting from treatment delivered "between 1 and 6 months after the incident". Yet that window would include the date on which Professor McFarlane considered that early intervention would have been appropriate (1 January 1999) and arguably the date when treatment did commence (27 May 1999). (Counsel for the plaintiff at trial treated this statement as relevant only to treatment undertaken wholly within the five month period: Tcpt, 12/12/06, p 280.) 80 Reading Professor McFarlane's evidence as whole, together with the documents upon which he placed reliance, it is clear that, in his opinion, and assuming that the plaintiff still had PTSD six years after the event, there was a 66% chance that treatment provided in January 1999 would have permitted an improvement in his condition. The chance of improvement would have increased, he concluded, to 80% if he had been removed from field operations prior to Nimmitabel. To place responsibility for the additional 14% at the feet of the State required a finding that his deployment to Nimmitabel was a consequence of the State's negligence; that matter will be considered further below. 81 Two things are apparent from Professor McFarlane's evidence at this point: first, he appears to have accepted that the plaintiff received appropriate treatment in May 1999 but expressed no view as to whether that had achieved an improvement over the likely outcome absent that treatment. Secondly, he made no comment as to the increase in improvement (if indeed that was what he was referring to in his evidence) by backdating the treatment by five months. Indeed, he appeared to accept that there was no scientific basis to express an opinion as to the effect of such a variation in the date of intervention. 82 This conclusion is consistent with the fact that Professor Tennant, Dr Brown and Dr Parmegiani were each unable to assess any chance of a better outcome from earlier intervention. Dr Wright thought there would be a better outcome, but was unable to put a percentage on the chance. This consistency is unsurprising since all five were reputable experts in the field and all recognised and relied upon the validity of the same underlying research. 83 It does not follow that the plaintiff failed to prove any chance of a better outcome resulting from psychological debriefing and monitoring, followed by earlier treatment. The fact that precise quantification is impossible and that the Court "must speculate to some extent" does not prevent the valuation of the chance: see Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638 at 639-640 (Brennan and Dawson JJ) and at 643-644 (Deane, Gaudron and McHugh JJ). 84 Dr Wright, consistently with his earlier evidence, adhered to the view that earlier intervention would have decreased the duration and severity of the condition. Professor Tennant agreed that the plaintiff, by six months after the incident when he saw Mr Cocco, had clearly established PTSD warranting appropriate treatment: Tcpt, 06/11/06, p 177. The following question and answer are recorded: "Q. You have expressed some opinions about whether or not there was any effect in the delay factor. … I think your answers were going to about a one month delay. I put this to you, that a one month delay compared with say a six month delay brings about a marked difference in the efficacy of effective psychological treatment in the case of the PTSD in terms of the duration and the severity? A. Only in the duration of symptoms up until the point where treatment beings, but after that, no." 85 Professor Tennant was critical of reliance upon studies as to the effectiveness of early intervention in respect of psychoses (such as schizophrenia), as opposed to affective disorders, such as that suffered by the plaintiff. 86 The trial judge said that Professor Tennant agreed with the proposition that if the plaintiff were significantly symptomatic before being deployed at Nimmitabel, and had at that time been in the hands of a psychologist or psychiatrist, "he would not necessarily have gone on to develop a PTSD": at [78]. The Court was not taken to any evidence which supported that finding in terms. In cross-examination, Professor Tennant gave the following evidence (Tcpt, 06/11/06, pp 173 and 177): "Q. … if he had been significantly symptomatic before Nimmitabel, it is the very kind of incident that you, the clinician, would want to keep him away from? A. If there has been the likelihood of him being shot at absolutely. If he was significantly symptomatic prior to that incident, you would not have sent him back into the second siege, that it clear. … Q. So if it is the case that the plaintiff probably had acute stress disorder … that early CBT would have had the effect of either minimising PTSD from developing or minimising its duration of severity, would you agree with that proposition? A. It would have alleviated the symptoms that pre-existed the institution of treatment, yes." 87 Dr Parmegiani stated in his report, prepared in June 2006 (at pp 4, 5): "At present, prevention of PTSD involves watching persons who have experienced a traumatic event, for the first four weeks. Those who present with Acute Stress Disorder are then treated with cognitive behaviour therapy. … Again, with the benefit of hindsight, a two-month delay probably increased the risk of an adverse medical outcome. There is however no accurate method of quantifying the loss of chance of a better medical outcome." 