Graham v Repatriation Commission

The proper approach 7 In McLean v Repatriation Commission [2001] FCA 1505 at [7 and 8], the Full Court (Whitlam, Madgwick & Dowsett JJ) observed: In Byrnes v Repatriation Commission (1993) 177 CLR 564, the High Court said at 571: 'The position may be summarized as follows: (1) First, sub-s. (3) of s 120 is applied: do all or some of the facts raised by the material before the Commission give rise to a reasonable hypothesis connecting the veteran's injury with the war service? The hypothesis will not be reasonable if it is contrary to known scientific facts or is obviously fanciful or untenable. If the hypothesis is not reasonable, the claim fails. Proof of facts is not in issue at this point. (2) If a reasonable hypothesis is established, sub-s. (1) of s 120 is applied. The claim will succeed unless: (a) one or more of the facts necessary to support the hypothesis are disproved beyond reasonable doubt; or (b) the truth of another fact in the material, which is inconsistent with the hypothesis, is proved beyond reasonable doubt, thus disproving, beyond reasonable doubt, the hypothesis.' Section 120A, where it operates, will modify this procedure. It is sufficient for present purposes to say that the section authorizes the Repatriation Medical Authority to issue Statements of Principles in connection with specified causes of death. When the cause of death is one to which such a statement applies, any hypothesis will only be reasonable if it satisfies the requirements of that statement … . 8 As the Full Court (Beaumont, Hill and O'Connor JJ) pointed out in Repatriation Commission v Deledio (1998) 83 FCR 82 at 97-8, where there is an applicable Statement of Principles ("SOP"), the hypothesis will only be reasonable if it falls within it. The applicant asserts that two different SOPs are arguably relevant for present purposes. One is headed "Statement of Principles Concerning Peptic Ulcer Disease" and the other is headed "Statement of Principles Concerning Non-Hodgkin's Lymphoma". They appear at AB 18 - 21 and AB 22 - 24 respectively. It may be that the former is not relevant for present purposes. I will return to that matter at a later stage. The applicant has formulated her hypothesis in order to conform with the latter. 9 An hypothesis will be reasonable if the material points towards it. It is not sufficient that it be merely open on the material. See East v Repatriation Commission (1987) 16 FCR 517 at 532-3; Repatriation Commission v Bey (1997) 79 FCR 364 at 371-3; and Bull v Repatriation Commission [2001] FCA 1832 at [14] - [20].

SUMMARY OF EVIDENCE 10 In view of the passage of time it is not surprising that there is little available detail as to the veteran's medical condition during his service and thereafter. The applicant gave evidence that he had commenced to smoke and drink heavily during his army service and that these habits continued until his death. She also claimed that following his discharge, and until his death, he suffered regularly from severe stomach pains. He experienced gastro-intestinal bleeding in 1967. The cause is unknown. In January 1977 the veteran sought medical treatment for abdominal pain and ankle-swelling. A secondary carcinoma was located in the neck. His doctor (Dr Dunn) discovered an epigastric mass. He noted that a barium meal had previously been reported as normal and inferred that there was probably a "pancreatic primary". Dr Milliner refers to a barium meal which appears to have been administered after the 1967 incident (AB 30, par 12). That seen by Dr Dunn in 1978 was presumably more recent. See Exhibit 2. Dr Dunn referred the veteran for further investigation. He was found to be suffering from cancer in the right supra-clavicular region and in the epigastrium. He died on 10 October 1978. Post-mortem examination disclosed wide-spread cancer (to use lay-terms), particularly in the vicinity of the stomach, adrenals, pancreas, liver, kidneys and bowel. I will hereafter use the words "cancer" and "malignancy" with their usual lay meanings as denoting any malignant growth. I will try to be more precise in the use of other medical terms.

