South Eastern Sydney Area Health Service & Anor v King [2006] NSWCA 2

New South Wales Court of Appeal

CITATION: South Eastern Sydney Area Health Service & Anor v King [2006] NSWCA 2

HEARING DATE(S): 16-18 November 2005

JUDGMENT DATE: 1 March 2006

JUDGMENT OF: Mason P at 1; McColl JA at 2; Hunt AJA at 3

DECISION: 1. The defendants' appeal against the judgment entered in favour of the plaintiff against the first defendant is dismissed. ; 2. The judgment entered in favour of the plaintiff against the second defendant for $7,000,000 is confirmed. ; 3. The plaintiff's cross-appeal is allowed. ; 4. The defendants are to pay the plaintiff's costs of the litigation in the Common Law Division up to and including 31 January 2005 on a party and party basis, and her costs thereafter on an indemnity basis. ; 5. The appellants are to pay the respondent's costs of the appeal and of the cross-appeal.

CATCHWORDS: Medical negligence case against hospital and treating specialist for whose negligence the hospital is vicariously liable — plaintiff (13 years old) being treated for cancer by way of radiotherapy, systemic chemotherapy and intrathecal chemotherapy — treatment prescribed by treating specialist radical, experimental and controversial procedure, rarely used in the hospital — protocol on which proposed treatment based in part amended before treatment administered because of risk of neurological damage — system whereby one doctor in the hospital's oncological team disseminated information he received from overseas concerning children's cancer — failure of team doctor to pass on to treating specialist information in his possession demonstrating that protocol had been amended — plaintiff becomes quadriplegic as result of treatment — trial judge finds team doctor negligent and hospital vicariously responsible, but finds treating specialist not negligent - Duty of treating specialist in particular circumstances of case to inform himself of the up-to-date information relating to proposed treatment — failure to ensure that such information obtained — treating specialist found negligent on appeal and hospital vicariously responsible - Refusal of trial judge to order indemnity costs after Calderbank letter sent by plaintiff offering to settle for smaller verdict — judge makes no reference to fact that defendants knew of evidence on which plaintiff's case depended at time of the letter — defendants' rejection of evidence unreasonable — indemnity costs ordered.

LEGISLATION CITED: Evidence Act 1995

Bolitho v City and Hackney Health Authority [1998] AC 232 Brymount Pty Ltd v Cummins [No 2] [2005] NSWCA 69 Calderbank v Calderbank [1976] Fam 93 Chappel v Hart (1998) 195 CLR 232 Flemington Properties Pty Ltd v Raine & Horne Commercial Pty Ltd, Federal Court (Lehane J), 2 February 1998 Forbes v Selleys Pty Ltd [2004] NSWCA 149 Fowdl v Fowdl, Court of Appeal, 4 November 1993, unreported Heydon v NRMA Ltd (2000) 51 NSWLR 1 Hillier v Sheather (1995) 36 NSWLR 414 CASES CITED: Jones v Bradley [No 2] [2003] NSWCA 258 Leichhardt Municipal Council v Green [2004] NSWCA 341 Maitland Hospital v Fisher [No 2] (1992) 27 NSWLR 721 March v Stramare (E & MH) Pty Ltd (1991) 171 CLR 506 Morgan v Johnson (1998) 44 NSWLR 578 Rogers v Whitaker (1992) 175 CLR 479 Rosenberg v Percival (2001) 205 CLR 434 SMEC Testing Services Pty Ltd v Campbelltown City Council [2000] NSWCA 323 TC v New South Wales [2001] NSWCA 380 Turner v State of South Australia (1982) 42 ALR 669 Woods v Multi-Sport Holdings Pty Ltd (2002) 208 CLR 460

South Eastern Sydney Area Health Service - First Appellant/Cross-Respondent PARTIES: Richard O'Gorman Hughes - Second Appellant/Cross-Respondent Monique Frances King - Respondent/Cross-Appellant

FILE NUMBER(S): CA 40358/05

COUNSEL: Appellants — PR Garling SC & SD Kalfas SC Respondent — DJ Higgs SC & DE Graham

SOLICITORS: Appellants — TressCox Respondent — Turner Freeman

LOWER COURT JURISDICTION: Supreme Court - Common Law Division

LOWER COURT FILE NUMBER(S): 18475/93

LOWER COURT JUDICIAL OFFICER: Newman AJ

- 38 - - 1 -

1 MASON P: I agree with Hunt AJA. 2 McCOLL JA: I agree with Hunt AJA. 3 HUNT AJA: Monique Frances King, the respondent to this appeal, was the plaintiff in proceedings in the Common Law Division of the Supreme Court. She sued the appellants, the South Eastern Sydney Area Health Service (which operated the Prince of Wales Children's Hospital) and the executrix of the estate of the late Professor Darcy O'Gorman-Hughes (a specialist paediatric haematologist and oncologist) practising in that hospital. Her claim was for damages for personal injuries. She alleged that, as a result of the treatment for cancer she underwent at the hospital in 1989, she became a quadriplegic because the treatment had been negligently performed by both the hospital and Professor O'Gorman-Hughes. She also alleged that Professor O'Gorman-Hughes had failed to warn her parents fully about the dangers of paraplegia or quadriplegia in the treatment plan he proposed. 4 The respondent was then 13 years old. She is now 29 years old. The presence of a tumour partly in the spinal canal and partly in the para-spinal muscles at the C7 level of her spine was confirmed during a surgical exploration carried out by Dr Warwick Stening (a neurosurgeon), who was treating the respondent for weakness and pain in her right arm. The tumour surrounded the right C 7 nerve root and extended to just inside the dura (the outermost fibrous membrane) covering the spinal cord. An immediate histological examination revealed the tumour to be a highly malignant one, and Dr Stening removed much but not all of it, the remaining one-third being inaccessible because the laminectomy operation he was performing at the time was from the posterior position. The respondent then came under the care of Professor O'Gorman-Hughes for the treatment of her cancer. 5 Professor O'Gorman-Hughes prescribed as treatment for the removal of the tumour a combination of: (1) radiotherapy, (2) systemic chemotherapy, which affects the body as a whole, and (3) intrathecal chemotherapy, which is administered by injection directly to the meninges (or membranes) of the spinal canal. The systemic chemotherapy included the administration of three agents: vincristine ("VCR"), cyclophosphamide ("CPA") and actinomycin D ("Act-D"). The intrathecal chemotherapy included the administration of three more agents: methotrexate ("MTX"), cytosine arabinoside ("Ara-C") and hydrocortisone. This procedure has been referred to generally as triple intrathecal therapy, and its acronym "TIT" has been used to refer to intrathecal chemotherapy generally (whether or not three agents are being administered). 6 Following the administration of this treatment plan, the respondent exhibited symptoms of myelopathy (damage to her spinal cord), which ultimately led to her quadriplegia emanating from the C6-C7 level.

The issues at the trial 7 The respondent's case was that: (1) Professor O'Gorman-Hughes and the other paediatric oncologists at the hospital failed to fulfil their professional obligation to keep themselves informed in relation to the management of the type of children's cancer from which she suffered. (2) No (or, alternatively, less) Act-D should have been administered by way of systemic chemotherapy, or alternatively it should have been administered at a later time than it was, because its administration enhanced the effect of the radiotherapy which had been given. (3) Less TIT should have been administered, or alternatively it should have been administered at a lesser frequency than it was. (The claim at the trial that no TIT should have been administered was not pursued on appeal.) (4) Less Ara-C should have been administered by way of TIT concurrently with the radiation. (5) The appellants failed to warn the respondent's parents fully about the risk that TIT could result in damage to her spinal cord, to enable them to obtain a second opinion or to seek further information about the proposed course of treatment. Part of the second element, and all of the third and fourth elements of the respondent's case were inextricably bound up with the first element. The respondent's case did not challenge the radiotherapy administered. 8 The appellants do not dispute that the respondent's quadriplegia resulted from the treatment she received. They nominated the cause as being the combination of radiotherapy, the administration of Act-D as part of the systemic chemotherapy, and TIT, but the appellants do dispute that any breach of duty on their part relating to their treatment of the respondent was causative of any damage. There was no suggestion that the respondent's quadriplegia resulted from the tumour. Nor was there any suggestion that the operative procedures performed by Dr Stening were responsible for that condition. 9 The appellants contended that: (1) In the light of medical knowledge as it existed in 1989, the treatment afforded to the respondent was reasonable, in that she was suffering from a life-threatening tumour which required "radical" treatment to prevent it from metastasising (spreading the cancer through the blood vessels) and in order to eradicate the tumour. (2) One of the risks of radical treatment is the chance of serious side effects such as those suffered by the respondent, of which the respondent's parents had been warned. (3) Quadriplegia was, by 1989 standards, an acceptable risk in the administration of such treatment. A contention by the appellants that, had the radical treatment not been carried out, the respondent may well have perished, was not pressed on appeal.