88 In cross-examination, he declined to stand by that statement, noting that during the experts' conference it had been pointed out that "after the first four weeks, where you can reduce the chance of a poor medical outcome by cognitive behaviour therapy, once the four weeks have expired and you qualify for post-traumatic stress disorder, … the timing of the treatment does not matter in terms of outcome": Tcpt, 02/11/06, p 58. He explained that the view expressed in the report had been based on extrapolating from a study of acute stress disorder to post-traumatic stress disorder. He declined to rely upon his own "clinical experience" as to the effects of delay, as being an unscientific approach. 89 Because the literature was inconclusive and the experts were at odds, his Honour noted at [119]: "Professor [McFarlane] disagreed with the statement in [the NICE Guidelines] that evidence suggested that the chances that a PTSD sufferer would benefit from treatment do not decrease with time elapsed from the traumatic event. Both Professor [McFarlane] and Dr Wright were of the opinion that clinical experience would contra-indicate such a statement." 90 At least in relation to Professor McFarlane, the Court was not taken to any evidence to support that proposition. Whilst no doubt an intuitive response based on clinical experience would not be dismissed out of hand, the overwhelming thrust of Professor McFarlane's evidence was that science-based medicine supported his views. With respect to the NICE Guidelines, he noted that there was a potential for misunderstanding if particular conclusions were read out of context. The first critical statement was that at par 5.6.3 under the heading "Predictors of response to trauma-focused psychological treatment". The Guidelines read: "A number of studies have investigated whether response to trauma-focused CBT can be predicted from patient or trauma characteristics. Overall, few predictors have been found. Some clinical important findings from these analyses were: * The time that had passed since the trauma was not related to treatment effectiveness …." 91 The Guidelines recommended that trauma-focused psychological treatment should be offered to PTSD sufferers regardless of the time that had elapsed since the trauma; that the duration should normally be eight to 12 sessions, usually at least once a week: par 5.7. In considering early psychological interventions in randomised control trials, the Guidelines noted that there was "limited evidence suggesting a difference favouring trauma-focused CBT over waiting list (random effects) on reducing the severity of PTSD symptoms (clinician-rated measures) …": par 7.6.1. 92 As noted above, other findings appeared to support intervention. Thus, at par 7.7, the Guidelines noted: "When trauma-focused CBT is delivered between 1 month and 6 months after the incident, there is evidence suggesting that it is effective for people at risk of developing chronic PTSD, compared with the effect of being on a waiting list, for PTSD diagnosis post-treatment and at 9-13 months' follow-up, as well as a number of other outcomes assessed post-treatment, which included self-report measures of PTSD severity, anxiety and quality of life and clinician assessed PTSD severity. However, the evidence is inconclusive for a number of outcomes assessed at 9-13 months' follow-up (self-report measures of PTSD severity, anxiety and quality of life) and the evidence suggests that there is no clinically important difference for clinician assessed PTSD severity at 9-13 months." 93 At par 7.10, the Guidelines set out an economic evaluation of early versus late delivery of psychological treatment. A summary at 7.10.4 stated: "Assuming a remission is worth more than £2420, then cognitive-behavioural therapy at 12 weeks is the most cost-effective option. Achieving faster recoveries by treating early, however, may provide intangible benefits to those who suffer severe initial PTSD symptoms, particularly by preventing the conditions from becoming chronic. Future early versus late intervention studies should include a waiting list control in order to reduce the uncertainty associated with similar treatments …." 94 In relation to the last material, Professor McFarlane commented (Tcpt, 30/10/06, p 104-5): "So what it's implying is that in fact if you provide earlier treatment in people who have got severe initial symptoms, you are in fact decreasing the risk of it becoming chronic, and I see it as somewhat at variance with some of the earlier matters raised." 