Ellis v Repatriation Commission

1. The decision of the Administrative Appeals Tribunal be set aside and the matter remitted to the Tribunal differently constituted to be determined according to law.
2. The Respondent pay the Applicant's costs. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011 (Cth).

IN THE FEDERAL COURT OF AUSTRALIA VICTORIA DISTRICT REGISTRY GENERAL DIVISION VID 721 of 2013

JUDGE: GORDON J DATE: 11 AUGUST 2014 PLACE: MELBOURNE

INTRODUCTION 1 The Applicant, Nora Ellis, is the widow of Mr Leslie Harold Ellis. Mr Ellis served in the Australian Army during the Second World War, rendering operational service from 8 April 1942 to 11 June 1947. Mr Ellis died on 3 January 2012. The Veterans' Entitlements Act 1986 (Cth) (the VE Act) provides for payments of pensions to dependants of veterans where the death of a veteran was war-caused. 2 Pursuant to s 14 of the VE Act, Mrs Ellis applied to the Repatriation Commission (the Commission) for a pension. Mrs Ellis' claim was rejected by a delegate of the Commission and, on review, by the Veterans' Review Board (the VRB). Mrs Ellis applied to the Administrative Appeals Tribunal (the AAT) for review of the VRB's decision. On 26 June 2013, the AAT affirmed the decision of the Commission by rejecting Mrs Ellis' claim for a pension: Ellis v Repatriation Commission [2013] AATA 433 (the AAT decision). 3 Mrs Ellis now appeals from the AAT decision pursuant to s 44 of the Administrative Appeals Tribunal Act 1975 (Cth). For the reasons below, the AAT decision should be set aside and the matter remitted to the AAT differently constituted to be determined according to law. 4 It is unfortunate that the hearing of this application on 27 May 2014 occurred approximately 10 months after Mrs Ellis lodged the application with the Court. The matter was placed in my docket on 20 May 2014.

LEGISLATIVE STRUCTURE 5 Section 13(1) of the VE Act provides that where a death of a veteran is war-caused, the Commonwealth is liable to pay a pension to the veteran's dependants. It was not in dispute that Mr Ellis rendered operational service and that he had suffered a war-caused knee injury. What was in dispute was whether his death was war-caused. 6 Section 8 of the VE Act sets out the circumstances in which a death is taken to be war-caused. Relevantly, s 8 provides: (1) Subject to this section and section 9A, for the purposes of this Act, the death of a veteran shall be taken to have been war-caused if: … (b) the death of the veteran arose out of, or was attributable to, any eligible war service rendered by the veteran; 7 Where a claim is related to operational service, the standard of proof to be applied to the question of whether the death was war-caused is prescribed by ss 120(1) and (3) of the VE Act. 8 Section 120 of the VE Act relevantly provides: (1) Where a claim under Part II for a pension in respect of … the death of a veteran, relates to the operational service rendered by the veteran, the Commission shall determine that … the death of the veteran was war-caused, as the case may be, unless it is satisfied, beyond reasonable doubt, that there is no sufficient ground for making that determination. Note: This subsection is affected by section 120A. … (3) In applying subsection (1) or (2) in respect of … the death of a person, related to service rendered by the person, the Commission shall be satisfied, beyond reasonable doubt, that there is no sufficient ground for determining: … (c) that the death was war-caused or defence-caused; as the case may be, if the Commission, after consideration of the whole of the material before it, is of the opinion that the material before it does not raise a reasonable hypothesis connecting the … death with the circumstances of the particular service rendered by the person. Note: This subsection is affected by section 120A. 9 The operation of s 120(3) is "affected by" s 120A of the VE Act. Section 120A of the VE Act applies to claims, amongst others, made on or after 1 June 1994 under Pt II relating to operational service rendered by a veteran. Section 120A(3) of the VE Act relevantly provides: For the purposes of subsection 120(3), a hypothesis connecting … the death of a person with the circumstances of any particular service rendered by the person is reasonable only if there is in force: (a) a Statement of Principles [SoP] determined under subsection 196B(2) … … that upholds the hypothesis. 10 Section 196B(2) relevantly provides: If the Authority is of the view that there is sound medical-scientific evidence that indicates that a particular kind of … death can be related to: (a) operational service rendered by veterans; or … the Authority must determine a [SoP] in respect of that kind of … death setting out: (d) the factors that must as a minimum exist; and (e) which of those factors must be related to service rendered by a person, before it can be said that a reasonable hypothesis has been raised connecting … death of that kind with the circumstances of that service. 11 Pursuant to s 196B(2) of the VE Act, the Repatriation Medical Authority had determined two SoPs relevant to Mr Ellis. As at 30 March 2011, a SoP concerning Malignant Neoplasm of the Liver, No 21 of 2011 (Malignant Neoplasm of the Liver SoP) and, as at 13 July 2011, a SoP concerning Diabetes Mellitus, No 89 of 2011 (Diabetes Mellitus SoP).