One basis of the treatment prescribed 10 Before stating the findings made by the trial judge (Newman AJ), it is necessary to describe a little more fully the basis on which Professor O'Gorman-Hughes, after consultation with other paediatric oncologists at the Prince of Wales Hospital, prescribed the treatment plan identified in par [5] supra. Professor O'Gorman-Hughes took into account many published writings, and his own substantial experience, but the significant issue in this case was that his treatment plan was based in part on a protocol known as IRS-II (to which further reference is made later). The approach of the trial judge, which is consistent with the evidence, was that this was a substantial basis for the treatment plan. In particular, Professor O'Gorman-Hughes followed the recommendations in the IRS-II protocol in determining to use TIT as part of the treatment plan and he took those recommendations into account in determining the size and number of doses of MTX, Ara-C and hydrocortisone to be administered by way of TIT and of Act-D to be administered by way of systemic chemotherapy. 11 The respondent's actual treatment was as follows: 3-23 March 1989 Radiotherapy over 21 days

3 March Intravenous VCR (1.8 mg) and CPA (700 mg)

6 March TIT MTX (12 mg) and hydrocortisone (50 mg)

10 March Intravenous CPA (700 mg)

14 March TIT MTX (12 mg), Ara-C* (70 mg) and hydrocortisone (50 mg)

16 March Intravenous VCR (1.8 mg) and CPA (700 mg)

23 March Intravenous CPA (700 mg), TIT MTX (12 mg), Ara-C (70 mg) and hydrocortisone (50 mg)

30 March–14 April Radiotherapy over 16 days

30 March Intravenous VCR (1.8 mg) and CPA (700 mg)

7 April Intravenous CPA (700 mg), TIT MTX (12 mg), Ara-C (70 mg) and hydrocortisone (50 mg)

14 April Radiotherapy (boost) on one day

26 April Intravenous VCR (1.8 mg) and oral CPA (100 mg) daily

3 May Intravenous VCR (1.5 mg) and oral CPA (100 mg) daily

10 May Intravenous VCR (1.5 mg) and oral CPA (100 mg) daily

24 May Intravenous VCR (1.5 mg) and oral CPA (100 mg) daily

31 May TIT MTX (12 mg), Ara-C (70 mg) and hydrocortisone (50 mg)

14 June Oral CPA (100 mg) daily

28 June Oral CPA (100 mg) daily until 11 July

17 July Intravenous VCR (1.5 mg) and Act-D (2800 g) over five days (600 g for four days and 400 g for one day)

9 August Intravenous VCR (1.5 mg) and oral CPA (100 mg) daily

16 August Intravenous VCR (1.5 mg)

* Ara-C is called cytarabine in the medical records. 12 The onset of the respondent's myelopathy is noted as having occurred on 21 July 1989. 13 The IRS-II protocol was published by a body called Intergroup Rhabdomyosarcoma Study Group ("IRS Group"). (A rhabdomyosarcoma is a highly malignant tumour affecting children which occurs frequently in the head and the neck.) The IRS Group, which is based principally in the United States of America, conducted a series of studies of different types of treatment for malignant tumours in the head. The protocol stipulated the dosages of chemotherapy agents to be administered and their frequency, and the amount of radiotherapy. 14 The specific study using the protocol employed by Professor O'Gorman-Hughes was carried out between 1978 and 1984, although the final report of the results of that study was not published until 1993. Professor O'Gorman-Hughes came to know of the existence of the protocol in 1987, from an article by Dr RB Raney and others published in a medical journal, "Cancer", which suggested that patients had a substantially improved survival rate when treated with an intensive therapy which included TIT. He took the view that, although the protocol was directed to tumours in the skull, it was also of value in treating tumours in the spinal area, with some modification. There has been no criticism of the view he took. 15 TIT had been included in the protocol as a specific measure to avoid the spread of the cancer. This is described as "prophylactic" TIT. (TIT could also be used for "salvage" purposes, when everything possible is used to achieve survival for desperate cases.) Prophylactic TIT was an experimental and controversial form of treatment. The appellants concede that it was radical (see [par [9] supra). According to the evidence, it had rarely been used previously for that purpose at the Prince of Wales Hospital. There was a difference of views between the IRS Group in the United States and study groups in Europe as to the use of TIT in the treatment of tumours. TIT was not used as a routine treatment in Europe. A Dutch group in 1985 warned that TIT was accompanied by a considerable risk of complications in the central nervous system (including paraplegia) because of the effects of its combination with radiation and systemic chemotherapy. 16 A new protocol, known as IRS–III, was published by the IRS Group and put into practice between 1984 and 1991 with a different group of patients. At first, the new protocol did not differ from IRS–II so far as is relevant here, but on 5 June 1987 an important amendment to the protocol was published. This document disclosed that neurological complications had been suffered by one patient being treated in accordance with IRS–II and by six patients being treated in accordance with IRS–III. All had received radiotherapy and TIT. Five of the seven patients suffered lower extremity weakness and bladder/bowel dysfunction. Two of those seven patients had already suffered spinal cord decompression, and were therefore neurologically impaired at the onset of treatment. The IRS Group stated that it was not apparent what factor, or factors, had caused these complications, but it continued: However, it is felt that the doses of prophylactic (neg CSF cytology) intrathecal chemotherapy [negative cerebrospinal fluid pathology TIT] should be cut back to those now used for CNS leukemia [central nervous system leukaemia] and that intrathecal therapy should be limited to 4 doses in 20 weeks, after which no further intrathecal chemotherapy is to be given. Patients already on CNS prophylaxis beyond 20 weeks should discontinue intrathecal therapy. Patients with positive CSF cytology or spinal cord involvement should be treated with the amended doses at the frequency called for in the protocol (Sect 8.72). In addition the "maximum" tolerance dose of radiation to the spinal cord should be no more than 4140-4200 cGy. The document then identified the amended (reduced) doses to be administered by way of TIT for the chemotherapeutical agents MTX, Ara-C and hydrocortisone. The dosages were also limited by reference to the patient's age rather than body surface area as previously. 17 This information was of vital importance in the treatment of the respondent. The treatment she received by way of TIT administration of MTX and Ara-C differed slightly from what had been recommended by IRS-II, but the respondent's case did not rely on those differences. Her treatment differed substantially from that advised by the 1987 amendment to IRS-III, and these differences formed the substance of the respondent's case. The differences were in relation to — (i) the frequency of the TIT, and its timing concurrently with the radiotherapy, as Ara-C enhanced the effect of the irradiation on the respondent's spinal cord (which was ultimately found by the judge to have materially contributed to, and thus caused, the respondent's quadriplegia), and (ii) the strength of the dosages of Ara-C administered. 18 The concurrency of the radiotherapy and the TIT differences in the frequency were as follows (the first week is nominated as week 0): TREATMENT Weeks nominated Radiotherapy — 0 1 2

