Court of Appeal Supreme Court New South Wales

• Medium Neutral Citation: Coote v Kelly [2013] NSWCA 357
• Hearing dates: 8 and 9 October 2013
• Decision date: 28 October 2013
• Before: Basten JA at [1]; Hoeben JA at [2]; Leeming JA at [3]

Decision

Appeal allowed, judgment and orders of the Court made on 14 March 2012 be set aside, and there be a new trial.

[Note: The Uniform Civil Procedure Rules 2005 provide (Rule 36.11) that unless the Court otherwise orders, a judgment or order is taken to be entered when it is recorded in the Court's computerised court record system. Setting aside and variation of judgments or orders is dealt with by Rules 36.15, 36.16, 36.17 and 36.18. Parties should in particular note the time limit of fourteen days in Rule 36.16.]

• Catchwords: APPEAL - challenge to findings of fact - failure to address inconsistencies in evidence - issues not resolved by findings as to credibility and reliability - need to address failure of four practitioners to observe key element of melanoma relied on by plaintiff - retrial required; NEGLIGENCE - medical practitioner - whether melanoma misdiagnosed as a plantar wart - whether breach of duty - whether finding that melanoma had probably metastasized before failure to diagnose supportable on the evidence
• Legislation Cited: Civil Liability Act 2002 (NSW), s 5D(1), s 5E; Civil Procedure Act 2005 (NSW); Uniform Civil Procedure Rules 2005 (NSW), r 51.53
• Cases Cited: Abalos v Australian Postal Commission [1990] HCA 47; 171 CLR 167; Fox v Percy [2003] HCA 22; 214 CLR 188; Hawkesbury District Health Service Ltd v Chaker [2010] NSWCA 320; Kirkwood v New South Wales Insurance Ministerial Corporation [1995] NSWCA 244; M v The Queen [1994] HCA 63; 181 CLR 487; Mitchell v Cullingral Pty Ltd [2012] NSWCA 389; MM Constructions (Aust) Pty Ltd v Port Stephens Council [2012] NSWCA 417; Paul v Cooke [2013] NSWCA 311; Pollard v RRR Corporation Pty Ltd [2009] NSWCA 110; State Rail Authority (NSW) v Earthline Constructions Pty Ltd (In liq) [1999] HCA 3; 73 ALJR 306; SS Hontestroom v SS Sagaporack [1927] AC 37; Tabet v Gett [2010] HCA 12; 240 CLR 537; Vagg v McPhee [2013] NSWCA 29; Voulis v Kozary (1975) 180 CLR 177; Warren v Coombes [1979] HCA 9; 142 CLR 531; Waterways Authority v Fitzgibbon [2005] HCA 57; 79 ALJR 1816; Watson v Foxman (1995) 49 NSWLR 315; Xu v Jinhong Design & Constructions Pty Ltd [2011] NSWCA 277
• Category: Principal judgment
• Parties: Melissa Anne Coote (Appellant); Dr Steven Kelly (Respondent)
• Representation: Counsel: BMJ Toomey QC / EG Romaniuk SC / OJ Dinkha (Appellant); TK Tobin QC / JM Morris (Respondent); Solicitors: Stacks Goudkamp (Appellant); Norton Rose (Respondent)
• File Number(s): CA 2012/100122

Decision under appeal

• Citation: Coote v Dr Kelly [2012] NSWSC 219
• Date of Decision: 2012-03-14 00:00:00
• Before: Schmidt J
• File Number(s): 2011/339988

Judgment

1 BASTEN JA: I agree, for the reasons given by Leeming JA, that the judgment below must be set aside and, assuming no extracurial resolution, that there must be a retrial. I also agree that the parties should bear their own costs in this court and that the costs at first instance should await that retrial.

2 HOEBEN JA: I agree with Leeming JA.

3 LEEMING JA: The appellant is the widow and executrix of the plaintiff Mr Malcolm Coote, who sued his doctor, the respondent Dr Steven Kelly. On five occasions, from 3 September 2009 until 20 May 2010, Dr Kelly treated what he had diagnosed as a plantar wart on the sole of Mr Coote's left foot by applying cryotherapy (liquid nitrogen). Subsequently, three other doctors inspected and treated Mr Coote's left foot. Dr Wall did so on 2 September 2010. Dr Hiddins examined his foot on five occasions between 11 January and 11 March 2011, on one of which she asked her more senior colleague Dr Cooke for advice. Dr Wall, and initially Dr Hiddins, diagnosed a plantar wart, which they treated by applying cryotherapy. Dr Hiddins pared back what she had diagnosed as a wart on each occasion, but on the second last consultation, she ceased paring it back, consulted Dr Cooke and did not otherwise treat it because it bled easily. On the last consultation, Dr Hiddins' evidence was that the lesion had grown and changed colour. A biopsy was taken, which identified melanoma. More precisely, it was an acral lentiginous melanoma, which is a relatively rare form of the cancer, accounting for 1-3% of all melanomas in Australia, which appears on the palms of the hands and the soles of the feet and which may or may not be pigmented (if not pigmented, it is an "amelanotic" melanoma).

4 The melanoma was removed shortly after it was diagnosed, but it had already metastasized (which is to say, cancerous cells had entered Mr Coote's blood system or lymphatic system and spread elsewhere in his body). Metastasis may be local or distant. Patients with distant metastases (which is "Stage IV" in a United States staging system for the disease approved in Australia) have survival rates measured in months rather than years. A study of 17,600 patients diagnosed with melanoma on which the primary judge relied showed that of the 1158 with distant metastases, some 863 had died within 2 years. On the other hand, if melanoma is diagnosed and removed early, a complete cure is highly likely.