OPERATION OF S 120(3) OF THE VE ACT 12 There are "settled and uncontroversial principles" in undertaking the task in s 120(3) as affected by a SoP under s 196B and by s 120A(3). It is often stated that the "Deledio steps" are adopted in the decision making process (Repatriation Commission v Deledio (1998) 83 FCR 82 (Deledio)), namely: 1. The [AAT] must consider all the material which is before it and determine whether that material points to a hypothesis connecting the injury, disease or death with the circumstances of the particular service rendered by the person. No question of fact finding arises at this stage. If no such hypothesis arises, the application must fail. 2. If the material does raise such a hypothesis, the [AAT] must then ascertain whether there is in force an SoP determined by the Authority under s 196B(2) or (11). …. 3. If an SoP is in force, the [AAT] must then form the opinion whether the hypothesis raised is a reasonable one. It will do so if the hypothesis fits, that is to say, is consistent with the "template" to be found in the SoP. The hypothesis raised before it must thus contain one or more of the factors which the Authority has determined to be the minimum which must exist, and be related to the person's service (as required by ss 196B(2)(d) and (e)). If the hypothesis does contain these factors, it could neither be said to be contrary to proved or known scientific facts, nor otherwise fanciful. If the hypothesis fails to fit within the template, it will be deemed not to be "reasonable" and the claim will fail. 4. The [AAT] must then proceed to consider under s 120(1) whether it is satisfied beyond reasonable doubt that the death was not war-caused, or in the case of a claim for incapacity, that the incapacity did not arise from a war-caused injury. If not so satisfied, the claim must succeed. If the [AAT] is so satisfied, the claim must fail. It is only at this stage of the process that the [AAT] will be required to find facts from the material before it. In so doing, no question of onus of proof or the application of any presumption will be involved. The Deledio steps are not applied mechanically and should not be substituted for the statutory test: Hill v Repatriation Commission (2005) 218 ALR 251 at [80] and [85]. 13 The relevant principles were also usefully addressed in Collins v Administrative Appeals Tribunal (2007) 163 FCR 35 at [48]: (a) The [AAT] must consider the whole of the material before it: s 120(3); (b) The Tribunal is to form an opinion whether the material raises a reasonable hypothesis connecting the injury, disease or death with the circumstances of service: s 120(3); (c) The formation of that opinion involves consideration as to whether [the] relevant SoP upholds the hypothesis: s 120A(3); (d) At the stage of [the] formation of the opinion in (b), involving the consideration in (c), no question of fact finding arises: Deledio 83 FCR at 97; (e) The formation of the opinion involves the reaching of a factual conclusion: [Bull v Repatriation Commission (2001) 4 AAR 326; 188 ALR 756 at [17]-[25]] and involves the assessment of all the material before the [AAT], but not the finding of facts or rejecting material: Bull 4 AAR 326; 188 ALR 756 at [22]. (Emphasis added.) 14 Section 120 "neither imposes a presumption that a death is war-caused, nor an onus on a claimant to prove that it is": Forrester v Repatriation Commission [2013] FCA 898 at [11] citing Deledio at 98. Section 120 requires the Commission to determine a death was war-caused "unless it is satisfied, beyond reasonable doubt, that there is no sufficient ground for making that determination": Forrester at [11] (emphasis added). Section 120(3) provides that the Commission is obliged to find that there is "no sufficient ground" that a death was war-caused if "after consideration of the whole of the material before it, [it] is of the opinion that the material before it does not raise a reasonable hypothesis connecting the … death with the circumstances of the particular service rendered by the person": Forrester at [12] (emphasis added). 15 The relevant hypothesis must "find some support" in the material, and the material must "point to, and not merely leave open" the hypothesis relied upon: Forrester at [14] citing the Full Court in East v Repatriation Commission (1987) 16 FCR 517 at 532-533 and Forrester at [31] citing Repatriation Commission v Bey (1997) 79 FCR 364 at 372, 375; see also Deledio at [97]. Whether material points to or supports a hypothesis is a matter which can be determined by inference or assumption: Forrester at [30] citing Elliott v Repatriation Commission (2002) 73 ALD 377 at [5]. As Emmett J stated in Collins at [14], "the first question is whether the [AAT], after consideration of the whole of the material before it, was of the opinion that that material did not raise a reasonable hypothesis connecting the death of the Veteran with the circumstances of his war service". 16 Next, the Tribunal is to determine the reasonableness of the hypothesis. The reasonableness of a hypothesis for the purposes of s 120 is assessed by reference to any relevant SoP in force: s 120A(3). A hypothesis connecting a death with war service will only be reasonable if the material that raises it includes all of the essential elements prescribed by the SoP: Repatriation Commission v Hill (2002) 69 ALD 581 at [55]. A hypothesis will not be reasonable if it is: "(i) contrary to proved or known scientific facts; (ii) obviously fanciful, impossible, incredible, absurd, ridiculous, not tenable, too remote or too tenuous; or (iii) (since 1994) inconsistent with (not upheld by) an applicable SoP": Deledio v Repatriation Commission (1997) 47 ALD 261 at 275; endorsed by the Full Court in Deledio at 96.

THE AAT DECISION 17 The AAT set out the relevant legislative provisions and the Deledio steps: at [28]-[31]. It relevantly stated at [38]-[39]: The hypothesis under consideration must be more than one which is left open on the material before me. It must be a hypothesis which is pointed to, or supported by, the material. For the reasons which I set out in Forrester and Repatriation Commission [2012] AATA 846 I am of the view that the requirement that the material before the [AAT] "points to a hypothesis connecting the ... death with the circumstances of the particular service" (emphasis added) means that the hypothesis advanced must be reasonable. The determination of whether the hypothesis is pointed to or supported by the material or is reasonable is not to be confused with the reasoning required at the third step of the Deledio process. That step requires the determination of whether the hypothesis fits the "template" to be found in any relevant Statement(s) of Principles. If the hypothesis does not fit the template it will be deemed not to be reasonable. (Original emphasis, citations omitted.) 18 Taking each of the Deledio steps: 1. First Step - the AAT stated that it was satisfied that the material before it pointed to two hypotheses connecting Mr Ellis' death to his operational service: at [40]; 2. Second Step - the AAT stated that it was satisfied that two relevant SoPs were in force: the Malignant Neoplasm of the Liver SoP and the Diabetes Mellitus SoP: at [43]; 3. Third Step - the AAT was not satisfied that either of the hypotheses fitted the template or was consistent with the relevant SoP: at [56]-[64]. 19 Although the AAT was satisfied that the material was consistent with the Malignant Neoplasm of the Liver SoP: at [51], the AAT was not satisfied that the whole of the material before it pointed to the requirements for the Diabetes Mellitus SoP having been met: at [59] and [62]. In particular, the AAT noted at [59]-[60] that the factors from the Diabetes Mellitus SoP relied upon in each hypothesis had specific requirements: Clause 6(b)(i) - [a Body Mass Index (BMI) of 25 or greater or a waist circumference of greater than 94 cm] for a period of at least five years before the clinical onset of diabetes mellitus; Clause 6(b)(ii) - an inability to undertake any physical activity greater than three METs for at least the five years before the clinical onset of diabetes mellitus. 20 The AAT considered that as neither hypothesis fitted the template of the Diabetes Mellitus SoP, the hypotheses were not reasonable for the purposes of s 120(3) of the VE Act. The AAT affirmed the decision of the Commission to reject Mrs Ellis' claim for a pension.