TIT administered to respondent — 0 1 2 5 8*

TIT recommended by IRS-II — 0 2 4 8 14

TIT advised by amended IRS-III — 0 6 12 20

* This is an approximation. The dosages of Ara-C administered were each of 70 mg, which was less than recommended by IRS -II , but considerably more than the 30 mg advised by the amendment to IRS -III . 19 Professor O'Gorman-Hughes died before the trial (in 2000), but there are a number of documents in evidence, including a statement he made, his answers to interrogatories and notes (called factual assumptions) prepared by his legal representatives for the purposes of the litigation, apparently with the assistance of Professor O'Gorman-Hughes. It is clear from these documents that he claimed to have been unaware of the relevant amendment made to the IRS protocol. He had no direct association with the IRS Group. However, one of the other specialist paediatric oncologists on the staff of the Prince of Wales Hospital, Dr Leslie White, was a "corresponding" member of the Children's Cancer Group in the United States, which was itself a member of the IRS Group. The only reference to Dr White's membership of that body in the written material from Professor O'Gorman-Hughes was in an answer he gave to the respondent's interrogatories, which were tendered by the appellants pursuant to s 63 of the Evidence Act 1995 on the basis that Professor O'Gorman-Hughes had died before the trial. The relevant part of the answer was successfully objected to by the respondent on the basis that it was non-responsive to the question, despite the facts that the non-responsive part could easily have been excised and the description of Dr White's membership of that body was directly relevant to the respondent's case. 20 Dr White nevertheless gave evidence in the trial on behalf of the appellants, and his evidence did, in the end, establish the existence of a system within the hospital's paediatric oncology department whereby Dr White normally circulated material he received relating to such oncology as part of his "usual communication with the [paediatric oncology] team". There was initially some inexactitude in his evidence as to the nature of his membership of the Children's Cancer Group and as to how the information he obtained which could be of interest to his colleagues was to be disseminated. He conceded that his membership was "for the hospital", although he also stated that it was an arrangement he had made personally with persons he had known from his training at a children's hospital in the United States. He denied that, at the relevant time in 1989, he received documents (including protocols) as a matter of course from the Children's Cancer Group as a result of his membership. The principal entitlement of a corresponding member, he said, was to attend the Group's twice yearly meetings. The minutes of those meetings were supplied to him "over this period in the 80s, leading up to 1989". He said that he usually retained the minutes of those meetings as "a useful source of general information". He said that, although there was a restriction imposed by the US body on any local implementation of the protocols referred to in the minutes (particularly the experimental ones), there was no restriction on the sharing of the information he gained from the US material, which included commentaries on the protocols. His own specialty did not include the treatment of sarcomas (Professor O'Gorman-Hughes was identified as the expert in that field). He did, however, accept that he knew that Professor O'Gorman-Hughes had based his schema (or prescription) for sarcoma patients at least in part on what was known about IRS–II, and he agreed that the current studies being undertaken by the IRS Group were regarded as being studies of importance. 21 In response to a subpoena, Dr White produced documents he had received in relation to a meeting of the Paediatric Oncology Group in Orlando, Florida, in April 1988. That group is separate from the Children's Cancer Group (of which he was a member), although it too was involved with the IRS Group. Amongst the "Orlando" documents, there were minutes of the meeting which contained a reference to the IRS–III protocol as one of various treatment protocols current at that time, and which indicated clearly enough to any reader interested in the IRS-II protocol that it had been replaced by another protocol. The minutes themselves were not tendered in evidence, but Dr White agreed that the frequency of the TIT then being advised in the minutes was in weeks 0, 6, 12 and 20. This was without any doubt a reference to the advice given in the 1987 amendment to IRS-III (see par [18] supra). This frequency is, as already stated (pars [17]–[18] supra), substantially less than the treatment subsequently administered to the respondent, and the frequency administered was found by the judge to have in part caused the respondent's quadriplegia. If Professor O'Gorman-Hughes had been informed of this fact, the respondent argued, he would have followed up what other changes had been made by and to the IRS-III protocol. With knowledge of that amendment, the argument ran, the respondent's treatment would not have included the same frequency of TIT or the excessive doses of Ara-C which caused her myelopathy and consequent quadriplegia. None of the many experts called at the trial suggested that it would have been unreasonable for Professor O'Gorman-Hughes to have followed the 1987 amendment. 22 Dr White said that he had received the documents in 1988 (around May). He said that the documents had been sent to him as a personal favour by an American medical practitioner, a member of the Paediatric Oncology Group in Orlando, whom he had met on a visit to the United States. He had expressed an interest in an alliance between that Group and the Children's Cancer Group, and the minutes of the most recent meeting of the Orlando Group had been sent to him in order to give him a better understanding of that Group. 23 Dr White said that he came across this reference in the Orlando minutes only at some time during preparation for the trial of the respondent's claim, and he had brought it to the attention of the hospital's legal advisors. Dr White nevertheless said that he would "normally" have circulated the document when it was received as part of his "usual communication with the team". He said, however, that this was a "large" document (the relevant reference to IRS-III was on page 239), although (he added) it certainly would have been available to all members of the team. How this was so was not explored at the trial. In his judgment, the judge (at [27]) drew the inference from Dr White's evidence that the information in the Orlando minutes had not been circulated. I would draw the same inference, as well as the further inference that it had not been circulated because the document had been a large one and Dr White had not got around to reading it. Dr White agreed that, had he been asked by Professor O'Gorman-Hughes whether he had any documentation in relation to the IRS-II protocol, he would have searched through his documents and would have produced this document.