95 Professor McFarlane also drew attention to par 7.11.2 in respect of early interventions for "acute PTSD". The Guidelines stated (at par 7.11.2.1): "Trauma-focused cognitive-behavioural therapy should be offered to those with severe post-traumatic symptoms or with severe PTSD in the first month after the traumatic event. These treatments should normally be provided on an individual out-patient basis." 96 As Professor McFarlane noted, that recommendation "actually even sort of goes inside the normal limit of one month's duration of symptoms to attract the diagnosis of PTSD … implying that if there is a concern that that has already emerged in the first month, that treatment should be instigated": Tcpt, p 106. The Guidelines also stated that such early intervention might reduce the number of sessions from eight to 12 to about five: par 7.11.2.3. 97 In further support of a science-based approach, Professor McFarlane provided a paper which he said addressed "the issue of sensitisation and the increasing reactivity of neurobiological systems": Tcpt, p 111. That paper, by Davidson JRT et al, "Posttraumatic Stress Disorder: Acquisition, Recognition, Course, and Treatment" J Neuropsychiatry Clin Neurosci 16:2; 135-147 (2004), dealt with a prospective study suggesting that the "response of those patients who go on to develop PTSD becomes abnormal between 1 and 4 months after the trauma, suggesting that this is the critical period during which the central nervous system adapts its response to ambiguous stimuli (such as loud noises) and determines whether PTSD develops": p 137. 98 Professor McFarlane did refer to the importance of "consensus statements" available in the scientific literature, and it may be that reference to which his Honour referred in [119]. That is, however, something different from individual clinical experience, upon which Dr Wright appeared to based his views. Findings 99 On the basis of this material, it is necessary to make findings as to the likelihood of a better outcome and the extent to which that outcome might have involved a reduction in the severity and duration of the symptoms. It is also necessary to consider how those conclusions would affect the assessment of damages. 100 As explained by Dr Wright, the first question to answer is the likelihood that the plaintiff would not have been deployed to Nimmitabel had he been appropriately debriefed and monitored after the incident at Coolabah. The answer to that question is not entirely one for medical opinion: it required consideration of whether the plaintiff would have revealed sufficient of his symptoms to allow a police psychologist to have decided that he should be placed on restricted duties or stood down. That matter should be assessed on the balance of probabilities. As indicated in the first judgment, the trial judge was not satisfied that the plaintiff would have allowed that to happen. There is no reason to interfere with that finding. It was based upon evidence that the plaintiff was anxious to achieve promotion, that the culture in the force was not conducive to doing otherwise than "toughing it out" and the plaintiff's hopes to continue in his position for the forthcoming Sydney Olympic Games. 101 The medical evidence supports the view that the provision of trauma-focused CBT early in 1999 had the potential to reduce the intensity and duration of the plaintiff's condition. However, whether therapy would have commenced within two weeks of the critical incident is quite uncertain. That step would have required a decision to treat for PTSD within one month, at a time when the clinical symptoms could not be satisfied. There was little evidence supporting the possibility of such a step being taken in early January 1999. A greater possibility was that, given appropriate debriefing and monitoring, which did not themselves constitute effective treatment, CBT would have been more likely to commence in late January 1999, some six weeks after the incident. If that had occurred, there would, in effect, have been an improvement of four months in the commencement date. 102 There was some material supporting the view that chronic PTSD may have an identifiable neurobiological explanation, suggesting not merely pre-incident susceptibility, but changes occurring between one and four months after the trauma: see reference above to Davidson et al. The fact that such evidence was not available to the profession in Australia in 1999 may mean that there would have been less urgency in commencing treatment, but it is, nevertheless, relevant to the present assessment of the likelihood of variation in result. 