Estate of Sheppard v Accident Compensation Corporation

IN THE DISTRICT COURT AT WELLINGTON [2013] NZACC 117 UNDER The Accident Compensation Act 2001 IN THE MATTER OF an appeal pursuant to section 149 of the Act (Appeal No. ACR 416/11) BETWEEN ESTATE OF IAN SHEPPARD Appellant AND ACCIDENT COMPENSATION CORPORATION Respondent Hearing: 12 March 2013 Appearances: Mr J Miller for appellant Mr A D Barnett for respondent Judgment: 24 April 2013 RESERVED JUDGMENT OF JUDGE D A ONGLEY [1] The appeal concerns a failure to adequately treat a melanoma which metastasised and caused the death of Mr Sheppard. A treatment failure is not in dispute because it has been well establish by expert opinion and by findings of the Heath and Disability Commissioner. The question in the appeal is the extent of the failure and whether it caused an injury. The respondent contends that correct treatment would not have materially altered the outcome. [2] The injury in question was untimely death, that is to say death earlier than would have occurred if the appellant had received proper treatment. The answer depends on probabilities, because it cannot be conclusively proved that death would not otherwise have occurred, or that life would have been significantly prolonged [3] Treatment injury is not covered where the personal injury is wholly or substantially caused by a person's underlying health condition. The appeal addresses the question whether earlier death was wholly or substantially caused by the cancer and only marginally by the failure to treat. [4] Mr Sheppard had an excision of a malignant melanoma on 24 September 2008. His general practitioner failed to arrange follow up treatment to detect possible micrometastasis. [5] The undisputed facts are substantially taken from the report of the Heath and Disability Commissioner issued on 23 April 2012. [6] In June 2006, general practitioner Dr Nicolaas Burnett completed a medical examination for lan Sheppard (then aged 45 years), as part of Mr Sheppard's application for New Zealand residency. For an Immigration Medical the patient is required to disclose any previous history of cancer including melanoma. Dr Burnett noted Mr Sheppard's relevant medical history, which included the removal of a melanoma in 1990. Mr Sheppard later enrolled as a regular patient at the practice where Dr Burnett worked, Thames Medical Centre. [7] In September 2008, Dr Burnett excised a mole from Mr Sheppard's back, and the specimen was sent to Pathlab Waikato Lid. The histology report, completed by pathologist Dr Kevin Dowden, confirmed that it was a melanoma and stated that it had been "completely excised". Dr Burnett was aware that wider excision was usually recommended for melanomas, and he telephoned Dr Dowden to confirm the result. He was advised again that the excision was complete, and recorded also that no further surgery was required. A few days later, Dr Burnett saw Mr Sheppard to remove the sutures and discuss the histology. No other follow-up was arranged. Dr Burnett said that he had been advised by Dr Dowden that no further follow-up was required. [8] In the appeal it is common ground that the outcome would probably not have been materially altered by a wider excision. The question is whether a proper follow-up would have significantly improved the outcome for Mr Sheppard. [9] In September 2009, Mr Sheppard went to see Dr Burnett about a lump in his left armpit. Dr Burnett referred Mr Sheppard to general surgeon Dr Chandra Reddy. There was no reference to Mr Sheppard's melanoma history in the referral letter. Ten days later, Dr Reddy reviewed Mr Sheppard and sent him for a fine needle aspirate test. The cytology was reported by Dr Dowden. It was noted that the sample was of "low cellularity" but that there was no sign of any malignant cells and that features were consistent with a lipoma (ie, benign). Dr Reddy arranged to see Mr Sheppard for a surgical excision of the lump on Mr Sheppard's return from a trip to South Africa. [10] When Dr Reddy reviewed Mr Sheppard in December 2009, the lump was firmer and Dr Reddy thought that it might be a lymph node. The lump was removed under general anaesthetic thirteen days later. The biopsy confirmed that it was malignant, and further tests revealed that Mr Sheppard had Stage IV malignant metastatic melanoma with brain, liver and lung metastases. Mr Sheppard died on 2 July 2010. Clinical Practice Guidelines [11] The Heath and Disability Commissioner recorded that the Clinical Practice Guidelines for the Management of Melanoma in Australia and New Zealand (Melanoma Guidelines) were issued by the New Zealand Guidelines Group and the Australian Cancer Network in November 2008. The Melanoma Guidelines reflect the accepted treatment and management of melanoma during the