QUESTIONS OF LAW 21 Mrs Ellis' appeal to this Court identified three questions of law: 1. Whether there was any material before the [AAT] which raised or pointed to a hypothesis that was consistent with [the Diabetes Mellitus SoP]. 2. Whether the [AAT] misconceived and/or misunderstood its function when applying s 120(3) of the [VE Act]. 3. Whether it was open for the [AAT] to find on the material before it that the material did not point to the veteran being overweight for at least five years before the clinical onset of his diabetes mellitus. 22 The central issue in this appeal is "whether the AAT misconceived and / or misunderstood its function when applying s 120(3) of the VE Act".

ANALYSIS 23 Mrs Ellis' argument was advanced as a series of separate propositions. It will be necessary to deal with those propositions separately. It is appropriate to address Question 2 before addressing Questions 1 and 3.

6.1 Whether the AAT misconceived and / or misunderstood its function when applying s 120(3) of the VE Act - Question 2 24 Mrs Ellis submitted the AAT made three errors of law in its application of s 120(3). Mrs Ellis described the errors in the following terms: First, the [AAT] misstated the first hypothesis put forward by Mrs Ellis linking Mr Ellis' death with his operational service and failed to distinguish between and correctly apply the first and third Deledio steps. Second, the [AAT] misapplied s 120(3) by requiring that each element of the Diabetes Mellitus SoP be supported by evidence tending to establish it when it was only necessary that the hypothesis be supported by the material before the [AAT]. Third, the [AAT] misapplied s 120(3) by embarking on fact finding at step 3 of the Deledio process. Each "error" will be taken in turn.

6.1.1 Identification of the hypothesis and the first and third Deleido steps 25 Mrs Ellis submitted that the AAT misstated the hypothesis put forward by Mrs Ellis linking Mr Ellis' death with his operational service and failed to distinguish between and correctly apply the first and third Deledio steps.

6.1.1.1.1 Relevant facts 26 In Mrs Ellis' Statement of Facts and Contentions before the AAT, the relevant hypothesis was described as: • Injury to the right knee (which it is noted has already been accepted by the Respondent as war caused) reduced ability to exercise weight gain "overweight" diabetes mellitus cancer of the liver death. 27 In her closing submissions before the AAT, Mrs Ellis described the hypothesis in these terms: … The first is that he was overweight and that meaning over 25 BMI for a period of five years before the clinical onset of his diabetes mellitus. Looking at [the Diabetes Mellitus SoP] and it's (sic) factor 6(b)(i). His being overweight in that period we say was related to his service-caused knee injury and his diabetes in turn was contributed by being overweight and then the diabetes in turn was a cause of his liver cancer. 28 In its reasons, the AAT described the relevant hypothesis at [37] as follows: (a) Mr Ellis' knee injury limited his physical activity to such an extent that he became overweight; (b) Mr Ellis' excessive weight contributed to his developing diabetes mellitus; (c) the diabetes mellitus contributed to the development of carcinoma of the liver; (d) carcinoma of the liver caused Mr Ellis' death. 29 The difference between the hypothesis stated by Mrs Ellis in her final submissions and that stated by the AAT is that the AAT did not refer to the requirements from the Diabetes Mellitus SoP - that Mr Ellis have a BMI of over 25 for more than five years before the clinical onset of the diabetes mellitus.

6.1.1.1.2 Parties' respective positions 30 Mrs Ellis submitted that, in breach of its duty to consider the particular hypothesis put forward, the AAT in fact misstated it. Counsel for Mrs Ellis referred to Plaintiff M61/2010E v Commonwealth (2010) 243 CLR 319 as authority for the proposition that the AAT is obliged to address or deal with the case put forward by an applicant. Mrs Ellis further submitted that it is necessary for the hypothesis raised by the material to include the elements prescribed by the SoP or it will not be upheld by the SoP: citing Repatriation Commission v Gosewinckel (1999) 59 ALD 690, 704 at [61]. 31 Mrs Ellis further submitted that as the first Deledio step required the AAT to determine whether the material raised or pointed to facts that supported the hypothesis, the AAT's misstatement of the first hypothesis meant that it failed to properly embark on the required task and did not consider the factual material relied on by Mrs Ellis by reference to the correct hypothesis. Further, Mrs Ellis submitted that when the AAT came to apply the third Deledio step, the same error infected the approach adopted by the AAT because when it embarked on the required matching exercise, the AAT used the wrong starting point, namely the wrong hypothesis. 32 The Commission adopted a different stance. It submitted that the alleged error was no more than a complaint about the degree of detail or precision with which the AAT stated the hypothesis.