The hypothesis 11 The applicant advanced the following hypothesis as linking the veteran's death to his army service: (i) The veteran was infected with Helicobacter pylori while serving in the army. (ii) That infection caused him to develop peptic ulcer disease. (iii) The peptic ulcer disease caused a primary cancer site to develop inside the stomach. The primary cancer was probably a gastric lymphoma of mucosal-associated lymphoid tissues (MALT). (iv) The primary gastric MALT lymphoma gave rise to the reticulum cell sarcoma which caused death. 12 It is common ground that Helicobacter pylori infection can cause peptic ulcers and, in due course, gastric MALT lymphoma which is a malignancy located in the stomach. 13 AAT rejected the hypothesis as being unreasonable for the following reasons: 1. There was no evidence that the veteran was ever infected with Helicobacter pylori. 2. There was no evidence that the veteran ever suffered from peptic ulcer disease. 3. The veteran had a normal barium meal in or about January 1977. 4. The post-mortem report showed that there was no primary cancer site inside the veteran's stomach.

The appeal 14 In arguing the appeal the applicant advanced three grounds, namely that: A. The Administrative Appeals Tribunal failed to decide the matter on the material before it, and erroneously attributed weight to misleading evidence. B. The Administrative Appeals Tribunal failed to make a finding that the veteran had peptic ulcer disease and Helicobacter infection despite the admissions of the respondent. C. The Administrative Appeals Tribunal erred in failing to identify material before it which satisfied the relevant factors and the relevant Statements of Principles and as such did raise a reasonable hypothesis as required by law.

Ground B 15 It seems that the AAT was in error in its findings that there was no evidence of Helicobacter pylori infection or peptic ulcer disease. In the AAT proceedings the respondent filed a document described as "Respondent's Facts and Contentions" in which it is recited that: 4.4 Shortly after returning home, (the veteran) suffered from gastric symptoms, which culminated in a gastric haemorrhage in 1967. The most likely diagnosis is peptic ulcer disease. 4.5 This being the case, it can be conceded that he suffered from Helicobacter pylori infection, which is more or less a necessary precursor to peptic ulcer disease. 16 In argument before me, the respondent sought to resile from these concessions, asserting that they had been made incorrectly and relying upon the decision in Calderaro v Secretary, Department of Social Security (1991) 33 FCR 244. Gray J was there concerned with a concession as to entitlement to a pension in certain circumstances. His Honour found that the AAT was entitled to reject that concession. However in the present case, the AAT did not consciously refuse to adopt the concession made by the respondent. It seems simply to have overlooked the concession. I see no reason why the parties to proceedings in the AAT should not agree as to a particular fact or facts. Indeed, the process of delivering statements of facts and contentions, which was adopted in this case, assumes that some facts will not be in dispute. It would unnecessarily extend proceedings if parties were to be prevented from agreeing facts. If the AAT intended to reject that concession (assuming that it was entitled to do so), then it should have given appropriate notice to the parties to enable them to lead such evidence as they thought appropriate and to make relevant submissions. The respondent should not now be allowed to depart from concessions which were presumably made after due consideration. It follows that the AAT was in error in its findings. This was not a case in which the AAT erred as to the evidence. It rather erred as to the issues of which it was seized. I would have little difficulty in characterizing that error as an error of law. However identification of the consequences of the error must depend upon an analysis of the case as a whole.