5 Mr Coote claimed that Dr Kelly had negligently failed to diagnose his melanoma. He commenced proceedings by statement of claim filed on 25 October 2011. He swore an affidavit on 24 November 2011 and gave evidence on commission on 16 December 2011. A defence was filed on 13 February 2012, and the trial ran for six days between 27 February and 5 March 2012. The primary judge delivered lengthy reasons with commendable and understandable expedition, on 14 March 2012. Mr Coote died two months later.

6 The primary judge found that Dr Kelly had breached his duty of care to Mr Coote by failing to observe and investigate a small black spot which her Honour found had been present on his foot from the beginning of September 2009. However, her Honour entered a verdict for the defendant, because she also found that it was probable that the melanoma had already metastasized prior to Mr Coote consulting his doctor, so that Dr Kelly's breach had not caused him harm.

7 Mrs Coote challenged the finding of no causation, and Dr Kelly challenged the finding of breach. The finding of breach turned on the resolution of conflicting testimonial evidence of primary fact, which fell to be evaluated in light of the contemporaneous documents as well as the inherent probabilities which were the subject of conflicting expert evidence. The question of causation was considerably more complicated, factually and legally, but plainly enough the starting point was the existence and extent of any melanoma on Mr Coote's foot in September 2009. It follows that the first issue arising on this appeal is Dr Kelly's challenge to the finding of breach.

8 Due allowance should be made for the fact that the litigation at first instance was prepared, conducted, and determined within five months of its commencement. However, for the reasons set out below, I have concluded that her Honour's finding of breach must be set aside. I first address the evidence bearing upon the issue, then the reasoning of the primary judge, and then explain why the respondent's challenge to her Honour's reasoning process must be accepted. Even so, if nothing Dr Kelly did, or failed to do, was causative of harm to Mr Coote, that would be an end of the matter. However, the force of the appellant's challenge to the findings of causation means that, regrettably, there must be a retrial.

Breach of duty

9 Mr and Mrs Coote gave evidence that there was a black, hard, slightly raised spot, like "a black piece of tar" on the sole of his left foot. Mrs Coote was firmly of the view that the spot had been there prior to her husband's first consultation with Dr Kelly; Mr Coote's recollection was that it was there from early September 2009. Dr Kelly conceded that if, as the plaintiff contended, there had been such a small black spot, he breached his duty in failing to observe and investigate it. But he denied there was such a spot, as did Drs Wall and Hiddins.

10 There was no issue that the defendant owed a duty of care to the plaintiff, nor that it had been breached if the lesion when seen by Dr Kelly was black. Thus, the issue of breach would be resolved, favourably to the plaintiff, if her Honour found that there was a black spot visible on the sole of Mr Coote's foot in September 2009. The case at trial could easily have been much more complex, if it had been contended that there had been a breach because (a) there was a black melanoma which was concealed by a plantar wart, or (b) there was an amelanotic melanoma which became pigmented by May 2010, or (c) although the melanoma was not present in September 2009, it was visible prior to Dr Kelly's last consultation in May 2010. That none of these issues arose was a consequence of the evidence of Mr and Mrs Coote.

11 The primary judge found, in accordance with the evidence of Mr and Mrs Coote (primarily Mrs Coote), that the lesion was present and pigmented when Mr Coote first saw Dr Kelly. It followed that there was a breach of duty by Dr Kelly undertaking no further investigations and diagnosing and treating his patient merely for a plantar wart. The finding of breach was challenged on appeal by Dr Kelly, purportedly by a cross-appeal. However the parties agreed that its grounds could be treated as grounds of a Notice of Contention (there were no orders adverse to Dr Kelly which could found a cross-appeal).

12 Dr Kelly's challenge was to a finding of fact made by the primary judge with the benefit of seeing the trial as a whole, including the testimonial evidence both of Mrs Coote and three doctors who failed to diagnose melanoma. That appellate judges are in the "permanent position of disadvantage" to which Lord Sumner referred in SS Hontestroom v SS Sagaporack [1927] AC 37 at 47 is clear. The deference reflecting that disadvantage is slightly differently expressed in Warren v Coombes [1979] HCA 9; 142 CLR 531; Abalos v Australian Postal Commission [1990] HCA 47; 171 CLR 167; State Rail Authority (NSW) v Earthline Constructions Pty Ltd (In liq) [1999] HCA 3; 73 ALJR 306 and Fox v Percy [2003] HCA 22; 214 CLR 188.

13 In Xu v Jinhong Design & Constructions Pty Ltd [2011] NSWCA 277 at [17], Basten JA noted that in some cases, where sufficient facts were incontrovertibly established by the evidence, a substituted judgment might be possible (Fox v Percy was such a case, as was Voulis v Kozary (1975) 180 CLR 177). His Honour also noted that it was possible that the appellate court could detect error without being satisfied that a contrary finding was appropriate. In that case, if the court were also satisfied that there had been a substantial miscarriage of justice, a retrial is required: Uniform Civil Procedure Rules 2005 (NSW), r 51.53. Earthline was an example, because in the language of Gaudron, Gummow and Hayne JJ at 321, "the substance of the matter is that there has not yet been a determination of the SRA's case upon a consideration of the real strength of the body of evidence it presented". So too here; there has not been a determination of the question of breach upon a consideration of the real strength of the evidence presented by Dr Kelly.