6.1.1.1.3 Analysis 33 The critical point in this Appeal can be stated relatively briefly. The AAT stated at [37] the first hypothesis for its consideration as being: (a) Mr Ellis' knee injury limited his physical activity to such an extent that he became overweight; (b) Mr Ellis' excessive weight contributed to his developing diabetes mellitus; (c) the diabetes mellitus contributed to the development of carcinoma of the liver; (d) carcinoma of the liver caused Mr Ellis' death. 34 Later in its reasons at [56] it said: Having considered all of the material before me, I have determined that the first hypothesis is not consistent with clause 6(b)(i) of the [SoP] as the material before me does not point to Mr Ellis being overweight (i.e. having a BMI of 25 or greater or a waist circumference greater than 94 cm) for a period of at least five years before the clinical onset of diabetes, which I have decided was sometime in 1975. (Original emphasis.) 35 Standing alone these two passages of the reasons could be read as indicating that the AAT's statement of the first hypothesis was to be understood as incorporating in the general references to "overweight" and "excessive weight" to the more precise elements of either a BMI exceeding 25 or a waist circumference exceeding 94 cm. 36 But that understanding of the two passages from the reasons set out above is not consistent with what the AAT said at [59] of its reasons: In considering the [Diabetes Mellitus SoP] I note that the requirements as to BMI and waist circumference are precise. … The first hypothesis does not meet the template set out in clause 6(b)(i) as it does not meet the necessary requirements for at least the five years before the clinical onset of diabetes. (Emphasis added.) 37 The question of identification of an hypothesis was considered by Mortimer J in Forrester at [23]: To engage the process in ss 120(1) and 120A, a particular hypothesis must be identified. The language of "hypothesis" is significant: it is a textual indication of at least two matters. First, the scheme's recognition of the difficulties associated with establishing an injury, disease or death as war-caused, sometimes long after the relevant service; second, the beneficial threshold set by ss 120 and 120A. A hypothesis is no more than a supposition or conjectural explanation of an ultimate fact: Repatriation Commission v Stares (1996) 66 FCR 594 at 601 per Black CJ, Ryan and Einfeld JJ. 38 "An hypothesis is a proposition made as a basis for reasoning or a supposition made as a starting point for further investigation from known facts. It is one connecting the death with service" (emphasis added): Bull v Repatriation Commission (2001) 188 ALR 756 at [38]. It is a starting point in a process which is administrative and inquisitorial: see for example Bushell v Repatriation Commission (1992) 175 CLR 408 at 425. Practical consequences follow from that proposition. 39 At the first stage in the Deledio process, there is an obligation or "duty to consider not merely the particular hypothesis put forward but also any other hypothesis fairly raised by the evidence": Hill (2005) 218 ALR 251 at [96]. The obligation or duty then cannot be undertaken without considering the whole of the material before the decision maker: see [13] above. Indeed as Emmett J stated in Collins "the first question is whether the [AAT], after consideration of the whole of the material before it, was of the opinion that that material did not raise a reasonable hypothesis connecting the death of the Veteran with the circumstances of his war service": at [14]. 40 The Deledio steps are a guide: see [12] above. The AAT is not required to follow them mechanically and should not rely upon them in substitution for the statutory test: see [12] above. Although there is nothing in the first Deledio step which requires that the hypothesis being considered refer to the SoP factors, as a matter of practice, a hypothesis will not be upheld at the third Deledio step unless it contains the essential elements of the relevant SoP that is in force. Why? Because at that stage the AAT needs to form an opinion as to whether the material raises a reasonable hypothesis connecting the death with the circumstances of the service, and this involves a consideration as to whether the relevant SoP upholds the hypothesis: see [12] and [13] above. 41 Here, in relation to the first limb of Deledio, the AAT identified an hypothesis but it was incorrect or incomplete: see [26]-[29] above. Relevantly, the AAT then stated: [40] Having considered all of the material before me I determine that it does point to each of the hypotheses put forward connecting Mr Ellis' death to his operational service. [41] The material points to Mr Ellis: • having suffered problems with his knee from the time he was injured in 1947; • not being active and to his engaging in limited physical exercise; • having gained substantial weight which he retained until he was diagnosed as suffering diabetes mellitus; • suffering diabetes mellitus which contributed to the development of carcinoma of the liver which caused his death. 42 Those passages demonstrate that at the first stage of the Deledio process the AAT did not consider the hypothesis put forward by Mrs Ellis exactly as it had been put forward, namely that of Mr Ellis being overweight with a BMI of over 25 for a period of five years before the clinical onset of his diabetes mellitus: see [26] and [27] above. The AAT's reasons for judgment in relation to the first limb are less than adequate. The hypothesis at [37] of the AAT's reasons for judgment is misstated or at the very least sloppily drafted. 43 What about the third limb? The AAT's reasons relevantly included the following at [56]-[59]: [56] Having considered all of the material before me, I have determined that the first hypothesis is not consistent with clause 6(b)(i) of the [SoP] as the material before me does not point to Mr Ellis being overweight (i.e. having a BMI of 25 or greater or a waist circumference greater than 94 cm) for a period of at least five years before the clinical onset of diabetes, which I have decided was sometime in 1975. [57] Of the recorded weight measurements in Mr Ellis' Army records, only that taken in April 1977 gives a BMI of 25 or greater. That measurement was taken after the clinical onset of diabetes. [58] The evidence of Mrs Ellis and of Mr Wells is that Mr Ellis was a large man during the time he lived at the hotel. Mr Richard Ellis gave evidence that Mr Ellis was about 18 stone sometime between 1972 and 1976, although he was unable to be more precise. Dr Prowse gave evidence that in his opinion Mr Ellis had a BMI of 25 or greater and a waist circumference of greater than 94 cm in January 1971. [59] In considering the [Diabetes Mellitus SoP] I note that the requirements as to BMI and waist circumference are precise. … The first hypothesis does not meet the template set out in clause 6(b)(i) as it does not meet the necessary requirements for at least the five years before the clinical onset of diabetes. (Emphasis added.) 44 As these passages in the reasons demonstrate, the AAT appears to have undertaken three quite separate tasks. It identified the first hypothesis, incorrectly (see [28] above), and carried that error through to the third stage: see [56] and [59] extracted above. Next, it looked to see whether material before the AAT pointed to Mr Ellis having a BMI of 25 or greater or a waist circumference greater than 94 cm for a period of at least five years before the clinical onset of diabetes: see [56]-[58] extracted above. (Whether the AAT considered all of the material before it is separately challenged by Mrs Ellis in Questions 1 and 3: see section 6.2 below). Then, the AAT addressed whether the first hypothesis met the template set out in cl 6(b)(i) and concluded that it did not do so. 45 The hypothesis put forward by Mrs Ellis referred to Mr Ellis being overweight with a BMI of over 25 for a period of five years before the clinical onset of his diabetes mellitus: see [26] and [27] above. This detail about Mr Ellis' BMI was an essential element of the relevant SoP: see [59] of the AAT's reasons. The AAT had a duty to consider the particular hypothesis put forward when undertaking the first Deledio step: see [39] above, and it did not do so. 46 I accept that the reasons for decision of the AAT "are not to be construed minutely and finely with an eye keenly attuned to the perception of error" and must be read as a whole: Minister for Immigration and Ethnic Affairs v Wu Shan Liang (1996) 185 CLR 259 at 272 and 291 and Collector of Customs v Pozzolanic Enterprises Pty Ltd (1993) 43 FCR 280 at 286-287. The difficulty here is that the AAT's reasons stated the first hypothesis incorrectly, and this infected other parts of its analysis: see [33]-[45] above. The AAT asked itself the wrong question. The appeal should be allowed. 47 That leads to the next complaint made by Mrs Ellis.