Grounds A and C 17 These grounds, as stated above, sound very much like questions of fact rather than of law. Putting the applicant's case at its highest for present purposes, it is that the AAT must have failed to apply the law as prescribed in Byrnes in that the evidence clearly demonstrated a reasonable hypothesis linking the veteran's death to his service, the first step prescribed by that decision. This exercise must commence with identification of the cause of the veteran's death. The applicant's hypothesis was that death was caused by a "B Cell lymphoma of the MALT-omer sub-type" in the stomach. The material suggests that a lymphoma may be either "nodal based" or "mucosal based" A MALT lymphoma is, by definition, mucosal based. A MALT lymphoma may originate elsewhere than in the stomach. At AB 49 there is reference to MALT lymphomas in the lung. The applicant submits that the relevant disease is one of a cluster now described as "Non-Hodgkin's Lymphoma" for which an SOP is in force. The applicant's hypothesis will be reasonable only if it is upheld by the SOP. It is common ground that the SOP will only support the hypothesis if: ® the veteran suffered primary B-cell lymphoma of the stomach; and ® at the time of the onset of Non-Hodgkin's lymphoma, he was suffering Helicobacter pylori infection. 18 It seems that in the AAT, the primary issue as to reasonableness of the hypothesis was whether the veteran had a primary lymphoma in his stomach as required by the SOP. That was by no means the only matter in dispute, but it was one of the bases upon which the AAT found that the applicant had not met the first requirement prescribed in Byrnes. The AAT concluded that the veteran had not suffered such a lymphoma. The other basis for rejecting the hypothesis involved the erroneous findings that the veteran had not suffered peptic ulcers and Helicobacter pylori infection. However if the AAT was correct in its views as to the stomach malignancy, then those errors would not matter. 19 The post-mortem report is the primary evidence as to cause of death. Although it does not expressly identify that cause, it describes one of its principal findings as "reticulum cell sarcoma" involving various organs including the stomach. Three other passages are presently relevant. The first appears under the heading "Post-Mortem Appearances". After reference to a large mass of tumour tissue extending from the lesser curvature of the stomach and the porta hepatis to the pelvic brim, it is said that: There were multiple seedings on surface of stomach, mesentery of small bowel, omentum of large bowel. 20 The second passage appears under the heading "Macroscopic Appearances of Organs". Of the intestines and stomach, it is noted: Neoplastic deposits on outer surface of stomach and omentum of large bowel. Mesentery of small bowel studded with deposits. 21 Finally, under the heading "Microscopic Findings on Sections Taken", concerning the stomach, it is noted: Deposits of reticulum cell sarcoma, some confined to the serosa, others invading all coats to the mucosa. 22 The initial decision to refuse the application was apparently based on the opinion of Dr Smith. The following is an extract from the letter advising rejection of the claim (AB 26): The only type of non-Hodgkin's lymphoma which may be caused by infection with Helicobacter pylori is primary B-cell lymphoma of the stomach. A Departmental Medical Officer has advised - '- The Non-Hodgkin's lymphoma was not a lymphoma of the stomach. There may be some confusion as to what constitutes the stomach but in medical terms it is the digestive organ between the oesophageus and the proximal small bowel. In colloquial terms the 'stomach' may refer to a wider area which is more accurately termed the abdomen. In this case the main tumour mass was within the abdomen and seeded to (among other sites) the serosa (outer layer) of the stomach, with some tumour deposits invading deeper into the stomach. In my opinion this is not a picture of a primary B-cell lymphoma of the stomach.' 23 Subsequently, Dr Grant indicated his agreement with this conclusion. In his report of 13 November 2000 he said: The cause of death is explicit - reticulum cell sarcoma. There are no autopsy findings to suggest concurrent malignant neoplasm of the pancreas, peptic ulcer disease or gastric or duodenal scarring. There appear to be no plausible grounds for a link to service via a primary pancreatic carcinoma or peptic ulcer disease. I agree with the assessment of Dr Smith, Compensation Medical Adviser, of 15 March 1999 that the stomach was involved secondarily rather than by primary tumour - the main sites affected were on the outer or serosal surface consistent with intraperitoneal metastases. 