The evidence relevant to breach

(a) The contemporaneous documentary evidence

14 When the melanoma was excised on 26 March 2011, it was strongly pigmented. There was complete agreement between the four experts (who gave evidence jointly) that it was not possible to say whether the lesion had originally been pigmented or non pigmented. The only contemporaneous documentary evidence in the period from September 2009 until May 2010 is found in the notes made by Dr Kelly in relation to five consultations on 3, 15 and 30 September 2009, 14 October 2009 and 20 May 2010. On two of those occasions, Dr Kelly was also examining varicose veins in Mr Coote's left leg, which is of no relevance to the issue. Otherwise, the entirety of the doctor's notes (which were handwritten save for the last entry) record the following:

3/9/09	Large (L) plantar wart → cryo
15/9/09	Cryo to wart
30/9/09	Cryo to wart
14/10/09	Cryo to wart
20/5/10	Debridement and cryo of left foot plantar wart

15 Dr Anthony Wall saw Mr Coote on 2 September 2010. His notes merely record "cryo plantar wart".

16 The two doctors' notes, which are unquestionably succinct, are important for the fact that they do not record any mention of pigmentation in what both doctors diagnosed as a plantar wart. Human papillomavirus, which causes warts, was detected in the tissue removed from Mr Coote's foot (there was uncontroversial evidence that the virus is found in all warts, but may also be found in the skin without the existence of warts). If indeed all there was to see was an unpigmented lesion which presented as a plantar wart, there was no occasion for more detailed notes to be made.

17 Dr Rosalind Hiddins saw Mr Coote for the first time on 4 January 2011 concerning a pain in his right elbow. She saw him again on 11 and 28 January 2011, 11 and 24 February 2011 and 11 March 2011 concerning his foot. She made notes on each occasion, which although concise are significantly more discursive than those of Drs Kelly and Wall.

18 Dr Hiddins' notes record administering cryotherapy for a plantar wart on 11 January 2011. On 28 January 2011 further cryotherapy was administered and her notes record "on exam - large 5 mm wart appears to be contracting" and that the patient had been using Wart Off (an over the counter topical treatment) between visits to her surgery. On 11 February 2011 her notes recorded that he "hasn't been using Wart Off in between" and "large wart measures 10 mm diameter" and "wart closer to heel 5 mm diameter - ? separately from normal tissue" and that cryotherapy for a plantar wart was administered. On 24 February 2011 her notes record:

"has had cryo treatment for last 8 weeks

r/v with Dr Cooke after debriding wart - ? proud flesh or granulation tissue

plan

not for treatment today - leave to heal for next 2 weeks and review"

Finally on 11 March 2011 her notes record:

"had 2 weeks off wart treatment - lesion has grown in meantime

soft, painful and vascular

2 x 3 mm biopsy removed from area and swab taken

no groin or popliteal fossa nodes"

Both her notes and the request for histology on 11 March 2011 record:

"History of wart-like lesion - atypical appearance? Infection? wart? pigmented lesions"

19 The histology report five days later identified melanoma, with "invasive malignancy very likely".

(b) The testimonial evidence of Drs Kelly, Wall and Hiddins

20 Each of those three doctors was trained to diagnose melanoma, including acral lentiginous melanomas on the sole of the foot. Each of them unquestionably examined Mr Coote's foot closely before and during the administration of cryotherapy. Dr Hiddins used a large magnifying glass with a fluorescent light around it to do so. Nowhere in the contemporaneous records made by any of the doctors is there a reference to what all of them understood to be an important diagnostic fact, if it was present, namely, that the lesion was black, until Dr Hiddin's reference on 11 March 2011 to "pigmented lesions".

21 Each of those three doctors was cross-examined as to their recollections, which were different. Dr Kelly candidly acknowledged that he had no specific recollection of any of the five consultations he had given. He conceded that it was possible that he might have failed to detect the melanoma even if it were pigmented, because, to use his words, "no one is infallible" and conceded that it followed as a matter of logic that what Mr and Mrs Coote saw was quite feasibly correct.

22 Dr Wall gave evidence that he had "quite a lot of experience with melanoma" and had offered melanoma screening as a service when he set up his clinic. He said that he had an actual recollection of Mr Coote's consultation with him on 2 September 2010, that he had told him about the usual causes of a plantar wart, that quite often they spontaneously regress within about a two year timeframe, that a reasonable treatment choice was to have no treatment, another was to apply salicylic acid to the lesion continuously for a period of four months, and that if he wanted to get rid of the wart more quickly he would recommend diathermy ablation and curettage under local anaesthetic. He recalled that Mr Coote rejected that advice and requested cryotherapy. Dr Wall said that he recalled observing a hyperkeratotic lesion (which refers to an abnormal thickening of the upper outermost layers of the skin), which image he could recall as he sat in the witness box some 18 months later. He said that he agreed with the diagnosis of a plantar wart. He was challenged strongly in cross-examination as to the accuracy of his detailed recollections (having given some 2500 consultations in the meantime) which were not supported by his notes. Nevertheless, Dr Wall rejected the propositions that he had been presented with a darkly pigmented lesion, and that he didn't have in truth the faintest clue as to what he saw on 2 September 2010 as he sat in the witness box a year and a half later. Part of Dr Wall's cross-examination was based on the fact that prior to giving evidence, Dr Kelly's solicitors had served a copy of Mr Coote's affidavit upon him, and notified him that they reserved the right to seek contribution from him, and that he had notified his insurer, whose claims manager had appeared when evidence was taken on commission from Mr Coote.