6.1.1.2 "Intertwining" of the first and third Deledio steps 48 Mrs Ellis submitted that in deciding whether the hypothesis was raised, the AAT impermissibly intertwined the first and third Deledio steps at [38] to [39] of its reasons: see [17] above. Mrs Ellis submitted that the AAT conflated reasonableness with the raising of the hypothesis, where there should be a distinction. Mrs Ellis submitted that s 120(3) requires that a hypothesis must be raised and, "separately, the hypothesis so raised must be reasonable". 49 In support of this proposition, Mrs Ellis referred to the decision of Mortimer J in Forrester at [31]-[32] and then again at [72]-[74]: 31 Importantly for the resolution of the issues in this case, the use of the verb "raise" in s 120(3), and the subsequent approach of asking whether material "points to" or "supports" a hypothesis, has been held to require more than that the material before the decision-maker leaves the propounded hypothesis open as a possibility. A possibility of connection between war service and death, injury or disease is not enough: Repatriation Commission v Bey (1997) 79 FCR 364 at 372, 375. 32 The reasonableness of a hypothesis in the context of a determination under s 120(3) therefore has two aspects: a medical or scientific aspect, and what I shall describe as a factual aspect. One looks to the factual circumstances relating to the particular veteran and the other looks to the medical or scientific basis for what is factually asserted. Since the introduction of s 120A in 1994, consideration of these aspects has become somewhat separated. However it is important to recall that they both form part of the assessment of whether a hypothesis is reasonable. A hypothesis will be reasonable if there are facts that point to or support it, but it also needs to be reasonable because a SoP determined under s 196B(2) or (3) (or medical or scientific opinion if there is no SoP) upholds it: see Repatriation Commission v Hill (2002) 69 ALD 581; [2002] FCAFC 192 at [53]. … … 72 To "fit" a SoP, the hypothesis must "fit" the causal aspect of the SoP and in that sense identify a relationship between the factor or factors relied on in the SoP and the veteran's war service. The hypothesis will then be reasonable (in a medical or scientific sense), because it is consistent with what the relevant medical experts have determined is a relationship between a disease, injury or death, and factors causative of that disease, injury or death and war service. 73 However this step in Deledio simply involves comparison between the hypothesis as articulated and the relevant SoP: after the introduction of s 120A, this aspect of determining reasonableness of the hypothesis largely centres on a matching exercise between the asserted hypothesis and the SoP, without a view being formed of particular facts. 74 In contrast, although it may not amount to fact finding, the first step in Deledio has a factual element particular to the material before the decision-maker about the veteran. It is, as I have pointed out at [32] above, a separate but integral aspect of determining the reasonableness of the asserted hypothesis. It involves a different comparison to the third step, and centres much more on the specific factual material relied on by the veteran. The different exercises involved mean, in my opinion, that a decision-maker could reach a conclusion adverse to a veteran on the first step, and - assuming against itself and moving to the following steps - a conclusion favourable to the veteran at the third step. That is because they are distinct aspects of reasonableness. (Emphasis added.) 50 Mrs Ellis also referred the Court to Bushell in support of this proposition. 51 Mrs Ellis' submission that the AAT was "conflating reasonableness with the raising" may be put to one side. As noted above, the Deledio steps are a guide and the AAT is not required to follow them mechanically and should not rely upon them as a substitute for the statutory test. 52 Section 120(3) requires the AAT to determine whether the material before it raises a reasonable hypothesis connecting the death with the circumstances of the particular service rendered by the person. The inquiry is whether there is a reasonable hypothesis. The section does not distinguish between the "raising" and the "reasonableness". As stated in Forrester at [74] the first Deledio step, the raising of a hypothesis, is an aspect of reasonableness. 53 In Repatriation Commission v Warren (2008) 167 FCR 511, a majority of the Full Court considered, and paraphrased, the Deledio steps. In relation to the third Deledio step, Lindgren and Bennett JJ held (at [26]): If there is an SoP in force, it must be asked whether the hypothesis of connection that is raised is a reasonable one. It will be reasonable if the hypothesis fits, that is to say, is consistent with, the "template" to be found in the SoP. To be consistent with that template, the hypothesis raised must contain the factors that the Authority has determined to be those that must as a minimum exist …, and of those factors the hypothesis must contain the factors that the Authority has determined to be the minimum related to the person's service that must exist … . If the hypothesis fits the template in these respects, it cannot be said to be contrary to proved known scientific facts or otherwise fanciful. If, on the other hand, the hypothesis fails to fit the template in these respects, it will be deemed not "reasonable" and the application fails. (Emphasis added.) 54 What these authorities demonstrate is two essential aspects - a hypothesis must be raised and it must be reasonable. It can be done together but it must be done. Here, the process was flawed because the AAT did not state the correct hypothesis (step 1), appeared to consider part of the correct hypothesis at some point but then returned to the incorrect hypothesis at step 3. That is an error.