103 Weight should be given to the acceptance in the NICE Guidelines, based on the study by Gillespie and others (concerning treatment of patients affected by a car bomb which exploded in the centre of Omagh, Northern Ireland, in August 1998) that the effectiveness of treatment was not dependent upon time that had passed since the trauma. Further, the finding that trauma-focused CBT delivered between one and six months after the incident was effective for people at risk of developing chronic PTSD, a group which must include the plaintiff, gave little guidance as to the effectiveness of treatment commencing one month after the incident, as compared with treatment commencing five or six months after the incident. Nor can any clear distinction be drawn between the cost effectiveness of treatment "at 12 weeks" and treatment commencing at five months. 104 Accepting that treatment has, to an extent, been valuable to the plaintiff, the evidence provides no basis for concluding that there was any real possibility that earlier commencement of the treatment would have permitted him to continue in the police force. There was some evidence that a continuation of his condition was in part due to his refusal to continue with prescribed medication (Zoloft). There is no support for a view that earlier treatment would have affected his attitude in respect of such medication. 105 Any attempt to assess a better outcome should be undertaken by reference to the consequences for both general damages and impact on earning capacity. The basis for any such assessment was hardly touched on in submissions and there is little assistance to be gained from the material before the Court. 106 Despite receiving treatment from late May 1999, the plaintiff left the police force in November of that year and was unemployed until February 2002 when he commenced work with Qantas as a baggage handler. Loss of earning capacity was agreed between the parties and no finding was made by his Honour in his first judgment, except to assess general damages at $90,000: see agreed schedule of damages handed up on 22/10/04 - Tcpt, p 1. That amount was not attributed to any particular period and it is not known what proportion, if any, might have been attributable to the period from January to May 1999, when the plaintiff was without the benefit of medical assistance. In the circumstances, the best that this Court can do is to work with the global figures and assume that any improvement in the severity or duration of the PTSD would be directly reflected in the damages as assessed or agreed. There is no basis on which to take any different course. 107 Difficulties of assessment of loss arose in LJP Investments Pty Ltd v Howard Chia Investments Pty Ltd (1989) 24 NSWLR 499. The defendant, in obtaining a significant financial benefit by building to the boundary of its land, was required to erect its scaffolding in a manner which encroached on the airspace of its neighbour. The expense which the defendant would have incurred in seeking to achieve the result by alternative means was less than half the sum demanded by the plaintiff for the period when the scaffolding was in place: at 507-508. Hodgson J stated at 508B: "It is impossible to be certain which course Chia would have taken, if it had refrained from trespassing: Chia's own unlawful act in trespassing, plus the absence of any of Chia's decision-makers from the witness box, have made such a decision highly problematic. I think in these circumstances, the Court is justified in taking the course taken in Armory v Delamirie (1722) 1 Stra 505; 93 ER 664, and resolving the question of value against the party whose action have made an accurate determination so problematic." 108 That approach was approved by this Court in Houghton v Immer (No. 155) Pty Ltd (1997) 44 NSWLR 46 at 59, Handley JA (Mason P and Beazley JA agreeing) stating that the Court should "assess the compensation in a robust manner, relying on the presumption against wrongdoers, the onus of proof, and resolving doubtful questions against the party 'whose actions have made an accurate determination so problematic'." The appropriateness of drawing inferences in favour of a plaintiff in such circumstances was approved in Murphy v Overton Investments (above) at [74]. 109 Because of the uncertain aetiology of the condition, combined with the absence of any clear medical evidence as to the effects of the four month delay in providing psychological treatment, the chance of some better outcome is highly speculative. Because Mr Cocco's intervention appears to have been of some assistance, it may be assumed that earlier intervention would have provided alleviation of the symptoms for a longer period. However, there is little chance that a chronic condition would have been avoided. 