23 The notes made by Dr Hiddins are significantly more discursive than those of Drs Kelly and Wall. She was a more junior doctor who had been working at the After Hours Medical Centre in Charlestown for seven months before she saw Mr Malcolm Coote. She had an understanding of what a melanoma was, and that it could occur on the foot, but had never seen a melanoma in real life until seeing Mr Coote. With the assistance of her notes, she recalled that on 11 January 2011 she examined the wart which "looked predominantly flesh-coloured" with "some tiny brown spots", which she said was quite similar to how a plantar wart would appear. She pared back the wart and applied hand-held liquid nitrogen. She said that she retained (one year later) an image in her head of what the lesion looked like as follows:

"the lesion was approximately 5 mm in diameter, it was fairly symmetrical, a nice circular appearance to it, it was predominantly flesh-coloured and it had very small brown spots scattered throughout. At that point there was only one lesion and it was on the sole of the foot."

24 On the second consultation (28 January 2011), again Dr Hiddins said that she had an actual recollection of its appearance, which was that it had improved and was shrinking. Once again she pared back the wart and applied the cryotherapy. She said that it was still a flesh-coloured lesion with small brown spots.

25 On the third consultation (11 February 2011), a second lesion appeared and once again Dr Hiddins recalled that it was predominantly flesh-coloured with brown spots.

26 On the fourth consultation (24 February 2011), Dr Hiddins said that the lesion had "dramatically changed": "it had gone from a flesh-coloured lesion to a more purple pink type lesion" which bled quite easily when she pared it. Dr Hiddins asked her supervisor Dr Cooke to pop in to give a second opinion. He recommended that the lesion not be treated but be left for two weeks.

27 On the fifth consultation (11 March 2011), Dr Hiddins said that the lesion had grown so that "it was completely different", being a "very dark purple" and "much larger". That was when she took some scrapings for biopsy.

28 Dr Hiddins was cross-examined to the effect that nowhere in her notes was the change of colour recorded. She was taken to Mr Coote's evidence that it was a "definite dark colour all over by January", which she denied. She was also taken to Mrs Coote's evidence that it was black, which she denied. She maintained, notwithstanding any contemporaneous records or photographs, that the change of colour was as she had deposed. She rejected the notion that her recollection might be mistaken; she said "I am quite confident in my description". No differently from Dr Wall, she had been served with a copy of Mr Coote's affidavit and a letter reserving Dr Kelly's rights to make a claim against her.

(c) The testimonial evidence of Mr and Mrs Coote

29 On the other hand, Mrs Coote was adamant that the lesion was black from September 2009, and steadily growing. She said in evidence in chief that she had herself seen a small roundish spot, a bit smaller than a matchstick head, "roundish and black in colour, and very slightly raised, about a millimetre", which she tried to scrape off with her fingernail, and when it didn't move, she told her husband that he should consult Dr Kelly. She was herself familiar with plantar warts, having had them herself, and as a PDPHE teacher had had experience with schoolchildren with plantar warts. She said that what she saw on the sole of her husband's foot "didn't look anything like" a plantar wart. She maintained that evidence in cross-examination:

"Q. Was it ever at a point non pigmented, I am talking about March, April?

A. Not since the word go. Right from the very beginning always ever black.

Q. You could not be mistaken about that?

A. No, not at all, I just don't think you are understanding. It was never anything except black."

30 She said she had not discussed with her husband what they remembered observing in August 2009. She said she had not read the transcript of his evidence on commission. She was asked whether she might be incorrect about seeing the black lesion and discussing it with her husband in August 2009:

"A. Not at all, I thought it was a black piece of tar that I tried to scrape off his foot with a fingernail.

Q. Is it possible that you are mistaken how the lesion looked in September and October 2009?

A. No.

Q. Is it possible that you are mistaken about how the lesion looked in May 2010?

A. No.

Q. Is it possible that you are mistaken about how the lesion looked in September 2010 when he went to see Dr Wall?

A. No.

Q. And is it possible you are mistaken about how the lesion looked in the early part of January 2011?

A. No, it was always black. It didn't heal, it continued to grow.

Q. Is it possible that the pigmentation came about some time in either late 2010?

A. Tar is black, it was always black. August 2009, it continued to remain black throughout its entire life.

Q. Is it possible that the pigmentation came on some time in January 2011?

A. No."

31 Mr Coote was less unequivocal. He said it had been grey and was going on black "between September and probably October, through those first three or four cryotherapies that we had" and thereafter black until finally excised and diagnosed. He said that it was not possible that he was mistaken about the colour when he saw Dr Wall in September 2010. In light of the fact that Mr Coote in part relied upon his wife's description, and that ultimately the primary judge placed decisive weight on Mrs Coote's evidence, it is not necessary to summarise Mr Coote's evidence in any greater detail.

The reasoning of the primary judge

32 Her Honour found that Mr Coote suffered from both a plantar wart and a melanoma, having noted the agreement between the experts that that was possible. Her Honour's reasons for concluding a black spot was present in September 2009 occupy [63]-[103] of her judgment.