time of the events covered in the investigation. Accepted practice included a clear emphasis on early diagnosis and a rigorous application of appropriate treatment. Other key aspects of accepted practice at the time were listed by the Commissioner as follows: "When a doctor identifies a lesion suspicious for melanoma, usual practice is to excise it with a narrow margin of normal looking skin The specimen is sent for histology. Histological examination will determine whether the lesion is a melanoma and, if so, provide necessary information to guide further management. Relevant information includes the 'level' of the melanoma, that is, the depth to which the melanoma cells have grown into the skin. The level is measured in millimetres and referred to as the 'Breslow' thickness. The 'stage' of a melanoma gives an indication as to how far the melanoma has spread. Stages I and II are confined to the skin, Stage III to the lymph nodes, and Stage IV indicates spread to internal organs. Other details may be included in what is known as a 'synoptic report'. Once a primary melanoma is confirmed, it is usual practice in most cases to undertake wider excision. The recommended excision margins are based on the maximum Breslow thickness and other prognostic features. Patients who have had a melanoma excised should be followed up at regular intervals. The main purpose of follow-up is to detect any recurrences or new suspicious skin lesions early, so that early treatment can be undertaken. The frequency and.duration of follow-up is dependent on how advanced the disease is at the time of presentation." [12] In the cover claim lodged on 26 April 2010, the appellant's general practitioner Dr Titchener noted that this was not this patient's first melanoma, which changed his risk stratification. He stated that an additional excision should have been done, and that there was also the need for surveillance and follow up by a specialist. He said that dermatology recommended follow-up every three months for the first year then every six months for the next two years. Medical reports and opinions Dr Beer, histopathologist [13] ACC obtained a report from Dr Beer, histopathologist of Pathlab. The pathologist who had examined the specimen had left New Zealand and comment from him could not be obtained. [14] Dr Beer considered that the histology report should have resulted in surgical review because the excision margins were less than 10mm and the Breslow thickness was 1.6mm. In his letter o 6 7 2010, he stated that: "There is close correlation between melanoma thickness and occult metastases in sentinel lymph nodes. This lesion (08/A4110231) was accurately measured as having a Breslow thickness of 1.6mm and actuarial studies indicate that this imparted a 50% risk of subsequent regional lymph node metastasis within 3 years. As can be surmised from the patients history, by the time of his skin excision in October 2008 the [die] was cast as this melanoma had already metastasised to the auxiliary lymph node which had taken a year to be clinically palpable. Meanwhile, blood born metastasises to liver, lung and brain had developed. This is in keeping with the natural biology of these lesions." [15] Dr Beer's reasoning was not fully explained. He referred to statistical information that predicted a 51% chance of regional lymph node metastasis. The summary did not explain whether the assessed risk distinguished between cases with or without followup treatment and Dr Beer did not explain how he "surmised" that metastasis had occurred by October 2008. Nevertheless this statement went unchallenged by other medical advisors, except to say that it could only have been made after the event. The respondent relies on Dr Beer's opinion as an argument that follow-up, while it should have been done, was unlikely to have affected the outcome. [16] Dr Beer wrote again on 12 August 2010 and said: "Melanomas of this profile (Breslow thickness between 1.01 and 2mm) have a 65% 10 year survival rate. This means that 35% of patients die within 10 years. The three year actuarial incidence of subsequent regional metastases in patients initially treated with wide local excision of melanomas between 1.50mm and 3.99mm was 51%. Sentinel lymph node biopsy is recommended for all melanomas measuring 1mm or more in thickness. Identifying sentinel lymph nodes in lesions arising from the back may be problematic in that metastases may occur in either axilla or either groin." [17] Perhaps this sheds some light on Dr Beer's reference to the patient's history. Events proved that he was not within the 65% group with a 10 year survival rate Mr Martin, general surgeon [18] The assertion that the die was cast is contentious in this appeal. The review and appeal were brought by Mr Sheppard's