6.1.1.3 Multiple determinations as to whether the material raised a hypothesis 55 Mrs Ellis' next complaint was that the AAT twice purported to determine whether the material raised a hypothesis (in accordance with the first Deledio step), with inconsistent outcomes. This is not an accurate description of the task undertaken by the AAT: see [44] above. However, the process adopted by the AAT was flawed: see [33]-[46] above.

6.1.2 Need for more precise evidence 56 Next, Mrs Ellis submitted that in determining that the first hypothesis was not consistent with the Diabetes Mellitus SoP, the AAT reasoned (at [58]-[59]) that: 1. The requirements as to BMI and waist circumference in the SoP are precise; 2. By implication, there was a need for more precise evidence that Mr Ellis was overweight (with a BMI of 25 or above) for at least five years before the clinical onset of diabetes; 3. The material before it did not include such evidence; and 4. Therefore the first hypothesis was not consistent with the SoP. 57 Mrs Ellis submitted that this reasoning involved a misapplication of s 120(3). It was submitted that s 120(3) of the VE Act requires no more than that the hypothesis connecting the injury or disease with the veteran's service be "raised" by the material before the AAT, and that the AAT had misconceived what was required for the material to raise a hypothesis by implying there was a need for more precise evidence. In particular, Mrs Ellis submitted that the decision-maker is not to engage in fact finding during the first three Deledio steps: Tunks v Repatriation Commission (2008) 102 ALD 274 at [35] citing Deledio at 97 and Bull at [18]-[23]. 58 Although it was necessary that the hypothesis be supported by some evidence, there was no requirement that there be evidence to support the hypothesis at every point: Repatriation Commission v Stares (1996) 66 FCR 594 at 599-600. In other words, Mrs Ellis submitted it was only necessary that the hypothesis, including any element based upon an assumption, be supported by the material before the AAT: Repatriation Commission v Glanville (2010) 114 ALD 616 at [57]; cf East at 533; and Hill (2002) 69 ALD 581 at [54]. Accordingly, Mrs Ellis submitted that a hypothesis may assume or infer the occurrence or existence of a "fact" or element of the hypothesis if the assumption or inference is pointed to by the material before the decision-maker: Elliott at [5] and Byrnes v Repatriation Commission (1993) 177 CLR 564 at 570-571. In that context, Mrs Ellis challenged the AAT's reliance upon East in holding that the hypothesis would only be reasonable if the material before it pointed to the precise requirements of cl 6(b)(i) having been met. Mrs Ellis submitted that the AAT took what was said in East at 533 out of context. 59 There is no identified error. It was common ground that it was not necessary for every element of the hypothesis to be supported, or pointed to, by the material before the AAT. It is the essential elements of the hypothesis that must be addressed. 60 In Hill (2002) 69 ALD 581 at [54]-[55], the Full Court held: If an essential element of a hypothesis is not raised (or pointed to) by the material before the decision-maker, then the hypothesis is not raised by that material ... Where s 120A(3) of the Act applies to a claim and there is a relevant SoP in force, whether or not an element is essential to a hypothesis will depend upon the terms of the SoP. As Goldberg J said in Repatriation Commission v McKenna (1998) 52 ALD 72 at 80: For the purposes of s 120A(3) of the Act the hypothesis which has to be upheld by a [SoP] is the hypothesis which connects the disease suffered by a veteran with the circumstances of his service. So stated, the hypothesis has to point to a connection which starts with the disease in respect of which the application is made and ends with the service. That connection will comprise a number of links or factors each of which must be upheld by a [SoP] ... On appeal, the Full Court approved these propositions: see McKenna v Repatriation Commission (1999) 86 FCR 144 at 150-1. Put another way, a hypothesis connecting a disease with war service will only be reasonable if the material that raises it includes all of the essential elements prescribed by the SoP … 61 The Full Court in Hill (2002) 69 ALD 581 (at [57]) then dealt with a submission about the validity of hypotheses built upon assumed facts and stated: Whatever the situation may have been in relation to claims before 1 June 1994, the effect of s 120A(3) (where there is an SoP under s 196B(2)) is that the hypothesis is reasonable only if it is upheld by the SoP. Pursuant to s 196B(2), the SoP must set out "the factors that must as a minimum exist" and "which of those factors must be related to service". The result is that, where it applies, the SoP prescribes the essential content of what is a reasonable hypothesis, for s 120(3) purposes, capable of connecting the particular kind of injury, disease or death with the circumstances of a veteran's particular service. In order to satisfy ss 120(3) and 120A(3), a hypothesis relied on by a veteran to support a pension claim must be supported by material pointing to each element that the SoP makes essential for the hypothesis to be reasonable. See also Connors v Repatriation Commission (2000) 59 ALD 61 at [18]-[19]. . 62 What then was the material and what were the facts that were said to be inferred from that material in relation to Mr Ellis? Mrs Ellis submitted that the material before the AAT raised the hypothesis that Mr Ellis was overweight for at least five years before the clinical onset of his diabetes. In particular, that there was material that Mr Ellis was overweight within the meaning of the SoP for more than five years before he was diagnosed with diabetes and thereafter changed his diet and started to lose weight. Mrs Ellis submitted that the AAT's failure to appreciate this was indicative of it having misconceived its function. Counsel for Mrs Ellis also placed considerable reliance on the decision in Tunks. In that case, the Federal Court identified two errors in the AAT's decision: 1. The conclusion that there was no evidence from which a possible assessment of the deceased's dietary habits before and after service could be made: s120(3); and 2. The conclusion that there was no evidence or no sufficient evidence to sustain a positive finding that there was a connection between the disease and the particular service: s 120(1). Mrs Ellis sought to draw an analogy between the first of these errors and the present case. 63 These submissions are rejected. If the AAT had approached its task in the manner contended by Mrs Ellis, the AAT would have been in error. It is not open to the AAT to infer or assume that the essential elements of a hypothesis are met. Its task was and remains that described earlier - to determine whether the whole material raises a reasonable hypothesis connecting the injury or disease with the circumstances of the particular service rendered by the veteran: see [12]-[16] above.