33 The starting point was correct. It was not possible to say from what was known of the strongly pigmented lesion in March 2011 how it presented in September 2009, although there was evidence that it would have been surprising if there was no pigmentation: at [68]. Her Honour noted the diametrically opposed evidence of Mr and Mrs Coote, on the one hand, and Drs Kelly, Wall and Hiddins on the other, and accepted Dr Kelly's notes as a contemporaneous record of at least part of what he had observed: at [69]-[73]. Her Honour correctly stated that what was to be assessed was not merely the credibility but the reliability of the divergent recollections of the witnesses. She noted that all had an interest in the outcome, that the doctors had been provided with Mr Coote's statement prior to their giving evidence, and that all made concessions when warranted.

34 Importantly for present purposes, her Honour conflated the assessment of the reliability of Dr Wall's and Dr Hiddins' recollections (at [82]):

"Dr Wall and Dr Hiddins gave evidence that they had a recollection of their consultations with Mr Coote and the state of the lesion, which Dr Kelly said he did not have. This evidence must be considered in the face of Dr Lynch and Dr Jammal's doubts that a general practitioner could have a reliable recall of a particular patient's lesion at a particular time, especially with the lapse of time after treatment. It seems to me that while it is possible that their recall is entirely reliable, it is unlikely, having in mind the lapse of time, the number of patients they have each treated since seeing Mr Coote and the number of lesions which they must have seen over that period."

Dr Lynch and Dr Jammal had given evidence that it was "preposterous" to assume a practitioner could remember a lesion from a time ago "without the aid of properly made notes".

35 Her Honour also observed that although the doctors' notes were contemporaneous records, all save the first "shed but little light on the presentation of the lesion after 3 September": at [83]. Her Honour then said (at [84], emphasis added)

"Given the controversy, it is necessary to consider all of this evidence, as well as giving close attention to the evidence which Mr and Mrs Coote gave as to what could be seen on Mr Coote's foot."

36 The following 19 paragraphs contain the balance of the reasoning supportive of her conclusion that Mrs Coote's recollection was to be preferred. They contain three headings: "Mr Coote", "Mrs Coote" and "Conclusion". The first two sections deal exclusively with the evidence of Mr and Mrs Coote respectively. The reasoning in the conclusion does not return at all to the conflicting evidence of Drs Kelly, Wall and Hiddins. Nor does it return to those doctors' notes. Reference is made only to one aspect of the evidence of Dr Hiddins: her evidence that the appearance of a second lesion in February 2011 admitted of the possibility that the melanoma appeared only then, which it was common ground was unlikely given its advanced stage when excised the following month.

37 Her Honour concluded her reasoning on breach at [102]-[103]:

"Having carefully considered all of the evidence, I am satisfied that it has been shown that it was probable that Mr Coote was suffering an ALM when Dr Kelly first saw him in 2009. I have concluded that Mr and Mrs Coote's evidence cannot be rejected as being unreliable, even though the presence of the small black spot which she described was not noted by Dr Kelly, who candidly admitted that he could not recall the lesion and that there was a possibility that he had made an error.

The acceptance of Mrs Coote's evidence must lead to the conclusion that Dr Kelly did err, even though I am entirely confident that his error was not deliberate and that what he was confronted with was extremely unusual, given that Mr Coote was suffering from both an ALM and a plantar wart. On the expert evidence, this lesion was difficult to diagnose. Nevertheless, I am of the view that the black spot which Mrs Coote described was present when Mr Coote saw Dr Kelly and had to be investigated."

38 There are two ways to read these 41 paragraphs of the judgment. The first is that the dispositive reasoning is in the last 20 paragraphs (from [84]-[103]), in which case there is no engagement whatsoever with the conflicting testimonial and documentary evidence of the doctors, such that her Honour considered that the only question was whether Mrs Coote's evidence was reliable. Although that is a natural reading of [84], I favour an alternative reading, which is that [82] and [83] are to be read as reasons which enabled her Honour thereafter to put entirely to one side the doctors' testimonial and documentary evidence. On that reading, the doctors' testimonial evidence was put to one side because her Honour formed the view, that it was likely to be unreliable (see [82]), and their notes were put to one side because they "shed but little light" on the lesion's presentation on 3 September 2009 (see [83]). That approach conflated the testimonial evidence of Drs Wall and Hiddins, and paid no attention to the appearance of the lesion after 3 September 2009.

The process of fact finding miscarried

39 With respect to the primary judge, who heard and determined in urgent circumstances what she candidly described as "an exceptionally difficult case", that reasoning cannot be permitted to stand, whichever way it is to be read. It is not that there was a failure to make requisite findings of fact, but rather a failure in the process of fact finding as disclosed by the reasons: see Waterways Authority v Fitzgibbon [2005] HCA 57; 79 ALJR 1816 at [129]-[130] (Hayne J). It is necessary to "engage with, or grapple or wrestle with the cases presented by each party": see Mitchell v Cullingral Pty Ltd [2012] NSWCA 389 at [116] and MM Constructions (Aust) Pty Ltd v Port Stephens Council [2012] NSWCA 417 at [134]. As explained by McColl JA in Pollard v RRR Corporation Pty Ltd [2009] NSWCA 110 at [66]:

"Where it is apparent from a judgment that no analysis was made of evidence competing with evidence apparently accepted and no explanation is given in the judgment for rejecting it, it is apparent that the process of fact finding miscarried."