widow Mrs Helga Sheppard. Mrs Sheppard obtained expert advice from Mr Richard Martin, a general surgeon with an FRACS and Master of Surgery in Surgical Oncology, specialising at the Sydney Melanoma Unit, which is the largest melanoma treatment and research facility in the world. Mr Martin currently practiced at Waitemata District Health Board and at the Melanoma Unit based in Auckland. In a letter of 30 June 2011, He advised: "My standard practice upon receipt of the excision biopsy that confirmed a 1.6 mm melanoma: I would have fully assessed the patient and this would have included a history of any previous skin cancers and family history of melanoma and general medical conditions, then a full clinical examination paying particular attention to the primary site and the draining lymph node fields would be undertaken. My recommended treatment would have been a wide excision with a 2cm margin of the primary site and a sentinel node biopsy. This would have detected micrometastases in the sentinel node, subsequently presenting as the palpable in the left axilla. Once a positive sentinel node is identified, I perform a staging CT scan prior to a completion lymphadenectomy. This is considered the standard of care in melanoma management around the world unless in a clinical trial. Mr Sheppard's prognosis at diagnosis was 91% (90.1 % - 92.8 %) 5-year survival; a positive sentinel node would decrease his prognosis to approximately 70-80%. However, the number and size of nodes impacts on prognosis and this could have been further estimated once a completion lymphadenectomy and pathology report was available for review. I do not believe that Mr Sheppard received the standard of care expected for a patient diagnosed with an intermediate thickness melanoma. While the data of sentinel node biopsy is still maturing, patients who have a positive sentinel node have a 20% absolute survival benefit compared to patients who present later with palpable disease and undergo treatment (this difference is seen again in the AJCC Stage III spectrum) Although I am reviewing this case retrospectively, it is impossible to tell whether early intervention would have ultimately affected the outcome. However, Mr Sheppard was denied this opportunity. The comment from the Pathologist that 'the die was cast' can only be said with retrospect. It is my expert opinion that upon receipt of an excision biopsy confirming melanoma, referral should have been made to a melanoma specialist. I do not believe it is the role of the pathologist to recommend clinical management. Depending on Mr Sheppard's staging and treatment he would have been under close observation for either recurrent melanoma or new primaries." [19] Significantly, Mr Martin said that it was impossible to tell whether early intervention would ultimately have affected the outcome. He provided a prognosis at diagnosis of about 91% 5 year survival, unless with a positive sentinel node biopsy, which would reduce the prognosis to 70%. [20] The survival advantage with detection and treatment of lymph node metastasis is a central question in the appeal. The Heath and Disability Commissioner made these findings concerning the width of excision and sentinel node biopsies: "My understanding is that wider excision of primary melanoma is to reduce the incidence of local recurrence (i.e. recurrence adjacent to the site of excision) and that it is unlikely that this makes any difference to the rate of metastasis, which is the main determinant of long term survival. I also note, from the Guidelines document, that there are two randomised controlled trials that show no evidence that a margin greater than I cm offers any survival advantage. From the same document, it is also noted that systematic review indicates there are inadequate data to confine a mortality difference between wider and narrower excision for primary invasive melanoma. In this case, the appearance of metastasis without any evidence of local recurrence would support the hypothesis that wider excision would have been unlikely to have altered Mr Sheppard's prognosis for survival. The Guidelines also recommend that sentinel node biopsies should be discussed with patients who have a primary tumour 1.2 to 3.5mm thick. In the Guidelines document, the multicenter selective lymphadenectomy trial has not demonstrated an overall survival advantage for patients undergoing sentinel node biopsies. Therefore I am unable to assert that a sentinel node biopsy performed on Mr Sheppard would have improved his prognosis." Dr Holland, general practitioner [21] The Corporation obtained independent advice from Dr Neil Holland, general practitioner and Dr David Roche, pathologist. [22] Dr Holland reported on 24 August 2010, and indicated that in his view, Dr Burnett