Judgment of Newman AJ 24 The judge held (at [35]) that a medical practitioner has a duty to exercise reasonable care and skill in the provision of professional advice and treatment, and that the standard expected of a medical practitioner is that of an ordinary skilled person exercising and professing to have that special skill: Rogers v Whitaker (1992) 175 CLR 479 at 483, 487. Professor O'Gorman-Hughes therefore owed the respondent in the present case a duty to exercise reasonable care and skill in the provision of paediatric oncology advice and treatment, and the standard expected of him was that of an ordinary skilled paediatric oncologist. The judge held (at [35]) that a similar duty was cast on the hospital. That ruling has not been challenged. The judge (at par [41]) necessarily implicitly identified an obligation to remain informed of any relevant changes occurring in the treatments being administered as within that duty. 25 In relation to the issue of breach of that duty of care, the judge — again following Rogers v Whitaker (at 489) — held (at [38]) that the factors to be considered will vary according to whether it is a case involving diagnosis, treatment or the provision of information or advice in relation to those matters. The respondent's submission, based on Woods v Multi-Sport Holdings Pty Ltd (2002) 208 CLR 460 at 482, was that, where it is possible to guard against a foreseeable risk (which, although perhaps not great, nevertheless cannot be called remote or fanciful) by adopting a means involving little difficulty or expense, the failure to adopt such means will in general be negligent. The judge accepted this submission (at [39]). See also Turner v State of South Australia (1982) 42 ALR 669 at 670–671. None of the principles stated in this and the preceding paragraph was in issue in this appeal. 26 An important issue resolved by the judge concerned the circumstances in which Professor O'Gorman-Hughes was unaware of the 1987 amendment to the IRS protocol. A reference to this amendment had been published before the respondent's treatment in a paper "Rhabdomyosarcoma in Children and Adolescents, a Review", written by Frederick B Ruymann, the Director of the Division of Haematology/Oncology in the Children's Hospital in the Ohio State University School of Medicine. He was a member of the IRS Group Committee at the time. The paper, which is said to have been written on behalf of the IRS Group, was included in a new medical journal, "Cancer in Children", dated December 1987. The paper stated that brain stem dysfunction in a small number of patients had prompted the amendment, which reduced TIT and avoided simultaneous TIT and radiotherapy. It identified in detail (in Tables 8 and 9) the reduced TIT as being administered in weeks 0, 6, 12 and 20 and then stopped, and the changes in the dosages for Ara-C already referred to (see par [18] supra). The paper made no reference to the particular neurotoxicity referred to in the 1987 amendment, of lower extremity weakness and bladder/bowel dysfunction resulting from treatment in accordance with the previous IRS protocol (see par [16] supra). 27 The judge described the medical journal in which this paper had been published as "obscure". He accepted the evidence of Dr Plowman, one of the appellants' expert medical witnesses, that the journal (at that time) had "an impact factor of virtually zero", and that the paper was not one Professor O'Gorman-Hughes would have been expected to have seen. There was no evidence of any other published reference to the 1987 amendment to the IRS protocol prior to the respondent's treatment, notwithstanding that other articles had been published on rhabdomyosarcoma involving the IRS Group. A more detailed article concerning the amendment was published in the "Cancer" journal in 1992, after the respondent had received her treatment. 28 The judge held (at [30]) that it was "entirely reasonable" for Professor O'Gorman-Hughes to expect that Dr White would have been in a position to advise him of any relevant protocol changes which had occurred since IRS-II, but that he had received no such advice from Dr White. This finding is a subject of the respondent's notice of contention. 29 Causation is established by a plaintiff if the defendant's negligence caused or materially contributed to his or her injuries: March v Stramare (E & MH) Pty Ltd (1991) 171 CLR 506 at 514, 524; Chappel v Hart (1998) 195 CLR 232 at [6]–[8]. If the defendant's negligent act or omission was so connected to the injuries that, as a matter of common sense and experience, it should be regarded as a cause of those injuries, the defendant is liable as having caused them: March v Stramare (E & MH) Pty Ltd (at 522). Once a plaintiff demonstrates that a breach of duty has occurred which is closely followed by damage, a prima facie causal connection will have been established. It is then for the defendant to show, by evidence and argument, that the plaintiff should not recover damages: Chappel v Hart, per Kirby J at [273]; see also Gummow J at [257]. Kirby J identified this as a shift in the evidentiary (but not legal) onus. In a detailed examination of the onus of proof in causation, in TC v New South Wales [2001] NSWCA 380 at [53]–[74], [83]–[101], Mason P has concluded, in a judgment with which Priestley & Beazley JJA agreed, that the law in Australia was correctly stated by Kirby J, and that a robust and pragmatic approach to proof permits (but does not compel) the tribunal of fact to proceed from breach and specific harm to causation; see also Forbes v Selleys Pty Ltd [2004] NSWCA 149 at [94]–[101]. 30 The judge held (at [33], [41]) that, had Professor O'Gorman-Hughes been aware of the amendment to the IRS protocol, he would have sought advice from members of the IRS Group as to the problems they had encountered by way of highly undesirable side effects from the administration of the previous regime, particularly as it related to the prescription of dosages of chemotherapeutical agents. Once he became aware of the amendment to the IRS protocol, the judge found (at [34]), Professor O'Gorman-Hughes would have modified the regime he prescribed for his patient. Applying the commonsense approach to causation, the judge held (at [44]–[49]) that the administration of Ara-C by way of TIT, especially the three doses given concurrently with the radiotherapy, enhanced the effect of the irradiation on the respondent's spinal cord. The doses of Ara-C in excess of those recommended by the amendment to the IRS protocol had materially contributed to, and thus caused, the respondent's myelopathy and consequent quadriplegia. 31 The judge also held (at [46]–[50]) that the quantity of Act-D administered to the respondent by the systemic chemotherapy exceeded both the amount stipulated in the IRS protocol and that actually prescribed by Professor O'Gorman-Hughes (according to Professor O'Gorman-Hughes, the dose he prescribed was 90% of the dose administered), and that this chemotherapeutical agent, having been administered after the radiotherapy, had contributed as well to the respondent's myelopathy by inducing a "recall" reaction to the radiotherapy. The appellants concede that Act-D had a known capacity to produce a radiation recall. 32 The judge held (at [41]) that Professor O'Gorman-Hughes's ignorance of the amendment to the IRS protocol was not the result of negligence on his part. He implicitly but necessarily accepted that Professor O'Gorman-Hughes had relied on the system for the dissemination by Dr White within the paediatric oncology department of the hospital of information relevant to the discharge of his own duty to keep himself informed of any relevant changes which had occurred to IRS-II. He concluded that Professor O'Gorman-Hughes had utilised a degree of diligence which would "guarantee the prescription of treatment in accordance with the most current and considered guidelines available in the world". That conclusion is also a subject of the respondent's notice of contention. 33 The judge held (at [41]) that Dr White (for whose acts and omissions the hospital was vicariously responsible) had either failed to make himself aware of such information in the relevant Orlando document that the IRS protocol had been amended or he had failed to establish a system for the proper dissemination of such information throughout the department. That breach of his duty of care by Dr White had materially contributed to the respondent's quadriplegia. The respondent's case against the hospital based on that breach of duty was thus successful. 34 The judge accepted that Professor O'Gorman-Hughes did not warn the respondent's parents fully about the dangers of paraplegia or quadriplegia in the proposed treatment, but he held (at [53]) that, even if they had been advised of the risk of paraplegia or quadriplegia, they would have given permission for the treatment to have been carried out. He therefore rejected the remaining part of the respondent's case against Professor O'Gorman-Hughes as well. That finding is made a subject of the respondent's amended notice of contention, which this Court ordered, without objection, to stand as the notice of contention. 35 The amount of damages had been agreed between the parties early in the trial at $7,000,000. Judgment in this amount was entered against both appellants, notwithstanding the failure of the respondent's case against Professor O'Gorman-Hughes. This was done pursuant to an agreement between the different insurers for the two appellants that a verdict against either of them would be a verdict against both appellants for the same amount, the liability for which they would then share pursuant to an undisclosed agreement between them. This explains the presence of the executrix of Professor O'Gorman-Hughes's estate as an appellant with the hospital. At all relevant times the two appellants have been jointly represented.