6.1.3 Impermissible fact finding 64 The AAT's task has been earlier described: see [12]-[16] above. The AAT stated that it had assessed the whole of the material before it and found that it did "not point to Mr Ellis being overweight (i.e. having a BMI of 25 or greater or a waist circumference greater than 94 cm) for a period of at least five years before the clinical onset of diabetes", and therefore the first hypothesis did not fit the template of cl 6(b)(i) of the Diabetes Mellitus SoP. 65 Mrs Ellis submitted that in arriving at this conclusion, the AAT must have discounted or rejected identified material and thereby embarked on impermissible fact finding at the first to third steps of the Deledio process. Given the views formed (see [44]-[46] above), this submission does not need to be addressed because the AAT's error was that it asked itself the wrong question and got the wrong answer. 66 Next Mrs Ellis submitted that the AAT impermissibly entered upon fact finding when it found that the date of clinical onset of diabetes mellitus was sometime in 1975: [35]-[36] and [56] of the AAT's reasons. Mrs Ellis submitted that contrary to [35] of the AAT's reasons, there was no agreement that the date of clinical onset was sometime in 1975. Instead, Mrs Ellis submitted that there was material before the AAT pointing to the date of clinical onset being sometime in the period 1973 to 1976. I accept that Mrs Ellis made this point in her closing submissions before the AAT. The question of fact finding in this context has been addressed in Deledio and Collins. The task for the AAT is to "consider all the material which is before it and determine whether that material points to a hypothesis connecting the injury, disease or death with the circumstances of the particular service rendered by the person. No question of fact finding arises at this stage". In this respect, the AAT was in error.

6.2 Failing to consider all the material before it - Questions 1 and 3 67 As we have seen, the AAT concluded that having considered all of the material before it, the first hypothesis was not consistent with clause 6(b)(i) of the Diabetes Mellitus SoP as the material did not point to Mr Ellis being overweight (i.e. having a BMI of 25 or greater or a waist circumference greater than 94 cm) for a period of at least five years before the clinical onset of diabetes. 68 On appeal, Mrs Ellis submitted that, contrary to the requirements in s 120(3), the AAT failed to consider all of the material before it in determining whether the material raised a reasonable hypothesis. Mrs Ellis submitted that despite the AAT's express statements at [40] and [56] that it had considered all of the material before it, it did not do so. 69 Mrs Ellis identified the omitted material as: (a) the evidence of Mrs Ellis that Mr Ellis was overweight and more particularly, the size he was in the January 1971 photograph during the whole time they operated the hotel business (1964 to 1973); (b) the evidence of Mr Richard Ellis pointing to Mr Ellis being diagnosed with diabetes sometime between about the time he left the pub (in 1973) to 1976; (c) the evidence of Mr Richard Ellis that it was only after Mr Ellis was diagnosed with diabetes that he started to lose weight on medical advice and that he became very careful about his diet after the diagnosis and lost weight rapidly; (d) a gap in the weights recorded in Mr Ellis' army records between 1953 and 1976; (e) a recorded weight in October 1976 of 13 stone which was very slightly under 25 BMI and which could have been either a weighed or stated record; (f) the evidence of Mrs Ellis that Mr Ellis had not lost any weight when she next saw him in about 1976; (g) the evidence of Mr Richard Ellis that he saw Mr Ellis two to three times per year until 1976 and that it was only from about 1976 that he began to lose weight. 70 I will deal with each item in turn.

6.2.1 Item (a): the evidence of Mrs Ellis that Mr Ellis was overweight and more particularly, the size he was in the January 1971 photograph during the whole time they operated the hotel business (1964 to 1973) 71 Mrs Ellis placed particular importance on the submission that the AAT had not referred to item (a) in its reasons for decision. There were two limbs to this item (a): 1. Mrs Ellis gave evidence that Mr Ellis was the size that he was in the January 1971 photograph during the whole time that they operated the hotel business (from 1964-1973); and 2. Dr Prowse had identified the man in the photograph (Mr Ellis) as having a BMI of 25 or greater in January 1971. Mrs Ellis submitted that this evidence taken together pointed to Mr Ellis having a BMI of 25 or greater from 1964 to 1973. 72 Counsel for Mrs Ellis submitted that the evidence of Dr Prowse was accepted by the AAT (at [24] and [58]), but Mrs Ellis' evidence that he was the same weight that he was in the photograph was not referred to in the reasons. It was submitted that taken together, this evidence was far more precise than just "he was overweight", and was the BMI evidence for which the AAT was looking. Counsel for Mrs Ellis submitted that this was only relevant in a context where the AAT was searching for more precise evidence as to Mr Ellis' BMI.