should not have relied on the advice of the pathologist alone to recommend no further treatment. He said that usual practice would be to make that kind of decision in conjunction with a general surgeon or plastic surgeon. He advised this was a failure to treat. While the initial excision by Dr Burnett had been appropriately done with the correct diagnosis, standard practice would have been to refer to a specialist surgeon for further management. He also indicated that the deceased's history of melanoma should have alerted Dr Burnett to a higher risk level with a need for closer surveillance. He said that the risk of the cancer spreading was probably increased by the deceased not being offered a wider excision. [23] Dr Holland concluded that there was an "inevitable degree of uncertainty as to whether adherence to the guidelines for the management of this melanoma would have influenced the outcome. Firstly, there is the uncertainty as to whether he died from late metastases of his first melanoma. Secondly, even with the closest of follow up melanoma has a high inherent risk of spreading." Dr Roche, pathologist [24] Dr Roche provided a report on 21 September 2010. He thought that the excision may have been adequate, there having been no local recurrence of the melanoma. It was unlikely that there was un-excised tumour left behind in the scar. However he advised that follow-up was indicated: "Even though the guidelines indicate that this melanoma ought to have had further management i.e. wider excision, follow up examinations, consideration of sentinel lymph node biopsy, there is no good evidence that these would have provided much, if any, extra survival benefit. ... The false negative auxiliary lymph node biopsy similarly would not have altered the prognosis, as the melanoma had already become metastatic by that stage. SLNB, wider excision, and closer follow up may help in controlling local and regional recurrence, but have not been shown to improve survival even according to the latest guidelines. Summary: It is quite likely that even with wider excision at the time, and earlier detection of the auxiliary metastases that the fatal brain metastases still may have occurred. Therefore there is no evidence to support the claim that injury was caused by failure to provide treatment." [25] In summary, Dr Roche said: 'In summary, wider excision, closer followup, and specialist referral for consideration of sentinel lymph node biopsy etc should have occurred. Had these been done, it may have allowed earlier detection of metastasis but there is not strong evidence to indicate that this would have altered the final prognosis. The depth of the melanoma and the mitotic rate alone strongly dictate the prognosis, and on average, one in four people with melanoma with the measurements of this one will die from the disease. There is no evidence to support the claim that following a different treatment path would have altered the progression and final outcome." [26] The Corporation's Complex Claim Panel considered the available evidence and, on 14 October 2010, recommended declining the claim. They offered only a very short opinion relying on the point that there was no local recurrence from the excision and there had been normal disease progress. On 2 November 2010 the Corporation declined the deceased's claim. The review [27] The decision was taken to review and Reviewer Ms L A Clark issued a decision on 31 May 2011. Mrs Sheppard appeared at the review and maintained that Dr Burnett should have referred her husband for specialist follow-up when he removed the mole in September 2008. The Reviewer decided that the test for causation was not met. The evidence of Drs Roche and Beer supported the view that the lack of follow up did not influence the progression of the disease. The Reviewer accepted the probability that Mr Sheppard's disease had already progressed by the time his initial lesion was removed in September 2008. Any failure to provide follow-up treatment could not be said to have caused the progression of the melanoma. The review was therefore dismissed. [28] Mrs Sheppard obtained Mr Martin's opinion after the review decision and the question is reargued with that additional evidence. The District Court is required to come to its own conclusion on its assessment and evaluation of the evidence, and the merits generally: Wildbore v ACC [2009] NZCA 34 (CA) [29] Mr Martin wrote again on 18 February 2013, as follows: "The issue is not that a wide excision would have affected Mr Sheppard's outcome, but more importantly a referral to a Melanoma Specialist would have instigated not only a wide excision, but a sentinel node biopsy (SNB). This would have identified the micrometastasis (at that time) in his axilla and then mandated