The appeal 36 The principal attack by the appellants on the judge's finding against Dr White (and thus the hospital) was that the information in the Orlando minutes on which the respondent's case depended was in an unsolicited document provided to him on a personal basis only, and for the general purpose of providing him with a better understanding of another oncology group in which he had expressed an interest. The appellants argued that there was no basis in the evidence for Professor O'Gorman-Hughes to have reasonably expected Dr White to give him any relevant information from that source. The appellants point to the absence of any suggestion of such an expectation in the written material from Professor O'Gorman-Hughes in evidence. They submit that Dr White owed no duty of care to make himself familiar with all of the material in his possession or to impart that information to his colleagues, including Professor O'Gorman-Hughes. 37 The interest Dr White had in the Paediatric Oncology Group in Orlando cannot fairly be described as a merely personal one. It was intimately bound up with the work Dr White was doing at the hospital, and it was directly relevant to the interchange of information which took place in the course of Dr White's employment in the hospital. The appellants' attack ignores the evidence from Dr White of the system whereby he normally circulated material he received relating to paediatric oncology as part of his "usual communication" with the team within the hospital's paediatric oncology department. That the members of that department operated as a team and to an extent collaborated with each other is clear from the evidence. A group which Professor O'Gorman-Hughes identified as the Children's Tumour Clinic Group operated in the Prince of Wales Hospital and its work is outlined in the material supplied by him and admitted into evidence. There were regular multidisciplinary sessions which included paediatric oncologists, paediatric surgeons and paediatric pathologists. They discussed the original formulation of the protocols, and they frequently and regularly reviewed the available information and literature on tumours such as rhabdomyosarcomas and soft tissue sarcomas. Regular weekly meetings of the hospital's paediatric oncology group also discussed the respondent's case concerning the formulation of her treatment. The plans for that treatment were outlined at those Friday morning meetings, and the members of the group agreed with the course of action proposed. 38 Dr White was a paediatric oncologist in the paediatric oncology group. The clear inferences from all of the evidence are that he took part in at least the regular weekly meetings of that group, and that he knew that his colleagues in that group needed all of the relevant information he had gleaned from his overseas connections for the treatment of their patients, including Professor O'Gorman-Hughes for the treatment of the respondent. Dr White knew that Professor O'Gorman-Hughes's protocol was based in part on IRS-II, and that studies of the IRS Group were regarded as being studies of importance (see par [20] supra). When Dr White did finally read the Orlando minutes, he realised the information they contained was significant to Professor O'Gorman-Hughes's treatment of the respondent, as he drew this information to the attention of the appellants' legal advisers, and he agreed that, if he had been asked by Professor O'Gorman-Hughes whether he had any information in relation to the IRS-II protocol, he would have searched through his documents and would have produced these particular minutes; he also said that he would normally have circulated them when the document was received (see par [23] supra). 39 It is true that there is no direct reference in the written material from Professor O'Gorman-Hughes to his reliance on the system described by Dr White in his evidence. This was one of the many disadvantages suffered by the parties on both sides of the record as a result of Professor O'Gorman-Hughes's death before the trial. According to the appellants' written submissions dated 5 December 2005, the fact that Dr White was a corresponding member of the Children's Cancer Study group was not raised as an issue in the case until it was mentioned in a report from Professor Bleyer in December 2002, two years after Professor O'Gorman-Hughes died. This has not been disputed by the respondent. The respondent's interrogatories must therefore have been drafted without any knowledge of the system's existence, and none of the interrogatories administered necessarily required Professor O'Gorman-Hughes to give details of that system. The only reference to Dr White by name in the answers to those interrogatories was excluded (see par [19] supra). The other, more discursive, material from Professor O'Gorman-Hughes is fully consistent with the existence of such reliance (see par [37] supra). 40 In my opinion, all of this material justified the judge's acceptance of a duty of care on the part of Dr White in relation to the dissemination of the Orlando minutes to the members of the oncology department team. He did not find that Dr White had a duty actively to seek out information from any particular overseas body; his implicit finding was merely that Dr White had a duty to pass on information he had received from such bodies. I am not persuaded that the judge's conclusion that Dr White had been negligent was erroneous. I would reach the same conclusion as the judge did for myself. 41 The appellants next attack the judge's finding (at [33]) that, if Professor O'Gorman-Hughes had been made aware of the amendment to the IRS protocol, he would have sought advice from members of the IRS Group as to the problems they had encountered by way of highly undesirable side effects from the chemotherapy being administered under the protocol. The appellants submit that the judge should have accepted the evidence of their witness, Dr Kellie, that in Sydney and Melbourne it was very rare in oncological circles to seek advice from overseas in the initial stages of treatment but relatively common to seek assistance from overseas if unusual side effects are encountered in the course of the treatment. 42 The two situations are quite different. Where there has been no question of reliance on an overseas protocol involving a radical, experimental and controversial means of treatment, used rarely in the hospital, the practice described by Dr Kellie may well be the usual one. In this case, however, the judge was fully entitled to conclude that Professor O'Gorman-Hughes would have made inquiries once he knew that there had been an amendment to the IRS protocol on which he had in part relied in formulating his own protocol. 43 The appellants then submit that there is doubt that the amendment to IRS-III would have been made available to Professor O'Gorman-Hughes by the IRS Group. However, the respondent's case ultimately accepted by the judge did not depend solely on the lack of an opportunity for Professor O'Gorman-Hughes to make inquiries of the IRS Group. Dr White gave evidence that there was no restriction on sharing the information he had obtained from the United States (see par [20] supra). The evidence discloses that members of various oncology groups willingly gave assistance to non-member medical practitioners. In any event, Dr Kellie (who was called by the appellants) said that, if he had read the Ruymann paper prior to treating a patient he would have called the IRS Group, and that it was likely that they would have shared the 1987 amendment with Professor O'Gorman-Hughes. 44 The case accepted by the judge necessarily depended on his acceptance of the evidence of several medical practitioners (some of them called by the appellants) that the dosages of Ara-C administered to the respondent by way of TIT, and the frequency of the TIT, together with the enhancement of the chemotherapy on the effects of the radiotherapy, had caused the respondent's quadriplegia. The amendment, as already stated (par [17] supra), recommended substantial changes in relation to these matters in the earlier protocol, and the judge concluded (at [34]) that the treatment based in part on the earlier IRS protocol would not have been administered if Professor O'Gorman-Hughes had been made aware of the amendment. It was the fact of the amendment — with its advice to lower the dosages and frequency of the treatments to substantially below what had been recommended in the earlier protocol because of the possibility that neurological complications had been caused by following those recommendations — that would have put Professor O'Gorman-Hughes on inquiry, wherever the inquiries may have been directed. 45 The appellants, however, challenge the status of the 1987 amendment on the basis that it was only part of an ongoing clinical trial, it had not been analysed, peer-reviewed or published, and it was still experimental. It was submitted that the amendment cannot therefore represent a standard of care which had to be applied by Professor O'Gorman-Hughes. The protocol itself, the appellants said, had merely been an experiment to test a hypothesis prior to being peer-reviewed in 1987, and therefore had not been binding on practitioners, nor had it necessarily been demonstrative of the best standards of practice. (The appellants' description of the 1987 article as being a peer review is inconsistent with its own description as an "initial" study.) The appellants next said that the amendment to that protocol in 1987 was also only a clinical trial with no proven record, no more than an experimental change, so that it too was not binding on Professor O'Gorman-Hughes, nor was it necessarily demonstrative of the best standards of practice. This challenge is misdirected. The issue was not whether Professor O'Gorman-Hughes should have adopted the amendment to the IRS Protocol for the first time. The relevant issue in the case was whether — having already adopted an earlier version of the IRS protocol at a time when it too had not been analysed or peer-reviewed, and advice having now been given to reduce the dosages and the frequency of the treatments it had recommended — Professor O'Gorman-Hughes would have followed the advice given in the amendment. 46 Another of the appellants' witnesses, Dr Plowman, expressed the view (which the judge accepted) that, if Professor O'Gorman-Hughes had been aware of the 1987 amendment to the protocol on which he had relied, he would have acted unreasonably if he had not followed the recommendations made by the amendment. That evidence was given in the face of the absence of any firm conclusion in the amendment to the protocol that the treatment previously recommended had in fact caused the neurological complications disclosed in that amendment, but it was no doubt based on the fact that the IRS Group had recommended reducing both the quantity of the chemotherapeutical agents administered by way of TIT and its frequency because of the risks of neurological complications, despite the absence of any such firm conclusion. Dr Plowman also said that the respondent's high risk of dying without the treatment had to be weighed against the risk involved in the treatment itself, and he then considered what he would himself have decided to do. Unfortunately, it is difficult to interpret this evidence (which was in re-examination), as it is unclear whether he says that he would have continued with the original recommendations in the protocol or with the protocol as amended. Counsel for the appellants (whose witness Dr Plowman was) did not ask him to make this evidence any clearer, and appears to have been content to leave the waters muddied. However, read in the light of the view which Dr Plowman had expressed immediately before he gave that evidence, I would interpret his evidence so that it was consistent with that view rather than contradictory of it. 47 It was open to the judge to conclude that the treatment based in part on the earlier IRS protocol would not have been administered if Professor O'Gorman-Hughes had been made aware of the amendment. My own conclusion is that Professor O'Gorman-Hughes, after making inquiries of members of the IRS Group or others in relation to the amendment, would have adopted the advice given by the amendment, and the respondent would not have been rendered quadriplegic by the treatment administered to her. That, as I understand the judgment, is the effect of what the judge held. There was no error made by him in doing so. 48 The appellants also attack the judge's finding on causation on another basis, that he had wrongly interpreted the medical evidence relevant to this issue. Even if the evidence could be interpreted in the way submitted by the appellants (of which I am not persuaded), the existence of two different interpretations of that evidence does not require a rejection of the interpretation which the judge has accepted. I have read the evidence to which reference was made in the submission, and I do not consider that any error has been demonstrated in the judge's finding in favour of the respondent on the issue of causation. 49 On this issue of causation, there was some debate between the parties concerning the decision of the House of Lords in Bolitho v City and Hackney Health Authority [1998] AC 232. This debate was started by the respondent in written submissions produced at the end of the hearing of the appeal, to which the appellants replied in writing after the hearing. In that case, the House of Lords accepted (at 240) the following statement concerning causation: The plaintiff can discharge the burden of proof on causation by satisfying the court either that the relevant person would in fact have taken the requisite action (although she would not have been at fault if she had not) or that the proper discharge of the relevant person's duty towards the plaintiff required that she take that action. The respondent relied on both alternatives. The appellants submitted that the statement did not reflect the law of Australia, and that in any event the second alternative requires the issue of negligence to be considered in an artificial context. 50 In my opinion, this debate does not need to be resolved in the present case. The judge held that Dr White was negligent in failing either to make himself aware of the information in his possession relating to the amendment to the IRS protocol on which Professor O'Gorman-Hughes relied or to establish a system for the proper dissemination of such information throughout the oncology department at the hospital. The judge then held that (1) the dosages and frequency of the chemotherapeutical agents administered by Professor O'Gorman-Hughes at a higher level than those advised by the amendment to the IRS protocol had caused the respondent's myelopathy and consequent quadriplegia, and (2) if Professor O'Gorman-Hughes had known of the relevant amendment, he would have modified the regime he had prescribed for the respondent. 51 As a consequence of all these findings, the case thus falls within the first alternative accepted in Bolitho, which in my opinion is part of the law of causation in Australia: see Chappel v Hart (at [14], [113]). (Although Hayne J dissented in the result (at [121]), his statement of principle at [113] is uncontroversial.) For this reason, it is unnecessary to consider whether the second alternative accepted by the House of Lords in Bolitho is available in Australia. 52 Finally, the appellants challenge the judge's finding that the quantity of Act-D, which exceeded both the amount stipulated in the original IRC protocol and that actually prescribed by Professor O'Gorman-Hughes, contributed to (and thus caused) the respondent's myelopathy and consequent quadriplegia. They submit that the judge failed to have regard to the fact that it was Professor O'Gorman-Hughes himself who deliberately determined the amended final dosage after it had been discovered that the earlier dosages had exceeded what he had prescribed. They point to the evidence of Dr Kellie to the effect that the total Act-D administered was not excessive by comparison with a number of other international protocols. The judge was, however, entitled to rely on the expert opinion of Professor Bleyer (and similar views expressed by three other medical practitioners, including one called by the appellants), that the Act-D "administered after radiotherapy contributed to the myelopathy by inducing a 'recall' reaction to the radiotherapy". I am not persuaded that the judge erred in that finding. 53 The judge does, however, appear (with respect) to have made a minor error in detail in relation to this matter, to which the parties have not referred, but it is as well to record it. In making the finding based on the evidence of Professor Bleyer, the judge relied in part on his own finding (at [50]) that the amount actually given to the respondent was in excess of that stipulated "by the amendment to IRS-III to be administered". That statement was erroneous, as the amendment does not refer to the dosage of Act-D. As that error in detail has not been relied on by the appellants, and as it is minor in nature, I do not propose that the conclusion be interfered with on appeal. 54 It follows, in my view, that the appeal should be dismissed. It remains necessary, however, to deal with the notice of contention filed by the respondent in relation to the rejection by the judge of her case against Professor O'Gorman-Hughes, and with her cross-appeal against the judge's refusal to order the appellants to pay indemnity costs.