40 This litigation was conducted in such a way that the persons who are entitled to feel a sense of grievance if the cases presented by the parties were not adequately addressed are not merely the litigants, but also Drs Wall and Hiddins. Although both were witnesses, and not parties, her Honour's findings amount, implicitly, to findings of serious breaches of duty by both practitioners. No less serious to Drs Wall and Hiddins is her Honour's rejection of their claimed actual recollection of Mr Coote's consultations with them. As Hoeben J said, with the agreement of Allsop P and Beazley JA, in Hawkesbury District Health Service Ltd v Chaker [2010] NSWCA 320 at [56]:

"Implicit in his Honour's complete rejection of Dr Tompsett's evidence was the proposition that the diagram in the respondent's clinical notes with its markings was not genuine, or had been placed there subsequently by Dr Tompsett. The same inference is implicit in relation to his Honour's rejection of the evidence concerning the booklet. These are very serious findings about any witness, here a medical professional. Before findings of that kind could be made, it was incumbent upon his Honour to set out clearly and logically the factual basis for them which inevitably would have involved a proper analysis of the evidence. This was not done. Accordingly, there was no proper factual basis for his Honour's acceptance of Dr Malouf's opinion."

41 I turn to how the process of fact finding as disclosed by the reasons miscarried.

42 First, there was her Honour's dismissal of the evidence of Drs Wall and Hiddins as unreliable. Although both doctors claimed to recall the consultations distinctly, and both squarely denied there was the black spot to which Mr and Mrs Coote deposed, in crucial respects their evidence was entirely unlike. Dr Wall had seen Mr Coote once, almost 18 months earlier, when Dr Kelly was not available. His notes were exiguous. And he gave evidence, which it was open to the primary judge to treat as unreliable, that he recollected a reasonably extensive conversation with Mr Coote, following which Mr Coote declined to follow his advice. On the other hand, Dr Hiddins was giving evidence about the first time she had ever seen a melanoma, presenting in a patient she saw six times in the first three months of 2011. She inspected his foot with a large magnifying glass with a fluorescent light around it. She made relatively detailed notes, which corroborate her recollection of a lesion which changed dramatically in that period.

43 The difference between those doctors' testimonial evidence was reflected in their cross-examination. It was squarely put that Dr Wall was lying when he said he had an actual recollection of the single occasion he had treated Mr Coote. The challenge to Dr Hiddins' recollection was, understandably, much milder.

44 Those differences needed to be addressed in order to engage with the "real strength" of the evidence presented by the defendant; cf Earthline at 321. It was not possible properly to discount Dr Hiddins' evidence in the same way and in the same sentence as Dr Wall's evidence.

45 Further, in dismissing the testimonial evidence of those doctors as unreliable, her Honour relied on the evidence of Drs Lynch and Jammal. Putting to one side the admissibility and weight of that evidence (as to which I do not express a view), that evidence, even taken at its highest, was not relevant to the recollection of Dr Hiddins. Their evidence was premised upon an inability to recall details of a lesion where a practitioner was unassisted by properly made notes. That opinion was inapplicable to Dr Hiddins' recollection.

46 Her Honour's findings amount, albeit implicitly, to a wholesale rejection of the testimonial evidence of Dr Wall and Dr Hiddins. It meant not only that Dr Wall and Dr Hiddins missed something which was obvious, but also either that they were lying when they said they had an actual recollection of the consultations, or at best were mistaken in their understanding of what an actual recollection (as opposed to a reconstruction) involves. It was open, having regard to the cross-examination, for her Honour to make such findings, but only if the reasoning process supporting it were exposed. One unpleasant aspect of curial adjudication is that on occasion, such findings need to be made. But they should be made expressly, not implicitly, and attended by a transparent and comprehensible reasoning process based on findings of fact reflective of the gravity of the court's conclusion.

47 Secondly, it was necessary to have regard to the doctors' notes. Although it is true that the later notes say little about what presented to Dr Kelly on 3 September 2009, the case advanced by the plaintiff was that there was always a visible black spot. The notes bore upon that central issue directly. Although there was occasion for recording in the notes the fact that the lesion was pigmented or black, if indeed it was, there was no occasion for recording in the doctors' notes the fact that the lesion was flesh-coloured.

48 The notes reflected three trained practitioners' observations, each of whom diagnosed and treated a plantar wart, and were aware of the significance of a pigmented lesion. Doctors Kelly and Wall maintained that they would have made a record of the fact that it was pigmented if they had seen it. It was possible (although the evidence suggested this was unlikely) that the consequence of the cryotherapy administered at the previous session would discolour the skin two weeks later, which would undermine Dr Kelly's observations on the second, third and fourth consultations, but that could not explain the final consultation with Dr Kelly on 20 May 2010, seven months since Mr Coote's previous visit, nor the consultation with Dr Wall, nor the first consultation with Dr Hiddins.

49 In short, the reasons did not properly engage at all with the remarkable fact that every contemporaneous record of Mr Coote's left foot was inconsistent with the testimonial evidence of Mrs Coote which the primary judge accepted. Once again, it was open to the primary judge to accept Mrs Coote's evidence in this respect, but only if first her Honour explained why those notes (particularly those made on 20 May 2010 by Dr Kelly, on 2 September 2010 by Dr Wall, and in January 2011 by Dr Hiddins) were to be discounted.