further treatment based on that. The only randomised control trial to date for SNB, MSLTI, has shown that there is at least a - 20% survival benefit - for detection of micrometastases in sentinel nodes vs waiting for a macrometastases in the 'control arm,' (See references) I totally disagree with the statement that statistics mean nothing and reference to the Ambros case. Evidence based medicine is the only way to practice surgery. It is the basis by which we make all of our decisions and provide prognosis for patients. As previously stated a wide excision does not affect the prognosis of a patient but referral to a specialist and sentinel node biopsy or further treatment may well have affected the outcome in this case. As recommended by the Clinical Practice Guidelines for the Management of Melanoma in Australia and New Zealand, Chapter 12 (see references). If one looks at the prognosis for different stage III melanoma (metastasised to lymph node) one can see quite large differences in prognosis and outlook. (See references). Had Mr Sheppard's melanoma metastasis been detected in the lymph node while small he may have had a different outcome. It is my expert opinion that a medical error occurred by not referring Mr Sheppard to a Melanoma Specialist for further assessment and management. He was denied the opportunity of potential cure. Clinicians must rely on statistics and evidence based medicine in order to practice Gold Standard treatment." The law [30] Mr Martin referred to ACC v Ambros [2007] NZCA 204. In that case, the Court of Appeal reaffirmed the need to prove actual causation: see Atkinson v ARCIC [2002] 1 NZLR 374 (CA). The relevance to the present appeal is that there is no clear proof that failure to follow up would have made a difference by at least slowing the course of the disease. The Court in Ambros considered various authorities from other jurisdictions, including observations in McGhee v National Coal Board [1973] 1 WLR 1, and said: [32] To the extent that McGhee treats the increase in risk as equivalent to a material contribution to injury as a rule of law, it must be seen as contrary to Atkinson and thus not applicable in New Zealand insofar as the assessment of causation under the accident compensation regime is concerned. However, it may still have relevance if it is seen as a case where the Court was prepared to draw a robust inference of causation. As Khoury says at 214, material increase of risk does in fact fit squarely within inferential reasoning, which typically infers from a known fact (e.g. the increase in risk) the existence of an unknown fact (e.g. material contribution). Whether the inference will be drawn would, however, depend on the totality of the circumstances. Professor Stapleton makes a similar point. She posits the situation where a particular surgical error more than doubles the risk of post- operative cardiac arrest. Where the error is made and post-operative arrest occurs, then, in her view, a court would be entitled to infer causation because the error has added more than the existing risk. See Stapleton "Scientific and Legal Approaches to Causation" in Freckelton and Mendelson (ed) Causation in Law and Medicine (2002) 14 at 22 - see at [65] - [70] below. [31] The reasoning here is that even a large increase in risk from a treatment failure may not necessarily demonstrate that the treatment failure caused an ensuing injury. However, the Court is permitted to draw a robust inference depending on all of the circumstances of an individual case. [32] Ambros also considered a case of negligent delay in a diagnosis of cancer in Gregg v Scott [2005] 2 AC 176. The delay in diagnosis and consequent treatment, reduced Mr Gregg's chance of survival from 42 per cent to 15 per cent. The Court in Ambros considered a minority opinion by Lord Nicholls that the law should fashion a meaningful remedy for the loss of even a poor prospect of recovery, even when a chance of survival with proper treatment involved hypothetical speculation on an event that still lay in the future. The Court in Ambros concluded: [46] Whatever the future developments in loss of chance in other jurisdictions, the loss of chance analysis seems to us to be incompatible with the accident compensation regime. Under a no fault regime, either there is cover or there is not. There is no ability to discount compensation and in a no fault regime no conceptual need to do so. Where the wrongdoer is not himself or herself financially liable for the injury, there is no need to discount the level of compensation to reflect the fact that other factors unrelated to the wrongdoer's conduct played a part in the injury. The focus is on "real compensation" - see The Woodhouse Report at [59]. Further, as indicated above in the discussion