The notice of contention 55 Many of the grounds of contention are directed to issues on which the respondent was successful in the case against the hospital, and refer to minute details in the evidence which are said to support that judgment. As the hospital's appeal should be dismissed, there is no point in my view in considering those details further.

Failure to keep informed 56 The respondent also contends that the judge erred in finding that Professor O'Gorman-Hughes was not negligent in failing to keep abreast of the literature and publications relevant to the treatment he prescribed for her and which had been published before that treatment commenced. She submits that the duty of care Professor O'Gorman-Hughes owed to her to do so could only have been achieved by Professor O'Gorman-Hughes undertaking for himself the obligation of conducting a literature search and making inquiries of the IRS Group. It is appropriate to consider this ground of contention. 57 The ground raises a number of related issues. The first is whether the duty exists in such absolute terms whatever the nature of the treatment being prescribed by a medical practitioner. The second issue is whether Professor O'Gorman-Hughes should have made the search and inquiries himself, rather than leaving it to Dr White to bring to his attention any relevant material he may randomly have received from his overseas sources. If it was not necessary for Professor O'Gorman-Hughes to perform this obligation himself, the third issue is whether the system which was in place — based on the obligation of Dr White to pass on any such information he received — was sufficient to discharge the obligation of Professor O'Gorman-Hughes. The fourth issue is whether a more proactive system would have discovered the only material published before the respondent underwent her treatment, the paper written by Dr Ruymann published in late 1987 (see par [26] supra). The final issue is whether the failure to set up such a system caused the respondent to become a quadriplegic. 58 I do not accept that the duty can be stated in the absolute terms for which the respondent contended. There have to be some practical limits to the obligation of a medical practitioner to keep abreast of the literature and publications relevant to the treatment he or she is prescribing. The obligation of medical practitioners to exercise a degree of skill and care expected of persons exercising and professing to have that special skill, as stated in Rogers v Whitaker (at [483]), was applied to lawyers (in particular, to barristers) in Heydon v NRMA Ltd (2000) 51 NSWLR 1 at [362]. But, because the material available to barristers at the present time through online searching was said in that case, perhaps only half-seriously, to be usually of marginal relevance and of even less real use to them, there is a limit to what may fairly be expected by way of such research in the little time usually available (Heydon v NRMA Ltd at [663]–[664]). It has not been suggested that the medical practitioners face a similar deluge of material available. Medical practitioners nevertheless have very busy practices and little time for the research which ideally should be done in many cases. Much will turn on the nature of the treatment being prescribed and on the nature of any risk which could reasonably be involved in that treatment. 59 Professor O'Gorman-Hughes was in a somewhat unusual position when he was treating the respondent. As already stated (in par [15] supra), he was in this case relying on an overseas protocol involving a radical, experimental and controversial procedure, and TIT had rarely been used in the hospital previously for prophylactic purposes. Each of those three adjectives — radical, experimental and controversial — is important in this particular case, as is their combination. Professor O'Gorman-Hughes had only used TIT once before, but for salvage rather than prophylactic purposes, on a very young child (referred to as Patient "S") some twelve months after her tumour had been diagnosed, and who did not survive. It was a very different situation to that facing Professor O'Gorman-Hughes in his treatment of the respondent. Salvage TIT is far more aggressive than prophylactic TIT, as it is used to achieve survival for desperate cases. That was not the case with the respondent. As part of any medical practitioner's duty of care to exercise reasonable care and skill in the provision of professional advice and treatment, a duty to take reasonable care to keep informed before embarking on the treatment of the nature prescribed in this case arose whether or not it was reasonably foreseeable that it could cause paraplegia or quadriplegia. The whole nature of this particular treatment, involving as it did injections directly to the membranes of the respondent's spinal canal, and thus involving her central nervous system, indicated the reasonable foreseeability of some form of neurological complication. 60 Professor O'Gorman-Hughes himself recognised the need to "review frequently" the information and literature on tumours such as rhabdomyosarcoma and soft tissue sarcoma. Accepting the existence of a duty on his part in the particular circumstances of this case to take reasonable care to inform himself of "the most current and considered guidelines available in the world" (to use the words of the judge at [41]), I do not accept that the performance of this duty could only have been achieved by Professor O'Gorman-Hughes by undertaking for himself the obligation to conduct a literature search and make inquiries of the IRS Group. Practising in a leading hospital as a member of a team permits a task such as this to be delegated, so that the same searches do not have to be made by many (if not all) of the members of that team. But this duty did require Professor O'Gorman-Hughes to ensure that, in the particular circumstances of this case (as outlined in the previous paragraph), he was placed in as good a position by a system based on collaborative teamwork as he would have been had he performed the duty himself. He had to do more in this case than merely rely in a purely reactive manner to whatever information Dr White may randomly have received from his overseas contacts. If he could not do it himself, he had to take reasonable care to ensure that someone else actively searched for material which was relevant to the task he was to undertake. 61 The evidence established that the journal in which Dr Ruymann's paper was published was available by February 1988 in the libraries of both the University of Sydney and the University of New South Wales. The evidence is that Professor O'Gorman-Hughes became aware in late 1989 that the IRS Group in America had recommended that central nervous system prophylaxis including TIT be changed because of known toxic consequences, and in particular that it had advised that TIT should be limited to four courses with limited maximum dosages of Ara-C (see par [18] supra), when he noted a reference to that fact in Dr Ruymann's paper — which he had found in the medical school library at the University of New South Wales. (The respondent's treatment took place between March and August 1989.) There was also available at that time (before online electronic searching was generally available) a publication called "Index Medicus" which enabled medical practitioners to search for specific articles of interest. There was also available in university libraries an internet search tool known as "Medline". 62 There was substantial evidence which was contrary to the view of Dr Plowman on which the judge relied (see par [27] supra) — that the journal had an impact value of virtually zero, and that the paper was not one Professor O'Gorman-Hughes would have been expected to have seen. The judge made no reference to this evidence in his judgment. 63 Professor Stevens (who was called by the respondent) said that, although the journal in which Dr Ruymann's paper was published was new, it belonged to a series of medical journals which was well established. He said that a person with qualifications to carry out research relating to the treatment of sarcomas, exercising reasonable care and skill, would have discovered Dr Ruymann's paper. He was aware of the Index Medicus, and said that, before the days of online electronic searching, it was the only way of searching for such articles. He had been aware of the article himself, and he had had no difficulty in accessing it. 64 Dr D'Angio (who was called by the appellants) said that the journal in which Dr Ruymann's paper was published was part of a well respected, but not widely read, publication. He referred to an "Impact Factor List", and he said that the journal in which Dr Ruymann's article was published did not, even as at the date of the trial, appear in the first 10 or 20 journals read by oncologists. That evidence did not lessen the weight to be given to Professor Stevens' evidence as to the accessibility of the paper when searches were made. 65 Dr Kellie (who was also called by the appellants) agreed that a search on Medline for "rhabdomyosarcoma" and "treatment" would have produced a number of articles which should be looked at. He also agreed that the journal in which Dr Ruymann's paper was published was part of a well regarded series, and that it would be an important source to access in relation to the treatment of a patient such as the respondent. However, Dr Kellie also said that, whilst the paper was an important review, he doubted whether he would have been expecting to find new information in a review, and that it was not the place he would go to for the sort of information for up-to-date references to treatment. 66 The circumstances in which Professor O'Gorman-Hughes was about to treat the respondent did not permit him to place such limited reliance on Dr Ruymann's paper merely because it was a review. Even Dr Kellie described it as an excellent review; he merely preferred to go back to the original research identified in the 167 references or citations in the endnotes. 67 Professor Bleyer (who was called by the respondent) said that Professor O'Gorman-Hughes should have taken steps to find the relevant literature, to read it, to understand it and to discuss it; a simple literature search would have revealed the Ruymann paper, which he described as a review of the experience of the IRS Group, and which should have been looked at before embarking on any consideration of TIT. The appellants point out that Professor Bleyer did not mention the Ruymann paper in any of his reports. That fact does not, however, meet his evidence that it was accessible by a simple literature search which was available at the relevant time. 68 This Court was not directed to any clear evidence as to the number of papers which would have been available if a proper search had been made in 1989, but it appears to have been generally accepted by the witnesses that there would not have been many. In my opinion, Professor O'Gorman-Hughes would have been obliged to read this paper if it had been discovered before treating the respondent. 69 If Professor O'Gorman-Hughes had read Dr Ruymann's paper, he would have become acutely aware of the advice given in the amendment to the IRS-III protocol in 1987 to lower the dosages and frequency of the treatments to substantially below what had been recommended in the IRS-II protocol on which he had relied. If what the paper disclosed of the reasoning for that advice did not satisfy him to change his own protocol, I am satisfied that he, like Dr Kellie, would have gone to the source of the information, the 1987 amendment to IRS-III. The judge held that, if Professor O'Gorman-Hughes had been aware of the amendment, the treatment based in part on the earlier IRS protocol would not have been administered. I have already expressed the view that there was no error in that finding (see par [47] supra). 70 The appellants, however, have put a further argument on causation so far as this part of the respondent's case is based on the failure to search for and find Dr Ruymann's paper. Professor Stevens conceded that, if the footnote in that paper to the reference to a small number of patients with brain stem dysfunction were followed up (see par [26] supra), it would have been discovered that only one person had in fact been affected by brain stem dysfunction, but he maintained that the information in Tables 8 and 9 relating to the reduction in the frequency and dosage of TIT to be administered (see [26] supra) was nevertheless of importance. He also pointed out that the significance of the article related to the effect of the interaction between the chemotherapy and the radiation, and that there was no distinction between the cases where radiation had been delivered to either the brain stem or the spinal cord. I do not consider that the discovery that only one patient had suffered brain stem dysfunction would have dissuaded Professor O'Gorman-Hughes from pursuing the reason for the reduction advised by the amendment. I reject the appellants' argument. 71 I conclude that the system in place for sharing information in the oncology department of the hospital was insufficient for the treatment prescribed, that Professor O'Gorman-Hughes was aware of the nature of the system in place, and that, by proceeding with the treatment of the respondent without ensuring that he had been placed in as good a position as if he had conducted a literature search and any necessary follow-up inquiries himself, he failed in his duty to the respondent in the particular circumstances of this case to be in possession of all the necessary up-to-date information in relation to that treatment. 72 I am satisfied that, if Professor O'Gorman-Hughes had performed that duty, he would have become aware of the 1987 amendment to the IRS protocol, and that he would have modified the regime he had prescribed for the respondent. As the excessive dosages of Ara-C by way of TIT and their frequency associated with the radiotherapy contributed to the respondent's myelopathy and thus her quadriplegia, causation is established as discussed earlier in this judgment. 73 There is already a judgment for the respondent against Professor O'Gorman-Hughes for $7,000.000, notwithstanding that no finding of negligence had been made against him at the trial. This was done pursuant to the agreement between the appellants referred to earlier in this judgment (see par [35] supra). That judgment against Professor O'Gorman-Hughes is now supported by this finding that his negligence was a cause of the respondent's injuries. The hospital is vicariously responsible for that negligence as well as for the negligence of Dr White. The judgment against the hospital is also supported by this finding.