50 Thirdly, the finding made by the primary judge that there was a plantar wart was inconsistent with the evidence of Mrs Coote, who maintained that the "piece of black tar" looked "nothing like a plantar wart". This does not merely involve an issue which detracts from the reliability of Mrs Coote's evidence. Instead, it goes to the heart of the case on breach. The last two sentences at [103] - the primary judge's conclusion on breach - cannot be reconciled with each other. The finding made by the primary judge of the unlikely but possible combination of a plantar wart and melanoma was, on the expert evidence, "difficult to diagnose". That describes something which is altogether different from the black spot which Mrs Coote said she recalled seeing. There was nothing to suggest any difficulty of diagnosing melanoma if Mr Coote's foot were as Mrs Coote had described and as her Honour found. Her Honour's reasons do not engage with, let alone resolve, the inconsistency.

51 Memory is all too fallible. McHugh J referred to "the everyday experience of the courts that honest witnesses are frequently in error about the details of events": M v The Queen [1994] HCA 63; 181 CLR 487 at 534. The process of conscious and subconscious reconstruction of what was actually said in a conversation, to which McLelland CJ in Eq referred in Watson v Foxman (1995) 49 NSWLR 315 at 319, is familiar. Although elaborate submissions were made about this at trial, and noted by her Honour noted at [79], her Honour's reasons do not explain why she came to the view that Mrs Coote's evidence was not unreliable. As her Honour noted, Mr and Mrs Coote were shown to have the same imperfect recollection of dates as most witnesses. There is no necessary reason for their visual recollection to be any less imperfect. In a case such as this where even the photographs of the final stages of Mr Coote's disease are distressing, and must have been much more so to Mrs Coote, the possible unreliability of her honest recollection needed to be engaged with in light of the balance of the evidence.

52 The effect of those matters is that the finding of breach cannot stand. However, it is not possible for this Court to find that there was no breach. Although I have been critical of aspects of the reasoning on breach above, it was open on the evidence to find that Mr Coote presented on 3 September 2009 in accordance with Mrs Coote's recollection. Her Honour was plainly impressed by Mrs Coote as a witness, and enjoyed advantages which this Court does not. The determination of the issue of breach was, to use the language of Tobias AJA, "a complex intellectual process involving the interaction of documentary material, elements of testimony from different witnesses and matters of emphasis none of which readily appear from reading a transcript": Vagg v McPhee [2013] NSWCA 29 at [85]. This appeal is quite clearly not a case like Fox v Percy where the relevant facts are incontrovertibly established; it is the opposite of such a case.

Causation

53 Only if it appears to the Court that some substantial wrong or miscarriage has been occasioned may a retrial be ordered: UCPR r 51.53. If the finding of no causation could not be impugned, then a retrial would not be necessary. However, just as there is force in the respondent's submissions on the finding of breach, so too is there force in the appellant's submissions impugning the finding of no causation.

54 The question was whether the failure to diagnose and investigate a melanoma in September 2009 caused "particular harm" in accordance with s 5D(1) of the Civil Liability Act 2002 (NSW). The particular harm alleged was metastasis of the melanoma. Metastasis had occurred prior to its excision on 26 March 2011, but just as it was not possible to say whether the melanoma was pigmented in September 2009 from its appearance in 2011, so too it was not possible to conclude whether it had metastasized by September 2009.

55 There is a correlation between survival after diagnosis with melanoma and the thickness of the tumour. For this purpose, the "Breslow" thickness of a melanoma is used, and the evidence was that thickness was the best prognostic indicator for melanoma. Breslow thickness can only be measured accurately with a microscope after excision. Where as here the melanoma is not ulcerated, its Breslow thickness is measured from the top of the granular cell layer of the epidermis to the base of the melanoma in the dermis or subcutaneous tissue. When Mr Coote's lesion was excised, it was found to have a Breslow thickness of 4.4 mm.

56 This is not an area of medical science where the causal mechanisms are entirely unknown. The evidence was that the thickness of the tumour was indicative of two things: first, the extent to which it came into contact with the lower layers of the skin, and (especially) the subcutaneous tissue, which would provide opportunities for metastasis, and secondly, the fact that the lesion had increased to that extent was itself an indication that its cells were more likely to metastasize.

57 Her Honour concluded that there was a "high likelihood" that the melanoma had metastasized even before Dr Kelly saw it, because it was "then 4 mm thick or more and that it was growing quickly, before any treatment was sought or given" (at [169]). Her Honour's reasons are contained in [164]:

"Having carefully considered all of the evidence as to the thickness of this lesion in 2009, I am not satisfied that it has been shown to be probable that the ALM was only 2 mm thick in 2009, when Dr Kelly first saw Mr Coote. Further, it has not been shown to be probable that this ALM had not already metastasised, at that time. It appears probable that even then, it had a thickness of 4 mm, or more. Given the rapid growth of the lesion to 3 September 2009, a growth consistent with Mr and Mrs Coote's description of continued growth between September 2009 and March 2010, estimated by Mrs Coote then to be about 17 mm in diameter and 20 mm by January 2011; its thickness in September 2009 and in 2011; when considered together with the metastases detected after excision in 2011, the first time that metastases was investigated, it is not possible to conclude that it is probable that the ALM had not already metastasised in September 2009."

58 I agree with the appellant that those sentences are to be read as conveying a finding that the tumour's Breslow thickness was 4 mm or more in September 2009. Critical to that reasoning were two things. The first was a reliance on Mr Coote's evidence that in 2009 it was 1-2 mm high, and the evidence that a melanoma generally extends as far below the skin as it does above. The appellant pointed out that that was difficult to reconcile with the evidence of Mrs Coote, which her Honour had previously found to be the most reliable, that it was raised by 1 mm. Why accept Mrs Coote's evidence as the best evidence of the black spot, yet apply Mr Coote's evidence of its height? Again, I am not saying that the finding was unavailable, but there is no reasoning process indicating why in this critical respect Mrs Coote's evidence has been discounted.