on McGhee, at [32], to the extent that risk of injury is considered a sufficient basis of liability (even in the limited circumstances of the Fairchild exception), this would be inconsistent with Atkinson. In terms of the causation principles set out in Atkinson, any risk must be realised in the occurrence of a personal injury and the personal injury must be proved to have been caused by the risk factor involved. In keeping with this principle, if the omission to treat causes an identifiable added injury, cover would be available for that injury. [33] The Court also dictated caution in using statistical information, at [71] et seq; [71] It would not be unusual in medical negligence cases for a claimant to be able to point only to a statistical link between particular events and the injury. Although Lord Nicholls was in the minority in Gregg v Scott, his comments on the use of such statistics are, in our view, of interest. He said that, in cases of medical negligence, assessment of a patient's loss may be hampered, to greater or lesser extent, by one crucial fact being unknown and unknowable: how the particular patient would have responded to proper treatment at the right time. The patient's previous or subsequent history may assist and there may be other indications, but there will be times when statistical evidence will be the main evidential aid - see at [27] per Lord Nicholls. [72] Lord Nicholls recognised that statistical evidence is not strictly a guide to what would have happened in a particular case. Statistics record retrospectively what happened to other patients in more or less comparable situations. They reveal trends of outcome and are general in nature. The different way other patients responded in a similar position says nothing about how the claimant would have responded. Statistics do not show whether the claimant patient would have conformed to the trend or been an exception from it. They are an imperfect means of assessing outcomes even of groups of patients undergoing treatment, let alone a means of providing an accurate assessment of the position of one individual patient - see at [28] of his judgment. [73] Lord Nicholls said that in practice statistical evidence of a diminution in perceived prospects will, despite its imperfection, often be the nearest one can get to evidence of diminution of actual prospects in a particular case. When there is nothing better, courts should, in his view, be able to use these figures and give them such weight as is appropriate in the circumstances. This conclusion is even more compelling when it is recalled that the reason why the actual outcome for the complainant patient if treated promptly is not known is that the defendant by his or her negligence prevented that outcome becoming known. The value of any statistics will of course depend on their quality: the methodology used in their compilation, how up to date they are, the number of patients involved in the statistics, the closeness of their position to that of the claimant, the clarity of the trend revealed by the figures and so on - see at [32]. . .. [76] We consider that statistical evidence may be of use in the assessment of causation but the limitations of such evidence must be clearly borne in mind. There is always a risk that statistical evidence gives an illusion of precision hat is lacking. Statistics as to what has happened to other patients in similar situations are also not necessarily a guide to what might have happened in the case at hand - see at [72]. Such evidence can, however, as pointed out by Spigelman CJ, provide evidence of possibility which might translate into the requisite degree of probability, depending on the strength of that statistical evidence and any other relevant evidence pointing to causation. ... [34] Under the accident compensation regime, loss of a chance is not a proper basis for causation when considering cover. Where the failure increases the risk of a poor outcome, the statistical evidence may permit an inference that the poor outcome was caused by the failure. That is to say a Court may draw a robust inference where there is a high probability in the individual case of a better outcome from treatment, or a worse outcome from failure to treat. [35] Treatment injury under the Accident Compensation Act 2001 is defined in s 32, the relevant parts of which are set out here: 32 Treatment injury (1) Treatment injury means personal injury that is - (a) suffered by a person - (i) seeking treatment from 1 or more registered health professionals; or (ii) receiving treatment from, or at the direction of, I or more registered health professionals; or b) caused by treatment; and (c) not a necessary part, or ordinary consequence, of the treatment, taking into account all the