24 Some parts of that report suggest that at the time, Dr Grant was not aware of the veteran's full history, but that is of no importance. In his report of 23 May 2001, Dr Grant observed: The diagnosis of peptic ulcer disease was not confirmed however either when the late veteran was alive or at post-mortem. In the latter case, the stomach wall was infiltrated with tumour and its anatomy greatly distorted, making any chronic scarring difficult to detect in any case. 25 As I have said, it is common ground that the veteran had suffered peptic ulcer disease at some stage. In his report of 31 May 2001, Dr Grant referred to a report provided by Dr Milliner (to which I will refer at a later stage) and offered the view that he was not persuaded by the arguments "favouring a primary stomach site over a non-gastric primary site" which appear in that report. I take the word "gastric" to refer to the stomach. 26 The matter was referred to Dr Ades, a Histopathologist. He agreed that the description in the post-mortem report of the main tumour mass as a "reticulum cell sarcoma" would include conditions now known as "diffuse large B-cell lymphoma", "T-cell lymphoma (anaplastic large cell and large cell NOS subtypes)" and certain other conditions. The two named conditions are examples of Non-Hodgkin's lymphoma as defined in the relevant SOP. The applicant's hypothesis, as advanced by Dr Milliner, was that the veteran's primary malignancy was a B-cell lymphoma in the stomach. Dr Ades concluded that it was most likely that the veteran had suffered a primary nodal disease which term does not include such a lymphoma. He relied for this conclusion upon: ® the veteran's presentation with a supra-clavicular lymph node; ® the normal barium meal; ® the autopsy findings which show "the bulk of disease to be nodal based on both sides of the diaphragm and mostly serosal involvement of the stomach with only focal mucosal involvement consistent with direct invasion. NB gastric MALT lymphoma tend to extensively involve the gastric mucosa." 27 He was asked if the veteran's tumour was "the same as a MALT tumour" and replied: In my view this is unlikely as this appears to be nodal based rather than mucosal based disease and MALT type lymphomas by definition require a low grade component to make the diagnosis. This would have been described at the time as either lymphocytic or poorly differentiated lymphocytic lymphoma. However diffuse large B cell lymphoma may arise from MALT lymphoma and then secondarily involve lymph nodes. This possibility would be difficult to exclude but would be unlikely. 28 It should be noted that Dr Ades was there addressing only the possibility of a mucosal based disease as opposed to a nodal based disease. He was not addressing the possibility of the veteran's disease originating in the stomach. He was also asked: What is the likelihood or otherwise that Helicobacter pylori played a part in its onset? 29 He replied: Helicobacter is a major and possibly necessary risk factor for gastric MALT lymphoma but not other lymphoma. This case, for the reasons given above, most probably did not arise from MALT lymphoma and therefore Helicobacter did not play a part in its onset. 30 Further questions were submitted for Dr Ades' consideration and he responded on 9 September 2001 as follows: 1. The normal barium meal referred to is from the letter from Dr P D Dunbar to the DVA dated 9.11.78. This letter was referring to initial investigations done by Dr Dunn following his presentation with supra-clavicular lymphadenopathy in January 1977. The absence of a radiological abnormality at this time before treatment suggests a primary gastric lymphoma is unlikely. 2. I am not aware of any reference to duodenal ulceration in the initial documents sent to me. The draft report by Dr Milliner, which I now have, refers to a long clinical history consistent with peptic ulceration. If this is so then Helicobacter Pylori infection is a likely underlying cause. Helicobacter infection is a major aetiological factor for primary gastric lymphoma of mucosa associated lymphoid tissue (MALT) type. Therefor if the lymphoma was a primary gastric lymphoma of this type there would be a link between these factors. However, as I have previously discussed, there is no evidence to suggest a primary gastric origin or type in this case and a nodal origin is most likely. Despite this, the possibility of undetected gastric lymphoma of MALT type with limited gastric involvement undergoing high grade transformation and extensive nodal spread is theoretical possible, although highly unlikely, in my opinion. 3. … 4. There is no evidence to indicate this was an indolent lymphoma. On the contrary, the reported histological features most probably corresponding to a large cell lymphoma including anaplastic cells and extensive necrosis, initial misdiagnosis is metastatic carcinoma, recurrence within 6 months following initial radiotherapy and death less than 2 years after presentation, indicate a high grade lymphoma. 5. As discussed previously, this is most probably primary nodal disease. A primary origin from gastric MALT lymphoma is theoretically possible, although highly unlikely. 