59 The second was her Honour's finding that the lesion had been picked at and repeatedly pared and pumiced and treated with cryotherapy (at [151]). Her Honour stated that had the melanoma been unpared, it could have been thicker than 4.4 mm (at [152]). That reflected the evidence of Professor Levi, who said that although debridement and pumicing the lesion would reduce the thickness of the lesion, "the amount of change couldn't be expected to be more than 0.1 to 0.2 mm, in my opinion". Her Honour recorded that evidence at [156]. Her Honour also said at [152] that "As the experts accepted, it follows that unpared, it could then have been thicker than 4.4 mm". Save for Professor Levi's evidence that that could make a difference of 0.1 to 0.2 mm, neither party was able to direct the Court to evidence supportive of that finding. As I read the evidence, it does not support the reasoning process to the effect that this was at all times a tumour whose Breslow thickness exceeded 4 mm, whose growth from September 2009 until March 2011 was kept in check by paring and pumicing. The paring and pumicing could make only a very small difference.

60 Her Honour found that the tumour was fast growing in 2009, and at that time probably had a Breslow thickness of 4 mm or more. Once the evidence as to the limited effects of paring and pumicing is borne in mind, it is very difficult to reconcile those findings with what was established in 2011 upon excision, namely, a Breslow thickness of 4.4 mm.

61 Finally, the respondent submitted that it was not possible for a finding of causation to have been reached based on the statistical data correlating life expectancy with Breslow thickness. The respondent said that the data had been collected for a different purpose: prognosis once a melanoma had been diagnosed and excised. That is correct, but the respondent went further and submitted that the data could not be used to support a conclusion that any given melanoma had or had not metastasized by a particular time ("It is not a valid use of these studies to establish when metastasis occurs").

62 There is a difficulty with that submission, because it is on its face inconsistent with the reasoning process of the primary judge as to the absence of causation which resulted in a verdict for the defendant. But in any event, I do not accept the submission.

63 The plaintiff had to show that Dr Kelly's breach of duty had caused "particular harm" within the meaning of s 5D(1). The particular harm on which the plaintiff's case was based was metastasis. It is known that the melanoma had metastasized before 26 March 2011. It would be sufficient, if the plaintiff could prove, to the civil standard, that it had not metastasized prior to May 2010. The burden of proof remained with him: s 5E. That forensic task was not without its difficulties, but it was not impossible. The statistical information demonstrates that melanomas with low Breslow thickness are unlikely to have metastasized (either regionally (which is Stage III of the disease) or distantly (which is Stage IV)). Of course, it is possible that even a thin melanoma has metastasized, but the plaintiff needed only to show on the balance of probabilities that his had not.

64 There is a further question as to whether there is a rational process by which one can infer Breslow thickness from how the lesion felt. Professor Levi was asked, twice, to express a view on that question, and said he could not do so. But it is at least reasonably arguable that there is some correlative relationship between the apparent depth of a melanoma and its microscopically measured Breslow thickness, and so the recollection of Mrs and Mr Coote was not incapable of supporting a finding that more probably than not his melanoma, had it been diagnosed and excised in September 2009, would have had a Breslow thickness of no more than, say, 2 mm. There was ample evidence of high survival rates for patients in those circumstances. Of the 3,285 patients in the United States study whose melanomas were unulcerated and of Breslow thickness between 1 and 2 mm, almost 90% were alive after five years and almost 80% were alive ten years later. Since the principal means by which melanoma kills its victims is metastasis, the same data would support the conclusion that more probably than not, the tumour had not metastasized by September 2009. There is nothing in Tabet v Gett [2010] HCA 12; 240 CLR 537 standing in the way of such a conclusion. The present is, emphatically, not a case where the doctor's negligence, if made out, left unaltered the risk of the patient suffering harm; cf Paul v Cooke [2013] NSWCA 311.

65 It is not necessary for present purposes to address other aspects of her Honour's reasoning, and in light of what is to occur in the future, it is inappropriate to do so. What has already been said demonstrates that her Honour's findings as to causation favourable to Dr Kelly are not such that the plaintiff must inevitably fail. It follows that there must be a retrial, notwithstanding its expense and inconvenience. That is the price accepted by our legal system for assuring to every litigant a manifestly fair trial, determined not only upon often fallible judicial impression but on the critical evidence adduced: Kirkwood v New South Wales Insurance Ministerial Corporation [1995] NSWCA 244 at 13 (Kirby P). It will be open to the parties, of course, to determine how that further hearing may be conducted most efficiently, in accordance with their obligations under s 56 of the Civil Procedure Act 2005 (NSW).

66 I propose that the appeal be allowed, the judgment and orders of the Court made on 14 March 2012 be set aside, and there be a new trial. Because both parties have enjoyed success on the appeal, there should be no order as to the costs of the appeal, with the intent that they each bear their own costs. The question of the costs of the proceedings at first instance should be left to the judge before whom the new trial is to be conducted.

**********

Amendments

30 October 2013 - Order amended pursuant to UCPR r36.17 to replace "14 March 2013" with "14 March 2012".
Amended paragraphs: Coversheet

Decision last updated: 30 October 2013