110 While any degree of precision must be misleading taken out of context, it is necessary to place figures on these factors. Accordingly, I would assess the long-term and short-term effects of providing early psychological treatment in January 1999 as possibly diminishing the severity of the condition by a figure somewhere between 20% and 60%. The evidence only allows the "better outcome" to be realistically identified as a range. The chance of a short-term improvement only, providing the bottom of the range, may be as high as 80%, but the chance of any long-term improvement (at the top end of the range) might realistically be as low as 20%. These figures would give a value to the loss of a chance as between 12% and 16%. However, there is no reason to suppose that the range is linear or that it can realistically be defined with such precision. Although the onus is on the plaintiff to identify the value of the lost chance, in accordance with the authorities referred to above, it is reasonable to err on the side of generosity, to avoid the negligent defendant obtaining too great a benefit of the doubt from uncertainty as to the causes and treatment of psychiatric conditions. Taking these factors into account, I would assess the value of the lost chance at 20%. 111 Accordingly, the judgment of the District Court should be set aside and in lieu thereof judgment should be entered in the sum of $80,000, such judgment to have effect from 7 June 2007. Costs of appeal 112 The primary contention of the State was that the plaintiff had failed to establish anything greater than a minimal amount by way of damages. The State has not been entirely successful in that contention, but has had substantial success. Subject to further argument it would seem appropriate that the respondent should pay 75% of the Appellant's costs of the appeal. Costs below 113 At the first trial, the State was ordered to pay the plaintiff's costs from the date of an offer of compromise (26 September 2002) on a solicitor/client basis. That order was set aside by this Court on the first appeal. When the matter came back before his Honour following judgment on the remittal hearing, his Honour repeated the order that the defendant pay the plaintiff's costs of the first trial, on a solicitor/client basis from 26 September 2002 and pay the costs of the remitted matter on a party/party basis. 114 Both of those orders were challenged in the notice of appeal, the State seeking an order that the plaintiff pay its costs "caused by or related to the plaintiff's failure at the first trial to adduce evidence sufficient to establish his claim for damages". Although no specific order was sought in respect of the costs of the remittal hearing, the order made by the trial judge was sought to be set aside and judgment was sought in favour of the defendant. It would have followed that the plaintiff would have been required to pay the defendant's costs of the remittal hearing. 115 In its written submissions, the State sought an opportunity to file further written submissions as to the appropriate costs order, in the event that this Court substituted a lesser sum for that awarded by the trial judge. The State did make submissions with respect to its challenge to the order as to the costs of the first trial, in the event that it was unsuccessful in interfering with the findings as to damages made on remitter. It sought costs in respect of the first trial on the basis that the plaintiff had failed to establish his case as originally presented. He had neither pleaded a case based on a loss of a chance, nor called appropriate evidence in support of such a case. Accordingly, it was submitted, he would have failed on the question of liability, because he failed to establish, on the balance of probabilities, that he had suffered an injury as a result of the negligence of the State. 116 There is substance in the submission that the trial judge did not deal with the arguments presented by the State in this respect, in simply remaking the original order. However, it is not possible for this Court to make any order as to the costs of the first trial as it does not know the details of the offer of compromise on which the order for costs was based. Accordingly, it is necessary to permit the parties an opportunity to put on further submissions with respect to the costs of the respective hearings in the District Court. 117 In addition, the judgment below in respect of costs, following the remittal hearing, indicated that a further offer had been made following the first appeal. As that material may affect the appropriate order to be made in respect of the costs of this appeal, the parties should also have leave to address the appropriate order as to costs in this Court. Conclusions 118 I would propose the following orders:

(1) Allow the appeal and set aside the orders made in the District Court on 7 June and 15 June 2007.

(2) In lieu thereof give judgment for the plaintiff in the sum of $80,000, such judgment to have effect on and from 7 June 2007.

(3) Give leave to the parties to file written submissions in respect of appropriate costs orders, if such orders cannot be agreed to, within the following timetable: (a) Appellant to file and serve submissions on or before 12 December 2008; (b) Respondent to file and serve submissions on or before 19 December 2008, such submissions to address the appropriate costs orders to be made in respect of the first trial, the remittal hearing and this appeal. 119 HANDLEY AJA: I agree with Basten JA. **********

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.