Failure to warn 74 In Rogers v Whitaker, the High Court held that, except in the case of an emergency or where disclosure would prove damaging to the patient, a medical practitioner has a duty to warn the patient of a material risk inherent in proposed treatment. A risk is material if, in the circumstances of the particular case, a reasonable person in the patient's position, if warned of the risk, would be likely to attach significance to it or if the medical practitioner is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it. 75 There was no dispute in this appeal that Professor O'Gorman-Hughes had a duty to warn the respondent's parents of the risk of paraplegia or quadriplegia in the treatment he proposed for their 13 year old daughter. The judge's finding that he did not perform that duty has not been challenged. The judge's finding challenged by the respondent in her notice of contention is it that, even if a warning had been given to her parents in relation to the treatment that it might cause paraplegia or quadriplegic, they would have given permission for the treatment it to be carried out. 76 The judge's reasons for his finding are expressed very briefly. The respondent's mother had given evidence that she and her husband had trusted Professor O'Gorman-Hughes. (The respondent's father had also died by the time of the trial.) The judge did no more than refer to that evidence and state his finding now challenged by the respondent. He did not refer to the evidence supporting that finding. 77 The respondent's mother gave evidence in chief that, had she been given this advice, she would probably have sought a second opinion in the United States. On the assumption that she had been advised by Professor O'Gorman-Hughes not to follow the treatment he proposed (because of the risk of paraplegia or quadriplegia), she agreed that, if she had been told that the IRS Group was one of the largest groups in the United States conducting studies in relation to the treatment he proposed and that the ISG Group advocated the treatment he proposed (but at lower dosages and frequency proposed by Professor O'Gorman-Hughes), she would have followed the advice given by that Group. 78 In cross-examination, the respondent's mother said that, if she had been advised that there was a risk that without the proposed treatment the cancer may grow or spread and cause the respondent's death, she felt that, because their daughter was so sick, she and her husband would probably have agreed to the proposed treatment to keep the respondent alive; if they had had to make the awful choice, she agreed, they would have chosen to keep her alive. 79 The answer the respondent's mother gave in chief is not necessarily inconsistent with her answer in cross-examination, as the respondent asserts, as she was not invited by the question asked in chief, nor did she deal in her answer, with the dilemma specifically posed in the cross-examination, of choosing between the risk of her daughter's life as a paraplegic or quadriplegic and the near certainty of her death from the cancer. 80 As McHugh J has pointed out — in Chappel v Hart (at [32] footnote 64) and in Rosenberg v Percival (2001) 205 CLR 434 at [26]–[27] — human nature being what it is, most persons who suffer harm as a result of a medical procedure and who sue for damages genuinely believe that they would not have undertaken the procedure if they had been warned of the risk of that harm. The same could be said of parents of young children who have had to make difficult decisions for their children. The present case presents the reverse of that situation. The answer given by the respondent's mother in cross-examination was obviously a concession in favour of the appellants' case, and was thus so transparently honest, that any conclusion reached other than that reached by the judge would have been perverse. In my opinion, no error has been demonstrated by the respondent in that finding. I would therefore reject this ground of the notice of contention.

Cross-appeal 81 In a separate judgment, the judge refused to order the appellants to pay indemnity costs to the respondent. The respondent's application was based on two offers of settlement made prior to the trial: (i) an offer pursuant to SCR Part 22 (Division 1) on 9 July 1999 to settle for $2,350,000 plus costs, and (ii) an offer by way of a Calderbank letter on 31 January 2005 to settle for $6,000,000 plus costs. The second offer was open for an acceptance until 4:30 pm on Friday 4 February 2005, which was the Friday before the trial commenced on Monday 7 February. The trial continued with breaks until 16 March, when judgment was reserved. The appellants did not respond to either offer. The parties advised the judge on 17 February, the sixth day of the trial, that agreement had been reached as to damages at $7,000,000 in the event that the respondent succeeded in her claim.