6. … 7. The pattern of organ involvement in the autopsy report supports a primary nodal origin for the lymphoma. In particular, the limited mostly serosal involvement of the stomach with the main mass described having a nodal distribution, most probably retroperitoneal, and presence of hilar lymphadenopathy supports this. 31 The reference in par 4 of Dr Ades' report to "indolent lymphoma" appears to relate to Dr Milliner's use, in par 25 of his report, of the expression "indolent lymphoid neoplasms", as describing "gastric cancer that has an untreated natural history in years". Dr Milliner's use of this term might suggest a belief that the veteran's disease had lain undetected for many years. Dr Ades' view (as expressed in par 4) appears to be to the contrary. 32 As I understand the evidence, Dr Smith, Dr Grant and Dr Ades consider that the post-mortem findings indicate that the malignancy originated outside of the stomach and spread to it. This is indicated by the seeding or deposits found on the outside of the stomach, with some penetration into that organ. Gastric MALT lymphoma originates inside the stomach, but it may spread beyond it. Dr Ades describes the likelihood of the latter scenario in the present case as "highly unlikely" and finds no evidence suggesting it. 33 I turn now to Dr Milliner's evidence. Most of his report is concerned with the likelihood of the veteran having contracted Helicobacter pylori infection during his military service and suffering from peptic ulcer disease thereafter. That is not directly relevant to the question of cause of death which must depend, for present purposes, upon interpretation of the post-mortem report. Dr Milliner deals with this matter at pars 25 - 37. His justification of the hypothesis that there may have been a primary malignancy in the stomach appears particularly at pars 28 - 30. In par 28 he refers to the evidence of Dr Smith and in particular, to the statement that the main tumour mass was within the abdomen and seeded to the serosa (the outer layer of the stomach) with some tumour deposits invading deeper in to the stomach. In par 29, Dr Milliner cavils with the use of the expression "seeded to", pointing out that the post-mortem report, under the heading "Post-Mortem Appearances", reads "multiple seedings on surface of stomach". He then refers to various passages in the post-mortem report, concluding that: Such seedings on the mucosal layers and surface layers of the stomach are CONSISTENT WITH MALT-omers that produce B Cell Non-Hodgkin's Lymphoma. 34 I take this to mean that MALT lymphomas produce seeding. I do not understand Dr Milliner to be asserting that such seeding is indicative of a gastric MALT lymphoma. 35 In par 30 Dr Milliner comments: It is my opinion that it would be unwise to rely implicitly on the autopsy report with respect to such 'niceties' as to whether the tumour was seeded to or on the stomach. It is not certain of the status of Dr. JAMIESON as already discussed and he certainly would have not been aware of the implications that we might draw from his description. If one reverts to the statistics, it then obtains that primary lymphoma of the stomach is relatively uncommon. It accounts for few than 15% of gastric malignancies and only about 2% of all lymphomas. The stomach is however the most frequent extranodal location for lymphoma. The disease is difficult to distinguish clinically from gastric carcinoma (even now let alone 1978). Both tumours are most often detected in the 6th decade of life, present with epigastric pain, early satiety and generalized fatigue. 36 Dr Milliner then suggests (par 31) that the veteran's age and symptoms on admission were "at least consistent with the clinical data available for Mr Graham". In pars 32, 33, 35 and 36 he suggests that numerous other aspects are consistent with a gastric MALT lymphoma. I do not take these observations to mean that the various matters in question are indicative of gastric lymphoma. Dr Milliner did not really answer the argument advanced by the other doctors that the post-mortem findings suggest that the malignancy developed outside of the stomach and subsequently spread to it. My own reading of the post-mortem report is that seeding or neoplastic deposits were found on the outer surface of the stomach by visual examination. Microscopic examination disclosed "invasions" through the wall of the stomach to the mucosa. The word "invasion" clearly demonstrates that the doctor conducting the examination (Dr Jamieson) intended to convey the conclusion that the cancer had originated outside of the stomach. No doubt it was for this reason that Dr Milliner felt compelled to discredit him. See pars 25, 26, 28 and 30. To say that "such seedings on the mucosal layers and surface layers of the stomach are consistent with MALT-omers" is to say nothing about the fact that there was no evidence of malignancy in the stomach (other than that of invasion) and ample evidence of wide-spread malignancy outside of it. It was suggested in argument that any evidence of cancer in the stomach may have disappeared as a result of radiotherapy and chemotherapy, but there is no evidence suggesting that possibility. Dr Milliner also does not deal with the normal barium meal which, in Dr Ades' view, suggested that a primary gastric lymphoma was unlikely. It is true that the precise date of this procedure is not known, but Dr Dunn presumably had that information and was content to make use of the results.