6.2.2 Item (b): the evidence of Mr Richard Ellis pointing to Mr Ellis being diagnosed with diabetes sometime between about the time he left the pub (in 1973) to 1976 73 The date of the onset of Mr Ellis' diabetes was significant, as under the Diabetes Mellitus SoP, the factors relied upon needed to exist for a period of at least five years before the clinical onset of the diabetes: see [19] above. 74 Mrs Ellis submitted that omitting evidence of Mr Richard Ellis pointing to Mr Ellis being diagnosed with diabetes sometime between about the time he left the pub (in 1973) to 1976 was significant, as it would have an impact on the question of whether Mr Ellis had been overweight for five years before the onset of the diabetes mellitus. This submission was also related to Mrs Ellis' contention that AAT had impermissibly embarked on fact-finding in relation to the time of onset of the diabetes mellitus, discussed at [66] above.

6.2.3 Item (c): the evidence of Mr Richard Ellis that it was only after Mr Ellis was diagnosed with diabetes that he started to lose weight on medical advice and that he became very careful about his diet after the diagnosis and lost weight rapidly 75 At the hearing, Counsel for Mrs Ellis accepted that reference to this material could be found in [20] of the AAT's reasons.

6.2.4 Item (d): a gap in the weights recorded in Mr Ellis' army records between 1953 and 1976 76 At the hearing, Counsel for Mrs Ellis accepted that reference to this material could be found at [20] and [27] of the AAT's reasons. However, Counsel submitted that there was a misstatement of the evidence in that paragraph, which lead to the omission identified at (e).

6.2.5 Item (e): a recorded weight in October 1976 of 13 stone which was very slightly under 25 BMI and which could have been either a weighed or stated record 77 The weight record in [27] of the AAT decision for 8 October 1976 should read 13 stone not 12 stone. I accept that this is a misstatement in the AAT's reasons. 78 Counsel for Mrs Ellis submitted that this was significant as Mr Ellis' BMI would only have been slightly under 25 if he weighed 13 stone, whereas his BMI would be notably lower if he weighed 12 stone. Mr Ellis' weight at this time was said to be relevant as Mrs Ellis submitted that the time of clinical onset of the diabetes mellitus was sometime between 1973 and 1976.

6.2.6 Item (f): the evidence of Mrs Ellis that Mr Ellis had not lost any weight when she next saw him in about 1976 79 Counsel for Mrs Ellis accepted that this material could be inferred from [12] of the AAT's reasons.

6.2.7 Item (g): the evidence of Mr Richard Ellis that he saw Mr Ellis two to three times per year until 1976 and that it was only from about 1976 that he began to lose weight 80 Counsel for Mrs Ellis submitted that the evidence of Mr Richard Ellis was that he saw Mr Ellis for two to three times a year from 1968 and that [19] of the AAT's reasons contained a misstatement. I accept that Mr Richard Ellis gave evidence that he saw his brother two or three times a year from 1968 to 1976. Mr Richard Ellis then gave evidence that he thought his brother weighed about 18 stone during that period, and he only started to lose weight from around 1976. This was said to be relevant as it would go further back into the five year period outlined under the Diabetes Mellitus SoP.

6.2.8 Contentions and Analysis 81 Mrs Ellis submitted that the failure to refer to this evidence was significant as the AAT's finding appeared to turn on the fact that there was not precise enough evidence. It must be recalled that at [59] the AAT stated that the requirements of the SoP were precise. Mrs Ellis submitted that it can be inferred that the AAT considered that there was a need for more precise evidence as to the BMI for the five years before onset than what was before the AAT (as the AAT understood the evidence to be). 82 Mrs Ellis submitted that the Court should infer from the AAT's failure to expressly refer to one or more of (a), (b), (e) or (g), or to give any weight to them, that they were not taken into account. 83 In support of that submission, Mrs Ellis referred to the following propositions: 1. Determining whether a matter has been considered is "essentially an evaluative process based exclusively on what the decision-maker has said or written": Anderson v Director General of the Department of Environmental and Climate Change (2008) 251 ALR 633 at [58] and Victoria Police Toll Enforcement v Taha; State of Victoria v Brookes [2013] VSCA 37 at [220]; 2. The Court is entitled to take the reasons of an administrative tribunal as setting out the findings on questions of fact which the tribunal considered were material to its decision: Minister for Immigration and Multicultural Affairs v Yusuf (2001) 206 CLR 323 at 331-2 [10], 338 [34] and 346 [68]; 3. Failing to make relevant findings can indicate that the tribunal misunderstood its duty and/or asked itself the wrong question: Yusuf at 338, 340, 346 and 349-50; 4. A failure to include reference to a matter in a statement of reasons may justify the inference that it was not taken into account. Whether the inference will be drawn in a particular case depends on all the circumstances: Bat Advocacy NSW Inc v Minister for Environment Protection, Heritage and the Arts [2011] FCAFC 59 at [46]-[47]; 5. Failure to refer to important evidence in reasons may more readily lead to an inference that the evidence was not considered: Applicant WAEE v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 75 ALD 630, 641 [47]. 84 Additionally, Mrs Ellis submits that as the evidence was referred to in oral submissions, and not referred to in the reasons, this gives a strong indication that the AAT did not consider that evidence: LVR (WA) Pty Ltd v Administrative Appeals Tribunal [2012] FCAFC 90 at [128]. 85 Standing alone these omissions may or may not justify the conclusion that the AAT made an error of law. However, it is not necessary to deal with these issues in isolation from the other errors upon which Mrs Ellis relies. It is enough to say that for at least the most part the alleged omissions might be seen as supporting the conclusion otherwise reached that the AAT did not properly identify the question to be asked and then answered: see section 6.1 above.

CONCLUSION 86 The appeal should be allowed with costs. Despite the parties' agreement that it was open to the Court to substitute a decision in relation to Mrs Ellis' entitlement to a pension under the VE Act, the appropriate course is for the matter to be remitted to the AAT to be determined according to law. I certify that the preceding eighty six (86) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Gordon.