Offer pursuant to SCR Part 22 82 The application based on the first offer of settlement was rejected by the judge because, as at that time, the material served on the appellants by the respondent did not demonstrate that she would necessarily succeed. The respondent's Statement of Claim did include her case based on the 1987 amendment to IRS-III, but the material served did not include much of what later became her case at the trial. Professor O'Gorman-Hughes was then still alive, and he contended that he had at all times carried out the treatment of the respondent in accordance with medical standards as he appreciated them at that time. 83 Part 52 rule 22 provides that, where a plaintiff who has made an offer of compromise in accordance with Part 22 (Division I) which is not accepted by the defendant, and where the plaintiff obtains a judgment no less favourable than the terms of the offer, then, unless the Court otherwise orders, the plaintiff will be entitled to an order against the defendant for costs on an indemnity basis in relation to his costs incurred from the day after the offer was made. The onus is on the defendant to persuade the Court that indemnity costs should not be ordered. He must demonstrate the basis on which an order should be made denying the plaintiff's entitlement to indemnity costs. He must establish that he had given serious thought to the risk involved in non-acceptance of the offer, and that he had assessed the plaintiff's case properly and in the context of the rule and the achievement of its purpose — to encourage the proper compromise of litigation, in the private interests of the litigants and in the public interest of the prompt and economical disposal of litigation. Generally, exceptional circumstances are required to justify such an order denying the plaintiff's entitlement. See, generally, Fowdl v Fowdl, Court of Appeal, 4 November 1993, unreported, per Kirby P at 12, 16; Hillier v Sheather (1995) 36 NSWLR 414 at 422-423; Morgan v Johnson (1998) 44 NSWLR 578 at 581-582. 84 Although the Court may take into account the particular features of a case and the difficulty of putting an accurate estimate on its value in advance of the litigation, those considerations do not provide a basis by themselves for denying the party making the offer an order for indemnity costs: Hillier v Sheather (at 423). In Maitland Hospital v Fisher [No 2] (1992) 27 NSWLR 721 at 725, it was said that a miscalculation in a case with large imponderables, where the course taken by the party to whom the offer was made may have been perfectly reasonable, would provide a basis for denying an order, but the subsequent decision in Hillier v Sheather suggests that, whilst "large" imponderables may be a relevant factor in the particular case in denying an order for indemnity costs, in the general case a party's reasonable belief as to his own chances of success does not alone provide a basis for denying an order for indemnity costs. 85 In the present case, the respondent has submitted that the judge had wrongly considered that a belief that the appellants had reasonable grounds for defending the claim constituted exceptional circumstances disentitling her from an order for indemnity costs. However, the fact that the plaintiff's case had changed significantly between the date of the plaintiff's offer and the trial in which the judgment obtained is higher than the amount of the offer does provide a sufficient basis for an order denying the plaintiff's entitlement to indemnity costs: Maitland Hospital v Fisher [No 2] (at 725). The very nature of the situation itself demonstrates that it would be unfair to a defendant to make an order for indemnity costs when the evidence at the trial is different from that known to the defendant at the time of the offer. Whether or not this is an "exceptional" situation does not matter. 86 In the present case, so far as the evidence goes, there can be no suggestion that at the time of the first offer the appellants knew of the information Dr White had at his disposal and of the system in place for the dissemination of that information to Professor O'Gorman-Hughes and others on which the respondent's successful case eventually depended (see par [39] supra). The judge's reference to a reasonable belief by the appellants at the time of the first offer that they had reasonable grounds for defending the claim was therefore appropriate in the circumstances. 87 I would not, accordingly, interfere with the judge's decision to deny the respondent indemnity costs based on the first offer.

The Calderbank letter 88 The application based on the second offer of settlement was rejected by the judge because, although he accepted that it was a genuine offer of compromise, the respondent had — […] not established the criteria which [the] authorities have indicated are to be considered when the court exercises its discretion to award indemnity costs following an offer made in a Calderbank letter. The judge gave no further reasons for his decision. 89 A letter written by one party to another (which may be endorsed "without prejudice except as to costs") offering to settle the litigation for a specified sum exclusive of costs, which has not been accepted by the other party, is relevant to the order for costs to be made where the judgment entered is more favourable to the party making the offer than the settlement offered if the Court is satisfied that the offer ought to have been accepted by the party to whom the offer was made: Calderbank v Calderbank [1976] Fam 93 at 106. 90 However, a Calderbank letter from a plaintiff to a defendant offering to settle the litigation for a specified amount is not an offer of compromise pursuant to Part 22. There is no presumption that indemnity costs will be ordered in favour of the plaintiff where the amount awarded is greater than the amount nominated in the letter. That factor is only one of the factors to which the court may have regard. The discretion to award indemnity costs following a Calderbank letter must be considered having regard to all of the circumstances of the case, including the relevant strengths and weaknesses of each party's case as they may have been apparent to the parties at the time the offer was made. See, generally, Flemington Properties Pty Ltd v Raine & Horne Commercial Pty Ltd, Federal Court (Lehane J), 2 February 1998, at 3; SMEC Testing Services Pty Ltd v Campbelltown City Council [2000] NSWCA 323 at [37]; Jones v Bradley [No 2] [2003] NSWCA 258 at [8]–[9], [16]; Leichhardt Municipal Council v Green [2004] NSWCA 341 at [19]; Brymount Pty Ltd v Cummins [No 2] [2005] NSWCA 69 at [13]. 91 The respondent has submitted that the offer of settlement was not only genuine but also a reasonable one. By this time, the appellants should have been fully aware from Dr White of the information he had at his disposal and of the system in place for the dissemination of that information to Professor O'Gorman-Hughes and others, so that her case on which she was ultimately successful was directly before them. The previous imponderables had considerably narrowed, and whatever belief the appellants may have had as to their chances of success, their decision to run the risk was unreasonable having regard to the relevant strengths and weaknesses of each party's case as they were apparent to the appellants at that time. The amount of the offer was $1,000,000 less than the figure agreed a short time later. The appellants' attitude was, the respondent submitted, indicative of an intention to litigate the matter "at all costs". 92 The appellants have submitted that, given the number and complexity of the issues involved, the respondent has failed to demonstrate that the rejection of the offer was imprudent. They point out that the offer was received at 5.30 pm on 31 January and was open for acceptance until 4.30 pm on 4 February 2005, and that it did not seek to advance any reasons in favour of its acceptance. They also point out that Professor O'Gorman-Hughes was exonerated by the judge. 93 There is no obligation on the party making an offer of compromise to advance reasons why the offer should be accepted. In any event, in the week before the trial was due to commence, it may safely be assumed that the parties are in possession of sufficient information as to the relevant strengths and weaknesses of each other's case. It is not suggested that the appellants sought further time to consider the offer; they simply ignored it. The appellants in their submissions did not deny the respondent's assertion that they knew of the information Dr White had at his disposal and of the system in place for the dissemination of that information to Professor O'Gorman-Hughes. Nor did they make any attempt to deal with that material, or with the absence of any mention by the judge of this very important material in his separate judgment dealing with costs. The exoneration of Professor O'Gorman-Hughes by the judge did not explain why an order for indemnity costs should not have been made against the hospital. The agreement between the appellants to share any verdict given against either of them makes the exoneration of Professor O'Gorman-Hughes irrelevant. 94 I am satisfied that the judge erred in the exercise of his discretion in relation to the second offer by failing to consider the appellants' knowledge of the strengths of the respondent's case at the time of the second offer. Exercising my own discretion, I would order that both appellants pay the respondent's costs of the litigation in the Common Law Division up to and including 31 January 2005 on a party and party basis, and her costs thereafter on an indemnity basis.

Costs of the appeal 95 The respondent has been successful in both the appeal (including obtaining judgment against Professor O'Gorman-Hughes in her notice of contention) and the cross-appeal. She did not succeed in relation to one ground in her notice of contention, the failure to warn. Next to no time was spent on that issue. I would order the appellants to pay the respondent's costs of the appeal and of the cross-appeal. To avoid any doubt, the costs of the appeal are intended to include the costs of the notice of contention.

Proposed orders 96 I propose that the following orders be made: 1. The defendants' appeal against the judgment entered in favour of the plaintiff against the first defendant is dismissed. 2. The judgment entered in favour of the plaintiff against the second defendant for $7,000,000 is confirmed. 3. The plaintiff's cross-appeal is allowed. 4. The defendants are to pay the plaintiff's costs of the litigation in the Common Law Division up to and including 31 January 2005 on a party and party basis, and her costs thereafter on an indemnity basis. 5. The appellants are to pay the respondent's costs of the appeal and of the cross-appeal.

---------------------------------

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.