Reasonableness of the hypothesis 37 It is important to keep in mind that the only step in the process prescribed by Byrnes which was undertaken by the AAT was to consider the reasonableness of the applicant's hypothesis. That process did not involve resolution of conflicts in the evidence. The only question was whether the material, or any part of it, pointed towards the hypothesis. In this case, the applicant's hypothesis must identify a possible cause of death and a possible basis for linking it to the veteran's service. The applicant's hypothesis is that death occurred as a result of a primary gastric lymphoma, but there is no evidence suggesting that the veteran suffered such a condition. Dr Jamieson's report reveals widespread malignancy outside of the stomach and invading it. He reported no other malignancy in the stomach. This finding was, in Dr Ades' view, supported by other considerations such as the barium meal result and the location of the secondary tumour, as well as the more detailed post-mortem findings. The applicant's hypothesis rejects these findings and the opinions of Drs Smith, Grant and Ades based upon them. It asserts a primary stomach lymphoma upon these bases: ® the veteran's history of Helicobacter pylori infection and peptic ulcer disease; ® certain factors which are said to be "consistent with" such a lymphoma; and ® statistical evidence, the direct relevance of which is not demonstrated. 38 Mere consistency with some aspects of the material does not point to the validity of the hypothesis. Dr Milliner identified only the possibility of gastric lymphoma. Dr Ades was unable to exclude such a bare possibility, but neither of them identified any material pointing in that direction. In particular, Helicobacter pylori infection and peptic ulcer disease may be relevant preconditions for gastric lymphoma, but they do not demonstrate its presence. 39 The applicant also sought to rely upon SOP 22 which appears at AB 18 - 21. However this appears to be an SOP issued for the purposes of s 120B rather than s 120A. We were informed by counsel for the respondent that SOP 21 may apply for present purposes. However that is of little assistance to the applicant because, as I understand her argument, any hypothesis for the purposes of SOP 21 also depends upon death as a result of cancer of the stomach. As I have said, none of the material points in that direction. 40 The respondent also submits that there is no evidence which would link any Helicobacter pylori infection and/or peptic ulcer disease to the veteran's service. However, given the concessions made by the respondent to which I have referred, that would be a matter to be determined having regard to the relatively short period of time between his returning from service and first suffering gastric symptoms in light of the following evidence from Dr Grant (AB 42): It is also important that he is unable to provide any objective findings to confirm that Mr Graham was infected with Helicobacter pylori let alone that this was probably acquired on service. As importantly, the history of epigastric pain beginning several months after discharge is more in keeping with acquisition after rather than during eligible service, in my opinion. The early research in Helicobacter pylori found that symptoms followed within minutes to hours after acquiring a large dose of the micro-organism, rather than weeks or months later. 41 This appears to be the only evidence as to the likely course taken by of Helicobacter pylori infection. It certainly does not point in the direction of the applicant's hypothesis, namely that the infection was contracted during the veteran's service. It is not necessary that I take this matter any further. 42 I should also mention that had I been otherwise minded to remit the matter to the AAT for further consideration, I would have left open for its consideration the question of whether or not at any further hearing, the respondent should be allowed to depart from the concessions to which I have referred.

Orders 43 In the circumstances the appeal should be dismissed. I will hear submissions as to costs. I certify that the preceding forty-three (43) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Dowsett.

For the Applicant: Mr N Millward (friend)

Solicitor for the Respondent: Australian Government Solicitor