circumstances of the treatment, including - (i) the person's underlying health condition at the time of the treatment; and (ii) the clinical knowledge at the time of the treatment. (2) Treatment injury does not include the following kinds of personal injury: (a) personal injury that is wholly or substantially caused by a person's underlying health condition: (3) The fact that the treatment did not achieve a desired result does not, of itself, constitute treatment injury. ... [36] The appellant's claim is that progress of the disease could have been delayed by an appropriate course of treatment. The claim is that the failure did in fact cause the consequence of unhindered progression of the disease and earlier death which was not an ordinary consequence of proper treatment. [37] Under Ambros, it is not enough to say that there was a better chance of delaying progress of the disease. However the statistical probability may permit a robust inference that intervention would probably have delayed progression of the disease. The reasoning requires a statistical probability. It is not enough that the deceased may have lost a chance of an improved outcome, but that he lost a good chance - in the nature of a probability. For example, Mr Martin's opinion that Mr Sheppard was "denied an opportunity" is not a basis for cover. Submissions [38] In argument in the appeal, the appellant did not rely on the width of incision that might have allowed cancerous cells to remain in the scar, In fact the localised tumour did not recur and there was insufficient evidence that the excision caused any added problem. [39] Mr Miller for the appellant aimed his submissions at Mr Martin's opinion that, if proper treatment had been given at diagnosis, statistically Mr Sheppard would have had at least a 5 year survival period (91% certain). [40] Mr Barnett submitted that the weight of medical evidence does not support a causal nexus between the treatment and avoidably rapid disease progression and death. While his treatment was not optimal, the independent specialists have concluded that it is likely that his melanoma would have progressed notwithstanding any treatment or delay in treatment in 2008. Dr Holland said that "There is an inevitable degree of uncertainty as to whether adherence to the guidelines for the management of this melanoma would have influenced the outcome". Dr Beer said that the die was cast because in October 2008 the melanoma had already metastasised to the auxiliary lymph node which had taken a year to be clinically palpable. [41] Mr Barnett submitted that Mr Martin, whilst critical of the care given, did not say that the outcome would have been different if the accepted standard of care had been provided. He did not disagree with the pathologist, Dr Beer, that "the die was cast"; he observed that it was the position in retrospect, that is to say on the evidence now available. [42] Mr Miller submitted that there was no evidence to support Dr Beer's statement that the melanoma had already metastasised. Mr Martin said that the only true indication of whether or not the melanoma had metastasised would be through a sentinel node biopsy. [43] Mr Miller submitted that Dr Martin's evidence placed a tactical burden on the respondent. That is to say that the established treatment failure and the statistical evidence shifts the tactical burden to the respondent. This again is a reference to Ambros at para 55: [55] As pointed out by the Court in Ithaca (Custodians) Limited v Perry Corporation [2004] 1 NZLR 731 at [44] - [47] (CA), the term burden of proof has been used in two quite distinct senses - see Williams "Burdens and Standards in Civil Litigation" (2003) 25 Syd LR 165. The first is a reference to the legal burden. The legal burden is what must ultimately be proven by a person in order to win the case. Equally, it can refer to the evidential burden. The term evidential burden is, in turn, used to refer to two quite distinct notions. In the first sense, it means the burden of adducing evidence on an issue on pain of having the trial Judge determine the issue in favour of the opponent. The second sense in which the phrase is used refers to the burden resting upon a party who appears to be at risk of losing on a given issue at a particular point in a trial. This merely involves a tactical evaluation of who is winning at a particular point which can shift depending upon the trial dynamics. This is often referred to as the tactical burden. [44] The Court said that while the shifting of the legal burden was rejected in Atkinson, the shifting of the evidential onus has met with a greater acceptance. The evidential burden may shift once a party has established a prima facie case. It was submitted that treatment failure was squarely shown and a statistical likelihood that a better outcome could otherwise have been achieved, bringing the claim to a point at which the Reviewer or the Court would have to be satisfied by other evidence that the treatment did not cause injury. That is reinforced by the fact that the treatment failure was the reason why the actual outcome for the complainant patient if treated promptly cannot be known (see Ambros para [73]). [45] Mr Miller submitted that there was an appalling treatment failure because the general practitioner had been informed of removal of an earlier melanoma, but still failed to refer the patient or arrange follow-up. He submitted that one can be mesmerised by the rule that loss of a chance is not the same as causation, but remember that every failure to treat involves a loss of chance. [46] Mr Miller emphasised Dr Martin's impressive practical and academic experience and referred to the conclusions to be drawn from Dr Martin's statement that one can see quite large differences in prognosis and outlook in the prognosis for different stage III melanoma so that if Mr Sheppard's melanoma metastasis had been detected in the lymph node while small he may have had a different come Decision [47] After carefully considering all the evidence, I think that Mr Barnett is correct in his submission that the expert advisors cannot say that the deceased's condition would probably have progressed in any other way had be received a different course of treatment in October 2008. [48] Dr Roche said that SLNB, wider excision, and closer follow up may help in controlling local and regional recurrence, but have not been shown to improve survival even according to the latest guidelines. He found no strong evidence to support the claim that injury was caused by failure to provide treatment, even with earlier detection of the axillary metastasis. [49] Dr Beer considered that the die was cast. In his first letter, Mr Martin commented on that view but did not contradict it. He said that it could only be said in retrospect and that Mr Sheppard was denied an opportunity, the outcome of which was impossible to predict. If Mr Martin proposed a retrospective probability of a better outcome if referral had been made to a melanoma specialist, he would have said so, but he did not. His focus was on the treatment error, not the probable outcome. [50] Dr Beer or Mr Martin did not expound an opinion based on the statistical information applied to the particular case of the deceased. The discussion in Ambros of the use of statistical information is relevant. The only direct discussion of the statistics with relation to Mr Sheppard's case is found in Dr Martin's second letter in February 2013. He naturally disagreed with a statement that statistics mean nothing. He stated that "The only randomised control trial to date for SNB, MSLTI, has shown that there is at least a - 20% survival benefit - for detection of micrometastases in sentinel nodes vs waiting for a macrometastases in the 'control arm '. " But Dr Martin's opinion stopped short of relating that to Mr Sheppard's history and the course of diagnosis and progression of disease. The information does not predict that Mr Sheppard would have lived 20% longer with treatment. His main argument in relation to statistics was that clinicians must rely on statistical evidence when making treatment decisions, and that Mr Sheppard was denied a chance of a better outcome. [51] Dr Martin said that Mr Sheppard may have had a different outcome with detection of metastasis in the lymph node while small, and that he was denied the opporunity of potential cure. It is not explained whether or not "cure" means an absolute cure, and it is likely to refer to a range of successful outcomes of treatment. Dr Martin's evidence is the best evidence supporting the appellant's claim, but it goes no further than loss of a chance. Ambros has decided that loss of a chance does not satisfy the test for causation. [52] I find that the evidence is insufficient to establish the kind of prima facic case that would shift the evidential burden to the respondent, even when it was because of the treatment failure that the possible outcome cannot be known. Causation remains a separate and essential element to prove treatment injury. The evidence does not reach a state of prima facie proof on the question of causation. [53] For those reasons, I reach the conclusion that the appeal must be dismissed. It is deeply distressing for the family that there should have been a treatment failure resulting in loss of a perhaps valuable chance at recovery or even temporary recovery. Regretfully, that is not enough in this case to establish accident compensation cover. [54] The appeal is therefore dismissed. Judge D A Ongley District Court Judge